1. science
2. chemistry

METHOD DEVELOPMENT ON SAMPLE PREPARATION FOR TRACE
METALS IN PETROLEUM PRODUCTS PRIOR TO THEIR DETERMINATION
USING INDUCTIVELY COUPLED PLASMA- SPECTROMETRIC TECHNIQUES
By
PHILISWA NOSIZO NOMNGONGO
Thesis in fulfilment of the requirement for the degree
PHILOSOPHIAE DOCTOR
in
CHEMISTRY
in the
FACULTY OF SCIENCE
of the
UNIVERSITY OF JOHANNESBURG
Supervisor
Co-supervisors
:
:
:
PROF. J. C. NGILA
PROF. T. A. M. MSAGATI
DR. B. MOODLEY
DECLARATION
I hereby declare that this dissertation, which I herewith submit for the research
qualification
DOCTORIAL DEGREE IN CHEMISTRY
to the University of Johannesburg, Department of Applied Chemistry, is, apart from the
recognised assistance of my supervisors, my own work and has not previously been
submitted by me to another institution to obtain a research diploma or degree.
__Philiswa N Nomngongo________ on this __20th__ day of ___March__2014__
(Candidate)
___Prof J Catherine Ngila__ on this _20th__ day of ___March__2014____
(Supervisor)
______Prof Titus Msagati__ on this _20th__ day of ___March__2014_
(Co-supervisor)
_______Dr Brenda Moodley_____ on this _20th__ day of ___March__2014_
(Co-supervisor)
i
DEDICATION
This thesis is dedicated to my mother (Makhawula Buyiswa Nomngongo), my husband
(Albert Molete Kaphe) and my sons (Sisipho and Kamohelo) for the understanding and
encouragement they provided me during the years of this study.
“And we know that all things work together for good to them that love God, to them who
are the called according to his purpose”. Romans 8:28
ii
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to my supervisor Prof. J.C. Ngila for the
continuous support of my PhD study and research, for her patience, motivation,
enthusiasm, and immense knowledge. Her guidance helped me through the duration of
research and writing of this thesis. I could not have imagined having a better supervisor
and mentor for my PhD study.
I sincerely thank Prof. T.A.M. Msagati and Dr. B. Moodley for their helpful advice and
having them as my co-supervisors. Their extensive discussions around my work and
interesting explorations in operations have been very helpful for this study.
I would like to acknowledge the University of Johannesburg (UJ) for giving me an
opportunity to pursue my PhD Degree at this institution. Special thanks to the Department
of Applied Chemistry, Department of Chemistry and Spectrau at UJ for providing all
facilities and particularly the instrumentation required for my research.
I express my deep gratitude and appreciation to my mentor Dr. J.L. Fischer of Sasol for
his input and help from the formulation to the completion of this study. I am also thankful
to Mrs. Eve Kroukamp, of Spectrau (central analytical facility at UJ), for her help with ICP
OES and ICP-MS instrumentation. I would also like to thank Mr M. Phali of Department
Chemistry at UJ, for assisting me with GFAAS operation.
I would like to thank my fellow colleagues in the Analytical Research Group: Joseph
Kamau, Isaac Mwangi, Stephen Musyoka, Monisola Ikhile, Banele Vatsha, Nomvano
Mketo, Geoffrey Bosire, Richard Nthumbi, Odwa Mapazi, Lwazi Ndlwana, Bonani Seteni,
Mogolodi Dimpe and Bhekie Mabhena for the motivating discussions, working together
before deadlines, and for all the fun we have had.
Many thanks to my family members especially my mother, my husband, my son and
siblings for their patience, love, support and encouragement during my period of study.
I recognize that this research would not have been possible without the financial
assistance of Sasol and NRF.
iii
PUBLICATIONS AND CONFERENCES
This study resulted in TEN (10) manuscripts that form part of this thesis. These include
published articles, accepted and submitted manuscripts, and manuscripts in preparation.
Philiswa N. Nomngongo, J. Catherine Ngila, Titus A.M. Msagati & Brenda Moodley.
2014. Chemometric optimization of hollow fiber-liquid phase microextraction for
extraction and preconcentration of trace elements in diesel and gasoline prior to their
ICP OES determination. Microchemical Journal, 114, 141-147.
Philiswa N. Nomngongo, J. Catherine Ngila, Joseph N. Kamau, Titus A.M. Msagati &
Brenda Moodley. 2013. Preconcentration of molybdenum, antimony and vanadium in
gasoline samples using Dowex 1-x8 resin and their determination with inductively
coupled plasma-optical emission spectrometry. Talanta, 110, 153-159.
Philiswa N. Nomngongo, J. Catherine Ngila, Stephen M. Musyoka, Titus A.M. Msagati &
Brenda Moodley. 2013. A solid phase extraction procedure based on electrospun
cellulose-g-oxolane-2,5-dione nanofibers for trace determination of Cd, Cu, Fe, Pb and
Zn in gasoline samples by ICP OES. Analytical Methods, 5, 3000-3008.
Philiswa N. Nomngongo, J. Catherine Ngila, Joseph N. Kamau, Titus A.M. Msagati,
Ljiljana Marjanovic & Brenda Moodley. 2013. Pre-concentration of trace elements in
short chain alcohols using different commercial cation exchange resins prior to ICP
OES detection. Analytica Chimica Acta, 787, 78-86
Philiswa N. Nomngongo, J. Catherine Ngila, Titus A.M. Msagati & Brenda Moodley.
2013. Simultaneous preconcentration of trace elements in water samples using Dowex
50W-x8 and Chelex-100 resins prior to their determination using inductively coupled
plasma optical emission spectrometry (ICP OES). Physics and Chemistry of the Earth,
66, 83–88
Philiswa N. Nomngongo, J. Catherine Ngila, Titus A.M. Msagati & Brenda Moodley.
2014. Kinetics and equilibrium studies for the removal of cobalt, manganese and silver
in ethanol using Dowex 50W-x8 cation exchange resin: Separation Science and
Technology (Accepted).
Philiswa N. Nomngongo, J. Catherine Ngila, Titus A.M. Msagati & Brenda Moodley.
Multivariate optimization of dual-bed solid phase extraction for preconcentration of Ag,
Al, As and Cr in gasoline prior to inductively coupled plasma optical emission
spectrometric determination: Submitted to Fuel.
iv
Publications and conferences
Philiswa N. Nomngongo, J. Catherine Ngila, Titus A.M. Msagati & Brenda Moodley. Full
factorial design for the optimization of simultaneous preconcentration of trace metal
ions in gasoline samples prior to their inductively coupled mass spectrometric
determination: Submitted to Microchimica Acta.
Philiswa N. Nomngongo, J. Catherine Ngila, Titus A.M. Msagati & Brenda Moodley.
Development and optimization of an offline hollow fiber solid phase microextraction
system for preconcentration of trace metal ions in fuel samples prior to their ICP-MS
determination: Submitted to Spectrochimica Acta Part B: Atomic Spectroscopy
Philiswa N. Nomngongo, J. Catherine Ngila, Titus A.M. Msagati & Brenda Moodley.
Preparation of titania-alumina hollow fiber membrane and its multivariate optimization
for simultaneous preconcentration of trace elements in diesel and gasoline samples prior
to ICP-MS determination: Submitted to Analytica Chimica Acta,
v
ABSTRACT
The main objective of this study was to develop sample preparation methods based on
separation and preconcentration of trace metals in organic matrices (alcohols, diesel and
gasoline). The presence of metals in organic matrices is undesirable, unless they are used
as additives. Therefore, it is important to control and monitor their concentrations in fuel
and petrochemical products as they tend to affect the quality of these products. Solid phase
extraction (SPE), solid phase microextraction (SPME) and hollow fiber-liquid phase
microextraction (HF-LPME were used for separation and preconcentration of Ag, Al, As,
Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, Sb, Ti, V and Zn prior to their inductively coupled
plasma optical emission/mass spectrometric (ICP OES/-MS) determination.
For solid phase extraction, the exchange efficiencies of different commercial ion
exchange resins, namely Dowex 50W-x8, Dowex 1-x8, Dowex MAC-3 and Chelex 100,
for preconcentration of metal ions in alcohol and fuel samples, were investigated. The
results obtained indicated that Dowex 50W-x8 was suitable for simultaneous
preconcentration of cations such as Cd, Co, Ni, Cu, Fe, Mn and Zn, among other metals
while Dowex 1-x8 was suitable for metal ions that exists in more than one oxidation states,
namely, As, Cr, Mo, Sb and V. Chelex-100 and Dowex MAC-3 were found to be selective
to a limited number of target metal analytes. For further applications, Dowex 50W-x8 and
Dowex 1-x8 were employed. Furthermore, the applicability of synthetic adsorbents such as
nanometer-sized alumina and functionalized cellulose nanofibers for preconcentration of
trace metals in gasoline samples, was investigated. Nanometer-sized alumina sorbent was
found to be suitable for simultaneous separation and preconcentration of Co, Cr, Mn, Ni
and Ti. Functionalized cellulose nanofibers on the hand, were suitable for simultaneous
preconcentration of Cd, Cu, Fe, Pb and Zn. The optimization of the experimental
parameters was achieved by both univariate and multivariate procedure.
The second preconcentration technique was solid phase microextraction which was
also used for extraction and enrichment of metal ions in diesel samples using two
approaches of the SPME method. The first approach was hollow fiber-solid phase
microextraction (HF–SPME) method using fiber-supported sol-gel combined with a cation
exchange resin (Dowex 50W-x8). This method showed satisfactory results for the
preconcentration of Cd, Cu, Fe, Pb and Zn in diesel and gasoline samples. The second
approach was based on membrane solid phase microextraction (MSPME) using titaniaalumina hollow fiber. The MSPME method applied was used for extraction and
vi
Abtract
preconcentration of trace amounts of Co, Cr, Mo, Ni, Sb and V in liquid fuel samples.
Multivariate techniques were used for optimization of experimental parameters for both
approaches.
The last preconcentration technique that was developed was hollow fiber- liquid phase
microextraction (HF-LPME). In this method fuel samples were first digested before being
subjected to HF-LPME system. Ammonium pyrrolidine dithiocarbamate (APDC) and
[C6MIM][PF6] ionic liquid were both used as chelating agent and acceptor phase,
respectively. Two level factorial and central composite designs were used for multivariate
optimization of experimental parameters. Satisfactory results were obtained for extraction
and preconcentration of Ag, Al, As, Mn and Ti.
vii
TABLE OF CONTENTS
TABLE OF CONTENTS ................................................................................................. viii
LIST OF FIGURES .......................................................................................................... xix
LIST OF TABLES .......................................................................................................... xxiii
LIST OF ABBREVIATIONS ...................................................................................... xxviii
CHAPTER ONE: ................................................................................................................. 1
INTRODUCTION ............................................................................................................... 1
1.1 BACKGROUND ........................................................................................................ 1
1.1.1 Petroleum Products ............................................................................................... 1
1.1.1.1 Petrochemical products.................................................................................. 3
1.2 METAL IONS IN PETROLEUM AND PETROCHEMICAL PRODUCTS ...... 4
1.3 PROBLEM STATEMENT ....................................................................................... 5
1.4 HYPOTHESIS ........................................................................................................... 6
1.5 OBJECTIVES OF THE STUDY ............................................................................. 6
1.5.1 Main Objective ..................................................................................................... 6
1.5.2 Specific Objectives ............................................................................................... 7
1.6 THESIS OVERVIEW ............................................................................................... 8
1.7 REFERENCES .......................................................................................................... 9
CHAPTER TWO: ............................................................................................................. 12
LITERATURE REVIEW ON DETERMINATION AND SAMPLE PREPARATION
METHODS FOR TRACE METALS ANALYSIS IN PETROLEUM BASED
PRODUCTS ....................................................................................................................... 12
2.1 SAMPLE PRETREATMENT METHODS .......................................................... 12
2.1.1 Non-Sorption Sample Preparation Methods ....................................................... 13
2.1.1.1 Conventional ashing and acid dissolution methods ..................................... 13
2.1.1.2 Microwave assisted-digestion method......................................................... 13
2.1.1.3 Electrothermal vaporization method ........................................................... 14
2.1.1.4 Dilution with organic solvents method........................................................ 14
2.1.1.5 Emulsion or microemulsion method ........................................................... 15
2.1.2 Preconcentration-Based Sample Preparation Techniques .................................. 15
2.1.2.1 Liquid-liquid extraction (LLE) .................................................................... 16
2.1.2.2 Co-precipitation ........................................................................................... 16
2.1.2.3 Cloud point extraction (CPE) ...................................................................... 16
viii
Table of contents
2.1.2.4 Stir-bar sorptive extraction (SBSE) ............................................................. 17
2.1.2.5 Solid phase extraction.................................................................................. 17
2.1.2.6 Solid phase microextraction (HF-SPME) .................................................... 21
2.1.2.7 Hollow fibre-liquid phase microextraction (HF-LPME) ............................. 21
2.2 ANALYTICAL TECHNIQUES FOR THE DETERMINATION OF METAL
IONS IN ORGANIC SAMPLES .................................................................................. 22
2.2.1 Flame Atomic Absorption Spectrometry............................................................ 23
2.2.2 Electrothermal Atomic Absorption Spectrometry .............................................. 23
2.2.3 Inductively Coupled Plasma (ICP) Techniques.................................................. 24
2.2.3.1 Inductively coupled plasma-optical emission spectrometry........................ 26
2.2.3.1.2 Spectrometer ......................................................................................... 27
2.2.3.1.3 Detector ................................................................................................ 28
2.2.3.1.4 Data processing system ........................................................................ 29
2.2.3.1.5 Interferences ......................................................................................... 29
2.2.3.2 Inductively coupled plasma-mass spectrometry .......................................... 30
2.2.3.2.1 Sample introduction for ICP-MS.......................................................... 30
2.2.3.2.2 Interfaces .............................................................................................. 31
2.2.3.2.3 Ion Focusing Systems ........................................................................... 31
2.2.3.2.4 Mass spectrometer/ mass analyzers ...................................................... 31
2.2.3.2.5 Reaction/ collision cell ......................................................................... 32
2.2.3.2.6 Detector ................................................................................................ 33
2.2.3.2.7 Interferences ......................................................................................... 33
2.2.4.3 Application of ICP OES and ICP-MS for determination of metal ions in fuel
samples .................................................................................................................... 34
2.3 CHEMOMETRIC TOOLS FOR OPTIMIZATION OF ANALYTICAL
METHODOLOGIES .................................................................................................... 36
2.3.1 First Order Designs ............................................................................................. 37
2.3.2 Second Order Designs ........................................................................................ 37
2.4 REFERENCES ........................................................................................................ 38
CHAPTER THREE: ......................................................................................................... 49
GENERAL METHODOLOGIES .................................................................................... 49
3.1 OVERVIEW OF EXPERIMENTAL DESIGN .................................................... 49
3.2 INSTRUMENTATION ........................................................................................... 49
ix
Table of contents
3.3 REAGENTS AND MATERIALS .......................................................................... 51
3.4 SEPARATION AND PRECONCENTRATION TECHNIQUES ....................... 51
3.4.1 Solid Phase Extraction ........................................................................................ 51
3.4.1.1 Electrospinning and functionalization of cellulose nanofibers with oxolane2,5-dione .................................................................................................................. 52
3.4.2 Solid Phase Microextraction (SPME)................................................................. 53
3.4.2.1 Preparation of nanometer-sized alumina and titania ................................... 53
3.4.3 Hollow Fiber-Liquid Phase Microextraction ...................................................... 54
3.4 ACID DIGESTION METHODS ............................................................................ 55
3.5 DETERMINATION OF METAL IONS ............................................................... 55
3.6 REFERENCES ........................................................................................................ 56
CHAPTER FOUR: ............................................................................................................ 58
PRECONCENTRATION OF TRACE MULTI-ELEMENTS IN WATER SAMPLES
USING DOWEX 50W-X8 AND CHELEX-100 RESINS PRIOR TO THEIR
DETERMINATION USING INDUCTIVELY COUPLED PLASMA ATOMIC
EMISSION SPECTROMETRY (ICP OES) ................................................................... 58
ABSTRACT ................................................................................................................... 58
4.1 INTRODUCTION ................................................................................................... 58
4.2 EXPERIMENTAL .................................................................................................. 60
4.2.1 Instrumentation ................................................................................................... 60
4.2.2 Reagents and Solutions....................................................................................... 60
4.2.3 Water Samples and Preparation .......................................................................... 60
4.2.4 Column Preparation ............................................................................................ 61
4.2.5 Preconcentration Procedure ................................................................................ 61
4.2.6 Optimization of Preconcentration Parameters .................................................... 61
4.3. RESULTS AND DISCUSSION ............................................................................. 62
4.3.1 Effect of pH ........................................................................................................ 62
4.3.2 Effect of Eluent Concentration ........................................................................... 63
4.3.3 Effect of Flow Rate............................................................................................. 64
4.3.4 Preconcentration of Multi-Element .................................................................... 65
4.3.5 Effect of Sample Volume ................................................................................... 65
4.3.6 Column Regeneration ......................................................................................... 66
4.3.7 Analytical Performances..................................................................................... 67
x
Table of contents
4.3.8 Application ......................................................................................................... 68
4.4 CONCLUSIONS ...................................................................................................... 71
4.5 REFERENCES ........................................................................................................ 71
CHAPTER FIVE: .............................................................................................................. 74
KINETICS AND EQUILIBRIUM STUDIES FOR THE REMOVAL OF COBALT,
MANGANESE AND SILVER IN ETHANOL USING DOWEX 50W-X8 CATION
EXCHANGE RESIN ......................................................................................................... 74
ABSTRACT ................................................................................................................... 74
5.1 INTRODUCTION ................................................................................................... 74
5.2 EXPERIMENTAL .................................................................................................. 76
5.2.1 Materials and Reagents....................................................................................... 76
5.2.2 Apparatus ............................................................................................................ 76
5.2.3 Adsorption Studies ............................................................................................. 77
5.2.4 Kinetic Studies.................................................................................................... 78
5.2.5 Adsorption Thermodynamics ............................................................................. 78
5.3 RESULTS AND DISCUSSION .............................................................................. 78
5.3.1 Effect of Contact Time ....................................................................................... 78
5.3.2 Effect of pH ........................................................................................................ 79
5.3.3 Effect of Resin Amount ...................................................................................... 80
5.3.4 Adsorption Isotherms and Comparison to Other Adsorbents ............................. 81
5.3.5 Adsorption Kinetics ............................................................................................ 87
5.3.6 Adsorption Thermodynamics ............................................................................. 90
5.3.7 Desorption Studies.............................................................................................. 91
5.3.8 Analytical Performance and Application of the Proposed Method .................... 92
5.4. CONCLUSIONS ..................................................................................................... 94
5.5 REFERENCES ........................................................................................................ 94
CHAPTER SIX: ................................................................................................................. 99
PRE-CONCENTRATION
OF
TRACE
ELEMENTS
IN
SHORT
CHAIN
ALCOHOLS USING DIFFERENT COMMERCIAL CATION EXCHANGE
RESINS PRIOR TO INDUCTIVELY COUPLED PLASMA-OPTICAL EMISSION
SPECTROMETRIC DETECTION ................................................................................. 99
ABSTRACT ................................................................................................................... 99
6.1 INTRODUCTION ................................................................................................... 99
xi
Table of contents
6.2 EXPERIMENTAL ................................................................................................ 101
6.2.1 Apparatus .......................................................................................................... 101
6.2.2 Reagents and Solutions..................................................................................... 102
6.2.3 Preparation of Column ..................................................................................... 103
6.2.4 Preconcentration and Recovery of the Metal Ions in Model Organic Solution 103
6.2.5 Effect of Matrix Ions Interferences .................................................................. 104
6.2.6 Procedure for the Dilution of Conostan Custom Made Multi-Element Oil
Standard ..................................................................................................................... 104
6.3 RESULTS AND DISCUSSION ............................................................................ 104
6.3.1 Effect of Sample Solution pH on Retention of Metal Ions............................... 105
6.3.2 Effect of Desorption Solution Concentration ................................................... 106
6.3.3 Effect of Sample Flow Rate.............................................................................. 107
6.3.4 Effect of Sample Volume ................................................................................. 108
6.3.5 Preconcentration of Multi-Element Using Different Sorbent Materials .......... 109
6.3.6 Effect of Matrix Ions Interferences .................................................................. 112
6.3.7 Analytical Parameters ....................................................................................... 114
6.3.8 The Effect of Column Regeneration ................................................................. 115
6.3.9 Accuracy and Validation of the Proposed Separation and Pre-Concentration
Procedure ................................................................................................................... 116
6.3.10 Application of the Proposed Separation and Pre-Concentration Procedure ... 118
6.4 CONCLUSIONS .................................................................................................... 120
6.5 REFERENCES ...................................................................................................... 121
CHAPTER SEVEN: ........................................................................................................ 125
PRECONCENTRATION OF MOLYBDENUM, ANTIMONY AND VANADIUM IN
GASOLINE
SAMPLES
USING
DOWEX
1-X8
RESIN
AND
THEIR
DETERMINATION WITH ICP OES ........................................................................... 125
ABSTRACT ................................................................................................................. 125
7.1 INTRODUCTION ................................................................................................. 125
7.2. EXPERIMENTAL ............................................................................................... 127
7.2.1 Instrumentation ................................................................................................. 127
7.2.2 Reagents, Solutions and Samples ..................................................................... 128
7.2.3 Preparation of Column ..................................................................................... 129
xii
Table of contents
7.2.4 Preconcentration and Recovery of Mo, Sb and V in Model Organic Solutions
and Real Gasoline Samples ....................................................................................... 130
7.2.5 Procedure for the Dilution of Certified Reference Material ............................. 130
7.2.6 Procedure for Acid Digestion of Gasoline Samples ......................................... 131
7.3. RESULTS AND DISCUSSION ........................................................................... 131
7.3.1 Selection of Stationary Phase ........................................................................... 131
7.3.2 Effect of Sample Solution pH on Retention of Metal Ions............................... 132
7.3.3 Effect of Eluent Concentration ......................................................................... 133
7.3.4 Effect of Sample Volume ................................................................................. 133
7.3.5 Analytical Performances................................................................................... 134
7.3.6 Effect of Matrix Ions Interferences .................................................................. 137
7.3.7 Regeneration Studies ........................................................................................ 137
7.3.8 Accuracy and Validation of the Separation and Preconcentration Procedure .. 138
7.3.9 Application of the Dowex 1-x8 Separation and Preconcentration Procedure in
Commercial Gasoline Samples.................................................................................. 139
7.4. CONCLUSION ..................................................................................................... 142
7.5 REFERENCES ...................................................................................................... 142
CHAPTER EIGHT: ........................................................................................................ 145
MULTIVARIATE
OPTIMIZATION
OF
DUAL-BED
SOLID
PHASE
EXTRACTION FOR PRECONCENTRATION OF Ag, Al, As AND Cr IN
GASOLINE PRIOR TO INDUCTIVELY COUPLED PLASMA OPTICAL
EMISSION SPECTROMETRIC DETERMINATION ............................................... 145
ABSTRACT ................................................................................................................. 145
8.1 INTRODUCTION ................................................................................................. 145
8.2 EXPERIMENTAL ................................................................................................ 148
8.2.1 Instrumentation ................................................................................................. 148
8.2.2 Reagents, Solutions and Samples ..................................................................... 148
8.2.3 Preparation of a Two Bed Column ................................................................... 149
8.2.4 Preconcentration and Recovery of Ag, Al, As and Cr in Synthetic Gasoline
Solution...................................................................................................................... 149
8.2.5 Optimization Approach .................................................................................... 150
8.2.6 Comparative Method ........................................................................................ 151
8.3 RESULTS AND DISCUSSION ............................................................................ 151
xiii
Table of contents
8.3.1 Factorial Design ................................................................................................ 151
8.3.2 Effect of Sample Volume ................................................................................. 155
8.3.3 Column Regeneration ....................................................................................... 156
8.3.4 Analytical Performances of the Dual-Bed SPE Method .................................. 156
8.3.5 Validation of the Dual-Bed SPE Method ......................................................... 157
8.3.6 Analysis of Real Samples ................................................................................. 159
8.4 CONCLUSION ...................................................................................................... 162
8.5 REFERENCES ...................................................................................................... 162
CHAPTER NINE: ........................................................................................................... 166
A SOLID PHASE EXTRACTION PROCEDURE BASED ON ELECTROSPUN
CELLULOSE-g-OXOLANE-2,5-DIONE
NANOFIBERS
FOR
TRACE
DETERMINATION OF Cd, Cu, Fe, Pb AND ZN IN GASOLINE SAMPLES BY ICP
OES ................................................................................................................................... 166
ABSTRACT ................................................................................................................. 166
9.1 INTRODUCTION ................................................................................................. 166
9.2 EXPERIMENTAL ................................................................................................ 168
9.2.1 Material and methods ....................................................................................... 168
9.2.2 Electrospinning and functionalization of cellulose nanofibers with oxolane-2,5dione .......................................................................................................................... 170
9.2.3 Column preparation .......................................................................................... 170
9.2.4 Preconcentration procedure .............................................................................. 170
9.2.5 Procedure for the dilution of Certified Reference Material .............................. 171
9.2.6 Procedure for acid digestion of gasoline samples ............................................ 171
9.3 RESULTS AND DISCUSSION ............................................................................ 172
9.3.1 Characterization of the adsorbent ..................................................................... 172
9.3.2 Effect of sample pH .......................................................................................... 174
9.3.3 Effect of eluent concentration........................................................................... 175
9.3.4 Effect of flow rate of sample solutions............................................................. 176
9.3.5 Effect of the sample volume ............................................................................. 177
9.3.6 Adsorption capacities ....................................................................................... 177
9.3.7 Column regeneration ........................................................................................ 178
9.3.8 Analytical Parameters ....................................................................................... 179
9.3.9 Accuracy and validation of the developed method .......................................... 181
xiv
Table of contents
9.3.10 Application of the cellulose-g-oxolane-2,5-dione SPE method ..................... 183
9.4 CONCLUSIONS .................................................................................................... 185
9.5 REFERENCES ...................................................................................................... 186
CHAPTER TEN: ............................................................................................................. 189
FULL FACTORIAL DESIGN FOR THE OPTIMIZATION OF SIMULTANEOUS
PRECONCENTRATION OF TRACE METAL IONS IN GASOLINE SAMPLES
PRIOR TO THEIR INDUCTIVELY COUPLED MASS SPECTROMETRIC
DETERMINATION ........................................................................................................ 189
ABSTRACT ................................................................................................................. 189
10.1 INTRODUCTION ............................................................................................... 189
10.2 MATERIALS AND METHODS ........................................................................ 191
10.2.1 Apparatus ........................................................................................................ 191
10.2.2 Reagents and Solutions................................................................................... 192
10.2.3 Preparation of Nanometer-Sized Alumina Using Sol-Gel Method ................ 193
10.2.4 Preparation of the Column.............................................................................. 193
10.2.5 Preconcentration and Recovery of Co, Cr, Mn, Ni, and Ti in a Synthetic
Gasoline Solution ...................................................................................................... 193
10.2.6 Optimization Approach .................................................................................. 194
10.2.7 Procedure for Microwave Acid Digestion of Gasoline Samples .................... 194
10.3 RESULTS AND DISCUSSION .......................................................................... 195
10.3.1 Characterization of the Nanometer-Sized Alumina........................................ 195
10.3.2 Factorial Design .............................................................................................. 196
10.3.3 Effect of Sample Volume ............................................................................... 200
10.3.4 Adsorption Capacities of Metal Ions .............................................................. 201
10.3.5 Regeneration of the Adsorbent ....................................................................... 202
10.3.6 Analytical Performance of the Nanometer-Sized Alumina SPE Method....... 202
10.3.7 Validation of the Nanometer-Sized Alumina SPE Method ............................ 203
10.3.8 Application of Nanometer-Sized Alumina SPE Method................................ 207
10.4 CONCLUSIONS .................................................................................................. 210
10.5 REFERENCES .................................................................................................... 210
CHAPTER ELEVEN: ..................................................................................................... 215
DEVELOPMENT AND MULTIVARIATE OPTIMIZATION OF AN OFFLINE
HOLLOW FIBER SOLID PHASE MICROEXTRACTION SYSTEM FOR
xv
Table of contents
PRECONCENTRATION OF TRACE METAL IONS IN FUEL SAMPLES PRIOR
TO THEIR ICP-MS DETERMINATION .................................................................... 215
ABSTRACT ................................................................................................................. 215
11.1 INTRODUCTION ............................................................................................... 215
11.2 EXPERIMENTAL .............................................................................................. 218
11.2.1 Instrumentation ............................................................................................... 218
11.2.2 Reagents, Solutions and Real Samples ........................................................... 219
11.2.3 Preparation of Sol-gel ..................................................................................... 219
11.2.4 Extraction and Pre-concentration Procedure .................................................. 220
11.2.5 Optimization Strategy ..................................................................................... 221
11.2.6 Comparative Procedure .................................................................................. 222
11.3 RESULTS AND DISCUSSION .......................................................................... 222
11.3.1 Preliminary Optimization Using Two Level Full Factorial Design ............... 222
11.3.2 Final optimization using a Central Composite Design ................................... 225
11.3.2.1 Analysis of Variance ............................................................................... 226
11.3.2.2 Optimization of Experimental Conditions............................................... 228
11.3.3 Regeneration Studies ...................................................................................... 230
11.3.4 Analytical Features ......................................................................................... 230
11.3.5 Effect of other Metal Ions on the HF-SPME Procedure ................................. 231
11.3.6 Validation and Application of the HF-SPME Method ................................... 232
11.4 CONCLUSION .................................................................................................... 236
11.5 REFERENCES .................................................................................................... 236
CHAPTER TWELVE: .................................................................................................... 240
PREPARATION OF TITANIA-ALUMINA HOLLOW FIBER MEMBRANE AND
MULTIVARIATE
OPTIMIZATION
FOR
SIMULTANEOUS
PRECONCENTRATION OF TRACE ELEMENTS IN DIESEL AND GASOLINE
SAMPLES PRIOR TO ICP-MS DETERMINATION ................................................ 240
ABSTRACT ................................................................................................................. 240
12.1 INTRODUCTION ............................................................................................... 240
12.2 EXPERIMENTAL .............................................................................................. 242
12.2.1 Instrumentation ............................................................................................... 242
12.2.2 Reagents and Solutions................................................................................... 243
12.2.3 Synthesis Titania-Alumina Sol ....................................................................... 244
xvi
Table of contents
12.2.4 Preparation of Titania-Alumina Hollow Fiber ............................................... 244
12.2.5 Preconcentration Method................................................................................ 245
12.2.6 Optimization Strategy ..................................................................................... 245
12.2.7 Comparative Method ...................................................................................... 246
12.3 RESULTS AND DISCUSSION .......................................................................... 246
12.3.1 Characterization of Titania-Alumina Hollow Fiber ....................................... 246
12.3.1.1 X-ray diffraction analysis ........................................................................ 247
12.3.1.2 Pore structure parameters ........................................................................ 248
12.3.1.3 Scanning electron microscopy (SEM) analysis ....................................... 248
12.3.2 Screening Analysis of Membrane Solid Phase Microextraction (MSPME)
Preconcentration System ........................................................................................... 249
12.3.3 Optimization of MSPME Preconcentration System ....................................... 253
12.3.4 Effect of Sample Volume ............................................................................... 255
12.3.5 Adsorption Capacities and Regeneration of the Hollow Fiber ....................... 256
12.3.6 Analytical Figure of Merit .............................................................................. 257
12.3.7 Validation, Application of MSPME Preconcentration System to Real Samples
and Comparison with a Standard Method ................................................................. 258
12.4 CONCLUSIONS .................................................................................................. 262
12.5 REFERENCES .................................................................................................... 262
CHAPTER THIRTEEN: ................................................................................................ 265
CHEMOMETRIC OPTIMIZATION OF HOLLOW FIBER-LIQUID PHASE
MICROEXTRACTION FOR PRECONCENTRATION OF TRACE ELEMENTS IN
DIESEL AND GASOLINE PRIOR TO THEIR ICP OES DETERMINATION ..... 265
ABSTRACT ................................................................................................................. 265
13.1 INTRODUCTION ............................................................................................... 265
13.2 EXPERIMENTAL .............................................................................................. 268
13.2.1 Reagent and Standard Solutions ..................................................................... 268
13.2.2 Instrumentation ............................................................................................... 268
13.2.3 Preparation of the HF-LPME ......................................................................... 269
13.2.4 Extraction Procedure ...................................................................................... 269
13.2.5 Optimization Strategy ..................................................................................... 270
13.3. RESULTS AND DISCUSSION ......................................................................... 271
13.3.1 Chemometric Optimization of HF-LPME ...................................................... 271
xvii
Table of contents
13.3.1.1 Factorial Design ....................................................................................... 271
13.3.1.2 Box–Behnken Design .............................................................................. 275
13.3.2 Interference Studies ........................................................................................ 279
13.3.3 Analytical Figure of Merit .............................................................................. 279
13.3.4 Validation and Application of the Proposed Method ..................................... 280
13.4 CONCLUSION .................................................................................................... 282
13.5 REFERENCES .................................................................................................... 282
CHAPTER FOURTEEN: ............................................................................................... 286
GENERAL CONCLUSIONS AND RECOMMENDATIONS ................................... 286
14.1 GENERAL CONCLUSION ............................................................................... 286
14.2 RECOMMENDATIONS .................................................................................... 289
xviii
LIST OF FIGURES
CHAPTER 1
Fig. 1.1. Simplified refinery process (Adopted from Marathon Petroleum Company)5 ....... 2
Fig. 1.2. Simplified block flow diagram of a CTL process9.................................................. 3
Fig. 1.3. End-products in which petrochemical products are used in our daily lives ............ 4
CHAPTER 2
Fig. 2.1. A schematic diagram of an ICP assembly showing the three concentric quarts
tubes composing the torch and the RF coil (adopted from Ref.83) ................................. 25
Fig. 2.2. Schematic diagram of ICP OES instrument showing its major components and
layout (adapted from Ref.85) ........................................................................................... 27
Fig. 2.3. Schematic diagram of inductively coupled plasma mass spectrometer. Diagram
reproduced from Linge and Jarvis96. .............................................................................. 30
CHAPTER 3
Fig. 3.1. Experimental design flow chart showing the summary of sample preparation
methods used for separation and preconcentration of metal ions in organic matrices and
detection techniques. ....................................................................................................... 50
Fig. 3.2. Reaction scheme for the functionalization of cellulose to cellulose-g-oxolane-2,5dione2 .............................................................................................................................. 53
CHAPTER 4
Fig. 4.1. The effect of pH on the recoveries of 20 µg L−1 metal ion solution: A) Dowex
50W-8, B) Chelex-100. sample volume 20 mL; amount of resin 1.5 g, flow rates of
sample and eluent 3.0 mL min−1, respectively n = 3 ...................................................... 63
Fig. 4.2. Influences of the eluent concentration on the recoveries of the analytes on Dowex
50w-x8 resin column. sample volume 20 mL; amount of resin 1.5 g, flow rates of
sample and eluent 3.0 mL min−1, respectively n = 3 ...................................................... 64
Fig. 4.3. Effect of sample volume on the recoveries of metal ions: pH 6.0; analyte
concentration 10 µg L-1; amount of sorbent 1.5 g; flow rates of sample and eluent 3.0
mL min−1; eluent volume 5 mL; replicates n=3 .............................................................. 66
CHAPTER 5
Fig. 5.1. Effect of contact time on retention of Ag, Co and Mn using Dowex 50W-x8 resin:
Initial concentration of metal ions 10 mg L-1; amount of resin 0.05 g; sample volume 20
mL; temperature 298 K; stirring rate 200 rpm; stirring time 0-60 min; initial pH 6 ...... 79
Fig. 5.2. Effect of pH on the adsorption of Ag, Co and Mn. Initial concentration of metal
ions 10 mg L-1; amount of resin 0.05 g; sample volume 20 mL; temperature 298 K;
stirring rate 200 rpm; stirring time 20 min; initial pH 4-10 ............................................ 80
xix
List of figures
Fig. 5.3. Effect of resin dosage on removal of Ag(I), Co(II) and Mn(II) by Dowex 50W-x8
cation exchange resin: Initial concentration of meat ions 10 mg L-1; amount of resin
0.02-1.0 g; sample volume 20 mL; temperature 298 K; stirring rate 200 rpm; stirring
time 20 min; initial pH 6 ................................................................................................. 81
Fig. 5.4. Sorption isotherm of (a) Ag, (b) Co and (c) Mn on Dowex 50W-x8 resin: Initial
concentration of meat ions 10 to 300 mg L-1; amount of resin 0.05 g; sample volume 20
mL; temperature, 293 to 313 K; shaking rate 200 rpm; shaking time 20 min; initial pH 6
........................................................................................................................................ 83
Fig. 5.5. Intraparticle diffusion plots for adsorption of silver, copper and manganese ....... 90
CHAPTER 6
Fig. 6.1. Effect of sample pH on retention of the analytes in ethanol onto Dowex 50W-x8
resin column: pH 6; analyte concentration 12 µg L-1; amount of resin 1.5 g; flow rates
of sample and eluent 3.0 mL min−1; eluent volume 5 mL; replicates n = 3 ................. 106
Fig. 6.2. Influences of the eluent concentration on the recoveries of the analytes on Dowex
50W-x8 resin column: pH 6; analyte concentration 12 µg L-1; amount of resin 1.5 g;
flow rates of sample and eluent 3.0 mL min−1; eluent volume 5 mL; replicates n = 3 107
Fig. 6.3. Effect of sample volume on the recoveries of metal ions: pH 6; analyte
concentration 12 µg L-1; amount of resin 1.5 g; flow rates of sample and eluent 3.0 mL
min−1; eluent volume 5 mL; replicates n = 3 ................................................................ 108
Fig. 6.4. Pre-concentration of metal ions from methanol, ethanol, iso-propanol and 2butanol Experimental conditions: pH 6; analyte concentration 12 µg L-1; amount of
resin 1.5 g; flow rates of sample and eluent 3.0 mL min−1; eluent volume 5 mL;
replicates n = 3 .............................................................................................................. 111
CHAPTER 7
Fig. 7.1. Effect of sample pH on retention of the analytes in ethanol onto Dowex 1-x8 resin
column. Sample volume: 20 mL; amount of resin 1.5 g; flow rates of sample and eluent
3.0 mL min−1; replicates n = 3) ..................................................................................... 132
Fig. 7.2. Effect of sample volume on the recoveries of metal ions: pH 6; analyte
concentration 14 µg L-1; amount of resin 1.5 g; flow rates of sample and eluent 3.0 mL
min−1; eluent volume 5 mL; replicates n = 3 ................................................................ 134
CHAPTER 8
Fig. 8.1. Pareto chart of standardized effects for variables in the separation and
preconcentration of silver (A); aluminium (B), arsenic (C) and Cr (D). ...................... 154
Fig. 8.2. Effect of sample volume on the recoveries of metal ions ................................... 155
CHAPTER 9
Fig. 9.2. Characteristic absorption peaks of (A) cellulose acetate, (B) deacetylated cellulose
and (C) cellulose-g-oxolane-2,5-dione ......................................................................... 172
xx
List of figures
Fig. 9.3. SEM micrographs: (I) cellulose nanofibers and (II) cellulose-g-oxolane-2,5-dione
nanofibers ..................................................................................................................... 174
Fig. 9.4. Effect of sample pH on retention of the analytes in ethanol onto cellulose-goxolane-2,5-dione column: Experimental conditions: analyte concentration 20 µg L-1;
amount of adsorbent 0.5 g; flow rates of sample and eluent 2.0 mL min−1; eluent
concentration 2.0 mol L-1, eluent volume 5 mL; replicates n = 3 ................................. 175
Fig. 9.5. Effect of sample volume on the recoveries of metal ions. Experimental conditions:
pH 6; analyte concentration 20 µg L-1; amount of cellulose-g-oxolane-2,5-dione
nanofibers 0.5 g; flow rates of sample and eluent 2.0 mL min−1; eluent concentration 2.0
mol L-1; eluent volume 5 mL; replicates n = 3 ............................................................. 177
CHAPTER 10
Fig. 10.1. Scanning electron microscopy images of alumina obtained by sol-gel methods
starting from AlCl3 as precursor, calcined at 1000°C for three hours. .......................... 195
Fig. 10.2. X-ray diffraction pattern of alumina obtained by sol-gel methods starting from
AlCl3 as precursor, calcined at 1000 °C for three hour ................................................ 196
Fig. 10.3. Pareto chart of standardized effects for variables in the separation and
preconcentration of Co, Cr, Mn, Ni and Ti. A = pH; B = eluent concentration (mol L-1)
and C = sample flow rate (mL min-1) ........................................................................... 199
Fig.10.4. Effect of sample volume on the recoveries of metal ions: pH 7.0; analyte
concentration 30 µg L-1; amount of sorbent 1.5 g; flow rates of sample and eluent 2.0
mL min−1; eluent volume 5 mL; replicates n = 3 .......................................................... 201
CHAPTER 11
Fig. 11.1. Pareto chart of standardized effects for variables related to the preconcentration
of (A) cadmium, (B) copper, (C) iron, (D) lead and (E) zinc ....................................... 225
Fig. 11.2. Response surface for percentage recovery of cadmium (A), copper (B), iron (C),
lead (D) and zinc (E) as function of acceptor phase amount (APA), mg mL-1 and
extraction time (ET), min at constant eluent concentration of 2.75 mol L-1 ................ 229
CHAPTER 12
Fig. 12.1. XRD spectra of nanometer-sized alumina powder (A), nanometer-sized titania
powder (B) and titania-alumina hollow fiber (C) calcined at 1000°C for 3 hours. (Theta
phases: α = alpha-phase Al2O3, γ = gamma-phase Al2O3, R = rutile TiO2) ................. 247
Fig. 12.2. SEM textural images of the titania-alumina (A) and polypropylene hollow fiber
(B). ................................................................................................................................ 248
Fig. 12.3. Pareto charts of standardized effects for variables in the cobalt and chromium
preconcentration. .......................................................................................................... 250
Fig. 12.4. Pareto charts of standardized effects for variables in the molybdenum and nickel
preconcentration. .......................................................................................................... 251
xxi
List of figures
Fig. 12.5.Pareto charts of standardized effects for variables in the preconcentration of
antimony and vanadium ................................................................................................ 252
Fig. 12.6. Response surface for percentage recovery of Cr (A), Co (B), (C), Mo (D), Ni (E)
and V (F) as function of extraction time (ET), min. ..................................................... 255
Fig. 12.7. Effect of sample volume on the recoveries of metal ions ................................. 256
CHAPTER 13
Fig. 13.1. Pareto charts of standardized effects for variables in the Ag and Al
preconcentration. .......................................................................................................... 273
Fig. 13.2. Pareto charts of standardized effects for variables in the As and Mn
preconcentration. .......................................................................................................... 273
Fig. 13.3. Pareto charts of standardized effects for variables in the Ti preconcentration. 274
Fig. 13.4. Response surfaces obtained for (A) Ag, (B) Al, (C) As, (D) Mn and (E) Ti after
extraction and preconcentration by HF-LPME ............................................................. 278
xxii
LIST OF TABLES
CHAPTER 2
Table 2.1. Selected applications of solid phase extraction for preconcentration of metal
ions in fuel samples ........................................................................................................ 20
Table 2.2. Application of ICP OES and ICP-MS for determination of metal ions in
petroleum products ......................................................................................................... 35
Table 2.3. Selected applications of chemometric tools for multivariate optimization of
analytical methodologies ................................................................................................ 37
CHAPTER 3
Table 3.1. Electrothermal AAS temperature programs for determination of metal ions .... 56
CHAPTER 4
Table 4.1. Recovery (%) of multi-element in aqueous solution using Dowex 50W-x8 and
Chelex-100 SPE methods ............................................................................................... 65
Table 4.2. Analysis of certified reference materials (mean of 3 replicates; concentration in
µg L-1) ............................................................................................................................. 68
Table 4.3. Concentration (µg L-1) of Cd, Co, Cr, Cu, Fe, Ni, Pb, and Zn in water samples
(replicates n=5, volume 100 mL, final volume 5 mL) .................................................... 70
Table 4.4. Guidelines for the presence of heavy metals in drinking water; concentration in
µg L-1 .............................................................................................................................. 70
CHAPTER 5
Table 5.1. Graphite furnace temperature program for the determination of silver, cobalt and
manganese in ethanol model solutions ........................................................................... 77
Table 5.2. Langmuir and Freundlich parameters for ion exchange adsorption of Ag, Co and
Mn onto Dowex 50W-x8 resin in ethanol ...................................................................... 84
Table 5.3. Temkin and Dubinin-Radushkevich parameters for ion exchange adsorption of
Ag, Co and Mn onto Dowex 50W-x8 resin in ethanol ................................................... 86
Table 5.4. Comparison of maximum adsorption capacities of Dowex 50W-x8 for Ag, Co
and Mn with other adsorbents reported in literature ....................................................... 87
Table 5.5. Kinetic parameters for the adsorption of Ag, Co and Mn onto Dowex 50W-x8 in
ethanol............................................................................................................................. 88
Table 5.6. Values of initial sorption rate (h) and half-adsorption time (t1/2) ....................... 89
Table 5.7. Thermodynamic parameters for the adsorption of Ag, Co and Mn on Dowex
50W-x8 ........................................................................................................................... 91
Table 5.8. Concentration of Ag, Co and Mn (µg L-1) in commercial ethanol samples ....... 93
xxiii
List of tables
CHAPTER 6
Table 6.1. Recovery (%) of multi-element in ethanol using Dowex 50W-x8 (Dow(a)),
Chelex-100 (Che1) and Dowex MAC-3 (Dow(b)) for SPE methods. .......................... 110
Table 6.2. Analysis of the metallo-organic Conostan standard for the determination of
analytes after application of the pre-concentration procedure; RSD= relative standard
deviation. ...................................................................................................................... 112
Table 6.3. Effect of potential interfering ions on the percentage recoveries of Cd, Cr, Cu,
Fe, Mn; Pb, Ti and Zn ................................................................................................... 113
Table 6.4. Analytical performances for the proposed Dowex 50W-x8 SPE method ........ 114
Table 6.5. Accuracy test results for spiked recovery (R): pH 6, sample volume: 100 mL, n
=3. ................................................................................................................................. 117
Table 6.6. Analysis of the certified reference material (CRM TMDW-500 drinking water)
for the determination of analytes after application of the pre-concentration procedure
...................................................................................................................................... 118
Table 6.7. Determination of metal ions (µg L-1) in commercial methanol, ethanol, isopropanol and 2-butanol samples after pre-concentration by the proposed method (pH 6,
sample volume: 100 mL, n = 3) and the comparative one (ETAAS) ........................... 119
CHAPTER 7
Table 7.1. Operation parameters and heating temperature program for ETAAS .............. 128
Table 7.2. Physical and chemical properties of the resins ................................................. 129
Table 7.3. Analytical performances for the proposed Dowex 1-x8 SPE method (sample
volume 100 mL) ........................................................................................................... 135
Table 7.4. Comparison of some methods used for determination of Mo, Sb and V ......... 136
Table 7.5. Effect of potential interfering ions on the recovery of metal ions .................... 137
Table 7.6. Percentage (%) recovery results when 1 mL gasoline sample 1-MFUG was
spiked with different metal concentrations (0-20 µg L-1) and made up in ethanol (100
mL) ............................................................................................................................... 138
Table 7.7. Concentrations (in µg L-1) of metal ions in gasoline samples determined by ICP
OES in sample solutions resulting from Dowex 1-x8 preconcentration procedure ..... 139
Table 7.8. Concentrations (in µg L-1) of metal ions in gasoline samples determined by ICP
OES in sample solutions resulting from Dowex 1-x8 preconcentration procedure and
GFAAS in sample solutions resulting from acid digestion procedure ......................... 141
CHAPTER 8
Table 8.1. The operating parameters of determination of elements by ICP OES ............. 148
Table 8.2. Factors and levels used in 23 factorial design for separation and preconcentration
of metal ions ................................................................................................................. 150
xxiv
List of tables
Table 8.3. Design matrix and the results of Ag, Al, As and Cr ......................................... 152
Table 8.4. Determination of Ag, Al, As and Cr (µg L-1) in gasoline sample spiked with
inorganic and organic standard solutions (mean ± standard deviation, n= 3) .............. 158
Table 8.5. The determination of Ag, Al, As and Cr in different gasoline samples using
dual-bed SPE/ICP OES and MAD/ ICP OES methods ................................................ 161
CHAPTER 9
Table 9.1. The operating parameters of determination of elements by ICP OES ............. 169
Table 9.2. 13C NMR chemical shifts for cellulose acetate, deacetylated cellulose and
cellulose-g-oxolane-2,5-dione and the corresponding assignments ............................. 173
Table 9.3. Influences of the eluent concentration on the recoveries of the analytes on
cellulose-g-oxolane-2,5-dione column. ........................................................................ 176
Table 9.4. Effect of flow rate of sample solutions: analytical results in terms of recovery
...................................................................................................................................... 176
Table 9.5. Column regeneration. ....................................................................................... 179
Table 9.6. Analytical performances for cellulose-g-oxolane-2,5-dione SPE method. ...... 180
Table 9.7. Analytical performances for acid digestion method ......................................... 181
Table 9.8. Accuracy test results for spike recovery test. ................................................... 182
Table 9.9. Results for the oil based certified reference material. ...................................... 183
Table 9.10. Concentrations (in µg L-1) of metal ions in commercial gasoline samples after
pre-concentration by the cellulose-g-oxolane-2,5-dione SPE method. ........................ 184
Table 9.11. Concentrations (in µg L-1) of metal ions in gasoline samples determined by
ICP OES in sample solutions resulting from acid digestion procedure ........................ 185
CHAPTER 10
Table 10.1. Operational ICP-MS parameters .................................................................... 192
Table 10.2. Factors and levels used in 23 factorial design for the separation and
preconcentration of metal ions...................................................................................... 194
Table 10.3. Design matrix and the results of metal ions ................................................... 197
Table 10.4. Column regeneration ...................................................................................... 202
Table 10.5. Analytical performances for the proposed nanometer sized Al2O3 SPE method
and microwave-assisted digestion method ................................................................... 205
Table 10. 6. Comparison of the proposed nanometer-sized SPE method with other methods
used for determination of trace metals in gasoline ....................................................... 205
Table 10.7. Determination of Co, Cr, Mn, Ni and Ti (µg L-1) in gasoline sample spiked
with inorganic and organic standard solutions (mean ± standard deviation, sample
volume = 100 mL, n= 3) ............................................................................................... 206
xxv
List of tables
Table 10.8. Concentrations (in µg L-1) of metal ions in gasoline samples determined by
ICP-MS in aqueous solutions resulting from nanometer-sized Al2O3 preconcentration
procedure (sample volume = 100 mL) and ICP OES in aqueous solutions resulting from
microwave-assisted digestion procedure ...................................................................... 209
CHAPTER 11
Table 11.1. Operational ICP-MS parameters .................................................................... 218
Table 11.2. Factors and levels used in 24 factorial design for separation and
preconcentration of metal ions in fuel samples ............................................................ 221
Table 11.3. Matrix of 24 full factorial design and the analytical response (% recovery) for
each experiment for extraction and preconcentration of metal ions ............................. 223
Table 11.4. List of experiments in the central composite design (actual values) for HFSPME optimization and the responses ......................................................................... 226
Table 11.5. Analytical performance of the HF-SPME system for preconcentration of metal
ions obtained under optimum conditions ...................................................................... 231
Table 11.6. Effect of potential interfering ions on the percentage recoveries of Cd, Cu, Fe,
Pb and Zn (mean % recovery ± standard deviation) ..................................................... 232
Table 11.7. Analytical results obtained in the analysis of spiked diesel sample. The
concentration and recovery values are expressed as the mean ± standard deviation of the
three replicates .............................................................................................................. 234
Table 11.8. Determination of Cd, Cu, Fe, Pb and Zn (µg L-1) in commercial diesel (D1 and
D2) and gasoline (G1 and G2) samples by proposed HF-SPME and comparative method
(n = 3, at 95% confidence level). .................................................................................. 235
CHAPTER 12
Table 12.1. Operational ICP-MS parameters .................................................................... 243
Table 12.2. Factors and levels used in 24 factorial design for separation and
preconcentration of metal ions in fuel samples ............................................................ 246
Table 12.3. List of experiments in the factorial design (actual values) for MSPME
optimization and the responses ..................................................................................... 249
Table 12.4. List of experiments in the central composite design (actual values) for MSPME
optimization and the responses ..................................................................................... 254
Table 12.5. Analytical figure of merit of the MSPME system for preconcentration of metal
ions obtained under optimum conditions ...................................................................... 257
Table 12.6. Analytical results obtained in the analysis of spiked diesel sample. The
concentration and recovery values are expressed as the mean ± standard deviation of the
three replicates .............................................................................................................. 258
xxvi
List of tables
Table 12.7. Determination of Co, Cr, Mo, Ni, Sb and V (µg L-1) in commercial diesel (D1
and D2) and gasoline (G1 and G2) samples by proposed MSPME and comparative
method MAD/ICP OES (n = 3, at 95% confidence level). ........................................... 261
CHAPTER 13
Table 13.1. Factors and levels used in 24 factorial design for extraction and
preconcentration of metal ions in fuel samples ............................................................ 271
Table 13.2. List of experiments in the factorial design (actual values) for HF-LPME
optimization and the response values ........................................................................... 272
Table 13.3. Experimental design using Box–Behnken design (CCD) and analytical
response values ............................................................................................................. 275
Table 13.4. Calculated critical point values ...................................................................... 278
Table 13.5. Analytical figure of merits for the proposed HF-LPME method ................... 280
Table 13.6. Analytical results obtained in the analysis of spiked diesel and gasoline
samples. The concentration and recovery values are expressed as the mean ± standard
deviation of the three replicates .................................................................................... 280
Table 13.7. The determination of Ag, Al, As, Mn and Ti in diesel and gasoline samples
using HF-LPME/ICP OES ............................................................................................ 281
Table 13.8. The determination of Ag, Al, As, Mn and Ti in diesel and gasoline samples
using MAD/ICP OES ................................................................................................... 282
xxvii
LIST OF ABBREVIATIONS
13
C NMR
ANOVA
AST
APDC
BZA
CCD
CCD
CID
CPE
CTL
DC
DDTC
DOE
EDXRF
EIE
ETAAS
ETV
FAAS
FT
FTIR
GTL
HF-LPME
HF-SPME
ICP
ICP-MS
ICP OES
LLE
LOD
LOQ
LPG
m/z
MAD
MIC
MSPME
PAN
PDMS
PMBP
PMT
PTFE
RF
RSA
Carbon-13 nuclear resonance spectroscopy
Analysis of Variance
Atomic spectrometric techniques
Ammonium pyrrolidine dithiocarbamate
Benzoylacetone
Charge-coupled device
Central composite design
Charge injection device
Cloud point extraction
Coal to liquid
Direct current
Diethyldithiocarbamate
Design of experiments
Energy dispersive X-ray fluorescence
Easily ionized element
Electrothermal atomic absorption spectrometry
Electrothermal vaporization
Flame atomic absorption spectrometry
Fischer–Tropsch
Fourier transform infra-red spectroscopy
Gas to liquid LPG
Hollow fiber liquid phase microextraction
Hollow fiber solid phase microextraction
Inductively coupled plasma
Inductively coupled plasma-mass spectrometry
Inductively coupled plasma-optical emission spectrometry
Liquid-liquid extraction
Limits of detection
Limits of quantification
Liquefied petroleum gas
Mass-to-charge ratio
Microwave-assisted digestion
Microwave-induced combustion
Solid phase microextraction membrane
1-(2-pyridylazo)-2-naphthol
Polydimethylsiloxane
1-phenyl-3-methyl-4-benzoyl-pyrazolone
Photomultiplier tube
Polytetrafluoroethylene
Radiofrequency
Republic of South Africa
xxviii
List of abbreviations
RSD
RTIL
SBSE
SEM
SPE
SPME
UV
WHO
XRD
XRF
Relative standard deviation
Room temperature ionic liquids
Stir-bar sorptive extraction
Scanning electron microscopy
Solid phase extraction
Solid phase microextraction membrane
Ultra Violet
World health organization
X-ray powder diffraction
X-ray fluorescence
xxix
CHAPTER ONE:
INTRODUCTION
1.1 BACKGROUND
In our modern and industrialized world, every country and society is totally dependent on
fossil fuels. For instance, crude oil and its derivatives are used for production of liquid
fuels that power vehicles, the preparation of medicines, the manufacture of plastics,
cosmetics, as well as other personal products that enhance human beings daily lives. The
subsections that follow below give background on petroleum and petrochemical products as
well the petrol additives. A statement on the problem that the PhD project sought to solve, the
objectives of the study and the structure of the thesis write up are presented in this section.
1.1.1 Petroleum Products
Petroleum products such as liquid fuels are chemical products derived from fossil
fuels.1 There are three well known types of fossil fuels; these include crude oil, coal and
natural gas. Crude oil is a complex mixture containing mostly hydrocarbons (compounds
with carbon and hydrogen) and traces of inorganic chemical species such metals, sulfur,
nitrogen and oxygen.1-3 Crude oil goes through a process called fractional distillation,
whereby different components of crude oil are separated so that they can be further
refined.4 The simplified refinery process is presented in Fig. 1.1. During factional
distillation, crude oil is put into a high-pressure steam boiler and the oil turns to vapor. The
latter then enters the bottom of the distillation tower (column) through a pipe. The vapors
rise in the column and condense to liquid as they reach their boiling points. These liquid
fractions flow through pipes and are collected in separate storage tanks.4
1
Chapter one:
Introduction
Fig. 1.1. Simplified refinery process (Adopted from Marathon Petroleum Company)5
Liquid fuel can be produced by the Fischer–Tropsch (FT) synthesis of a syngas (a
mixture of hydrogen and carbon monoxide), which can be obtained from natural gas
reforming or coal gasification. The resulting processes are known as gas to liquid (GTL)
and coal to liquid (CTL).6 The gas to liquid process consists of autothermal reforming,
from which methane, steam and oxygen reacts to form syngas. The by-products of this
process, which are mainly water and carbon dioxide are then removed from syngas before
they are introduced to the FT reactor where syngas is converted into hydrocarbons. The
products from the FT reactor are separated in the separation column called a hydrocracking reactor (first separation). The latter (hydro-cracking reactor) converts heavy
hydrocarbons into gasoline and diesel. The products such as liquefied petroleum gas
(LPG), gasoline and diesel as are then further separated in a second separation section.6
Fig. 1.2 shows a simplified block flow diagram of a CTL process. The CTL process can be
achieved either by direct or indirect liquefaction. The major conversion steps between
energy forms when dealing with indirect liquefaction are gasification (that is, converting
solid coal to syngas) and FT synthesis (that is converting the syngas to hydrocarbon
products).7 The block flow diagram for indirect CTL process is similar to the GTL block
flow diagram; however, the main differences are the starting materials.6 The direct
liquefaction involves the use of catalysts to covert coal to liquid products.8
2
Chapter one:
Introduction
Fig. 1.2. Simplified block flow diagram of a CTL process9
The liquid fuels under investigation in this study, are gasoline and diesel. Gasoline is a
volatile and flammable liquid that is widely used in internal combustion engines and is
formed by a mixture of hydrocarbons (4-12 carbons). These hydrocarbons include
paraffins, naphthenes, olefins, and aromatic hydrocarbons.10,11 In addition, small amounts
of additives are added to improve its stability, control deposit formation in engines,
improve performance and modify other characteristics.10
Diesel is a blend of petroleum-derived compounds called middle distillates and is
heavier than gasoline but lighter than lubricating oil and unlike gasoline; diesel may or
may not contain metal additives.10 It consists of hydrocarbons ranging from 10 to 24
carbons with the most notorious contaminant in diesel fuel which is sulfur.
1.1.1.1 Petrochemical products
Petrochemical products are chemicals that are derived from petroleum products such as
naphtha.12,13 In petrochemical industries, the organic chemicals with the largest production
volume are olefins (ethylene, butylene and propylene), aromatics (benzene, toluene and
xylenes) and methanol.13 The latter can be produced by chemical cracking and catalytic
reforming.13 Chemical cracking refers to the catalytic break down of large hydrocarbon
molecules into simpler molecules at high temperatures. Catalytic reforming is used to
convert low-octane naphthas into high-octane gasoline blending components and often to
benzene, toluene and xylene aromatics for petrochemical use.12,13 Olefins, aromatic and
methanol serve as precursors to a wide range of petrochemical products, such as polymers,
3
Chapter one:
Introduction
alcohols, resins, ketones, esters and carboxylic acids, among others. Petrochemicals can be
used in the automotive and aviation industries, in explosives, plastics, soaps and
detergents, cosmetics, dyes, food packaging, solvents in pharmaceutical preparations,
home furnishings, carpet backing, refrigerators and chemical intermediates, among
others.12 Other uses of these chemicals are listed in Fig.1.3.
Fig. 1.3. End-products in which petrochemical products are used in our daily lives
1.2 METAL IONS IN PETROLEUM AND PETROCHEMICAL PRODUCTS
Trace metal ions in petroleum and petrochemical products come from many sources,
for example, some of the metals such as nickel and vanadium are natural constituents of
petroleum and they can therefore be present in all its derivatives.11,14,15 Iron and zinc are
the main construction materials for fuel tanks and as a result they might be transferred to
the fuel during transport and storage. Copper on the other hand may be introduced during
the distillation and refinement processes.11,16,17 In addition, some of the elements can be
introduced into the petroleum derivatives from catalysts used to process crude oil and its
fractions, during the distillation.18 Moreover, these elements can be added as additives to
improve or promote specific characteristics of the products.11 Additives for gasoline,
diesel, and other petroleum products may contain a wide range of elements including
aluminium, calcium, cerium, chromium, cobalt, copper, lanthanum, lead, lithium,
4
Chapter one:
Introduction
magnesium, manganese, molybdenum, nickel, silicon, silver, sodium, thallium, tin,
tungsten, vanadium, zinc and zirconium.11,19 Even though metallic species in petroleum
and petrochemical products are present in trace levels, the overall emission relative to its
burning contributes significantly to environmental pollution, particularly in urban areas.
Therefore, their analysis is essential in the different sectors of economy from refining to
environmental risk assessment.
1.3 PROBLEM STATEMENT
The presence of metals in petroleum and petrochemical products is undesirable, unless
they are used as additives. Therefore, it is important to control and monitor their
concentrations in fuel and petrochemical products as they tend to affect the quality of these
products. For instance, the presence of Cu can catalyze oxidation reactions and
significantly increase the gum content of gasoline leading to fuel decomposition and poor
engine performance.11,20,21 Additionally, the presence of Ni and Pb species in fuel reduces
the efficiency of catalytic reactors used in vehicle exhaust systems, thus increasing the
emission of carbon monoxide and oxides of sulfur and nitrogen.11,22 Therefore, the quality
control of petroleum and petrochemical products is a subject of economic and
environmental importance, in view of the fact that fuels outside of the quality
specifications can bring direct problems to the consumer through the malfunctioning of the
vehicle engine, increase in fuel consumption and maintenance costs.23 Other petrochemical
products (organic solvents in particular) are used in pharmaceutical, food industries and
paint formulations. The presence of trace elements in these solvents has health effects and
most metal impurities are detrimental to catalytic processes used in industry. These metals
include Ag, Al, As, Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, Sb, Ti, V and Zn, among
others.17,24 Therefore, monitoring of these toxic elements is important, since they are
released into the atmosphere by fuel combustion and concerning concern to human
health.11,21
Since metal concentrations in fuels and petrochemical products are generally in trace
levels, sensitive and fast techniques with low detection limits are required for monitoring.
The problem with the commonly used techniques (flame atomic absorption spectrometry
and energy dispersive X-ray fluorescence spectrometry) is that they have poor/ high
detection limits compared to inductively coupled plasma optical emission/mass
spectrometry (ICP OES/-MS) or electrothermal atomic absorption spectrometry (ETAAS)
5
Chapter one:
Introduction
and therefore cannot detect at low levels. In addition, direct determination of metals in
diesel and gasoline, by most analytical techniques is difficult due to their volatility, low
viscosity, corrosivity and immiscibility with water.25 Despite of this challenge, procedures
based on direct determination using ETAAS have widely reported in literature because of
its high sensitivity and tolerance to high organic matrix loads.26-27 However, the routine
analysis using ETAAS is disadvantageous because of its low sample throughput compared
to inductively coupled plasma-based methods.28
Inductively coupled plasma techniques are well-established multi-elemental and
sensitive techniques of analysis.29 Besides the many advantages they offer, these
techniques suffer from interferences problems which can be manifested in different ways.
For example, if the sample is diluted with organic solvents, the sensitivity decreases due to
the effect of cooling of the plasma. Moreover, the direct introduction of organic samples
into the plasma requires special care because direct loading of organic samples to the
plasma may destabilize or extinguish the latter.22,29 Therefore, analytical methods that can
solve the problem of matrix effect in ICP-MS/OES techniques are required. These methods
may include separation and pre-concentration methods that can extract, separate and
concentrate the metals from complex matrices prior to their determination.
1.4 HYPOTHESIS
Separation and preconcentration of trace metals in organic matrices can be achieved by
solid phase extraction using different adsorbents, hollow fiber-solid phase microextraction
and hollow fiber- liquid phase microextraction prior to their spectrometric determination.
The application of the developed methods will allow accurate assessment of the levels of
metal ions in organic solvent, diesel and gasoline samples. In addition, the elimination of
the organic matrix contributed to the minimize matrix effects and ultimately to improve
sensitivity and accuracy.
1.5 OBJECTIVES OF THE STUDY
1.5.1 Main Objective
The main objective of this project was to develop sample preparation methods based on
separation and pre-concentration of trace metals in organic matrices, using solid phase
extraction (SPE), solid phase microextraction (SPME) and hollow fiber-liquid phase
6
Chapter one:
Introduction
microextraction (HF-LPME prior to their ICP OES/-MS determination. The use of the
sample preparation methods was to eliminate organic matrix thus minimizing matrix
effects and ultimately to improve sensitivity and accuracy.
1.5.2 Specific Objectives
The specific objectives were to:
1. Develop solid phase extraction methodologies for separation and preconcentration of
trace elements in organic solvents, diesel and gasoline samples. Studies on SPE method
involved the following:
 Evaluate the SPE exchange and adsorption capabilities of different commercially
available cation exchange resins and synthetic adsorbents for simultaneous
preconcentration of metal ions in organic matrices.
 Use of univariate and multivariate techniques for optimization of analytical
parameters (performance characteristics) such as effect of pH, effect of sample and
eluent flow rate, among others factors that affect offline SPE of metal ions prior to
their determination using either ICP OES or ICP-MS.
 Application of offline pre-concentration SPE system for preconcentration of metal
ions in real organic solvent and fuel samples.
2. Develop
solid
phase
microextraction
methodologies
for
extraction
and
preconcentration of trace elements in diesel and gasoline samples. The SPME method
was carried out as follows:
 Preparation of an HF-SPME system for extraction and preconcentration of metal
ions in fuel samples.
 Preparation of membrane SPME (MSPME) for separation and preconcentration of
metal ions in fuel samples.
 Use of multivariate methodologies for optimization of HF-SPME and MSPME
parameters. These parameters include sample pH, acceptor phase concentration,
eluent concentration, sample volume, and extraction time, among others.
 Application of HF-SPME and MSPME systems for separation and preconcentration
of metal ions in real fuel samples prior to their determination using ICP-MS.
3. Develop hollow fiber-liquid phase microextraction methodologies for extraction and
preconcentration of trace elements in diesel and gasoline samples.
7
Chapter one:
Introduction
 Preparation of an HF-LPME system for extraction and preconcentration of metal
ions in fuel samples.
 Use of multivariate methodologies for optimization of HF-SPME and MSPME
parameters. These parameters include sample pH, extraction time; stirring rate;
chelating agent concentration, among others.
 Application of HF-SPME preconcentration system in real fuel samples prior to their
determination using ICP OES.
1.6 THESIS OVERVIEW
This thesis is structured into five different parts arranged according to the themes as
presented in form of 14 chapters. A brief description of each part and its constituent
chapters is presented in Part 1 to 5
Part 1: General Introduction and Thesis Overview, is composed of three chapters,
which include introduction, literature review and overview of materials and methods. The
general introduction or background to the topic, problem statement as well as the
objectives of the study is described in Chapter 1. Chapter 2 reviews the literature on the
analytical techniques used for preconcentration and determination of metal ions. Chapter 2
also highlights the use of design of experiments (DOE) in the optimization of
preconcentration technique. Chapter 3 describes an overview of research design, materials
and methods.
Part 2: Solid Phase Extraction, presents the results and discussion that were obtained
when it was used a sample cleanup method. In this part, the component exchange
efficiencies of different commercial ion exchange resins for preconcentration of metal ion
in water, alcohol and fuel samples were investigated. In addition, the applicability of
synthetic adsorbents such as nanometer-sized alumina and functionalized cellulose
nanofibers for preconcentration of trace metals in gasoline samples was investigated. These
results and discussion are covered in Chapter 4-10.
Part 3: Hollow Fiber-Solid Phase Microextraction, describes the results and
discussions on hollow fiber-solid phase microextraction as an extraction and
preconcentration method. There two chapters that cover this section, these include (i)
Hollow fiber solid phase microextraction as a preconcentration technique for simultaneous
microextraction of metal ions in fuel samples prior to their ICP-MS determination (Chapter
8
Chapter one:
Introduction
11) and (ii) Preparation of alumina-titania hollow fiber as a preconcentration system and its
optimization by factorial design for the trace determination of heavy metal ions in gasoline
and diesel samples using ICP-MS (Chapter 12).
Part 4: Hollow Fiber- Liquid Phase Microextraction, describes the results and
discussion obtained when hollow fiber- liquid phase microextraction was used as a
preconcentration method. In this part, a novel method based on hollow-fiber liquid-phase
microextraction and ICP OES for the measurement of metal ions in liquid fuel samples is
described (Chapter 13).
Part 5: General Conclusions, comprised of one chapter (Chapter 13) and it highlights
the major findings and the conclusions drawn from the results obtained from Chapters 4-13
as well as suggested further studies.
1.7 REFERENCES
1. World Health Organization (WHO). 2008. Petroleum products in drinking-water.
Background document for preparation of WHO Guidelines for drinking-water quality.
Geneva, World Health Organization (WHO/SDE/WSH/03.04/74). Available at:
http://www.who.int/water_sanitation_health/dwq/chemicals/antimonysum.pdf Accessed
5 September 2011
2. de Souza, R. M., Meliande, A. L. S., Da Silveira, C. L. P. & Aucélio, R. Q. 2006.
Determination of Mo, Zn, Cd, Ti, Ni, V, Fe, Mn, Cr and Co in crude oil using
inductively coupled plasma optical emission spectrometry and sample introduction as
detergentless microemulsions. Microchemical Journal, 82, 137-141.
3. Khuhawar, M.Y., Mirza A. M. & Jahangir, T.M. 2012. Determination of Metal Ions in
Crude Oils, Crude Oil Emulsions- Composition Stability and Characterization, Prof.
Manar El-Sayed Abdul-Raouf (Ed.), ISBN: 978-953-51-0220-5, InTech.
4. The refinery process. Available at http://www.afpm.org/The-Refinery-Process. Accessed
29 April 2013.
5. Sutton, D. 2012. Refining 101: Marathon Petroleum Company. Available at
http://www.api.org/events-andtraining/proceedings/proceedings/~/media/Files/Events/Conference%20Proceedings/Re
fining101forCustoms-March2012v6.ashx. Accessed 18 March 2013
6. Sudiro, M. & Bertucco, A. 2009. Production of synthetic gasoline and diesel fuel by
alternative processes using natural gas and coal: Process simulation and optimization.
Energy, 34, 2206-2214.
7. Hao, X., Dong, G., Yang, Y., Xu, Y. & Li, Y. 2007. Coal to Liquid (CTL):
Commercialization Prospects in China. Chemical Engineering Technology, 30, 1157–
1165.
9
Chapter one:
Introduction
8. Höök, M. & Aleklett, K. 2010. A review on coal-to-liquid fuels and its coal
consumption. International Journal of Energy Research, 34, 848-864.
9. Miller, L., Ackiewicz, M. & Cicero D. C. 2008. Coal to liquid technology: Clean liquid
fuel from coal (2-28-08_CTL_Brochure.pdf). National Energy Technology Laboratory,
Office of Fossil Energy, US Department of Energy. Available at
http://www.fossil.energy.gov/programs/fuels/publications/2-28-08_CTL_Brochure.pdf
Accessed 10 December 2012.
10. Pereira, R. C. C. & Pasa, V. M. D. 2005. Effect of Alcohol and Copper Content on the
Stability of Automotive Gasoline. Energy & Fuels, 19, 426-432.
11. Korn, M. D. G. A., Dos Santos, D. S. S., Welz, B., Vale, M. G. R., Teixeira, A. P.,
Lima, D. D. C. & Ferreira, S. L. C. 2007. Atomic spectrometric methods for the
determination of metals and metalloids in automotive fuels - A review. Talanta, 73, 111.
12.
Dawn,
M.
2009.
Petrochemical
compounds:
An
introduction.
http://toxicbeauty.co.uk/blog/2009/02/09/petrochemical-compounds-an-introduction.
Accessed 19 May 2011.
13. Davis, S. 2011. Petrochemical industry overview. Available at:
http://www.sriconsulting.com/CEH/Public/Reports/350.0000/ accessed 31 Aug. 11.
14. Turunen, M., Peräniemi, S., Ahlgrén, M. & Westerholm, H. 1995. Determination of
trace elements in heavy oil samples by graphite furnace and cold vapour atomic
absorption spectrometry after acid digestion. Analytica Chimica Acta, 311, 85-91.
15. Vale, M. G. R., Damin, I. C. F., Klassen, A., Silva, M. M., Welz, B., Silva, A. F.,
Lepri, F. G., Borges, D. L. G. & Heitmann, U. 2004. Method development for the
determination of nickel in petroleum using line-source and high-resolution continuumsource graphite furnace atomic absorption spectrometry. Microchemical Journal, 77,
131-140.
16. Saint'pierre, T. D., Dias, L. F., Maia, S. M. & Curtius, A. J. 2004. Determination of Cd,
Cu, Fe, Pb and Tl in gasoline as emulsion by electrothermal vaporization inductively
coupled plasma mass spectrometry with analyte addition and isotope dilution calibration
techniques. Spectrochimica Acta Part B: Atomic Spectroscopy, 59, 551-558.
17. Chaves, E. S., F. G. Lepri, J. S. A. Silva, D. P. C. De Quadros, T. D. Saint’pierre And
A. J. Curtius 2008. Determination of Co, Cu, Fe, Mn, Ni and V in diesel and biodiesel
samples by ETV-ICP-MS. Journal of Environmental Monitoring, 10, 1211-1216.
18. Aucelio, R. Q. & Curtius, A. J. 2002. Evaluation of electrothermal atomic absorption
spectrometry for trace determination of Sb, As and Se in gasoline and kerosene using
microemulsion sample introduction and two approaches for chemical modification.
Journal of Analytical Atomic Spectrometry, 17, 242-247.
19. Du, B., Wei, Q., Wang, S. & Yu, W. 1997. Application of microemulsions in
determination of chromium naphthenate in gasoline by flame atomic absorption
spectroscopy. Talanta, 44, 1803-1806.
10
Chapter one:
Introduction
20. Teixeira, L. S. G., Bezerra, M. D. A., Lemos, V. A., Santos, H. C. D., De Jesus, D. S.
& Costa, A. C. S. 2005. Determination of Copper, Iron, Nickel, and Zinc in Ethanol
Fuel by Flame Atomic Absorption Spectrometry Using On-Line Preconcentration
System. Separation Science and Technology, 40, 2555 - 2565.
21. dos Santos, D. S. S., Teixeira, A. P., Korn, M. G. A. & Teixeira, L. S. G. 2006.
Determination of Mo and V in multiphase gasoline emulsions by electrothermal atomic
absorption spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 61, 592595.
22. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
23. Dias, J. C., Kubota, L. T., Nesterenko, P. N., Dicinoski, G. W. & Haddad, P. R. 2010.
A new high-performance chelation ion chromatographic system for the direct
determination of trace transition metals in fuel ethanol. Analytical Methods, 2, 15651570.
24. Tormen, L., E. S. Chaves, T. D. Saint’pierre, V. L. A. Frescura And A. J. Curtius 2008.
Determination of trace elements in fuel ethanol by ICP-MS using direct sample
introduction by a microconcentric nebulizer. Journal of Analytical Atomic
Spectrometry, 23, 1300-1304.
25. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A.
M. & Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in
automotive gasoline by X-ray fluorescence after pre-concentration on cellulose paper.
Talanta, 72, 1073-1076.
26. Reboucas, M. V., Domingos, D., Santos, A. S. O. & Sampaio, L. 2010. Determination
of trace metals in naphtha by graphite furnace atomic absorption spectrometry:
Comparison between direct injection and microemulsion pretreatment procedures. Fuel
Processing Technology, 91, 1702-1709.
27. Reboucas, M. V., Ferreira, S. L. C. & Neto, B. D. B. 2003. Arsenic determination in
naphtha by electrothermal atomic absorption spectrometry after preconcentration using
multiple injections. Journal of Analytical Atomic Spectrometry, 18, 1267-1273.
28. Donati, G. L., Amais, R. S., Schiavo, D. & Nobrega, J. A. 2013. Determination of Cr,
Ni, Pb and V in gasoline and ethanol fuel by microwave plasma optical emission
spectrometry. Journal of Analytical Atomic Spectrometry, 28, 755-759.
29. Bettinelli, M., Spezia, S., Baroni U. & Bizzarri, G. 1995. Determination of trace
elements in fuel oils by inductively coupled plasma mass spectrometry after acid
mineralization of the sample in a microwave oven. Journal of Analytical Atomic
Spectrometry, 10, 555-560.
11
Chapter two:
Literature review
CHAPTER TWO:
LITERATURE REVIEW ON DETERMINATION AND SAMPLE PREPARATION
METHODS FOR TRACE METALS ANALYSIS IN PETROLEUM BASED
PRODUCTS
This chapter presents a literature review on techniques for sample preparation and
quantification of trace metals in fuel samples. An overview of inductively coupled plasma
techniques (that is, inductively coupled plasma optical emission/ mass spectrometry) is given,
along with their principles, analytical capabilities, sample introduction methods, ion
separation, detection and applications. In addition, the use of multivariate (chemometric)
techniques in the optimization of separation and preconcentration procedures is also covered.
2.1 SAMPLE PRETREATMENT METHODS
The most important stage in fuel analysis is the sample pretreatment or the preparation
step, because it prepares the sample and make it more appropriate for analysis using
analytical techniques. Sample pretreatment is an analysis stage where most errors occur
which may affect reproducibility, precision, limits of detection and quantification of various
analytical techniques. In addition, this can be the most time consuming step which then leads
to an increase in the cost of the analysis.1-4 Like any other analytical method, various sample
preparation procedures used prior to the determination of metals in fuels have advantages and
disadvantages. Therefore, before developing a sample pretreatment procedure, there are some
important factors that have to be taken to consideration. These factors include (i) the
analytical technique to be used, (ii) the nature of the sample, (iii) the analyte and its
concentration, (iv) the degree of accuracy and precision required, (v) the availability of
equipment, materials and reagents and (vi) cost of the analysis.2 There are different sample
pretreatment methods that have been developed for metal and metalloid quantification in fuel
samples which may be categorized into preconcentration and non-preconcentration based
sample preparation techniques.
The next section will briefly review the non-preconcentration or non-sorption based
sample preparation techniques.
12
Chapter two:
Literature review
2.1.1 Non-Sorption Sample Preparation Methods
The non-preconcentration based sample preparation methods includes conventional
ashing and acid dissolution, microwave digestion, electrothermal vaporization, dilution with
organic solvents and emulsion/microemulsion
2.1.1.1 Conventional ashing and acid dissolution methods
In this approach, the fuel sample is usually placed in a quartz beaker or a platinum
crucible and heated over a Bunsen gas burner or placed in a furnace to achieve total ashing of
the material. The ash is then dissolved in the mineral acid resulting in an aqueous solution
that can, in principle, be analyzed by any spectrometric technique.5 However, this methods is
the loss of elements such as Hg, halides of As, Se, V and Pb, by volatilization. In addition,
the method is associated with the risk of contamination due to the larger amount of reagents
used, the open environment, as well as the precipitation of sparingly soluble salts (e.g.,
sulfates of Pb, Ba and Sr) during the final dissolution step.5 In addition, the conventional
ashing and acid dissolution methods are time consuming and unsafe due to the evolution of
toxic vapors.2,5,6
2.1.1.2 Microwave assisted-digestion method
Microwave digestion methods are commonly used for analysis of total metal ion content
in organic samples by destruction of carbon metal bonds.7 In these methods, fuel samples are
mixed with mineral acids and oxidizing agents such as nitric acid and hydrogen peroxide in a
closed high-pressure polytetrafluoroethylene (PTFE) vessel at temperatures above the boiling
point of nitric acid.8 The aforementioned method is known as microwave assisted digestion
(MAD), which uses HNO3 and H2O2 for element dissolution in fuel samples.9,10 The use of
concentrated acids in MAD methods show good efficiency of sample digestion, however, it
increases the blank values and cannot be supported by some analytical techniques such as
inductively coupled plasma spectrometry (ICP-MS) and inductively coupled plasma optical
emission spectrometry (ICP OES). Therefore, a subsequent step could be necessary to dilute
or remove the excess acid.11,12 However, as much as MAD methods solve the problem of
volatilization, they as well increase the risks of cross-contamination and incomplete
mineralization of organic matrix.13
13
Chapter two:
Literature review
Microwave-induced combustion (MIC) technique has been proposed to overcome the
problems related to MAD. The MIC procedure involves the use of ammonium nitrate and
dilute nitric acid (as an absorbing solution). The MIC procedures were developed in order to
combine the advantages of microwave assisted digestion and combustion techniques,
allowing the digestion of samples that are difficult to bring into solution with a high sample
throughput.
12,14-16
Furthermore, the application of a reflux after combustion improves the
analyte recoveries. The MIC procedure avoids the use of concentrated acids and significantly
reduces laboratory waste and analysis time, which is an important aspect for routine
analysis.12
2.1.1.3 Electrothermal vaporization method
Electrothermal vaporization (ETV) is an alternative sample introduction technique for
inductively coupled plasma techniques that has been widely studied and applied.17,18 The
main advantages of ETV are the ability of the technique to process semi-solid samples (e.g
slurry), improvement of sensitivity and the ability to analyze microliter volumes of a wide
variety of sample matrices such as fuel, highly acidic and saline solution and solids (Xia et al.
2007).17 Another attractive feature of this method is that in the case of organic samples, it
minimizes carbon formation on components of the equipment, eliminate oxygen addition and
reduces interfering polyatomic species.18,19 For the determination of trace elements in organic
samples, ETV has been coupled with other sample pretreatment techniques (such as
microemulsion and dilution with organic solvents).18,19 Nevertheless, its parameters have to
be optimized for each element and therefore discrete sampling makes sample throughput
significantly slow, even when automatic samplers are used, thus extending the experimental
procedure.6,13 In addition, the multielement capability can also be limited by the differential
volatilization of the element species.18
2.1.1.4 Dilution with organic solvents method
Dilution with organic solvents is one of the simplest sample pretreatment procedures
where a sample is diluted with an appropriate organic solvents such as xylene,20,21
alcohols,22,23 n-hexane, kerosene and methyl isobutyl ketone.24 The use of alcohols for
sample dilution is much safer as compared to other organic solvents. This is because the
procedure involved is simple and the calibration against inorganic standard solutions diluted
with alcohol is possible.23 However, this sample pretreatment approach displays some
14
Chapter two:
Literature review
disadvantages, which increase of analyte concentration due to evaporation of the solvent and
difficulty in handling of some organic solvents with conventional laboratory equipment due
to their different physical properties.2,25 Moreover, this pretreatment method does not reduce
the problem of organic loading and plasma destabilization or extinction in the case of ICP
techniques.2,25 In addition, the stability of the sample solutions depends on the container
material and the analyte concentration may change rapidly after dilution due to adsorption
onto the walls of the recipients.2,25
2.1.1.5 Emulsion or microemulsion method
Emulsion or microemulsion (three-component system) sample pretreatment method
appears to be the most promising approach, because it involves short sample preparation time
and low risk of analyte losses by volatilization or sorption.26 Microemulsion can be prepared
using either detergent or detergentless route. In the case of detergent microemulsion, the fuel
is dispersed in the aqueous phase as micro-drops stabilized by micelles generated by the
addition of a surfactant (such as Triton-100).27,28 The disadvantage of detergent
microemulsion is the low stability of the emulsions which depends on the chemical
conditions, such as pH, characteristics and concentration of the emulsifier, thus affecting the
sensitivity and reproducibility of the analytical instrumental signal.13
Detergentless microemulsion is used to overcome the stability challenge. In this method,
a co-solvent (normally low molecular weight alcohol) allows for the formation of a
homogeneous and long-term stable three component solution containing the aqueous and
organic phase.2,29,30
2.1.2 Preconcentration-Based Sample Preparation Techniques
Extraction of the analyte from organic matrices is an additional way of sample
pretreatment that combines the advantages of separating the analyte from the complex matrix,
by transferring it to an aqueous phase and preconcentrating it at the same time.2 Several
techniques have been reported for the separation and preconcentration of traces of elements
in various samples matrices. These include liquid-liquid extraction,31 co-precipitation,32 cloud
point extraction,33 stir-bar sorptive extraction,34 solid phase extraction,35 hollow fibre-liquid
phase microextraction31 and solid phase microextraction.36 The focus of this study is to search
for suitable preconcentration techniques or procedures that are faster, easier, safer, and less
expensive, to provide accurate and precise data with reasonable detection and quantification
15
Chapter two:
Literature review
limits. Therefore, the methods of choice in this study are solid phase extraction, solid phase
microextraction and hollow fiber-liquid phase microextraction. The other four techniques will
be briefly mentioned while the methods of choice will be discussed in detail.
The following non-focus methods are discussed in brief.
2.1.2.1 Liquid-liquid extraction (LLE)
The principle of LLE is based on the transfer of analyte from aqueous sample to waterimmiscible organic solvent.31 In the case of metals, analyte extraction from aqueous solution
to organic phase takes place after a complexation reaction.37 Liquid–liquid extraction is the
most widely used reliable and efficient technique. This is due to its simplicity, convenience
and wide scope. However, it is a time, reagent and labour consuming procedure, which
cannot be easily automated.38 In addition, the method is unattractive due to the emulsion
formation and the use of toxic organic solvents which can generate large amounts of making
it expensive and environmentally unfriendly.31
2.1.2.2 Co-precipitation
There are three basic steps that are involved in this sample preparation technique. In the
first step, an analyte of interest (metal ions) reacts with an organic or inorganic compound to
forms solid phase. product. In the second step, the major product (precipitate) reacts with
other metals to form chelates. In the third step, solid particles are separated from the aqueous
media and re-dissolved in acid or in an appropriate organic solvent, such as isobutyl methyl
ketone.38 The advantages of co-precipitation method include its simplicity and the fact that
various analyte ions can be pre-concentrated and separated simultaneously from the matrix.
Both inorganic and organic co-precipitants have been used as efficient collectors of trace
elements.37 The main drawback of this technique is that, it is slow and sometimes samples
have to be kept over-night for complete co-precipitation.
2.1.2.3 Cloud point extraction (CPE)
The mechanism of CPE method is as follows: if temperature or pressure is changed or an
appropriate substance is added to the solution, an aqueous solution of some surfactant attains
the cloud point. It then becomes turbid and separates into two isotropic phases. The latter is
composed of a surfactant phase of small volume, which is rich in the surfactant and contains
16
Chapter two:
Literature review
the analyte or metal complex trapped by micellar structures, and a bulk diluted aqueous
phase.37 It should noted that the surfactant-rich phase is hydrophobic and it does not contain
the hydrophilic metal, therefore, chelating agents are normally used to form metal complexes
that can be trapped by micellar structures. This because Separation and preconcentration
based cloud point extraction is an analytical tool that has a great potential yet to be explored
in improving detection limits and other analytical characteristics over those of diverse
methods. Cloud point extraction has been proven to cause lower toxic threats to the
environment.39 Furthermore, CPE has many advantages, such as low cost, safety and speed,
and is a simple procedure with a high capacity to concentrate a wide variety of analytes of
widely varying nature with high recoveries and high preconcentration factors.40 However, the
major limitation of using CPE is that the viscous surfactant-rich phase prevents a smooth and
instant injection to conventional analytical instruments.39
2.1.2.4 Stir-bar sorptive extraction (SBSE)
In this extraction technique, the stir bar is coated with a layer of polydimethylsiloxane
(PDMS) and then used to stir aqueous samples. During the stirring process, the analyte is
being extracted and enriched into the PDMS coating. After extraction, desorption of the
analyte of interest can be achieved either by thermal desorption or liquid desorption.33 The
advantages of using SBSE include high sample capacity, greater sensitivity, straightforward
quantification, higher extraction recoveries of polar and nonpolar compounds, and lower
detection limits. However, the main drawback is that, the technique is not fully automated.41
The following methods have been chosen for investigation in this study.
2.1.2.5 Solid phase extraction
Solid phase extraction (SPE) is an attractive separation and preconcentration procedure
for determination of trace metals in fuel samples by spectrometric techniques.2,34 The
principle of SPE is based on the partitioning of the analytes between liquid (sample matrix)
and solid (sorbent) phase.35 Solid phase extraction enables the preconcentration and
purification of analytes from complex sample matrix by sorption onto a solid sorbent prior to
detection. The fundamental procedure of SPE involves passing the liquid sample through a
column (cartridge) containing a solid sorbent. In the process, the analyte is retained. After the
entire sample has been passed through, the analytes are eluted or stripped off the sorbent
using an appropriate eluent e.g. nitric acid in the case of metals.35 Solid phase extraction
17
Chapter two:
Literature review
displays some important advantages compared to LLE. These include simplicity, flexibility,
high selectivity, automation, rapidity, higher enrichment factors, absence of emulsion, use of
different sorbent materials, low cost because of lower consumption of reagents and more
importantly environmentally friendly. In addition, SPE provides matrix separation, reduces
matrix effects and improves the detection limits of spectrometric techniques such as FAAS
and ICP-MS/OES.42-45
The retention of analytes on the solid sorbent is the most important process for
preconcentration. There are different retention mechanisms that can be used. These include
adsorption, chelation, ion-pairing and ion exchange.35 The retention mechanisms depend on
the nature of the sorbent. Adsorption occurs when the analytes are adsorbed onto the solid
sorbent through van der Waals forces or hydrophobic interactions. The latter occurs when the
highly non-polar solid (reversed phase) sorbent is used. The limitation of hydrophobic
interaction is that analytes (such as trace elements) cannot be adsorbed on the solid sorbent.
This is because trace element species are ionic, thus they cannot be retained by non-polar
sorbents.35 In chelation, sorption with chelating resins is mostly due to the complexation of
trace metals with functional groups. Chelating resins are organic polymers with chemically or
noncovalently bonded functional groups containing nitrogen, oxygen and sulfur.4,35 In the
ion-pairing process, an ion-pair reagent is added to a non-polar sorbent. Ion pair reagents
contain a non-polar portion (such as long chain of aliphatic hydrocarbon) and polar portion
(such as an acid or base). The non-polar portion of the reagent interacts with the non-polar
sorbent whereas the polar portion forms an ion pair with the ionic species present in the
sample. Examples of ion-pair reagents include quaternary ammonium salts and sodium
dodecyl-sulfate, among others 35
In ion exchange, the primary retention mechanism of the analyte is based mostly on the
electrostatic attraction of the charged analyte to the charged functional groups on the surface
of the sorbent. Ion exchange SPE can be used for analytes that are charged when in solution
(usually aqueous, but sometimes organic). Generally, ionic exchange sorbents contain
cationic or anionic functional groups that can exchange the associated counter-ion.35 An ion
exchange sorbent can have strong or weak sites. Strong sites are sites that are present as ionexchange sites at any pH, while weak sites are only present as ion-exchange sites at pH
values greater or less than the pKa of the sorbent. Strong sites include sulfonic acid groups for
cation-exchange and quaternary amines for anion-exchange. Weak sites on the other hand
18
Chapter two:
Literature review
consist of carboxylic acid groups for cation-exchange or primary, secondary and tertiary
amines for anion-exchange.35
Solid phase extraction has applied for separation and preconcentration of metal ions in
different sample matrices, such as environmental, fuel and biological samples, among others.
Procedures based on SPE for metal pre-concentration from fuel samples have been developed
for quantification by spectrometric techniques. In these procedures, different solid sorbents
modified with organofunctional groups (e.g. chelating groups) to extract metal ions have been
used.2,45 A number of selected applications of SPE to the separation and preconcentration of
metal ions in fuels are summarized in Table 2.1.
19
Chapter two:
Literature review
Table 2.1. Selected applications of solid phase extraction for preconcentration of metal ions in fuel samples
Sample
Solid phase material
Analytes
Alcohol fuel
Alcohol fuel
Alcohol fuel
Gasoline
Moringa oleifera
Moringa oleifera
Vermicompost
Silica gel modified with 2aminothiazole groups
Silica gel modified with 2aminothiazole groups
Silica gel modified with 2aminothiazole groups
Amberlite XAD-4
functionalized with 3,4dihydroxybenzoic acid
Cellulose paper
Fuel
kerosene
Fuel ethanol
Ethanol Fuel
Automotive
gasoline
Ethanol fuel
Gasoline
Chromatography paper
XAD- 3,4-dihydroxybenzoic
acid
Ethanol fuel 2,2′-dipyridylamine bonded
silica
Fossil fuels Chitosan microspheres
and biofuels
a
N.I= not included
Analytical technique
References
Zn
Cd
Cd
Cu, Fe, Ni and Zn
Concentration levels
(µg L-1)
N.I.a
N.I.
N.I.
2.8-8.4
FAAS
FAAS
FAAS
FAAS
Alves et al.46
Alves et al.47
Bianchin et al.42
Roldan et al.48
Cu, Fe, Ni and Zn
3.0-11.0
FAAS
Roldan et al.49
Cu, Ni and Zn
4.9-8.1
FAAS
Roldan et al.50
Cu, Fe, Ni and Zn
7.8-44
FAAS
Teixeira et al.51
Cu and Fe
98-446
XRF
Teixeira et al.52
Cu, Fe, Ni and Zn
Cu, Fe, Pb and Zn
129-309
2.3-22.4
EDXRF
FS-FAAS
Teixeira et al.53
Santos et al.45
FAAS
Vieira et al.54
FAAS
Pradoa et al.55
Fe, Cr, Cu, Co, Pb, 11-66
Ni and Zn
Cu, Ni and Zn
0.4-4.8
20
Chapter two:
Literature review
2.1.2.6 Solid phase microextraction (HF-SPME)
Solid-phase microextraction (SPME) was introduced in 1990 by Arthur and Pawliszyn to
tackle the need to ease rapid sample preparation both in the laboratory and on-site where the
investigated system is located.36 In this technique, a sorbent is coated on a fused silica fibre.
The latter is then fitted inside the needle of a syringe-like SPME holder. Solid phase
microextraction is used to extract and preconcentrate analytes of interest by immersing the
fibre into the sample. There are two basic steps involved in SPME process: (i) partitioning of
analytes between the extraction phase and the sample matrix and (ii) desorption of
concentrated extracts into an analytical instrument.39
The attractive features of using SPME technique include short sample preparation times;
small sample volumes; analyte concentration from liquid, gaseous and solid samples; solventfree extraction technique and easily automated to allow high-throughput analysis.39 However,
the main limitation of this technique is related to polymeric extractant phase and the
desorption process (Pena-Pereira et al. 2009).31 This limitation can be avoided by the use of a
hollow fibre membrane. Hollow fibre-SPME involving the use of a membrane as the
adsorption material integrates sampling, extraction and concentration into a single step and
inherits the advantages of SPME and membrane separation.56 In this study HF-SPME will be
used for the separation and preconcentration of metal ions from organic matrices.
Applications of HF-SPME for preconcentration of metal ions in aqueous samples are
reported by Cui et al.,56 Mester et al.,57 Bravo-Sanchez et al.,58 Es’haghi et al.59 and Huang
and Hu,60 among others. However, there is limited or no information about the application of
SPME for preconcentration of trace elements in fuel samples. Therefore, the interest of this
study is in the development of HF-SPME that would be used as a sample pretreatment
method for preconcentration of metal ions in fuel samples.
2.1.2.7 Hollow fibre-liquid phase microextraction (HF-LPME)
Hollow fibre liquid-phase microextraction (HF-LPME) allows extraction and
preconcentration of analytes from complex sample matrices in both a simple and inexpensive
way (Pena-Pereira et al. 2009).36 The principle of this technique is based on the extraction of
analytes from an aqueous sample, through the organic solvent immobilized as a supported
liquid membrane and into the acceptor solvent placed inside the lumen of the hollow
fiber.61,62 In HF-LPME, the hollow fibre is first soaked in an organic solvent to immobilize
21
Chapter two:
Literature review
the solvent into the pores, so that microextractant solvent is not in direct contact with the
sample solution. According to Rasmussen and Pedersen-Bjergaard63 the organic solvent used
within the pores of the hollow fibre has to satisfy the following criteria: (i) It should be
immiscible with water; (ii) it should be strongly immobilized in the pores of the hollow fibre
and (iii) it should provide appropriate extraction selectivity and high extraction recoveries.
However, HF-LPME is more suitable for extraction of analytes from aqueous media.
The analytes are extracted from the sample through the organic phase in the pores of the
hollow fiber, and then into an acceptor solution inside its lumen.64,65 The use of small pore
size serves for two reasons. Firstly, to prevent large molecules and particles present in the
donor solution from entering the accepting phase. Secondly, to prevent most macromolecules
that are not soluble in the organic phase present from entering the hollow fibre, thus yielding
very clean extracts.17
The extraction of analytes by HF-LPME system is based on the diffusion process. The
HF-LPME is facilitated by high partition coefficients for the target analytes which posses a
challenge when hydrophilic analytes such as metal ions are to be extracted. The poor partition
coefficient prevents hydrophilic analytes from being extracted in systems based on diffusion
alone.64 This challenge can be eliminated by using an active transport HF-LPME process,
whereby the carrier (chelating agent) is added to the sample solution66 and the complexed
analyte is extracted into the acceptor solution.31
To the best of our knowledge there is a limited or no information about the application of
HF-LPME for preconcentration of trace elements in fuel samples. Therefore, the interest of
this study is in the development of HF-LPME method for preconcentration of metal ions in
fuel samples.
2.2 ANALYTICAL TECHNIQUES FOR THE DETERMINATION OF METAL IONS
IN ORGANIC SAMPLES
The majority of analytical techniques reported in the literature for determination of metal
ions in organic sample are based on atomic spectrometric methods.67 These techniques
include inductively coupled plasma-optical emission spectrometry (ICP OES) flame atomic
absorption spectroscopy (FAAS) and electrothermal atomic absorption spectrometry
(ETAAS). In addition, electroanalytical
68
and XRF53 techniques have also been reported.
This study focuses on reviewing atomic spectrometric techniques. (AST). In this study we
selected a number of AST techniques that were investigated in detail while others were not.
22
Chapter two:
Literature review
For those techniques in which no investigation were done, we refer to them as non-focus
techniques
The first part of the review of atomic spectrometric techniques describes the non-focus
analytical detection techniques namely FAAS and ETAAS as follows:
2.2.1 Flame Atomic Absorption Spectrometry
The advantages of FAAS include high sample throughput, relatively easy to use and high
precision.69 The main disadvantages of using FAAS include poor limits of detections,
element limitations, several interference (of which some are severe, thus limiting the method
application), 1-10 elements per determination and lack of screening ability.69 Despite the
above mentioned limitations, FAAS combined with independent sample pretreatment
procedures such as SPE,69 have been successfully and extensively applied for the
determination of metal ions in fuel samples.
Roldan et al.48 developed FAAS method for the determination of Cu, Fe, Ni and Zn in
gasoline. Solid phase extraction based on silica gel modified with 2-aminothiazole groups
was used for preconcentration of metal ions prior to their FAAS determination. In another
report, Santos et al.45 developed an Amberlite XAD-4 functionalized with 3,4dihydroxybenzoic acid SPE method for separation and preconcentration of metal ions prior to
sequential multi-element flame atomic absorption spectrometry (FS-FAAS). A number of
procedures for analyte separation and preconcentration of trace elements from fuel samples
prior to FAAS determination have also been reported by different authors (Table 2.1). Other
methods for sample preparation method used for determination of metal ions using FAAS are
reported by Pignalosa et al.70 and Reis et al.71
2.2.2 Electrothermal Atomic Absorption Spectrometry
Electrothermal atomic absorption spectrometry (ETAAS) with a graphite furnace is one
of the analytical techniques designed to perform the quantitative analysis of metals in a wide
variety of samples. The major advantages of ETAAS over FAAS include: (i) capability of
using relatively small volumes or masses of samples, (ii) direct determination of analytes in
differences sample matrices with minimal sample preparation and (iii) high sensitivity and
capability to deal with high organic loads. The high sensitivity is due to the longer residence
time of the sample in the source and production or higher proportion of atoms. In addition,
the high sensitivity of ETAAS (low detection limits) is related to the 100% introduction of
23
Chapter two:
Literature review
the sample into the analytical volume as compared to 5% in the case of FAAS.2,69 In addition,
ETAAS technique allows complete elimination of the organic matrix and semi-solid (slurry)
matrices if an appropriate heating program and a suitable chemical modifier are used.
Therefore, direct analysis of fuel samples is possible.2
De Jesus72 described the determination of arsenic and cadmium in crude oil using a direct
sample introduction ETAAS, where a mixture of Pd, Mg and Triton X-100 was employed as
a chemical modifier, and the results obtained by direct sampling method were in agreement
with the comparative method. Procedures using direct determination of metal ions in fuel
samples using ETAAS have been reported in the literature.73,74 The direct introduction of fuel
samples to ETAAS for the determination of trace metals is associated with some drawbacks,2
such as volatility, flammability, immiscibility with water, low viscosity and excessive
spreading of fuel during thermal pretreatment. The spreading is due to the temperature
gradient when longitudinally heated atomizers are used.2,74
To overcome this problem, the use of emulsions or microemulsions combined with
ETAAS appears to be the most promising approach.74 Microemulsion methods combined
with ETAAS have been reported in several papers for the determination of trace elements in
fuel samples.28,75-77 Some metals such as Hg, As and Sb are present in fuel samples in trace
concentrations that are lower than the limits of detection of ETAAS,2 thus, their
determination can be achieved using hydride generation ETAAS techniques.2,26,78. Even
though ETAAS have been extensively applied for fuel analysis, the main drawbacks of this
technique includes slower analysis time, chemical interferences, elemental limitations (1-6
elements per determination), lack of screening ability, limited dynamic range and loss of
analyte during ashing. In addition, because of low sample throughput when using ETAAS
compared to inductively coupled plasma-based methods, this method is not common in
routine analysis (Donati et al. 2013).79
The following analytical techniques have been chosen (focus techniques) for
determination of the analytes in this study.
2.2.3 Inductively Coupled Plasma (ICP) Techniques
There are two widely used ICP techniques, these included inductively coupled plasma
optical emission spectrometry (ICP OES) and inductively coupled plasma mass spectrometry
(ICP-MS). These techniques have the same ion excitation source which is ICP. However,
they have different optics and detectors.
24
Chapter two:
Literature review
An inductively coupled plasma is a well known, high-temperature source (6000 to 10 000
K) that is suitable for the atomization, ionization and excitation of elemental species.80 For
both ICP OES and ICP-MS, the plasma is formed in precisely the same way.81 The plasma is
formed within a quartz torch, which consists of three concentric quartz tubes, through which
argon gas flows as shown in Fig. 1.1. The forms due to the interaction of an intense magnetic
field (produced by radiofrequency (RF) passing through a copper coil) on a tangential flow of
gas at about 15 L min-1 flowing through a torch.81 The outer (plasma) gas flow (10-15 L min1
) is introduced into the torch tangentially to enable it to form a vortex flow. The plasma gas
flow sustains and stabilizes the high temperature ICP and positions the plasma relative to the
outer walls and the induction coil, preventing the walls from melting and facilitating the
observation of emission signals. The central (auxiliary) gas flow (0.5-1.5 L min-1) determines
the position of the plasma above the intermediate ring the torch and within the RF coil. The
inner (nebulizer) gas flow (0.7-1.5 L min-1) transports the analyte (sample aerosols) to the
plasma.80,82
Fig. 2.1. A schematic diagram of an ICP assembly showing the three concentric quarts tubes
composing the torch and the RF coil (adopted from Ref.83)
Although the formation of the plasma is similar for both techniques, its role in ICP OES
is different compared to ICP-MS. In ICP OES, the plasma torch (normally positioned
vertically) is used to generate photons of light, by the excitation of electrons from the groundstate atom to a higher energy level. The atomic and ionic excited state species are then
relaxed to the ground state via the emission of photons. Therefore, the wavelength of the
25
Chapter two:
Literature review
photons can be used to identify the elements of interest.81,82 In ICP-MS on the other hand, the
plasma torch (normally mounted horizontally) is used to generate positively charged ions, not
photons. In ICP-MS, photons are stopped from reaching the detector because they have the
potential to increase signal noise.81
2.2.3.1 Inductively coupled plasma-optical emission spectrometry
Inductively coupled plasma-optical emission spectrometry is one of the most widely used
analytical techniques for determination of trace elements in different sample matrices such as
food, environmental, petroleum products and biological samples, among others.82 The ICP
OES technique is used for both identification and quantification of elements in a sample. Fig.
2.2 shows the major components and layout of a typical ICP OES instrument.
As mentioned in Section 2.2.4, the principle of ICP is based on the spontaneous emission
of photons from atoms and ions that have been excited in RF discharge.82 In ICP OES, the
liquid sample is converted to an aerosol spray and directed to the central channel of argon gas
plasma contained by a strong magnetic field. Since the plasma is at high temperature, the
aerosol droplets are quickly vaporized. The elements in the sample become excited and the
electrons emit energy at a characteristic wavelength as they return to ground state. The
emitted light is then measured by optical spectrometry.84
In comparison with FAAS and ETAAS, ICP OES has a higher atomization temperature, a
more inert environment, and the natural ability to provide simultaneous determinations for up
to 70 elements. These advantages (mainly high temperature) make the ICP OES less
vulnerable to matrix interferences, and better able to correct for interferences when they
occur. Typically, limits of detection can range from parts per million (ppm) to parts per
billion (ppb). In cases where sample volume is not limited, ICP OES can provide LODs as
low as, or lower than its best competitor, GFAAS.82
Fig. 2.2 shows the major components and layout of a typical ICP OES instrument. It
consists of a sample introduction chamber, an ICP (plasma source) spectrometer, a detector
and a data processing system.
26
Chapter two:
Literature review
Fig. 2.2. Schematic diagram of ICP OES instrument showing its major components and
layout (adapted from Ref.85)
2.2.3.1.1 Sample introduction for ICP OES
An ideal sample introduction system should meet the following requirements; (i) amenity
to samples in all phases (solid, liquid, or gas), (ii) tolerance to complex matrices, (iii) the
ability to analyze very small amounts of samples, (iv) excellent stability and (v)
reproducibility, high transport efficiency, simplicity, and low cost.82 As seen from Fig. 2.2, an
ICP OES sample introduction system is composed of a peristaltic pump, spray chamber and
nebulizer. The liquid samples are pumped by a peristaltic pump into the instrument and then
the liquid is converted into an aerosol or mist through a nebulization process. After the
sample aerosol is produced by the nebulizer, it is transported to the plasma where its
desolvation, vaporization, atomization, ionization and excitation take place. It should be
noted that only very small droplets in the aerosol are suitable for injection into the plasma.
Larger droplets in the aerosol are removed by the spray chamber which is placed between the
nebulizer and the torch. An additional purpose of the spray chamber is to minimize signal
pulsation caused by the peristaltic pump.86,87
2.2.3.1.2 Spectrometer
The spectrometer is used to separate the light emitted the excited species in the plasma
into the individual wavelengths so that the emission from each excited species can be
identified and its intensity can be measured without interference from emission at other
27
Chapter two:
Literature review
wavelengths. This is achieved by the use of either monochromator or polychromator
spectrometers. For this reason, ICP OES system can be categorized according to the method
it uses for analysis and it can be divided into three different systems. These include single
channel (sequential), simultaneous (multi-element analysis) and simultaneous/sequential
systems. Sequential ICP OES monochromator system allows for the analysis of only one
analytical line at a time. In this system, to scan the whole region of the electromagnetic
spectrum, the detector has to be held at a fixed position and the grating is turned sequentially.
The advantage of using monochromator-based ICP OES systems is their spectral flexibility
which allows for the determination of any element whose emission can be measured by the
technique. However, the major drawback is that they require large amounts of sample, time
consuming and have a lower sample throughput.82,86,87
In the Simultaneous ICP OES polychromator system, each emission line can be observed
during the entire sample introduction period, and theoretically more samples can be analyzed
in a shorter period of time. The advantage of this system is that it has a high sample
throughput rate and can be programmed for 20 to 30 spectral lines. In addition, due to the fast
speed of analysis, all elements can be analysed at the same time and minimal sample volume
is required. The third system combines the advantages of both simultaneous and sequential
ICP OES systems.82,86,87
2.2.3.1.3 Detector
The detector converts light energy (photons) from analyte emissions generated in the
plasma into an electrical signal that can be quantified. Conventional ICP OES instruments
commonly used photomultiplier tube (PMT) detection systems. The latter converts the light
intensity into an electrical signal that can be quantified. The electrical signal is therefore
related to the concentration of the elements in the sample solution.88 The disadvantages of
PMT detectors include less flexibility in adapting to a laboratory’s changing requirements
and limited wavelength selection such that foreknowledge of analyte wavelengths was
required before use. To overcome the problem of wavelength selection, additional PMT
detectors were necessary to access additional analyte wavelengths.82,89 Recently, instruments
using solid state detection systems in charge-coupled device (CCD) and charge injection
device (CID) have become commercially available. These type of detectors overcame many
of the deficiencies inherent in PMT-based ICP OES and they can observe all wavelengths at
the same time.82,88,89 Charge-coupled devices in particular are well-known for their high
28
Chapter two:
Literature review
sensitivity and low noise characteristics. Thus, providing low detection limits and superior
signal-to-noise ratio performance.89
2.2.3.1.4 Data processing system
In ICP OES, computers are used to control the automatic sampler, the instrument and data
collection. Data processing systems can be divided into (i) a method file (to run the
instrument) and (ii) sequence file (to tell the instrument where a sample is in the autosampler
tray, when to run it and where to store the collected data file).
2.2.3.1.5 Interferences
An ICP OES is one of the atomic spectrometry techniques with the least number of
interferences.82 This is because the argon plasma is inert when compared to the chemical
reactivity of a flame. In addition, the high temperature of the plasma helps to reduce chemical
interferences.82 However, chemical interferences in ICP OES caused by easily ionized
elements (EIEs) such as alkaline elements, do exist. The easy ionizability of these elements is
due to their low ionization potential. Depending upon the analyte species, high concentrations
of alkali metals can suppress or enhance emission signals.82 This kind of interference can be
eliminated by diluting the sample solution to the point that the EIE effect is not measurable.
Occasionally, higher RF power or mathematical correction may be used to compensate for
EIE interference. In addition, instrumental conditions such as observation width, viewing
height and viewing volume can be chosen to minimize such interference.90 Apart from
chemical interferences, other interferences such as matrix effects (nebulizer interferences)
and spectral (background) interferences are common in ICP OES system. Matrix effect
interferences arise from physical and chemical differences between reference standards and
samples, or between samples. The latter is due to the inconsistence presence of matrix salts
and organic compounds or different viscosities and surface tension of the liquids. The use of
internal standards can be used to eliminate matrix effects such as viscosity differences
between samples and calibration standards.91
Spectral interferences are most common in ICP OES and they are due to the multielement
nature of the ICP. These interferences can be divided into two types, namely spectral line
coincidence or overlap and enhanced spectral background caused by recombination
continuum radiation or stray light from intense emission lines, normally those of alkaline
earth elements.92 The spectral overlap can be eliminated by using monochromators with
29
Chapter two:
Literature review
higher resolution. In addition, it can be solved by monitoring a secondary emission line for
that particular analyte or use a background correction technique. The second type of spectra
interferences can be avoided by adjusting the temperature of the plasma (in some
instruments) or monitoring alternate lines.
2.2.3.2 Inductively coupled plasma-mass spectrometry
Inductively coupled plasma-mass spectrometry is a combination of an atomization source
and analytical technique, that is, inductively coupled plasma (ICP) and mass spectrometer
(MS). It is a well-established, fast, precise and accurate multi-element analytical technique
for the determination of trace elements in liquid and solid samples.93 The ICP-MS technique
was first developed in the early 80's by Houk et al.94 Since then, it has become increasingly
popular for the analysis of trace elements in geological, chemical, environmental, biological,
industrial, metallurgical, petrochemical, medical and archaeological materials.93,95 Figure 3
shows a schematic diagram of an ICP-MS instrument reproduced from Linge and Jarvis96.
Fig. 2.3. Schematic diagram of inductively coupled plasma mass spectrometer. Diagram
reproduced from Linge and Jarvis96.
2.2.3.2.1 Sample introduction for ICP-MS
The sample introduction system of an ICP-MS is similar to that of ICP OES and consists
of a peristaltic pump, autosampler, nebulizer and spray chamber.81 Depending on the sample
introduction system, ICP-MS can accept both solid and liquid samples.81,97 The main function
30
Chapter two:
Literature review
of sample introduction system is to produce a fine aerosol of the sample using a nebulizer and
spray chamber for liquid samples and laser ablation for solid samples.81 Since the majority of
ICP-MS applications focuses on liquid samples, the mechanism for the formation of aerosols
for liquid samples will be described briefly little bit in detail compared to laser ablation.
For liquid samples analysis, the sample is pumped via peristaltic pumps into the nebulizer
where it gets aspirated with high velocity argon to form a fine mist known as aerosols. The
fine droplets pass through the spray chamber before they enter the plasma and the small
droplets are separated from larger droplets. The spray chamber also removes solvent from the
aerosol, thus improving the ionization efficiency.96 In general, only 2% of the original aerosol
is required to produce droplets small enough to be vaporized in the plasma torch.81,97-99
2.2.3.2.2 Interfaces
The ICP works at high temperature (6000 to 10 000 K) and atmospheric pressure (760
Torr) while MS requires vacuum (approximately 10-6 Torr). Thus, the interface between the
ICP and MS components becomes very crucial. The role of the interface is to transport the
positively charged ions efficiently, consistently and with electrical integrity from the ICP to
the mass spectrometer region.81 The interface consists of two metallic cones, that is sampler
and skimmer cones, which are maintained under vacuum with a mechanical roughing pump.
These cones have a very small orifice diameter, 0.8-1.2 mm and 0.4-0.8 mm for sampler and
skimmer cones, respectively.81 The plasma gas expands as a supersonic jet and the sample ions
are passed into the MS system at high speeds, expanding in the vacuum system. 96,97
2.2.3.2.3 Ion Focusing Systems
The ion focusing system is composed of ion optics (lenses) and the latter is situated
between the skimmer cone and the mass analyzer. The ion focusing system has two important
functions. Firstly, its role is to transport the analyte ion extracted from the interface cones and
focus them into the mass analyzer. Secondly, it rejects the matrix component and non-analyte
species such particulates, neutral species, and photons from getting through to the mass
analyzer and the detector. This is because these species cause signal instability and contribute
to background noise levels, which ultimately affect the performance of the system.81
2.2.3.2.4 Mass spectrometer/ mass analyzers
The role of the mass analyzer is to separate the analyte ions according to their mass to
charge (m/z) ratio. There are three different kinds of commercially available mass analyzers
31
Chapter two:
Literature review
which can be employed to separate isotopes. These include quadrupole mass filters, double
focusing magnetic sector and time-of-flight.81 The Perkin-Elmer Sciex ELAN 6000 used for
our study has a quadrupole mass analyser; therefore the latter will be further discussed. A
quadrupole mass analyzer consists of four hyperbolic rods (each opposing rod pair connected
electrically) that are parallel to and equidistant from the ion beam.81,96 A direct current (DC)
component is applied to two rods and the radio frequency (RF) component is applied to the
other pair of rods.81 Analyte ions produced in the ICP are focused and passed in between the
rods. The ions travel down the central axis and the voltages applied to the rods cause the ions
to oscillate.96 The extent of the oscillations of ions is influenced by the charge and the mass
of the analyte ions. For instance, extreme oscillations cause the unstable ions to collide with
the rods and be ejected from the stable transmission region. Only ions of a single m/z
(resonant ions) that have stable oscillatory paths through the rods can emerge and exit the
quadrupole.96
2.2.3.2.5 Reaction/ collision cell
A number of elements are known as having poor detection limits by ICP-MS. The reason
for poor detection limits is due to spectral interferences (polyatomic interferences) generated
by ions derived from the plasma gas, matrix components, or the solvent–acid used to get the
sample into solution.81 The traditional way of overcoming this problem is use of the cold/cool
plasma approach; whereby low temperatures are applied to the plasmas to reduce the
formation of the interferences. However, this approach can be difficult to optimize, time
consuming to change back and forth between normal- and cool-plasma conditions.
Furthermore, it is effective for a few of the interferences and susceptible to more severe
matrix effects.81 The development of ICP-MS equipped with collision/reaction cell
technology is another way to overcome problems linked with polyatomic interferences. The
collision/ reaction cell is positioned between the ion focusing system and mass analyzer. In
collision/reaction cell techniques, different gases (usually H2, NH3, CH4 or He) or a
combination of them are used to eliminate polyatomic ions such argon-based, O2+, N2+ and
CO+, among others, without affecting the analyte ion of interest via chemical reactions or
interactions.100
32
Chapter two:
Literature review
2.2.3.2.6 Detector
Ions signals are measured with the detector after they have passed the quadrupole. The most
common type of ion detector used in an ICP-MS system is the channel electron multiplier. The
latter is an open glass cone coated with a semi conductor type material that generates electrons
when ions hit its surface.96 The front of the cone has a negative voltage applied to it to attract
positively charged ions.96,97 When each ion strikes the front surface, additional secondary
electrons are formed and are attracted towards the grounded end. As these electrons strike the
surface of the tube, more electrons are formed and the process continues to form a discrete
pulse of about 108 electrons.96
2.2.3.2.7 Interferences
Inductively coupled plasma-MS displays a number of advantages such as fast quantitative
and semi-quantitative trace element analysis, multi-element analysis, analysis of both solid
and liquid samples and small sample quantities are required (approximately 3-5 mL for liquid
samples). In addition, ICP-MS offers a wide analytical range (µg L-1-mg L-1) for liquid
samples), high sensitivity, good precision and accuracy and the possibility of measuring
isotopes.93 Besides the many advantages of this technique offers, the interferences are its
biggest limitation. These interferences can be divided into two major groups, namely spectral
and non-spectral interferences.98 Some of the spectral interferences can be avoided. For
instance, elemental isobaric interferences can be avoided by choosing alternative, noninterfered analyte isotopes. However, this is not applicable for other elements such as arsenic.
Furthermore, isobaric interferences can be corrected mathematically by monitoring the
intensity of an isotope of the interfering element which is free from spectral interferences.
Polyatomic interferences on the other hand can be avoided by using interference free analyte
isotopes, removing matrix, using alternative sample introduction, mathematical correction
equations, as well as, using cool plasma.98 However, mathematical corrections rarely work in
real applications. For this reason, collision/reaction cell is the best strategy for the of
elimination polyatomic interferences. In addition, high resolution mass analyzers are used to
correct for these kinds of interefrences.98 The doubly charged interferences can be avoided by
optimizing the instrument conditions.
Non-spectral interferences can be divided into two categories; physical signal suppression
as a result of the presence of organic or undissolved solid in the sample matrix and matrix
interferences. They can be manifested in different ways, for example, if the sample is diluted
33
Chapter two:
Literature review
with organic solvents, the sensitivity decreases due to the effect of cooling of the plasma.
Moreover, the direct introduction of organic samples into the plasma requires special care
because it may destabilize or extinguish the plasma.10,18 This problem can be avoided by
using alternative sample preparation methods such as separation and preconcentration
techniques to remove organics to minimize matrix effect.
2.2.4.3 Application of ICP OES and ICP-MS for determination of metal ions in fuel
samples
Inductively coupled plasma-MS and OES techniques have an advantage of multielement
detection capabilities technique. In both techniques, the direct introduction of organic
solvents requires special care, as the organic load may de-stabilize or extinguish the plasma.18
In addition, other problems that appear in ICP-MS are formation of carbon deposits on the
sampler and skimmer cones and in the ion lens of the mass spectrometer.2,18 For these
reasons, sample introduction techniques (such as electrothermal vaporization) and
pretreatment methods (such as microwave-assisted digestion) have been developed and
reported in the literature.10,101 Moreover, Several methods for the determination of metal ions
in petroleum products have been developed using ICP OES and ICP-MS and selected
applications are summarized in Table 2.2.
34
Chapter two:
Literature review
Table 2.2. Application of ICP OES and ICP-MS for determination of metal ions in petroleum products
Sample
Sample treatment
Biodiesel and vegetable oil Alcohol dilution
Detection technique
ICP OES
Gum deposits
Microwave-assisted digestion
ICP OES
Petroleum products
Direct introduction using
ICP OES and ICP-MS
pneumatic, ultrasonic and
microflow pneumatic
nebulizers
Ultrasonic nebulization of ICP-MS
toluene solutions
Crude oil
Crude oil distillation
products
Gasoline
Diesel and biodiesel
Ethanol fuel
Fuels and light petroleum
Asphaltene
Crude fuel oil
Crude oil
Analytes
Ca, Cu, Fe, K, Mg, Na, P,
S and Zn
Al, Ca, Co, Cu, Fe, Mg,
Mn, Ni, Pb and Zn
Al, Ba, Ca, Co, Cr, Cu, Fe,
Hg, Mn, Mo, Ni, P, Pb, S,
Si, Sn, V and Zn
References
Chaves et al.23
Dantas et al.102
Lienemann et al.103
Al, Ti, Fe, Zn, Sr, Ag, Sn, Duyck et al.104
Pb V, Ni, Co, Y, Mo, Cd,
Ba and La,
Cu
Kowalewska et al.105
Microwave-assisted digestion
ICP-MS
Emulsion
Emulsion
Direct introduction
Direct flow injection using
microflow nebulizer and
heated spray chamber
Sonication or vortex agitation
Detergentless microemulsions
ETV-ICP-MS
ETV-ICP-MS
ETV-ICP-MS
ICP-MS
Cu, Mn, Ni and Sn
Co, Cu, Fe, Mn, Ni and V
Ag, Cd, Cu, Pb and Tl
Ti, V, Fe, Ag, Cd, Cd, Hg,
Pb, Cr, Ni and Mo
ICP OES
ICP OES
Fe, Ni and V
de Souza et al.108
Mo, Zn, Cd, Ti, Ni, V, Fe, de Souza et al.109
Mn, Cr and Co
Pereira et al.12
Microwave-induced
combustion
35
Saint’Pierre et al.18
Chaves et al.106
Saint’Pierre et al.19
Caumette et al.107
Chapter two:
2.3 CHEMOMETRIC TOOLS
Literature review
FOR OPTIMIZATION OF
ANALYTICAL
METHODOLOGIES
Optimization of an analytical method refers to improvement of its performance in order
to obtain the maximum benefit from it.110,111 The term “optimization” has been widely used
in analytical chemistry to discover optimum conditions at which the developed method can
produce the best possible analytical response.110,111 Conventionally, optimization in
analytical chemistry has been performed using a univariate technique, which means,
monitoring one factor at time. The disadvantages of this method are as follows; (i) it may
lead to ambiguous results and interpretation because the interactive effects among the
variables are not examined. (ii) Univariate optimization increases the number of
experiments to be conducted. Therefore, this leads to an increase in analysis time as well as
an increase in the consumption of reagents and materials.110,112
Chemometric tools (multivariate statistic techniques) have commonly been used to
overcome the problems associated with univariate techniques. The advantages of
multivariate statistical techniques include reduction in the number of required experiments,
thus, resulting in lower reagent consumption and significantly less laboratory work.
Consequently, multivariate techniques are faster to implement and more cost-effective than
traditional univariate approaches. In addition, multivariate statistic techniques allow for the
simultaneous study of several experimental variables and the development of mathematical
models that permit the assessment of the relevance and statistical significance of factors
being studied.110,113,114 Furthermore, these techniques facilitate the evaluation of interaction
effects between factors.110 There are two different types of variables that exist in
multivariate techniques (designs), namely analytical responses and factors. The analytical
response can be qualitative or quantitative. Optimization of analytical methods using
multivariate designs can be accomplished by the use of experimental designs of two types,
namely first and second order.
The use of chemometric tools for optimization of analytical methodologies has been
widely reported.7,22,47,73,116 They have been used for optimization of sample preparation
strategies and analyte determination using microwave-assisted digestion,7 instrumental
variables22,73,115,116 and solid-phase extraction conditions.47 Table 2.3 summarizes relevant
examples of first-order and second-order designs for the optimization of some analytical
methodologies.
36
Chapter two:
Literature review
Table 2.3. Selected applications of chemometric tools for multivariate optimization of
analytical methodologies
Analytes
Cu
Zn
Cu, Fe and Pb
Cu, Pb, Ni and
Cd
As
Al, Cu, Fe, Ni
and Zn
Al, Ca, Cu, Fe,
Mg, Ni, Pb and
Zn
As
Cd
Techniques
Spectrophotometry
Emulsion breaking
Detergent emulsion
Experimental conditions
in GFAAS
Acid digestion
Microwave-oven
digestion
Microwave-assisted
digestion
Samples
Sugar cane spirit
Diesel oil
Naphtha
Biodiesel
References
Caldas et al.117
Cassella et al.118
Brum et al.76
Lobo et al.112
Gasoline
Diesel oil
Becker et al.26
Sant’Ana et al.119
Gum deposits
Dantas et al.102
Experimental conditions Gasoline
in HGFAAS
Online SPE
Fuel alcohol
Trindade et al.78
Bianchin et al.42
2.3.1 First Order Designs
First order designs are commonly used in exploratory studies when a large number
of factors need to be considered or screened. In actual fact, these designs are used in an
attempt to identify factors that demonstrate large main effects and to discard any factors,
from further study, that have no noticeable effects.113 In these designs, the crucial
assumption is that all interactions are insignificant, including two-factor interactions.113
The equation for first order designs is as follows:113,114
(1)
Y  a1 b1 A  c1 B  d1 AB
where Y is the experimental analytical response, A and B represent the variables to be
optimized, a1 is an independent term, b1 and c1 coefficients of the linear terms and d1 is the
coefficient of the interaction term. There are two commonly used first order designs,
namely factorial and Plackett-Burman designs and this study focuses on factorial design.
The latter was used because only 3-4 variables were monitored.
2.3.2 Second Order Designs
In cases where the linear model (first order design) is not sufficient to represent the
experimental data adequately, second order designs are performed in addition to those of
37
Chapter two:
Literature review
factorial design. The results obtained from the second order designs can be used to
determine a quadratic response surface113,115 and the latter can be described by Eq. 2.114
Y  a2  b2 A  c2 B  d 2 A2  e2 B 2  f 2 AB
(2)
where Y is the experimental analytical response, A and B represent the variables to be
optimized, a2 is an independent term, b2 and c2 are coefficients of the linear terms, d2 and e2
coefficients of the quadratic terms and f2 is the coefficient of the interaction term. The
advantage of using second order designs is that they determine influence of the variables to
be optimized on the response. In addition, they enable the response function to be obtained
and optimized.114 The three commonly used second order designs include Box-Behnken
design, central composite design and Doehlert matrix. The software used in this study had
Box-Behnken and central composite designs. Therefore, this study used central composite
and Box-Behnken designs.
2.4 REFERENCES
1. Oliveira, E. D. 2003. Sample preparation for atomic spectroscopy: evolution and future
trends. Journal of the Brazilian Chemical Society, 14, 174-182.
2. Korn, M. D. G. A., De Andrade, J. B., De Jesus, D. S., Lemos, V. A., Bandeira, M. L. S. F.,
Dos Santos, W. N. L., Bezerra, M. A., Amorim, F. A. C., Souza, A. S. & Ferreira, S. L. C.
2006. Separation and preconcentration procedures for the determination of lead using
spectrometric techniques: A review. Talanta, 69, 16-24.
3. Zhao, L., H. K. Lee & Majors, R. E. 2010. The Use of Hollow Fibers in Liquid-Phase
Microextraction.
LCGC
North
America,
29
(3).
Available
at
http://www.chromatographyonline.com/lcgc/article/articleDetail.jsp?id=682847 accessed
11 April 2011
4. Türker, A. R. 2007. New sorbents for solid-phase extraction for metal enrichment. Clean –
Soil, Air, Water, 35, 548-557.
5. Duyck, C., Miekeley, N., Porto Da Silveira, C. L., Aucélio, R. Q., Campos, R. C.,
Grinberg, P. & Brandão, G. P. 2007. The determination of trace elements in crude oil and
its heavy fractions by atomic spectrometry. Spectrochimica Acta Part B: Atomic
Spectroscopy, 62, 939-951.
6. Wang, T., Jia, X. & Wu, J. 2003. Direct determination of metals in organics by inductively
coupled plasma atomic emission spectrometry in aqueous matrices. Journal of
Pharmaceutical and Biomedical Analysis, 33, 639-646.
7. Santelli, R. E., Bezerra, M. D. A., De Santana, O. D., Cassella, R. J. & Ferreira, S. L. C.
2006. Multivariate technique for optimization of digestion procedure by focussed
38
Chapter two:
Literature review
microwave system for determination of Mn, Zn and Fe in food samples using FAAS.
Talanta, 68, 1083-1088.
8. Soylak, M., Tuzen, M., Souza, A. S., Korn, M. D. G. A. & Ferreira, S. L. C. 2007.
Optimization of microwave assisted digestion procedure for the determination of zinc,
copper and nickel in tea samples employing flame atomic absorption spectrometry. Journal
of Hazardous Materials, 149, 264-268.
9. Munoz, R. A. A., Correia, P. R. M., Nascimento, A. N., Silva, C. S., Oliveira, P. V. &
Angnes, L. 2007. Electroanalysis of Crude Oil and Petroleum-Based Fuel for Trace
Metals: Evaluation of Different Microwave-Assisted Sample Decompositions and
Stripping Techniques. Energy & Fuels, 21, 295-302.
10. Bettinelli, M., S. Spezia, U. Baroni And G. Bizzarri. 1995. Determination of trace
elements in fuel oils by inductively coupled plasma mass spectrometry after acid
mineralization of the sample in a microwave oven. Journal of Analytical Atomic
Spectrometry, 10, 555-560.
11. Stewart, I. I. & W. Olesik, J. 1998. Steady state acid effects in ICP-MS. Journal of
Analytical Atomic Spectrometry, 13, 1313-1320.
12. Pereira, J. S. F., Moraes, D. P., Antes, F. G., Diehl, L. O., Santos, M. F. P., Guimarães, R.
C. L., Fonseca, T. C. O., Dressler, V. L. & Flores, É. M. M. 2010. Determination of metals
and metalloids in light and heavy crude oil by ICP-MS after digestion by microwaveinduced combustion. Microchemical Journal, 96, 4-11.
13. Aguirre, M. A., Kovachev, N., Hidalgo, M. & Canals, A. 2012. Analysis of biodiesel and
oil samples by on-line calibration using a Flow Blurring[registered sign] multinebulizer in
ICP OES without oxygen addition. Journal of Analytical Atomic Spectrometry, 27, 21022110.
14. Flores, É. M. D. M., Barin, J. S., Paniz, J. N. G., Medeiros, J. A. & Knapp, G. 2004.
Microwave-assisted sample combustion: a technique for sample preparation in trace
element determination. Analytical Chemistry, 76, 3525-3529.
15. Moraes, D. P., Mesko, M. F., Mello, P. A., Paniz, J. N. G., Dressler, V. L., Knapp, G. &
Flores, É. M. M. 2007. Application of microwave induced combustion in closed vessels for
carbon black-containing elastomers decomposition. Spectrochimica Acta Part B: Atomic
Spectroscopy, 62, 1065-1071.
16. Duarte, F. A., Pereira, J. S. F., Barin, J. S., Mesko, M. F., Dressler, V. L., Flores, E. M. D.
M. & Knapp, G. 2009. Seafood digestion by microwave-induced combustion for total
arsenic determination by atomic spectrometry techniques with hydride generation. Journal
of Analytical Atomic Spectrometry, 24, 224-227.
17. Xia, L., Y. Wu & Hu. B. 2007. Hollow-fiber liquid-phase microextraction prior to lowtemperature electrothermal vaporization ICP-MS for trace element analysis in
environmental and biological samples. Journal of Mass Spectrometry, 42, 803-810.
18. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
39
Chapter two:
Literature review
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
19. Saint'pierre, T. D., Dias, L. F., Maia, S. M. & Curtius, A. J. 2004. Determination of Cd,
Cu, Fe, Pb and Tl in gasoline as emulsion by electrothermal vaporization inductively
coupled plasma mass spectrometry with analyte addition and isotope dilution calibration
techniques. Spectrochimica Acta Part B: Atomic Spectroscopy, 59, 551-558.
20. Giusti, P., Nuevo Ordonez, Y., Philippe Lienemann, C., Schaumloffel, D., Bouyssiere, B.
& Lobinski, R. 2007. [small mu ]Flow-injection-ICP collision cell MS determination of
molybdenum, nickel and vanadium in petroleum samples using a modified total
consumption micronebulizer. Journal of Analytical Atomic Spectrometry, 22, 88-92.
21. Pohl, P., Vorapalawut, N., Bouyssiere, B., Carrier, H. & Lobinski, R. 2010. Direct multielement analysis of crude oils and gas condensates by double-focusing sector field
inductively coupled plasma mass spectrometry (ICP MS). Journal of Analytical Atomic
Spectrometry, 25, 704-709.
22. dos Santos, E. J., Herrmann, A. B., Chaves, E. S., Vechiatto, W. W. D., Schoemberger, A.
C., Frescura, V. L. A. & Curtius, A. J. 2007. Simultaneous determination of Ca, P, Mg, K
and Na in biodiesel by axial view inductively coupled plasma optical emission
spectrometry with internal standardization after multivariate optimization. Journal of
Analytical Atomic Spectrometry, 22, 1300-1303.
23. Chaves, E. S., De Loos-Vollebregt, M. T. C., Curtius, A. J. & Vanhaecke, F. 2011.
Determination of trace elements in biodiesel and vegetable oil by inductively coupled
plasma optical emission spectrometry following alcohol dilution. Spectrochimica Acta Part
B: Atomic Spectroscopy, 66, 733-739.
24. Khuhawar, M.Y., Mirza A. M. & Jahangir, T.M. 2012. Determination of Metal Ions in
Crude Oils, Crude Oil Emulsions- Composition Stability and Characterization, Prof. Manar
El-Sayed Abdul-Raouf (Ed.), ISBN: 978-953-51-0220-5, InTech.
25. Campos, R. C., Dos Santos, H. R. & Grinberg, P. 2002. Determination of copper, iron,
lead and nickel in gasoline by electrothermal atomic absorption spectrometry using threecomponent solutions. Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 15-28.
26. Becker, E. M., Dessuy, M. B., Boschetti, W., Vale, M. G. R., Ferreira, S. L. C. & Welz, B.
2012. Development of an analytical method for the determination of arsenic in gasoline
samples by hydride generation–graphite furnace atomic absorption spectrometry.
Spectrochimica Acta Part B: Atomic Spectroscopy, 71–72, 102-106.
27. Sanz-Medel, A., Fernandez De La Campa, M. D. R., Gonzalez, E. B. & FernandezSanchez, M. L. 1999. Organised surfactant assemblies in analytical atomic spectrometry.
Spectrochimica Acta Part B: Atomic Spectroscopy, 54, 251-287.
28. Aucelio, R. Q. & Curtius, A. J. 2002. Evaluation of electrothermal atomic absorption
spectrometry for trace determination of Sb, As and Se in gasoline and kerosene using
microemulsion sample introduction and two approaches for chemical modification.
Journal of Analytical Atomic Spectrometry, 17, 242-247.
40
Chapter two:
Literature review
29. Brandão, G., de Campos, R., Luna, A., de Castro, E. & de Jesus, H. 2006. Determination
of arsenic in diesel, gasoline and naphtha by graphite furnace atomic absorption
spectrometry using microemulsion medium for sample stabilization. Analytical and
Bioanalytical Chemistry, 385, 1562-1569.
30. Brandão, G. P., de Campos, R. C., de Castro, E. V. R. & de Jesus, H. C. 2008.
Determination of manganese in diesel, gasoline and naphtha by graphite furnace atomic
absorption spectrometry using microemulsion medium for sample stabilization.
Spectrochimica Acta Part B: Atomic Spectroscopy, 63, 880-884.
31. Pena-Pereira, F., Lavilla, I. & Bendicho, C. 2009. Miniaturized preconcentration methods
based on liquid-liquid extraction and their application in inorganic ultratrace analysis and
speciation: A review. Spectrochimica Acta Part B: Atomic Spectroscopy, 64, 1-15.
32. Bezerra, M. A., Dos Santos, W. N. L., Lemos, V. A., Korn, M. D. G. A. & Ferreira, S. L.
C. 2007. On-line system for preconcentration and determination of metals in vegetables by
Inductively Coupled Plasma Optical Emission Spectrometry. Journal of Hazardous
Materials, 148, 334-339.
33. David, F. & Sandra, P. 2007. Stir bar sorptive extraction for trace analysis. Journal of
Chromatography A, 1152, 54-69.
34. Yang, G., Fen, W., Lei, C., Xiao, W. & Sun, H. 2009. Study on solid phase extraction and
graphite furnace atomic absorption spectrometry for the determination of nickel, silver,
cobalt, copper, cadmium and lead with MCI GEL CHP 20Y as sorbent. Journal of
Hazardous Materials, 162, 44-49.
35. Camel, V. 2003. Solid phase extraction of trace elements. Spectrochimica Acta Part B:
Atomic Spectroscopy, 58, 1177-1233.
36. Risticevic, S., Niri, V., Vuckovic, D. & Pawliszyn, J. 2009. Recent developments in solidphase microextraction. Analytical and Bioanalytical Chemistry, 393, 781-795.
37. Korn, M. D. G. A., Dos Santos, D. S. S., Welz, B., Vale, M. G. R., Teixeira, A. P., Lima,
D. D. C. & Ferreira, S. L. C. 2007. Atomic spectrometric methods for the determination of
metals and metalloids in automotive fuels - A review. Talanta, 73, 1-11.
38. Komjarova, I. & Blust, R. 2006. Comparison of liquid-liquid extraction, solid-phase
extraction and co-precipitation preconcentration methods for the determination of
cadmium, copper, nickel, lead and zinc in seawater. Analytica Chimica Acta, 576, 221-228.
39. Jiang, H., Hu, B., Chen, B. & Xia, L. 2009. Hollow fiber liquid phase microextraction
combined with electrothermal atomic absorption spectrometry for the speciation of arsenic
(III) and arsenic (V) in fresh waters and human hair extracts. Analytica Chimica Acta, 634,
15-21.
40. Ojeda, C. & Rojas, F. 2009. Separation and preconcentration by a cloud point extraction
procedure for determination of metals: an overview. Analytical and Bioanalytical
Chemistry, 394, 759-782.
41. Barri, T., S. Bergstrom, J. Norberg, And J.A. Jonsson 2004. Miniaturized and Automated
Sample Pretreatment for Determination of PCBs in Environmental Aqueous Samples
41
Chapter two:
Literature review
Using an On-Line Microporous Membrane Liquid-Liquid Extraction-Gas Chromatography
System. Analytica Chemistry, 76, 1928-1934.
42. Bianchin, J. N., Martendal, E., Mior, R., Alves, V. N., Araújo, C. S. T., Coelho, N. M. M.
& Carasek, E. 2009. Development of a flow system for the determination of cadmium in
fuel alcohol using vermicompost as biosorbent and flame atomic absorption spectrometry.
Talanta, 78, 333-336.
43. Liang, P., Qin, Y., Hu, B., Peng, T. & Jiang, Z. 2001. Nanometer-size titanium dioxide
microcolumn on-line preconcentration of trace metals and their determination by
inductively coupled plasma atomic emission spectrometry in water. Analytica Chimica
Acta, 440, 207-213.
44. Bulut, V. N., Gundogdu, A., Duran, C., Senturk, H. B., Soylak, M., Elci, L. & Tufekci, M.
2007. A multi-element solid-phase extraction method for trace metals determination in
environmental samples on Amberlite XAD-2000. Journal of Hazardous Materials, 146,
155-163.
45. Santos, D. S. S., M.G. A. Korn, M. A. B. Guida, G. L. Dos Santos, V. A. Lemos And L. S.
G. Teixeira 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by Sequential
Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase Extraction.
Journal of Brazzilian Chemical Society, 22, 552-557.
46. Alves, V., Mosquetta, R., Carasek, E. & Coelho, N. 2012. Determination of Zn(II) in
alcohol fuel by flame atomic absorption spectrometry after on-line preconcentration using
a solid phase extraction system. Journal of Analytical Chemistry, 67, 448-454.
47. Alves, V. N., Mosquetta, R., Coelho, N. M. M., Bianchin, J. N., Di Pietro Roux, K. C.,
Martendal, E. & Carasek, E. 2010. Determination of cadmium in alcohol fuel using
Moringa oleifera seeds as a biosorbent in an on-line system coupled to FAAS. Talanta, 80,
1133-1138
48. Roldan, P. S., Alcântara, I. L., Padilha, C. C. F. & Padilha, P. M. 2005. Determination of
copper, iron, nickel and zinc in gasoline by FAAS after sorption and preconcentration on
silica modified with 2-aminotiazole groups. Fuel, 84, 305-309.
49. Roldan, P. S., Alcântara, I. L., Rocha, J. C., Padilha, C. C. F. & Padilha, P. M. 2004.
Determination of Copper, Iron, Nickel and Zinc in fuel kerosene by FAAS after adsorption
and pre-concentration on 2-aminothiazole-modified silica gel. Ecl. Quím., São Paulo, 29,
33-40.
50. Roldan, P. S., Alcântara, I. L., Castro, G. R., Rocha, J. C., Padilha, C. C. F. & Padilha, P.
M. 2003. Determination of Cu, Ni, and Zn in fuel ethanol by FAAS after enrichment in
column packed with 2-aminothiazole-modified silica gel. Analytical and Bioanalytical
Chemistry, 375, 574-577.
51. Teixeira, L. S. G., Bezerra, M. D. A., Lemos, V. A., Santos, H. C. D., De Jesus, D. S. &
Costa, A. C. S. 2005. Determination of Copper, Iron, Nickel, and Zinc in Ethanol Fuel by
Flame Atomic Absorption Spectrometry Using On-Line Preconcentration System.
Separation Science and Technology, 40, 2555 - 2565.
42
Chapter two:
Literature review
52. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A. M.
& Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in automotive
gasoline by X-ray fluorescence after pre-concentration on cellulose paper. Talanta, 72,
1073-1076.
53. Teixeira, L. S. G., Santos, E. S. & Nunes, L. S. 2012. Determination of copper, iron,
nickel and zinc in ethanol fuel by energy dispersive X-ray fluorescence after preconcentration on chromatography paper. Analytica Chimica Acta, 722, 29-33.
54. Vieira, E. G., Soares, I. V., Dias Filho, N. L., Da Silva, N. C., Garcia, E. F., Bastos, A. C.,
Perujo, S. D., Ferreira, T. T., Rosa, A. H. & Fraceto, L. F. 2013. Preconcentration and
determination of metal ions from fuel ethanol with a new 2,2′-dipyridylamine bonded
silica. Journal of Colloid and Interface Science, 391, 116-124.
55. Pradoa, A. G. S., Pescaraa, I. C., Evangelistaa, S. M., Holandaa, M. S., Andradeb, R. D.,
Suareza, P. A. Z. & Zarac, L. F. 2011. Adsorption and preconcentration of divalent metal
ions in fossil fuels and biofuels: Gasoline, diesel, biodiesel, diesel-like and ethanol by
using chitosan microspheres and thermodynamic approach. Talanta, 84, 759–765.
56. Cui, C., He, M. & Hu, B. 2011. Membrane solid phase microextraction with alumina
hollow fiber on line coupled with ICP OES for the determination of trace copper,
manganese and nickel in environmental water samples. Journal of Hazardous Materials,
187, 379-385.
57. Mester, Z., Sturgeon, R. E. & Lam, J. W. 2000. Sampling and determination of metal
hydrides by solid phase microextraction thermal desorption inductively coupled plasma
mass spectrometry. Journal of Analytical Atomic Spectrometry, 15, 1461-1465.
58. Bravo-Sánchez, L. R., Ruiz Encinar, J., Fidalgo Martínez, J. I. & Sanz-Medel, A. 2004.
Mercury speciation analysis in sea water by solid phase microextraction-gas
chromatography-inductively coupled plasma mass spectrometry using ethyl and propyl
derivatization. Matrix effects evaluation. Spectrochimica Acta Part B: Atomic
Spectroscopy, 59, 59-66.
59. Es’haghi, Z., Khalili, M., Khazaeifar, A. & Rounaghi, G. H. 2011. Simultaneous
extraction and determination of lead, cadmium and copper in rice samples by a new preconcentration technique: Hollow fiber solid phase microextraction combined with
differential pulse anodic stripping voltammetry. Electrochimica Acta, 56, 3139-3146.
60. Huang, C. & Hu, B. 2011. Synthesis and characterization of titania hollow fiber and its
application to the microextraction of trace metals. Analyst, 136, 1425-1432.
61. Abulhassani, J., Manzoori, J. L. & Amjadi, M. 2010. Hollow fiber based-liquid phase
microextraction using ionic liquid solvent for preconcentration of lead and nickel from
environmental and biological samples prior to determination by electrothermal atomic
absorption spectrometry. Journal of Hazardous Materials, 176, 481-486.
62. Es’haghi, Z. & Azmoodeh, R. 2010. Hollow fiber supported liquid membrane
microextraction of Cu2+ followed by flame atomic absorption spectroscopy determination.
Arabian Journal of Chemistry, 3, 21-26.
43
Chapter two:
Literature review
63. Rasmussen, K. E. & Pedersen-Bjergaard, S. 2004. Developments in hollow fibre-based,
liquid-phase microextraction. TrAC Trends in Analytical Chemistry, 23, 1-10.
64. Dadfarnia, S. & Haji Shabani, A. M. 2010. Recent development in liquid phase
microextraction for determination of trace level concentration of metals--A review.
Analytica Chimica Acta, 658, 107-119.
65. Psillakis, E. & Kalogerakis, N. 2003. Developments in liquid-phase microextraction.
TrAC Trends in Analytical Chemistry, 22, 565-574
66. Ho, T. S., Halvorsen, T. G., Pedersen-Bjergaard, S. & Rasmussen, K. E. 2003. Liquidphase microextraction of hydrophilic drugs by carrier-mediated transport. Journal of
Chromatography A, 998, 61-72.
67. Aucélio, R. Q., De Souza, R. M., De Campos, R. C., Miekeley, N. & Da Silveira, C. L. P.
2007. The determination of trace metals in lubricating oils by atomic spectrometry.
Spectrochimica Acta Part B: Atomic Spectroscopy, 62, 952-961.
68. Martiniano, L. C., Abrantes, V. R., Neto, S. Y., Marques, E. P., Fonseca, T. C. O., Paim,
L. L., Souza, A. G., Stradiotto, N. R., Aucélio, R. Q., Cavalcante, G. H. R. & Marques, A.
L. B. Direct simultaneous determination of Pb(II) and Cu(II) in biodiesel by anodic
stripping voltammetry at a mercury-film electrode using microemulsions. Fuel. Update Vol
and page?
69. Broekaert, J. A. C. & Evans, E. H. 2008. Atomic Spectroscopy. In: Gunzler, H. &
Williams., A. (eds.) Handbook of Analytical Techniques. Wiley-VCH Verlag GmbH.
70. Pignalosa, G. & Knochen, M. 2001. Determination of wear metals in lubricating oils using
flow injection AAS, Atomic Spectroscopy, 22, 150–157.
71. Reis, B. F., Knochen, M., Pignalosa, G., Cabrera, N. & Giglio, J. 2004. A multicommuted
flow system for the determination of copper, chromium, iron and lead in lubricating oils
with detection by flame AAS. Talanta, 64, 1220-1225.
72. de Jesus, A., Zmozinski, A. V., Damin, I. C. F., Silva, M. M. & Vale, M. G. R. 2012.
Determination of arsenic and cadmium in crude oil by direct sampling graphite furnace
atomic absorption spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 71–72,
86-91.
73. Sousa, J. K. C., Dantas, A. N. D. S., Marques, A. L. B. & Lopes, G. S. 2008. Experimental
design applied to the development of a copper direct determination method in gasoline
samples by graphite furnace atomic absorption spectrometry. Fuel Processing Technology,
89, 1180-1185.
74. Becker, E., Rampazzo, R. T., Dessuy, M. B., Vale, M. G. R., Da Silva, M. M., Welz, B. &
Katskov, D. A. 2011. Direct determination of arsenic in petroleum derivatives by graphite
furnace atomic absorption spectrometry: A comparison between filter and platform
atomizers. Spectrochimica Acta Part B: Atomic Spectroscopy, 66, 345-351.
75. Reboucas, M. V., Domingos, D., Santos, A. S. O. & Sampaio, L. 2010. Determination of
trace metals in naphtha by graphite furnace atomic absorption spectrometry: Comparison
44
Chapter two:
Literature review
between direct injection and microemulsion pretreatment procedures. Fuel Processing
Technology, 91, 1702-1709.
76. Brum, D. M., Lima, C. F., Robaina, N. F., Fonseca, T. C. O. & Cassella, R. J. 2011.
Multiple response optimization for Cu, Fe and Pb determination in naphtha by graphite
furnace atomic absorption spectrometry with sample injection as detergent emulsion.
Spectrochimica Acta Part B: Atomic Spectroscopy, 66, 338-344.
77. Cunha, F. A. S., Sousa, R. A., Harding, D. P., Cadore, S., Almeida, L. F. & Araújo, M. C.
U. 2012. Automatic microemulsion preparation for metals determination in fuel samples
using a flow-batch analyzer and graphite furnace atomic absorption spectrometry.
Analytica Chimica Acta, 727, 34-40.
78. Trindade, J. M., Marques, A. L., Lopes, G. S., Marques, E. P. & Zhang, J. 2006. Arsenic
determination in gasoline by hydride generation atomic absorption spectroscopy combined
with a factorial experimental design approach. Fuel, 85, 2155-2161.
79. Donati, G. L., Amais, R. S., Schiavo, D. & Nobrega, J. A. 2013. Determination of Cr, Ni,
Pb and V in gasoline and ethanol fuel by microwave plasma optical emission spectrometry.
Journal of Analytical Atomic Spectrometry, 28, 755-759.
80. Sutton, K. L. & Caruso, J. A. 1999. Liquid chromatography-inductively coupled plasma
mass spectrometry. Journal of Chromatography A, 856, 243-258.
81. Thomas, R. 2004. Practical Guide to ICP-MS, New York, U.S. A., Marcel Dekker, Inc.
82. Hou, X. & Jones, B. T. 2000. Inductively Coupled Plasma/Optical Emission
Spectrometry. In: MEYERS, R. A. (ed.) Encyclopedia of Analytical Chemistry.
Chichester: John Wiley & Sons Ltd.
83. Environmental Measurement I: Gas-Solution Analytical Center. Inductively coupled plasma
spectrometer
(ICP):
A
typical
plasma
source.
Available
at
http://em1.stanford.edu/Schedule/ICP/abouticp.htm Accessed 12 September 2011
84. Skoog, D. A. 2004. Fundamentals of analytical chemistry, Thomson-Brooks/Cole.
85. http://www.aandb.com.tw/Page0001/icp_oes_01_optima_7x00_dv.html Accessed
86. Wang, T. 2005. Inductively coupled plasma optical emission spectrometry. In: Dekker, M.
(ed.) Analytical Instrumentation Handbook. Rahway, NJ, USA: Merck Research
Laboratories.
87. Boss, C.B., Fredeen, K.J. Concept, Instrumentation and Techniques in Inductively
Coupled Plasma Optical Emission Spectrometry, 2nd edition, Perkin-Elmer, Norwalk, CT,
1997.
88. Morishige, Y. & Kimura, A. 2008. Ionization interference in inductively coupled plasmaoptical emission spectroscopy, Sei Technical Review, 66, 106-11
89. Agilent Technologies, Inc. 2012. CCD and CID solid-state detectors: Technical overview.
http://www.chem.agilent.com/Library/technicaloverviews/Public/59910881EN_TechOvie
w_700_CCDvsCID.pdf. Accessed 24 May 2011.
45
Chapter two:
Literature review
90. Galley, P. J. & Hieftje, G. M. 1994. Easily ionizable element (EIE) interferences in
inductively coupled plasma atomic emission spectrometry: II. Minimization of EIE effects
by choice of observation volume. Spectrochimica Acta Part B: Atomic Spectroscopy, 49,
703-724.
91. Tyler, G. ICP OES, ICP-MS and AAS Techniques Compared. Technical Note 05. Jobin
Yvon
publication,
(2005).
Available
at
http://www.jobinyvon.com/usadivisions/Emission/applications/TN05.pdf Accessed 12
September 2011.
92. Miles, D.L., 1987. The application of inductively coupled plasmas to the analysis of
natural waters and acidic deposition. In: Rowland, A.P. (Ed.), Chemical Analysis in
Environmental Research. Institute of Freshwater Ecology, Symposium No. 18. NERC, UK,
pp. 40-49.
93. Potts, P. J. (1987). Inductively coupled plasma-mass spectrometry. In: Handbook of
Silicate Rock Analysis. Blackie, Glasgow, pp. 575-586.
94.Houk, R. S., Fassel, V. A., Flesch, G. D., Svec, H. J., Gray, A. L. & Taylor, C. E. 1980.
Inductively coupled argon plasma as an ion source for mass spectrometric determination of
trace elements. Analytical Chemistry, 52, 2283-2289.
95. Ammann, A. A. 2007. Inductively coupled plasma mass spectrometry (ICP MS): a
versatile tool. Journal of Mass Spectrometry, 42, 419-427.
96. Linge, K. L. & Jarvis, K. E. 2009. Quadrupole ICP-MS: Introduction to Instrumentation,
Measurement Techniques and Analytical Capabilities. Geostandards and Geoanalytical
Research, 33, 445-467.
97. Jarvis, K. E., Gray, A. L. & Houk, R. S. 1992. Handbook of Inductively Coupled Plasma
Mass Spectrometry. Chapman and Hall: New York.
98. Brouwers, E. E. M., Tibben, M., Rosing, H., Schellens, J. H. M. & Beijnen, J. H. 2008.
The application of inductvely coupled plasma mass spectrometry in clinical
pharmacological oncology research. Mass Spectrometry Reviews, 27, 67-100.
99. Broekaert, J. A. C. 2005. Analytical atomic spectrometry with flames and plasmas,
Weinheim, Wiley-VCH Verlag GmbH &Co. KGaA.
100. Colon, M., Hidalgo, M. & Iglesias, M. 2011. Arsenic determination by ICP-QMS with
octopole collision/reaction cell. Overcome of matrix effects under vented and pressurized
cell conditions. Talanta, 85, 1941-1947.
101. Saint'pierre, T. D., Frescura, V. L. A. & Curtius, A. J. 2006. The development of a
method for the determination of trace elements in fuel alcohol by ETV-ICP-MS using
isotope dilution calibration. Talanta, 68, 957-962.
102. Dantas, A. N. S., Costa, R. S., Gouveia, S. T. & Lopes, G. S. 2010. Development of a
microwave-assisted digestion method using ICP OES to measure metals in gum deposits of
internal combustion engines. Fuel Processing Technology, 91, 1422-1427.
103. Lienemann, C., P., Dreyfus, S., Pecheyran, C. & Donard, O., F.X. 2007. Trace Metal
Analysis in Petroleum Products: Sample Introduction Evaluation in ICP OES and
46
Chapter two:
Literature review
Comparison with an ICP-MS Approach. Oil & Gas Science and Technology - Rev. IFP,
62, 69-77.
104. Duyck, C., Miekeley, N., Porto Da Silveira, C. L. & Szatmari, P. 2002. Trace element
determination in crude oil and its fractions by inductively coupled plasma mass
spectrometry using ultrasonic nebulization of toluene solutions. Spectrochimica Acta Part
B: Atomic Spectroscopy, 57, 1979-1990.
105. Kowalewska, Z., Ruszczyńska, A. & Bulska, E. 2005. Cu determination in crude oil
distillation products by atomic absorption and inductively coupled plasma mass
spectrometry after analyte transfer to aqueous solution. Spectrochimica Acta Part B:
Atomic Spectroscopy, 60, 351-359.
106. Chaves, E. S., F. G. Lepri, J. S. A. Silva, D. P. C. De Quadros, T. D. Saint’pierre And A.
J. Curtius 2008. Determination of Co, Cu, Fe, Mn, Ni and V in diesel and biodiesel
samples by ETV-ICP-MS. Journal of Environmental Monitoring, 10, 1211-1216.
107. Caumette, G., Lienemann, C.-P., Merdrignac, I., Paucot, H., Bouyssiere, B. & Lobinski,
R. 2009. Sensitivity improvement in ICP MS analysis of fuels and light petroleum matrices
using a microflow nebulizer and heated spray chamber sample introduction. Talanta, 80,
1039-1043.
108. de Souza, R. M., Saraceno, A. L., Duyck, C., Da Silveira, C. L. P. & Aucélio, R. Q.
2007. Determination of Fe, Ni and V in asphaltene by ICP OES after extraction into
aqueous solutions using sonication or vortex agitation. Microchemical Journal, 87, 99-103.
109. de Souza, R. M., Meliande, A. L. S., Da Silveira, C. L. P. & Aucélio, R. Q. 2006.
Determination of Mo, Zn, Cd, Ti, Ni, V, Fe, Mn, Cr and Co in crude oil using inductively
coupled plasma optical emission spectrometry and sample introduction as detergentless
microemulsions. Microchemical Journal, 82, 137-141.
110. Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S. & Escaleira, L. A. 2008.
Response surface methodology (RSM) as a tool for optimization in analytical chemistry.
Talanta, 76, 965-977.
111. Araujo, P. W. & Brereton, R. G. 1996. Experimental design II. Optimization. TrAC
Trends in Analytical Chemistry, 15, 63-70.
112. Lobo, F. A., Goveia, D., Oliveira, A. P. D., Pereira-Filho, E. R., Fraceto, L. F., Filho, N.
L. D. & Rosa, A. H. 2009. Comparison of the univariate and multivariate methods in the
optimization of experimental conditions for determining Cu, Pb, Ni and Cd in biodiesel by
GFAAS. Fuel, 88, 1907-1914.
113. Tarley, C. R. T., Silveira, G., Dos Santos, W. N. L., Matos, G. D., Da Silva, E. G. P.,
Bezerra, M. A., Miró, M. & Ferreira, S. L. C. 2009. Chemometric tools in electroanalytical
chemistry: Methods for optimization based on factorial design and response surface
methodology. Microchemical Journal, 92, 58-67.
114. Ferreira, S. L. C., Dos Santos, H. C., Fernandes, M. S. & De Carvalho, M. S. 2002.
Application of Doehlert matrix and factorial designs in optimization of experimental
variables associated with preconcentration and determination of molybdenum in sea-water
47
Chapter two:
Literature review
by inductively coupled plasma optical emission spectrometry. Journal of Analytical Atomic
Spectrometry, 17, 115-120.
115. Ferreira, S. L. C., Bruns, R. E., Da Silva, E. G. P., Dos Santos, W. N. L., Quintella, C.
M., David, J. M., De Andrade, J. B., Breitkreitz, M. C., Jardim, I. C. S. F. & Neto, B. B.
2007. Statistical designs and response surface techniques for the optimization of
chromatographic systems. Journal of Chromatography A, 1158, 2-14.
116. Caldas, L. F. S., De Paula, C. E. R., Brum, D. M. & Cassella, R. J. 2013. Application of
a four-variables Doehlert design for the multivariate optimization of copper determination
in petroleum-derived insulating oils by GFAAS employing the dilute-and-shot approach.
Fuel, 105, 503-511.
117. Caldas, L. F. S., Francisco, B. B. A., Netto, A. D. P. & Cassella, R. J. 2011. Multivariate
optimization of a spectrophotometric method for copper determination in Brazilian sugarcane spirits using the Doehlert design. Microchemical Journal, 99, 118-124.
118. Cassella, R. J., Brum, D. M., De Paula, C. E. R. & Lima, C. F. 2010. Extraction induced
by emulsion breaking: a novel strategy for the trace metals determination in diesel oil
samples by electrothermal atomic absorption spectrometry. Journal of Analytical Atomic
Spectrometry, 25, 1704-1711.
119. Sant’Ana, F. W., Santelli, R. E., Cassella, A. R. & Cassella, R. J. 2007. Optimization of
an open-focused microwave oven digestion procedure for determination of metals in diesel
oil by inductively coupled plasma optical emission spectrometry. Journal of Hazardous
Materials, 149, 67-74.
48
CHAPTER THREE:
GENERAL METHODOLOGIES
In this chapter, an overview of the general aspects of sample preparation methods used
throughout this study is given. Instrumentations as well as some instrumental parameters,
employed during quantification of trace elements are discussed. However, the detailed
procedures of sample preparation techniques are presented in the respective chapters. It
should be noted that, only the detailed description of procedures that are not fully discussed
in respective chapters, are given in this chapter. Therefore, Chapter 3 discusses details of
various aspects of experimental procedures not covered adequately in the results and
discussion chapters 4-13.
3.1 OVERVIEW OF EXPERIMENTAL DESIGN
Fig. 3.1 presents the flow chart that summarises the research design performed in this
study. The experimental design shows the analytical methods used for the separation and
preconcentration of metal ions in organic samples and their determination using ICP OES and
ICP-MS. In addition, the design shows the different stages carried out.
3.2 INSTRUMENTATION
A number of analytical techniques were used for characterization of the synthesized
adsorbents, sample preparation and detection of metal ions. Inductively coupled plasma
optical emission spectrometer (ICP OES, SPECTRO Analytical Instruments, GmbH,
Germany), electrothermal atomic absorption spectrometer (ETAAS, Perkin-Elmer, USA) and
inductively coupled plasma mass spectrometer (ICP-MS, Perkin-Elmer SCIEX Instruments,
Concord, Canada) were used for quantification of metal ions in the samples. The
morphological characteristics of adsorbents were determined by scanning electron
microscopy (SEM, JSM-6360LVSEM, JEOL Co., Japan). Characterization of the attached
functional groups on the adsorbent was performed using Fourier transform infra red
spectroscopy (FTIR, PerkinElmer, USA) and solid state
13
C NMR (Bruker, GmbH,
Germany). The specific surface area value was determined from adsorption isotherms by
using the multipoint method using Surface Area and Porosity Analyzer (ASAP2020 V3.00H,
Micromeritics Instrument Corporation, Norcross, USA). X-ray powder diffraction (XRD)
measurements were carried out with a Philips X-ray generator model PW 3710/31 a
49
Chapter 3:
General methodologies
diffractometer with automatic sample changer model PW 1775 (scintillation counter, Cutarget tube and Ni-filter at 40 kV and 30 mA. Solid phase extraction was carried out in a
VacMaster-24 sample SPE station (Supelco, PA, USA). Microwave assisted digestion was
carried out in an Ethos D (Milestone, Sorisole, Italy) with maximum pressure 1450 psi and
maximum temperature 300°C.
Samples in organic phase
Microwave- assited
digession (MAD)
Solid phase extraction using
different adsorbents
Solid phase microextraction
(SPME) using two categories
Hollow fiber-SPME
Hollow fiber liquid phase
microextraction (HF-LPME)
Membrane -SPME
Determination of metal ions in aqueous phase using
ICP-OES or ICP-MS
Fig. 3.1. Experimental design flow chart showing the summary of sample preparation
methods used for separation and preconcentration of metal ions in organic matrices and
detection techniques.
50
Chapter 3:
General methodologies
3.3 REAGENTS AND MATERIALS
All reagents were of analytical grade unless otherwise stated and double distilled
deionized waster (Millipore, Bedford, MA, USA) was used throughout the experiments.
Model solutions were prepared using ethanol and synthetic gasoline to maintain the organic
matrix. The model solutions were prepared in plastic volumetric flasks and working
solutions, as per the experimental requirements, were freshly prepared from the stock solution
for each experimental run. Commercial alcohol samples were purchased from local chemical
suppliers. Fuel samples were collected from different local filling stations and stored for short
period (maximum of 8 hours) in polypropylene bottle. It should be noted that before
sampling, the polypropylene bottles were cleaned properly and soaked in 1% nitric solution
for 24 hour to reduce the possibility of contamination. All ion exchange resins (Dowex 50Wx8, Dowex MAC-3, Chelex-100 and Dowex 1-x8) were purchased from one chemical
supplier. Adsorbents used were prepared by mixing appropriate amounts starting materials.
The pH of the sample solutions were adjusted with glacial acetic acid and ammonia solutions.
Solution of nitric acid was used for the elution of the analytes. The polypropylene hollow
fiber membranes obtained from Membrana (Wuppertal, Germany) were used for preparation
of hollow fiber solid phase microextraction and hollow fiber liquid phase microextraction.
Concentrated nitric acid and hydrogen peroxide were used in the digestion of fuel samples.
3.4 SEPARATION AND PRECONCENTRATION TECHNIQUES
3.4.1 Solid Phase Extraction
Solid phase extraction was carried out using different solid phase materials namely
commercially available ion exchange resins, metal oxide adsorbent and functionalised
cellulose nanofibers. Dowex 50W-x8, Chelex-100 and Dowex MAC-3 resins were used as
solid phase materials for separation and preconcentration of metal ions in alcohol sample.
Dowex 1-x8 resin was used for preconcentration of metal ions in gasoline samples. A SPE
based on the dual resin column was also used for enrichment of metal and metalloids in
gasoline samples. Nanometer-sized alumina and functionalised cellulose nanofibers, prepared
according to Rogojan et al.1 and Musyoka et al.,2 respectively, were used as packing
materials for preconcentration of metal ions in gasoline samples. The effect of some
experimental parameters such as sample pH, eluent concentration and flow rates were
51
Chapter 3:
General methodologies
optimised used either univariate or multivariate procedures. Detergentless emulsion and
alcohol dilution were used to prepare fuel samples before they are percolated to the SPE
packed column. The detailed procedure describing the preparation of functionalized cellulose
nanofibers is given in Section 3.4.1.1.
3.4.1.1 Electrospinning and functionalization of cellulose nanofibers with oxolane-2,5dione
Electrospinning of cellulose acetate was performed according to the procedure described
by Ma et al.3 and Musyoka et al.2 Briefly, the polymer solution was prepared by stirring
cellulose
acetate,
16%
(w/v)
in
acetone/N,N-dimethylacetamide
solvent
mixture,
(Me)2CO:DMAc, (2:1) for 14 h to ensure solution homogeneity. The electrospinning set-up
consisted of a grounded aluminium foil electrode, high voltage direct current power supply
glass syringe with a stainless steel needle and programmable syringe pump (NE-1000 single
syringe pump, New Era pump systems). The glass syringe was filled with cellulose acetate
solution. The grounded aluminium foil collector was placed 14 cm from the tip of a stainless
steel needle. The programmable syringe pump was used to pump the solution through a 20
gauge (bore diameter = 0.45 mm) stainless steel needle at a flow rate of 0.60 mL h−1. The
needle was connected to a high voltage direct current source with a maximum output of 25
kV. The electrospinning voltage for this work was fixed at 15 kV unless otherwise stated.
The electrospun fiber mat was heated in an oven at 200 ◦C for 1 h and carefully peeled off
from the aluminium foil. Cellulose acetate nanofibers were deacetylated by soaking the mat
in 0.3 M NaOH solution for 8 h followed by washing with double distilled water to obtain
neutral pH.
Prior to functionalization of the cellulose nanofibers, electrospun fiber mat was sonicated
twice with ethanol followed by toluene to remove water. Five grams of cellulose nanofibers
were added to the mixture of 90 mL of toluene and 12.5 mL pyridine then heated at 60 °C for
1 h. Fifteen grams of oxolane-2,5-dione was then added to the mixture and heated overnight
at 90 °C. The mixture was filtered while hot and soxhlet extracted using ethanol to remove
unreacted oxolane-2,5-dione, then dried under vacuum at 40 °C, yielding a white fibrous mat.
The reaction scheme as proposed by Musyoka et al.2 is shown in Fig. 3.2.
52
Chapter 3:
General methodologies
O
HO
O
OH
*
OHO
i) C7H8/ C5H5N; 60 0C; 1 h
O
O
0
ii) C4H4O3; 90 C; 16 h
OH
*
O
OHO
n
O
O
OH
n
Fig. 3.2. Reaction scheme for the functionalization of cellulose to cellulose-g-oxolane-2,5dione2
3.4.2 Solid Phase Microextraction (SPME)
Solid phase microextraction was carried out using hollow fiber-SPME and membrane-SPME.
The HF-SPME was prepared by hollow fiber-supported sol-gel combined with cation
exchange resin. The procedure for the preparation of sol-gel was according to Es’haghi et al.4
For membrane based solid phase microextraction (MSPME), alumina-titania hollow fiber
was used for extraction and preconcentration of metal ions in fuels. The synthesis of aluminatitania sol was prepared according to Jung et al.5 The structural characteristics of hollow fiber
membrane were evaluated by XRD, SEM and BET. The XRD patterns of alumina and titania
powder, prepared according to Rogojan et al.1 and Li et al.,6 were used for the assignment of
their respective characteristics peak. The detailed procedures for the preparation of Al2O3 and
TiO2 are described in Section 3.4.2.1.
The optimization of the HF-LPME and MSPME preconcentration system was carried
out using two-level (24) full factorial design with a central point and central composite design
(CCD). Four variables for both methods were regarded as important factors. For HF-SPME,
sample solution pH, acceptor phase amount, extraction time and eluent concentration were
regarded as factors. For MSPME on the other hand, the important factors were sample pH,
eluent concentration, extraction time and eluent volume.
3.4.2.1 Preparation of nanometer-sized alumina and titania
The nanometer alumna and titania powders were prepared by sol-gel method. The latter is
based on the phase transformation of a sol obtained from metallic alkoxides or
53
Chapter 3:
General methodologies
organometallic precursors. The sol contains particles in suspension and it is polymerized at low
temperature, in order to form a wet gel. The resulting gel is dried in order to remove the solvent
(e.g. alcohol).1 To obtain desired crystalline structure of the metallic oxide, the dried gel is treated
by heating it at a specific temperature. A Sol-gel method was chosen because of it many
advantages. These include versatility and the possibility to obtain high purity materials,
allowance of the synthesis of special materials, energy savings by using low processing
temperature, higher surface areas, well-defined pore size distribution and superior
homogeneity.1,7
Nanometer-sized alumina was prepared as follows; a mass (2.66 g) of AlCl3 was
dissolved in 25 ml absolute ethanol followed by drop wise addition of 28% ammonium
solution. The addition of the latter was done in order to for a sol gel to form. The resulting sol
gel was left to maturate for 30 hours at room temperature and then dried for 24 hours at
100°C. Finally, the gel was calcined by heating in a furnace at a rate of 20°C min -1 to 1000
°C and holding it for three hours.
The synthesis of nanometer-sized titania was carried as follows; 10 mL tetrabutyl
butoxide was dissolved in a mixture solvent containing ethanol (45 mL) and acetic acid (3
mL) under vigorous stirring at room temperature. double distilled deionized waster ((1.25)
was added dropwise to the above solution under vigorous stirring for about 10 min. The
resulting sol-gel was dried in the oven at 80°C overnight. Finally, the gel was calcined by
heating in a furnace at a rate of 20°C min-1 to 1000 °C and holding it for three hours.
3.4.3 Hollow Fiber-Liquid Phase Microextraction
Extraction and preconcentration of metal ions in organic matrices using HF-LPME is
difficult. This is because the organic phase in the lumen of the hollow fiber membrane is
miscible with donor phase (sample). Therefore, sample pretreatment that will first convert
fuel sample to aqueous phase before it is subjected to HF-LPME system is required. For this
reason, diesel and gasoline sample samples were first digested using microwave-assisted
digestion method. The digested samples were the subject to HF-LPME preconcentration
system. The latter, which is a room temperature ionic liquid combined with ammonium
pyrrolidinecarbodithioate as a chelating agent, was used for extraction and preconcentration
of metal ions in digested fuel samples.
54
Chapter 3:
General methodologies
The preparation of the HF-LPME procedure was adopted from Ghasemi et al.8 The
extraction procedure was carried out according to Xia et al.9 and Ghasemi et al.8 Full
factorial and central composite designs were used for screening and optimization of the
effective factors that influence preconcentration and stripping of the metal ions. The factors
included sample pH, concentration of the chelating agent, extraction time and stripping
solution concentration.
3.4 ACID DIGESTION METHODS
Acid digestions of fuel samples were carried out using two methods, namely acid
digestion in a hot plate using Teflon beakers and microwave-assisted digestion. The modified
procedure for acid digestion on a hot plate was adopted from Amorim et al.10 Microwaveassisted digestion on the other hand was carried out according to Kowalewska et al.11 The
details of these methods are given in the respective chapters.
3.5 DETERMINATION OF METAL IONS
Determination of metal ions in the samples after pretreatment was carried by using either ICP
OES or ICP-MS or both techniques. The instrumental operational settings and parameters are
presented in the respective chapters. Electrothermal AAS was used as comparative method
for the determination of metal ion in alcohols. The analytical procedure for ETAAS analysis
reported by Anselmi et al.,12 Reboucas et al.13 and de Oliveira et al.14 were modified in order
to suit the sample matrix. Table 1 presents electrothermal AAS temperature programs for
determination of metal ions. It should be noted that the ramp, hold time and gas flow rate
conditions for pyrolysis and atomization steps were the same for all the metal ions.
55
Chapter 3:
General methodologies
Table 3.1. Electrothermal AAS temperature programs for determination of metal ions
Step
T (°C)
Ramp (°C s-1)
Hold time (s)
Drying 1
Drying 2
Pyrolysis
Cd
Cr
Cu
Fe
Mn
Pb
Zn
Atomization
Cd
Cr
Cu
Fe
Mn
Pb
Zn
Cleaning
90
200
1
1
5
5
10
5
Gas flow rate
(mL min-1)
300
300
300
0
6
0
1
3
300
300
1600
1300
800
1000
600
100
1100
2400
2300
2000
2300
1500
2300
2500
3.6 REFERENCES
1. Rogojan, R., Andronescu, E., Ghitulica, C. & Vasile, B.S. Synthesis and characterisation
of Alumina nano-powder obtained by sol-gel method, U.P.B. Sci. Bull. Series B 73 (2011)
65-76.
2. Musyoka, S., Ngila, C., Moodley, B., Kindness, A., Petrik, L. & Greyling, C. 2011.
Oxolane-2,5-dione modified electrospun cellulose nanofibers for heavy metals adsorption.
Journal of Hazardous Materials, 192, 922-927.
3. Ma, Z., Kotaki, M. & Ramakrishna, S. 2005. Electrospun cellulose nanofiber as affinity
membrane. Journal of Membrane Science, 265, 115-123.
4. Es’haghi, Z., Khalili, M., Khazaeifar, A. & Rounaghi, G. H. 2011. Simultaneous extraction
and determination of lead, cadmium and copper in rice samples by a new preconcentration technique: Hollow fiber solid phase microextraction combined with
differential pulse anodic stripping voltammetry. Electrochimica Acta, 56, 3139-3146.
5. Jung, Y.-S., Kim, D.-W., Kim, Y.-S., Park, E.-K. & Baeck, S.-H. 2008. Synthesis of
alumina–titania solid solution by sol–gel method. Journal of Physics and Chemistry of
Solids, 69, 1464-1467.
56
Chapter 3:
General methodologies
6. Li, J., Qi, H.-Y. & Shi, Y.-P. 2009. Applications of titania and zirconia hollow fibers in
sorptive microextraction of N,N-dimethylacetamide from water sample. Analytica
Chimica Acta, 651, 182–187.
7. Akbarnezhad, S., Mousavi, S. M., & Sarhaddi, R. 2010. Sol-gel synthesis of alumina-titania
ceramic membrane: Preparation and characterization. Indian Journal of Science and
Technology, 3, 1048-1051
8. Ghasemi, E., Najafi, N. M., Raofie, F. & Ghassempour, A. 2010. Simultaneous speciation
and preconcentration of ultra traces of inorganic tellurium and selenium in environmental
samples by hollow fiber liquid phase microextraction prior to electrothermal atomic
absorption spectroscopy determination. Journal of Hazardous Materials, 181, 491-496.
9. Xia, L., Y. Wu And B. Hu 2007. Hollow-fiber liquid-phase microextraction prior to lowtemperature electrothermal vaporization ICP-MS for trace element analysis in
environmental and biological samples. Journal of Mass Spectrometry, 42, 803-810.
10. Amorim, F. A. C., Lima, D. C., Amaro, J. A. A., Valea, M. G. R. & Ferreira, S. L. C.
2007. Methods for vanadium determination in fuel oil by GFAAS with
microemulsification and acid digestion sampling. Journal of Brazzilian Chemical Society,
18, 1566-1570.
11. Kowalewska, Z., Ruszczyńska, A., Bulska E. 2005. Cu determination in crude oil
distillation products by atomic absorption and inductively coupled plasma mass
spectrometry after analyte transfer to aqueous solution, Spectrochimica Acta part B 60,
351-359.
12. Anselmi, A., Tittarelli, P.& Katskov, D. A. 2002. Determination of trace elements in
automotive fuels by filter furnace atomic absorption spectrometry. Spectrochimica Acta
Part B 57, 403–411.
13. Reboucas, M. V., Domingos, D., Santos, A. S. O. & Sampaio, L. 2010. Determination of
trace metals in naphtha by graphite furnace atomic absorption spectrometry: Comparison
between direct injection and microemulsion pretreatment procedures. Fuel Processing
Technology, 91, 1702-1709.
14. de Oliveira, A. P., de Moraes, M., Neto, J. A. G. & Lima, E. C. 2002. Simultaneous
determination of Al, As, Cu, Fe, Mn, and Ni in fuel ethanol by GFAAS. Atomic
Spectroscopy, 23, 39-43.
57
CHAPTER FOUR:
PRECONCENTRATION OF TRACE MULTI-ELEMENTS IN WATER SAMPLES
USING DOWEX 50W-X8 AND CHELEX-100 RESINS PRIOR TO THEIR
DETERMINATION USING INDUCTIVELY COUPLED PLASMA ATOMIC
EMISSION SPECTROMETRY (ICP OES)
ABSTRACT
This work presents a solid phase extraction (SPE) method for simultaneous preconcentration
of trace elements in water samples prior to their ICP OES determination. Dowex 50w-x8 and
Chelex-100 resins were used as SPE sorbent materials for preconcentration of trace Cd, Co,
Cr, Cu, Fe, Ni, Pb and Zn. The optimum sample pH, eluent concentration and sample flow
rates were found to 6, 3.0 mol L-1 and 3.0 mL min-1, respectively. In terms of multi-element
preconcentration capabilities, Dowex 50W-x8 appeared to be a better sorbent. The recoveries
for all the tested analytes were >95%. Under optimized conditions using Dowex 50W-x8, the
relative standard deviations for different metals were < 3%. The limits of detection and limits
of quantification ranged from 0.01-0.39 µg L-1 and 0.05-0.1.3 µg L-1, respectively. The
accuracy of the preconcentration method was confirmed by spike recovery test and the
analysis of certified reference materials. The SPE method was applied for preconcentration of
the analyte ions in tap water, bottled water and wastewater samples.
Keywords: Trace multi-element, simultaneous preconcentration, ion exchange resins, Chelex100, Dowex 50w-x8, Drinking water
4.1 INTRODUCTION
Water plays an important in role various human activities; these include agriculture,
sanitation, industrial production, energy, and transportation. Furthermore, it plays a vital role
in sustaining ecosystems that provide important services to both the environment and the
humans.1 Water scarcity is a global concern. However, the current water concern is the crisis of
governance rather than a crisis of physical scarcity. This is because (i) scarce water resources are
allocated inefficiently, (ii) water pollution is unregulated thus compromising water quality and
(iii) uneffective water service providers fail to serve the public. In addition, social and
environmental concerns are left unaddressed.1 Due to these inadequacies with conventional water
resources management, Integrated Water Resources Management (IWRM), assists in
addressing the issues such as environmental protection; promotion of economic growth,
58
Chapter 4:
Preconcentration of trace multi-elements in water
sustainable agricultural development and democratic participation in governance and
improving human health.1 Therefore, in this study we propose and develop a method that can
be used as a quantitative tool to monitor drinking water quality.
Monitoring trace metals in environmental samples is crucial since most of these metals
have negative or positive effects on human health depending on their concentration levels and
chemical form (e.g. Cr (VI)).2,3 Even though some of the metals such as Co, Cr, Cu, Fe, Ni
and Zn have the range of biochemical functions in living organisms, they can be toxic when
taken in excess.2 In contrast, the presence of non-essential metals such as lead and cadmium,
even in trace levels, damages central nervous function, lowers energy levels, damages the
blood composition, lung, kidneys, liver, and other vital organs.2,4 Therefore, monitoring of
trace elements in the environment is extremely important because it helps to control exposure
of humans and animals to these substances.5
Due to the high toxicity of heavy metals, it is crucial to detect ultra-low levels especially
in drinking water. The determination of metal ions in water samples has been successfully
carried out with different analytical techniques.6,7 These include flame atomic absorption
spectrometry (FAAS), graphite furnace atomic absorption spectrometry (GFAAS),
inductively coupled plasma optical emission spectrometry (ICP OES) and inductively
coupled plasma-mass spectrometry (ICP-MS).6,8,9 Among the above mentioned analytical
techniques, ICP OES is extensively used for the determination of metal ions. This is because
ICP OES displays attractive features such as multi-element detection capacity, wide linear
range, low limits of detection and high sample throughput.10 However, ICP OES is not
suitable for direct analysis of trace levels. Therefore, prior to detection with ICP OES an
effective pre-concentration step such as solid phase extraction (SPE) is required. The latter is
a widely applied and powerful pre-treatment approach prior to analysis.11 Among other
advantages, SPE is commonly used because a variety of extraction materials (sorbents) are
available, and the extraction can be tuned depending on how these sorbents interact with the
analytes. As a result, various adsorbents such as Amberlite XAD resins, Chelex-100 and
Dowex 50W-x4, among others, have been used for the separation/pre-concentration of trace
elements in various complex samples.7,12,13
The present study seeks to determine the most suitable cation exchange resin that will
have high metal retention efficiency over a wide operating pH range. Therefore, performance
of Chelex-100 and Dowex 50w-x8 sorbents for simultaneous pre-concentration of cadmium,
cobalt, chromium, copper, iron, nickel, lead and zinc in aqueous solutions was investigated.
59
Chapter 4:
Preconcentration of trace multi-elements in water
Various factors affecting the cation exchange process, such as sample volume, concentration
of the eluent, sample and eluent flow rates as well as the accuracy of the method, were
investigated.
4.2 EXPERIMENTAL
4.2.1 Instrumentation
Analyte metal ions were determined using Spetro Arcos ICP OES equipped with Cetac
ASX-520 autosampler. Solid phase extraction was carried out in a VacMaster-24 sample SPE
station (Supelco, PA, USA). The latter was used to control the sample loading and elution
flow rate to 3.0 ml min-1.
4.2.2 Reagents and Solutions
All reagents were of analytical reagent grade unless otherwise stated and double distilled
deionized waster (Millipore, Bedford, MA, USA) was used throughout the experiments.
Spectrascan stock solutions (1000 mg L-1) of Cd, Co, Cr, Cu, Fe, Ni, Pb and Zn (Teknolab
A/S, Dröbak, Norway) were used to prepare the working solutions for SPE at concentrations
of 6 µg L-1 (Cr, Co, Ni ), 10 µg L-1 (Cd), 12 µg L-1 (Pb), 30 µg L-1 (Cu, Fe and Zn). Working
solutions, as per the experimental requirements, were freshly prepared from the stock solution
for each experimental run. A Spectrascan multi-element standard solution at concentration of
100 mg L-1 was used to prepare working standard solutions at concentrations of 10-70 µg L-1
for Cd, Co, Cr, Fe, Ni and Pb; and 30-180 µg L-1 for Cu and Zn in measurements of
concentrations of analytes in all model and sample solutions. Solutions of nitric acid at
concentrations of 0.5, 1.0, 2.0, 3.0 and 4.0 mol L-1 used for the elution of the analytes from
the column were prepared from ultrapure concentrated acid (65%, Sigma-Aldrich, St. Loius,
MO, USA). The pH adjustments were performed with 1.0 M HNO3 and NaOH solutions. The
cation exchangers used in this study as packing materials were Chelex-100 and Dowex 50wx8 (sodium forms) purchased from Sigma Aldrich (St. Loius, MO, USA).
4.2.3 Water Samples and Preparation
Tap water samples were obtained from University of Johannesburg (Doornfotein and
Kingsway campuses). Effluent wastewater samples were collected from Johannesburg Water.
The wastewater samples were filtered through a 0.45 µm pore-size double distilled deionised
60
Chapter 4:
Preconcentration of trace multi-elements in water
cellulose nitrate membrane to remove any fine particulate matter present. Bottled water
samples were obtained from a local supermarket.
4.2.4 Column Preparation
Supelco polyethylene columns (1.35 cm in diameter and 6.5 cm in length) with frits were
employed for SPE. The columns were soaked in 5% HNO3 solution and then rinsed
successively with double distilled deionized waster (Millipore, Bedford, MA, USA).
Afterwards, slurries of 1.5 g of Chelex-100 or Dowex 50W-x8 resin in double distilled water
were loaded into the columns. Porous frits were placed at the bottom and at the top of the
column for allowing the adsorbent to settle properly. The resin columns were washed using
triple distilled water followed by conditioning with 10 mL ammonium acetate buffer (1.0 mol
L-1, pH 9.0). After each use, the resin in the column was washed with 20 mL of water
followed by 10 mL of 1.0 mol L-1 NaOH. This was done in order to keep the resin in sodium
form.
4.2.5 Preconcentration Procedure
The pH values of the model solutions of Cd, Co, Cr, Cu, Fe, Ni, Pb and Zn were adjusted
to 6. The solutions were then each passed through a SPE column packed with either Chelex100 or Dowex 50w-x8 at a flow rate of 2.0 and 3.0 mL min-1, respectively. Metal ions
retained on the resins were eluted with 5.0 mL of 0.5-4.0 mol L-1HNO3 at a flow rate of 3.0
mL min-1. The metal concentrations in the final solution were determined using ICP OES.
The same procedure was applied to the blank solutions. After each run, the columns were
conditioned as per Section 2.3.
4.2.6 Optimization of Preconcentration Parameters
The SPE system was optimized in order to determine the best retention/ elution
conditions for trace metal ion determination with good sensitivity and precision.6 Several
experimental variables affecting the pre-concentration system such as eluent concentrations,
sample and eluent flow rates, among other parameters, were evaluated and optimized. To
obtain these conditions, preliminary tests were performed to investigate factors that exert
significant influence on the retention of the analytes by cation exchange resin. The factors
selected include eluent concentration, sample volume as well as sample flow rate. In previous
61
Chapter 4:
Preconcentration of trace multi-elements in water
study,6 the maximum retention of the analytes onto the cation exchange resin was observed at
pH 6. We decided to use the same pH value at 6 in the present study, for all the experiments.
The optimization of the sample flow rate was carried out to ensure the quantitative retention
of the analytes of interest. The effect of flow rate of the sample solution on the retention of
the studied metal ions on the Chelex-100 and Dowex 50w-x8 resins was carried out with a
column packed with 1.5 g of resin. Sample solutions were passed through the column at
various flow rates (1.0-5.0 mL min-1). The flow rates less than 1.0 mL min-1 were not studied
to avoid long analysis time. The percentage recoveries were calculated by relating the final
obtained concentration (Cf) of the analyte to the original concentration (Ci) of the metal ion in
the model solution.
%R 
Cf
Ci
 100
(1)
4.3. RESULTS AND DISCUSSION
4.3.1 Effect of pH
The sample pH for quantitative preconcentration of Cd, Co, Cr, Cu, Fe, Ni, Pb and Zn in
the Dowex 50W-x8 and Chelex 100 columns is one of the most important factors.14 This is
because, highly acidic solutions may lead to protonation of resin’s functional group while
highly alkaline solution may result in the precipitation of metal ions as hydroxides. This may
results in the underestimation of metal ion concentrations in drinking water samples.
Therefore, the effect of sample pH on the retention of the analytes onto Dowex 50W-x8 and
Chelex 100 resins was carried out between pH 4 and 10. The influence of the sample pH on
the pre-concentration of Cd, Co, Cr, Cu, Fe, Ni, Pb and Zn is presented in Fig. 1. It was
observed that for both resins, lower recoveries (<95%) were obtained for all metal ions at pH
4 and this can be attributed to competition between the metal ions and the hydronium ions for
the active sites on the adsorbent surface. However, quantitative recoveries were obtained for
all analytes at pH range 6-7 and 5-7 for Dowex 50W-x8 and Chelex 100 resins, respectively.
In this study, quantitative recoveries defined as the percentages that are more than or equal to
95%. Therefore, pH 6 was chosen for subsequent investigations.
62
Chapter 4:
Preconcentration of trace multi-elements in water
Fig. 4.1. The effect of pH on the recoveries of 20 µg L−1 metal ion solution: A) Dowex 50W8, B) Chelex-100. sample volume 20 mL; amount of resin 1.5 g, flow rates of sample and
eluent 3.0 mL min−1, respectively n = 3
4.3.2 Effect of Eluent Concentration
The desorption of the analytes bound onto the surface of the Dowex 50w-x8 and Chelex100 resins is expected to be achieved by proton exchange from the acid solution.12 The
desorbing eluent should also be compatible with the pre-concentration (SPE) procedure. For
this reason, HCl was not used in this work due to the risk of the formation of insoluble
chloride complexes with some of the metal ions e.g. Pb. Nitric acid is preferred compared to
other acids (such as HCl and H2SO4) because all nitrate salts are soluble compared to chloride
and sulfate salts. The desorption/ elution of metal ions from Dowex 50w-x8 and Chelex-100
using various nitric acid concentrations (0.5-4.0 mol L-1) has been investigated. From the
results in Fig. 4.2, it was observed that in order to desorb the metal ions from Dowex 50w-x8,
a higher concentration of nitric acid as compared to Chelex-100 was used. This implied that
Dowex 50w-x8 strongly binds the metal ions compared to Chelex-100. The results indicated
that metal ions were quantitavely recovered from Chelex-100 when the concentration of
HNO3 was between 1.0 and 2.0 mol L-1 while in the case of Dowex 50w-x8 a flow rate of 3.0
mol L-1 was used. This should be expected because Chelex-100 (iminodiacetic acid
functional group) is a weakly acidic cation exchanger whereas Dowex 50w-x8 (sulfonic acid
functional group) is a strongly acidic cation exchanger.
63
Chapter 4:
Preconcentration of trace multi-elements in water
Fig. 4.2. Influences of the eluent concentration on the recoveries of the analytes on Dowex
50w-x8 resin column. sample volume 20 mL; amount of resin 1.5 g, flow rates of sample and
eluent 3.0 mL min−1, respectively n = 3
4.3.3 Effect of Flow Rate
The optimization of the sample flow rate was carried out to ensure the quantitative
retention of the analytes of interest. The effect of flow rate of the sample solution on the
retention of the studied metal ions on the Dowex 50w-x8 resin was carried out with a column
packed with 1.5 g of resin. Sample solutions (20 mL) were passed through the column at
various flow rates (1.0-5.0 mL min-1). The flow rates less than 1.0 mL min-1 were not studied
to avoid long analysis time. The optimum flow rate for this work was defined as the rate of
flow of the sample solution through the column at which more than 95% retention of metal
ions takes place. The results showed that the optimum flow rate for quantitative sorption of
metal ions onto the resin was between 1.0 and 3.0 mL min-1. Such observation were expected
because, ideally lower flow rate should give highest recovery. The increase of flow rate more
than 3.0 mL min-1 caused a gradual decrease in sorption due to insufficient contact time
between the resin and the metal ions, 2.0 and 3.0 mL min-1 flow rates were chosen as the
optimum flow rate for sample loading onto Chelex-100 and Dowex 50W-x8 resins,
respectively.
64
Chapter 4:
Preconcentration of trace multi-elements in water
4.3.4 Preconcentration of Multi-Element
The efficiency of studied cation exchange resins for pre-concentration of multi-elements
(concentration of each analyte equal to 10 µg L-1) in aqueous solution was investigated under
optimum conditions. The results indicated that the highest retention of the analytes from
aqueous model solutions was observed on Dowex 50W-x8 resin (Table 4.1). This might be
due to the larger exchange capacity (1.7 meq mL-1) and its functional groups (sulfonic acid).
The recoveries of metal ions from Dowex 50W-x8 ranged from 95 to 101%. It can be
concluded that the affinity of studied analytes towards Dowex 50W-x8 was very similar.
Therefore, they could be pre-concentrated with the same efficiency.15 The results in Table 4.1
indicated that Chelex-100 was only suitable for the removal of Cu, Fe and Zn at an optimum
flow rate of 2.0 mL min-1. The rest of the metals were not quantitatively recovered at this
optimum flow rate. It was then concluded that Chelex-100 was not suitable for preconcentration of multi-element in aqueous matrices. Therefore, Dowex 50W-x8 at an
optimum flow rate of 3.0 mL min-1 was used for further analysis.
Table 4.1. Recovery (%) of multi-element in aqueous solution using Dowex 50W-x8 and
Chelex-100 SPE methods
Resins
Recovery, (%)
Dowex
Chelex
Cd
Co
Cr
Cu
Fe
Ni
Pb
Zn
99.2±1.4 97.4±1.3 96.3±1.2
101±1.2 99.3±4.2 96.4±1.4 95.1±1.2 97.9±2.1
88.9±1.2 80.6±3.8 85.3.1±4.0 95.8±2.4 97.5±2.4 78.1±1.2 91.0±1.2 96.5±3.8
Experimental conditions: sample volume = 20 mL amount of resin = 1.5 g; flow rates = 2.0
and 3.0 mL min−1 for Chelex-100 and Dowex 50W-x8, respectively; eluent volume = 5 mL;
replicates = 3
4.3.5 Effect of Sample Volume
The influence of sample volume on the recoveries of analyte ions on the solid phase was
studied in order to obtain high preconcentration factor.16,17 Therefore, the effect of sample
volume on the retention of Cd, Co, Cr, Cu, Fe, Ni, Pb and Zn onto Dowex 50W-x8 resin was
investigated in the range of 50-1000 mL, while keeping the metal ion concentration fixed at
10 µg L-1. It is seen from the Fig. 4.3 that the retention of metal ions can be achieved
quantitatively (≥95%) by sample volume up to 700 mL. As the volume the sample Therefore,
65
Chapter 4:
Preconcentration of trace multi-elements in water
the highest preconcentration factor was found to be 140 when the adsorbed metal ions were
eluted with 5 mL of 3 mol L−1 HNO3. At volumes higher than 700 mL, a decrease in
quantitative recoveries of metal ions was observed. This might be due to the saturation of the
active sites of the adsorbent.
Fig. 4.3. Effect of sample volume on the recoveries of metal ions: pH 6.0; analyte
concentration 10 µg L-1; amount of sorbent 1.5 g; flow rates of sample and eluent 3.0 mL
min−1; eluent volume 5 mL; replicates n=3
4.3.6 Column Regeneration
In order to investigate the recyclability of Dowex 50W-x8 column, successive retention
and elution cycles were performed by passing 20 mL of copper, iron and zinc solutions
through the column. The regeneration of Dowex 50W-x8 column were evaluated by
monitoring the changes in the recoveries of copper, iron and zinc through 200 retentionelution cycles. The Dowex 50W-x8 column was reused after regeneration with 20 mL double
distilled water and 10 mL of 1.0 mol L-1 NaOH, respectively. It was found to be stable up to
150 retention/elution cycles without observable decrease in the recoveries of copper, iron and
zinc (> 95%). It should be noted that this column regeneration can be affect by the matrix of
66
Chapter 4:
Preconcentration of trace multi-elements in water
the real samples. This implies that in the case of real samples, the retention/elution cycles
might lower that when model solutions are used.
4.3.7 Analytical Performances
The analytical performance of the SPE-Dowex 50W-x8 method under optimum
conditions for pre-concentration of metal ions was evaluated. The dynamic linear range of the
method was evaluated and obtained as 10-70 µg L-1 for Cd, Cr, Co, Ni and Pb; 30-160 µg L-1
for Cu, Fe and Zn. The correlation coefficients (R2) of the calibration curves were in the
range 0.9991–0.9997. The limits of detection (LOD) and limits of quantification (LOQ) of
the SPE method were investigated under optimum experimental conditions by applying the
blank solution procedure. They were calculated according to Eqs. 1 and 2.18
LOD=
3  SD
m
(1)
LOQ=
10  SD
m
(2)
where SD is standard deviation of the blank signal (n = 20) and m is the gradient of the
calibration curve. For a 100 mL, the LOD of Cd, Co, Cr, Cu, Fe, Ni, Pb, and Zn were found
to be 0.06, 0.08, 0.05, 0.02, 0.01, 0.39 and 0.02 µg L-1, respectively; and LOQ were 0.19,
0.26, 0.11, 0.08, 0.05, 1.3 and 0.08 µg L-1 for Cd, Co, Cr, Cu, Fe, Ni, Pb, and Zn,
respectively. The LOD and LOQ values obtained in this study can be improved by increasing
the volume of the sample. The instrumental detection limits (IDL) were 0.1, 0.2, 0.2, 0.4, 0.1,
1.0 and 0.2 µg L-1 for Cd, Co, Cr, Cu, Fe, Ni, Pb, and Zn, respectively. It can be seen that
SPE/ICP-OES method has improved LODs.
The precision (reproducibility) of the SPE method was studied by performing 15
successive measurements at a concentration level of 10 µg L-1 of multi-element aqueous
solution (containing Cd, Co, Cr, Cu, Fe, Ni, Pb, and Zn). The overall reproducibility of preconcentration procedure expressed in terms of relative standard deviation (%RSD) was
reasonably good (<3%).
The analytical procedure was validated by analysis of certified reference materials, BCR713 (Effluent wastewater) and CRM TMDW-500 drinking water. The results in Table 4.2
displayed a good agreement between the obtained values and certified values with
satisfactory recoveries ranging 97-103%. These results confirmed the validity of the Dowex
50W-x8 SPE method for the preconcentration of the metal ions from aqueous matrices. In
addition, they confirm that SPE-Dowex 50W-x8 is a suitable sample preparation method that
67
Chapter 4:
Preconcentration of trace multi-elements in water
can be applied an a last step in drinking water purification plants to check the quality of
drinking water before it is distributed for human consumption.
Table 4.2. Analysis of certified reference materials (mean of 3 replicates; concentration in µg
L-1)
Cations
BCR-713 Effluent wastewater
Certified
5.1±0.6
NCa
21.9±2.4
68.4±3.3
398.3±32.0
30.6±4.6
47.0±4
216.2±32.13
Cd
Co
Cr
Cu
Fe
Ni
Pb
Zn
a
NC= not certified
Obtained
5.0±0.8
15.3±1.3
22.1±0.7
66.8±1.3
383.5±3.5
29.7±2.1
48.3±1.3
213.5±1.8
CRM TMDW-500 drinking water
Recovery
97.5±1.1
101±1
97.7±2.4
96.3±1.4
97.1±1.3
103±1
98.8±3.1
Certified
10.0±0.05
25.0±0.1
20.0±0.1
20.0±0.1
100.0±0.5
60.0±0.3
40.0±0.2
70.0±0.4
Obtained
9.7±0.7
24.3±0.6
19.6±0.3
20.1±0.2
97.8±0.6
57.9±0.9
38.7±0.4
70.3±0.1
Recovery (%)
97.0±2.1
97.2±1.4
98.0±1.1
101±1
97.8±1.7
96.5±1.3
96.8±2.4
100±1
4.3.8 Application
Water quality is one of the important aspects in human and environmental health. This is
because majority of diseases that cause morbidity and mortality in population are water
related.19 For this reason, different organisations such as USEPA, WHO and South African
National Standards (SANS) have set some guidelines for the concentration limits of heavy
metals in drinking water.20-22
The Dowex 50W-x8 SPE method was applied for the determination of trace Cd, Co, Cr,
Cu, Fe, Ni, Pb, and Zn in tap water, bottled water and wastewater (effluent) samples. For
analysis, 100 mL of water samples were studied by the Dowex 50W-x8 SPE method. The
results of analysis are given in Table 4.3. Copper, Fe, Ni and Zn were present in all water
samples investigated. In the case of drinking water, the highest concentrations of Zn, Cu, Fe
and Ni were found in TW 1, TW 3 and TW 5 samples, respectively. The level of the
cadmium was found to be below the detection limit in TW 1, TW 2, TW 3 and BW 1 samples
whereas Pb was not detected in TW 1, TW 2, BW 1 and BW 2. Chromium on the other hand
was found to be below the detection limit in TW 2 and BW2 samples whereas TW 4 and TW
5 were found to contain highest Cr concentration compared to other water samples.
Generally, bottled water samples were found to contain lower metal ion content compared to
68
Chapter 4:
Preconcentration of trace multi-elements in water
tap water samples. The differences in metal levels between tap and bottled water samples
might be attributed to different types of filters or/and adsorbents used and to how often these
filters or/and adsorbents are changed or cleaned. As expected, wastewater contained all the
studied metal ions with highest Fe content, followed by Zn and Cd. Johannesburg is a mining
area and cadmium occurs naturally on the earth’s crust and can be found in natural deposits
such as ores containing other elements. Therefore, the high concentration of Cd in wastewater
might be due to release of this element during mining activities.
The metal ion concentrations obtained were compared against the allowed maximum
contamination levels (MCLs) by WHO,20 USEPA21 and SANS 24122 in drinking water. The
MCL values for the analytes of interest are given in Table 4.4. Based on the drinking water
samples analysed, all samples investigated in this study showed no pollution Co, Cr, Cu, Fe,
Ni, Pb, and Zn except for TW 4 and TW 5 samples which showed pollution of Cd.
69
Chapter 4:
Preconcentration of trace multi-elements in water
Table 4.3. Concentration (µg L-1) of Cd, Co, Cr, Cu, Fe, Ni, Pb, and Zn in water samples (replicates n=5, volume 100 mL, final volume 5 mL)
Samples
Cd
Co
Cr
Cu
Fe
Ni
Pb
Zn
TWa 1
ND
ND
0.36±0.01
60.2±0.9
59.9±0.4
6.12±0.16
ND
4801±2
TW 2
ND
ND
ND
125±0.3
10.12±0.2
9.37±0.21
ND
54.6±0.8
TW 3
ND
0.80±0.02
2.78±0.05
277±1.9
16.2±1.0
4.70±0.27
0.62±0.11
166±0.1
TW 4
9.21±0.74
12.4±0.2
23.5±0.9
22.5±0.3
93.7±1.1
27.4±0.64
1.82±0.05
78.7±0.7
TW 5
9.83±0.21
11.3±0.1
26.3±0.9
20.8±0.7
103±1
26.44±0.6
2.12±0.44
76.5±0.8
BW 1
ND
ND
1.16±0.01
34.4±0.5
8.25±0.2
8.95±0.14
ND
21.8±0.1
BW 2
1.22±0.32
2.10±0.03
ND
12.3±0.1
19.8±0.28
3.48±0.22
ND
23.4±0.5
WWEc
40.3±0.2
2.91±0.59
5.48±0.11
22.2±0.2
265±3
3.54±0.33
9.15±0.42
91.5±0.2
Pb
˂20
15.0
10
Zn
˂5000
5000
3000
b
a
TW= tap water; bBW= bottled water; cWWE= wastewater Effluent; ND = not detectable
Table 4.4. Guidelines for the presence of heavy metals in drinking water; concentration in µg L-1
Organizations
SANS
USEPA
WHO
Cd
˂5.0
5.0
3.0
Co
˂500
50.0
Cr
˂100
100
5.0
Cu
˂1000
1000-1300
2000
70
Fe
˂200
300
300
Ni
˂150
70
Chapter 4:
Preconcentration of trace multi-elements in water
4.4 CONCLUSIONS
In this study, the efficiency of Chelex-100 and Dowex 50W-x8 cation exchange resins
for the separation and pre-concentration of multi-element in aqueous solutions was
investigated and the results demonstrated that Dowex 50W-x8 resin has good capability
and efficiency for the simultaneous preconcentration of metal ions. In comparison, Chelex100 showed limited performance (preconcentration with percentage recovery ≥ 95%) to
only few metals namely Cu, Fe and Zn whereas Dowex 50W-x8 had the best overall
performance for a wider range of metals.
The limits of detection (0.01-0.39 µg L-1) and quantification (0.05-1.3 µg L-1) were
relatively low, suggesting that the method may be applied for trace analysis of these
analytes in drinking water and wastewater samples. The accuracy (% recovery) and
precision (% RSD) of the Dowex 50W-x8 SPE method ranged from 95-105% and 1.22.2%, respectively. The proposed procedure was applied to the determination of trace
metals in CRMs drinking water and wastewater samples. The results revealed that the
method can be used for routine monitoring or spot analysis of metal ion contaminants in
the drinking water supplies. In addition, the results indicated that all except for TW 4 and
TW 5 passed the drinking water standards (guidelines) for the studied trace metals set by
WHO, USEPA and SANS.
4.5 REFERENCES
1. Xie, M. 2006. Integrated Water Resources Management (IWRM)-Introduction to
Principles and Practices. World Bank Institute (WBI). Found online at
http://www.pacificwater.org/userfiles/file/IWRM/Toolboxes/introduction%20to%20iwr
m/IWRM%20Introduction.pdf Accessed 5 September 2012.
2. Tufekci, M., Bulut, V., Elvan, H., Ozdes, D., Soylak, M. & Duran, C. 2012.
Determination of Pb(II), Zn(II), Cd(II), and Co(II) ions by flame atomic absorption
spectrometry in food and water samples after preconcentration by coprecipitation with
Mo(VI)-diethyldithiocarbamate. Environmental Monitoring and Assessment, 1-9.
3. Kazi, T.G., Arain, M.B. Baig, J.A. Jamali, M.K. Afridi, H.I. Jalbani, N. Sarfraz, R.A.
Shah, A.Q. & Niaz, A. 2009. The correlation of arsenic levels in drinking water with the
biological samples of skin disorders. Science and Total Environment, 407, 1019-1026.
4. Afridi, H. I., Kazi, T. G., Jamali, M. K., Kazi, G. H., Arain, M. B., Jalbani, N., Shar, G.
Q., Sarfaraz, R. A., 2006. Evaluation of toxic metals in biological samples (scalp hair,
71
Chapter 4:
Preconcentration of trace multi-elements in water
blood and urine) of steel mill workers by electrothermal atomic absorption
spectrometry. Toxicology and Industrial Health, 22, 381-393.
5. dos Santos Silva, E., Correia, L., Dos Santos, L., Dos Santos Vieira, E., Lemos, V.,
2012. Dispersive liquid-liquid microextraction for simultaneous determination of
cadmium, cobalt, lead and nickel in water samples by inductively coupled plasma
optical emission spectrometry. Microchimica Acta, 178, 269-275.
6. Soylak, M., 2004. Solid phase extraction of Cu(II), Pb(II), Fe(III), Co(II) and Cr(III) on
Chelex-100 column prior to their flame atomic spectrometric determinations. Analytical
Letter, 37, 1203-1217.
7. Tuzen, M., Soylak M., Elci, L., 2005. Multi-element pre-concentration of heavy metal
ions by solid phase extraction on Chromosorb 108. Analytica Chimica Acta, 548, 101108.
8. Komjarova, I., Blust, R., 2006. Comparison of liquid-liquid extraction, solid-phase
extraction and co-precipitation pre-concentration methods for the determination of
cadmium, copper, nickel, lead and zinc in seawater. Analytica Chimica Acta, 576, 221228.
9. Bezerra, M. A., Dos Santos, W. N. L., Lemos, V. A., Korn, M. D. G. A., Ferreira, S. L.
C., 2007. On-line system for pre-concentration and determination of metals in
vegetables by Inductively Coupled Plasma Optical Emission Spectrometry. Journal of
Hazardous Material, 148, 334-339.
10. Xiong, C., Qin, Y. Hu, B., 2010. On-line separation/preconcentration of V(IV)/V(V) in
environmental water samples with CTAB-modified alkyl silica microcolumn and their
determination by inductively coupled plasma-optical emission spectrometry. Journal of
Hazardous Material, 178, 164-170.
11. Hennion, M.-C., 1999. Solid-phase extraction: method development, sorbents, and
coupling with liquid chromatography. Journal of Chromatography A 856, 3-54.
12. Malla, M. E., Alvarez, M. B., Batistoni D.A., 2002. Evaluation of sorption and
desorption characteristics of cadmium, lead and zinc on Amberlite IRC-718
iminodiacetate chelating ion exchanger. Talanta, 57, 277-287.
13. Ramesh A., Mohan, K.R., Seshaiah, K., 2002. Pre-concentration of trace metals on
Amberlite XAD-4 resin coated with dithiocarbamates and determination by inductively
coupled plasma-atomic emission spectrometry in saline matrices. Talanta, 57, 243-252.
14. Jiménez, M. A. S., Velarte, R., Castillo, J. R. 2002. Performance of different
preconcentration columns used in sequential injection analysis and inductively coupled
plasma-mass spectrometry for multielemental determination in seawater.
Spectrochimica Acta B: Atomic Spectroscopy, 57, 391-402.
15. Pyrzyñska, K., Joñca, Z., 2000. Multi-element pre-concentration and removal of trace
metals by solid-phase extraction. Analytical Letters, 33, 1441 - 1450
72
Chapter 4:
Preconcentration of trace multi-elements in water
16. Shishehbore, M., Afkhami, A., Bagheri, H., 2011. Salicylic acid functionalized silicacoated magnetite nanoparticles for solid phase extraction and preconcentration of some
heavy metal ions from various real samples. Chemistry Central Journal,5, 1-10.
17. Aydin, F. A., Soylak, M., 2010. Separation, preconcentration and inductively coupled
plasma-mass spectrometric (ICP-MS) determination of thorium(IV), titanium(IV),
iron(III), lead(II) and chromium(III) on 2-nitroso-1-naphthol impregnated MCI GEL
CHP20P resin. Journal of Hazardous Material, 173, 669-674.
18. Ingle Jr., J.D., Grouch, S.R. (1988) Spectrochemical Analysis, Prentice-Hall Inc. 173174.
19. El-Harouny, M. A., El-Dakroory, S. A., Attalla, S. M., Hasan, N. A., Hegazy, R., 2008.
Chemical quality of tap water versus bottled water: evaluation of some heavy metals
and elements content of drinking water in dakahlia governorate – Egypt. Mansoura
Journal of Forensic Medicine and Clinical Toxicology 16, 1-15.
20. World Health Organization (WHO). 2008. Guidelines for drinking-water quality. 3rd
edition incorporating the 1st and 2nd Addenda, Vol. 1: Recommendations, Geneva,
Switzerland.
21. United States Environmental Protection Agency (USEPA). 2011. Edition of the
drinking water standards and health advisories. EPA 820-R-11-002, Office of water, US
Environmental Protection Agency, Washington, D.C.
22. South African National Standards (SANS) 241. 2005. Drinking water specification.
Edition 6.
73
CHAPTER FIVE:
KINETICS AND EQUILIBRIUM STUDIES FOR THE REMOVAL OF COBALT,
MANGANESE AND SILVER IN ETHANOL USING DOWEX 50W-X8 CATION
EXCHANGE RESIN
ABSTRACT
Removal of Ag, Co and Mn ions in ethanol by cation exchange resin, Dowex 50W-x8,
was investigated. The adsorption characteristics of metal ions onto Dowex 50W-x8 resin
were described by Langmuir isotherms. The maximum sorption exchange capacities at 298 K
were found to be 47.4 mg g-1, 52.6 mg g-1 and 57.0 mg g-1 for Ag, Co and Mn, respectively.
The data was also fitted to Temkin and Dubinin-Radushkevich adsorption isotherm models to
evaluate other adsorption properties. The ion exchange of silver, cobalt and manganese on
cation exchange resin followed pseudo-second-order kinetics and the intraparticle diffusion
was rate-determining step. The thermodynamic parameters indicated that the sorption of
metal ions onto Dowex 50W-x8 resin was spontaneous (negative ∆G0) and endothermic in
nature (positive ∆H0) implying that the adsorption capacity increased with increasing
temperature. Ion exchange capacities of Ag, Co and Mn on the Dowex 50W-x8 were studied
in the model ethanol solutions, and the resin can be regenerated by eluting metal ions with 3.0
mol L-1 HNO3 followed by washing it with 10 ml of double distilled water and 10 ml of 2.0
M NaOH, respectively. The proposed method was applied for metal ion removal in real
ethanol samples.
Keywords: Kinetics; ion exchange; metal ions; ethanol; Dowex 50W-x8 resin
5.1 INTRODUCTION
Ethanol finds a wide range of applications in pharmaceutical industries, food industries,
petroleum industries and paint formulations, among others.1,2 It is used as an extractant in the
production of medicine, antibiotics and as a component in food additives and food
processing. In addition, it is used as a fuel additive, antifreeze agent, solvent for gums,
chemical intermediate and solvent in chemical industries.2-4
The presence of trace metals in pharmaceutical products, fuel, food and paint is
undesirable. In pharmaceutical products, metal ions have an ability to decompose the material
of interest or catalyse the degradation of the active pharmaceutical ingredient. The
decomposition may lead to potential toxic effects.5,6 In addition, ethanol is used as fuel
74
Chapter five:
Kinetics and equilibrium studies
additive, therefore, metal ions present in it may be responsible for the decomposition and
poor performance of the fuel, leading to corrosion of vehicle parts and formation of
precipitates.7 Furthermore, monitoring toxic metals in ethanol is also important, since they
are released into the atmosphere by the fuel combustion, thus causing air pollution8.
Therefore, it is crucial to investigate the levels of trace metals in ethanol because their
presence in these products has negative health effects and most metal impurities are
detrimental to catalytic processes used in industry.
Conventional methods that have been used for removal of metal ions from different
matrices include precipitation, ion exchange and liquid-liquid extraction, among other
methods.9-11 Among these traditional methods, ion exchange receives considerable interest
because it can effectively separate and preconcentrate some anions and cations and can easily
be regenerated and reused by a regeneration process.10,12,13 In literature, there have been
various investigations about removal of heavy metals from aqueous solution by ion exchange
resins such as Dowex10, Chelex-10012 and Amberlite.13 However, the application of Dowex
50W-x8 for removal of metal ions from an organic matrix has not yet been explored.
Moreover, ion exchange in organic matrices offers interesting possibilities for the separation
and extraction of metal ions in petroleum processing.14
The literature search shows extensive focus on the methods for analysis Cu2+, Zn2+, Ni2+,
Pb2+, Cd2+ and Fe2+, particularly in organic matrices (16-18). Due to very few studies
reported on the analysis of Co(II), Mn(II) and Ag(I), in organic phase samples. There is an
increasing interest in these metals because they occur naturally in coal and crude oil. In view
of this, the aim of this work was to investigate the removal of Ag, Co and Mn in ethanol
solution using Dowex 50W-x8 at different temperatures. This study was carried out in order
to evaluate the performance of ion exchange process, predictions of the analyte partitioning
between an organic solution and ion exchanger surface. The information obtained in this
work will help in further application of ion exchange resins in solid phase extraction of metal
ions in gasoline samples particularly where the concentrations are too low and may not be
detectable with GFAAS or ICP. For this purpose, a batch technique was used in order to
determine the equilibrium data. The influences of experimental parameters such as contact
time, pH and the amount of resin, were studied. Adsorption isotherms, kinetics and
thermodynamics were studied to understand the adsorption mechanism of metal ions on
Dowex 50W-x8 resin.
75
Chapter five:
Kinetics and equilibrium studies
5.2 EXPERIMENTAL
5.2.1 Materials and Reagents
All reagents were of analytical grade unless otherwise stated and double distilled deionized
waster (Millipore, Bedford, MA, USA) was used throughout the experiments. Dowex 50Wx8 (sodium form), and ammonia were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Absolute ethanol and glacial acetic acid were obtained from Merck (Darmstadt, Germany)
whereas Spectrascan stock solutions (1000 mg L-1) of Ag(I), Co (II)and Mn(II) were
purchased from Industrial Analytical (Pty) Ltd, (Johannesburg, South Africa). Absolute
ethanol was used to prepare model solutions. Spectrascan stock solutions of Ag(I), Co(II) and
Mn(II), were used to prepare the working solutions at different concentrations. Working
solutions, as per the experimental requirements, were freshly prepared from the stock solution
for each experimental run. The pH adjustments were performed with glacial acetic acid and
ammonia solutions. The cation exchanger used in this study as the adsorbent was Dowex
50W-x8.
5.2.2 Apparatus
Perkin-Elmer (Shelton, Ct, USA) AAnalyst 400 Atomic Absorption Spectrometer
equipped with HGA-900 graphite furnace and AS-800 autosampler was used to analyse the
concentration of metal ions in ethanol solution. Lumina hollow cathode lamps from PerkinElmer were used in these experiments. The heating program of the electrothermal atomic
absorption spectrometry (ETAAS) used for the determination of Ag, Co and Mn in ethanol
model solutions is shown in Table 5.1.
76
Chapter five:
Kinetics and equilibrium studies
Table 5.1. Graphite furnace temperature program for the determination of silver, cobalt and
manganese in ethanol model solutions
Step
Drying 1
Drying 2
Pyrolysis
Ag
Co
Mn
Atomisation
Ag
Co
Mn
Clean out
Temperature (°C)
110
200
Remap (s)
1
1
Hold (s)
5
10
1200
800
1400
5
5
5
20
20
20
2300
2300
2000
2450
0
0
0
1
6
6
6
6
5.2.3 Adsorption Studies
Before batch experiments, three successive elution-washing cycles using 3 mol L-1 HNO3,
H2O and 2 mol L-1 NaOH were applied to the resin in order to remove impurities. The last
step of the conditioning which consisted of percolating excess NaOH through the column was
carried out in order to convert the resin to sodium form. The batch ion experiments were
carried out to monitor the removal of Ag, Co and Mn by Dowex 50W-x8. Different pH
values, contact time and resin amounts were optimized at controlled temperature.
The effect of each factor was investigated by keeping other variables constant. For the
effect of pH experiments, 20 ml of ethanol model solutions containing 10 mg L-1 of Ag, Co
and Mn were added into 100 ml plastic bottles (previously soaked in 1% nitric acid) with
constant amount of resin (0.05 g). Te pH of the ethanol model solutions was first adjusted to
different pH values (4-10) using acetic acid or dilute ammonium solution. Solutions were
agitated at 200 rpm for a predetermined period. Temperatures were controlled by a thermostat
shaker bath. All batch experiments were performed in triplicate. At the end of agitation time,
the resin was filtered and metal concentrations in the solutions were analysed by ETAAS. To
study the effect of the resin amount, different masses (0.02 to 1.0 g) were added into ethanol
solution containing 10 mg L-1 of each metal ion. Equilibrium adsorption experiments were
performed with initial concentrations of the metal ions corresponding to 10 to 300 mg L-1
while maintaining the resin mass, pH and equilibration time constant. After reaching
77
Chapter five:
Kinetics and equilibrium studies
equilibrium, solutions were filtered and analyzed. The adsorption capacity at different
concentrations was calculated using Eq. (1).
qe 
(C 0  C e )V
m
(1)
where qe (mg g-1) is the equilibrium adsorption capacity, C0 and Ce (mg L-1) are the initial and
equilibrium concentration of metal ions in solution, V (L) is the volume, and m (g) is the
amount of the resin.
5.2.4 Kinetic Studies
Kinetic experiments were performed by using 20 mL of metal ion solutions at a
concentration of 10 mg L-1. Samples were agitated at different time intervals (0–60 min) and
remaining metal ion concentrations were analysed by ETAAS. The rate constants were
calculated using conventional rate expressions. Eq. (2) was used to determine adsorption
capacity (qt) at time t:
qt 
(C 0  Ct )V
m
(2)
where qt (mg g-1) is the adsorption capacity at time t, C0 (mg L-1) is the initial metal
concentration, Ct (mg L-1) is the concentration of metal ions in solution at time t, V (L) is the
volume, and m (g) is the amount of the resin.
5.2.5 Adsorption Thermodynamics
The effect of temperature on the adsorption of metal ions onto Dowex 50W-x8 resin was
studied under five different temperatures: 20, 25, 30, 35 and 40 °C. Thermodynamic studies
were conducted with initial concentrations of 10-300 mg L-1 of Ag, Co and Mn ethanol
solutions at pH 6 and were agitated using a thermostat shaker bath for 20 min at a speed of
200 rpm.
5.3 RESULTS AND DISCUSSION
5.3.1 Effect of Contact Time
Contact time plays an important role in the adsorption of metal ion onto a solid material.
Fig. 5.1 shows the effect of contact time on Dowex 50W-x8 resin of Ag, Co and Mn. The
results indicated that the removal efficiency (%) of metal ions adsorbed increased with
78
Chapter five:
Kinetics and equilibrium studies
increasing time of equilibration and it reached a steady state value at 20 min for all the
studied metal ions. Therefore, 20 min was selected for further investigations.
Fig. 5.1. Effect of contact time on retention of Ag, Co and Mn using Dowex 50W-x8 resin:
Initial concentration of metal ions 10 mg L-1; amount of resin 0.05 g; sample volume 20 mL;
temperature 298 K; stirring rate 200 rpm; stirring time 0-60 min; initial pH 6
5.3.2 Effect of pH
The pH of the sample solution is one of the most important parameter that affects the ionexchange process. This is probably because hydrogen ions compete with the analytes for the
exchange sites of the adsorbent. In addition, the solution pH influences the ionization of the
resin’s functional groups and the speciation of metal ions.10,15 The effect of pH on the
removal of metal ions in ethanol using Dowex 50W-x8 resin were studied at room
temperature by varying the initial pH of metal solution from 4.0 to 10. Constant resin amount
(0.05 g) was added to all reaction bottles and solutions were agitated at 200 rpm for 20 min.
Fig. 2 shows the uptake of Ag, Co and Mn by the resin (removal efficiency) as a function of
pH value. It can be seen from Fig. 5.2 that the optimal uptake of metal ions occurred at pH 67 for Ag and Co and 4-7 for Mn. At high pH values, a decrease in removal efficiency was
observed. This decrease could be explained by the formation of metal ion precipitates. For
further investigations, pH 6 was selected.
79
Chapter five:
Kinetics and equilibrium studies
Fig. 5.2. Effect of pH on the adsorption of Ag, Co and Mn. Initial concentration of metal ions
10 mg L-1; amount of resin 0.05 g; sample volume 20 mL; temperature 298 K; stirring rate
200 rpm; stirring time 20 min; initial pH 4-10
5.3.3 Effect of Resin Amount
The amount of resin is an important parameter to attain the quantitative removal of metal
ions from sample solutions.
10
It was observed in Fig 5.3 that that the retention of the metal
ions increased with increasing amount of resin. This is because the number of available
adsorption sites increased with increasing the resin amount. Consequently, the removal
efficiency for Ag, Co and Mn ions is also increased.10,15,17 A mass of 0.05 g was selected for
further investigations.
80
Chapter five:
Kinetics and equilibrium studies
Fig. 5.3. Effect of resin dosage on removal of Ag(I), Co(II) and Mn(II) by Dowex 50W-x8
cation exchange resin: Initial concentration of meat ions 10 mg L-1; amount of resin 0.02-1.0
g; sample volume 20 mL; temperature 298 K; stirring rate 200 rpm; stirring time 20 min;
initial pH 6
5.3.4 Adsorption Isotherms and Comparison to Other Adsorbents
Adsorption isotherms were used to describe how analytes are distributed between sample
solution (liquid phase) and the resin (solid phase) when the ion exchange process reaches
equilibrium.17,18 The models were also used to describe the interaction of the analytes with
the sorbent. The four well-known isotherms, i.e. Langmuir, Freundlich Temkin and DubininRadushkevich were used to carry out the adsorption isotherm study. The Langmuir isotherm
model describes a homogeneous monolayer chemical adsorption process; while, Freundlich
isotherm model describes a heterogeneous physical adsorption process.18-20 Unlike Langmuir
and Freundlich models, Temkin isotherm model is based on the assumption that the free
energy of adsorption is a function of surface coverage due to the adsorbent-metal ion
interactions.21 Dubinin-Radushkevich isotherm on the other hand, is based on the
heterogeneous surface of the metal ion in order to distinguish between physisorption and
chemisorptions.22 The linearized equations of the isotherm models are represented by Eqs. (36).23-26
Langmuir equation:
81
Chapter five:
Kinetics and equilibrium studies
Ce
1
1
Ce 

qe q max
K L q max
(3)
where qe is the equilibrium adsorption capacity of ions on the adsorbent (mg g−1); Ce, the
equilibrium ions concentration in solution (mg L−1); qmax, the maximum capacity of the
adsorbent (mg g−1); and KL, the Langmuir adsorption constant (L mg−1).
Freundlich equation:
ln qe  ln K F 
1
ln Ce
n
(4)
where equilibrium capacity qe and Ce are defined as above, KF is the Freundlich constant (L
mg−1), and n is the heterogeneity factor.
Temkin equation:
qe 
RT
RT
ln K T 
ln Ce
bT
bT
(5)
where KT is Temkin constant representing adsorbent–adsorbate interactions, R is the gas
constant (8.314 J mol-1 K) , T is the temperature (K), bT is Temkin isotherm constant;
B
RT
is another constant related to adsorption heat.10
bT
Dubinin-Radushkevich equation:
ln qe  ln K DR  B 2
(6)
where KDR is the maximum adsorption capacity in mg g-1, B is a constant related to the
adsorption energy in mol2 kJ-2 and

  RT ln 1 

1
Ce

is a Polanyi potential calculated from Eq.(7).



(7)
where R is the gas constant (8.314 J mol-1 K) and T is the temperature (K).
The ion exchange of Ag, Co and Mn ions was carried out at different initial metal ion
concentrations and temperature (Fig. 5.4). The adsorption data was fitted to Langmuir,
Freundlich, Temkin and Dubinin and Radushkevich models. These isotherms relate the metal
ion uptake per unit weight of adsorbent qe to the equilibrium metal ion concentration in the
bulk solution Ce.17,27 The correlation coefficients (R2) were used to judge the fitness of the
isotherm equation.
82
Chapter five:
Kinetics and equilibrium studies
Fig. 5.4. Sorption isotherm of (a) Ag, (b) Co and (c) Mn on Dowex 50W-x8 resin: Initial
concentration of meat ions 10 to 300 mg L-1; amount of resin 0.05 g; sample volume 20 mL;
temperature, 293 to 313 K; shaking rate 200 rpm; shaking time 20 min; initial pH 6
The values of the parameters of the four isotherms and the correlation coefficients are
presented in Tables 5.2 and 5.3. It can be seen in Table 5.2 that the ion exchange adsorption
of Ag, Co and Mn onto Dowex 50W-x8 was better described by Langmuir model than
Freundlich model. This implied that Ag, Co and Mn in were chemically adsorbed on the
surface of Dowex 50W-x8 resin. It should be noted that there was no data available on
adsorption of Ag, Co and Mn in ethanol solutions using different cationic exchange resins.
Therefore, comparison of adsorption capacities of different resins for Ag, Co and Mn with
other method reported in literature was not performed.
83
Chapter five:
Kinetics and equilibrium studies
Table 5.2. Langmuir and Freundlich parameters for ion exchange adsorption of Ag, Co and
Mn onto Dowex 50W-x8 resin in ethanol
Cations
T(K)
293
298
303
308
313
293
298
303
308
313
293
298
303
308
313
Ag
Co
Mn
Langmuir
qmax(mg/g)
KL
43.8
5.65
47.8
7.59
51.3
9.29
56.0
11.3
61.7
15.7
48.1
5.54
52.0
8.10
55.3
10.06
57.3
13.86
63.5
19.2
55.4
5.37
58.5
9.20
61.7
11.46
65.4
13.2
71.6
16.4
2
R
0.9999
0.9996
0.9998
0.9999
0.9999
0.9999
0.9999
0.9997
0.9997
0.9999
0.9996
0.9996
0.9998
0.9999
0.9999
Kf
39.9
38.3
44.2
45.5
51.8
36.6
39.2
38.8
46.6
46.4
39.8
41.7
42.8
47.1
45.2
Freundlich
n
R2
7.64
0.8165
7.64
0.8280
7.24
0.9572
6.82
0.9595
5.85
0.7557
6.11
0.7457
6.01
0.7629
5.24
0.8495
5.33
0.9618
4.85
0.7559
6.61
0.7485
6.24
0.7091
5.89
0.9049
5.77
0.8245
6.63
0.7585
Langmuir isotherm can be represented in terms of a dimensionless constant separation
factor (RL). The latter is equal to the ratio of the unused adsorbent capacity to the maximum
adsorbent capacity and thus it can be a measurement of the adsorbent capacity used and the
affinity between the adsorbate and adsorbent.28 The dimensionless constant separation factor,
RL, value can be calculated from Eq. (8)
RL 
1
1  K L C0
(8)
where KL is the Langmuir constant and C0 is the highest initial concentration. The value of RL
indicates the type of the isotherm to be either unfavorable (RL > 1), linear (RL = 1), favorable
(0 < RL < 1), or irreversible (RL = 0).28,29 The RL values obtained were between 0 and 1
indicating that the adsorption of Ag, Co and Mn onto Dowex 50W-x8 resin is favorable.
The Temkin adsorption isotherm model was used to evaluate the adsorption potentials of
Dowex 50W-x8/metal ion interactions. It can be seen from Table 5.3 that values of KT
followed this trend Mn > Co > Ag at all temperatures. This means in cation exchange
processes using Dowex 50W-x8, adsorbent/metal ion interactions were more effective for Mn
84
Chapter five:
Kinetics and equilibrium studies
compared to Co and Ag. The lower adsorption potential for Ag+ could be due to their large
ionic radius.30
It can be seen in Table 5.2, the adsorption data fits well according to the Langmuir
isotherm model at all temperatures, however, it is insufficient to explain the chemical or
physical properties of the adsorption process. Therefore, the adsorption data was subjected to
the Dubinin-Radushkevich isotherm which is an analogue of Langmuir to determine the
nature of the adsorption process.29,31 The mean adsorption energy (E, kJ mol-1) calculated
from B value of Dubinin-Radushkevich isotherm using Eq. (9) provides information about
the physical or chemical properties.29,31
E
1
(9)
 2B
If the numerical values of E < 8 kJ mol-1, this means that the adsorption mechanism is
dominated by physisorption. If E values are in the range of 8–16 kJ mol-1, the adsorption
process follows chemisorption mechanism.22,31 The E value is independent of temperature but
it varies according to the nature of adsorbent and metal ion interactions.22 The DubininRadushkevich isotherm parameters are presented in Table 5.3. It can be seen from this table
that the values of E were greater than 8, this indicated that the adsorption of Ag, Co and Mn
onto Dowex 50W-x8 resin followed the chemisorption mechanism due to the formation of
complexes of sulfonate oxygen atoms with the metal ions. In addition, the maximum
adsorption capacity (KDR) increased with temperature.
85
Chapter five:
Kinetics and equilibrium studies
Table 5.3. Temkin and Dubinin-Radushkevich parameters for ion exchange adsorption of Ag,
Co and Mn onto Dowex 50W-x8 resin in ethanol
Cations
Ag
Co
Mn
T(K)
293
298
303
308
313
293
298
303
308
313
293
298
303
308
313
Temkin
Dubinin-Radushkevich
B
KT
R2
2.67
3.25
3.51
3.91
4.39
3.69
3.77
3.90
4.55
3.87
4.50
4.57
4.67
5.09
5.46
30.5
34.7
37.6
39.2
43.2
31.7
36.3
37.6
40.3
44.5
36.4
40.0
42.6
45.5
50.9
0.5530
0.6667
0.6982
0.7334
0.7359
0.6380
0.7198
0.7076
0.6864
0.7438
0.7332
0.7942
0.7661
0.7679
0.7459
kDR
(mol g-1)
47.5
50.6
53.9
54.9
55.7
50.1
54.4
57.1
55.6
60.2
57.1
60.5
61.1
64.5
70.0
B×108
(mol2J2)
-0.75
-0.71
-0.65
-0.58
-0.53
-0.68
-0.63
-0.60
-0. 57
-0.45
-0.76
-0.73
-0.64
-0.55
-0.49
E (kJ
mol-1)
8.16
8.39
8.77
9.28
9.71
8.57
8.91
9.13
9.37
10.5
8.11
8.33
8.83
9.52
10.1
R2
0.7606
0.7525
0.7952
0.7188
0.7143
0.8207
0.8094
0.8151
0.7963
0.7801
0.9206
0.9357
0.9370
0.9493
0.8820
The comparison of adsorption capacity of Dowex 50W-x8 for some of the studied metal
ions from organic and aqueous matrices with that of different adsorbents in literature
46-53
is
presented in Table 5.4. It can be seen from this Table that the maximum sorption capacities of
Dowex 50W-x8 are comparable, either slightly higher or lower than the adsorption capacities
of other adsorbents reported in the literature.
46-53
The variations in maximum adsorption
capacities might be due to the differences on the experimental conditions and the structure,
functional groups and porosity of the sorbent materials. 50
86
Chapter five:
Kinetics and equilibrium studies
Table 5.4. Comparison of maximum adsorption capacities of Dowex 50W-x8 for Ag, Co and
Mn with other adsorbents reported in literature
Adsorbents
MBT/SDSACMNPs
CSIS
SH–DETA–
PDBMPA
CSMO
MCR
ATZ-SSQ
ATZ-SG
AEPE-PS-MPs
MFT
T8-Pr-DPA
Dowex
50W-x8
resin
Ag
11.6
Adsorption capacities (mg g-1)
Co
Mn
-
References
Karimi et al.42
-
60.0
4.5
-
Monier et al.43
Yin et al.44
-
53.5
3.9
5.3
2.4
106.1
5.3
57.0
Monier et al.45
Guzel et al.46
Dias Filho et al.47
Dias Filho et al.47
Jainae et al.48)
Yirikoglu et al.49
Soares et al.50
Present work
47.5
60.1
47.2
AEPE-PS-MPs = 2-(3-(2-aminoethylthio)propylthio)ethanamine polystyrene-coated CoFe2O4 magnetic
particles; ATZ-SG = 3-amino-1,2,4-triazole-propyl modified silica gel; ATZ-SSQ = 3-(3-amino-1,2,4triazole)propyl]octasilsesquioxane; CSIS = Cross-linked magnetic chitosan-isatin Schiff’s base resin; CSMO =
Cross-linked magnetic chitosan–diacetylmonoxime Schiff’s base resin; MCR = Modified carrot residues; MFT=
Melamine-formaldehyde-thiourea
chelating
resin;
SH–DETA–PDBMPA
=
poly(diethylenetriamine
bis(methylene phosphonic acid)); T8-Pr-DPA= octakis[3-(2,2′-dipyridylamine)propyl] octasilsesquioxane
5.3.5 Adsorption Kinetics
Adsorption kinetics provides important information about the adsorption mechanism.10
The ion exchange adsorption kinetics of each metal ion with Dowex 50W-x8 were
investigated by two kinetic models: Lagergren pseudo-first-order (Eq. 10) and pseudosecond-order (Eq. 11).10,32,33
ln(qe  qt )  ln qe  k1t
(10)
t
1
1
 t
q1 qe
k 2 qe2
(11)
87
Chapter five:
Kinetics and equilibrium studies
where qt (mg g−1) is the adsorption capacity at time t (min); qe (mg g−1) is the adsorption
capacity at adsorption equilibrium; and k1 (min−1) and k2 (g mg−1 min−1) are the kinetic rate
constants for the pseudo-first-order and the pseudo-second-order models, respectively.
The applicability of each kinetic model was judged by correlation coefficient (R 2) as well
as the agreement between experimental and calculated qe values. The results in Table 5.5
indicated that the correlation coefficient values for pseudo-first order model were low (< 0.5)
and the experimental qe values do not agree with the ones calculated from the linear plots.
The correlation coefficients for the pseudo-second-order kinetic model were greater than 0.9
and the calculated qe values were in agreement with the experimental values (Table 5.5). The
results indicated the applicability of pseudo-second-order kinetic equation and the secondorder nature of the adsorption process of metal ion on Dowex 50W-x8 resin.18 Similar to
Langmuir isotherm model, pseudo-second- order kinetics assumes that the rate limiting step
may be chemisorptions involving valence forces through sharing and exchange of
electrons.15,1
Table 5.5. Kinetic parameters for the adsorption of Ag, Co and Mn onto Dowex 50W-x8 in
ethanol
Metal
ion
Ag
Co
Mn
qe, expt
(mg g-1)
qe, expt
(mg g-1)
47.4
52.6
57.0
Pseudo-first- order
k1 (min-1)
1.55×10-2
6.83×10-3
3.33×10-3
qe, cal
(mg g-1)
15.1
17.2
19.8
Pseudo-second- order
R2
0.3516
0.2222
0.2167
k2 (g mg−1
min−1)
1.95×10-2
5.63×10-2
3.26×10-2
qe, cal
(mg g-1)
47.1
52.7
57.2
R2
0.9992
0.9998
0.9999
Pseudo-second-order kinetic rate constant (k2) and adsorption capacity (qe) were used to
calculate the initial sorption rate (h), and the half-adsorption time (t1/2) given by Eqs. 12 and
13. It can be seen from Table 5.5 and 5.6 that pseudo-second-order rate constant and initial
sorption rate followed the order as Co > Mn > Ag. Half-adsorption time is defined as the time
required to remove half of the amount of the analyte at equilibrium29,34 and is considered as a
measure of the adsorption rate.29 It can be seen from Table 5.6 that the half adsorption times
for all metal ions were short (ranging 0.34-1.09 min). This indicated high affinity between the
adsorbent and metal ions.
88
Chapter five:
Kinetics and equilibrium studies
h  k 2 qe2
t1 
2
(12
1
k 2 qe
(13)
Table 5.6. Values of initial sorption rate (h) and half-adsorption time (t1/2)
Metal ions
h(mg g-1 min-1)
t 1 (min)
2
Ag
Co
Mn
43.2
156
118
1.09
0.34
0.54
The estimation of the rate limiting step is one of the most important factors to be
considered in the adsorption process.35,36 In a solid–liquid sorption process, the transfer of
analyte is normally characterized by external mass transfer, or intraparticle diffusion, or both
[35]. The most commonly used technique for finding the mechanism involved in the
adsorption process is intraparticle diffusion plot21,35-37 whereby adsorption capacity at time t
(qt) is plotted against the square root of time (Eq. 14).
qt  K id t
1
(14)
2
The plots of qt versus t1/2 for Ag, Co and Mn presented in Fig.5.5 showed that two steps
occurred in the ion exchange adsorption process. An initial steep-sloped portion at
intraparticle diffusion followed by the plateau at equilibrium.35,38 According to Rengaraj et
al.38, the first steep-sloped portion (0 to 1.5 min) was attributed to the external surface
adsorption or the instantaneous adsorption. The second portion (gentle sloped, 1.5 to 8 min)
was assigned to the gradual adsorption stage where intraparticle diffusion was rate-limiting.
These observations revealed that the intraparticle diffusion is rate-controlled and leads to a
plateau at equilibrium.35 The intraparticle diffusion rate constants were obtained from the
slope of the steep-sloped portion. The kid values were found to be 12.5, 7.84 and 4.12 mg g-1
h-1/2 for Ag, Co and Mn, respectively.
89
Chapter five:
Kinetics and equilibrium studies
Fig. 5.5. Intraparticle diffusion plots for adsorption of silver, copper and manganese
5.3.6 Adsorption Thermodynamics
The thermodynamic parameters such as change in standard free energy (ΔG0), enthalpy
(ΔH0) and entropy (ΔS0) were determined by using Eqs. (15) and (16).29
(15)
G   RT ln K L
ln K L 
S 0 H 0

R
RT
(16)
where R (8.314 J mol-1 K) is the gas constant, T (K) the absolute temperature and KL (L g-1)
is the Langmuir constant. The values ΔH0 and ΔS0 were estimated from the slopes and
intercepts of a graph of lnKL against T-1.
The results in Table 5.7 showed that ion exchange process of the metal ions was
endothermic; this was confirmed from the positive values of enthalpy change (ΔH0). The
possible explanation of endothermic nature of heats of adsorption (enthalpy) is that the metal
ions are well solvated. Therefore, in order for the metal ions to be adsorbed, they have to lose
part of their hydration sheath. The dehydration process of the metal ions requires energy. The
energy of dehydration replaced the exothermic nature of the metal ions getting attach to the
Dowex 50W-x8 surface.22,39,40 The positive ΔH0 can be interpreted based on the noticeably
strong interaction between metal ions and Dowex 50W-x8 surface. Furthermore, the positive
90
Chapter five:
Kinetics and equilibrium studies
enthalpy values support the earlier observation that the adsorption capacity of Dowex 50Wx8 resin for Ag, Co and Mn increased with increasing temperature due to the enhanced
mobility of analyte molecules.40
The positive value of ΔS0 revealed the affinity of metal ions towards the Dowex 50W-x8
resin.41 In addition, the positive value can originate from the redistribution of energy between
the metal ion and the adsorbent. For instance, before adsorption process takes place, the
heavy metal ions near the surface of Dowex 50W-x8 will be more ordered, and the ratio of
free metal ions to interact with the adsorbent will be higher than in the successive adsorbed
state. Thus, the distribution of rotational and translational energy among a small number of
molecules increases with increasing adsorption.28
It can be seen from Table 5.7 that the ΔG0 values decreased with increased temperature.
This indicated the spontaneity of the process at higher temperatures.22,41 Furthermore, the
negative values of ΔG0 confirmed the feasibility of the adsorption process.41
Table 5.7. Thermodynamic parameters for the adsorption of Ag, Co and Mn on Dowex 50Wx8
Cations ΔH (kJ mol-1)
Ag
Co
Mn
37.22
46.09
39.73
ΔS (J mol-1)
141.5
171.3
150.6
ΔG (kJ mol-1)
293 K
-4.240
-4.170
-4.095
298 K
-5.022
-5.183
-5.498
303 K
-5.615
-5.816
-6.144
308 K
-6.209
-6.732
-6.607
313 K
-7.165
-7.690
-7.229
5.3.7 Desorption Studies
Desorption and regeneration studies were carried out in different solutions. This was done
in order to test the possibility reusability of resin for various cycles and check out the longterm performance Dowex 50W-x8 resin. The desorption of Ag, Co and Mn was carried out
by eluting them from the metal ion-loaded resin with 3.0 mol L-1 HNO3 solution. It was
observed that up to 99% of the adsorbed metal ions can be recovered in 3.0 mol L-1 solution
of HNO3 and there was no recovery observed in double distilled deionised water indicating
the absence of physical bonding. After desorption with acid, the resin was regenerated by
washing it with 10 ml of double distilled deionised water followed by 10 ml of 2.0 M NaOH.
Regenerated Dowex 50w-x8 resin was effective for uptake of metal ions comparable to the
91
Chapter five:
Kinetics and equilibrium studies
fresh one over 50 cycles of adsorption/desorption. It should be noted that the amount of resin
used was 0.05 g. Therefore, it was concluded that the Dowex 50W-x8 can be reused for
several times without significantly decreasing its exchange capacities.
5.3.8 Analytical Performance and Application of the Proposed Method
The detection limits (LOD) of the proposed method for Ag, Co and Mn based on three
times the standard deviations of the blank (n = 3, n = 20) were 0.23, 0.18 and 0.10 µg L-1,
respectively. The precision expressed as relative standard deviations (RSD) values (n = 15),
was 2.5, 1.2, 1.5, for Ag, Co and Mn, respectively. Commercial samples of ethanol were
obtained from four different chemical suppliers. The real samples were analyzed before and
after of the treatment with Dowex 50W-x8 to determine the amount of Ag, Co and Mn
present, and the results found are presented in Table 5.8. To check accuracy of the method,
the metal ions adsorbed onto the resin were eluted with 3.0 HNO3 and the resulting solution
was analysed with ICP OES. The obtained results are also presented in Table 5.8. It can be
seen from this table that the Dowex 50W-x8 was very efficient in the removal of metal ions
in ethanol samples since it was able to remove more than 95% of the studied analytes. The
results obtained in this study indicated that due to the high cation exchange capacities of
Dowex 50W-x8 can be used to reduce the residual concentration of unwanted metal species
in ethanol and other organic solvents below the discharge limits. In addition, the regeneration
properties of Dowex 50W-x8 provide economical benefits. Moreover, the purpose of this
study was to develop sample preparation and pre-concentration of trace level analyses, and
not so much the toxic effects of the metals. In the introduction section it is indicated that
metal elements are undesirable in fuels and solvents used in sensitive applications such as
pharmaceuticals and high purity analytical reagents.
92
Chapter five:
Kinetics and equilibrium studies
Table 5.8. Concentration of Ag, Co and Mn (µg L-1) in commercial ethanol samples
Ag (µg L-1)
Samples
BA
AA
Co (µg L-1)
Mn (µg L-1)
AEa
BA
AA
EA
BA
AA
EA
1
1.91±0.03 0.07±1×10-3
1.83±0.01
2.56±0.05
ND
2.57±0.06
3.61±0.11
ND
3.59±0.12
2
1.38±0.01 0.03±2×10-3
1.33±0.03
5.33±0.13
0.04±2×10-3
5.28±0.18
1.42±0.05
ND
1.43±0.01
3
2.87±0.03 0.04±1×10-3
2.84±0.04
7.15±0.23
0.05±1×10-3
7.08±0.15
2.22±0.10
ND
2.21±0.03
4
1.56±0.02 0.06±3×10-3
1.51±0.03
6.37±0.37
0.02±0.1×10-3
6.34±0.36
12.31±0.41 0.03±1×10-3
BA= before adsorption; AA= after adsorption, EA= Eluted with acid; aComparative method using ICP OES
93
12.26±0.75
Chapter five:
Kinetics and equilibrium studies
5.4. CONCLUSIONS
This study investigated the ion exchange adsorption of Ag, Co and Mn from ethanol
using Dowex 50W-x8 cation exchange resin. The adsorption properties of Dowex 50W-x8
resin for removal metal ions in ethanol solution were investigated. The experimental data
fitted well to Langmuir Isotherm model. The maximum sorption exchange capacities at
298 K were 47.4 mg g-1, 52.6 mg g-1 and 57.0 mg g-1 for Ag, Co and Mn, respectively. The
data was also fitted to Temkin and Dubinin-Radushkevich adsorption isotherm models to
evaluate other adsorption properties. Based on Temkin isotherm, it was concluded that
adsorbent/metal ion interactions are stronger for Mn removal due to larger values of
Temkin constant (KT). The mean free energy values estimated from the DubininRadushkevich isotherm showed that the ion exchange adsorption followed the
chemisorption process. The studies showed that the ion exchange adsorption reaction
followed the pseudo-second-order reaction kinetics. The thermodynamic parameters (∆S0,
∆H0 and ∆G0) showed that the ion exchange adsorption process was spontaneous with the
endothermic nature. The overall experimental and theoretical results of the present study
proved that Dowex 50W-x8 cation exchange resin was suitable for ion exchange
adsorption of silver, cobalt and manganese from ethanol solution. The precision and
accuracy of the method were satisfactory. Therefore, Dowex 50W-x8 resin can be an
effective solid-phase material for preconcentration of trace metal ions in organic matrix.
5.5 REFERENCES
1. Grodowska, K. And Parczewski, A. 2010. Organic Solvents in the Pharmaceutical
Industry. Acta Poloniae Pharmaceutical-Drug Research, 67, 3-12.
2. Pritchard, J.D. 2007. Methanol–general information. CHAPD HQ, HPA. Available at
http://www.hpa.org.uk/webc/HPAwebFile/HPAweb_C/1194947361312 accessed 20
June 2011.
3. Gnansounou, E. & Dauriat, A. 2005. Ethanol fuel from biomass: a review. Journal of
Scientific Industrial Reearch, 60, 809-921.
4. Anton, R., Barlow, S., Boskou, D., Castle, L., Crebelli, R., Dekant, W., Engel, K.-H.,
Forsythe, S., Grunow, W., Larsen, J.C., Leclercq, C., Mennes, W., Milana, M.-R., Pratt,
I., Rietjens, I., Svensson, K., Tobback, P. & Toldrá, F. 2005. Opinion of the Scientific
panel on food additives, flavourings, processing aids and materials in contact with food
94
Chapter five:
Kinetics and equilibrium studies
on a request from the commission related to propan-2-ol as a carrier solvent for
flavourings. The EFSA Journal, 202, 1-10.
5. Hussain, S., Liba, A. & McCurdy E. 2011. Validating ICP-MS for the analysis of
elemental impurities according to Draft USP General Chapters 232 and 233. Available
at http://www.spectroscopyonline.com/spectroscopy/Articles/Validating-ICP-MS-forthe-Analysis-of-Elemental-Im/ArticleStandard/Article/detail/749104 accessed 14 May
2012.
6. Fliszar, K. A., Walker, D. & Allain, L. 2006. Profiling of metal ions leached from
pharmaceutical packaging materials PDA. Journal Pharmaceutical Science and
Technology, 60, 337-342.
7. Chaves, E. S., F. G. Lepri, J. S. A. Silva, D. P. C. De Quadros, T. D. Saint’pierre And A.
J. Curtius 2008. Determination of Co, Cu, Fe, Mn, Ni and V in diesel and biodiesel
samples by ETV-ICP-MS. Journal of Environmental Monitoring, 10, 1211-1216.
8. Santos, D. S. S., M.G. A. Korn, M. A. B. Guida, G. L. Dos Santos, V. A. Lemos And L.
S. G. Teixeira 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by
Sequential Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase
Extraction. Journal of Brazzilian Chemical Society, 22, 552-557.
9. Komjarova, I. & Blust, R. 2006. Comparison of liquid-liquid extraction, solid-phase
extraction and co-precipitation preconcentration methods for the determination of
cadmium, copper, nickel, lead and zinc in seawater. Analytica Chimica Acta, 576, 221228.
10. Alyüz, B. & Veli, S. 2009. Kinetics and equilibrium studies for the removal of nickel
and zinc from aqueous solutions by ion exchange resins. Journal of Hazardous
Materials, 167, 482-488.
11. Korn, M. D. G. A., De Andrade, J. B., De Jesus, D. S., Lemos, V. A., Bandeira, M. L.
S. F., Dos Santos, W. N. L., Bezerra, M. A., Amorim, F. A. C., Souza, A. S. &
Ferreira, S. L. C. 2006. Separation and preconcentration procedures for the
determination of lead using spectrometric techniques: A review. Talanta, 69, 16-24.
12. Lee, I. H., Kuan, Y.-C. & Chern, J.-M. 2007. Equilibrium and kinetics of heavy metal
ion exchange. Journal of the Chinese Institute of Chemical Engineers, 38, 71-84.
13. Pehlivan, E. & Altun, T. 2006. The study of various parameters affecting the ion
exchange of Cu2+, Zn2+, Ni2+, Cd2+, and Pb2+ from aqueous solution on Dowex
50W synthetic resin. Journal of Hazardous Materials, 134, 149-156.
14. Inglezakis, V. J. & Loizidou, M. D. 2007. Ion exchange of some heavy metal ions from
polar organicsolvents into zeolite. Desalination, 211, 238-248.
15. Liu, F., Li, L., Ling, P., Jing, X., Li, C., Li, A. & You, X. 2011. Interaction mechanism
of aqueous heavy metals onto a newly synthesized IDA-chelating resin: Isotherms,
thermodynamics and kinetics. Chemical Engineering Journal, 173, 106-114.
95
Chapter five:
Kinetics and equilibrium studies
16. Gode, F. & Pehlivan, E. 2003. A comparative study of two chelating ion-exchange
resins for the removal of chromium(III) from aqueous solution. Journal of Hazardous
Materials, 100, 231-243.
17. Edebali, S. & Pehlivan, E. 2010. Evaluation of Amberlite IRA96 and Dowex 1×8 ionexchange resins for the removal of Cr(VI) from aqueous solution. Chemical
Engineering Journal, 161, 161-166.
18. Hameed, B. H., Ahmad, A. A. & Aziz, N. 2007. Isotherms, kinetics and
thermodynamics of acid dye adsorption on activated palm ash. Chemical Engineering
Journal, 133, 195-203.
19. Pérez, N., Sánchez, M., Rincón, G. & Delgado, L. 2007. Study of the behavior of metal
adsorption in acid solutions on lignin using a comparison of different adsorption
isotherms. Latin American applied research, 37, 157-162.
20. Weber, W. J. 1972. Physicochemical processes for water quality control, WileyInterscience.
21. Chen, Z., Ma, W. & Han, M. 2008. Biosorption of nickel and copper onto treated alga
(Undaria pinnatifida): Application of isotherm and kinetic models. Journal of
Hazardous Materials, 155, 327-333.
22. Donat, R., Akdogan, A., Erdem, E. & Cetisli, H. 2005. Thermodynamics of Pb 2+ and
Ni2+ adsorption onto natural bentonite from aqueous solutions. Journal of Colloid and
Interface Science, 286, 43-52.
23. Langmuir, I. 1918. The adsorption of gases on plane surfaces of glass, mica and
platinum. Journal of American Chemical Society, 40, 1361–1403.
24. Freundlich, H. M. F. 1906. Über die adsorption in lösungen. Z. Phys. Chem. 57, 385470.
25. Temkin, M. J. V. & Pyzhev. 1940. Recent modifications to Langmuir isotherms, Acta
Physiochim. USSR, 12, 217-222.
26. Dubinin, M. M. & Radushkevich L.V. 1947. Equation of the characteristics curve of
activated charcoal. Chem. Zent. 1, 875.
27. Rengaraj, S., Joo, C. K., Kim, Y. & Yi, J. 2003. Kinetics of removal of chromium from
water and electronic process wastewater by ion exchange resins: 1200H, 1500H and
IRN97H. Journal of Hazardous Materials, 102, 257-275.
28. Shin, K.-Y. Hong, J.-Y. & Jang, J. 2011. Heavy metal ion adsorption behavior in
nitrogen-doped magnetic carbon nanoparticles: Isotherms and kinetic study. Journal of
Hazardous Materials, 190, 36–44.
29. Qu, R., Sun, C., Ma, F., Cui, Z., Zhang, Y., Sun, X., Ji, C., Wang, C. & Yin, P. 2012.
Adsorption kinetics and equilibrium of copper from ethanol fuel on silica-gel
functionalized with amino-terminated dendrimer-like polyamidoamine polymers.
Fuel, 92, 204-210.
96
Chapter five:
Kinetics and equilibrium studies
30. Horsfall, M. & Spiff, A. I. 2005. Equilibrium sorption study of Al3+, Co2+ and Ag+ in
aqueous solutions by fluted pumpkin (Telfairia Occidentalis HOOK f) waste biomass.
Acta Chimica of Slov. 52, 174–181.
31. Donia, A. M., Atia, A. A. & Abouzayed, F.I. 2012. Preparation and characterization of
nano-magnetic cellulose with fast kinetic properties towards the adsorption of some
metal ions. Chemical Engineering Journal, 191, 22–30.
32. Ho, Y. S. Mckay, G. 1999. Pseudo-second order model for sorption processes. Process
Biochemistry, 34, 451-465.
33. Ho, Y. S. 2006. Review of second-order models for adsorption systems. Journal of
Hazardous Materials, 136, 681–689.
34. Yu, Z., Qi, T., Qu, J., Wang, L. & Chu, J. 2009. Removal of Ca(II) and Mg(II) from
potassium chromate solution on Amberlite IRC 748 synthetic resin by ion exchange.
Journal of Hazardous Materials, 167, 406-412.
35. Rengaraj, S., Yeon, J.-W., Kim, Y., Jung, Y., Ha, Y.-K. & Kim, W.-H. 2007.
Adsorption characteristics of Cu(II) onto ion exchange resins 252H and 1500H:
Kinetics, isotherms and error analysis. Journal of Hazardous Materials, 143, 469-477.
36. Sarkar, M. P. Acharya, K. & Battacharya, B. 2003. Modeling the adsorption kinetics of
some priority organic pollutants in water from diffusion and activation energy
parameters. Journal of Colloid and Interface Science, 266, 28-32.
37. Atia, A. A., Donia, A. M. &Yousif, A. M. 2008. Synthesis of magnetic chelating resins
functionalized with tetraethylenepentamine for adsorption of molybdate anions from
aqueous solutions, Journal of Hazardous Materials, 155, 100–108.
38. Rengaraj, S. Kim, Y.C. Joo, K. Choi, K. & Yi, J. 2004. Batch adsorptive removal of
copper ions in aqueous solutions by ion exchange resins: 1200H and IRN97H. Korean
Journal Chemical Engineering, 21, 187-194.
39. Naseem, R. & Tahir, S.S. 2001. Removal of Pb (II) from aqueous/acidic solutions by
using bentonite as an adsorbent, Water Research, 35, 3982–3986.
40. Hefne, J. A., Mekhemer, W. K., Alandis, N. M., Aldayel, O. A. & Alajyan, T. 2008.
Kinetic and thermodynamic study of the adsorption of Pb (II) from aqueous solution to
the natural and treated bentonite. International Journal of Physical Science, 3, 281288.
41. Mittal, A. Kurup, L. & Mittal, J. 2007. Freundlich and Langmuir adsorption isotherms
and kinetics for the removal of Tartrazine from aqueous solutions using hen feathers,
Journal of Hazardous Materials, 146,243–248.
42. Karimi, M. A., Mohammadi, S. Z., Mohadesi, A., Hatefi-Mehrjardi, A., MazloumArdakani, M., Sotudehnia Korani, L. & Askarpour Kabir, A. 2011. Determination of
silver(I) by flame atomic absorption spectrometry after separation/preconcentration
using modified magnetite nanoparticles. Scientia Iranica, 18, 790-796.
97
Chapter five:
Kinetics and equilibrium studies
43. Monier, M., Ayad, D. M., Wei, Y. & Sarhan, A. A. 2010. Adsorption of Cu(II), Co(II),
and Ni(II) ions by modified magnetic chitosan chelating resin. Journal of Hazardous
Materials, 177, 962-970.
44. Yin, P., Tian, Y., Wang, Z., Qu, R., Liu, X., Xu, Q. & Tang, Q. 2011. Synthesis of
functionalized silica gel with poly(diethylenetriamine bis(methylene phosphonic acid))
and its adsorption properties of transition metal ions. Materials Chemistry and
Physics, 129, 168-175.
45. Monier, M., Ayad, D. M., Wei, Y. & Sarhan, A. A. 2010. Preparation and
characterization of magnetic chelating resin based on chitosan for adsorption of
Cu(II), Co(II), and Ni(II) ions. Reactive and Functional Polymers, 70, 257-266.
46. Güzel, F., Yakut, H. & Topal, G. 2008. Determination of kinetic and equilibrium
parameters of the batch adsorption of Mn(II), Co(II), Ni(II) and Cu(II) from aqueous
solution by black carrot (Daucus carota L.) residues. Journal of Hazardous Materials,
153, 1275-1287.
47. Dias Filho, N. L., Costa, R. M. & Marangoni, F. 2008. Adsorption of transition-metal
ions in ethanol solution by a nanomaterial based on modified silsesquioxane. Colloids
and Surfaces A: Physicochemical and Engineering Aspects, 317, 625-635.
48. Jainae, K., Sanuwong, K., Nuangjamnong, J., Sukpirom, N. & Unob, F. 2010.
Extraction and recovery of precious metal ions in wastewater by polystyrene-coated
magnetic particles functionalized with 2-(3-(2-aminoethylthio)propylthio)ethanamine.
Chemical Engineering Journal, 160, 586-593.
49. Yirikoglu, H. & Gülfen, M. 2008. Separation and Recovery of Silver(I) Ions from Base
Metal Ions by Melamine‐formaldehyde‐thiourea (MFT) Chelating Resin. Separation
Science and Technology, 43, 376-388.
50. Soares, I. V., Vieira, E. G., Do Carmo, D. R. & Dias Filho, N. L. 2013. Solid-phase
extraction of metal ions from fuel ethanol with a nanostructured adsorbent.
Microchemical Journal, 110, 120-126.
98
CHAPTER SIX:
PRE-CONCENTRATION OF TRACE ELEMENTS IN SHORT CHAIN ALCOHOLS
USING DIFFERENT COMMERCIAL CATION EXCHANGE RESINS PRIOR TO
INDUCTIVELY COUPLED PLASMA-OPTICAL EMISSION SPECTROMETRIC
DETECTION
ABSTRACT
Chelex-100, Dowex 50W-x8 and Dowex MAC-3 exchange resins were investigated for
separation and pre-concentration of trace amounts of Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn in
alcohols with respect to retention and desorption characteristics. Dowex 50W-x8 was found
to be the best sorbent with percentages recoveries >95%. In addition, Chelex-100 appeared to
be suitable for the pre-concentration of Cu, Fe and Zn, whereas Dowex MAC-3 was selective
for Cu and Fe. Therefore, Dowex 50W-x8 was used for further investigations. The relative
standard deviations <4% (n =20), limits of detection and quantification were 0.1-1.2 µg L-1
and 0.3-1.5 µg L-1, respectively. The SPE method was validated against a certified reference
material and the results were in agreement with certified values. The accuracy of the
optimized method was verified by the recovery test in the spiked alcohol samples. The
accuracy and spike recovery test for different metal ions were in the range 98-102% and 95105%, respectively. The optimised method was applied to the separation and preconcentration of metal ions in different commercial alcohol samples.
Keywords: Cation exchange resins, trace metals, separation, pre-concentration, alcohols,
Dowex 50W-x8
6.1 INTRODUCTION
Organic solvents such as alcohols, find a wide range of applications in pharmaceutical
industries, food industries and paint formulations, among others.1,2 For instance, methanol,
ethanol, iso-propanol and 2-butanol are used as the extractants in the production of medicine,
antibiotics and as components in food additives and food processing. In addition, these
liquids are used as fuel additives, antifreeze agents, solvents for gums, chemical
intermediates, solvents in chemical industries and denitrification agents in wastewater
treatment.2-4
The presence of trace metals in pharmaceutical products, fuel, food and paint is
undesirable. In pharmaceutical products, metal ions have an ability to decompose the material
99
Chapter six:
Preconcentration of trace elements in alcohols
of interest or catalyze the degradation of the active pharmaceutical ingredient. The
decomposition may lead to potential toxic effects.5,6 In addition, since some alcohols are used
as fuel additives, they may be responsible for the decomposition and poor performance of the
fuel, leading to corrosion of vehicle parts and formation of precipitates.7 Furthermore,
monitoring toxic elements is also important, since they are released into the atmosphere by
the fuel combustion, thus causing air pollution.8 Therefore, it is crucial to investigate the
levels of trace metals in alcohols because their presence in these products has negative health
effects and most metal impurities are detrimental to catalytic processes used in the industry.
Many metals occur naturally in fossil materials and, as a result, they can be present in
petroleum based products. The presence of the metals in alcohols can also be due to their
incorporation during the production process, by contact with refinement or distillation
equipment, storage and transport.8,9 Metal concentrations in organic solvents are generally in
trace levels, therefore sensitive and fast techniques with low detection limits are required.
The direct determination of trace elements in organic solvents by inductively coupled plasmaoptical emission spectrometry (ICP OES) poses a challenge with respect to the operating
parameters of the instrument.10,11 This is because direct loading of organic samples to the ICP
can destabilize or extinguish the plasma.12,13 Therefore, sample pretreatment prior to metal
ion determination is required.
Techniques involving sample pretreatment for the quantification of trace elements in
organic matrices have been reported in literature. These methods include microwave-assisted
acid digestion,14 microwave-induced combustion,15 conventional ashing and acid
dissolution16 and electrothermal vaporization.12 The limitation of conventional ashing and
acid dissolution methods is that they are time-consuming and volatile elements may be lost.14
Microwave digestion methods solve the problem of volatilization, but they increase the risk
of cross-contamination. Electrothermal vaporization is known to eliminate oxygen addition
and reduce the organic matrix interference. Nevertheless, its parameters have to be optimized
for each element thus lengthening the experimental procedure.14 Therefore, an accurate and
reliable analytical procedure based on simultaneous separation and preconcentration of
analytes prior to analysis in fuel samples, is required.
Separation and preconcentration techniques such solid phase extraction (SPE) has been
used for enrichment of heavy metal ions in both aqueous and organic matrices.8,17-24 The
advantages of using SPE method include high sensitivity, possibility of performing
simultaneous
preconcentration
step,
reduced
100
matrix
interferences,
reasonable
Chapter six:
Preconcentration of trace elements in alcohols
preconcentration factors with relatively rapid separation, reusability and low cost.23,24 In
addition, the use different solid phase materials can provide a better separation of
interferences, high efficiency and higher rate of process, and the possibility of combining
with different detection techniques.22 Various solid phase materials including Amberlite
XAD resins,8 modified silica gel17-19 biosorbent20 and ion exchange resins24 have been used
for the solid-phase extraction of traces heavy metals in aqueous and organic samples prior to
their instrumental analysis. Amberlite XAD resins and silica gel have been anchored by
organo-functional groups for metal ion binding and extraction from complex matrices. Due to
the leaching behaviour of organic samples such as ethanol, the main limitation of using a
modified sorbent for metal ions in organic matrices is the difficulty in maintaining the
organofunctional group attached to the solid phase.8
Therefore, in this work the possibility of using commercially available cation exchange
resins for separation and pre-concentration of metal ions in methanol, ethanol, iso-propanol
and 2-butanol, was investigated. The reason for choosing commercially available resins is
that they are the most commonly utilized cation exchangers for the removal of metal ions
from aqueous solutions and contain functional groups for metal ion binding and hence
effective extraction from organic phase matrices. In addition, procedures involving preconcentration of Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn in short chain alcohols (C1-C4) using
commercial cation exchange have not been reported in the literature. The influence of
experimental parameters on the retention of metal ions by the resins was studied and the
optimised procedure was applied to the determination of metal ions in commercial alcohol
samples.
6.2 EXPERIMENTAL
6.2.1 Apparatus
Analyte metal ions (Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn) were determined using a Spectro
Arcos 165 ICP OES (SPECTRO Analytical Instruments, GmbH, Germany) equipped with
Cetac ASX-520 autosampler. The operating conditions on the ICP OES spectrometer during
the measurements were as follows: forward power: 1400 W, plasma argon flow rate: 13 L
min-1, auxiliary argon flow rate: 2.00 L min-1, nebulizer argon flow rare: 0.95 L min-1. The
most prominent atomic and ionic analytical lines of metal ions were selected for
investigation, that is, Cd 228.802 nm, Cr 267.716 nm, Cu 324.754 nm, Fe 259.940 nm, Mn
101
Chapter six:
Preconcentration of trace elements in alcohols
294.921, Pb 220.353 nm, Ti 334.940, and Zn 213.856 nm. Solid phase extraction was carried
out in a VacMaster-24 sample SPE station (Supelco, PA, USA). The latter was used to
control the sample loading and elution flow rates.
Comparative experiments for the determination of metal ion were performed using
AAnalyst 400 Atomic Absorption Spectrometer (Perkin-Elmer, USA). Appropriate hollow
cathode lamps from Perkin-Elmer were used in these experiments. The analytical procedure
for electrothermal atomic absorption spectrometry (ETAAS) analysis reported by Anselmi et
al. [25], Reboucas et al. [26] and de Oliveira et al. [27] were modified in order to suit the
sample matrix.
6.2.2 Reagents and Solutions
All reagents were of analytical grade unless otherwise stated and double distilled
deionised water (Millipore, Bedford, MA, USA) was used throughout the experiments.
Absolute ethanol, methanol, iso-propanol (Merck, Darmstadt, Germany), and 2-butanol
(Sigma-Aldrich, St. Loius, MO, USA) were used to prepare model solutions. Spectrascan
stock solutions (1000 mg L-1) of Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn (Industrial Analytical Pty
Ltd, Johannesburg, South Africa) were used to prepare the working solutions for solid phase
extraction (SPE) at concentrations of 10 µg L-1 (Cd and Cr) and 12 µg L-1 for other metal
ions. Working solutions, as per the experimental requirements, were freshly prepared from
the stock solution for each experimental run. A Spectrascan multi-element standard solution
at a concentration of 100 mg L-1 (Industrial Analytical Pty Ltd, Johannesburg, South Africa)
was used to prepare working standard solutions at concentrations of 10-120 µg L-1 (Cd, Cr,
Cu, Fe, Mn, Pb. Ti and Zn) for measurements of concentrations of analytes in all model and
sample solutions. The certified reference materials, drinking water standard (CRM-TMDW500) obtained from High-Purity Standard Inc (Charleston, South Carolina) was used to
validate the SPE method. Conostan custom made multi-element oil standard used in the
experiment studies was obtained from SCP Science (Quebec, Canada). Solutions of nitric
acid at concentrations of 0.5, 1.0, 2.0, 3.0 and 4.0 mol L-1 were prepared from ultrapure
concentrated acid (65%, Sigma-Aldrich, St. Loius, MO, USA). These solutions were used for
the elution of the analytes from the column. The pH adjustments were performed with glacial
acetic acid (Merck, Darmstadt, Germany) and ammonia (Sigma-Aldrich, St. Loius, MO,
USA) solutions.
102
Chapter six:
Preconcentration of trace elements in alcohols
The cation exchanger sorbents used in this study as packing materials included Chelex100, Dowex 50w-x8 (sodium forms) and Dowex MAC-3 (hydrogen form) were purchased
from Sigma Aldrich (St. Loius, MO, USA).
6.2.3 Preparation of Column
Polyethylene columns of diameter 1.35 cm and 6.5 cm in height were used for preconcentration. Slurries of 1.5 g of Dowex 50W-x8, Dowex MAC-3 and Chelex 100 resin in
double distilled deionised water were prepared and the columns were packed to a height of 3
cm. A porous frit was placed at the bottom of the column and at the top of the packing
material to hold and confine the adsorbent within the designated capacity/volume. The
entrapment of the packing material serves to eliminate dead volume. The columns were
washed with double distilled deionised water followed by conditioning with 10 mL
ammonium acetate buffer (1.0 M, pH 9.0) and then 10 mL of appropriate organic solvent.
The latter was to ensure that the organic media was not mixed with aqueous solution,
particularly at the metal retention step of the preconcentration process. This was also done to
ensure that there is no mixing of organic and aqueous phases, in case this would affect the
retention mechanism of the column.
6.2.4 Preconcentration and Recovery of the Metal Ions in Model Organic Solution
Model metal ion solutions were prepared as follows: 1.0 mL of 1.0 mg L-1 Cd and Cr
solutions was separately transferred into 100 mL volumetric flasks and made up to the mark
with ethanol to obtain 10 µg L-1 concentration of each metal ion. The procedure was repeated
for Cu, Fe, Mn, Pb, Ti, V and Zn to obtain 12 µg L-1. Ethanol solutions of each metal ion (20
mL) were percolated through the ion exchange resin column at a flow rate of 3.0 mL min-1. It
should be noted that the sample volume of 20 mL was employed only in the initial
optimization experiments. For the real samples, a volume of 100 mL was employed. The
column was washed with 10 mL of double distilled deionised water to remove excess alcohol
solution, followed by 5.0 mL of ammonium acetate buffer solution (1.0 mol L-1, pH 9) to
remove major cations (Na, Ca, K, etc). It should be noted that the washings with double
distilled deionised water and ammonium acetate buffer solution were discarded. Lastly the
metal ions were eluted with 5 mL of 3.0 mol L-1 HNO3 solution. All fractions obtained during
the elution stage were collected separately and analysed by ICP OES. The same procedure
103
Chapter six:
Preconcentration of trace elements in alcohols
was applied to the blank solutions. After each use, the resin in the column was washed with
20 mL of double distilled deionised water followed by 10 ml of 1.0 M NaOH (this was done
in order to keep the resin in sodium form) and stored for the next experiment. The effect of
sample pH, sample volume, eluent concentration, sample and eluent flow rates were
investigated. It should be noted that the procedure for pH measurement in alcoholic medium
was similar to that of aqueous medium and that there was no damage of the pH meter probe
when using alcoholic samples. The pH meter was calibrated after each pH measurement to
check if there were changes on the calibration. All analyses were performed in triplicate.
6.2.5 Effect of Matrix Ions Interferences
The effect of potential interfering ions on the determination of Cd, Cr, Cu, Fe, Mn, Pb, Ti
and Zn was also investigated by known amounts of cations (K+, Na+, Ca2+, Mg2+, Ni2+, Co2+,
Al3+ and Ag+ ) were added to the ethanol solution. The concentration of the studied metal
ions was fixed at 20 µg L-1 of and the general preconcentration procedure was applied. The
concentration of interfering ions ranged from 2 to 1000 mg L-1.
6.2.6 Procedure for the Dilution of Conostan Custom Made Multi-Element Oil Standard
The trace element forms in petroleum products such as gasoline is not fully known, such
that various species may display different sorption behaviours.8 Therefore, an Conostan
custom made multi-element oil standard obtained from SCP Science (Quebec, Canada)
containing 1.0 mg L-1 of each metal ions, was used to study the resins sorption efficiency for
different metal species. The dilution of the organic standard was performed as follows: a 1.0
mL aliquot of 1.0 mg L-1 Conostan custom made multi-element oil standard dissolved in 10
mL of hexane. The solution was quantitatively transferred in to 100 mL volumetric flask and
made to the mark with acetone to obtain 10 µg L-1 of each metal ion. Suitable aliquots (20
mL) of the solution were taken and pre-concentrated by the proposed procedure and analysed
with ICP OES. The same procedure was applied to the preparation of the blank solutions.
6.3 RESULTS AND DISCUSSION
The trace metal pre-concentration method described in this work was optimized in order
to determine the best retention/elution conditions with good sensitivity (highest slope) and
precision (%RSD <5%).28 To obtain these conditions, preliminary tests were performed to
104
Chapter six:
Preconcentration of trace elements in alcohols
investigate factors that exert significant influence on the retention of the analytes by cation
exchange resin. The factors selected include pH, eluent concentration as well as sample flow
rate. The type of eluent was HNO3 and the mass of 1.5 g the sorbent was used. The
percentage recoveries (%R) were calculated by relating the obtained (final) concentration of
the analytes after pre-concentration to the original (initial) concentration of the metal ion in
the model solution.
6.3.1 Effect of Sample Solution pH on Retention of Metal Ions
In solid-phase extraction, one of the most important parameters for obtaining quantitative
retention of trace elements is sample solution pH.29 This is because the retention of metal ion
by of the adsorbent is highly dependent on pH.20 In previous studies,28 pH of the sample
solution equal to 6.0 was found to be suitable for the retention of metal ions onto Chelex-100.
Therefore, the preliminary investigation on influence of sample solution on the retention
efficiency of Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn onto Dowex 50W-x 8 resins was examined in
the pH range of 4.0-10. The model organic solutions of each metal ion were adjusted to the
desired pH using glacial acetic acid or 1.0 mol L-1 ammonium hydroxide solution. The
experiments were carried out by passing 20 mL of 10 µg L-1 for Cd and Cr; 12 µg L-1 for Cu,
Fe, Mn, Pb, Ti and Zn through Dowex 50W-x8 resin column while keeping the flow rate at a
constant value 3.0 mL min-1. The retained metal ions were stripped from the column with, 5.0
mL of 3.0 mol L-1 HNO3 at a flow rate of 3.0 mL min-1. The recoveries of the metal ions are
presented in Fig. 6.1. The same method was repeated for Dowex MAC-3. However, an eluent
concentration of 2.0 mol L-1 (instead of 3.0 mol L-1) was used. Even though the recoveries of
some of the metal ions were below 70%, the results follow the same trend with maxima at pH
6. The optimum pH for quantitative recoveries of the analytes from the resin was 6.
Therefore, all subsequent studies were carried out at pH 6.
105
Chapter six:
Preconcentration of trace elements in alcohols
Fig. 6.1. Effect of sample pH on retention of the analytes in ethanol onto Dowex 50W-x8
resin column: pH 6; analyte concentration 12 µg L-1; amount of resin 1.5 g; flow rates of
sample and eluent 3.0 mL min−1; eluent volume 5 mL; replicates n = 3
6.3.2 Effect of Desorption Solution Concentration
The desorption of the analytes bound onto the surface of the Dowex 50W-x8 cation
exchange resin is achieved by proton exchange in the acid solution.30 Various concentrations
of HNO3 were evaluated for stripping of the retained metal ions from Dowex 50w-x8 resin.
In this work, the best eluent concentration was defined as the concentration of the eluent that
eluted more than 95% of the retained metal ions. The results in Fig. 6.2 indicates that the
recoveries of Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn increased with increasing concentration of
the eluent up to 3.0 mol L-1. It is noteworthy to point out that metal ions appeared to be bound
by Dowex 50W-x8 strongly such that eluent concentration less than 3.0 mol L-1 were not
suitable for quantitative elution except for copper. It was also observed that at concentrations
higher than the optimum eluent concentration, the recoveries decreased. This may possibly be
explained by the oxidizing property of nitric acidic at high concentrations. The latter may
oxidize some of the metals to different oxidation states thus altering their speciation and
hence column retention behavior. Similar phenomena were observed in literature.24,30-34
Therefore, for further applications 3.0 mol L-1 HNO3 was selected for elution of metal ion
from Dowex 50w-x8 resin.
106
Chapter six:
Preconcentration of trace elements in alcohols
Fig. 6.2. Influences of the eluent concentration on the recoveries of the analytes on Dowex
50W-x8 resin column: pH 6; analyte concentration 12 µg L-1; amount of resin 1.5 g; flow
rates of sample and eluent 3.0 mL min−1; eluent volume 5 mL; replicates n = 3
6.3.3 Effect of Sample Flow Rate
The optimization of the sample flow rate was carried out to ensure the quantitative
retention of the analytes of interest. The effect of sample flow rate of the sample (ethanol)
solution on the retention of metal ions on the Dowex 50w-x8 resin was done in a column
packed with 1.5 g of resin. Sample solutions were passed through the column at various flow
rates (1.0-5.0 mL min-1). Flow rates less than 1.0 mL min-1 were not studied so as to avoid
long analysis time periods. The optimum flow rate for this work was defined as the rate of
flow of the sample solution through the column that gave at least 95% retention of metal ions.
The studies showed that the optimum flow rate for quantitative sorption of metal ions onto
the resin was between 1.0 and 3.0 mL min-1. Flow rates greater than 3.0 mL min-1 caused a
gradual decrease in sorption due to insufficient contact time between the resin and the metal
ions. Therefore, 3.0 mL min-1 flow rate was chosen as the optimum flow rate for sample
loading.
107
Chapter six:
Preconcentration of trace elements in alcohols
6.3.4 Effect of Sample Volume
Using large sample volumes at a defined flow rate improves the pre-concentration factor
of the SPE method.31 In order to investigate the possibility of enriching low concentrations of
analytes from large volumes, the maximum applicable sample volume must be determined.
For this purpose, the concentrations of each metal ion were kept constant while increasing the
sample volume. The effect of sample volume was investigated by passing 5 to 700 mL of
ethanol model solution containing fixed amount of analytes (12 µg L-1) through Dowex 50Wx8 resin column under optimum conditions. The recoveries of the analytes (Fig. 6.3) were
quantitative (≥ 95%) for all analyte ions in the sample volume ranging 5–500 mL.
Fig. 6.3. Effect of sample volume on the recoveries of metal ions: pH 6; analyte concentration
12 µg L-1; amount of resin 1.5 g; flow rates of sample and eluent 3.0 mL min−1; eluent
volume 5 mL; replicates n = 3
At volumes higher than 500 mL, the recoveries for the metal ions decreased, possibly due
to the excess metal ions loaded over the column capacity which results in the saturation of the
active sites. The pre-concentration factor, which has been defined as the ratio of the sample
volume loaded onto the column to that of the eluent volume used for stripping of the retained
metal ions,35 was found to be a maximum of 100 when a volume of 500 mL was used.
However, in the actual experiment a compromise sample volume of 100 mL was used for
108
Chapter six:
Preconcentration of trace elements in alcohols
optimization of analytical parameters as well as in the real sample analysis. This was done in
order to speed up the analysis time. Even though the recoveries were quantitative, as the
volume increases the stripped metal concentration increases such that analytical techniques
with high limits of detection like FAAS can be used.
6.3.5 Preconcentration of Multi-Element Using Different Sorbent Materials
The efficiency of cation exchange resins for pre-concentration of multi-elements
(concentration of each analyte equal to 12 µg L-1) in ethanol was investigated under optimum
conditions. The results indicated that the highest retention of the analytes from ethanol model
solutions was observed on Dowex 50W-x8 resin (Table 6.1). This might be due to the larger
exchange capacity (1.7 meq mL-1) and its functional groups (sulfonic acid). The recoveries of
metal ions from Dowex 50W-x8 ranged from 95 to 101%. It can be concluded that the
affinity of analytes towards Dowex 50W-x8 is very similar and therefore, they could be preconcentrated with the same efficiency.36
The experimental conditions for Dowex 50W-x8 were tried on Chelex 100. However, the
quantitative retentions of metal ions under these conditions were found to be < 70% for Cd,
Cr Fe, Mn, Pb and Ti; and < 90% for Cu and Zn. This was due to the fact that, the
equilibration time was not sufficient enough to allow an effective retention of the metal ions.
Therefore, the flow rate of the sample solution was reduced to 1.0 mL min-1. When the latter
was applied, an increase in the metal ion recoveries was noticed. For instance, the recoveries
for Cu, Fe and Zn were ≥95%. In addition, ≥ 40% increase was observed for Cd, Cr, Mn, Pb
and Ti recoveries (Table 6.1, Chelex-100 at 1.0 mL min-1 designated as Che2). The results
indicated that Chelex-100 was suitable for removal of Cu, Fe and Zn at a flow rate of 1.0 mL
min-1 rather than 3.0 mL min-1. The rest of the metals were not quantitatively recovered at
this flow rate. It was therefore predicted that the retention of metal ions onto Chelex-100 is
governed by slow kinetics. It was then concluded that Chelex-100 was not suitable for preconcentration of multi-element in organic matrices, for the list of metals studied.
Dowex MAC-3 resin, which contains carboxylic acid as a functional group, revealed
great selectivity towards Cu and Fe (Table 6.1). The retention of Cd, Cr, Mn, Pb, Ti and Zn in
the multi-element solution was poor. This was confirmed by the lower recoveries. The lower
retention of the above metal ions onto Dowex MAC-3 might be due to relatively lower
affinity for the sorbent compared to Cu and Fe. This suggests that, although this resin had
109
Chapter six:
Preconcentration of trace elements in alcohols
larger exchange capacity (3.8 meq mL-1) but because of its large particle size (12-50 mesh),
the overall retention characteristics were somehow poor. This may possibly be explained by
relatively smaller surface area as a result of large particle size.
Table 6.1. Recovery (%) of multi-element in ethanol using Dowex 50W-x8 (Dow(a)),
Chelex-100 (Che1) and Dowex MAC-3 (Dow(b)) for SPE methods.
Resins
Recovery (%)
Cd
Cr
Cu
Fe
Mn
Pb
Ti
Zn
Dow(a) 99.2±1.4 97.4±1.3
101±1.2
99.3±1.2 97.9±4.2 96.4±1.4 95.1±1.2 97.9±2.1
Dow(b) 53.4±2.1 38.2±3.2
98.6±2.4
99.3±2.1 56.7±2.1 75.8±1.4 63.2±1.2 82.6±1.2
Che1
40.8±1.4 39.3±1.2 88.2.4±1.2 68.5±1.2 56.2±1.3 36.5±1.3 56.3±1.4 88.9±2.4
Che2
88.9±1.2 80.6±3.8
96.1±4.0
95.8±2.4 87.5±2.4 78.1±1.2 91.0±1.2 96.5±3.8
Experimental conditions: sample volume; 20 mL; amount of resin 1.5 g; flow rates of sample
and eluent: 3.0 mL min−1; eluent volume 5 mL; replicates = 3; Chelex 100 (Chen2, 1.0 mL
min−1)
In comparison to Dowex MAC-3 and Chelex-100 resins, the Dowex 50W-x8 with a mesh
particle size of 100-200, had the most uniform particle distribution that supported the best
flow of solutions through SPE columns. In addition, the fine particle size of Dowex 50W-x8
resin reduced the equilibration time required for the adsorption of metal ions as compared to
other resins. For these reasons, Dowex 50W-x8 resin was selected for pre-concentration of
multi-elements in different alcohols.
The capabilities of Dowex 50W-x8 resin to retain metal ions from methanol, ethanol, isopropanol and 2-butanol were investigated under optimum conditions. The results obtained are
presented in Fig. 6.4. Dowex 50W-x8 was found to be suitable for sample cleanup for all the
metal ions as percentage recoveries were all ≥95%. It can be seen from Fig.6.4 that metal
ions had different recoveries in various alcohols. For instance, the recovery of Mn was the
highest (≥100%) in methanol and iso-propanol matrices, whereas Cu showed the highest
percentage recoveries (>99%) in ethanol and 2-butanol matrix and Zn had the highest %
recoveries (>99) in ethanol, iso-propanol and 2-butanol matrix. In addition, Cd, Cr, Fe, Pb,
and Ti had the highest recoveries in ethanol, iso-propanol, 2-butanol and methanol,
respectively. The variation in the uptake of the analytes by the cation exchange resin in
110
Chapter six:
Preconcentration of trace elements in alcohols
organic solvents might be attributed to the differences in dielectric constants of the alcohol
solutions as well as the size, charge and polarizability of the metal ions.37
Fig. 6.4. Pre-concentration of metal ions from methanol, ethanol, iso-propanol and 2-butanol
Experimental conditions: pH 6; analyte concentration 12 µg L-1; amount of resin 1.5 g; flow
rates of sample and eluent 3.0 mL min−1; eluent volume 5 mL; replicates n = 3
The sorption efficiency of Dowex 50W-x8 resin was also investigated by
preconcentrating metallo-organic Conostan standard solutions. Table 6.2 shows the recovery
results obtained from the preconcentration and determination of metal ions in diluted metalloorganic Conostan standard solutions. The results obtained were compared by evaluating the
percentage recovery for each metal ion. It should be noted that the percentage recovery was
evaluated with respect to the Conostan certified value. As it can be seen in Table 6.2, the
determined concentration values were not significantly different from the certified values at
95% confidence level and the percentage recoveries ranged from 98.3-102%. These results
imply that the Dowex 50W-x8 SPE system can be used for the preconcentration of trace
elements in their inorganic or metal-organic forms.
111
Chapter six:
Preconcentration of trace elements in alcohols
Table 6.2. Analysis of the metallo-organic Conostan standard for the determination of
analytes after application of the pre-concentration procedure; RSD= relative standard
deviation.
Concentration (µg L-1)
Elements
Cd
Cr
Cu
Fe
Mn
Pb
Ti
Zn
Certified
RSD (%)
Found (n=3)
RSD (%)
Recovery (%)
1000
1000
1000
1000
1000
1000
1000
1000
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
983.3
991.7
1016
991.7
983.3
983.3
1008
991.7
1.5
1.5
1.4
1.5
1.5
1.5
1.4
1.5
98.3
99.2
102
99.2
98.3
98.3
101
99.2
6.3.6 Effect of Matrix Ions Interferences
The effect of potential interfering ions was also investigated. The tolerance limit was set
as the concentration of the ion required to cause ≤5% error. The results are presented in Table
6.3. It was found that ions normally present in ethanol samples did not interfere with the
recoveries of the analytes. This suggests that the Dowex 50W-x8 preconcentration method
can be applied for the removal of trace amounts of the studied metal ions in alcohol samples
in the presence higher concentration of other cations and anions.
112
Chapter six:
Preconcentration of trace elements in alcohols
Table 6.3. Effect of potential interfering ions on the percentage recoveries of Cd, Cr, Cu, Fe, Mn; Pb, Ti and Zn
Ions
K+
Na+
Ca2+
Mg2+
Ni2+
Co2+
Al3+
Ag+
[Interfering ion] (mg L-1)
1000
1000
1000
1000
5
5
50
2
Cd
98.1±1.3
99.5±1.5
99.0±0.9
98.5±2.5
96.0±2.2
98.2±1.1
98.3±1.7
99.4±1.7
Cr
99.3±1.2
98.7±1.4
100±1.1
97.0±1.8
99.8±0.3
95.9±3.1
97.6±2.1
96.3±0.5
Cu
99.5±0.6
98.9±1.2
97.9±0.9
98.3±1.1
95.9±3.1
95.3±2.7
98.5±1.2
99.4±0.5
113
Fe
97.3±1.6
98.3±1.9
97.9±1.3
98.4±1.7
95.9±2.6
97.8±2.2
98.4±1.8
97.4±2.5
Mn
99.3±0.8
98.9±2.1
99.6±0.5
98.9±0.7
96.1±2.4
96.1±1.4
97.8±1.9
98.9±2.4
Pb
98.3±1.2
98.9±2.1
97.8±2.5
99.0±2.7
95.3±2.8
96.0±2.6
97.7±1.2
96.9±1.0
Ti
100±2.1
99.2±1.1
98.9±2.3
99.3±0.9
100±0.2
98.1±1.1
98.6±2.1
97.1±3.8
Zn
99.4±0.5
98.1±1.1
98.3±1.6
98.9±1.9
96.8±2.3
97.8±1.5
99.4±0.4
98.8±1.1
Chapter six:
Preconcentration of trace elements in alcohols
6.3.7 Analytical Parameters
The analytical performance of the Dowex 50w-x8 SPE method under optimum conditions
for pre-concentration of Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn in ethanol was evaluated. The
linearity of the methods was studied by pre-concentrating 100 mL of ethanol spiked with
multi-element standard at a concentration range 0.005 to 150 µg L-1. A linear range 0.2 to
2400 µg L-1 after pre-concentration was achieved for all the investigated analytes. The
sensitivity of the pre-concentration system was defined as the gradient (slope) of the
calibration graph. The results in Table 6.4 indicated that the SPE method was more sensitive
to Ti, Cu, Mn and Zn compared to the rest of metal ions. Thus the highest slope obtained was
103.7 L µg-1 for Ti while the lowest was 14.4 L µg-1 for Cr.
The IUPAC limit of detection (LOD) and limit of quantification (LOQ) under optimized
conditions were obtained from the signals of 16 successive measurements of the blank and
the slope (m) of the calibration curve. The LOD was defined as the lowest concentration of an
analyte giving signals equal to three times the standard deviation (3SD) of blank signal
divided by the slope of the calibration curve that the analytical technique can detect (3SD/m).
The LOQ, on the other, was defined as the to the smallest concentration of an analyte giving
signals equal to ten times the standard deviation of blank signal divided by the slope of the
calibration curve which can be accurately and precisely measured with an analytical
procedure (10SD/m). For 100.0 mL of sample solution used, the calculated LOD and LOQ
are presented in Table 6.4.
Table 6.4. Analytical performances for the proposed Dowex 50W-x8 SPE method
Cations
Slope
R2
LOD
IDL
LOQ
21.0
0.9978
0.4
0.1
1.3
Cd
14.4
0.9979
0.4
0.2
1.2
Cr
79.0
0.9990
0.1
0.4
0.4
Cu
23.8
0.9975
0.5
0.1
1.5
Fe
60.7
0.9990
0.1
0.1
0.5
Mn
30.8
0.9987
0.3
1.0
0.9
Pb
103.7
0.9976
0.1
0.4
0.3
Ti
57.8
0.9994
0.1
0.2
0.5
Zn
-1
NB. The units for LOD, IDL and LOQ are in µg L while the slope is in cps L µg-1
114
%RSD
2.3
3.2
1.4
1.7
2.1
1.2
3.3
1.8
Chapter six:
Preconcentration of trace elements in alcohols
The precision (reproducibility) of the SPE method was studied by performing twenty
successive measurements at a concentration level of 10 µg L-1 of multi-element organic
solution (containing Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn). The overall reproducibility of preconcentration procedure expressed in terms of relative standard deviation (%RSD) was
reasonably good (<5%) shown in Table 6.4.
The analytical parameters (LOD and %RSD) obtained with the Dowex 50w-x8 SPE
method were compared with ones reported in literature. The LOD and %RSD values were
similar to or even better than those obtained by de Oliveira et al.27 applying direct
determination using ETAAS. For Cu Fe and Ni, the LOD values were better than those
obtained by Santos et al.,8 Roldan et al.,19 Teixeira et al.38 and Teixeira et al.40 For Pb, the
LOD value obtained by the proposed method were lower than those obtained by Santos et al.8
and Santos et al.39 using ETAAS and SPE/FS-AAS, respectively. In addition, the %RSD
obtained in this study were better than those reported by Santos et al.8, Teixeira et al.17 and
Teixeira et al.38 For Cu and Fe, the LOD and RSD value obtained by the proposed method
were similar those obtained by Rocha et al.41 applying ICP OES direct sample introduction
by an ultrasonic nebulizer and membrane desolvator. These analytical parameters
demonstrated that the Dowex 50w-x8 SPE for separation and preconcentration of metal ions
in alcohols is better than in previously published methods. Moreover, comparing with the
published methods, the proposed method is broader in terms of the number of investigated
analytes; thus, allowing the simultaneous preconcentration and determination of all target
metal ions in different alcohol samples.
6.3.8 The Effect of Column Regeneration
The regeneration of the column is one of the important parameters in evaluating the
cation exchange resin material. In order to investigate the stability and recyclability of Dowex
50W-x8 column, successive retention and elution cycles were performed by passing 20 mL of
ethanol solutions (containing Cu, Mn and Ti) through the column. The stability and
regeneration of Dowex 50W-x8 column were evaluated by monitoring the changes in the
recoveries of Cu, Mn and Ti through 150 retention-elution cycles. The Dowex 50W-x8
column can be reused after regeneration with 20 mL double distilled deionised water and 10
mL of 1.0 mol L-1 NaOH, respectively. The column was found to be stable up to 100
115
Chapter six:
Preconcentration of trace elements in alcohols
retention/elution cycles without observable decrease in the recoveries of metal ions (>95%).
Therefore, repeated use of the resin is possible.
6.3.9 Accuracy and Validation of the Proposed Separation and Pre-Concentration
Procedure
The accuracy of the present method was tested by performing the spike recovery tests in
methanol, ethanol, iso-propanol and 2-butanol. Known amounts of each metal ion (5 and 10
µg L-1) were added to methanol, ethanol, iso-propanol and 2-butanol samples. The results
obtained are shown in Table 6.5. It can be seen in Table 6.5 that there is good agreement
between the added and recovered analyte concentration. The percentage recoveries of
analytes ranged from 95 to 105% and the results showed that the different organic sample
matrixes did not affect the recovery of the trace metals. In addition, other metal ions (such as
Cd, Mn, Pb and Ti) were present in minor concentration in the analyzed samples and can be
retained by the resin. Therefore, it can be concluded that Dowex 50W-x8 SPE is a suitable
method for separation and pre-concentration of trace metal ions in organic solvents.
116
Chapter six:
Preconcentration of trace elements in alcohols
Table 6.5. Accuracy test results for spiked recovery (R): pH 6, sample volume: 100 mL, n =3.
Element
-1
Added (µg L )
0
Cd
5
10
0
Cr
5
10
0
Cu
5
10
0
Fe
5
10
0
Mn
5
10
0
Pb
5
10
0
Ti
5
10
0
Zn
5
10
a
ND= not detected
Methanol
Found
R (%)
a
ND
4.9±0.3
97.8±1.2
9.8±0.1
98.4±2.1
5.3±0.1
10.1±0.3
97.0±1.5
15.1±0.2
98.7±1.4
12.3±0.5
17.2±0.6
99.2±1.1
22.1±0.6
98.9±0.9
41.4±0.3
46.3±0.7
98.6±2.4
51.2±0.4
99.1±1.3
1.4±0.4
6.4±0.2
99.2±1.6
11.3±0.5
98.7±2.1
3.3±0.7
8.1±0.2
97.8±1.9
13.1±0.4
98.6±1.2
1.5±0.1
6.4±0.4
98.4±1.7
11.4±0.5
98.7±2.4
21.4±0.6
26.4±0.6
99.7±1.1
31.3±0.2
99.3±1.4
Ethanol
Found (µg L-1)
R (%)
2.3±0.2
7.1±0.7
95.0±3.1
12.1±0.1
97.5±1.1
7.2±0.3
12.0±0.2
96.6±0.6
17.0±0.9
98.3±2.1
18.3±0.3
23.0±0.2
95.0±1.2
28.8±0.1
105±1.6
38.4±0.4
43.2±1.0
96.0±2.1
48.0±0.8
96.0±1.8
3.7±0.6
8.6±1.2
98.0±1.4
13.4±1.4
97.5±1.2
2.9±0.6
7.7±0.7
95.2±1.6
12.4±1.1
95.0±2.1
4.0±0.2
8.8±0.1
95.0±3.4
13.5±0.6
95.0±3.8
21.3±1.3
26.4±1.4
102±2.1
31.8±1.8
105±1.4
117
Iso-propanol
Found (µg L-1)
R (%)
ND
4.9±0.3
98.4±1.2
9.9±0.9
99.2±2.1
7.1±0.1
12.1±1.0
99.6±1.8
16.8±0.4
96.8±1.4
34.3±0.3
39.4±0.4
102±2.1
44.2±0.3
99.0±1.2
29.7±0.2
34.5±0.6
96.0±1.4
39.8±1.4
101±1.4
8.6±1.0
13.4±0.8
96.4±2.1
18.2±0.9
96.2±1.2
ND
4.9±0.7
97.4±1.6
9.5±0.7
95.0±1.1
ND
4.9±0.3
97.4±2.1
9.5±0.4
95.0±1.1
26.8±0.6
32.0±0.7
104±1.1
36.7±0.6
99.0±2.1
2-butanol
Found (µg L-1)
R (%)
ND
4.9±0.5
97.0±2.1
9.8±0.2
97.9±3.1
1.8±0.6
6.7±0.2
98.2±1.8
11.7±0.3
99.1±1.5
36.3±0.3
41.4±0.7
102.0±0.8
49.3±0.7
99.8±1.6
15.5±0.6
20.4±0.3
98.4±1.3
25.4±0.9
99.2±1.8
1.3±0.8
6.2±0.5
99.0±1.4
11.1±0.2
98.6±2.1
2.9±0.1
7.8±0.9
97.6±1.8
12.9±0.3
99.3±0.9
ND
4.9±0.6
97.0±2.4
10.0±0.4
99.6±1.3
21.1±0.6
26.0±0.4
98.8±1.2
31.0±0.6
99.6±0.8
Chapter six:
Preconcentration of trace elements in alcohols
Due to the absence of certified reference material (CRM) for the type of investigated
samples, the validity of the proposed separation and preconcentration method was examined
by analysing CRM TMDW-500 drinking water standard after diluting it with ethanol. The
results of the CRM supplied values and those obtained with our procedure for the
investigated metal ions, are summarised in Table 6.6. Satisfactory recoveries in the range of
95.5% to 102% were obtained. The precision of the measurements (n=6) expressed as %
RSD ranged between 0.1 to 2.0 %. Applying the student t-test at the 95% confidence level,
there was no significance differences between the certified and obtained concentration values.
Therefore, the agreement between certified and found concentration values of the analytes of
interest, demonstrated that the Dowex 50W-x8 SPE method was accurate for trace analysis of
Cd, Cr, Cu, Fe, Mn, Pb Ti and Zn in organic matrices.
Table 6.6. Analysis of the certified reference material (CRM TMDW-500 drinking water) for
the determination of analytes after application of the pre-concentration procedure
Elements
Cd
Cr
Cu
Fe
Mn
Pb
Zn
Certified (µg L-1)
10.0±0.05
20.0±0.1
20.0±0.1
100.0±0.5
40.0±0.2
40.0±0.2
70.0±0.4
Obtained (µg L-1)
9.8±0.2
19.7±0.1
20.1±0.4
98.7±0.6
38.9±0.3
38.2±0.4
70.1±0.1
Recovery (%)
98.0±1.2
98.5±0.5
100.5±1.1
98.7±1.2
97.3±0.9
95.5±2.1
100.3±0.5
6.3.10 Application of the Proposed Separation and Pre-Concentration Procedure
The proposed separation and pre-concentration procedure was applied to the
determination of metal ions in commercial methanol, ethanol, iso-propanol and butan-2-ol
samples from different solvent distributors. The concentrations of the metal ions in the
samples are shown in Table 6.7. It can be seen from this table that Cd does not occur in
methanol and 2-butanol or its concentration is lower than the LOD of the method and in other
alcohols it is present in trace levels (1.0-2.3 µg L-1). The concentrations of Cu, Fe and Zn
were the highest for all the analysed alcohol samples compared to other metal ions.
118
Chapter six:
Preconcentration of trace elements in alcohols
Table 6.7. Determination of metal ions (µg L-1) in commercial methanol, ethanol, iso-propanol and 2-butanol samples after pre-concentration by
the proposed method (pH 6, sample volume: 100 mL, n = 3) and the comparative one (ETAAS)
Elements
Methanol
SPE/ICP
ETAAS
OES
NDa
ND
5.3±0.1
5.1±0.5
13.3±0.5
12.9±0.5
41.4±0.3
41.8±0.6
1.4±0.4
1.4±0.3
3.3±0.7
3.4±0.6
1.5±0.1
1.5±0.2
21.4±0.6
22.1±0.3
Cd
Cr
Cu
Fe
Mn
Pb
Ti
Zn
a
ND= not detected
Ethanol
SPE/ICP
ETAAS
OES
2.3±0.2
2.5±0.9
7.2±0.3
7.0±0.7
18.3±0.3
18.6±0.3
38.4±0.6
38.1±0.5
3.7±0.5
3.8±0.7
2.9±0.6
2.9±0.9
4.0±0.2
4.0±0.4
21.3±0.4
21.5±0.3
119
Iso-propanol
SPE/ICP
ETAAS
OES
1.0±0.3
1.0±0.7
7.1±0.2
7.1±0.4
39.9±0.3
35.8±0.5
29.7±0.8
30.3±0.34
8.6±0.1
8.6±0.22
3.9±0.5
4.0±0.9
1.8±0.3
1.8±0.2
26.8±0.3
26.9±0.2
2-butanol
SPE/ICP
ETAAS
OES
ND
ND
1.8±0.6
1.8±0.7
36.3±0.3
36.2±0.1
15.1±0.6
15.5±0.2
1.3±0.8
1.2±0.8
2.9±0.1
3.0±0.3
ND
ND
21.7±0.6
22.2±0.5
Chapter six:
Preconcentration of trace elements in alcohols
The determination of analytes in the samples was performed by electrothermal atomic
absorption spectrometry (ETAAS) for the comparison of the results found by the proposed
separation and pre-concentration method. This was done in order further evaluate the
accuracy of the proposed method. In addition, ETAAS was chosen because of its capability
for the direct determination of metal ions in various matrices [25, 26]. It should be noted
that ETAAS determination was taken as the standard method in this study. According to
student t-test, the results obtained by the Dowex 50W-x8 pre-concentration method and
ETAAS (Table 6.7) were in agreement with each other at the 95% confidence level. The
agreement between the two set of results confirmed the reliability of Dowex 50W-x8 preconcentration method. Furthermore, the precision of both methods was comparable in that
the standard deviations did not differ significantly at 95% confidence level. Although
ETAAS has the ability to accomplish direct determination of metal ions in alcohol matrix,
the entire process of optimizing the parameters for each metal ion is time consuming.
Therefore, the main advantage of the proposed method over ETAAS is the capability of
performing a simultaneous pre-concentration and determination of metals (the analytes of
interest) within a short time. Thus SPE method in conjunction with ICP OES procedure
developed in this study, offers the advantage of time saving and a less tedious procedure.
6.4 CONCLUSIONS
In this study, the efficiency of Chelex-100, Dowex MAC-3 and Dowex 50W-x8 cation
exchange resins for the separation and pre-concentration of multi-element from ethanol
was investigated. The experimental conditions (such as sample pH, eluent concentration
and sample flow rates, among others) for quantitative pre-concentration and recovery of
metals prior to ICP OES detection were studied. The results demonstrated that Dowex
50W-x8 (strong cation-exchange) resin has good capability and efficiency for the
simultaneous pre-concentration of Cd, Cr, Cu, Fe, Mn, Pb, Ti and Zn in organic media. In
comparison, Chelex-100 showed limited performance (pre-concentration with percentage
recovery > 95%) for only few metals namely Cu, Fe and Zn whereas Dowex MAC-3 could
only show performance for Cu and Fe. In view of these results, Dowex 50W-x8 which had
the best overall performance for a wider range of metals was employed in further
experiments.
120
Chapter six:
Preconcentration of trace elements in alcohols
The optimized Dowex 50W-x8 solid phase extraction method was fast and at least a
pre-concentration factor of 100 for 500 mL sample was achieved for all metal ions. It was
observed that 1.5 g of Dowex 50W-x8 can be used as high as 100 pre-concentration cycles
without any loss in its adsorption efficiency. The metal ions appeared to be strongly bound
onto the resins and as such the elution was not quantitative when the eluent concentrations
were less than 3.0 mol L-1 except for Cu. Therefore, the elution of the metal ions adsorbed
on Dowex 50W-x8 was obtained with 3.0 mol L-1.
The analytical performances of the proposed Dowex 50W-x8 SPE method was
comparable or even better than other pre-concentration methods reported in the literature.
The accuracy and precision of the SPE method were reported in terms of recovery (%)
ranging 95-104%, and %RSD ranging 1.2-3.3%. The proposed SPE method was applied
for the determination of trace metal ions in reagent grade solvents; methanol, ethanol, isopropanol and 2-butanol.
6.5 REFERENCES
1. Grodowska, K. & Parczewski, A. 2010. Organic Solvents in the Pharmaceutical
Industry. Acta Poloniae Pharmaceutical-Drug Research, 67, 3-12.
2. Pritchard, J.D. 2007. Methanol–general information. CHAPD HQ, HPA. Available at
http://www.hpa.org.uk/webc/HPAwebFile/HPAweb_C/1194947361312. Accessed 20
June 2011.
3. Gnansounou, E. & Dauriat, A. 2005. Ethanol fuel from biomass: a review. Journal of
Scientific Industrial Reearch, 60, 809-921
4. Anton, R., Barlow, S., Boskou, D., Castle, L., Crebelli, R., Dekant, W., Engel, K.-H.,
Forsythe, S., Grunow, W., Larsen, J.C., Leclercq, C., Mennes, W., Milana, M.-R., Pratt,
I., Rietjens, I., Svensson, K., Tobback, P. & Toldrá, F. 2005. Opinion of the Scientific
panel on food additives, flavourings, processing aids and materials in contact with food
on a request from the commission related to propan-2-ol as a carrier solvent for
flavourings. The EFSA Journal, 202, 1-10.
5. Hussain, S., Liba, A. & McCurdy, E. 2011. Validating ICP-MS for the Analysis of
Elemental Impurities According to Draft USP General Chapters 232 and 233 (2011).
Available at http://www.spectroscopyonline.com/spectroscopy/Articles/Validating-ICPMS-for-the-Analysis-of-Elemental-Im/ArticleStandard/Article/detail/749104. Accessed
14 May 2012.
6. Fliszar, K. A., Walker, D. & Allain, L. 2006. Profiling of metal ions leached from
pharmaceutical packaging materials PDA. Journal Pharmaceutical Science and
Technology, 60, 337-342.
121
Chapter six:
Preconcentration of trace elements in alcohols
7. Chaves, E. S., F. G. Lepri, J. S. A. Silva, D. P. C. De Quadros, T. D. Saint’pierre And A.
J. Curtius 2008. Determination of Co, Cu, Fe, Mn, Ni and V in diesel and biodiesel
samples by ETV-ICP-MS. Journal of Environmental Monitoring, 10, 1211-1216.
8. Santos, D. S. S., M.G. A. Korn, M. A. B. Guida, G. L. Dos Santos, V. A. Lemos And L.
S. G. Teixeira 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by
Sequential Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase
Extraction. Journal of Brazzilian Chemical Society, 22, 552-557.
9. Takeuchi, R., Santos, A., Medeiros, M. & Stradiotto, N. 2009. Copper determination in
ethanol fuel samples by anodic stripping voltammetry at a gold microelectrode.
Microchimica Acta, 164, 101-106.
10. Kishi, Y. & Kawabata, K. 2004. Determination of trace metallic impurities in organic
solvents by DRC-ICP-MS. Semiconductor pure water and chemical conference.
Available at http://www.perkinelmer.co.kr/files/AP00030.pdf. Accessed 19 March
2012.
11. Dionex Application no. 72. Determination of Trace Metals in Water Miscible Organic
Solvents by Ion Chromatography/Inductively Coupled Argon Plasma Spectroscopy
(IC/ICAP).
Available
at
http://www.dionex.com/en-us/webdocs/4671AN72_LPN034619-02.pdf. Accessed 19 March 2012.
12. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
13. Bettinelli, M., S. Spezia, U. Baroni And G. Bizzarri. 1995. Determination of trace
elements in fuel oils by inductively coupled plasma mass spectrometry after acid
mineralization of the sample in a microwave oven. Journal of Analytical Atomic
Spectrometry, 10, 555-560.
14. Wang, T., Jia, X. & Wu, J. 2003. Direct determination of metals in organics by
inductively coupled plasma atomic emission spectrometry in aqueous matrices. Journal
of Pharmaceutical and Biomedical Analysis, 33, 639-646..
15. Pereira, J. S. F., Moraes, D. P., Antes, F. G., Diehl, L. O., Santos, M. F. P., Guimarães,
R. C. L., Fonseca, T. C. O., Dressler, V. L. & Flores, É. M. M. 2010. Determination of
metals and metalloids in light and heavy crude oil by ICP-MS after digestion by
microwave-induced combustion. Microchemical Journal, 96, 4-11.
16. Ekanem, E. J., Lori, J. A. & Thomas, S. A. 1997. The determination of wear metals in
used lubricating oils by flame atomic absorption spectrometry using sulphanilic acid as
ashing agent. Talanta, 44, 2103-2108.
17. Teixeira, L. S. G., Bezerra, M. D. A., Lemos, V. A., Santos, H. C. D., De Jesus, D. S.
& Costa, A. C. S. 2005. Determination of Copper, Iron, Nickel, and Zinc in Ethanol
Fuel by Flame Atomic Absorption Spectrometry Using On-Line Preconcentration
System. Separation Science and Technology, 40, 2555 - 2565.
18. Roldan, P. S., Alcântara, I. L., Rocha, J. C., Padilha, C. C. F. & Padilha, P. M. 2004.
Determination of Copper, Iron, Nickel and Zinc in fuel kerosene by FAAS after
122
Chapter six:
Preconcentration of trace elements in alcohols
adsorption and pre-concentration on 2-aminothiazole-modified silica gel. Ecl. Quím.,
São Paulo, 29, 33-40.
19. Roldan, P. S., Alcântara, I. L., Padilha, C. C. F. & Padilha, P. M. 2005. Determination
of copper, iron, nickel and zinc in gasoline by FAAS after sorption and preconcentration
on silica modified with 2-aminotiazole groups. Fuel, 84, 305-309.
20. Alves, V. N., Mosquetta, R., Coelho, N. M. M., Bianchin, J. N., Di Pietro Roux, K. C.,
Martendal, E. & Carasek, E. 2010. Determination of cadmium in alcohol fuel using
Moringa oleifera seeds as a biosorbent in an on-line system coupled to FAAS. Talanta,
80, 1133-1138.
21. Sharma, R. K. & Pant, P. 2009. Preconcentration and determination of trace metal ions
from aqueous samples by newly developed gallic acid modified Amberlite XAD-16
chelating resin. Journal of Hazardous Materials, 163, 295-301.
22. Yin, J., Jiang, Z., Chang, G. & Hu, B. 2005. Simultaneous on-line preconcentration and
determination of trace metals in environmental samples by flow injection combined
with inductively coupled plasma mass spectrometry using a nanometer-sized alumina
packed micro-column. Analytica Chimica Acta, 540, 333-339.
23. Abollino, O., Aceto, M., Sarzanini, C. & Mentasti, E. 2000. The retention of metal
species by different solid sorbents: Mechanisms for heavy metal speciation by
sequential three column uptake. Analytica Chimica Acta, 411, 223-237.
24. Pohl, P. & Prusisz, B. 2010. Chemical fractionation of Cu, Fe and Mn in canned Polish
beers. Journal of Food Composition and Analysis, 23, 86-94.
25. Anselmi, A., Tittarelli, P.& Katskov, D. A. 2002. Determination of trace elements in
automotive fuels by filter furnace atomic absorption spectrometry. Spectrochimica Acta
Part B: Atomic Spectroscopy, 57, 403–411
26. Reboucas, M. V., Domingos, D., Santos, A. S. O. & Sampaio, L. 2010. Determination
of trace metals in naphtha by graphite furnace atomic absorption spectrometry:
Comparison between direct injection and microemulsion pretreatment procedures. Fuel
Processing Technology, 91, 1702-1709.
27. de Oliveira, A. P., de Moraes, M., Neto, J. A. G. & Lima, E. C. 2002. Simultaneous
determination of Al, As, Cu, Fe, Mn, and Ni in fuel ethanol by GFAAS. Atomic
Spectroscopy, 23, 39-43.
28. Soylak, M. 2004. Solid phase extraction of Cu(II), Pb(II), Fe(III), Co(II) and Cr(III) on
Chelex-100 column prior to their flame atomic spectrometric determinations Analytical
Letters, 37, 1203-121.
29. Camel, V. 2003. Solid phase extraction of trace elements. Spectrochimica Acta Part B:
Atomic Spectroscopy, 58, 1177-1233.
30. Malla, M. E., Alvarez, M. B. & Batistoni, D. A. 2002. Evaluation of sorption and
desorption characteristics of cadmium, lead and zinc on Amberlite IRC-718
iminodiacetate chelating ion exchanger. Talanta, 57, 277-287.
31. Pohl, P., Bogdal, Z. & Prusisz, B. 2005. Preconcentration and Fractionation of Cd, Co,
Cu, Ni, Pb and Zn in Natural Water Samples Prior to Analysis by Inductively Coupled
Plasma Atomic Emission Spectrometry. Microchimica Acta, 150, 253-259.
123
Chapter six:
Preconcentration of trace elements in alcohols
32. Yildiz, O., Citak, D., Tuzen, M. & Soylak, M. 2011. Determination of copper, lead and
iron in water and food samples after column solid phase extraction using 1phenylthiosemicarbazide on Dowex Optipore L-493 resin. Food and Chemical
Toxicology, 49, 458-463.
33. Pohl, P. & Prusisz, B. 2009. Application of tandem column solid phase extraction and
flame atomic absorption spectrometry for the determination of inorganic and
organically bound forms of iron in wine. Talanta, 77, 1732-1738.
34. Pohl, P. & Prusisz, B. 2004. Pre-concentration of Cd, Co, Cu, Ni and Zn using different
off-line ion exchange procedures followed by the inductively coupled plasma atomic
emission spectrometric detection. Analytica Chimica Acta, 502, 83-90.
35. Parham, H., Pourreza, N. & Rahbar, N. 2009. Solid phase extraction of lead and
cadmium using solid sulfur as a new metal extractor prior to determination by flame
atomic absorption spectrometry. Journal of Hazardous Materials, 163, 588-59.
36. Pyrzyñska, K. & Joñca, Z. 2000. Multielement Preconcentration and Removal of Trace
Metals by Solid-Phase Extraction. Analytical Letters, 33, 1441 - 145.
37. Inglezakis, V. J. & Loizidou, M. D. 2007. Ion exchange of some heavy metal ions from
polar organicsolvents into zeolite. Desalination, 211, 238-248.
38. Teixeira, L. S. G., Santos, E. S. & Nunes, L. S. 2012. Determination of copper, iron,
nickel and zinc in ethanol fuel by energy dispersive X-ray fluorescence after preconcentration on chromatography paper. Analytica Chimica Acta, 722, 29-33.
39. Santos, L. N., Neto, J. A. G. & Caldas, N. M. 2012. Simultaneous determination of Cu
and Pb in fuel ethanol by graphite furnace AAS using tungsten permanent modifier with
co-injection of Ir. Fuel, 99, 9-12..
40. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A.
M. & Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in
automotive gasoline by X-ray fluorescence after pre-concentration on cellulose paper.
Talanta, 72, 1073-1076.
41. Rocha, M. S., Mesko, M. F., Silva, F. F., Sena, R. C., Quaresma, M. C. B., Araujo, T.
O. & Reis, L. A. 2011. Determination of Cu and Fe in fuel ethanol by ICP OES using
direct sample introduction by an ultrasonic nebulizer and membrane desolvator. Journal
of Analytical Atomic Spectrometry, 26, 456-46.
124
CHAPTER SEVEN:
PRECONCENTRATION OF MOLYBDENUM, ANTIMONY AND VANADIUM IN
GASOLINE SAMPLES USING DOWEX 1-X8 RESIN AND THEIR
DETERMINATION WITH ICP OES
ABSTRACT
Strong ion exchangers (Dowex 50W-x8 and Dowex 1-x8) were used for the separation
and preconcentration of trace amounts of Mo, Sb and V in gasoline samples. Dowex 1-x8
resin was found to be suitable for the quantitative retention of these metal ions from organic
matrices. The elution of the metal ions from Dowex 1-x8 resin was achieved by using 2.0 mol
L-1 HNO3 solution. The Dowex 1-x8 preconcentration and separation method gave an
enrichment factor of 120 with limits of detection equal to 0.14, 0.05 and 0.03 µg L-1 for Mo,
Sb and V, respectively. The limits of quantification were found to be 0.48, 0.18 and 0.10
µg L-1 for Mo, Sb and V, respectively. Under optimized conditions, the relative standard
deviations of the proposed method (n =20) were <4%. The accuracy of Dowex 1-x8
preconcentration procedure was verified by the recovery test in the spiked samples of
gasoline sample. The Dowex 1-x8 preconcentration method was applied to Conostan custom
made oil based certified reference material for the determination of Mo, Sb and V. The
results of the paired t-test at a 95% confidence level showed no significant difference. The
separation and preconcentration procedure was also applied to the gasoline samples collected
from different filling stations.
Keywords: Preconcentration; gasoline; Dowex 1-x8; metal ions; ICP OES
7.1 INTRODUCTION
The knowledge of metal ion concentrations in fuel is of great interest with respect to
economic and environmental issues.1,2 For instance, the presence of metal ions such as
antimony (Sb) and vanadium (V) in gasoline causes catalyst poisoning during the catalytic
cracking of naphtha and gasoline.3 Generally, metal ions in gasoline play a significant role in
engine maintenance, since metallic species can catalyse the corrosion of engines or promote
the formation of gums and sediments.4-6 In addition, some metal ion compounds, especially
vanadium, in gasoline are of environmental concern due to their potential impact in human
health since they may cause mutagenic and carcinogenic effects.7-9 Furthermore, their
presence in gasoline causes fuel degradation, air pollution (especially in big cities) and
125
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
reduces the efficiency of catalytic reactors in vehicle exhaust systems, thus increasing the
emission of exhaust gases.4,6
Many metals occur naturally in fossil materials and, as a result, they can be present in
petroleum based products. The presence of metal ions in petrochemical compounds (e.g. fuel)
can also be due to their incorporation during the production process, by contact with
refinement or distillation equipment, storage and transport. Another source of these elements
is that some can be added to the fuel improve its characteristics of the products.5,10 Vanadium
and molybdenum (Mo) are widely used as catalysts in the desulfurisation of petroleum,
petrochemicals and coal-derived liquids to minimise sulfur dioxide emission from fuel
combustion.11 As such, residues of trace amounts of these metal ions can be found in the final
product. The presence of metals in petrochemical organic products is undesirable, unless they
are used as additives. Methods for the analysis of vanadium in fuel has been one of the most
studied elements.1,8,12 However, there is an increasing interest towards a number of other
elements, such as Mo and Sb, among others, because they occur naturally in fossil fuels.
Direct determination of metals in fuel samples such as gasoline, by most analytical
techniques is difficult. This is because of its volatility, low viscosity, corrosivity and
immiscibility with water.13 Inductively coupled plasma-optical emission spectrometry (ICP
OES) is a sensitive multi-element technique, but direct introduction of gasoline requires a
special care.13,14 This is because direct loading of organic samples to the ICP can destabilize
or extinguish the plasma.14,15 Therefore, a sample preparation step that will separate and
preconcentrate trace metals in gasoline prior to ICP OES detection, is required. Techniques
involving separation and preconcentration procedures for the determination of trace elements
in gasoline and fuel kerosene are reported in the literature.6,13,16,17 In some of these
preconcentration techniques, various sorbent materials have been functionalised with organofunctional groups in order to extract metal ions from complex matrices. Due to the leaching
action of organic samples, the main limitation of these procedures in the analysis of metals in
gasoline is to maintain the organofunctional group attached to the solid phase.6 In addition,
these techniques focus only on the preconcentration of Cu, Ni, Fe and Zn. Therefore, to the
best of our knowledge, there are no reports on Mo, Sb and V.
The aim of the present study was to explore the applicability of commercially available
ion exchange resins for the preconcentration and separation of Mo, Sb and V in gasoline prior
to their determination using ICP OES. The reason for choosing commercial resins is that they
are the most commonly utilized cation exchangers for the removal of many metal ions from
126
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
aqueous solutions and therefore well studied. The resins contain functional groups for metal
ion binding and hence are effective in the extraction of heavy metals from organic phase
matrices. In addition, procedures involving separation and preconcentration methods using
commercially available ion exchange resins combined with ICP OES for the determination of
Mo, Sb and V in gasoline have not been reported in the literature. The retention performance
of Mo, Sb and V ions on strongly acidic cation exchanger resin (Dowex 50W-x8) and
strongly basic anion exchanger resin (Dowex 1-x8), was studied at different solution pH
values for different resin columns. For the selected resin, that is, Dowex 1-x8, experimental
conditions for retention/desorption of Mo Sb and V ions prior to their ICP OES detection,
were optimized. The validity of the separation and preconcentration procedure was verified
by comparing the solid phase extraction (SPE)/ICP OES results with those obtained by
GFAAS after acid digestion. The spike-recovery experiments were conducted to evaluate the
accuracy of the method. The procedure was applied for the determination of traces of Mo, Sb
and V in Conostan custom made oil based certified reference material (CRM) and
commercial gasoline samples.
7.2. EXPERIMENTAL
7.2.1 Instrumentation
Metal ions were determined using a Spetro Arcos ICP OES (SPECTRO Analytical
Instruments, GmbH, Germany) equipped with a Cetac ASX-520 autosampler. The operating
conditions were as follows: forward power 1400 W, plasma argon flow rate 13 L min-1,
auxiliary argon flow rate 2.00 L min-1, and nebulizer argon flow rare 0.95 L min-1. The most
prominent atomic and ionic analytical spectral lines of the metals studied, were selected for
investigation, i.e. Mo 202.030 nm, Sb 206.833 nm and V 292.402 nm. Solid phase extraction
was carried out in a VacMaster-24 sample SPE station (Supelco, PA, USA). The latter was
used to control the sample loading and elution flow rate at 3.0 ml min-1. Comparative
experiments for the determination of metal ions were performed using A Perkin-Elmer
(Norwalk, CT, USA) A Analyst 100 atomic absorption spectrometer equipped with a HGA800 graphite furnace and an AS-72 autosampler. High purity nitrogen (99.996%, Afrox,
South Africa) was used as purging gas. Appropriate hollow cathode lamps (HCL) from
Perkin-Elmer were used in these experiments. The graphite furnace atomic absorption
127
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
spectrometry (GFAAS) operation parameters and heating temperature program is presented
in Table 7.1.
Table 7.1. Operation parameters and heating temperature program for ETAAS
Spectrometer setup
Wavelength (nm)
Lamp type
Lamp current (mA)
Step
Drying 1
Drying 2
Pyrolysis
Atomization
Cleaning
Mo
313.3
HCL
10
Sb
217.6
HCL
15
Heating program for the atomizer
Step T (°C)
Ramp (°C s-1)
90
1
200
1
Mo
Sb
V
1800 , 1000 , 1000
5
Mo
Sb
V
2450 , 1500 , 2400
0
2500
1
V
318.40
HCL
15
Hold time (s)
5
10
20
6
5
7.2.2 Reagents, Solutions and Samples
All reagents were of analytical grade unless otherwise stated and double distilled
deionised water (Millipore, Bedford, MA, USA) was used throughout the experiments.
Absolute ethanol (Merck, Darmstadt, Germany), was used to prepare model solutions.
Spectrascan stock solutions (1000 mg L-1) of Mo Sb and V (Industrial Analytical Pty,
Johannesburg, Ltd, South Africa) were used to prepare the working solutions for SPE at
concentrations of 10 µg L-1 for Mo and Sb and 12 µg L-1 for V. Working solutions, as per the
experimental requirements, were freshly prepared from the stock solution for each
experimental run. A Spectrascan multi-element standard solution at concentration of 100 mg
L-1 (Industrial Analytical Pty, Johannesburg, Ltd, South Africa) was used to prepare working
standard solutions of 10-120 µg L-1 for measurements of concentrations of analytes in all
model and sample solutions. Solutions of nitric acid at concentrations of 0.5, 1.0, 2.0 and 3.0
mol L-1 used for the elution of the analytes from the column, were prepared from ultrapure
concentrated acid (65%, Sigma-Aldrich, St. Loius, MO, USA). The pH adjustments were
performed with glacial acetic acid and diluted ammonia solutions (Sigma-Aldrich, St. Loius,
MO, USA). Suprapur 30% hydrogen peroxide (H2O2, Merck, Darmstadt, Germany) was used
for acid digestion procedure.
128
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
Ten gasoline samples from different local filling stations were used for method
development and validation. Gasoline samples with and without additives are described as
metal-containing unleaded gasoline (MCUG) and metal-free unleaded gasoline (MFUG),
respectively.
The cation exchangers used in this study as packing materials namely Dowex 1-x8
(Chloride form) and Dowex 50W-x8 (sodium form), were purchased from Sigma-Aldrich,
(St. Loius, MO, USA). The properties of the resins are given in Table 7.2.
Table 7.2. Physical and chemical properties of the resins
Type
Functional group
Matrix
Standard mesh size
Total exchange capacity
(meq mL-1)
Water
retention
capacity%
pH range
Maximum
operating
temperature
Dowex 50W-x8
Dowex 1-x8
Strong acidic cation exchanger
Sulfonic acid
Styrene-divinylbenzene
100-200
1.7
Strong basic anion exchanger
Quaternary amine
Styrene-divinylbenzene
200-400
1.2
50-58
39-45
0-14
120 °C
0-14
66 °C
7.2.3 Preparation of Column
Polyethylene columns of diameter 1.35 cm and 6.5 cm in height were used for
preconcentration. Slurries of 1.5 g of Dowex 50W-x8 or Dowex 1-x8 in double distilled
deionised water were prepared and packed into columns to heights of about 3-4 cm. A porous
frit was placed at the bottom of the column and at the top of the packing material to hold and
confine the adsorbent within the designated capacity/volume. The entrapment of the packing
material serves to eliminate the dead volume. The columns were washed with double distilled
deionised water followed by conditioning with 10 mL ammonium acetate buffer (1.0 M, pH
9.0) and then 10 mL of ethanol.
129
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
7.2.4 Preconcentration and Recovery of Mo, Sb and V in Model Organic Solutions and
Real Gasoline Samples
Model metal ion solutions were prepared as follows: 1.0 mL of 1.0 mg L-1 of Mo and Sb
solutions were separately transferred into 100 mL volumetric flasks and made up to the mark
with ethanol to obtain 10 µg L-1 of each metal ion. The procedure was repeated for V to
obtain 12 µg L-1. Ethanol solutions of each metal ion (20 mL) were percolated through the
ion exchange resin column with a flow rate of 3.0 mL min-1. The column was washed with 10
mL of double distilled deionised water to remove excess organic solution, followed by 5.0
mL of ammonium acetate buffer solution, to remove major cations (Na, Ca, K, etc). Lastly,
the metal ions were eluted with 5 mL of 2.0 mol L-1 HNO3 solution. All fractions obtained
during the elution stage were collected separately and analysed with ICP OES. It should be
noted that the washings with double distilled deionised water and ammonium acetate buffer
solution were discarded. The same procedure was applied to the blank solutions. In between
the experiments, the resin was washed as described in Section 7.2.3. The effect of pH sample
solution, sample volume, eluent concentration and sample and eluent flow rates were
investigated. In order to measure metal ions in real samples, 1.0 mL aliquot of gasoline was
placed in a 100 ml volumetric flask and diluted with ethanol. The resulting solution was then
subjected to the above mentioned preconcentration procedure. All analyses were performed
in triplicate.
7.2.5 Procedure for the Dilution of Certified Reference Material
To validate the preconcentration method described in this study, a Conostan custom made
oil based certified reference material (CRM) obtained from SCP Science (Quebec, Canada)
containing 1.0 mg L-1 of each metal ions was used. The dilution of the CRM was performed
as follows: a 1.0 mL aliquot of 1.0 mg L-1 CRM was dissolved in 10 ml of hexane. The
solution was quantitatively transferred to a 100 mL volumetric flask and made to the mark
with acetone to obtain 10 µg L-1 of each metal ion. Suitable aliquots (20 ml) of the solution
were taken and pre-concentrated by the proposed procedure and analysed with ICPOES. The
same procedure was applied to the preparation of blank solutions.
130
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
7.2.6 Procedure for Acid Digestion of Gasoline Samples
The acid digestion procedure was carried out according to Amorim et al.1 A brief
Description of the procedure is as follows: 5.0 ml of gasoline sample was placed into a 100
mL Teflon beaker followed by the addition of 2.0 mL H2O2 (30%) and 6 mL concentrated
HNO3 and heated at 170 ± 10 °C in hot plate for 10 min. The digested content was left to cool
down to room temperature, quantitatively transferred to a volumetric flask and then diluted
with double distilled deionised water to a final volume of 50 mL. double distilled deionised
water, applied to the same procedure, was used as the blank. The digested samples were then
analyzed by GFAAS.
7.3. RESULTS AND DISCUSSION
In order to achieve quantitative adsorption of Mo, Sb and V onto the solid sorbent, the
preconcentration method was optimized for various analytical parameters (such as the sorbent
material selection, pH and eluent concentration). Flow rates, eluent volume and the amount of
the sorbent fixed to 3.0 mL min, 5.0 mL and 1.5 g, respectively. The experimental conditions
for the preconcentration of metal ions were investigated using absolute ethanol model
solutions. The percentage recovery of analytes retained on the column was calculated from
the concentration of the metal ions in the starting sample and the amounts of analytes eluted
from Dowex 1-x8 and Dowex 50W-x8 columns.
7.3.1 Selection of Stationary Phase
In order to choose a suitable SPE stationary phase for the preconcentration of Mo, Sb and
V, anionic (Dowex 1-x8) and cationic (Dowex 50W-x8) exchange resins were tested. Since
the pH of the sample solution plays a vital role in the retention of metal ions, a comparison of
recoveries of Mo, Sb and V at different pH values (4, 7 and 9) were carried out. The
%recovery values for Mo, Sb and V were less than 50% at the three pH values when Dowex
50W-x8 resin was used. The lowest %recovery was 9.8% for Mo at pH 9 while the highest
was 44.6% for Sb at pH 7. This shows that these elements were partially retained on the
surface of the Dowex 50W-x8 resin. The reason for the partial adsorption might be the fact
that these elements exist in various oxidation states and different ionic species in aqueous
solutions. Pyrzynska and Jonca19 studied the behavior of Mo in aqueous samples. They found
that Mo was partially retained on cation exchange resin because in solution, the dominant
131
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
species is MoO42-. When using the stronger anionic resin namely Dowex-1x8, improved
results were obtained with the lowest recovery value being 54.9% for V at pH 4 while the
highest value was 97.4 % for Sb at pH 7. Therefore, Dowex-1x8 was used for further
investigations.
7.3.2 Effect of Sample Solution pH on Retention of Metal Ions
Owing to competition for the exchange sites of the resin, between metal ions and
hydrogen ions in solution, the effect of pH of the sample solution was studied as a significant
factor for quantitative retention of the analyte.20 The effect of the pH of the ethanol model
solution on the retention of Mo, Sb and V onto Dowex 1-x8 was investigated in the pH range
4–10. The results are shown in Fig. 7.1. The recovery values for Mo, Sb and V were not
quantitative at the pH values below 6. The recoveries increased with increasing pH and
reached quantitative values at the pH range 6–8 for Mo and V ions and 6-7 for Sb. Therefore,
pH 6 was selected for further investigations.
Fig. 7.1. Effect of sample pH on retention of the analytes in ethanol onto Dowex 1-x8 resin
column. Sample volume: 20 mL; amount of resin 1.5 g; flow rates of sample and eluent 3.0
mL min−1; replicates n = 3)
132
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
7.3.3 Effect of Eluent Concentration
The effect of eluent (HNO3) concentration on elution of metal ions from Dowex 1-x8
resin was investigated by carrying out the elution with 0.5-3.0 mol L-1 HNO3. In this work,
the optimum eluent concentration was defined as the concentration of the eluent that can elute
more than 95% of the retained metal ions. The results obtained indicated that elution of metal
ions is dependent on the eluent concentration. Thus when [HNO3] was increased from 0.5 to
2.0 mol L-1 the elution efficiency of the acid improved from 92.4 to 100.7% for Mo; 83.2 to
98.3% for Sb; and 91.0 to 96.5% for V. However, there was a slight decrease in elution
efficiency when the acid concentration was increased from 2.0 to 3.0 mol L-1 in all the cases
except for V which showed an increase from 96.5 to 100.1%. Therefore, for further
investigations 5.0 ml of 2.0 mol L-1 HNO3 was selected for elution of metal ions.
7.3.4 Effect of Sample Volume
The preconcentration of larger sample volumes at a defined sample flow rate improves
the preconcentration factor of the SPE method.21 In order to investigate the possibility of
enriching low concentrations of analytes from large volumes, the maximum applicable
sample volume must be determined. For this purpose, the concentrations of each metal ion
were kept constant while increasing the sample volume. The recoveries of the Mo, Sb and V
ions from different volumes of ethanol solutions containing 14 µg L-1 of each metal ion, are
presented in Fig. 7.2.
133
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
Fig. 7.2. Effect of sample volume on the recoveries of metal ions: pH 6; analyte concentration
14 µg L-1; amount of resin 1.5 g; flow rates of sample and eluent 3.0 mL min−1; eluent
volume 5 mL; replicates n = 3
As it can be seen from this figure, the recoveries were found to be stable upto 500 mL for
Sb and 600 mL for Mo and V. At higher volumes, the recoveries for analytes decreased. This
decrease may be due to the saturation of the exchange sites of Dowex 1-x8 resin. Therefore,
the highest preconcentration factor of 120 was achieved when using 600 mL of the sample
and 5.0 mL of final volume.
7.3.5 Analytical Performances
The linearity of the method was checked by preconcentrating 20 mL aliquot sampled
from 100 mL of 0.05 to 140 µg L-1 of Mo, Sb and V standards in ethanol solution, to obtain a
final volume of 5 mL (preconcentration factor of 4). The analytes showed good linearity
according to the results in Table 7.3. The sensitivity of the preconcentration method was
defined as the slope of the calibration curve. The results in Table 7.3 indicated that the
Dowex 1-x8 SPE method was more sensitive to V compared to the rest of metals studied.
Thus the highest slope obtained was 113.1 cps L µg-1 for V while the lowest was 26.39 cps L
µg-1 for Mo. Note that the ICP OES signal intensity readings are given as counts per second
(cps).
134
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
The reproducibility (precision) of the SPE method, calculated as the relative standard
deviation (n = 12), in model sample solutions containing 10 µg L-1 of Mo, Sb and V was in
the range 1.1-1.9%. The IUPAC limit of detection (LOD) and limit of quantification (LOQ)
under optimized conditions were obtained from the signals of 16 successive measurements of
the blank and the slope (m) of the calibration curve. The LOD was defined as the lowest
concentration of an analyte giving signals equal to three times the standard deviation (3SD) of
blank signal divided by the slope of the calibration curve that the analytical technique can
detect (3SD/m). The LOQ, on the other, was defined as the to the smallest concentration of an
analyte giving signals equal to ten times the standard deviation of blank signal divided by the
slope of the calibration curve which can be accurately and precisely measured with an
analytical procedure (10SD/m). The calculated LOD and LOQ as well as instrumental
detection limits (IDL) are presented in Table 7.3. It can be seen from this Table that SPE
method has improved detection capabilities as compared to ICP-OES method.
The Dowex 1-x8 SPE method was compared with the other previous works (Table 7.4).
Comparison of analytical features of the present method with other sample preparation
techniques indicated that the LOD and LOQ of the Dowex 1-x8 SPE are better than or
comparable with other methods.
Table 7.3. Analytical performances for the proposed Dowex 1-x8 SPE method (sample
volume 100 mL)
-1
Slope (cps L µg L )
Correlation efficient
LOD (µg L-1)
IDL (µg L-1)
LOQ (µg L-1)
%RSD
Mo
26.39
0.9998
0.14
0.5
0.48
1.9
Sb
53.43
0.9987
0.05
2.0
0.18
1.2
135
V
113.1
0.9999
0.03
0.5
0.10
1.1
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
Table 7.4. Comparison of some methods used for determination of Mo, Sb and V
Analytes
LOD (µg L-1)
4.0
LOQ (µg L-1)
N.Ia.
Ref.
[3]
Tungsten
coated-GFAAS
ETAAS
using ETAAS
N.I.
N.I.
[22]
2.5
0.8
N.I.
N.I.
[23]
[24]
GFAAS
300
800
45
[1]
GFAAS
100
250
14
ETAAS
0.05
0.18
[25]
TXRF
65 and 75
N.I.
[26]
ETAAS
ICP OES
0.9 and 4.7
4.1 and 2.4
N.I.
13.7 and 7.9
[27]
[28]
ICP OES
0.14, 0.05 and 0.03
0.48, 0.18 and 0.10
This work
Sample Preparation method
and Surfactant microemulsion
V
Matrix
Gasoline
kerosene
Heavy oils
Sb
V
Naphta
Petroleum
V
Fuel oils
V
Microemulsion
Direct determination
solid sampling accessory
Microemulsion
Acid digestion
Detergent emulsion
Petroleum
condensate, diesel
and gasoline
Diesel oil
Extraction
induced
by
emulsion breaking
Petroleum
Direct analysis
products
Gasoline
Multiphase emulsion
Crude oil
Detergentless microemulsion
Sb
V
Mo and V
Mo and V
Mo and V
Mo, Sb and Gasoline
V
a
N.I. = not included
Direct determination
SPE-Dowex 1-x8
Detection
ETAAS
136
[8]
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
7.3.6 Effect of Matrix Ions Interferences
The effect of potential interfering ions was investigated in order to examine the
possibility of selective recovery of Mo, Sb and V on Dowex 1-x8 resin in the presence of
some anions and cations in the gasoline samples. Known amounts of anions and cations
were added to the ethanol solution containing 20 µg L-1 of Mo, Sb and V and the general
preconcentration procedure was applied. The tolerance limit was set as the concentration of
the ion required to cause ≤5% error. The results are presented in Table 7.5. It should be
noted that the effect of Group I and II cations were not investigated because they were
removed by washing the column with ammonium acetate buffer solution. It was found that
ions normally present in gasoline samples did not interfere with the recoveries of the
analytes. This suggests that the Dowex 1-x8 preconcentration method can be applied for
selective removal of trace amounts of Mo, Sb and V in fuel samples that contains higher
concentration of secondary cations and anions.
Table 7.5. Effect of potential interfering ions on the recovery of metal ions
Ions
SO42PO43-
Concentration of
interfering ions (mg L-1)
1000
1000
1000
1000
1000
25
25
25
25
Recovery (%)
Mo
99.3±1.2
98.7±1.4
100±1.1
97.0±1.8
99.8±0.3
95.9±3.1
97.6±2.1
99.3±0.5
96.8±1.4
Sb
97.6±0.8
97.9±2.1
99.6±0.5
96.9±0.7
99.1±2.4
99.1±1.4
97.8±1.9
98.9±2.4
97.4±2.8
Cl
CO32NO3Fe3+
Cu2+
Mn2+
Al3+
Experimental conditions: sample volume = 50 mL, replicates n = 3
V
100±2.1
99.2±1.1
98.9±2.3
99.3±0.9
100±0.2
98.1±1.1
99.6±2.1
99.1±3.8
98.9±0.9
7.3.7 Regeneration Studies
The regeneration of the column is one of the important parameters in evaluating the
stability of the cation exchange resin material. In order to investigate the stability and
recyclability of Dowex 1-x8 column, successive retention and elution cycles were
performed by passing 20 mL of ethanol solutions (containing Mo, Sb and V) through the
137
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
column. The stability and regeneration of Dowex 1-x8 column were evaluated by
monitoring the changes in the recoveries of Mo, Sb and V through retention-elution cycles.
The results showed that Dowex 50W-x8 column can be reused after regeneration with 20
mL double distilled deionised water followed by 10 mL of 1.0 mol L-1 NaOH. The column
was found to be stable up to 150 retention/elution cycles without any observable decrease
in the recoveries of metal ions (> 95%). Therefore, recycling of the Dowex 1-x8 resin is
possible.
7.3.8 Accuracy and Validation of the Separation and Preconcentration Procedure
An addition/recovery test was performed on the gasoline sample (1-MFUG) to estimate
the accuracy of the Dowex 1-x8 SPE procedure. The results given in Table 7.6 showed
good agreement between the added and found metal ion content. The recovery values for
Mo, Sb and V ions were quantitative (≥ 95%). Therefore, the Dowex 1-x8 SPE method can
be applied for the separation and preconcentration of analyte ions in gasoline samples.
Table 7.6. Percentage (%) recovery results when 1 mL gasoline sample 1-MFUG was
spiked with different metal concentrations (0-20 µg L-1) and made up in ethanol (100 mL)
Element
Added (µg L-1)
Found (µg L-1)
Recovery (%)
Mo
0
5
10
20
0
5
10
20
0
5
10
20
20.10±0.76
24.96±0.35
30.03±0.54
39.77±0.10
70.95±0.82
75.73±0.25
80.64±1.0
90.75±0.12
NDa
4.89±0.91
9.98±0.34
19.7±0.54
97.2±1.6
99.3±1.2
98.4±1.3
95.6±2.1
96.9±3.1
99.0±0.6
97.8±1.2
99.8±1.4
98.5±2.1
Sb
V
a
ND = Not detectable; Experimental conditions: sample pH = 6, sample volume = 20 mL,
replicates n =3
138
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
The validity of the Dowex 1-x8 SPE method was investigated by analyzing a Conostan
custom made oil based CRM. The results of the CRM certified values and those
determined with Dowex 1-x8 SPE method for Mo and V are presented in Table 9.7.
Satisfactory recoveries in the range 99.8% to 101% were obtained. The precision of the
measurements (n=3) expressed as % RSD ranged between 1.3% and 1.5 %. According to
the student t-test at the 95% confidence level, there was no significance difference between
the certified and determined concentration values.
Table 7.7. Concentrations (in µg L-1) of metal ions in gasoline samples determined by ICP
OES in sample solutions resulting from Dowex 1-x8 preconcentration procedure
Concentration (µg L-1)
Elements
Mo
V
a
Certified
RSDa (%)
Found
RSD (%)
1000
1000
1.0
1.0
998.2
1005
1.3
1.5
Recovery (%)
99.8
101
RSD= relative standard deviation, replicates n = 6
7.3.9 Application of the Dowex 1-x8 Separation and Preconcentration Procedure in
Commercial Gasoline Samples
Dowex 1-x8 SPE method was used to separate and preconcentrate Mo, Sb and V ions
in commercial gasoline samples collected from different petrol filling stations in
Johannesburg (South Africa). The results obtained are presented in Table 7.8. In general,
the concentrations of Sb were high (above 60 µg L-1) in almost all gasoline samples
irrespective of the source (manufacturer) except for 3-MCUG and 3-MFUG samples.
Molybdenum and vanadium concentrations were the highest in 5-MCUG and 2-MFUG
samples, respectively.
For comparison, the concentration of Mo, Sb and V in ten gasoline samples were also
determined by GFAAS after acid digestion (Table 7.9). The analyte concentrations
obtained by Dowex 1-x8 preconcentration method (Table 7.8) were in agreement with the
results obtained by GFAAS after acid digestion according to the paired t-test at 95%
confidence level: tcal= 0.14, 0.72 and 0.87 for Mo, Sb and V, respectively. In all the cases
tcal was lower than tcrit= 2.26 for Mo and Sb (n=10); tcrit= 2.44 for V (n=7). The
139
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
determination of Mo, Sb and V by GFAAS after acid digestion was used as an additional
procedure for quality check of the Dowex 1-x8 separation and preconcentration method.
The main advantage of the Dowex 1-x8 column method described in this study is that it
does not require rigorous acid digestion unlike the acid digestion method. In addition, the
column method is advantageous because it minimizes the risks of cross-contamination
during acid digestion. It should be noted that acid digestion followed by GFAAS
determination was taken as the standard method in this study.
140
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
Table 7.8. Concentrations (in µg L-1) of metal ions in gasoline samples determined by ICP OES in sample solutions resulting from Dowex 1x8 preconcentration procedure and GFAAS in sample solutions resulting from acid digestion procedure
Mo
c
1-MFUG
2-MCUGd
2-MFUG
3-MCUG
3-MFUG
4-MFUG
5-MCUG
5-MFUG
6-MCUG
6-MFUG
a
X  st
Dowex 1-x8-SPE
20.1±0.8
17.1±0.4
36.9±0.4
22.0±0.4
22.0±0.4
32.6±0.1
70.7±0.01
50.1±0.1
47.7±0.3
19.6±0.2
GFAAS
19.8±0.7
16.8±0.4
37.1±0.7
22.3±1.0
22.0±0.5
32.8±0.1
70.4±0.1
49.6±0.1
48.5±0.2
19.8±0.9
Sb
Dowex 1-x8-SPE
71.0±0.8
70.2±1.8
64.6±0.8
27.7±1.0
30.1±0.6
91.0±0.9
71.8±0.01
63.6±0.02
94.1±0.5
80.0±1.0
V
GFAAS
71.2±0.7
66.6±0.02
85.4±0.2
28.4±0.03
29.3±0.5
90.8±0.9
71.7±0.4
63.4±0.8
93.7±0.1
79.4±1.0
Dowex 1-x8-SPE
NDb
16.1±0.1
43.9±0.07
ND
ND
2.78±0.08
14.7±0.1
3.96±0.09
9.89±0.04
4.30±0.09
GFAAS
NDb
15.8±0.1
48.2±0.1
ND
ND
2.69±0.79
14.5±0.5
4.10±0.41
9.89±0.51
4.30±0.43
n , X: Average value (n = 3); t: student‘s t (P < 0.05); s: Estimation of the standard deviation, n: number of determinations; bND =
Not detectable; cMFUG = metal-containing unleaded gasoline; dMCUG = metal-free unleaded gasoline; 1-6 are the numbers allocated to the
six gasoline filling stations
141
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
7.4. CONCLUSION
The separation and preconcentration of Mo, Sb and V presented in this study
contributes to the growing field of fuel analysis and purification. The results obtained in
the preliminary studies demonstrated that Dowex 1-x8 was suitable for preconcentration of
Mo, Sb and V in organic matrices. Instead of using fresh resin for each analysis, the
reusability of the Dowex 1-x8 was found to be about 150 cycles after desorption and
regeneration treatment without any loss in its initial sorption performance. In addition, the
Dowex 1-x8 SPE method was successful in pre-concentrating metal ions from large sample
volume with a preconcentration factor of 120 achieved when using 600 mL of the sample
and 5.0 mL of final volume. The elution of metal ions from the resin column was
performed by 2.0 mol L-1 HNO3. The positive features of the present separation and
preconcentration method include relatively high selectivity, good precision and accuracy.
The Dowex 1-x8 SPE procedure was successfully applied for simultaneous determination
of trace amounts (µg L-1 range) of molybdenum, antimony and vanadium at in gasoline
samples.
7.5 REFERENCES
1. Amorim, F. A. C., Lima, D. C., Amaro, J. A. A., Valea, M. G. R. & Ferreira, S. L. C.
2007. Methods for vanadium determination in fuel oil by gf aas with
microemulsification and acid digestion sampling. Journal of Brazzilian Chemical
Society, 18, 1566-1570.
2. Brandão, G. P., De Campos, R. C., De Castro, E. V. R. & De Jesus, H. C. 2008.
Determination of manganese in diesel, gasoline and naphtha by graphite furnace
atomic absorption spectrometry using microemulsion medium for sample stabilization.
Spectrochimica Acta Part B: Atomic Spectroscopy, 63, 880-884.
3. Aucelio, R. Q. & Curtius, A. J. 2002. Evaluation of electrothermal atomic absorption
spectrometry for trace determination of Sb, As and Se in gasoline and kerosene using
microemulsion sample introduction and two approaches for chemical modification.
Journal of Analytical Atomic Spectrometry, 17, 242-247.
4. Takeuchi, R., Santos, A., Medeiros, M. & Stradiotto, N. 2009. Copper determination in
ethanol fuel samples by anodic stripping voltammetry at a gold microelectrode.
Microchimica Acta, 164, 101-106.
5. Chaves, E. S., F. G. Lepri, J. S. A. Silva, D. P. C. De Quadros, T. D. Saint’pierre And A.
J. Curtius 2008. Determination of Co, Cu, Fe, Mn, Ni and V in diesel and biodiesel
samples by ETV-ICP-MS. Journal of Environmental Monitoring, 10, 1211-1216.
142
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
6. Santos, D. S. S., M.G. A. Korn, M. A. B. Guida, G. L. Dos Santos, V. A. Lemos And L.
S. G. Teixeira 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by
Sequential Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase
Extraction. Journal of Brazzilian Chemical Society, 22, 552-557.
7. Pereira, J. S. F., Moraes, D. P., Antes, F. G., Diehl, L. O., Santos, M. F. P., Guimarães,
R. C. L., Fonseca, T. C. O., Dressler, V. L. & Flores, É. M. M. 2010. Determination of
metals and metalloids in light and heavy crude oil by ICP-MS after digestion by
microwave-induced combustion. Microchemical Journal, 96, 4-11.
8. Santelli, R. E., Bezerra, M. A., Freire, A. S., Oliveira, E. P. & De Carvalho, M. D. F. B.
2008. Non-volatile vanadium determination in petroleum condensate, diesel and
gasoline prepared as detergent emulsions using GF AAS. Fuel, 87, 1617-1622.
9. Bettinelli, M. & Tittarelli, P. 1994. Evaluation and validation of instrumental procedures
for the determination of nickel and vanadium in fuel oils. Journal of Analytical Atomic
Spectrometry, 9, 805-812.
10. Reyes, M. N. M. & Campos, R. C. 2005. Graphite furnace atomic absorption
spectrometric determination of Ni and Pb in diesel and gasoline samples stabilized as
microemulsion using conventional and permanent modifiers. Spectrochimica Acta
Part B: Atomic Spectroscopy, 60, 615-624.
11. Zeng, L. & Cheng, C. Y. 2009. A literature review of the recovery of molybdenum and
vanadium from spent hydrodesulphurisation catalysts: Part I: Metallurgical processes.
Hydrometallurgy, 98, 1-9.
12. de Souza, R. M., Saraceno, A. L., Duyck, C., Da Silveira, C. L. P. & Aucélio, R. Q.
2007. Determination of Fe, Ni and V in asphaltene by ICP OES after extraction into
aqueous solutions using sonication or vortex agitation. Microchemical Journal, 87, 99103.
13. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A.
M. & Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in
automotive gasoline by X-ray fluorescence after pre-concentration on cellulose paper.
Talanta, 72, 1073-1076.
14. Bettinelli, M., S. Spezia, U. Baroni And G. Bizzarri. 1995. Determination of trace
elements in fuel oils by inductively coupled plasma mass spectrometry after acid
mineralization of the sample in a microwave oven. Journal of Analytical Atomic
Spectrometry, 10, 555-560.
15. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
16. Roldan, P. S., Alcântara, I. L., Rocha, J. C., Padilha, C. C. F. & Padilha, P. M. 2004.
Determination of Copper, Iron, Nickel and Zinc in fuel kerosene by FAAS after
143
Chapter seven:
Preconcentration of Mo, Sb and V in gasoline
adsorption and pre-concentration on 2-aminothiazole-modified silica gel. Ecl. Quím.,
São Paulo, 29, 33-40.
17. Roldan, P. S., Alcântara, I. L., Padilha, C. C. F. & Padilha, P. M. 2005. Determination
of copper, iron, nickel and zinc in gasoline by FAAS after sorption and
preconcentration on silica modified with 2-aminotiazole groups. Fuel, 84, 305-309.
18. Pyrzyñska, K. & Joñca, Z. 2000. Multielement Preconcentration and Removal of Trace
Metals by Solid-Phase Extraction. Analytical Letters, 33, 1441 - 1450.
19. Zhang, L., Chang, X., Hu, Z., Zhang, L., Shi, J. & Gao, R. 2010. Selective solid phase
extraction and preconcentration of mercury(II) from environmental and biological
samples using nanometer silica functionalized by 2,6-pyridine dicarboxylic acid.
Microchimica Acta, 168, 79-8.
20. Pohl, P., Bogdal, Z. & Prusisz, B. 2005. Preconcentration and Fractionation of Cd, Co,
Cu, Ni, Pb and Zn in Natural Water Samples Prior to Analysis by Inductively Coupled
Plasma Atomic Emission Spectrometry. Microchimica Acta, 150, 253-259.
21. Y. Nakamoto, T. Ishimaru, N. Endo, K. Matsusaki, Anal. Sci. 20 (2004) 739–741.
22. Brandão, G. P., De Campos, R. C., De Castro, E. V. R. & De Jesus, H. C. 2007.
Determination of copper, iron and vanadium in petroleum by direct sampling
electrothermal atomic absorption spectrometry. Spectrochimica Acta Part B: Atomic
Spectroscopy, 62, 962-96.
23. Cassella, R. J., Brum, D. M., De Paula, C. E. R. & Lima, C. F. 2010. Extraction
induced by emulsion breaking: a novel strategy for the trace metals determination in
diesel oil samples by electrothermal atomic absorption spectrometry. Journal of
Analytical Atomic Spectrometry, 25, 1704-1711.
24. Cinosi, A., Andriollo, N., Pepponi, G. & Monticelli, D. 2011. A novel total reflection
X-ray fluorescence procedure for the direct determination of trace elements in
petrochemical products. Analytical and Bioanalytical Chemistry, 399, 927-933.
25. dos Santos, D. S. S., Teixeira, A. P., Korn, M. G. A. & Teixeira, L. S. G. 2006.
Determination of Mo and V in multiphase gasoline emulsions by electrothermal
atomic absorption spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy,
61, 592-595.
26. Cassella, R., Barbosa, B. S., Santelli, R. & Rangel, A. 2004. Direct determination of
arsenic and antimony in naphtha by electrothermal atomic absorption spectrometry
with microemulsion sample introduction and iridium permanent modifier. Analytical
and Bioanalytical Chemistry, 379, 66-71.
27. de Souza, R. M., Meliande, A. L. S., Da Silveira, C. L. P. & Aucélio, R. Q. 2006.
Determination of Mo, Zn, Cd, Ti, Ni, V, Fe, Mn, Cr and Co in crude oil using
inductively coupled plasma optical emission spectrometry and sample introduction as
detergentless microemulsions. Microchemical Journal, 82, 137-141.
144
CHAPTER EIGHT:
MULTIVARIATE OPTIMIZATION OF DUAL-BED SOLID PHASE EXTRACTION
FOR PRECONCENTRATION OF Ag, Al, As AND Cr IN GASOLINE PRIOR TO
INDUCTIVELY COUPLED PLASMA OPTICAL EMISSION SPECTROMETRIC
DETERMINATION
ABSTRACT
In this work, a dual-bed resin solid phase extraction (SPE) for preconcentration of Ag, Al,
As and Cr prior to their inductively coupled plasma-optical emission spectroscopy (ICP OES)
determination has been developed. Dowex 50W-x8 and Dowex 1-x8 packed in a column
were used as metal ion sorbents. The optimization of the dual-bed SPE procedure was carried
out using a two level full factorial design with three central points. Under optimized
conditions, the limits of detection and quantification (n = 21) ranged from 0.16-0.22 and
0.52-0.76 µg L-1, respectively. Enrichment factors of 100, 130, 130 and 150 and relative
standard deviations (n = 15) of 1.2, 2.0, 1.8 and 1.3% were obtained in the determination of
Ag, Al, As and Cr, respectively. The validity of the proposed method was checked by
applying the standard addition method and the recoveries at the 20 μg L−1 level using both
inorganic and organic metal standards ranged from 95 to 99%. The proposed method
presented an analytical throughput of about 18 samples per hour and was applied for the
determination of metal ions in ten gasoline samples. In addition, the accuracy of the method
was evaluated using microwave-assisted digestion method and the results were not
significantly different (at 95% confidence level).
Keywords: Dual-bed resin, metal ions, factorial design, separation and preconcentration, ICP
OES, gasoline
8.1 INTRODUCTION
Controlling the levels of metals in petroleum products such as gasoline is a critical step in
the petrochemical industry because they act as catalyst poisons thus cause deleterious effects
on the refinery and processing operations unless they are added as additives.1,2 Therefore, it is
crucial to accurately determine metal content in liquid fuels which are also the main sources
of energy for vehicles. Other effects of metal ions (even in trace concentrations) in liquid
fuels are reported in the literature.1,3-5 These include (i) poor fuel performance (ii) decrease in
the engine durability and efficiency and (iii) environmental pollution caused by the release of
145
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
toxic metals into the atmosphere during fuel combustion.1,3-5 Therefore, the development of
sensitive and selective analytical techniques for the determination of metals in gasoline is one
of the most important aspects of quality control in petroleum industries.1 These techniques
must be fast, simple, precise, accurate and economical to be easily employed in routine
procedures. In addition, since metal ions in liquid fuel samples are usually present in trace
levels, the analytical methods must be capable of resulting in high enrichment/preconcentration factors enough to cope with the demands.1
Analytical methods based on electrothermal atomic absorption spectrometry (ETAAS)
are popular because they are associated with high sensitivity and tolerance to high organic
matrix loads.2 However, ETAAS is not common in routine analysis because of its low sample
throughput as compared to inductively coupled plasma-based techniques.2 Inductively
coupled plasma optical emission spectrometry (ICP OES) is widely used in routine
quantification of metal ions in different sample matrices. This technique is attractive due to
its multielement capability, relative sensitivity, wide linear range and high sample
throughput. However, the direct introduction of fuels into the plasma requires special care, as
the organic load may de-stabilize or extinguish the plasma.6-9
Hitherto, different sample preparation approaches for determination of metal ions in fuels
has been developed to overcome the problems associated with ICP OES and are reported in
the literature. These methods include conventional ashing and acid dissolution,10 microwave
digestion,11,12
dilution
with
organic
solvents,13
emulsion/microemulsion14,15
and
preconcentration using solid phase extraction.16,17 However some of these sample preparation
methods have some limitations, for instance conventional ashing and acid dissolution
methods are time-consuming and also volatile elements may easily escape (be lost).18
Microwave digestion methods may be a good alternative to these methods and solve the
problem of volatilization, but they increase the risks of cross-contamination. In addition, the
use of some concentrated acids (except ultra pure) could increase the blank values and cannot
be supported by some analytical techniques such as ICP OES.19 To solve the problem of
introducing concentrated acid the digested sample are normally diluted before introduction to
ICP OES. However, this becomes a challenge because the analytes of interest in real samples
are in trace levels. Since small amount of samples (normally 1 mL) are used, the
concentration of analytes in the samples become too diluted to be determined by ICP OES.
Emulsion/microemulsion technique is one of the most promising approaches due to its short
preparation time and the low risk of analyte losses by volatilization or sorption. However, its
146
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
disadvantage is the low stability which then affects the sensitivity and reproducibility of the
analytical instrumental signal.19 Dilution with organic solvents is one of the simplest sample
pretreatment procedures but does not reduce the problem of organic loading and plasma
destabilization or extinction in case of the ICP techniques.9
To overcome these difficulties associated with sample pretreatment methods, an accurate
and reliable analytical procedure based on separation and preconcentration of analytes prior
to analysis in fuel samples, is required. Preconcentration of the analytes from the organic
matrices combines the advantages of separating the analyte from the complex fuel matrix,
transferring it to an aqueous phase and enriching it at the same time.9 Procedures based on
solid phase extraction (SPE) for the separation and preconcentration of trace elements in
gasoline and fuel kerosene are reported in the literature.16,17,20
Recently, chemometric techniques have been used for optimization of different analytical
methods. This is because, these techniques allow more than one variable to be optimized
simultaneously.21 The advantages of multivariate techniques include reduction in the number
of required experiments, thus, lowering reagent consumption and significantly less laboratory
work. They are faster to implement and more cost-effective than traditional univariate
approaches.22,23 In addition, chemometric methods are able to generate mathematical models
that permit assessment of the relevance and statistical significance of factors being studied,
and evaluation of interaction effects between them (factors). 21,22 Full factorial design is one
of the well-known statistical processes for multivariate optimization and is widely applied in
analytical chemistry. This is due to its effectiveness in the identification of significant
variables and the best conditions of an experimental procedure.21
The aim of this work was to investigate the analytical performance and the potential
applicability of a dual-bed resin column for Ag, Al, As and Cr determination in gasoline via
off-line SPE/ICP-MS system. Information about the specific forms of elements in the fuel
samples is limited. In addition, some element can exist in more than one oxidation states and
also the chemical form of an element might change due the change in sample pH. Therefore,
the use a dual-bed column proposed in this study will be advantageous because different
metal species can be retained by either the cationic or anionic exchange resin, thus, enabling
the total metal analysis in fuel sample. A full two-level factorial design with a central point
was used for optimization of experimental variables (pH, eluent concentration and sample
flow rate) that affect the retention/desorption of metal ions. To the best of our knowledge,
this is the first time that dual-bed resin column and the optimized preconcentration method
147
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
are proposed for Ag, Al, As and Cr determination in gasoline. In addition, this study offers a
simple system with no need of acid digestion prior to metal ion determination.
8.2 EXPERIMENTAL
8.2.1 Instrumentation
Metal ions (Ag, Al, As and Cr) were determined using a Spectro Arcos 165 ICP OES
(SPECTRO Analytical Instruments, GmbH, Germany) equipped with Cetac ASX-520
autosampler. The ICP OES operating conditions are listed in Table 8.1. Sample introduction
was achieved using a pneumatic cross-flow nebulizer mounted onto a Scott double-pass spray
chamber. Sample solutions were pumped to the nebulizer using a built in four channel
peristaltic pump. The most prominent atomic and ionic analytical spectral lines of the metal
ions were selected for investigation, that is, Ag 329.068 nm, Al 167.078 nm, As 193.759 nm
and Cr 283.563 nm. Solid phase extraction was carried out in a VacMaster-24 sample SPE
station (Supelco, PA, USA). The latter was used to control the sample loading and elution
flow rates. The microwave digestions were carried out in an Ethos D (Milestone, Sorisole,
Italy) with maximum pressure 1450 psi and maximum temperature 300 ◦C.
Table 8.1. The operating parameters of determination of elements by ICP OES
RF power
1400 W
Plasma argon flow rate
Auxiliary argon flow rate
Nebulizer argon flow rare
Sample aspiration rate
Replicate measurements (n)
13 L min-1
2.00 L min-1
0.95 L min-1.
2.0 mL min-1
3
8.2.2 Reagents, Solutions and Samples
All reagents were of analytical grade unless otherwise stated and double distilled
deionised water (Millipore, Bedford, MA, USA) was used throughout the experiments.
Absolute ethanol (99.9%) used to prepare model solutions and suprapur 30% hydrogen
peroxide (H2O2) used for the acid digestion procedure were obtained from Merck,
(Darmstadt, Germany). Spectrascan stock solutions (1000 mg L-1) of Ag, Al, As and Cr
(Industrial Analytical Pty, Johannesburg, Ltd, South Africa) were used to prepare the
148
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
working solutions (prepared in organic phase) for SPE at concentrations of 10 µg L-1 for each
analyte. Working solutions, as per the experimental requirements, were freshly prepared from
the stock solution for each experimental run. A Spectrascan multi-element 100 mg L-1
standard solution (Industrial Analytical Pty, Ltd, Johannesburg, South Africa) was used to
prepare working standard solutions at concentrations of 0-120 µg L-1 for measurements of
concentrations of analytes in all model and sample solutions. The cation exchangers used in
this study as packing materials, that is, Dowex 1-x8 (Chloride form) and Dowex 50W-x8
(sodium form) as well as solutions of nitric acid at concentration range of 1.0-4.0 mol L-1
used for the elution of the analytes from the columns, were prepared from ultrapure
concentrated acid (65%), were purchased from Sigma-Aldrich (St. Loius, MO, USA). The pH
adjustments were performed with glacial acetic acid and ammonia solutions (Sigma-Aldrich,
St. Loius, MO, USA). Ten gasoline samples from different local filling stations were used for
method development and validation.
8.2.3 Preparation of a Two Bed Column
The two bed resin was prepared in a 6.5 mL polyethylene column. The preparation
procedure was carried out as follows: the first bed was prepared by placing the slurry
(prepared in double distilled deionised water) of 750 mg Dowex 50W-x8 resin in the column
that has a porous frit at the bottom. Another frit was then placed on the top of the packing
material. The second bed was prepared by placing the slurry of 750 mg Dowex 1-x8 resin on
the top of the first bed and a porous frit was placed on top of the packing material to hold and
confine the adsorbents within the designated capacity/volume. The total height of the duo
sorbent bed was approximately 4 cm in the column. The columns were sequentially washed
with 10 mL of 3.0 mol L-1 HNO3 and double distilled deionised water followed by
conditioning with 10 mL ammonium acetate buffer (1.0 M, pH 9.0) and then 10 mL of
ethanol.
8.2.4 Preconcentration and Recovery of Ag, Al, As and Cr in Synthetic Gasoline
Solution
The model solutions were prepared as follows: 10 mL of synthetic gasoline was placed in
a 100 mL polyethylene volumetric flask followed by addition of 1.0 mL of 1.0 mg L-1 of
metal ions solution and made up to the mark with ethanol to obtain 10 µg L-1 of each metal
149
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
ion. The mixture was homogenized by shaking. An aliquot of 50 mL of the model metal
solutions were passed through an ion exchange column at an appropriate flow rate. The
columns were washed with 10 mL of double distilled deionised water to remove excess
organic solution, followed by 5.0 mL of ammonium acetate buffer solution to remove major
cations (Na, Ca, K, etc) [17]. Lastly the metal ions were eluted with 5 mL of appropriate
concentration of HNO3 solutions. All fractions obtained during the elution stage were
collected separately and analyzed by ICP OES. The same procedure was applied to the blank
solutions. In the case of real sample analysis, a gasoline–ethanol mixture was prepared
according to Chaves et al.13 An aliquot of 1.0 mL of gasoline sample was placed in a 100 mL
polypropylene volumetric flask and 500 μL of concentrated HNO3 was added to the sample
which was then diluted with ethanol. In between experiments, the resins in the column were
washed with 20 mL of double distilled deionised water followed by 10 ml of 1.0 M NaOH
(this was done in order to keep the resin in sodium cation and hydroxide anion forms) and
stored for the next experiment.
8.2.5 Optimization Approach
The optimization of the separation and preconcentration method was carried out using a
full factorial design (23) involving three variables i.e. pH, eluent concentration (EC) and
sample flow rate (SFR). The latter were considered as factors. The latter had two levels
namely minimum and maximum including a central point, as shown in Table 8.2. Each factor
was chosen according to data from previous experiments. All the experiments were carried
out in random. The experimental data was processed by using the Minitab version 15
software program.
Table 8.2. Factors and levels used in 23 factorial design for separation and preconcentration
of metal ions
Variable
pH
EC (mol L-1)
SFR (mL min-1)
Low level (-1)
3.0
1.0
1.0
Central point (0)
6.0
2.5
2.0
150
High level (+1)
9.0
4.0
3.0
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
8.2.6 Comparative Method
The Microwave acid digestion procedure was carried out according to Kowalewska et
al.24 Briefly, 5.0 mL of the gasoline sample was placed into a Teflon vessel followed by 6 mL
HNO3 (65%) and 2.0 mL H2O2 (30%). The vessels were inserted into a microwave unit and
heated according to the conditions recommended by the manufacturer. The microwave
program used was as follows: Step 1: 25-200C for 10 min at 1000 W; Step 2: 200C for
1000 W for 10 min giving a total time of 20 min and cool down time was approximately 10
min. The digested material was left to cool down to room temperature. After cooling, the
vessels were opened and 2 mL of concentrated HNO3 and 2 mL of hydrogen peroxide were
added. The heating program was then repeated. This step was done in order minimize
incomplete mineralization of the organic matrix. Finally, the Teflon vessel contents were
cooled down to room temperature and quantitatively transferred to a 50 ml calibration flask.
The samples were spiked with 20 µg L-1 followed by the addition of 1 mL of concentrated
nitric acid and the flask was filled up to the mark using double distilled deionised water. The
samples were spiked in order to increase the concentration levels of metal ions (such as Cr) in
the final solution so that they can be detected by ICP OES. This was done because small
volumes of the gasoline samples were used. The distilled deionised was submitted to the
same procedure and used as the blank. The samples were then analyzed with ICP OES.
8.3 RESULTS AND DISCUSSION
8.3.1 Factorial Design
The factors affecting the performance of dual-bed solid phase extraction for separation
and preconcentration of Ag, Al, As and Cr ions in gasoline samples were investigated. The
variables (factors) chosen for the optimization of the preconcentration system included
sample pH, eluent concentration (EC) and sample flow rate (SFR). In order to determine the
main factors of the preconcentration system, a two-level full factorial design (23) with three
replicates of the central point (CP) was performed. The percentage recovery of each metal ion
was used as the analytical response. Table 8.3 shows the experimental design matrix and the
results derived from each run for Ag, Al, As and Cr, respectively. It can be seen from this
table that the design matrix resulted to 11 sets of experiments and the preconcentration and
determination of analytes is done on the same day. It should be noted that if univariate
151
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
technique, which simply means monitoring one factor at time, was used, each variable will
require about 8 sets of experiments which translate to a total of about 24 experiments. In
addition, after each set of experiments, detection of the analytes is required in order to
proceed to the next variable and this may take a week to complete the optimization process.
Since with classical design, the optimization of each variable is done when other are kept
constant, the results obtained by this process may lead to ambiguous results and interpretation
because the interactive effects among the variables are not examined.
Table 8.3. Design matrix and the results of Ag, Al, As and Cr
Run
pH
EC (mol L-1)
SFR (mL min-1)
+1
+1
+1
1
+1
+1
-1
2
+1
-1
+1
3
+1
-1
-1
4
-1
+1
+1
5
-1
+1
-1
6
-1
-1
+1
7
-1
-1
-1
8
0
0
0
9
0
0
0
10
0
0
0
11
EC= eluent concentration; SFR= sample flow rate
Recovery (%)
Ag
30.34
32.45
38.59
40.25
71.29
77.02
78.87
79.82
97.48
97.48
97.31
Al
44.71
33.02
20.18
60.67
49.73
71.09
72.05
55.30
98.56
98.45
99.01
As
60.82
59.87
98.96
75.76
45.91
19.01
32.92
42.12
60.25
61.17
62.16
Cr
99.11
96.62
46.51
36.88
65.44
70.32
37.43
44.78
59.04
59.41
59.26
Analysis of variance (ANOVA) and p-values were used to investigate the significance of
the effects on the dual-bed preconcentration system. The Pareto chart of main effects and
their interactions produced from ANOVA results, is shown in Fig. 8.1. The bar lengths of the
Pareto chart are proportional to the absolute value of the estimated effects and they help in
comparing the relative importance of effects [25]. It can be seen from Fig. 8.1 that sample pH
was highly significant for all the metal ions studied except for Al. This is because ion
exchange is chiefly governed by the pH of the solution. This is partly because hydrogen ions
also strongly compete for active sites and the solution pH influences the ionization of surface
functional groups. Therefore, in can be concluded that sample pH played a significant role in
the retention of the analytes. Other factors such as sample flow rate and eluent concentration
had little or no statistical significant effect on the extraction of Ag, Al, As and Cr. In view of
the information obtained from Table 8.3 and ANOVA results, silver and aluminium can be
152
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
preconcentrated simultaneously. This is because they both had the highest percentage
recoveries at the same experimental conditions (experiments 9-11). Therefore, the optimum
sample pH, eluent concentration and sample flow rate chosen for simultaneous separation and
preconcentration of Ag and Al in a gasoline matrix concur with the conditions established by
experiment (run) 9-11. For As and Cr, the highest recoveries were observed at experiments 1
and 3, respectively. Therefore, the optimum sample pH, eluent concentration and sample flow
rate for preconcentration of As were 9.0, 1.0 mol L-1 and 3.0 mL min-1, respectively. For
separation and preconcentration of Cr on the other hand, the optimum sample pH, eluent
concentration and sample flow rate were selected to be 9.0, 4.0 mol L-1 and 3.0 mL min-1,
respectively. It can be seen from the optimum conditions that the eluent concentration was
differed depending on the analyte. This might be due to the different behaviour of metal ions
which results from the difference in charge density. For instance, metal ions that have a
higher charge density are strongly bound by the resins functional groups; therefore, high acid
concentrations are required to strip off the retained metal ions.
153
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
Fig. 8.1. Pareto chart of standardized effects for variables in the separation and preconcentration of silver (A); aluminium (B), arsenic
(C) and Cr (D).
154
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
8.3.2 Effect of Sample Volume
The concentrations of metals in real gasoline samples are typically at trace levels. To
solve the problem of detectability of trace levels, a preconcentration procedure can be
employed where large sample volumes are used to obtain high enrichment factors. In
addition, the optimization of sample volume helps in evaluating the saturation point of an
absorbent. Therefore, the capacity of the column was examined by loading, 50-1000 mL
volumes of synthetic gasoline solutions containing 15 µg L-1 of metal ions. The recoveries of
the analytes from different volumes of synthetic gasoline solutions are presented in Fig. 8.2.
The recoveries were found to be stable up to 500, 650, 650 and 750 mL for Ag, Al, As and
Cr, respectively. Therefore, the experimental preconcentration factors, defined as the ratio of
the analyte concentrations before and after preconcentration, were calculated to be 100, 130,
130 and 150 for Ag, Al, As and Cr, respectively. The observed decrease in recoveries for
each metal ion was probably due to the excess analytes, loaded over the column capacity as
the sample volume increased. As a compromise, 100 mL was chosen for further investigation.
This was done in order to speed up sample analysis.
Fig. 8.2. Effect of sample volume on the recoveries of metal ions
155
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
8.3.3 Column Regeneration
The regeneration of the dual bed column was investigated by monitoring the changes in
the recoveries of Ag, Al and As through several retention-elution cycles. In each cycle, 50
mL multi-element solution (20 µg L−1) was passed through the column and then eluted with 5
mL of 3.0 mol L-1 HNO3. The procedure was carried out 40 times per day for 5 subsequent
days (a maximum of 200 runs) without any changes in the performance.
8.3.4 Analytical Performances of the Dual-Bed SPE Method
Calibration solutions were prepared with multi-element standards containing 0, 5, 20, 40,
80, 100 and 120 µg L−1 in 100 mL volume. Each solution was passed through the column and
collected in 5 mL of HNO3. The calibration curve was linear (r2 = 0.9981–0.9994) for Ag, Al,
As and Cr. The IUPAC limit of detection (LOD) and limit of quantification (LOQ) under
optimized conditions were calculated from the signals of 21 successive measurements of the
blank (100 mL) and the slope (m) of the calibration curve. The LOD was defined as the
lowest concentration of an analyte giving signals equal to three times standard deviation (SD)
of blank signal divided by the slope of the calibration curve that the analytical technique can
detect (3SD/m). The LOQ, on the other, was defined as the to the smallest concentration of an
analyte giving signals equal to ten times the standard deviation of blank signal divided by the
slope of the calibration curve which can be accurately and precisely measured with an
analytical procedure (10SD/m). Under the optimum conditions, the LOD were found to be
0.17, 0.16, 0.18 and 0.22 µg L-1, for Ag, Al, As and Cr, respectively. On the other hand, the
LOQ values we determined as 0.57, 0.52, 0.59 and 0.76 µg L−1 for Ag, Al, As and Cr,
respectively. The instrumental detection limits (IDL) were 0.6, 1.0, 2.0 and 0.2 µg L−1 for Ag,
Al, As and Cr, respectively. The precision of the preconcentration system, calculated as the
relative standard deviation (% RSD; n= 15), was less than or equal to 2% with mean
recoveries of 97.3±1.2%, 99±2.0%, 98.7 ±1.8 % and 97.6±1.3%, for Ag, Al, As and Cr
respectively. The time required for preconcentration of 100 mL of sample was obtained to be
6 under the following conditions: percolation for 300 s at a flow rate 2-3 mL min−1; elution
for 30 s at a flow rate 3.0 mL min−1; washing and conditioning for 30 s. However, the sample
preconcentration was performed in triplicate and they were all carried out at the same time.
Therefore, the overall time for preconcentration of triplicates was approximately 10 min.
Hence, the throughput sample was approximately 18 samples h-1.
156
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
8.3.5 Validation of the Dual-Bed SPE Method
Due to the absence of certified reference material (CRM) that is similar to the
investigated samples, the validity of the proposed dual-bed SPE system was examined by
using addition/recovery experiments. Gasoline sample (G1) was spiked with organic and
inorganic standard solutions. Additionally, the purpose of spiking the gasoline sample with
organic and inorganic standard solutions was to evaluate the ion exchange efficiency of resins
to different metal species in fuel samples. This is because the speciation of trace elements in
petroleum products is not fully known and different species may display different adsorption
behaviors.16 The recoveries of analytes spiked into the gasoline sample are presented in Table
8.4. It can be seen from this table that the recovery values calculated for the standard
additions (organic and inorganic forms) for the investigated metal ions were greater than or
equal to 95%. These results confirmed the accuracy of the proposed method and insignificant
matrix effects, taking into consideration that the recoveries were in the range from 95-99%.
In addition, since similar percentage recoveries were obtained for organic and inorganic
forms, this implies that a dual-bed SPE system may be used for preconcentration of trace
elements in their inorganic or organic forms.
157
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
Table 8.4. Determination of Ag, Al, As and Cr (µg L-1) in gasoline sample spiked with inorganic and organic standard solutions (mean
± standard deviation, n= 3)
Ag
ISa
MOSb
a
Addedc
0
5
10
5
10
Foundc
27.0±0.1
31.8±1.0
36.7±1.0
31.8±1.1
36.6±0.9
Al
Recovery
96.4±1.5
97.1±1.4
95.2±1.2
95.7±1.1
Found
57.0±0.1
61.9±0.6
66.8±1.3
61.9±0.8
66.9±1.6
As
Recovery
97.8±1.3
98.3±1.3
98.2±1.2
98.7±1.8
Found
157±2
162±2.
167±2
162±2
167±2
IS: Inorganic standard; bMOS = metallo-organic standard; cConcentration in µg L-1
158
Cr
Recovery
99.0±2.3
98.8±1.8
95.4±1.7
96.8±1.2
Found
60.6±0.4
65.5±1.2
70.5±1.3
65.4±0.8
70.5±0.8
Recovery
97.4±2.1
99.1±2.5
96.8±1.8
98.9±1.1
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
8.3.6 Analysis of Real Samples
The proposed method was applied for determination of Ag, Al, As and Cr in ten gasoline
samples. The obtained metal ion concentrations are presented in Table 8.5. In general, the
concentrations of Al were higher in all the samples compared to other studied metal ions. The
higher concentration might be due to the abundance of Al in the Earth’s crust. It was observed
that Al and As were always present in all the samples, whereas Cr was not detectable in G7 and
G8 samples. It can be seen from Table 8.5 that the concentrations of Ag in most of the samples
are below the LOQ (calculated using 10SD/m) of the proposed method. The use of dual-bed SPE
method prior to ICP OES determination showed an improvement for both sensitivity and LOD
for metal ions those that were in ultra-trace levels. This can be an important feature in the
analysis of the fuel samples.
In view of the fact that there is no gasoline or similar reference material available with
certified values for the studied metal ions, it was crucial to use an independent sample
pretreatment technique for further validation (microwave-assisted digestion). It should be noted
that ICP OES after microwave-assisted digestion was used as a reference method. The results
obtained for gasoline samples by the proposed method did not differ significantly from the
values obtained by the reference method according to the paired Student's t-test at the 95%
confidence level (tcal = 0.98, 0.46, 0.16 and 1.40 for Ag, Al, As and Cr, respectively). In all the
cases, tcal was lower than tcrit = 3.18, n = 4 for Ag; tcrit = 2.26 for Al and As, n=10; tcrit = 2.44 for
Cr, n = 7. In addition, the statistical F-test showed that the precisions of the proposed analytical
methods were not significant at 95% confidence level. The overall results are satisfactory and
show that the dual-bed SPE method has provided accurate results.
Although the results obtained by the proposed method were not significantly different to
those obtained by the comparative method, dual-bed SPE method displays more advantages
compared to microwave-assisted digestion method. For instance, the proposed method does not
require the sample to be subjected to any drastic pretreatment such as concentrated acid
heating.17 Furthermore, the use of concentrated acids in microwave-assisted digestion could
increase the blank metal values. In addition, the use of concentrated acids is not suitable for use
ICP OES. Therefore, a subsequent step could be necessary to dilute or remove the excess
159
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
acid.12,26 However, this becomes a challenge because the analytes of interest in real samples are
in trace levels and since small amount of samples (1 mL) are used, the concentration of analytes
in the samples become too diluted to be determined by ICP OES. In contrast, the samples
prepared by the proposed method are compatible with ICP OES without further dilutions. In
view of the above limitations of microwave digestion method, the dual-bed preconcentration
method is advantageous because it minimizes the risks of incomplete mineralization of the
organic matrix and cross-contamination.19 Furthermore, in terms of sample through put, the dualbed SPE method had a higher throuput (18 samples h-1) compared to microwave-assisted
digestion method (10 samples h-1).
160
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
Table 8.5. The determination of Ag, Al, As and Cr in different gasoline samples using dual-bed SPE/ICP OES and MAD/ ICP OES
methods
Samples
Ag (µg L-1)
G1
G2
G3
G4
G5
G6
G7
G8
G9
G10
DB-SPEa
27.0± 0.1
ND
4.67±0.16
4.14±0.22
22.8±0.4
NDc
ND
ND
ND
ND
a
Al (mg L1)
MADb
26.6±0.1
ND
4.48±0.12
3.83±0.34
23.2±0.7
ND
ND
ND
ND
ND
DB-SPE
1469±20
538±5
799±10
710±8
626±7
1021±7
1089±19
748±10
1018±14
854±12
As (µg L-1)
MAD
1481±21
582±5
801±12
696±8
631±8
1018±10
1091 ±20
753±9
1015±14
848±11
DB-SPE
157±1.8
133±2
112±3
98.4±1.0
47.9±0.4
69.3±0.9
83.6±0.5
52.4±0.7
79.3±0.7
57.9±0.3
Cr (µg L-1)
MAD
162±2
131±3
111±3
98.6±1.1
48.1±0.5
70.0±0.8
82.8±0.8
51.9±0.5
78.8±0.6
58.2±0.3
DB-SPE
60.6±0.4
75.1±0.2
12.1±0.2
64.1±0.5
ND
3.28±0.10
ND
ND
31.1±0.2
20.5±0.3
DC-SPE = dual column solid phase extraction; bMAD = microwave- assisted digestion; cND = not detectable
161
MAD
59.9±0.5
75.5±0.4
11.9±0.2
63.9±0.6
ND
2.91±0.34
ND
ND
31.3±0.2
19.7±0.3
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
8.4 CONCLUSION
A full factorial design used for optimization of dual-bed SPE column system allowed
the establishment of optimum conditions for separation and preconcentration of metal ions
in gasoline samples. In addition, factorial design helped in evaluating the interaction
between the investigated factors and their effect on the analytical response (recovery). The
optimum conditions for retention and elution of metal ions with respect to sample pH,
eluent concentration and sample flow rate were as follows:(i) for Ag and Al-6.0, 2.5 mol L1
and 2.0 mL min-1; (ii) for As-9.0, 1.0 mol L-1 and 3.0 mL min-1 and for Cr- 9.0, 4.0 mol
L-1 and 3.0 mL min-1. The optimized dual-bed SPE procedure proved to be suitable for
total preconcentration of metal ions in gasoline samples. In addition, the preconcentration
step permitted the elimination of the organic matrix, thus, avoiding the need for digestion
of the samples before ICP OES determination. The proposed method was applied in the
determination of Ag, Al, As and Cr in ten real gasoline samples purchased from different
filling stations in Johannesburg, South Africa. The dual-bed SPE method can be considered
as a superior method compared to or to other sample pretreatment techniques such as acid
digestion because it combines relatively low LOD and LOQ values obtained in the range
0.16-0.22 and 0.52-0.76 µg L-1, respectively for all metals, with higher sample throughput
of 18 samples h-1.
8.5 REFERENCES
1. Cassella R. J., Brum D. M., Lima C.F., Caldas L. F. S. & de Paula C. E. R. 2011.
Multivariate optimization of the determination of zinc in diesel oil employing a novel
extraction strategy based on emulsion breaking. Analytica Chimica Acta, 690, 79-85.
2. Donati, G. L., Amais, R. S., Schiavo, D. & Nobrega, J. A. 2013. Determination of Cr,
Ni, Pb and V in gasoline and ethanol fuel by microwave plasma optical emission
spectrometry. Journal of Analytical Atomic Spectrometry, 28, 755-759.
3. Saint'pierre, T. D., Dias, L. F., Maia, S. M. & Curtius, A. J. 2004. Determination of Cd,
Cu, Fe, Pb and Tl in gasoline as emulsion by electrothermal vaporization inductively
coupled plasma mass spectrometry with analyte addition and isotope dilution calibration
techniques. Spectrochimica Acta Part B: Atomic Spectroscopy, 59, 551-558.
4. Sousa, J. K. C., Dantas, A. N. D. S., Marques, A. L. B. & Lopes, G. S. 2008.
Experimental design applied to the development of a copper direct determination
method in gasoline samples by graphite furnace atomic absorption spectrometry. Fuel
Processing Technology, 89, 1180-1185.
162
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
5. Sousa, J. K. C., Dantas, A. N. D. S., Marques, A. L. B. & Lopes, G. S. 2008.
Experimental design applied to the development of a copper direct determination
method in gasoline samples by graphite furnace atomic absorption spectrometry. Fuel
Processing Technology, 89, 1180-1185.
6. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001..
7. Bettinelli, M., S. Spezia, U. Baroni And G. Bizzarri. 1995. Determination of trace
elements in fuel oils by inductively coupled plasma mass spectrometry after acid
mineralization of the sample in a microwave oven. Journal of Analytical Atomic
Spectrometry, 10, 555-560.
8. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A.
M. & Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in
automotive gasoline by X-ray fluorescence after pre-concentration on cellulose paper.
Talanta, 72, 1073-1076.
9. Korn, M. D. G. A., Dos Santos, D. S. S., Welz, B., Vale, M. G. R., Teixeira, A. P.,
Lima, D. D. C. & Ferreira, S. L. C. 2007. Atomic spectrometric methods for the
determination of metals and metalloids in automotive fuels - A review. Talanta, 73, 111.
10. Ekanem, E. J., Lori, J. A. & Thomas, S. A. 1997. The determination of wear metals in
used lubricating oils by flame atomic absorption spectrometry using sulphanilic acid as
ashing agent. Talanta, 44, 2103-2108.
11. Sant’ana, F. W., Santelli, R. E., Cassella, A. R. & Cassella, R. J. 2007. Optimization of
an open-focused microwave oven digestion procedure for determination of metals in
diesel oil by inductively coupled plasma optical emission spectrometry. Journal of
Hazardous Materials, 149, 67-74.
12. Pereira, J. S. F., Moraes, D. P., Antes, F. G., Diehl, L. O., Santos, M. F. P., Guimarães,
R. C. L., Fonseca, T. C. O., Dressler, V. L. & Flores, É. M. M. 2010. Determination of
metals and metalloids in light and heavy crude oil by ICP-MS after digestion by
microwave-induced combustion. Microchemical Journal, 96, 4-11.
13. Chaves, E. S., De Loos-Vollebregt, M. T. C., Curtius, A. J. & Vanhaecke, F. 2011.
Determination of trace elements in biodiesel and vegetable oil by inductively coupled
plasma optical emission spectrometry following alcohol dilution. Spectrochimica Acta
Part B: Atomic Spectroscopy, 66, 733-739.
14. de Souza, R. M., Meliande, A. L. S., Da Silveira, C. L. P. & Aucélio, R. Q. 2006.
Determination of Mo, Zn, Cd, Ti, Ni, V, Fe, Mn, Cr and Co in crude oil using
inductively coupled plasma optical emission spectrometry and sample introduction as
detergentless microemulsions. Microchemical Journal, 82, 137-141.
163
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
15. Sousa, J. K. C., Dantas, A. N. D. S., Marques, A. L. B. & Lopes, G. S. 2008.
Experimental design applied to the development of a copper direct determination
method in gasoline samples by graphite furnace atomic absorption spectrometry. Fuel
Processing Technology, 89, 1180-1185.
16. Santos, D. S. S., M.G. A. Korn, M. A. B. Guida, G. L. Dos Santos, V. A. Lemos And
L. S. G. Teixeira 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by
Sequential Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase
Extraction. Journal of Brazzilian Chemical Society, 22, 552-557.
17. Nomngongo, P. N., Ngila, J. C., Kamau, J. N., Msagati, T. A. M. & Moodley, B. 2013.
Preconcentration of molybdenum, antimony and vanadium in gasolsine samples using
Dowex 1-x8 resin and their determination with inductively coupled plasma–optical
emission spectrometry. Talanta, 110, 153-159.
18. Wang, T., Jia, X. & Wu, J. 2003. Direct determination of metals in organics by
inductively coupled plasma atomic emission spectrometry in aqueous matrices. Journal
of Pharmaceutical and Biomedical Analysis, 33, 639-646.
19. Aguirre, M. A., Kovachev, N., Hidalgo, M. & Canals, A. 2012. Analysis of biodiesel
and oil samples by on-line calibration using a Flow Blurring[registered sign]
multinebulizer in ICP OES without oxygen addition. Journal of Analytical Atomic
Spectrometry, 27, 2102-2110.
20. Roldan, P. S., Alcântara, I. L., Padilha, C. C. F. & Padilha, P. M. 2005. Determination
of copper, iron, nickel and zinc in gasoline by FAAS after sorption and preconcentration
on silica modified with 2-aminotiazole groups. Fuel, 84, 305-309.
21. Escudero, L. A., Cerutti, S., Olsina, R. A., Salonia, J. A. & Gasquez, J. A. 2010.
Factorial design optimization of experimental variables in the on-line
separation/preconcentration of copper in water samples using solid phase extraction and
ICP OES determination. Journal of Hazardous Materials, 183, 218-223.
22. Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S. & Escaleira, L. A. 2008.
Response surface methodology (RSM) as a tool for optimization in analytical
chemistry. Talanta, 76, 965-977.
23. Tarley, C. R. T., Silveira, G., Dos Santos, W. N. L., Matos, G. D., Da Silva, E. G. P.,
Bezerra, M. A., Miró, M. & Ferreira, S. L. C. 2009. Chemometric tools in
electroanalytical chemistry: Methods for optimization based on factorial design and
response surface methodology. Microchemical Journal, 92, 58-67.
24. Kowalewska, Z., Ruszczyńska, A. & Bulska, E. 2005. Cu determination in crude oil
distillation products by atomic absorption and inductively coupled plasma mass
spectrometry after analyte transfer to aqueous solution. Spectrochimica Acta Part B:
Atomic Spectroscopy, 60, 351-359.
25. Ferreira, S. L. C., Queiroz, A. S., Fernandes, M. S. & Dos Santos, H. C. 2002.
Application of factorial designs and Doehlert matrix in optimization of experimental
variables associated with the preconcentration and determination of vanadium and
164
Chapter eight:
Dual-bed resin SPE for preconcentration of Ag, Al, As and Cr
copper in seawater by inductively coupled plasma optical emission spectrometry.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1939-1950.
26. Stewart, I. I. & W. Olesik, J. 1998. Steady state acid effects in ICP-MS. Journal of
Analytical Atomic Spectrometry, 13, 1313-1320.
165
CHAPTER NINE:
A SOLID PHASE EXTRACTION PROCEDURE BASED ON ELECTROSPUN
CELLULOSE-g-OXOLANE-2,5-DIONE NANOFIBERS FOR TRACE
DETERMINATION OF Cd, Cu, Fe, Pb AND ZN IN GASOLINE SAMPLES BY ICP
OES
ABSTRACT
Cellulose-g-oxolane-2,5-dione nanofibers were prepared as an adsorbent for
simultaneous separation and preconcentration of trace amounts of Cd, Cu, Fe, Pb and Zn
ions in gasoline samples prior to ICP OES detection. The nanofibers were chemically and
morphologically characterized by FTIR, solid-state 13C NMR, BET, SEM techniques. The
influences of experimental parameters such as sample pH, HNO3 concentration on metal
ion elution from the nanofibers, flow rate and sample volume were investigated. The metal
ions were retained on 0.5 g of the adsorbent at pH 6 and recovered with 5.0 mL of 2.0 mol
L−1 HNO3. The adsorption capacities for the adsorbent were 273.1, 183.6, 195.5, 236.2 and
182.4 mg g-1 for Cd, Cu, Fe, Pb and Zn, respectively. The relative standard deviation was
<3% (n =15), limits of detection and quantification were 0.13-0.68 µg L-1 and 0.42-2.2 µg
L-1, respectively, and the maximum preconcentration factor was 60. It was observed that
cellulose-g-oxolane-2,5-dione nanofibers can be used for more than 30 adsorption-elution
cycles without decreasing the extraction efficiency. The accuracy of the method was
confirmed by analyzing certified reference material and by performing spike recovery test.
The accuracy and recovery for different metal ions were in the range 97-102% and 9699%, respectively. The optimized method was applied for the separation and
preconcentration of metal ions in gasoline samples.
Keywords: Cellulose-g-oxolane-2,5-dione nanofibers, preconcentration, gasoline, trace
metal ions, ICP OES
9.1 INTRODUCTION
One of the problems that petroleum industries are faced is the presence of metal ions in
fuels. Metals occur naturally in fossil materials and, as a result, their presence can be
transferred to petroleum based products. Their presence can also be due to their
incorporation during the production process, by contact with refinement or distillation
equipment, storage and transport.1,2 Metal ions found in fuel have an ability to catalyze
166
Chapter nine:
Solid phase extraction using electrospun nanofibers
oxidative reactions in hydrocarbon mixtures and degrading thermal stability of fuel.3 For
instance, copper ions can catalyze the oxidation reactions of unsaturated hydrocarbons in
gasoline with oxygen leading to formation of gums, which then results in the
decomposition of fuel leading to poor engine performance.3-5 In addition, the presence of
Pb in fuel can reduce the efficiency of catalytic reactors used in vehicle exhaust systems,
thus increasing the emission of exhaust gases such as carbon monoxide and oxides of
sulfur and nitrogen into the atmosphere.3,6 Therefore, it is important to control and monitor
their concentrations in gasoline. Metal concentrations in fuel are generally in trace
levels,2,6,7 therefore sensitive and fast techniques with low detection limits are required.
Direct determination of fuel samples such as gasoline by most analytical techniques is
difficult. This is because of its volatility, low viscosity, corrosivity and immiscibility with
water.8 Inductively coupled plasma-optical emission spectrometry (ICP OES) is a sensitive
multi-element technique. However, direct introduction of organic matrices poses a
challenge with respect to the operating parameters of the instrument.7,8 This is because
direct loading of organic samples on to the ICP can destabilize or extinguish the plasma. 6,7
Therefore, a sample preparation step that will separate and pre-concentrate trace metals in
gasoline prior to ICP OES detection is required. Techniques involving separation and preconcentration procedures for the determination of trace elements in organic matrices, such
as ethanol fuels and gasoline, are reported in literature.4,5,8-12 However, the development of
methodologies to separate and preconcentrate metal ions in different fuel matrices prior to
their spectrometric determination is still a gap in the scientific literature.13
Recently, the use of nanometer-sized materials in solid phase extraction (SPE) as
metals ion adsorbents has become an active area of research in the field of separation
science because of their special properties.14 The latter include small diameter, large
specific surface area, high degree of macromolecular orientation and the resultant superior
mechanical
properties.
These
nanometer-sized
materials
can
be
obtained
by
electrospinning process which is a technique that uses electric force to make the spinning
process producing polymer fibers with diameters in the nanometer range (10–1000 nm).15
Cellulose is one the most abundant biopolymers that has been intensely studied for
possible application in removal of toxic metal ions in the environmental samples.16,17
However, cellulose is relatively inert because the hydroxyl groups, which act as active
groups, are involved in inter- and intramolecular hydrogen bonding.18 Therefore,
functionalization approaches have been employed to improve the surface, reactivity and
167
Chapter nine:
Solid phase extraction using electrospun nanofibers
stability of this biopolymer.18,19 The functionalization strategies have been attained through
two main routes. These include (i) introduction of functional groups into the raw cellulose
backbone; and (ii) electrospinning of celluloses to form nanofibers (this is done to increase
its surface area and pore volume) followed by change of the functional groups on the
cellulose surface.19,20 The second route, which involves electrospinning, has recently
received much attention. However, difficulties to find suitable solvents to dissolve
cellulose make it difficult to directly prepare cellulose nanofiber by electrospinning.21 To
overcome these challenges, cellulose derivatives such as cellulose acetate has been used to
prepare nanofibers. Cellulose acetate nanofiber is then treated with alkaline solution to
remove the acetyl groups completely to obtain regenerated cellulose nanofiber.19,21-23
Functionalized cellulose and cellulose nanofibers have been applied for the adsorption of
metal ions in the environmental samples including wastewater.19
The objective of this work was to develop a SPE procedure for simultaneous separation
and preconcentration of trace amounts of Cd, Cu, Fe, Pb, and Zn in gasoline using
cellulose-g-oxolane-2,5-dione electrospun nanofibers prior to their ICP OES detection. It
should be noted that manganese, which used as organic manganese fuel additive, was
deliberately omitted in this study because it is already present in large quantities in some
gasoline samples and therefore it does not qualified to be trace. It is important to highlight
that there are no reported data about the application of cellulose-g-oxolane-2,5-dione
nanofibers for separation and preconcentration of metal ions in gasoline. The effects of
pH, flow rates, eluent concentration and sample volume were studied and optimized. The
validity of the SPE procedure was studied and was verified by analyzing a certified
reference material (CRM). In addition, spike-recovery experiments were carried out and
corresponding recoveries were evaluated.
9.2 EXPERIMENTAL
9.2.1 Material and methods
All reagents were of analytical grade unless otherwise stated and double distilled
deionised water (Millipore, Bedford, MA, USA) was used throughout the experiments.
Cellulose acetate (Mw = 30,000, % acetyl content = 39.8), N,N-dimethylacetamide
(DMAc), acetone, toluene and oxolane-2,5-dione obtained from Sigma–Aldrich (South
Africa) were used in the preparation of the solid phase material. Absolute ethanol (Merck,
168
Chapter nine:
Solid phase extraction using electrospun nanofibers
Darmstadt, Germany) and was used to prepare model solutions. Spectrascan stock
solutions (1000 mg L-1) of Cd, Cu, Fe, Pb, and Zn (Industrial Analytical Pty Ltd,
Johannesburg, South Africa) were used to prepare the working solutions for solid phase
extraction (SPE) at concentrations of 20 µg L-1 for all metal ions. Working solutions
(prepared in organic phase), as per the experimental requirements, were freshly prepared
from the stock solution for each experimental run. A Spectrascan multi-element standard
solution at concentration of 100 mg L-1 (Industrial Analytical Pty Ltd, Johannesburg, South
Africa) was used to prepare working standard solutions at concentrations of 10-120 µg L-1
(Cd, Cu, Fe, Pb and Zn) for measurements of concentrations of analytes in all model and
sample solutions. Solutions of nitric acid at concentrations of 0.5, 1.0, 2.0, and 3.0 mol L-1
were prepared from ultrapure concentrated acid (65%, Sigma-Aldrich, St. Loius, MO,
USA). These solutions were used for the elution of the analytes from the column. The pH
adjustments were performed with glacial acetic acid (Merck, Darmstadt, Germany) and
ammonia (Sigma-Aldrich, St. Loius, MO, USA) solutions. Ten gasoline samples from
different local filling stations were used for method development and validation. Gasoline
samples with additives and without additives are described as metal-containing unleaded
gasoline (MCUG) and metal-free unleaded gasoline (MFUG), respectively.
Metal ions (Cd, Cu, Fe, Pb and Zn) were determined using a Spectro Arcos 165 ICP
OES (SPECTRO Analytical Instruments, GmbH, Germany) equipped with Cetac ASX-520
autosampler. The ICP OES operating conditions listed in Table 9.1. Sample introduction
was achieved using a pneumatic cross-flow nebulizer mounted onto a Scott double-pass
spray chamber. Sample solutions were pumped to the nebulizer using a built in four
channel peristaltic pump. The most prominent atomic and ionic analytical spectral lines of
the metal ions were selected for investigation, that is, Cd 228.802 nm, Cu 324.754 nm, Fe
259.941 nm, Pb 220.353 nm and Zn 213.856 nm. Solid phase extraction was carried out in
a VacMaster-24 sample SPE station (Supelco, PA, USA). The latter was used to control
the sample loading and elution flow rates.
Table 9.1. The operating parameters of determination of elements by ICP OES
RF power
1400 W
Plasma argon flow rate
Auxiliary argon flow rate
Nebulizer argon flow rare
Sample aspiration rate
Replicate measurements
13 L min-1
2.00 L min-1
0.95 L min-1.
2.0 mL min-1
3
169
Chapter nine:
Solid phase extraction using electrospun nanofibers
Infrared spectra were recorded using Spectrum 100 FT-IR (PerkinElmer, USA)
spectrometer equipped with Universal Attenuated Total Reflectance (ATR) attachment
with a diamond crystal in the range 4000-500 cm-1. Solid-state cross polarization magic
angle spin Carbon-13 nuclear magnetic resonance (CP-MAS
13
C NMR) spectra of the
nanofibers were obtained on a Bruker Avance III 600 MHz (GmbH, Germany)
spectrometer installed with the software Bruker Topspin 2.1 software. Spectra were
acquired at frequencies of 75.47 MHz with a magic angle spinning of 10 kHz. The
morphological characteristics of nanofibers were measured by scanning electron
microscopy (Leo 1450 using SmartSEM software version 5.03.06), operating at 10 kV.
9.2.2 Electrospinning and functionalization of cellulose nanofibers with oxolane-2,5dione
Electrospinning of cellulose acetate was performed according to the procedure
described by Ma et al.21 and Musyoka et al.19 Deacetylation of cellulose acetate nanofibers
and functionalization of cellulose nanofibers with oxolane-2,5-dione was carried out as per
Musyoka et al.19
9.2.3 Column preparation
A mass (0.5 g) of cellulose-g-oxolane-2,5-dione nanofibers was introduced into the
polyethylene columns (diameter 1.35 cm and 6.5 cm in height) plugged with a small
portion of porous frits at both ends. Porous frits were used to hold and confine the
adsorbent within the designated capacity/volume. Before use, 1.0 mol L-1 HNO3 solution
and doubly distilled water were passed through the column in order to clean it. The
columns were conditioned to the desired pH with ammonium buffer solution.
9.2.4 Preconcentration procedure
Model solutions were prepared as follows: 2.0 mL of 1.0 mg L-1 Cd, Cu, Fe, Pb and Zn
solutions were separately transferred into 100 mL volumetric flasks and made up to the
mark with ethanol to obtain 20 µg L-1 concentration of each metal ion. Ethanol model
solutions of metal ions (20 mL) were percolated through the cellulose-g-oxolane-2,5-dione
columns at a flow rate of 3.0 mL min-1. The column was washed with 10 mL of double
distilled deionised water to remove excess alcohol solution, followed by 5.0 mL of
170
Chapter nine:
Solid phase extraction using electrospun nanofibers
ammonium acetate buffer solution to remove major cations (Na+, Ca2+, K+, etc). Lastly, the
metal ions were eluted with 5 mL of 2.0 mol L-1 HNO3 solution. All fractions obtained
during the elution stage were collected separately and analyzed by ICP OES. It should be
noted that the washings with double distilled deionised water and ammonium acetate
buffer solution were discarded. The same procedure was applied to the blank solutions.
After each use, the sorbent in the column was washed with 20 mL of double distilled
deionised water and stored for the next experiment. The effect of sample pH, sample
volume, eluent concentration and sample and eluent flow rates were investigated. All
analyses were performed in triplicate.
9.2.5 Procedure for the dilution of Certified Reference Material
To validate the preconcentration method described in this study, a Conostan custom
made oil based certified reference material (CRM) obtained from SCP Science (Quebec,
Canada) containing 1.0 mg L-1 of each metal ion was used. The dilution of the CRM was
performed as follows: a 2.0 mL aliquot of 1.0 mg L-1 CRM dissolved in 10 ml of hexane.
The solution was quantitatively transferred in to 100 mL volumetric flask and made to the
mark with acetone to obtain 20 µg L-1 of each metal ion. Suitable aliquots (20 ml) of the
solution were taken and pre-concentrated by the cellulose-g-oxolane-2,5-dione SPE
procedure and analyzed with ICP OES. The same procedure was applied to the preparation
of the blank solution.
9.2.6 Procedure for acid digestion of gasoline samples
The acid digestion procedure was carried out according to Amorim et al.24. Briefly
description of the procedure involved is as follows: 5.0 ml of gasoline sample was
transferred into 100 mL Teflon beaker followed by 2.0 mL H2O2 (30%) and 6 mL HNO3
(65%) and heated at 170 ± 10 °C in hot plate for 10 min. The digested content was left to
cool down to room temperature and quantitatively transferred to a volumetric flask and
diluted with double distilled deionised water to a final volume of 50 mL. double distilled
deionised water applied to the same procedure was used as the blank. The samples were
then analyzed with ICP OES.
171
Chapter nine:
Solid phase extraction using electrospun nanofibers
9.3 RESULTS AND DISCUSSION
9.3.1 Characterization of the adsorbent
Chemical changes (deacetylation and functionalization) of the nanofibers materials
were investigated by ATR-FTIR spectroscopy and the results are presented in Fig. 1.
Cellulose acetate nanofibers showed a characteristic strong carbonyl absorption band at
1750 cm-1 which is due to the presence of acetyl groups.19,21 After the nanofibers were
treated with alkaline solution, the disappearance of the strong carbonyl absorption was
noticed. This showed that the deacetylation process was achieved. In addition, the
hydroxyl absorption band at 3321 cm-1 was stronger for deacetylated cellulose nanofibers
compared to cellulose acetate nanofibers. This might be due to the increase of the hydroxyl
groups in the deacetylated cellulose.15 After surface functionalization of cellulose
nanofibers, the FTIR spectrum in Fig. 9.1 shows absorption band at 1730 cm-1. The latter
suggested that the deacetylated cellulose nanofibers was successfully functionalized with
oxolane-2,5-dione.19
Fig. 9.1. Characteristic absorption peaks of (A) cellulose acetate, (B) deacetylated cellulose
and (C) cellulose-g-oxolane-2,5-dione
172
Chapter nine:
Solid phase extraction using electrospun nanofibers
The findings obtained by FTIR were confirmed by CP-MAS 13C-NMR and the results
are summarized in Table 9.2. The assignment of the chemical shifts of cellulose, cellulose
acetate and cellulose-g-oxolane-2,5-dione was done according to Splendore et al.18
Musyoka et al.19 Biermann et al.25 Kono et al.26 and Kono et al.27. The spectrum of
cellulose acetate nanofibers showed a downfield signal at δ 172 ppm which was attributed
to the CO of the acetyl group. The up field signal at δ 22 ppm was assigned to the methyl
carbons present in the acetate chain.19 The NMR spectrum of deacetylated cellulose
nanofibers showed the characteristic signal pattern of cellulose. However, the
functionalized cellulose nanofibers showed new signals at 174 and 30 ppm, which were
attributed to the carbonyl and methylene groups, respectively.19 The deacetylation of
cellulose acetate to cellulose and the functionalization of the latter were successfully
confirmed by FTIR and CP-MAS 13C-NMR analyses.
Table 9.2.
13
C NMR chemical shifts for cellulose acetate, deacetylated cellulose and
cellulose-g-oxolane-2,5-dione and the corresponding assignments
Nanofibers
Cellulose acetate
Deacetylated cellulose
Cellulose-g-oxolane-2,5dione
δ (ppm)
172
105,67
75
22
105,67
91
71-82
174
106,66
90
70-81
30
Assignments
CO
C1,C6
C2-C5
CH3
C1, C6
C4
C2,C3 and C5
CO
C1, C6
C4
C2,C3 and C5
CH2
Morphological structures of cellulose and cellulose-g-oxolane-2,5-dione nanofibers by
SEM image are shown in Fig. 9.2. The unmodified cellulose nanofiber had a mean
diameter of 334 nm. After functionalization, the mean diameter of cellulose-g-oxolane-2,5dione nanofibers did not change. The specific surface area and the pore size of cellulose-goxolane-2,5-dione nanofibers were determined using the BET equation applied to the
adsorption data on nitrogen adsorption/desorption experiments. The results showed that the
average specific surface area and the pore size were 14.31 m2 g-1 and 113.56 Å,
173
Chapter nine:
Solid phase extraction using electrospun nanofibers
respectively. The relatively large specific surface area and small pore structures confirmed
that cellulose-g-oxolane-2,5-dione nanofibers can serve as an adsorbent with large
adsorption capacity.
Fig. 9.2. SEM micrographs: (I) cellulose nanofibers and (II) cellulose-g-oxolane-2,5-dione
nanofibers
9.3.2 Effect of sample pH
Sample pH is a very important factor for efficient adsorption of analyte ions on the
solid phase material.28 This is because, sample pH affects adsorbent surface charge, degree
of ionization and speciation of the analyte. Consequently, the effects of pH of ethanol
solution on the retentions of Cd, Cu, Fe, Pb and Zn ions onto cellulose-g-oxolane-2,5dione nanofibers were examined at the pH range of 4.0-10.0. The pH was adjusted using
glacial acetic acid and/or 1.0 mol L-1 NH3. It can be seen from Fig. 9.3 that the percentage
recovery increased with increased pH values and all the analytes were quantitatively
recovered in the pH range 6.0–7.0. The increase in metal ion removal as pH increased was
due to the decrease in competition between protons (H+) and positively charged metal ions
at the surface sites. Therefore, a decrease in positive charge results in a lower repulsion of
the adsorbing analyte. However, a decrease was observed at ph values greater than 7, this
decrease might be due to the formation of insoluble metal ion hydroxides and this
observation is in close agreement with Musyoka et al.19 Therefore, for all subsequent
174
Chapter nine:
Solid phase extraction using electrospun nanofibers
experiments pH 6.0 was adopted as optimum for the quantitative separation and
preconcentration of metal ions.
Fig. 9.3. Effect of sample pH on retention of the analytes in ethanol onto cellulose-goxolane-2,5-dione column: Experimental conditions: analyte concentration 20 µg L-1;
amount of adsorbent 0.5 g; flow rates of sample and eluent 2.0 mL min−1; eluent
concentration 2.0 mol L-1, eluent volume 5 mL; replicates n = 3
9.3.3 Effect of eluent concentration
Effect of eluent concentration on metal ion elution from the adsorbent was
investigated. Various concentrations of HNO3 were used for desorption of adsorbed
analyte ions from the cellulose-g-oxolane-2,5-dione column. It can be seen from Table 9.3
that lower recoveries were obtained with 0.5-1.0 mol L-1 HNO3. The lower recoveries of
metal ions might be due to the strong interaction between metal and adsorbent.28
Therefore, it may be concluded that 0.5-1.0 mol L-1 HNO3 was not suitable for quantitative
elution of retained metal ions. Quantitative recoveries were obtained with 2.0-3.0 mol L-1
HNO3. Under optimized conditions, 2.0 mol L-1 was used for further investigations.
175
Chapter nine:
Solid phase extraction using electrospun nanofibers
Table 9.3. Influences of the eluent concentration on the recoveries of the analytes on
cellulose-g-oxolane-2,5-dione column.
Elements
Nitric acid concentration (mol L-1)
0.5
1.0
2.0
3.0
Cd
65.3±2.5
94.3±1.9
97.2±1.0
97.5±1.6
Cu
66.2±2.2
90.1±1.5
98.0±1.1
97.6±1.1
Fe
67.6±1.5
88.7±1.5
98.9±0.9
98.1±1.3
Pb
70.1±2.1
93.7±1.7
97.4±1.4
97.8±1.4
Zn
61.3±1.8
89.6±1.4
97.2±1.5
96.8±1.2
Experiment conditions: sample volume 20 mL; amount of adsorbent 0.5 g, flow rates of sample and eluent
2.0 mL min−1, replicates n = 3)
9.3.4 Effect of flow rate of sample solutions
The optimization of the sample and eluent flow rates was carried out to ensure the
quantitative retention and desorption of the analytes. The effect of sample and eluent flow
rates of the sample solution (20 mL) on the retention of the metal ions onto cellulose-goxolane-2,5-dione nanofibers was done on a column packed with 0.5 g of the adsorbent.
The influence of flow rates was investigated in the range of 1–5 mL min-1. It can be seen
from Table 9.4 that the quantitative recovery values were obtained in the flow rate range of
1–2 mL min-1 for sample and eluent solution. Flow rates greater than 2.0 mL min-1 caused
a gradual decrease in sorption due to insufficient contact time between the cellulose-goxolane-2,5-dione nanofibers and the metal ions. Therefore, 2.0 mL min-1 flow rate was
chosen as the optimum flow rate for sample loading and analyte elution.
Table 9.4. Effect of flow rate of sample solutions: analytical results in terms of recovery
Follow rates
(min L-1)
1.0
2.0
3.0
4.0
5.0
Recovery (%)
Cd
98.3±1.1
97.8±2.1
94.5±0.5
88.7±1.4
77.4±3.1
Cu
99.8±0.9
99.4±1.1
95.4±1.8
90.1±1.9
76.5±2.5
Fe
98.3±1.5
98.5±0.9
95.1±0.4
91.2±0.8
79.4±2.4
176
Pb
97.8±1.0
98.4±0.9
94.8±1.2
87.8±1.8
80.1±2.0
Zn
99.6±0.1
99.4±0.7
96.5±0.5
93.1±0.8
86.7±1.8
Chapter nine:
Solid phase extraction using electrospun nanofibers
9.3.5 Effect of the sample volume
The effect of the sample volume on the recoveries of metal ions on cellulose-goxolane-2,5-dione column was investigated by varying the sample volume from 20 to 500
mL. The results are shown in Fig. 9.4. The sample volume did not affect quantitative
recoveries in the range of 20–300 mL of the sample volume for the investigated metal ions.
Above 300 mL, the recoveries decreased for all the metal ions. The decrease in the
recoveries of the analytes is probably due to the excess analytes loaded over the column
capacity with increasing sample volume. The preconcentration factor defined as the ratio
of the analyte concentrations before and after preconcentration, was calculated as 60 when
eluent volume is 5.0 mL.
Fig. 9.4. Effect of sample volume on the recoveries of metal ions. Experimental conditions:
pH 6; analyte concentration 20 µg L-1; amount of cellulose-g-oxolane-2,5-dione nanofibers
0.5 g; flow rates of sample and eluent 2.0 mL min−1; eluent concentration 2.0 mol L-1;
eluent volume 5 mL; replicates n = 3
9.3.6 Adsorption capacities
The adsorption capacity is an important factor, because it determines how much
sorbent is required to quantitatively concentrate the analytes from a given solution.14
177
Chapter nine:
Solid phase extraction using electrospun nanofibers
Preliminary adsorption studies with cellulose-g-oxolane-2,5-dione nanofibers revealed that
60 minutes is sufficient time for the system to reach equilibrium. Therefore, 0.05 g
cellulose-g-oxolane-2,5-dione nanofibers was equilibrated in 20 mL of Cd, Cu, Fe, Pb and
Zn ethanol solutions at concentrations 30 to 250 mg L-1 by shaking for 60 minutes at pH 6.
The experimental data were fitted into the general equation of the modified Langmuir
model presented in Eq. 1.19 The later was used to calculate the maximum adsorption
capacity.
Ce
1
1

Ce 
qe qmax
K L qmax
(1)
where qe is the equilibrium adsorption capacity of ions on the adsorbent (mg g−1); Ce, the
equilibrium ions concentration in solution (mg L−1); qmax, the maximum capacity of the
adsorbent (mg g−1); and KL, the Langmuir adsorption constant (L mg−1). The results
showed that adsorption capacity of the analytes probably differ due to their size, degree of
hydration and the value of their binding constant with cellulose-g-oxolane-2,5-dione.14 The
maximum adsorption capacities were found to be 273.1, 183.6, 195.5, 236.2 and 182.4 mg
g-1 for Cd, Cu, Fe, Pb and Zn, respectively. The sorption capacities were higher than those
reported by Qu et al.,3 and Prado et al.13. Due to the different behaviour of metals in
various matrices, the results were not compared to those studies carried out in aqueous
phase.
9.3.7 Column regeneration
In order to investigate the possibility of reusing the cellulose-g-oxolane-2,5-dione
column, adsorption/desorption experiments were conducted. The results in Table 9.5
showed that the column can be reused after elution of metal ions with 5.0 mL 2.0 mol L-1
HNO3 and rinsed with 20 mL distilled water, respectively. The column was stable over 30
adsorption/elution cycles without obvious decrease in the recoveries for the metal ions.
The recoveries of the metal ions were remained greater than 95%. It should be noted that 5
cycles were done per day over 6 days to give a total of 30 cycles.
178
Chapter nine:
Solid phase extraction using electrospun nanofibers
Table 9.5. Column regeneration.
Analytes
No. of cycles
Cd
Cu
Fe
Pb
Zn
1
97.2±2.2
98.0±1.2
98.9±1.4
97.4±2.4
97.2±2.1
10
96.8±2.4
98.3±1.1
99.1±0.87
97.9±1.8
97.1±2.4
30
97.2±2.4
98.1±2.1
98.3±1.2
97.1±1.8
96.9±2.4
Experimental conditions: pH 6; analyte concentration 20 µg L-1; amount of cellulose-g-oxolane-2,5-dione
nanofibers 0.5 g; flow rates of sample and eluent 2.0 mL min−1; eluent concentration 2.0 mol L-1 ; volume 5
mL; replicates n = 3
9.3.8 Analytical Parameters
The analytical performance of the cellulose-g-oxolane-2,5-dione SPE method under
optimum conditions for pre-concentration of Cd, Cu, Fe, Mn, Pb and Zn was evaluated and
the results are presented in Table 9.6. The linearity of the methods was investigated by
preconcentrating seven portions of ethanol solution spiked with multi-element standard at
concentration ranging from 0.5 to 150 µg L-1 and analyzed by ICP OES. It was observed
that a linear range of 15 to130 µg L-1 after pre-concentration was achieved for all the
investigated analytes. Correlation coefficients are presented in Table 9.6. The sensitivity of
the pre-concentration system was defined as the gradient (slope) of the calibration graph.
The results in Table 9.6 indicated that the SPE method was more sensitive to Cu and Fe
compared to the rest of metals ions. Thus the highest slope obtained was 84.4 L µg-1 for Cu
while the lowest was 14.9 L µg-1 for Pb.
The limits of detection (LOD) and limits of quantification (LOQ) for Cd, Cu, Fe Pb
and Zn ions were obtained from the signals of fifteen blank samples (n = 15) and the slope
of the calibration curve. The LODs were defined as the lowest concentration of the analyte
giving signals equal to three times the standard deviation of blank signal (Eq. 2). While,
LOQs were defined as the lowest concentration of analyte equal to ten times the standard
deviation of blank signal (Eq.3) that can be accurately and precisely analyzed.29 The LOD
and LOQ were calculated according to Eq. 2 and 3.30
LOD 
3  SD
m
(2)
LOQ 
10  SD
m
(3)
179
Chapter nine:
Solid phase extraction using electrospun nanofibers
where SD is standard deviation of the blank signal and m is the gradient of the calibration
curve. For 50.0 mL of sample solution used in the preconcentration step, LOD and LOQ
obtained for Cd, Cu, Fe Pb, and Zn are presented in Table 9.6. The instrument detection
limits (IDL) for Cd, Cu, Fe Pb, and Zn can seen in Chapter 6 (Table 6.4)
The precision (reproducibility) of the SPE method was studied by performing fifteen
successive measurements at a concentration of 20 µg L-1 of multi-element organic solution
(containing Cd, Cu, Fe, Pb and Zn). The overall reproducibility (precision) of preconcentration procedure expressed in terms of relative standard deviation (%RSD) was
reasonably good (<3%) as shown in Table 9.6. The overall time required for
preconcentration of 20 mL of sample (percolation was 180 s at flow rate of 3 mL min−1;
elution was 30 s at flow rate of 3.0 mL min−1 and washing and conditioning was 30 s) was
about 4 min. it should be noted that the sample preconcentration was performed in
triplicates and they were all carried out at the same time. Therefore, the overall time for
preconcentration of triplicates was approximately 7 min. Hence, the throughput sample
was approximately 26 samples h-1.
Table 9.6. Analytical performances for cellulose-g-oxolane-2,5-dione SPE method.
Analyte
Sensitivity
(cps L µg-1)
Correlation
efficient (R2)
LOD (µg L-1)
LOQ (µg L-1)
Precision
(%)
Cd
Cu
Fe
Pb
Zn
20.4± 1.3
84.4±2.1
36.6±1.7
14.9±0.9
24.8±1.6
0.9995
0.9989
0.9974
0.9913
0.9986
0.48
0.13
0.31
0.68
0.32
1.61
0.42
1.01
2.18
1.06
1.5
1.1
1.8
2.1
1.3
Experimental conditions: pH 6; analyte concentration 20 µg L-1; amount of cellulose-g-oxolane-2,5-dione
nanofibers 0.5 g; flow rates of sample and eluent 2.0 mL min −1; eluent concentration 2.0 mol L-1 ;eluent
volume 5 mL; replicates n = 3
The analytical performances of the proposed method were compared with the ones
obtained using a comparative method (acid digestion method). Table 9.7 shows the
analytical characteristics for acid digestion procedure. It should be noted that acid
digestion method does not employ the SPE preconcentration step. The SPE method based
on the electrospun cellulose-g-oxolane-2,5-dione nanofibers was more sensitive compared
to acid digestion method. In terms of correlation efficient, the acid digestion had better R2
compared to the preconcentration method. The LOD and LOQ of the proposed
180
Chapter nine:
Solid phase extraction using electrospun nanofibers
preconcentration method were better than those of the acid digestion method. TheThe
precision of the two sample pretreatment methods were not significantly different at 95%
confidence level using statistical F-test. The samples (triplicates) were digested for ten min
and cooled for about 20 min to room temperature. This means that the overall time for acid
digestion of triplicates was 30 min. Therefore, the throughput sample was about 6 sample
h-1. When comparing the sample throughput using acid digestion with that obtained by the
proposed pre-concentration method, the latter had relatively higher sample throughput.
Table 9.7. Analytical performances for acid digestion method
Analyte
Sensitivity
(cps L µg-1)
Correlation
efficient (R2)
LOD (µg L-1)
LOQ (µg L- Precision
1
)
(%)
Cd
Cu
Fe
Pb
Zn
16.6±1.1
64.9±3.1
29.7±2.1
9.5±0.6
18.3±1.1
0.9996
0.9999
0.9994
0.9991
0.9993
0.75
0.41
0.56
0.96
0.48
2.50
1.37
1.87
3.21
1.60
1.4
1.2
1.6
2.1
1.4
A comparison of the proposed method with other separation and preconcentration
methods in terms of selected analytical parameters such as LOD and %RSD was also
carried out. Roldan et al.9 developed a method for Cu, Fe, Ni and Zn determination in
gasoline after preconcentration in silica modified with 2-aminotizole group. The LODs
obtained were 0.8, 3, 2 and 0.1 µg L-1 for Cu, Fe, Ni and Zn, respectively. Santos et al.5
developed a method for preconcentration of metal ions in gasoline samples using XAD
modified with 3,4 dihydroxybenzoic acid and the LODs for Fe, Cu, Pb and Zn ranged from
2.2-3.1 µg L-1 and the %RSD ranged from 5.8 to % 9.7. Teixeira and coworkers8 used
cellulose paper for the separation and preconcentration of Cu and Fe in automotive
gasoline. The LOD and %RSD ranged from 10 to 15 µg L-1 and 7.8 to 8.1, respectively. In
most cases, the proposed SPE method has low RSD and relatively low LOD values when
compared the other methods reported in literature.
9.3.9 Accuracy and validation of the developed method
The accuracy of the present method was tested by performing the spike recovery tests in
gasoline samples (1-MFUG and 2-MCUG). Known amounts of each metal ion (20 and 40
µg L-1) were added into the gasoline samples. The results obtained are shown in Table 9.8.
181
Chapter nine:
Solid phase extraction using electrospun nanofibers
These results show a good agreement between the added and recovered analyte
concentration. The percentage recoveries (%R) of analytes ranged from 96 to 99 % and the
results showed that the different gasoline sample matrices did not affect the recovery of
trace metals. In addition, the sorbent was found to retain metal ions at relatively low
concentrations (Such as Cd). Therefore, it can be concluded that cellulose-g-oxolane-2,5dione SPE is a suitable method for separation and pre-concentration of trace metal ions in
gasoline samples.
Table 9.8. Accuracy test results for spike recovery test.
1-MFUGa
Element
Cd
Cu
Fe
Pb
Zn
Added
(µg L-1)
0
20
40
0
20
40
0
20
40
0
20
40
0
20
40
Found
(µg L-1)
c
ND
20.0±1.4
39.3±1.2
50.3±1.6
69.8±0.6
89.8±1.1
350±1
370±1
390±1
11.9±0.6
31.3±0.7
51.4±0.7
267±1
287 ±0.1
307±1
2-MCUGb
R (%)
99.8±0.6
98.1±2.1
97.4±1.0
98.8±1.6
97.8±2.4
99.2±1.9
97.2±0.9
98.9±1.1
98.3±1.3
99.5±1.0
Found
(µg L-1
1.08±0.11
20.7±1.0
40.3±0.6
5.33±1.56
25.19±1.14
44.6±0.2
187±1
207±2
226±1
8.92±0.90
28.3±0.7
47.9±1.1
102±0.1
122±0.2
142±0.1
R (%)
98.2±1.4
98.1±1.2
99.3±1.1
98.2±0.7
96.9±1.0
98.3±2.1
97.1±1.2
97.4±2.4
99.4±0.1
99.4±0.4
a
MFUG = metal-free unleaded gasoline; bMCUG = metal-containing unleaded gasoline; cND = not
detectable; 1&2 are the numbers allocated to the six gasoline filling stations. Experimental conditions: pH 6;
analyte concentration 20 µg L-1; amount of cellulose-g-oxolane-2,5-dione nanofibers 0.5 g; flow rates of
sample and eluent 2.0 mL min−1; eluent concentration 2.0 mol L-1; eluent volume 5 mL; replicates n = 3
The validity of the method was checked by analyzing Conostan custom made oil based
CRM for Cd, Cu, Fe, Pb and Zn. The certified and determined concentration values of the
investigated metal ions are given in Table 9.9. The determined values agree well with the
certified values reported for the CRM. A paired t-test was applied to evaluate these results,
and at 95% confidence level, there was no significant difference between the means for
both certified and determined results. In addition, satisfactory recoveries in the range of
182
Chapter nine:
Solid phase extraction using electrospun nanofibers
97.5% to 102% were obtained. The precision of the measurements (n = 6) expressed as %
RSD ranged between 0.54% and 1.4 %.
Table 9.9. Results for the oil based certified reference material.
Elements
Concentration (µg L-1)
Cd
Cu
Fe
Pb
Zn
Certified
1000
1000
1000
1000
1000
RSD (%)
1.0
1.0
1.0
1.0
1.0
Found (n=3)
973.5
1016
998.1
963.4
1011
RSD (%)
1.2
0.5
1.0
1.4
0.6
R (%)
97.5
102
99.2
96.4
101
Experimental conditions: pH 6; analyte concentration 20 µg L-1; amount of cellulose-g-oxolane-2,5-dione
nanofibers 0.5 g; flow rates of sample and eluent 2.0 mL min−1; eluent concentration 2.0 mol L-1, eluent
volume 5 mL; replicates n = 3
9.3.10 Application of the cellulose-g-oxolane-2,5-dione SPE method
The cellulose-g-oxolane-2,5-dione SPE procedure was applied to 10 gasoline samples
obtained from 6 different gasoline filling stations in Johannesburg (South Africa). The
results are summarized in Table 9.10. Generally, the total metal ion content in gasoline
samples followed this trend; 3-MCUG > 6-MCUG > 1-MFUG > 2-MFUG > 5-MCUG >
3-MFUG > 5-MFUG > 2-MCUG > 4-MFUG > 6-MFUG. It can be seen that the
concentration of Fe in gasoline samples was higher (above 100 µg-1) than other metal ions
regardless of the gasoline type. The highest Fe content was 649.50 µg L-1 for 3-MCUG
while the lowest was 114.34 µg L-1 for 6-MFUG. The concentration of Pb in gasoline was
≤ 24 µg L-1 except for the 3-MCUG and 6-MFUG samples. Cadmium content was the
lowest in almost all the samples except for 6-MCUG sample. Samples 1-MFUG, 4-MFUG
and 6-MFUG had the highest concentrations of Cu compared to other samples. In addition,
1-MFUG and 6-MCUG has the highest Zn concentration and 6-MFUG has the lowest Zn
content.
183
Chapter nine:
Solid phase extraction using electrospun nanofibers
Table 9.10. Concentrations (in µg L-1) of metal ions in commercial gasoline samples after
pre-concentration by the cellulose-g-oxolane-2,5-dione SPE method.
Samples
Cd
Cu
Fe
Pb
Zn
1-MFUGa
2-MCUGb
2-MFUG
3-MCUG
3-MFUG
4-MFUG
5-MCUG
5-MFUG
6-MCUG
6-MFUG
NDc
1.08±0.11
1.04±0.21
2.03±0.32
1.97±1.11
11.87±1.05
5.02±016
3.39±0.17
318.89±1.61
2.72±0.09
50.25±1.56
5.33±0.93
16.08±1.01
8.08±0.41
5.92±0.62
42.94±1.74
12.12±0.38
3.90±0.16
15.19±1.66
38.79±0.31
350.08±1.09
187.08±1.14
475.67±1.41
649.50±1.69
309.88±0.48
154.30±1.21
308.85±1.88
225.97±0.91
149.75±0.45
114.34±1.17
11.83±0.61
8.92±0.09
23.96±1.02
105.42±0.89
3.08±0.65
5.38±1.12
20.73±1.01
ND
15.06±0.06
61.98±1.00
267.33±1.10
102.33±1.09
94.78±1.21
105.75±1.19
62.42±0.77
69.84±0.81
85.81±0.35
100.87±0.15
280.98±0.73
16.42±0.85
a
MFUG = metal-free unleaded gasoline; bMCUG = metal-containing unleaded gasoline; cND = Not
detectable; 1-6 are the numbers allocated to the six gasoline filling stations. Experimental conditions: pH 6;
amount of cellulose-g-oxolane-2,5-dione nanofibers 0.5 g; flow rates of sample and eluent 2.0 mL min −1;
eluent concentration 2.0 mol L-1; eluent volume 5 mL; replicates n = 3
To verify the results of the cellulose-g-oxolane-2,5-dione SPE procedure, gasoline was
digested in a hot plate with a mixture of H2O2 and HNO3. The concentrations of analytes
were determined by ICP OES. The results shown in Table 9.11 suggested that there is no
significant difference between the results of the two methods at 95% confidence level. The
determination of Cd, Cu, Fe, Pb and Zn by ICP OES after acid digestion was used as an
additional procedure for quality check of the cellulose-g-oxolane-2,5-dione separation and
preconcentration method. It should be noted that acid digestion followed by ICP OES
determination was adopted as the standard method (cross-check method) in this study. The
main advantage of the cellulose-g-oxolane-2,5-dione SPE procedure described in this study
is that it does not require rigorous acid digestion unlike the acid digestion method. In
addition, the SPE method had relatively low LOD and LOQ, high sample throughput and
better sensitivity compared to the standard method. The column method is also
advantageous because it minimizes the risks of cross-contamination during acid digestion.
184
Chapter nine:
Solid phase extraction using electrospun nanofibers
Table 9.11. Concentrations (in µg L-1) of metal ions in gasoline samples determined by
ICP OES in sample solutions resulting from acid digestion procedure
Sample
Cd
NDc
1.10±0.57
0.97±0.18
1.87±0.23
2.05±1.13
11.91±1.13
4.47±0.11
3.29±0.23
320.53±1.78
2.70±0.14
1-MFUGa
2-MCUGb
2-MFUG
3-MCUG
3-MFUG
4-MFUG
5-MCUG
5-MFUG
6-MCUG
6-MFUG
Cu
49.87±1.78
5.29±1.38
16.27±0.89
7.82±0.29
5.50±0.44
42.76±1.87
11.59±0.45
4.16±0.18
14.56±1.83
39.08±0.23
Fe
354.67±0.95
186.55±1.18
470.23±1.30
652.99±1.77
308.12±0.55
152.28±1.30
310.86±2.00
224.73±0.89
149.04±0.92
114.85±1.22
Pb
12.09±0.58
9.15±0.17
24.27±1.68
106.74±1.01
2.94±0.56
5.58±1.20
21.52±0.97
ND
14.50±0.12
62.01±0.91
Zn
263.66±0.90
101.92±1.17
95.69±1.21
106.93±1.21
61.34±0.79
71.19±0.66
85.93±0.50
101.52±0.13
279.1±0.85
16.62±1.01
a
MFUG = metal-free unleaded gasoline; bMCUG = metal-containing unleaded gasoline; cND = Not
detectable; 1-6 are the numbers allocated to the six gasoline filling stations
9.4 CONCLUSIONS
Cellulose nanofibers functionalized with oxolane-2,5-dione was prepared as an
adsorbent for the removal of metal ions from gasoline samples. Fourier transform-IR and
solid state
13
C NMR spectroscopy confirmed the success of deacetylation of cellulose
acetate nanofibers to cellulose nanofibers and anchoring of oxolane-2,5-dione to the
deacetylated cellulose nanofibers. The morphological structures of the cellulose and
cellulose-g-oxolane-2,5-dione nanofibers were determined by SEM and the fiber diameter
was found to be 334 nm. The average specific surface area and the pore size of cellulose-goxolane-2,5-dione nanofibers (obtained from the BET analysis) were 14.31 m2 g-1 and
113.56 Å. The solid phase extraction of Cd, Cu, Fe, Pb and Zn in gasoline samples using
cellulose-g-oxolane-2,5-dione was investigated. Separation and preconcentration occurred
efficiently, resulting in reasonably high preconcentration factor of 60 and the low LOD and
LOQ ranging from 0.13-0.68 µg L-1 and 0.42-2.2 µg L-1, respectively.
The SPE method provided relatively good precision with %RSD lower than 3%. The
elution of metal ions from the solid material was performed with 2.0 mol L-1 HNO3. The
cellulose-g-oxolane-2,5-dione column could be reused over 30 adsorption/elution cycles
without any loss of its adsorption properties for analyte ions. The sorption capacities were
273.1, 183.6, 195.5, 236.2 and 182.4 mg g-1 for Cd, Cu, Fe, Pb and Zn, respectively. The
accuracy of the separation and preconcentration procedure was accessed by analysis of the
185
Chapter nine:
Solid phase extraction using electrospun nanofibers
CRM and spike recovery test. Satisfactory recoveries for spike recovery test and CRM
were as follows: 96-99 % and 97.5%-102%, respectively. The results obtained in this study
showed that cellulose-g-oxolane-2,5-dione nanofibers have a good potential for separation
and preconcentration of trace metal ions from organic matrices. In addition, the application
of cellulose-g-oxolane-2,5-dione nanofibers sorbent packed in SPE column (flow mode)
had an advantage over to batch mode method, such as reuse of nanofibers for several
cycles, simplicity and elimination of matrix effects.
9.5 REFERENCES
1. Reyes, M. N. M. & Campos, R. C. 2005. Graphite furnace atomic absorption
spectrometric determination of Ni and Pb in diesel and gasoline samples stabilized as
microemulsion using conventional and permanent modifiers. Spectrochimica Acta Part
B: Atomic Spectroscopy, 60, 615-624.
2. Chaves, E. S., F. G. Lepri, J. S. A. Silva, D. P. C. De Quadros, T. D. Saint’pierre And A.
J. Curtius 2008. Determination of Co, Cu, Fe, Mn, Ni and V in diesel and biodiesel
samples by ETV-ICP-MS. Journal of Environmental Monitoring, 10, 1211-1216.
3. Qu, R., Sun, C., Ma, F., Cui, Z., Zhang, Y., Sun, X., Ji, C., Wang, C. & Yin, P. 2012.
Adsorption kinetics and equilibrium of copper from ethanol fuel on silica-gel
functionalized with amino-terminated dendrimer-like polyamidoamine polymers. Fuel,
92, 204-210.
4. Teixeira, L. S. G., Bezerra, M. D. A., Lemos, V. A., Santos, H. C. D., De Jesus, D. S. &
Costa, A. C. S. 2005. Determination of copper, iron, nickel, and zinc in ethanol fuel by
flame atomic absorption spectrometry using on-line preconcentration system.
Separation Science and Technology, 40, 2555 - 2565.
5. Santos, D. S. S., M.G. A. Korn, M. A. B. Guida, G. L. Dos Santos, V. A. Lemos And L.
S. G. Teixeira 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by
Sequential Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase
Extraction. Journal of Brazzilian Chemical Society, 22, 552-557.
6. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
7. Bettinelli, M., S. Spezia, U. Baroni And G. Bizzarri. 1995. Determination of trace
elements in fuel oils by inductively coupled plasma mass spectrometry after acid
mineralization of the sample in a microwave oven. Journal of Analytical Atomic
Spectrometry, 10, 555-560.
8. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A.
M. & Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in
186
Chapter nine:
Solid phase extraction using electrospun nanofibers
automotive gasoline by X-ray fluorescence after pre-concentration on cellulose paper.
Talanta, 72, 1073-1076.
9. Roldan, P. S., Alcântara, I. L., Rocha, J. C., Padilha, C. C. F. & Padilha, P. M. 2004.
Determination of Copper, Iron, Nickel and Zinc in fuel kerosene by FAAS after
adsorption and pre-concentration on 2-aminothiazole-modified silica gel. Ecl. Quím.,
São Paulo, 29, 33-40.
10. Roldan, P. S., Alcântara, I. L., Padilha, C. C. F. & Padilha, P. M. 2005. Determination
of copper, iron, nickel and zinc in gasoline by FAAS after sorption and preconcentration
on silica modified with 2-aminotiazole groups. Fuel, 84, 305-309.
11. Roldan, P. S., Alcântara, I. L., Castro, G. R., Rocha, J. C., Padilha, C. C. F. & Padilha,
P. M. 2003. Determination of Cu, Ni, and Zn in fuel ethanol by FAAS after enrichment
in column packed with 2-aminothiazole-modified silica gel. Analytical and
Bioanalytical Chemistry, 375, 574-57.
12. Alves, V. N., Mosquetta, R., Coelho, N. M. M., Bianchin, J. N., Di Pietro Roux, K. C.,
Martendal, E. & Carasek, E. 2010. Determination of cadmium in alcohol fuel using
Moringa oleifera seeds as a biosorbent in an on-line system coupled to FAAS. Talanta,
80, 1133-1138.
13. Prado, A. G. S., Pescaraa, I. C., Evangelistaa, S. M., Holandaa, M. S., Andradeb, R. D.,
Suareza, P. A. Z. & Zarac, L. F. 2011. Adsorption and preconcentration of divalent
metal ions in fossil fuels and biofuels: Gasoline, diesel, biodiesel, diesel-like and
ethanol by using chitosan microspheres and thermodynamic approach. Talanta, 84,
759–765.
14. Shishehbore, M., Afkhami, A. & Bagheri, H. 2011. Salicylic acid functionalized silicacoated magnetite nanoparticles for solid phase extraction and preconcentration of some
heavy metal ions from various real samples. Chemistry Central Journal, 5, 1-10.
15. Zhang, L., Menkhaus, T. J. & Fong, H. 2008. Fabrication and bioseparation studies of
adsorptive membranes/felts made from electrospun cellulose acetate nanofibers. Journal
of Membrane Science, 319, 176-184.
16. Tarley , C. R. T. & Arruda, M. A. Z. 2004. Biosorption of heavy metals using rice
milling by-products. Characterisation and application for removal of metals from
aqueous effluents. Chemosphere, 54, 987-995.
17. O’connell, D. W., Birkinshaw, C. & O’dwyer, T. F. 2008. Heavy metal adsorbents
prepared from the modification of cellulose: A review. Bioresource Technology, 99,
6709-6724.
18. Splendore, G., Benvenutti, E. V., Kholin, Y. V. & Gushikem, Y. 2005. Cellulose
acetate-Al2O3 hybrid material coated with N-Propyl-1,4-diazabicyclo [2.2.2] octane
chloride: preparation, characterization and study of some metal halides adsorption from
ethanol solution. Journal of the Brazilian Chemical Society, 16, 147-152.
187
Chapter nine:
Solid phase extraction using electrospun nanofibers
19. Musyoka, S., Ngila, C., Moodley, B., Kindness, A., Petrik, L. & Greyling, C. 2011.
Oxolane-2,5-dione modified electrospun cellulose nanofibers for heavy metals
adsorption. Journal of Hazardous Materials, 192, 922-927.
20. Huang, Z.-M., Zhang, Y. Z., Kotaki, M. & Ramakrishna, S. 2003. A review on
polymer nanofibers by electrospinning and their applications in nanocomposites.
Composites Science and Technology, 63, 2223-2253.
21. Ma, Z., Kotaki, M. & Ramakrishna, S. 2005. Electrospun cellulose nanofiber as affinity
membrane. Journal of Membrane Science, 265, 115-123.
22. Son, W. K., Youk, J. H., Lee, T. S. & Park, W. H. 2004. Electrospinning of ultrafine
cellulose acetate fibers: Studies of a new solvent system and deacetylation of ultrafine
cellulose acetate fibers. Journal of Polymer Science Part B: Polymer Physics, 42, 5-11.
23. Liu, H. & Hsieh, Y.-L. 2002. Ultrafine fibrous cellulose membranes from
electrospinning of cellulose acetate. Journal of Polymer Science Part B: Polymer
Physics, 40, 2119-2129.
24. Amorim, F. A. C., Welz, B., Costa, A. C. S., Lepri, F. G., Vale, M. G. R. & Ferreira, S.
L. C. 2007. Determination of vanadium in petroleum and petroleum products using
atomic spectrometric techniques. Talanta, 72, 349-359.
25. Biermann, C. J., Chung, J. B. & Narayan, R. 1987. Grafting of polystyrene onto
cellulose acetate by nucleophilic displacement of mesylate groups using the
polystyrylcarboxylate anion. Macromolecules, 20, 954-957.
26. Kono, H., Yunoki, S., Shikano, T., Fujiwara, M., Erata, T. & Takai, M. 2002. CP/MAS
13C NMR study of cellulose and cellulose derivatives. 1. Complete assignment of the
CP/MAS 13C NMR spectrum of the native cellulose. Journal of the American Chemical
Society, 124, 7506-7511.
27. Kono, H., Erata, T. & Takai, M. 2003. Complete Assignment of the CP/MAS 13C
NMR Spectrum of Cellulose IIII. Macromolecules, 36, 3589-3592.
28. Aydin, F. A. & Soylak, M. 2010. Separation, preconcentration and inductively coupled
plasma-mass spectrometric (ICP-MS) determination of thorium(IV), titanium(IV),
iron(III), lead(II) and chromium(III) on 2-nitroso-1-naphthol impregnated MCI GEL
CHP20P resin. Journal of Hazardous Materials, 173, 669-674.
29. Tuzen, M., Soylak, M., Citak, D., Ferreira, H. S., Korn, M. G. A. & Bezerra, M. A.
2009. A preconcentration system for determination of copper and nickel in water and
food samples employing flame atomic absorption spectrometry. Journal of Hazardous
Materials, 162, 1041-1045.
30. Ingle, J. D. & Crouch, S. R. 1988. Spectrochemical Analysis, Prentice-Hall.
188
CHAPTER TEN:
FULL FACTORIAL DESIGN FOR THE OPTIMIZATION OF SIMULTANEOUS
PRECONCENTRATION OF TRACE METAL IONS IN GASOLINE SAMPLES
PRIOR TO THEIR INDUCTIVELY COUPLED MASS SPECTROMETRIC
DETERMINATION
ABSTRACT
The presence of trace elements in gasoline, unless they are used as additives, is
detrimental and they usually occur in very low concentrations in gasoline, requiring sensitive
detection techniques or preconcentration step prior to their determination. The use of full
two-level factorial design for optimization of simultaneous separation and preconcentration
of trace metal ions in gasoline samples using nanometer-sized alumina prior to determination
by ICP-MS is reported. The metal ions at a concentration of 20 µg L-1(20 mL) were retained
on 1.5 g of nanometer-sized Al2O3 at pH 8.0 and recovered with 5.0 mL of 2.0 mol L−1
HNO3. The adsorption capacities for the adsorbent were found to be 12.51, 10.74, 14.63,
13.11 and 11.65 mg g-1 for Co, Cr, Mn, Ni and Ti, respectively. Under optimized
experimental conditions, the precision was ≤ 2% (n =15), limits of detection and
quantification ranged from 0.03-0.10 µg L-1 and 0.10-0.22 µg L-1, respectively, and the
maximum preconcentration factor was 120. The accuracy of the SPE method was confirmed
by performing spike recovery tests with organic and inorganic standards. The optimized
nanometer-sized Al2O3 SPE procedure was then applied for the separation and
preconcentration of metal ions in commercial gasoline samples.
Keywords: Preconcentration, nanometer-sized alumina, factorial design, trace metals,
gasoline
10.1 INTRODUCTION
The presence of metal ions in petroleum products such as gasoline is undesirable, not
only due to the possibility of damaging vehicle part, gum formation, catalytic poisoning and
poor fuel performance, but also because of the pollution caused by the release of toxic metals
into the environment during fuel combustion.1 Metallic elements are generally present in very
low concentrations in fuel samples. Therefore, sensitive detection techniques or
preconcentration step prior to their determination are required.1 Notwithstanding the
sensitivity of spectrometric techniques such as inductively coupled plasma mass spectrometry
189
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
(ICP-MS) and electrothermal atomic absorption spectrometry (ETAAS), preliminary sample
preparation is required. Therefore, there is a crucial need for the separation and
preconcentration of trace elements before their determination.1,2 In addition, the direct
introduction of organics in ICP-MS is detrimental to the instrument; therefore, sample
pretreatment prior to metal ion determination is required.
Techniques involving sample pretreatment for the quantification of trace elements in fuels
have been reported in literature. These methods include microwave-assisted acid digestion,3
microwave-induced
combustion,4
conventional
ashing
and
acid
dissolution5
and
electrothermal vaporization.2 The limitation of conventional ashing and acid dissolution
methods is that they are time-consuming and volatile elements may be lost.3 Microwave
digestion methods solve the problem of volatilization, but they increase the risk of crosscontamination. Electrothermal vaporization is known to eliminate oxygen addition and
reduce the organic matrix interference. Nevertheless, its parameters have to be optimized for
each element thus lengthening the experimental procedure.3 Therefore, an accurate and
reliable analytical procedure based on simultaneous separation and preconcentration of
analytes prior to analysis in fuel samples, is required. Few studies have reported on the
separation and preconcentration of trace metals in organic phases using simple methods such
as solid phase extraction (SPE).2,6-11 The advantages of using SPE method include high
sensitivity, possibility of performing simultaneous preconcentration step, reduced matrix
interferences, reasonable preconcentration factors with relatively rapid separation and low
cost.12,13 In addition, SPE has become more attractive due to the use of different adsorbent
materials with high sorption capacities such as nanometer-sized metal oxides.
Metal oxide adsorbents such as alumina, titania and zirconia have been reported for the
separation and preconcentration of metal ions.12,14,15 They are attractive as adsorbents due to
their large surface area, intrinsic surface reactivity and ability to chemisorb many
substances.16 The use of nanometer-sized Al2O3 as SPE sorbent material for preconcentration
of trace metals has recently received more attention.12,17,18
Recently, multivariate techniques have been used for optimization of preconcentration
methods.19-22 The advantages of using multivariate optimization include the reduction of the
number of experiments needed to be carried out, thus, resulting in lower reagent consumption
and significantly less laboratory work.19 Furthermore, multivariate methods make it possible
to understand conditions that cannot be explained by the traditional univariate approach; for
instance, the interactions between the factors that influence the analytical response.22-26
190
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Therefore, this study reports on the preparation and application of nanometer-sized
alumina for separation and preconcentration of trace metals in gasoline samples prior to ICPMS detection. A full two-level factorial design with a central point was used for optimization
of experimental variables (pH, eluent concentration and sample flow rate) that affect the
retention/desorption of metal ions. To the best of our knowledge, this is the first time that the
optimized nanometer-sized Al2O3 SPE methodology is proposed for separation and
preconcentration of Co, Cr, Mn, Ni and Ti in gasoline samples.
10.2 MATERIALS AND METHODS
10.2.1 Apparatus
A Perkin-Elmer Sciex ELAN 6000 (Perkin-Elmer SCIEX Instruments, Concord, Canada)
inductively coupled plasma mass spectrometer (ICP-MS) was used for all measurements. The
ICP-MS instrument was optimized daily and operated as recommended by the manufacturer.
The operating conditions are presented in Table 10.1. Argon of 99.996% purity (Afrox, South
Africa) was used. Comparative experiments for the determination of metal ions were
performed using a Spectro Arcos 165 ICP OES (SPECTRO Analytical Instruments, GmbH,
Germany) equipped with Cetac ASX-520 autosampler after acid digestion. The operating
conditions were as follows: forward power: 1400 W, plasma argon flow rate: 13 L min-1,
auxiliary argon flow rate: 2.00 L min-1, nebulizer argon flow rare: 0.95 L min-1. The most
prominent atomic and ionic analytical spectral lines of the metal ions were selected for
investigation, that is, Co 230.786, Cr 267.716 nm, Mn 257.611, Ni 231.604 and Ti 334.941
Morphological structure of the Al2O3 was observed using scanning electron microscope
(SEM) (JSM-6360LVSEM, JEOL Co., Japan) after gold coating. In SEM analysis, most
samples need to be coated to make them conductive. Therefore, the gold coating was used
because the alumina material under study does not conduct electricity. The specific surface
area value was determined from adsorption isotherms by using the Brunauer, Emmett and
Teller (BET) multipoint method using Surface Area and Porosity Analyzer (ASAP2020
V3.00H, Micromeritics Instrument Corporation, Norcross, USA). All the gases used for
analysis were instrument grade. X-ray powder diffraction (XRD) measurements were carried
out with a Philips X-ray generator model PW 3710/31 a diffractometer with automatic
sample changer model PW 1775 (scintillation counter, Cu-target tube and Ni-filter at 40 kV
and 30 mA).
191
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Table 10.1. Operational ICP-MS parameters
RF power
Gas flow rates
Outer
Intermediate
Carrier
Resolution
Sweeps per reading
Dwell time
Readings per replicate
Replicates
Auto lens
Isotopes
a
IS = Internal standard
1100
15 L min-1
1.2 L min-1
0.95 L min-1
0.7 a.m.u. (10% of the peak height)
1
25 ms
100
3
On
59
Co, 52Cr, 55Mn, 60Ni and 48Ti, 45Sc (ISa)
10.2.2 Reagents and Solutions
All reagents were of analytical grade unless otherwise stated and double distilled
deionised water (Millipore, Bedford, MA, USA) was used throughout the experiments.
Anhydrous aluminium chloride (Sigma-Aldrich, St. Loius, MO, USA) was used as a
precursor for the preparation of nanometer-sized alumina. Synthetic gasoline was prepared
by mixing 91% isooctane and 9% n-heptane (Sigma Sigma-Aldrich, St. Loius, MO, USA).
Spectrascan stock solutions (1000 mg L-1) of Co, Cr, Mn, Ni, and Ti (Industrial Analytical
(Pty) Ltd, Johannesburg, South Africa) were used to prepare the working solutions for SPE at
concentrations of 30 µg L-1 for other metal ions. Working solutions (prepared in an organic
medium), as per the experimental requirements, were freshly prepared from the stock solution
for each experimental run. A Spectrascan multi-element standard solution at a concentration
of 100 mg L-1 (Industrial Analytical (Pty) Ltd, Johannesburg, South Africa) was used to
prepare working standard solutions at concentrations of 10-150 µg L-1 (Co, Cr, Mn, Ni, and
Ti) for measurements of concentrations of analytes in all model and sample solutions.
Scandium was used for internal standardization. Conostan custom made multi-element oil
standard used in the experiment studies was obtained from SCP Science (Quebec, Canada).
Solutions of nitric acid at concentrations of 1.0, 2.0 and 3.0 mol L-1 (used for the elution of
the analytes from the column) were prepared from ultrapure concentrated acid (65%, Sigma-
192
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Aldrich, St. Loius, MO, USA). The pH adjustments were performed with glacial acetic acid
(Merck, Darmstadt, Germany) and ammonia (Sigma-Aldrich, St. Loius, MO, USA) solutions.
10.2.3 Preparation of Nanometer-Sized Alumina Using Sol-Gel Method
Nanometer-sized alumina was prepared according to Rogojan et al.27A mass (2.66 g) of
AlCl3 was dissolved in 25 ml absolute ethanol followed by drop wise addition of 28%
ammonium solution. The addition of the latter was done in order to for a sol gel to form. The
resulting sol gel was left to maturate for 30 hours at room temperature and then dried for 24
hours at 100°C. Finally, the gel was calcined by heating in a furnace at a rate of 20°C min-1 to
1000 °C and holding it for three hours.
10.2.4 Preparation of the Column
Polyethylene columns (6.5 cm length and 1.35 cm i.d.) were used for separation and
preconcentration of metal ions. A total of 1.5 g of alumina was slurried in water and then
loaded onto the columns. A porous frit was placed at the bottom of the column and on top of
the packing material to avoid disruption of the sorbent during sample percolation. The
column bed height was approximately 3 cm. The columns were washed with double distilled
deionised water followed by conditioning with 10 mL ammonium acetate buffer (1.0 M, pH
9.0) and then 10 mL of ethanol.
10.2.5 Preconcentration and Recovery of Co, Cr, Mn, Ni, and Ti in a Synthetic Gasoline
Solution
The procedure for the preparation of gasoline–ethanol–water mixture was carried out
according to Ozcan and Akman.28 A 10 ml aliquot of synthetic gasoline sample was placed in
a 100 ml polyethylene volumetric flask followed by the addition of 5 ml of concentrated
HNO3 and10 ml of water. The mixture was spiked with 3.0 mL of a 1.0 mg L-1 multi-element
oil standard solution and made up to the mark with ethanol to obtain 30 µg L-1 concentration
of each metal ion. The mixture was homogenized by shaking and a single phase solution was
obtained. It should be noted that the stability of the gasoline-ethanol- water mixture was not
monitored. This is because the resulting mixture was passed through the column immediately
after homogenization. A 20 ml aliquot of the model metal solution was passed through the
Al2O3 column with a flow rate of 2.0 mL min-1. The column was washed with 10 mL of
193
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
double distilled deionised water to remove excess organic solution, followed by 5.0 mL of
ammonium acetate buffer solution to remove major cations (Na, Ca, K, etc); lastly the metal
ions were eluted with 5 mL of 2.0 mol L-1 HNO3 solution. It should be noted that off-line
SPE system was used. This means the fraction were collected and then brought to the ICPMS for analysis. The same procedure was applied to the blank solutions and real samples. For
gasoline samples, 1.0 mL of the sample was used instead of 10 ml. In between the
experiments, the column was washed as described in Section 10.2.4.
10.2.6 Optimization Approach
The optimization of the separation and preconcentration method was carried out using a
23 full factorial design involving three variables i.e. pH, eluent concentration (EC) and
sample flow rate (SFR) which were considered as factors. Maximum, central point and
minimum levels in Table 10.2 for each factor were chosen according to data from previous
experiments. All the experiments were carried out in random order. The experimental data
was processed by using the Minitab version 15 statistic software programs.
Table 10.2. Factors and levels used in 23 factorial design for the separation and
preconcentration of metal ions
Variable
pH
EC (mol L-1)
SFR (mL min-1)
Low level (-1)
5
1.0
1.0
Central point (0)
7.5
2.0
2.0
High level (+1)
10
3.0
3.0
10.2.7 Procedure for Microwave Acid Digestion of Gasoline Samples
The Microwave acid digestion procedure was carried out according to Kowalewska et
al.29 Briefly, 5.0 mL of the gasoline sample was placed into a Teflon vessel followed by 6 mL
HNO3 (65%) and 2.0 mL H2O2 (30%). The vessels were inserted into a microwave unit and
heated according to the conditions recommended by the manufacturer. The digested content
was left to cool down to room temperature. After cooling, the vessels were opened and 2 ml
of concentrated HNO3 and 2 mL of hydrogen peroxide were added, and the heating program
was repeated. Finally, the Teflon vessel contents were cooled down to room temperature and
quantitatively transferred to a 50 mL calibration flask, 1 mL of concentrated nitric acid was
194
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
added and the flask was filled up to the mark using double distilled deionised water. The
latter submitted to the same procedure was used as a blank. The samples were then analyzed
with ICP OES.
10.3 RESULTS AND DISCUSSION
10.3.1 Characterization of the Nanometer-Sized Alumina
The surface and textural morphology (SEM image) of nanometer-sized Al2O3 obtained by
the sol-gel method starting from aluminum chloride as a precursor is illustrated in Fig. 10.1.
The SEM image showed fine particles and their diameter was estimated to range from 40-60
mn.
Fig. 10.1. Scanning electron microscopy images of alumina obtained by sol-gel methods
starting from AlCl3 as precursor, calcined at 1000°C for three hours.
The BET results revealed that the nanometer-sized Al2O3 adsorbent has porous
characteristics with a remarkable specific surface area of 312 m2 g−1. The relative high
specific surface area revealed the availability of adsorbent sites for metal ions
preconcentration (adsorption). This phenomenon is significant for the application of Al2O3 in
the separation and preconcentration system for the determination of metals in an organic
195
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
matrix. Furthermore, the relatively high specific surface area represents substantial
developments over ion-exchange adsorbents such Chelex-100.
X-ray diffraction pattern (for 2θ diffraction angles from 10° to 80°) of the nanometersized alumina calcined at 1000°C for 3 hours is presented in Fig. 10.2. The XRD pattern
showed the crystalline structure of the nanometer-sized particles indicating various peaks
indexed to alumina.30,31 The peaks were attributed to two crystallization phases of alumina,
that is, α-Al2O3 and γ-Al2O3.
Fig. 10.2. X-ray diffraction pattern of alumina obtained by sol-gel methods starting from
AlCl3 as precursor, calcined at 1000 °C for three hour
10.3.2 Factorial Design
The nanometer-sized alumina SPE method had several experimental variables to be
optimized. However, the parameters (factors) that may possibly affect the separation and
preconcentration process are sample pH, eluent concentration and sample flow rate. Eluent
type, eluent flow rate and adsorbent mass were fixed as nitric acid aqueous solution, 2 mL
min-1 and 1.5 g, respectively. In order to determine the effect of pH, eluent concentration,
sample flow rate and their interactions on the preconcentration method, a two-level full
factorial design (23) with three replicates of the central point (CP) was performed. Replicates
of the central point were carried out in order to determine the experimental errors and
196
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
curvature tests.19 Table 10.3 shows the experimental design matrix and the results obtained
from each experiment. The percentage recovery was used as the analytical response and it
was used for evaluation of the factorial design. The percentage recoveries were calculated by
relating the obtained concentration (Cf) of the analyte to the original concentration (Ci) of the
metal ion in the model solution (Eq. 1).
%R 
Cf
Ci
 100
(1)
It can be seen from Table 3 that quantitative recoveries for all five metal ions are obtained
in experiments 9-11, which are the central points. In this study, the quantitative recovery
were defined as the percentage recovery of a target analyte that is more than or equal to 95%.
It can be seen from Table 10.3 that quantitative recoveries for all five metal ions are obtained
in experiments 9-11, which are the central points. The significance (p-value = 0.05) of the
experimental factors in the performance of the SPE system was checked by performing an
analysis of variance (ANOVA). The estimated main effects and their interactions can be seen
in the Pareto charts presented in Fig. 10.3. The bars that exceed a vertical reference line (95%
confidence interval) are significant values with respect to the response.
Table 10.3. Design matrix and the results of metal ions
Experiment pH
1
2
3
4
5
6
7
8
9 (CP)
10(CP)
11(CP)
+1
+1
+1
+1
-1
-1
-1
-1
0
0
0
EC (mol SFR (mL
L-1)
min-1)
+1
+1
-1
-1
+1
+1
-1
-1
0
0
0
+1
-1
+1
-1
+1
-1
+1
-1
0
0
0
Recovery (%)
Co
63.6
60.2
65.3
56.9
33.9
27.3
34.1
26.8
98.5
98.7
99.2
197
Cr
70.0
67.9
69.2
64.7
34.9
41.3
42.4
26.5
96.9
97.6
97.3
Mn
90.2
89.7
88.3
65.1
30.1
25.8
23.6
35.4
97.3
98.6
97.3
Ni
88.7
85.9
48.8
47.0
38.3
40.0
35.2
26.0
101.5
100.4
101.7
Ti
94.9
80.4
64.6
75.9
50.8
75.1
63.3
50.8
101.1
100.7
102.2
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Sample pH was, in general, the most important parameter (p = 0.05) for all the metal ions.
It can be seen from Fig. 10.3 that the significant factor for retention of Cr and Mn onto the
Al2O3 column was the pH of the solution. When the pH of the model solution was 7.5, the Cr
and Mn recoveries were greater than 95%. The effect of other factors, that is, eluent
concentration and sample flow rate were not significant at 95% confidence level. According
to the Pareto chart (Fig. 10.3) for Co, sample pH and sample flow rate are statistically
significant at the 95% confidence level. However, the pH effect was significantly higher than
the eluent concentration. This implies that, the later had lower influence in the
preconcentration process of cobalt. The results (Fig. 10.3) for preconcentration of Ni
demonstrated that sample pH and eluent concentration as well as their interaction were
statistically significant at the 95% confidence level. In the case of Ti retention in investigated
levels, all the variables and their interactions are not statistically significant at 95%
confidence level. Although, all the main effects were not statistically significant, pH had a
major effect on the retention of Ti and analytical response. This is supported by higher
recoveries at higher pH values (Table 10.3) as compared to other analytes. In addition it can
be seen from Table 10.3 that Ti can be retained at a wide pH range with recoveries> 50% as
compared to other analyses. Since sample pH played a major role on the analytical response,
it is possible that it influenced the overall effects of the studied factors.
198
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Fig. 10.3. Pareto chart of standardized effects for variables in the separation and
preconcentration of Co, Cr, Mn, Ni and Ti. A = pH; B = eluent concentration (mol L-1) and C
= sample flow rate (mL min-1)
The ANOVA results (data not shown) showed that the curvature was significant at 95%
confidence level. This means that there is an experimental region for maximum sorption of
metal ions from its central point to the value that represents the pH variable at its high level.19
In addition, the surface of alumina is positively charged when the pH is lower than its
isoelectric point (≈7.3),32 this results in an electrostatic repulsion for the target metal ions.33 It
is reported that when pH is above the isoelectric point, the surface of the nanometer-sized
alumina powder attracts the analytes of interest and lead to an enhancement of the adsorption
199
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
efficiency.33 In view of the information, the optimum eluent concentration and sample flow
rate chosen for simultaneous separation and preconcentration of Co, Cr, Mn, Ni and Ti in a
gasoline matrix concur with the conditions established by experiment 9-11. However, due to
the significance of the curvature, sample solution acidity was found to be the factor with
highest effect; optimum sample pH was set at 8.0. Since the sample flow rate had a little or
no significant effect at 95% confidence level, 3.0 mL min-1 was used. Therefore, the
simultaneous SPE procedure was carried under the following conditions; sample pH, eluent
concentration and sample flow rate were 8.0, 3.0 mol L-1 and 2.0 mL min-1, respectively.
10.3.3 Effect of Sample Volume
Due to the low concentrations of trace metals in gasoline samples, it is crucial to transfer
these analytes into smaller volumes for a high preconcentration factor by using sample
solutions with large volumes.34,35 Therefore, the effect of sample volume on the adsorption of
Co, Cr, Mn, Ni and Ti onto nanometer-sized Al2O3 was investigated in the range of 50-700
mL, while keeping the metal ion concentration fixed at 30 µg L-1. It can be seen from
Fig.10.4 that the retention of metal ions can be achieved quantitatively (≥95%) by up to 600
mL of the sample. Therefore, the highest preconcentration factor was found to be 120 when
the adsorbed metal ions were eluted with 5 mL of 2 mol L−1 HNO3. At volumes higher than
600 mL, a decrease in quantitative recoveries of metal ions was observed. This might be due
to the saturation of the active sites of the adsorbent. Therefore, as a compromise, a sample
volume of 100 mL was used for was used in real sample analysis. This volume was chosen to
speed up the analysis time.
200
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Fig.10.4. Effect of sample volume on the recoveries of metal ions: pH 7.0; analyte
concentration 30 µg L-1; amount of sorbent 1.5 g; flow rates of sample and eluent 2.0 mL
min−1; eluent volume 5 mL; replicates n = 3
10.3.4 Adsorption Capacities of Metal Ions
The adsorption capacity is an important factor, because it determines how much sorbent is
required to quantitatively concentrate the analytes from a given solution.34 Preliminary
adsorption studies with nanometer-sized alumina revealed that 35 minutes is adequate time
for the system to reach equilibrium. Therefore, 0.1 g nanometer-sized Al2O3 was equilibrated
in 50 mL of Co, Cr, Mn, Ni and Ti ethanol solutions at concentrations 30 to 250 mg L-1 by
shaking for 35 minutes at pH 8.0. The amount of metal ions in solution was determined by
GFAAS. The experimental data were fitted into the general equation of the modified
Langmuir model presented in Eq. 1.36 The later was used to calculate the maximum
adsorption capacity.
Ce
1
1

Ce 
qe qmax
K L qmax
(1)
The results showed that adsorption capacity of the analytes probably differ due to their
size, degree of hydration and the value of their binding constant with nanometer-sized
alumina. The maximum adsorption capacities were found to be 12.51, 10.74, 14.63, 13.11
201
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
and 11.65 mg g-1 for Co, Cr, Mn, Ni and Ti, respectively. The adsorption capacities of
nanometer-sized alumina for metal ions were similar or better than those reported in Ref.12
10.3.5 Regeneration of the Adsorbent
The stability and regeneration possibility of the Al2O3 adsorbent were investigated and
the results are summarised in Table 10.4. The adsorbent can be reused after regeneration with
5.0 ml of a 2.0 mol L−1 HNO3 solution and 20 ml distilled water, respectively, and is
relatively stable up to 45 runs without obvious decrease in the recoveries for the studied ions.
Table 10.4. Column regeneration
Analytes
No. of cycles
1
99.2±0.6
97.9±1.3
98.6±0.8
100.3±0.2
100.7±0.4
Co
Cr
Mn
Ni
Ti
22
98.7±1.1
97.6±1.3
97.9±0.8
99.4±1.0
98.3±1.4
45
96.2±1.4
97.1±1.1
95.3±1.2
97.1±0.5
97.9±0.4
10.3.6 Analytical Performance of the Nanometer-Sized Alumina SPE Method
The analytical performance of the nanometer-sized alumina SPE method under optimum
conditions for separation and preconcentration of metal ion was evaluated and the results are
presented in Table 10.5. The sensitivity of the SPE was defined as the gradient (slope) of the
calibration graph. The results indicated that the SPE method was more sensitive to Ti, Ni and
Mn as compared to the Co and Cr. Thus the highest slope obtained was 127.94 cps L µg-1 for
Ti while the lowest was 21.13 cps L µg-1 for Cr. The low sensitivity of Cr might be due to the
spectral interferences from
36
Ar16O. This can be avoided by using ICP-MS that is equipped
with collision cell. It should be noted that the ICP-MS used in this study uses mathematical
equations to correct for spectral interferences.
The limit of detection (LOD) and the limit of quantification (LOQ) were defined as
LOD  3Sd
m
and LOQ  10Sd , respectively, where m and Sd are the slope of the
m
analytical curve and the standard deviation of 20 consecutive measurements of the blank
signal, respectively.37 For 100.0 mL of sample solution used, LOD and LOQ for Co, Cr, Mn,
202
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Ni, and Ti are presented in Table 10.5. The overall precision (reproducibility) of the SPE
method, expressed as relative standard deviation (n = 15, 20 µg L-1), was found to be ≤2%. In
the preliminary trials, we had to establish how long it takes to load the sample and elute. We
found that the time required for preconcentration of 50 mL sample was about 19 min (17 min
percolation, 60 s elution and 60 s conditioning). However, the SPE manifold used in this
study could handle up 24 cartridges (24 samples) and they were all processed within 23 min.
Hence, the throughput sample was approximately 63 samples h-1.
The analytical figure of merits of the nanometer-sized Al2O3 SPE method were compared
with the ones obtained using a comparative method (microwave acid digestion method).
According to the results in Table 10.5, the SPE method was more sensitive compared to acid
digestion method. In addition, as expected, the LOD and LOQ of the proposed
preconcentration method were better than those of the microwave-assisted digestion method.
This is because the LOD and LOQ of MAD technique are the same as the instrumental limits
of detection while the preconcentration enhances the LOD and LOQ of the method. At 95%
confidence level, the statistical F-test results showed that the precision of the two sample
pretreatment methods were not significantly different. In terms of correlation efficient, the
acid digestion had better R2 compared to the SPE method. The SPE method had a higher
throuput compared to microwave-assisted digestion method (10 samples h-1).
A comparison of the proposed method with other sample preparation techniques in terms
of selected analytical parameters such as LOD, LOQ and %RSD was also carried out (Table
10.6). Comparison of analytical figures of merit for the present method with other sample
preparation techniques indicated that the LOD, LOQ and %RSD of the nanometer-sized
alumina SPE were comparable or even better than the reported methods.
10.3.7 Validation of the Nanometer-Sized Alumina SPE Method
Due to the absence of certified reference material (CRM) that is similar to the
investigated samples, the accuracy of the proposed separation and preconcentration method
was examined by standard addition method. Gasoline sample (1-MFUG) was spiked with
organic and inorganic standard solutions. In addition, the aim of spiking the gasoline sample
with organic and inorganic standard solutions was to evaluate the nanometer-sized alumina
sorption efficiency to different metal species in gasoline. This because trace element forms in
petroleum products is not fully known and different species may display different adsorption
behaviors. 6 As it can be seen in Table 10.7, similar percentage recoveries were obtained for
203
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
organic and inorganic forms. This implied that nanometer-sized Al2O3 preconcentration
system may be used for the sorption of trace elements in their inorganic or metal-organic
forms. In addition, the results obtained (Table 10.7), confirmed the accuracy of the
preconcentration method, taking into consideration that the recoveries were in the range from
97-101%.
204
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Table 10.5. Analytical performances for the proposed nanometer sized Al2O3 SPE method and microwave-assisted digestion method
Analytes
Sensitivity (cps L µg-1)
SPEa
21.1
34.21
70.7
109
128
MADb
19.3
25.3
52.0
88.4
90.5
Correlation
coefficient
SPE
MAD
0.9987
0.999
0.9976
0.9995
0.9980
0.9999
0.9971
0.9996
0.9911
0.9994
LOD (µg L-1)
LOQ (µg L-1)
Precision (%RSD)
SPE
0.10
0.04
0.03
0.07
0.05
SPE
0.22
0.18
0.10
0.22
0.15
SPE
1.5
1.3
1.0
1.2
1.1
Co
Cr
Mn
Ni
Ti
a
SPE: Solid phase extraction method; MAD: Microwave acid digestion method.
MAD
0.82
0.55
0.36
0.29
0.30
MAD
2.74
1.84
1.20
0.97
1.02
MAD
1.8
1.1
1.7
1.5
1.1
Table 10. 6. Comparison of the proposed nanometer-sized SPE method with other methods used for determination of trace metals in gasoline
Analytes
Ni
Mn
Cr and Ni
Cr
Ni
Mn and Ni
Co, Cr, Mn, Ni
and Ti
Sample treatment method
Microemulsions
Microemulsion
Microemulsion
Automatic microemulsion
Modified silica gel SPE
Emulsion
Nanometer-sized alumina SPE
Detection
ETAAS
GFAAS
MIP-OES
GFAAS
FAAS
ETV-ICP-MS
ICP-MS
LOD (µg L-1)
0.8
0.5
0.9 and 20
0.024
2
0.02 and 0.38
0.10, 0.04, 0.03,
0.07 and 0.05
LOQ(µg L-1)
2.6
1.7
3.0 and 70
0.070
6.6
0.07 and 1.3
0.22,
0.18,
0.10, 0.22 and
0.15
Precision (%)
NI
6
>10
NI
NI
NI
1.5, 1.3, 1.0, 1.2 and
1.1
Ref.
[38]
[39]
[40]
[41]
[11]
[1]
This work
GFAAS= graphite furnace atomic absorption spectrometry; ETAAS = electrothermal atomic absorption spectrometry; MIP-OES = microwave
plasma optical emission spectrometry; FAAS= flame atomic absorption spectrometry; ICP-MS = inductively coupled plasma mass spectrometry
205
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Table 10.7. Determination of Co, Cr, Mn, Ni and Ti (µg L-1) in gasoline sample spiked with inorganic and organic standard solutions (mean ±
standard deviation, sample volume = 100 mL, n= 3)
Co
c
IS
a
MOSb
a
Added
0
5
20
0
5
20
c
Found
6.20±1.31
11.2±1.1
25.9±0.5
6.20±1.31
11.1±1.2
26.1±1.4
d
Recovery
99.4±1.3
98.3±2.4
97.8±1.3
99.4±2.3
Found
74.6±1.6
79.5±1.4
94.4±1.2
74.6±1.6
79.5±1.0
94.5±1.0
Cr
Recovery
97.8±1.6
99.1±2.1
98.6±3.4
99.3±1.4
Found
85.9±0.9
90.8±1.1
105±2
85.9±0.9
90.6±1.2
106±1
Mn
Recovery
99.2±1.5
97.3±1.7
97.8±2.1
98.8±2.0
Found
44.9±1.1
49.8±1.2
64.9±1.4
44.9±1.1
49.8±1.0
64.6±1.0
IS: Inorganic standard; bMOS = metallo-organic standard; cConcentration in µg L-1; Recovery in %
206
Ni
Recovery
97.4±2.4
100±1
97.6±1.6
98.3±1.9
Ti
Found
2.22±0.56
7.13±1.74
22.10±1.31
2.22±0.56
7.18±1.12
21.9±1.5
Recovery
98.2±1.5
101±0.7
99.2±0.8
98.3±1.2
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
10.3.8 Application of Nanometer-Sized Alumina SPE Method
The preconcentration method was applied to the analysis of ten gasoline samples from six
different filling stations in Johannesburg, as presented in Table 10.8. As shown in this table,
the concentrations of Co are quite low (˂ 20 µg L-1) for almost all the samples except for 2MCUG sample (23.10 µg L-1). The concentration of Cr was found to be relatively high (˂ 50
µg L-1) in 1-MFUG, 2-MCUG and 3-MCUG samples. The Mn concentrations were generally
high in all the samples. Thus, the highest concentration was 38.23 mg L-1 and the lowest was
85.83 µg L-1. The concentration of Ni ranged from 6.74 to 195.36 µg L-1 and Ti content was
relatively low except for 2-MCUG and 6-MCUG samples. In addition, it can be seen from
Table 10.8 that Co, Cr and Ti could not be quantified in some of the samples, as their
concentrations were found to be below the LOD. It is worthwhile mentioning that Mn in
samples 2-MCUG, 3-MCUG and 6-MCUG was used as a fuel additive, and it was therefore
present in both free and organic compound forms (methylcyclopentadienyl manganese
tricarbonyl, MMT). It should be noted that for determination of Mn in 2-MCUG, 3-MCUG
and 6-MCUG samples, the latter were further diluted to obtain µg L-1 range.
The results obtained by the developed preconcentration method were compared to the
results obtained by a comparative method using ICP OES after microwave-assisted digestion
(Table 10.8). It should be noted that acid digestion followed by ICP OES determination was
taken as the standard method in this study. The paired t-test (95% confidence level) showed
that the results obtained by the proposed procedure were not significantly different with those
obtained by the comparative method. Although the results obtained by the two methods were
not significant different at 95% confidence level, SPE/ICP-MS showed better analytical
performances compared to ICP OES after microwave-assisted digestion. In addition, the
other advantage of the proposed method described in this study is that it does not require
rigorous acid digestion unlike the microwave-assisted digestion method. The column method
is also advantageous because it minimizes the risks of cross-contamination during acid
digestion. It should be noted that since samples are diluted, preconcentration prior to analyte
determination, is required even if techniques with low detection limits like ICP-MS are used.
In addition, the advantage of using SPE techniques prior ICP-MS detection is that it
eliminated organic matrix completely. For instance, if the sample with high content of
organics is injected into the ICP-MS, this will results in a deterioration of the precision of
207
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
measurements, which in turn leads to the high of LOD for most elements because of the
formation of carbon-containing ions (C2+, CO2+ and ArC+)
208
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
Table 10.8. Concentrations (in µg L-1) of metal ions in gasoline samples determined by ICP-MS in aqueous solutions resulting from nanometersized Al2O3 preconcentration procedure (sample volume = 100 mL) and ICP OES in aqueous solutions resulting from microwave-assisted
digestion procedure
Samples
Co
Cr
Mn
Ni
Ti
ICP-MS
ICP OES
ICP-MS
ICP OES
ICP-MS
ICP OES
ICP-MS
ICP OES
ICP-MS
ICP OES
a
1-MFUG 6.28±0.51 6.22±0.43 74.73±0.89 75.01±0.83 85.83±1.10 85.77±1.13 44.87±0.32 44.50±0.25 2.15±0.06 2.21±0.03
2-MCUGb 23.10±0.11 22.82±0.22 59.27±0.31 59.10±0.29 15.68±1.15d 15.56±2.05d 74.33±0.56 73.96±0.61 24.77±0.21 24.98±0.18
ND
2-MFUG 4.43±0.23 4.21±0.21 11.85±0.11 11.80±0.16 94.09±0.98 94.10±1.00 21.89±0.07 21.78±0.05 ND
d
d
6.03±0.22
5.91±0.31
63.83±0.78
63.78±0.81
38.23±2.56
38.42±3.65
66.46±0.71
65.89±0.68
14.36±0.44
14.43±0.51
3-MCUG
c
ND
ND
ND
149.12±1.31 148.74±1.21 17.11±0.03 16.81±0.01 ND
ND
3-MFUG ND
ND
4-MFUG 4.42±0.15 4.33±0.10 3.17±0.05 2.95±0.07 91.20±0.56 90.81±0.66 22.65±0.08 22.36±0..10 ND
ND
109.49±1.27 110.02±1.30 195.36±1.01 194.75±0.97 ND
ND
5-MCUG 16.11±0.36 15.85±0.41 ND
ND
ND
ND
103.57±1.13 103.45±1.10 6.47±0.01
6.50±0.04
ND
ND
5-MFUG ND
d
d
ND
30.96±0.25 31.05±0.31 22.01±1.95 21.50±3.12 54.97±0.33 55.17±0.30 35.96±0.69 36.12±0.75
6-MCUG ND
ND
20.58±0.21 20.32±0.32 115.39±1.05 114.60±1.02 31.37±0.44 31.51±0.51 12.58±0.12 12.41±0.11
6-MFUG ND
a
MFUG: metal-free unleaded gasoline; bMCUG: metal-containing unleaded gasoline; cND: Not detectable; dConcentration in mg L-1; 1-6 are the
numbers allocated to the six gasoline filling stations
209
Chapter ten
Optimization of simultaneous preconcentration of trace metal ions
10.4 CONCLUSIONS
A SPE method for simultaneous separation and preconcentration of trace metal ions in
gasoline samples using nanometer-sized alumina prior to ICP-MS determination has been
investigated. The application of a two-level full factorial design for the optimization of the
variables affecting the separation and preconcentration method was found to be efficient,
requiring a reduced number of experiments. Under optimized conditions, the quantitative
retention and elution of metal ion was achieved when sample solution pH, eluent
concentration and flow rates were 8.0, 2.0 mol L-1 and 2.0 mL min-1, respectively. The
nanometer-sized alumina could be used for more than 45 adsorption/elution cycles without
a significant change in the recoveries (≤5%). The simultaneous separation and
preconcentration of metal ions occurred efficiently, resulting in a reasonably high
preconcentration factor of 120 with low LOD and LOQ values ranging from 0.03-0.10 µg
L-1 and 0.10-0.22 µg L-1, respectively. The preconcentration system provided relatively
good precision with %RSD lower than 2%. The adsorption capacities of nanometer-sized
Al2O3 for Co, Cr, Mn, Ni and Ti were found to be 12.31, 10.54, 14.13, 13.01 and 11.50 mg
g-1, respectively. The developed nanometer-sized alumina SPE is simple, cheap, efficient,
precise and accurate since results obtained with the analysis of spiked samples presented
good agreement with the added values. In addition, SPE/ICP-MS showed better analytical
performances compared to ICP OES after microwave-assisted digestion. The optimized
nanometer-sized alumina SPE method was applied in the determination of Co, Cr, Mn, Ni
and Ti in ten real gasoline samples obtained from six local filling stations.
10.5 REFERENCES
1. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
2. Roldan, P. S., Alcântara, I. L., Rocha, J. C., Padilha, C. C. F. & Padilha, P. M. 2004.
Determination of Copper, Iron, Nickel and Zinc in fuel kerosene by FAAS after
adsorption and pre-concentration on 2-aminothiazole-modified silica gel. Ecl. Quím.,
São Paulo, 29, 33-4.
3. Wang, T., Jia, X. & Wu, J. 2003. Direct determination of metals in organics by
inductively coupled plasma atomic emission spectrometry in aqueous matrices. Journal
of Pharmaceutical and Biomedical Analysis, 33, 639-646.
210
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
4. Pereira, J. S. F., Moraes, D. P., Antes, F. G., Diehl, L. O., Santos, M. F. P., Guimarães,
R. C. L., Fonseca, T. C. O., Dressler, V. L. & Flores, É. M. M. 2010. Determination of
metals and metalloids in light and heavy crude oil by ICP-MS after digestion by
microwave-induced combustion. Microchemical Journal, 96, 4-11.
5. Ekanem, E. J., Lori, J. A. & Thomas, S. A. 1997. The determination of wear metals in
used lubricating oils by flame atomic absorption spectrometry using sulphanilic acid as
ashing agent. Talanta, 44, 2103-2108.
6. Santos, D. S. S., Korn, M. G. A., Guida, M. A. B., Dos Santos, G. L., Lemos, V. A. &
Teixeira, L. S. G. 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by
Sequential Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase
Extraction. Journal of Brazzilian Chemical Society, 22, 552-557.
7. Teixeira, L. S. G., Bezerra, M. D. A., Lemos, V. A., Santos, H. C. D., De Jesus, D. S. &
Costa, A. C. S. 2005. Determination of Copper, Iron, Nickel, and Zinc in Ethanol Fuel
by Flame Atomic Absorption Spectrometry Using On-Line Preconcentration System.
Separation Science and Technology, 40, 2555-2565.
8. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A.
M. & Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in
automotive gasoline by X-ray fluorescence after pre-concentration on cellulose paper.
Talanta, 72, 1073-1076.
9. Roldan, P. S., Alcântara, I. L., Castro, G. R., Rocha, J. C., Padilha, C. C. F. & Padilha,
P. M. 2003. Determination of Cu, Ni, and Zn in fuel ethanol by FAAS after enrichment
in column packed with 2-aminothiazole-modified silica gel. Analytical and
Bioanalytical Chemistry, 375, 574-577.
10. Alves, V. N., Mosquetta, R., Coelho, N. M. M., Bianchin, J. N., Di Pietro Roux, K. C.,
Martendal, E. & Carasek, E. 2010. Determination of cadmium in alcohol fuel using
Moringa oleifera seeds as a biosorbent in an on-line system coupled to FAAS. Talanta,
80, 1133-1138.
11. Roldan, P. S., Alcântara, I. L., Padilha, C. C. F. & Padilha, P. M. 2005. Determination
of copper, iron, nickel and zinc in gasoline by FAAS after sorption and preconcentration
on silica modified with 2-aminotiazole groups. Fuel, 84, 305-309.
12. Yin, J., Jiang, Z., Chang, G. & Hu, B. 2005. Simultaneous on-line preconcentration and
determination of trace metals in environmental samples by flow injection combined
with inductively coupled plasma mass spectrometry using a nanometer-sized alumina
packed micro-column. Analytica Chimica Acta, 540, 333-339.
13. Abollino, O., Aceto, M., Sarzanini, C. & Mentasti, E. 2000. The retention of metal
species by different solid sorbents: Mechanisms for heavy metal speciation by
sequential three column uptake. Analytica Chimica Acta, 411, 223-237.
14. Hu, W. Hu B. & Jiang, Z. 2006. On-line preconcentration and separation of Co, Ni and
Cd via capillary microextraction on ordered mesoporous alumina coating and
211
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
determination by inductively plasma mass spectrometry (ICP-MS). Analytica Chimica
Acta, 572, 55-62.
15. Zheng, F.-Y., Li, S.-X., Lin, L.-X. & Cheng, L.-Q. 2009. Simple and rapid
spectrophotometric determination of trace titanium (IV) enriched by nanometer size
zirconium dioxide in natural water. Journal of Hazardous Materials, 172, 618-622.
16. Liu, Y., Liang, P. & Guo, L. 2005. Nanometer titanium dioxide immobilized on silica
gel as sorbent for preconcentration of metal ions prior to their determination by
inductively coupled plasma atomic emission spectrometry. Talanta, 68, 25-30.
17. Hua, M., Zhang, S., Pan, B., Zhang, W., Lv, L. & Zhang, Q. 2012. Heavy metal
removal from water/wastewater by nanosized metal oxides: A review. Journal of
Hazardous Materials, 211–212, 317-331.
18. Chang, G., Jiang, Z., Peng, T. & Hu, B. 2003. Preparation of High-Specific-SurfaceArea Nanometer-sized Alumina by Sol-Gel Method and Study on Adsorption Behaviors
of Transition Metal Ions on the Alumina Powder with ICP-AES. Acta Chimica Sinica,
61, 100-103.
19. Tuzen, M., Soylak, M., Citak, D., Ferreira, H. S., Korn, M. G. A. & Bezerra, M. A.
2009. A preconcentration system for determination of copper and nickel in water and
food samples employing flame atomic absorption spectrometry. Journal of Hazardous
Materials, 162, 1041-1045
20. Ferreira, S. L. C., Queiroz, A. S., Fernandes, M. S. & dos Santos, H. C. 2002.
Application of factorial designs and Doehlert matrix in optimization of experimental
variables associated with the preconcentration and determination of vanadium and
copper in seawater by inductively coupled plasma optical emission spectrometry.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1939-1950.
21. Amais, R. S., Ribeiro, J. S., Segatelli, M.G., Yoshida, I. V. P., Luccas, P. O. & Tarley,
C.R.T. 2007. Assessment of nanocomposite alumina supported on multi-wall carbon
nanotubes as sorbent for on-line nickel preconcentration in water samples. Separation
and Purification Technology, 58, 122-128.
22. Cerutti, S., Salonia, J. A., Ferreira, S. L. C., Olsina, R. A. & Martinez, L. D. 2004.
Factorial design for multivariate optimization of an on-line preconcentration system for
platinum determination by ultrasonic nebulization coupled to inductively coupled
plasma optical emission spectrometry. Talanta, 63, 1077–1082.
23. Soylak, M., Narin, I., Bezerra, M. D. A. & Ferreira, S. L. C. 2005. Factorial design in
the optimization of preconcentration procedure for lead determination by FAAS.
Talanta, 65, 895-899.
24. dos Santos, W. N. L., Dias, F. D. S., Fernandes, M. S., Reboucas, M. V., Vale, M. G.
R. Welz, B. Ferreira, S. L. C. 2005. Application of multivariate technique in method
development for the direct determination of copper in petroleum condensate using
graphite furnace atomic absorption spectrometry. Journal of Analytical Atomic
Spectrometry, 20, 127–129.
212
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
25. Soylak, M., Tuzen, M., Souza, A. S., Korn, M. D. G. A. & Ferreira, S. L. C. 2007.
Optimization of microwave assisted digestion procedure for the determination of zinc,
copper and nickel in tea samples employing flame atomic absorption spectrometry.
Journal of Hazardous Materials, 149, 264-268.
26. Bezerra, M. A., Bruns, R. E. & Ferreira, S. L. C. 2006. Statistical design-principal
component analysis optimization of amultiple response procedure using cloud point
extraction and simultaneous determination of metals by ICP OES. Analytica Chimica
Acta, 580, 251–257.
27. Rogojan, R., Andronescu, E. Ghitulica, C. & Vasile, B.S. 2011. Synthesis and
characterization of Alumina nano-powder obtained by sol-gel method. U.P.B. Scientific
Bulletin Series B, 73, 65-76.
28. Ozcan, M. & Akman, S. 2005. Determination of Cu, Co and Pb in gasoline by
electrothermal atomic absorption spectrometry using aqueous standard addition in
gasoline–ethanol–water three-component system. Spectrochimica Acta Part B: Atomic
Spectroscopy, 60, 399-402.
29. Kowalewska, Z., Ruszczyńska, A. & Bulska, E. 2005. Cu determination in crude oil
distillation products by atomic absorption and inductively coupled plasma mass
spectrometry after analyte transfer to aqueous solution. Spectrochimica Acta Part B:
Atomic Spectroscopy, 60, 351-359.
30. Shek, C. H., Lai, J. K. L., Gu, T. S. & Lin, G.M. 1997. Transformation evolution and
infrared absorption spectra of amorphous and crystalline nano–Al2O3 powders,
Nanostructured Materials, 8, 605–610.
31. Saleh, T. A. & Gupta, V. K. 2012. Synthesis and characterization of alumina nanoparticles polyamide membrane with enhanced flux rejection performance. Separation
and Purification Technology, 89, 245-251.
32. Paulhiac, J. L. & Clause, O. 1993. Surface coprecipitation of cobalt(II), nickel(II), or
zinc(II) with aluminum(III) ions during impregnation of .gamma.-alumina at neutral pH.
Journal of the American Chemical Society, 115, 11602-11603.
33. Cui, C., He, M. & Hu, B. 2011. Membrane solid phase microextraction with alumina
hollow fiber on line coupled with ICP OES for the determination of trace copper,
manganese and nickel in environmental water samples. Journal of Hazardous
Materials, 187, 379-385.
34. Shishehbore, M., Afkhami, A. & Bagheri, H. 2011. Salicylic acid functionalized silicacoated magnetite nanoparticles for solid phase extraction and preconcentration of some
heavy metal ions from various real samples. Chemistry Central Journal, 5, 1-10.
35. Aydin, F. A. & Soylak, M. 2010. Separation, preconcentration and inductively coupled
plasma-mass spectrometric (ICP-MS) determination of thorium(IV), titanium(IV),
iron(III), lead(II) and chromium(III) on 2-nitroso-1-naphthol impregnated MCI GEL
CHP20P resin. Journal of Hazardous Materials, 173, 669-674.
213
Chapter ten:
Optimization of simultaneous preconcentration of trace metal ions
36. Qu, R., Sun, C., Ma, F., Cui, Z., Zhang, Y., Sun, X., Ji, C., Wang, C. & Yin, P. 2012.
Adsorption kinetics and equilibrium of copper from ethanol fuel on silica-gel
functionalized with amino-terminated dendrimer-like polyamidoamine polymers. Fuel,
92, 204-210.
37. IUPAC. 1978. Analytical Chemistry Division, Nomenclature, symbols, units and their
usage in spectrochemical analysis-II. Data interpretation analytical chemistry division.
Spectrochimica Acta B: Atomic Spectroscopy 33 241–245.
38. Campos, R. C., Dos Santos, H. R. & Grinberg, P. 2002. Determination of copper, iron,
lead and nickel in gasoline by electrothermal atomic absorption spectrometry using
three-component solutions. Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1528.
39. Brandão, G. P., De Campos, R. C., De Castro, E. V. R. & De Jesus, H. C. 2008.
Determination of manganese in diesel, gasoline and naphtha by graphite furnace
atomic absorption spectrometry using microemulsion medium for sample stabilization.
Spectrochimica Acta Part B: Atomic Spectroscopy, 63, 880-884.
40. Donati, G. L., Amais, R. S., Schiavo, D. & Nobrega, J. A. 2013. Determination of Cr,
Ni, Pb and V in gasoline and ethanol fuel by microwave plasma optical emission
spectrometry. Journal of Analytical Atomic Spectrometry, 28, 755-759.
41. Cunha, F. A. S., Sousa, R. A., Harding, D. P., Cadore, S., Almeida, L. F. & Araújo, M.
C. U. 2012. Automatic microemulsion preparation for metals determination in fuel
samples using a flow-batch analyzer and graphite furnace atomic absorption
spectrometry. Analytica Chimica Acta, 727, 34-40.
214
CHAPTER ELEVEN:
DEVELOPMENT AND MULTIVARIATE OPTIMIZATION OF AN OFFLINE
HOLLOW FIBER SOLID PHASE MICROEXTRACTION SYSTEM FOR
PRECONCENTRATION OF TRACE METAL IONS IN FUEL SAMPLES PRIOR TO
THEIR ICP-MS DETERMINATION
ABSTRACT
A simple and efficient hollow fiber-solid phase microextraction (HF–SPME) method using
hollow fiber-supported sol-gel combined with cation exchange resin was developed for the
preconcentration of Cd, Cu, Fe, Pb and Zn in diesel and gasoline samples. The cationic
exchanger used in this study was Dowex 50-x8 resin. The optimization of HF-SPME procedure
was carried out using two-level full factorial and central composite designs. Four factors
(variables), including sample solution pH, acceptor phase amount, extraction time and eluent
concentration were considered as factors in the optimization. The four factor and two-level full
factorial design (24 with 19 runs) results, based on the analysis of variance (ANOVA),
demonstrated that acceptor phase amount, eluent concentration and extraction time led to a more
significant improvement of the analytical response at 95% confidence level. Central composite
design was then applied in order to determine the optimum conditions for metal ion
preconcentration. Under optimized experimental conditions, the precision was ≤ 3% (C = 10 µg
L-1, n =12), limits of detection and quantification ranged from 0.08-0.28 µg L-1 and 0.28-0.93 µg
L-1, respectively, and the maximum preconcentration factor was 30.
Keywords: HF-SPME, multivariate optimization, preconcentration, trace metals, ICP-MS, liquid
fuels
11.1 INTRODUCTION
The presence of metal ions in liquid fuels is undesirable, not only because of the possibility
of damaging vehicle parts, catalytic poisoning and poor fuel performance, but also because of the
pollution caused by the release of toxic metals into the atmosphere during fuel combustion.1,2
Depending on the concentration of metal ions in liquid fuels, poor engine performance and
increased levels of pollution can be observed.3 In addition, considering the large number of
215
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
vehicles circulating in metropolitan areas, toxic metal ions such as Pb, has a potential of
presenting severe public health issues.3-5 Taking into consideration the aforementioned aspects,
accurate determination and knowledge of metal ions in liquid fuels is necessary to guarantee the
quality of the product, good performance of vehicle engines and reduced environmental
pollution.6
The concentrations of metal ions in liquid fuel samples such as diesel and gasoline are
usually at trace levels. For this reason, sensitive techniques or pre-concentration for their
determination are required.7 Direct determination of metals in diesel and gasoline, by most
analytical techniques is difficult.8 This is because of their volatility, low viscosity, corrosivity
and immiscibility with water.8,9 Despite this challenge, procedures based on direct determination
using electrothermal atomic absorption spectrometry (ETAAS) are reported in literature.10,11 This
is because of the high sensitivity and tolerance to high organic matrix loads.3 In addition, the use
of emulsion or microemulsion procedures combined with ETAAS has recently become attractive
alternative sample preparation methods. This is due to the fact that these procedures had short
sample preparation time and low risk of analyte losses by volatilization or sorption.12 However,
the disadvantage of these procedures is the stability of microemulsion between the fuel oils and
surfactant.13 For instance, it has been reported that microemulsion can be stable for 20 to 60
minutes.14,15 In addition, the routine analysis using ETAAS is disadvantageous because of its low
sample throughput compared to inductively coupled plasma-based methods.3
Inductively coupled plasma mass spectrometry (ICP-MS) is a well-known spectrometric
technique that is used in routine analysis of metal ions in different sample matrices such as
environmental, food, biological and fuel samples, among others. The main advantages of using
ICP-MS include multielement capability, low detection limits, sensitivity, wide linear range and
high sample throughput. However, the direct introduction of liquid fuels into the plasma requires
special care, as the organic load may de-stabilize or extinguish the plasma.7,9,16 Therefore, a
sample preparation step that will extract trace metals in diesel and gasoline samples prior to ICP–
MS determination is required. One sample preparation method that is commonly coupled to ICPMS is electrothermal vaporization (ETV).6,7,17,18 The latter uses a temperature program for
volatilization of matrix components before introducing the analyte into the plasma. An
electrothermal vaporization technique minimizes carbon formation on components of the
216
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
equipment and reduces interference due to polyatomic species.7 However, its parameters have to
be optimized for each element thus lengthening the experimental procedure.19 Therefore, an
accurate and reliable analytical procedure based on simultaneous separation and preconcentration
of analytes prior to analysis in fuel samples, is required.
Solid phase microextraction (SPME) is an ideal alternative preconcentration technique that
can be used to eliminate organic matrices prior to ICP-MS determination. This technique was
introduced in 1990 by Arthur and Pawliszyn to tackle the need to ease rapid sample preparation
both in the laboratory and on-site.20 The benefits of using SPME technique include short sample
preparation times, small sample volumes, analyte preconcentration from liquid, gaseous and
solid samples and easy automation to allow high-throughput analysis.20 However, the main
limitation of this technique is related to polymeric extractant phase and the desorption process.21
This limitation can be avoided by the use of a hollow fiber membrane. Hollow fiber-SPME
involves the use of a membrane as the adsorbing material that integrates sampling, extraction and
preconcentration into a single step. Additionally, it inherits the advantages of SPME and
membrane separation.22,23
This study explored the use of hollow fiber solid phase microextraction (HF-SPME)
technique using a hollow fiber-supported sol-gel combined with cation exchange resin (Dowex
50W-x8 for extraction and preconcentration of trace elements (Cd, Cu, Fe, Pb and Zn) in diesel
oil and gasoline samples. The combination of sol-gel and cation exchange resin provides a
versatile preconcentration method for the metal ion analysis. The advantage of incorporating a
cation exchange inside the hollow fiber membrane includes the possibility of fiber regeneration
and high preconcentration factor. The HF-SPME system was carried out in a batch mode.
However, the latter involves time-consuming steps. To overcome this problem, multivariate
techniques were used for optimization of factors influencing preconcentration conditions for
metal ion determination. This is because multivariate techniques are faster, more economical and
effective than the traditional univariate method. In addition, multivariate optimization makes it
possible to understand interactions between experimental variables that are not explained by the
traditional approach.24 In this work, two-level full factorial and central composite designs were
used for optimization of the factors affecting the preconcentration system. To the best of our
knowledge, the assessment of hollow fiber-supported sol-gel combined with cation exchange
217
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
resin for preconcentration of metal ions in diesel and gasoline, has not been reported in the
literature.
11.2 EXPERIMENTAL
11.2.1 Instrumentation
A Perkin-Elmer Sciex ELAN 6000 (Perkin-Elmer SCIEX Instruments, Concord, Canada)
inductively coupled plasma mass spectrometer (ICP-MS) was used for all measurements. The
ICP-MS instrument was optimized daily and operated as recommended by the manufacturer. The
operating conditions are presented in Table 11.1. Argon of 99.996% purity (Afrox, South Africa)
was used. The Accurel S6/2 polypropylene hollow fiber membrane used here was obtained from
Membrana (Wuppertal, Germany). The wall thickness of the fiber was 450 μm, the inner
diameter was 1800 µm, and the pore size was 0.2 µm.
Table 11.1. Operational ICP-MS parameters
RF power
Gas flow rates
Outer
Intermediate
Carrier
Resolution
Sweeps per reading
Dwell time
Readings per replicate
Replicates
Auto lens
Isotopes
a
1100
15 L min-1
1.2 L min-1
0.95 L min-1
0.7 a.m.u. (10% of the peak height)
1
25 ms
100
3
On
111
Cd, 63Cu, 56Fe, 208Pb, 66Zn 45Sc (ISa), 72Ge
(IS), 115In (IS), 209Bi (IS)
IS = Internal standard
For comparative method, metal ions (Cd, Cu, Fe, Pb and Zn) were determined using a
Spectro Arcos 165 ICP OES (SPECTRO Analytical Instruments, GmbH, Germany) equipped
with Cetac ASX-520 autosampler. The operating conditions on the ICP OES spectrometer during
218
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
the measurements were as follows: forward power: 1400 W, plasma argon flow rate: 13 L min-1,
auxiliary argon flow rate: 2.00 L min-1, nebulizer argon flow rare: 0.95 L min-1. The most
prominent atomic and ionic analytical lines of metal ions were selected for investigation, that is,
Cd 214.438 nm, Cu 324.754 nm, Fe 238.204 nm, Pb 220.353 nm and Zn 213.856. The
microwave digestions were carried out in an Ethos D (Milestone, Sorisole, Italy) with maximum
pressure 1450 psi and maximum temperature 300◦C.
11.2.2 Reagents, Solutions and Real Samples
All reagents were of analytical grade unless otherwise stated and double distilled deionised
water (Millipore, Bedford, MA, USA) was used throughout the experiments. The sol–gel
precursor tetraethylorthosilicate (TEOS), absolute ethanol (EtOH), 2-amino-2-hydroxynmethylpropane-1,3-diol (TRIS) and ammonium hydroxide were obtained from Sigma-Aldrich (St.
Loius, MO, USA). Dowex 50w-x8 cation exchange resins (sodium form) with a mesh size of
200-400, was obtained from Sigma-Aldrich (St. Loius, MO, USA). Synthetic gasoline was
prepared by mixing 91% isooctane and 9% n-heptane (St. Loius, MO, USA). Spectrascan stock
solutions (1000 mg L-1) of Cd, Cu, Fe, Pb and Zn (Industrial Analytical (Pty) Ltd, Johannesburg,
South Africa) were used to prepare the working solutions for HF-LPME at concentrations of 15
µg L-1 for each metal ion. Working solutions (prepared in organic phase), as per the experimental
requirements, were freshly prepared from the stock solution for each experimental run. A
Spectrascan multi-element standard solution at a concentration of 100 mg L-1 (Industrial
Analytical (Pty) Ltd, Johannesburg, South Africa) was used to prepare working standard
solutions for measurements of concentrations of analytes in model and sample solutions.
11.2.3 Preparation of Sol-gel
The sol–gel solution was prepared by a method proposed by Es’haghi et al.25 Describing the
procedure briefly, 640 µL of TEOS, 130 µL of TRIS aqueous solution (5%) as base catalyst and
500 µL of EtOH were added into a Teflon beaker. The mixture stirred and heated at 50°C for 2–3
h until turbidity appeared. Thereafter, 20 µL of concentrated ammonium hydroxide was added
into the mixture. The latter was then centrifuged at 1200 rpm for 5 min. The sol (top clear liquid
solution) was removed and the gel (white precipitate at the bottom) was washed twice, in
219
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
succession, with deionized water and once with ethanol to remove the un-reacted reactant and
excess catalyst. The resulting gel was transferred to a clean vial and dispersed in 1 mL 1octanol/ethanol (1:1 v/v) mixture and then used for metal extraction study.
11.2.4 Extraction and Pre-concentration Procedure
An appropriate amount of Dowex 50w-x8 resin was added to the gel which was previously
dispersed 1 mL 1-octanol/ethanol (1:1 v/v) mixture. Before, the extraction and preconcentration
procedure for metal ions in fuel samples, the hollow fiber (450 μm, the inner diameter was 1800
µm, and the pore size was 0.2 µm) was cut into segments with a length of 4 cm and one end of
the fiber was sealed using a hot plate. The fiber segments were cleaned with acetone to remove
impurities and directly dried in air. Then, the fiber was submerged in the 1-octanol for a few
seconds to fill the membrane pores of the hollow fiber wall. After that, appropriate
concentrations of the acceptor phase (sol–gel/Dowex 50W-x8) were injected into the lumen of
the hollow fiber with a pasture pipette. The fiber surface was washed with water to remove
excess organic solvent. Then the other end the hollow fiber was sealed to prevent the leaking of
the acceptor phase.
The procedure for the preparation of gasoline–ethanol–water mixture was carried out
according to Ozcan and Akman.26 A 10 ml aliquot of synthetic gasoline sample was placed in a
100 ml polyethylene volumetric flask followed by the addition of 5 ml of concentrated HNO3
and 10 ml of water. The mixture was spiked with 1.5 mL of a 1.0 mg L-1 multi-element oil
standard solution and made up to the mark with ethanol to obtain 15 µg L-1 concentration of each
metal ion. The mixture was homogenized by shaking and a single phase solution was obtained. It
should be noted that the stability of the gasoline-ethanol-water mixture was not monitored. This
is because the resulting mixture was subjected the preconcentration system immediately after
homogenization. The hollow fiber was placed into the sample solution present in plastic bottles
containing 50 mL of the above mentioned model solutions. The bottles were covered and shaken
at 500 rpm. During this procedure, the analytes from the sample solution diffuses through the
porous polypropylene membrane into the acceptor phase.25 At the end of the extraction, the
hollow fiber was taken out from the vial and rinsed with double distilled deionised water then
transferred into a polypropylene centrifuge vial containing 5.0 mL of eluent solution. The
220
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
analytes were eluted from the fiber with ultrasonic agitation for 5 min. The same procedure was
applied to the blank solutions. In the case of analysis of real samples, an aliquot of 2.5 mL of
gasoline sample was placed in a 250 mL polypropylene volumetric flask followed by the
addition of 1.25 mL of concentrated HNO3 and 25 mL double distilled deionised water,
respectively. The mixture was then diluted to the mark with ethanol. Between experiments, the
hollow fibers were washed with double distilled deionised water and stored in 1.0 M NaOH
solution (this was done in order to keep the resin in sodium form) for the next experiment.
11.2.5 Optimization Strategy
The optimization of the HF-SPME was carried out considering four variables, namely,
sample pH, eluent concentration (EC), acceptor phase amount (APA) and extraction time (ET).
In all experiments, 50 mL of metal ion solutions with a final concentration of 15 μg L−1 were
used. The optimization was carried out by using the multivariate strategy. Firstly, a screening of
the influential variables on the analytical response was tested by employing a two-level (24) full
factorial design with three central points. The factors and their levels are presented in Table 10.2.
The second step of the optimization strategy involved the application of a RSM based on a
central composite design. The latter was applied in the optimization of the level of the variables
that were considered as significant according to the results obtained in 24 full factorial design.
The design of experiments was performed using Minitab 15 and Design Expert 8.0.7.1 Software
programs.
Table 11.2. Factors and levels used in 24 factorial design for separation and preconcentration of
metal ions in fuel samples
Variable
pH
EC (mol L-1)
APA (mg mL-1)
ET (min)
Low level (-1)
4.0
1.0
50
10
221
Central point (0)
6
2.5
100
25
High level (+1)
8
4.0
150
40
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
11.2.6 Comparative Procedure
The Microwave acid digestion procedure was carried out according to Kowalewska et al.27
Briefly, 5.0 ml of the gasoline sample was placed into a Teflon vessel followed by 6 mL HNO3
(65%) and 2.0 mL H2O2 (30%). The vessels were inserted into a microwave unit and heated
according to the conditions recommended by the manufacturer. The digested content was left to
cool down to room temperature. After cooling, the vessels were opened and 2 ml of concentrated
HNO3 and 2 ml of hydrogen peroxide were added, and the heating program was repeated.
Finally, the Teflon vessel contents were cooled down to room temperature and quantitatively
transferred to a 50 mL calibration flask, 1 mL of concentrated nitric acid was added and the
samples were spiked with 20 µg L-1 of the target analytes. The flask was then made up to the
mark with double distilled deionised water. The latter water was submitted to the same procedure
and used as a blank. The samples were then analyzed with ICP OES.
11.3 RESULTS AND DISCUSSION
We note that the ICP-MS instrument used in this study did not have collision/reaction cell.
Therefore, in all analyses, mathematical equations were incorporated to correct potential spectral
interferences such as 40Ar16O on 56Fe, and 36Ar2 on 72Ge. In addition, the use of pre-concentration
procedure prior to metal ion determination increased the concentration levels of the metal ions in
the final solution. Furthermore, model solutions were also used to monitor the performance of
the instruments. According to the results obtained there was no sign of interferences and the
concentration levels for blank solutions were low, as expected.
11.3.1 Preliminary Optimization Using Two Level Full Factorial Design
The optimization of the HF-SPME preconcentration system was performed using a 24 full
factorial design with three replicates of the center point, totalizing to 19 experiments. Table 11.3
shows the factorial design matrix and the results derived from each run for cadmium, copper,
iron, lead and zinc, respectively. The effect of factors in the HF-SPME system was investigated
by using analysis of variance (ANOVA) considering the recovery as the analytical response. The
information obtained from ANOVA, generally main effects and their interactions, is represented
222
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
in a Pareto Chart (Fig. 11.1). The length of each bar in the chart is proportional to the absolute
value of its estimated effect. When the bar length exceeds the vertical reference line (p = 0.05) it
implies that the effect of the variable or interaction is significant.28
Table 11.3. Matrix of 24 full factorial design and the analytical response (% recovery) for each
experiment for extraction and preconcentration of metal ions
Experiments pH
APA
ET
EC
Cd
Cu
Fe
Pb
Recovery (%)
58.6
43.2
61.2
47.0
65.8
60.0
67.6
62.5
71.8
52.9
73.3
60.8
83.4
72.9
85.1
79.87
76.5
60.9
77.0
63.0
84.4
69.0
89.0
70.8
82.7
72.3
85.5
74.1
96.0
95.0
98.1
95.7
82.2
71.1
82.1
71.1
82.1
71.2
-1
-1
-1
-1
39.5
53.6
1
1
-1
-1
-1
40.2
64.8
2
-1
1
-1
-1
53.6
75.6
3
1
1
-1
-1
54.1
84.5
4
-1
-1
1
-1
59.6
67.8
5
1
-1
1
-1
60.6
75.3
6
-1
1
1
-1
70.0
84.0
7
1
1
1
-1
69.9
90.4
8
-1
-1
-1
1
65.3
64.2
9
1
-1
-1
1
66.3
68.6
10
-1
1
-1
1
73.7
82.4
11
1
1
-1
1
74.7
89.3
12
-1
-1
1
1
85.3
90.3
13
1
-1
1
1
86.0
93.1
14
-1
1
1
1
94.4
93.8
15
1
1
1
1
95.2
95.0
16
0
0
0
0
70.7
81.5
17
0
0
0
0
70.5
81.7
18
0
0
0
0
70.5
81.7
19
APA= Acceptor phase amount; ET= extraction time; EC = eluent concentration
Zn
42.9
45.9
58.8
59.9
61.8
63.7
73.3
79.5
55.4
57.7
86.9
89.9
73.4
75.0
93.3
94.9
71.6
71.5
71.7
Fig. 11.1 A shows that eluent concentration (24.31), extraction time (19.06) and acceptor
phase amount (10.20) are statistically significant at 95% confidence level. Sample pH and the
main effect interaction were not significant at 95% confident level. For preconcentration of Cu
(Fig. 11.1 B), acceptor phase amount (14.71), extraction time (13.24), eluent concentration
(10.19) and sample pH (6.24) were statistically significant at the 95% confidence level.
Moreover, interactions between acceptor phase amount and extraction time (-5.46), acceptor
223
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
phase amount and eluent concentration (-3.68) as well as between extraction time and eluent
concentration (3.62) are also significant. In the case of Fe extraction (Fig. 11.1 C), eluent
concentration (24.31), extraction time (19.06) and acceptor phase amount 10.20) as well as
interaction between extraction time and eluent concentration were statistically significant at the
95% confidence level. Fig. 11.1 D shows that ET (16.31), APA (16.01), EC (15.53), sample pH
(3.85) and interaction between APA and ET (4.74) are significant at 95% confidence level.
Lastly, for preconcentration of Zn (Fig 11.1 E), APA (19.59), EC (17.10) and ET (14.21) were
statistically significant at the 95% confidence level. In addition, interactions between APA and
EC (6.28) are also significant.
The positive values signify that increasing the factors from minimum to maximum will lead
to the increase in the analytical response (% recovery). In contrast, the negative values indicate
an enhancement in the analytical response when the level changes from maximum to minimum
.29 The ANOVA results demonstrated that for extraction of Cd, Fe and Zn, the sample pH was
not significant (at 95% confidence level). Although sample pH was significant for the
preconcentration of Cu and Pb, considering the effect value of the sample pH compared to other
main effects, one can conclude that acceptor phase volume, eluent concentration and extraction
time lead to a more pronounced improvement of the analytical response. Therefore, the sample
pH was fixed at 7.0. The overall results obtained for the full factorial design showed that the
variables such as APA, ET and EC required a final optimization. Therefore, these variables were
optimized using a central composite design. The latter is associated with the response surface
method and is appropriate for the location of an optimum set of experimental conditions from a
very good fitting of a quadratic model.30
224
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
Fig. 11.1. Pareto chart of standardized effects for variables related to the preconcentration of (A)
cadmium, (B) copper, (C) iron, (D) lead and (E) zinc
11.3.2 Final optimization using a Central Composite Design
In view of the fact that the significant factors were identified, central composite design was
used for the final optimization of the HF-SPME method. A central composite design matrix
containing a total of 20 experiments and response based on each of the experimental runs, are
shown in Table 11.4. The observed percentage recoveries of each metal ion ranged from 45.01-
225
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
98.10%, 51.23-99.12%, 38.18-98.60%, 4.75-98.13% and 42.28-99.01% for Cd, Cu, Fe, Pb and
Zn, respectively.
Table 11.4. List of experiments in the central composite design (actual values) for HF-SPME
optimization and the responses
Experiment APA
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
50
150
50
150
50
150
50
150
100
100
100
100
15.9
184.1
100
100
100
100
100
100
ET
15
15
40
40
15
15
40
40
27.5
27.5
27.5
27.5
27.5
27.5
6.5
48.5
27.5
27.5
27.5
27.5
EC
Cd
Cu
1.5
1.5
1.5
1.5
4
4
4
4
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
0.65
4.85
2.75
2.75
Recovery (%)
48.23
53.91
70.77
70.14
66.10
59.97
97.87
92.91
52.18
54.40
71.80
69.75
67.58
70.13
96.18
99.12
88.09
87.02
87.96
87.09
88.00
86.95
87.99
87.12
45.01
51.23
97.16
96.13
57.11
63.01
98.10
96.12
74.41
63.30
75.88
71.93
88.01
86.97
88.12
86.98
Fe
Pb
Zn
44.01
90.22
69.43
94.48
41.98
72.75
59.94
74.88
84.25
84.18
84.37
84.29
38.18
95.89
61.28
98.56
75.18
58.41
83.89
84.13
47.38
73.87
54.63
98.13
60.76
69.02
68.03
86.86
82.69
82.65
82.70
82.71
43.75
93.75
72.91
95.82
62.77
75.81
82.68
82.74
50.13
85.48
66.01
91.65
54.77
82.44
74.93
96.37
86.21
86.55
86.36
86.75
42.26
91.91
75.52
99.01
64.45
73.81
86.52
86.75
11.3.2.1 Analysis of Variance
The analysis of variance (ANOVA) parameters of the predicted response surface quadratic
model for the recoveries of Cd, Cu, Fe, Pb and Zn, were obtained. It should be noted that the
ANOVA results are not included for simplicity purposes. However, they can be viewed in the
Supplementary data. The p-values less than 0.05 indicated that the model terms are significant at
95% confidence level, whereas values greater than 0.05 indicate that the model terms are not
226
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
significant.31,32 The model F-values and p-values were less than 0.05; this demonstrated that the
model was significant for extraction and preconcentration of metal ions in gasoline model
solutions. The ANOVA results (Supplementary data) showed the lack of fit for F-values was
statistically significant at 95% confidence level. Bashir et al.31 reported that a significant lack of
fit suggests that there may be some systematic variation unaccounted for in the hypothesized
models. This might be due to the precise replicate values of the independent variable in the
model that provide an estimate of pure error.31
According to Joglekar and May,33 the minimum correlation coefficient for a good fit model
should be higher than 0.80. The correlation coefficients (0.9824-0.9993) obtained in the present
study for preconcentration of Cd, Cu, Fe, Pb and Zn were higher than 0.80. High correlation
coefficients values demonstrate a relatively good agreement between the predicted and observed
results within the range of experimental runs.31
The data obtained (Table 11.4) were analysed using Design Expert 8.0.7.1 software and
resulted in Eq. (1)-(5) for the models to illustrate the dependence of the analytical response (%
recovery) with respect to the evaluated variables, that is, acceptor phase amount, extraction time
and eluent concentration. The final regression models were expressed in terms of actual factors.
% Re cov ery(Cd )   32.57  0.72 A  2.18 B  20.94 C  4.19  10 3 AB
 1.76  10  2 AC  5.07  10 2 BC  2.60  10 3 A 2
(1)
 2.85  10 2 B 2  3.19 C 2
% Re cov ery(Cu )   8.85  0.51 A  0.85 B  24.58 C  6.07  10 3 AB
 9.66  10 3 AC  0.13 BC  2.01  10 3 A 2
(2)
 1.91  10 2 B 2  4.61C 2
% Re cov ery( Fe)   61.81  1.18 A  2.29 B  26.98 C  6.60  10 3 AB
5.91  10 2 AC  0.11 BC  2.56  10 3 A 2
2
 1.18  10 B  4.15 C
2
(3)
2
% Re cov ery( Pb )   27.52  0.81A  0.23 B  31.53 C  5.52  10 3 AB
8.58  10 2 AC  5.12  10 2 BC  2..31  10 3 A 2
3
 1.59  10 B  3.57 C
2
2
227
(4)
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
% Re cov ery( Zn)   21.53  0.95 A  0.42 B  22.19 C  3.19  10 3 AB
2.38  10 2 AC  9.63  10  2 BC  2.56  10 3 A 2
(5)
 3.49  10 3 B 2  3.75 C 2
In Eqs (1)-(5), A, B and C correspond to independent variables of aceptor phase amount,
extraction time and eluent concentration, respectively, while the terms AB, BC and AC
correspond to the interactions of the variables.
11.3.2.2 Optimization of Experimental Conditions
The 3D response surface plots in Fig. 11.2 were used to access the interactive relationship
between independent variables and analytical response.31 It can be seen from Fig. 11.2 that, in
each plot eluent concentration was kept constant while the other two variables were varied within
the experimental ranges. The reason why eluent concentration was kept constant during the
evaluation is that the perturbation plots (data not included) for all the metal ions showed a semiflat curvature for eluent concentration compared to other independent variables. The semi-flat
curvature indicates that, the influence of eluent concentration on the analytical response was less
significant compared to acceptor phase amount and extraction time, which had a relatively
significant effect on % recovery. Fig. 11.2 shows the surface response plots of percentage
recovery versus the acceptor phase amount (APA) and extraction time (ET) at constant eluent
concentration of 2.75 mol L-1 for optimization of metal ion preconcentration using HF-SPME. It
was observed that for all the metal ions, the maximum percentage recoveries of metal ions at
APA 150 mg L-1 and ET 40 min, were 101.10%, 101.33%, 101.32%, 98.28% and 95.99% for
Cd, Cu, Fe, Pb and Zn, respectively. On the other hand, the minimum recoveries (47.51%,
51.23%, 38.15%, 43.75% and 43.28% for Cd, Cu, Fe, Pb and Zn, respectively) were obtained at
APA 50 mg mL-1 and ET 15 min. Based on results obtained from 24 full factorial and central
composite designs the optimum conditions that led to quantitative retention and elution of metal
ion were as follows: pH = 7.0, acceptor phase amount = 150 mg mL-1, extraction time = 40 min
and eluent concentration = 2.75 mol L-1.
228
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
Fig. 11.2. Response surface for percentage recovery of cadmium (A), copper (B), iron (C), lead
(D) and zinc (E) as function of acceptor phase amount (APA), mg mL-1 and extraction time (ET),
min at constant eluent concentration of 2.75 mol L-1
The total number of experiments performed in full factorial and central composite designs to
attain optimum conditions for simultaneous preconcentration of Cd, Cu, Fe, Pb and Zn using HFSPME were 39. The latter corresponds to an average of 19 experiments for full factorial design
exploring four factors and 20 experiments for CCD exploring three factors. In comparison with
the traditional univariate optimization procedure, the total number of experiments carried out in
229
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
this study was relatively low. In addition, the reduced number of experiments allows improved
data interpretation at greater speed and efficiency of the proposed analytical method. Unlike the
multivariate optimization procedure, the traditional univariate procedure requires at least 8–10
experimental points for optimization of each variable per metal ion. In terms of data
interpretation, the univariate procedure may lead to misleading results and analysis because the
effects of interactions among the variables cannot be examined.34
11.3.3 Regeneration Studies
The regeneration of the column is one of the important parameters in evaluating the stability
of the HF-SPME supported sol-gel combined with Dowex 50W-x8 resin. The recyclability of
HF-SPME method was achieved by performing adsorption and desorption repeatedly. The
stability and regeneration of HF-SPME was evaluated by monitoring changes in the recoveries of
the target metal ions through retention-elution cycles. It was observed that the HF-SPME was
stable up to 75 retention/elution cycles (15 replicates in 5 consecutive days) without obvious loss
of its analytical performance. Therefore, the recycling of hollow fiber-supported sol-gel
combined with cation exchange resin is possible.
11.3.4 Analytical Features
Under the optimum conditions, calibration curves were constructed for the preconcentration
of Cd, Cu, Fe, Pb and Zn ions according to the general procedure. Linearity was obtained
between 0.4-140, 0.1-140, 0.5-120, 0.8-100 and 0.1-120 µg L-1 for Cd, Cu, Fe, Pb and Zn,
respectively, in a 50 mL sample. The sensitivity of the HF-SPME method was defined as the
gradient (slope) of the calibration graph. The results in Table 11.5 indicated that the sensitivity
trend of the HF-SPME method was Zn >Cu > Cd > Fe > Pb. Thus the highest sensitivity
obtained was 106.8 cps L µg-1 for Zn while the lowest was 34.0 cps L µg-1 for Pb. The possibility
of enriching low concentrations of analytes from large volumes, the maximum applicable sample
volume should be investigated.8 It was observed that quantitative recoveries for all target metal
ions were stable (≥ 95%) when sample volume was 150 mL. Therefore, the highest
preconcentration factor, defined as the ratio of the sample volume loaded onto the column to the
230
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
eluent volume used for stripping of the retained metal ions, for this method was 30. Therefore,
150 ml was used for further investigations.
The limits of detection (LOD) and limits of quantification (LOQ) were calculated according
to IUPAC definition; that is,
LOD  3Sd
m and
LOQ  10Sd
m , respectively, where m and Sd
are the slope of the respective analytical curve and the standard deviation of 20 consecutive
measurements of the blank signal, respectively. It should be noted that model (synthetic)
gasoline samples (prepared as described in Section 2.2) were used as blank solutions. The LOD
and LOQ results as well as extraction efficiency are presented in Table 11.5.
The precision of the HF-SPME method, calculated as the relative standard deviation (RSD, n
= 15), in sample solutions containing 10 µg L-1 of each metal ion was in the range of 0.4-2.8%.
The time required for preconcentration of 150 mL of sample (40 min extraction and 5 min
elution) was about 45 min. It should be noted the thermostat shaker can handle up to 32 samples
at the same time. Therefore, the throughput sample was approximately 32 samples h-1.
Table 11.5. Analytical performance of the HF-SPME system for preconcentration of metal ions
obtained under optimum conditions
Analyte
Cd
Cu
Fe
Pb
Zn
Sensitivity
(cps L µg-1)
79.8
103.6
75.5
34.0
103.8
R2
LOD (µg L-1)
LOQ (µg L-1)
0.9965
0.9997
0.9988
0.9967
0.9989
0.1
0.1
0.2
0.3
0.08
0.4
0.3
0.5
0.9
0.3
Precision
(%RSD)
0.4
0.7
2.8
2.5
0.6
Recovery
(%)
101±0.1
102±0.4
99.8±0.1
98.7±0.3
99.2±0.1
11.3.5 Effect of other Metal Ions on the HF-SPME Procedure
Liquid fuel samples normally contain a number of metal ions that exist naturally in the Earth’s
crust. The interference induced by commonly coexisting ions is listed in Table 11.6.
Preconcentration of the investigated analytes (10 µg L-1) in the presence of potential interfering
ions, were investigated under the optimum conditions. The effect of interfering ions was
investigated in order to assess the applicability of the proposed HF-SPME procedure. The
231
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
tolerance limit was set as the concentration of the ion required to cause ≤5% error. If the
presence of an interfering ion caused a variation more than 5% in the recoveries of metal ions, it
was then considered as an interferent. The results are presented in Table 11.6. All studied ions
were found not to affect the retention and recoveries of the target analytes. In addition, higher
concentrations of group I and II metals can be tolerated. These results suggest that HF-SPME
method can be applied for the determination of trace levels of Cd, Cu, Fe, Pb and Zn in liquid
fuel samples that contains higher concentrations of secondary cations.
Table 11.6. Effect of potential interfering ions on the percentage recoveries of Cd, Cu, Fe, Pb
and Zn (mean % recovery ± standard deviation)
Ions
K+
Na+
Ca2+
Mg2+
Mn2+
Ni2+
Co2+
Cr3+
Al3+
Ag+
[Interfering ion] (mg
L-1)
1000
1000
1000
1000
5
5
5
2
50
2
Cd
Cu
Fe
Pb
Zn
97.1±1.3
96.5±1.5
96.0±0.9
98.5±2.5
95.1±2.4
96.0±2.2
98.2±1.1
96.0±1.6
98.3±1.7
99.4±1.7
99.5±0.6
98.9±1.2
97.9±0.9
98.3±1.1
95.7±2.5
95.9±3.1
95.3±2.7
97.6±1.7
98.5±1.2
99.4±0.5
97.3±1.6
98.3±1.9
97.9±1.3
98.4±1.7
96.6±2.1
95.9±2.6
97.7±2.2
96.6±1.4
98.4±1.8
97.4±2.2
98.3±1.2
96.9±2.1
96.8±2.5
97.0±2.7
95.6±3.1
95.3±2.8
96.0±2.6
95.5±2.3
97.7±1.2
98.9±1.0
99.4±0.5
98.1±1.1
97.3±1.6
96.9±1.9
96.6±2.1
96.8±2.3
97.8±1.5
97.2±1.8
99.4±0.4
98.8±1.1
11.3.6 Validation and Application of the HF-SPME Method
Due to the lack of CRMs for liquid fuel, the accuracy of the proposed method was verified
through addition/recovery experiments, by adding 20 µg L-1 of each metal ion to diesel and
gasoline samples. In addition, one of the main problems associated with the preconcentration and
determination of trace elements in liquid fuels is the lack of knowledge about the form of the
analyte in the sample.35,36 Therefore, the standard addition method was carried out using both
metal-organic standards and aqueous metal standards. The results obtained as the average of
three replicates of each metal ion are presented in Table 11.7. As it can be seen for this table, the
recoveries obtained using either inorganic or organic standards were greater than or equal to
232
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
95%. This means that the Cu, Fe Ni and Zn present in liquid fuel samples can be determined
through the calibration technique using either inorganic or organic standards. Moreover, as it can
be seen in Table 11.7, the method has relatively good accuracy and the recoveries were between
95% and 103%.
The HF-SPME method was applied for extraction and preconcentration of metal ions in
commercial diesel and gasoline samples collected from different filling stations in Johannesburg
(South Africa). For the comparative method (reference method), the samples were digested using
microwave-assisted digestion (MAD) method and the concentrations of the analytes were
determined using ICP OES. It should be noted that, the digested samples were spiked with a
known amount of each analyte before their ICP OES determination. The obtained results using
the proposed method are presented in Table 11.8. Generally, the total metal content trend was
1160, 857, 734.8 and 332.7 µg L-1 for D1, G1, D2 and G2, respectively. As can be seen in Table
11.8, the concentration of Cd were quite low (˂ 10 µg L-1) for all the samples. Copper
concentration in diesel (D1 and D2) and G1 samples were higher than G2 sample. Iron and Zn
concentrations on the other hand were higher in D1 and G1 samples. The highest concentration
of lead was observed in the D1 sample. The results obtained indicated that Cd, Cu, Fe Pb and Zn
in liquid fuels could be quantitatively extracted and preconcentrated using HF-SPME method
before their inductively coupled plasma mass spectrometric determination.
The accuracy of the method was also evaluated by analysis of target analytes in diesel and
gasoline samples using the reference method (MAD/ICP OES) and the results are presented in
Table 11.8. Applying the paired student t-test, the results obtained were not significantly
different at 95% confidence level. In addition to the results not being significantly different,
another advantage of the proposed method is that it minimizes the risks of incomplete
mineralization of the organic matrix and cross-contamination. Furthermore, it avoids the use of
concentrated acids and significantly reduces the laboratory waste and analysis time (sample
throughput ≈ 32 samples h-1), which is an important aspect for routine analysis work.
233
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
Table 11.7. Analytical results obtained in the analysis of spiked diesel sample. The concentration and recovery values are expressed as
the mean ± standard deviation of the three replicates
Sample
Analytes
Added (µg L-1)
D1
Cd
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
0
20
Cu
Fe
Pb
Zn
G1
Cd
Cu
Fe
Pb
Zn
234
Inorganic standard
Found (µg L-1)
Recovery (%)
3.1±0.5
22.9±0.8
99.0±1.7
361±4
381±4
99.4±2.3
468±5
488±3
98.4±2.1
153±4
173±2
98.7±3.4
175±4
195±3
101±3.4
5.2±0.6
24.9±0.5
98.7±1.3
272±4
291±4
98.3±2.3
424±4
444±4
99.1±3.0
ND
19.5±0.5
97.5±1.1
157±2
176±2
97.8±2.6
Metallo-organic standard
Found (µg L-1) Recovery (%)
3.1±0.5
22.4±1.2
96.5±2.1
361±4
379±2
97.5±1.5
468±5
488±4
99.5±2.4
153±4
172±2
95.5±0.9
175±4
195±4
98.5±1.3
5.2±0.6
24.3±1
95.5±1.8
272±4
292±3
101±3.4
423±4
443±2
95.5±1.3
ND
19.3±1.0
96.5±0.8
157±2
177±2
103±2.8
Chapter 11:
Hollow fiber solid phase microextraction of trace metal ions in fuel
Table 11.8. Determination of Cd, Cu, Fe, Pb and Zn (µg L-1) in commercial diesel (D1 and D2) and gasoline (G1 and G2) samples by
proposed HF-SPME and comparative method (n = 3, at 95% confidence level).
Techniques
Analytes
D1
D2
G1
G2
-1
HF-SPME/ICP-MS
MAD/ICP OES
Cd
Cu
Fe
Pb
Zn
Cd
Cu
Fe
Pb
Zn
3.1±0.5
361.0±3.7
468.2±5.1
152.8±3.9
174.9±1.1
2.9±0.9
358.6±4.0
470.3±6.3
151.9±4.2
175.1±1.3
235
Concentrations (µg L )
6.95±0.4
5.2±0.6
486.1±4.3
271.6±4.0
131.2±2.3
423.7±4.1
60.0±2.5
ND
50.5±1.1
156.5±1.6
7.1±0.3
4.8±1.2
484.8±4.8
269.3±3.7
129.5±3.4
425.1±4.5
60.2±3.0
ND
51.2±1.4
155.7±1.8
8.1±0.3
89.4±2.7
113.4±1.3
61.9±1.7
59.9±1.8
7.9±0.5
90.2±2.6
112.5±1.6
62.1±2.0
59.3±1.5
Chapter eleven
Hollow fiber solid phase microextraction of trace metal ions in fuel
11.4 CONCLUSION
An offline hollow fiber solid phase microextraction system based on fiber-supported
sol-gel combined with cation exchange resin for preconcentration of trace metal ions in
liquid fuel samples prior to their ICP-MS determination has been developed. The
experimental conditions were optimized using two-level factorial and central composite
designs, resulting in the use of sample pH = 7.0, acceptor phase amount = 150 mg mL-1,
extraction time = 40 min and eluent concentration = 2.75 mol L-1. The accuracy of the
proposed method was confirmed by analysis of the spiked diesel and gasoline samples. The
measured concentrations were in good agreement at 95% confidence level with the added
values. The precision, expressed as relative standard deviation, was less than or equal to
3%. The optimized method was applied for simultaneous preconcentration of the target
analytes in commercial diesel and gasoline samples. The advantage of the proposed
method over MAD/ICP OES is that it minimizes the risks of incomplete mineralization of
the organic matrix and cross-contamination. In addition, HF-SPME/ICP-MS uses dilute
acids and significantly reduces the laboratory waste and analysis time (sample throughput
≈ 32 samples h-1), which is an important aspect for routine analysis work.
11.5 REFERENCES
1. Saint'pierre, T. D., Dias, L. F., Maia, S. M. & Curtius, A. J. 2004. Determination of Cd,
Cu, Fe, Pb and Tl in gasoline as emulsion by electrothermal vaporization inductively
coupled plasma mass spectrometry with analyte addition and isotope dilution calibration
techniques. Spectrochimica Acta Part B: Atomic Spectroscopy, 59, 551-558.
2. Sousa, J. K. C., Dantas, A. N. D. S., Marques, A. L. B. & Lopes, G. S. 2008.
Experimental design applied to the development of a copper direct determination
method in gasoline samples by graphite furnace atomic absorption spectrometry. Fuel
Processing Technology, 89, 1180-1185.
3. Donati, G. L., Amais, R. S., Schiavo, D. & Nobrega, J. A. 2013. Determination of Cr,
Ni, Pb and V in gasoline and ethanol fuel by microwave plasma optical emission
spectrometry. Journal of Analytical Atomic Spectrometry, 28, 755-759.
4. Oliveira, M., Saczk, A., Okumura, L., Fernandes, A., Moraes, M. & Stradiotto, N. 2004.
Simultaneous determination of zinc, copper, lead, and cadmium in fuel ethanol by
anodic stripping voltammetry using a glassy carbon–mercury-film electrode. Analytical
and Bioanalytical Chemistry, 380, 135-140.
236
Chapter eleven: Hollow fiber solid phase microextraction of trace metal ions in fuel
5. Cunha, F. A. S., Sousa, R. A., Harding, D. P., Cadore, S., Almeida, L. F. & Araújo, M.
C. U. 2012. Automatic microemulsion preparation for metals determination in fuel
samples using a flow-batch analyzer and graphite furnace atomic absorption
spectrometry. Analytica Chimica Acta, 727, 34-40.
6. Chaves, E. S., Lepri, F. G., Silva, J. S. A., de Quadros, D. P. C., Saint’pierre, T. D. & A.
Curtius J. 2008. Determination of Co, Cu, Fe, Mn, Ni and V in diesel and biodiesel
samples by ETV-ICP-MS. Journal of Environmental Monitoring, 10, 1211-1216.
7. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
8. Nomngongo, P. N., Ngila, J. C., Kamau, J. N., Msagati, T. A. M. & Moodley, B. 2013.
Preconcentration of molybdenum, antimony and vanadium in gasolsine samples using
Dowex 1-x8 resin and their determination with inductively coupled plasma–optical
emission spectrometry. Talanta, 110, 153-159.
9. Teixeira, L. S. G., Rocha, R. B. S., Sobrinho, E. V., Guimarães, P. R. B., Pontes, L. A.
M. & Teixeira, J. S. R. 2007. Simultaneous determination of copper and iron in
automotive gasoline by X-ray fluorescence after pre-concentration on cellulose paper.
Talanta, 72, 1073-1076.
10. Reboucas, M. V., Domingos, D., Santos, A. S. O. & Sampaio, L. 2010. Determination
of trace metals in naphtha by graphite furnace atomic absorption spectrometry:
Comparison between direct injection and microemulsion pretreatment procedures. Fuel
Processing Technology, 91, 1702-1709.
11. Reboucas, M. V., Ferreira, S. L. C. & Neto, B. D. B. 2003. Arsenic determination in
naphtha by electrothermal atomic absorption spectrometry after preconcentration using
multiple injections. Journal of Analytical Atomic Spectrometry, 18, 1267-1273.
12. Becker, E. M., Dessuy, M. B., Boschetti, W., Vale, M. G. R., Ferreira, S. L. C. & Welz,
B. 2012. Development of an analytical method for the determination of arsenic in
gasoline samples by hydride generation–graphite furnace atomic absorption
spectrometry. Spectrochimica Acta Part B: Atomic Spectroscopy, 71–72, 102-106.
13. Aguirre, M. A., Kovachev, N., Hidalgo, M. & Canals, A. 2012. Analysis of biodiesel
and oil samples by on-line calibration using a Flow Blurring [registered sign]
multinebulizer in ICP OES without oxygen addition. Journal of Analytical Atomic
Spectrometry, 27, 2102-2110.
14. Cassella, R., Barbosa, B. S., Santelli, R. & Rangel, A. 2004. Direct determination of
arsenic and antimony in naphtha by electrothermal atomic absorption spectrometry with
microemulsion sample introduction and iridium permanent modifier. Analytical and
Bioanalytical Chemistry, 379, 66-71.
15. Meeravali, N. N. & Jai Kumar, S. 2001. The utility of a W-Ir permanent chemical
modifier for the determination of Ni and V in emulsified fuel oils and naphtha by
237
Chapter eleven: Hollow fiber solid phase microextraction of trace metal ions in fuel
transverse heated electrothermal atomic absorption spectrometer. Journal of Analytical
Atomic Spectrometry, 16, 527-532.
16. Bettinelli, M., Spezia, S., Baroni, U. & Bizzarri. G. 1995. Determination of trace
elements in fuel oils by inductively coupled plasma mass spectrometry after acid
mineralization of the sample in a microwave oven. Journal of Analytical Atomic
Spectrometry, 10, 555-560.
17. Saint'pierre, T. D., Maranhão, T. D. A., Frescura, V. L. A. & Curtius, A. J. 2005. The
development of a method for the determination of trace elements in fuel alcohol by
electrothermal vaporization–inductively coupled plasma mass spectrometry using
external calibration. Spectrochimica Acta Part B: Atomic Spectroscopy, 60, 605-613.
18. Saint'pierre, T. D., Frescura, V. L. A. & Curtius, A. J. 2006. The development of a
method for the determination of trace elements in fuel alcohol by ETV-ICP-MS using
isotope dilution calibration. Talanta, 68, 957-962.
19. Wang, T., Jia, X. & Wu, J. 2003. Direct determination of metals in organics by
inductively coupled plasma atomic emission spectrometry in aqueous matrices. Journal
of Pharmaceutical and Biomedical Analysis, 33, 639-646.
20. Risticevic, S., Niri, V., Vuckovic, D. & Pawliszyn, J. 2009. Recent developments in
solid-phase microextraction. Analytical and Bioanalytical Chemistry, 393, 781-795.
21. Pena-Pereira, F., Lavilla, I. & Bendicho, C. 2009. Miniaturized preconcentration
methods based on liquid-liquid extraction and their application in inorganic ultratrace
analysis and speciation: A review. Spectrochimica Acta Part B: Atomic Spectroscopy,
64, 1-15.
22. Cui, C., He, M. & Hu, B. 2011. Membrane solid phase microextraction with alumina
hollow fiber on line coupled with ICP OES for the determination of trace copper,
manganese and nickel in environmental water samples. Journal of Hazardous
Materials, 187, 379-385.
23. Yang, R. Q. & Xie W.L. 2006 Determination of cannabinoids in biological samples
using a new solid phase micro-extraction membrane and liquid chromatography–mass
spectrometry, Forensic Sci. Int. 162, 135–139.
24. Cerutti, S., Salonia, J. A., Ferreira, S. L. C., Olsina, R. A. & Martinez, L.D. 2004
Factorial design for multivariate optimization of an on-line preconcentration system for
platinum determination by ultrasonic nebulization coupled to inductively coupled
plasma optical emission spectrometry. Talanta 63, 1077–1082.
25. Es’haghi, Z., Khalili, M., Khazaeifar, A. & Rounaghi, G. H. 2011. Simultaneous
extraction and determination of lead, cadmium and copper in rice samples by a new preconcentration technique: Hollow fiber solid phase microextraction combined with
differential pulse anodic stripping voltammetry. Electrochimica Acta, 56, 3139-3146.
26. Ozcan, M. & Akman, S. 2005. Determination of Cu, Co and Pb in gasoline by
electrothermal atomic absorption spectrometry using aqueous standard addition in
238
Chapter eleven: Hollow fiber solid phase microextraction of trace metal ions in fuel
gasoline–ethanol–water three-component system. Spectrochimica Acta Part B: Atomic
Spectroscopy, 60, 399-402.
27.Kowalewska, Z., Ruszczyńska, A. & Bulska, E. 2005. Cu determination in crude oil
distillation products by atomic absorption and inductively coupled plasma mass
spectrometry after analyte transfer to aqueous solution. Spectrochimica Acta Part B:
Atomic Spectroscopy, 60, 351-359.
28. Somera, B., Corazza, M., Yabe, M., Segatelli, M., Galunin, E. & Tarley, C. 2012. 3mercaptopropyltrimethoxysilane-Modified Multi-walled Carbon Nanotubes as a New
Functional Adsorbent for Flow Injection Extraction of Pb(II) from Water and Sediment
Samples. Water, Air, & Soil Pollution, 223, 6069-6081.
29. Martendal, E., Maltez, H. F. & Carasek, E. 2009. Speciation of Cr(III) and Cr(VI) in
environmental samples determined by selective separation and preconcentration on
silica gel chemically modified with niobium(V) oxide. Journal of Hazardous Materials,
161, 450-456.
30. Costa, H., Fátima Lima, G., Nacano, L. & Tarley, C. 2011. Preconcentration/cleanup
studies of tin from environmental water samples by oxidized multiwall carbon
nanotubes packed column and its determination by ETAAS. Water, Air, & Soil
Pollution, 217, 557-565.
31. Bashir, M. J. K., Aziz, H. A., Yusoff, M. S. & Adlan, M. N. 2010. Application of
response surface methodology (RSM) for optimization of ammoniacal nitrogen removal
from semi-aerobic landfill leachate using ion exchange resin. Desalination, 254, 154161.
32. Körbahti, B. K. & Tanyolac, A. 2008. Electrochemical treatment of simulated textile
wastewater with industrial components and Levafix Blue CA reactive dye: Optimization
through response surface methodology. Journal of Hazardous Materials, 151, 422-431.
33. Joglekar, A. M. & May, A. T. 1987. Product excellence through experimental design.
Cereal Food World, 32, 857–868.
34. Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S. & Escaleira, L. A. 2008.
Response surface methodology (RSM) as a tool for optimization in analytical
chemistry. Talanta, 76, 965-977.
35. Nunes, L. S., Barbosa, J. T. P., Fernandes, A. P., Lemos, V. A., Santos, W. N. L. D.,
Korn, M. G. A. & Teixeira, L. S. G. 2011. Multi-element determination of Cu, Fe, Ni
and Zn content in vegetable oils samples by high-resolution continuum source atomic
absorption spectrometry and microemulsion sample preparation. Food Chemistry, 127,
780-783.
36. Santos, D. S. S., Korn, M. G. A., Guida, M. A. B., Dos Santos, G. L., Lemos, V. A. &
Teixeira, L. S. G. 2011. Determination of Copper, Iron, Lead and Zinc in Gasoline by
Sequential Multi-Element Flame Atomic Absorption Spectrometry after Solid Phase
Extraction. Journal of Brazzilian Chemical Society, 22, 552-557.
239
CHAPTER TWELVE:
PREPARATION OF TITANIA-ALUMINA HOLLOW FIBER MEMBRANE AND
MULTIVARIATE OPTIMIZATION FOR SIMULTANEOUS PRECONCENTRATION
OF TRACE ELEMENTS IN DIESEL AND GASOLINE SAMPLES PRIOR TO ICP-MS
DETERMINATION
ABSTRACT
A titania-alumina hollow fiber membrane was synthesized using the template method
coupled with a sol–gel process. The crystal forms of the mixed oxides hollow fiber was
evaluated using X-ray diffraction (XRD). The morphological structure and surface
characteristics of the titania-alumina hollow fiber was characterized by scanning electron
microscope (SEM) and nitrogen adsorption/desorption BET technique. The synthesized titaniaalumina hollow fiber membrane was used for extraction and preconcentration of trace amounts
of Co, Cr, Mo, Ni, Sb and V in diesel and gasoline samples. The optimization of the
experimental parameters was performed using a full 24 factorial design involving the variables:
sample pH, eluent concentration (EC), extraction time (ET) and eluent volume (EV). The full
factorial design was used to screen for significant variables. The optimum conditions were
determined by central composite design. These two independent multivariate designs led to the
following optimum conditions: pH = 8.0; EC = 2.75 mol L-1; ET = 25 min and EV = 5 mL.
Under optimized conditions, a preconcentration factor of 50 was obtained; LOD and LOQ
ranged from 0.01-0.12 and 0.08-0.22 µg L-1, respectively. The preconcentration method was
applied in the determination of trace elements in real diesel and gasoline samples.
Keywords: Titania-alumina hollow fiber, trace elements, multivariate optimization, diesel,
gasoline, ICP-MS
12.1 INTRODUCTION
The occurrence of metal ions in petroleum fractions such as diesel and gasoline is of
substantial importance because of its effects on the use and performance characteristics of the
desired products.1 For instance, elements like copper, antimony, nickel and vanadium are known
to catalyse oxidative reactions, degrading the thermal stability of the petroleum fractions and
only low concentrations of such metals can be tolerated especially in diesel. 1,2 Therefore,
240
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
accurate determination of trace metal ions in diesel and gasoline is a very important step in the
industrial production processes to assure their subsequent use. In addition, their quantification
helps in atmospheric pollution monitoring. For these reasons, different procedures have been
developed for the elemental analysis of diesel and gasoline. These methods involve various
sample pretreatments such as alcohol dilution,3 microemulsion4 and microwave-assisted
digestion5, among others. Techniques employing preconcentration procedures to extract metal
ions in fuel samples prior to their determination are also reported in the literature.6-8 The benefit
of using preconcentration techniques is that they combine the advantages of separating the
analyte from the complex matrix, by transferring it to an aqueous phase and preconcentrating it
at the same time.9
Membrane solid phase microextraction (MSPME) has been reported for preconcentration of
metal ions in various sample matrices such as human serum and environmental water
samples.10,11 This technique integrates sampling, extraction and preconcentration into a single
step.10,11 Furthermore, it inherits the advantages of both the solid phase microextraction (SPME)
and membrane separation.10 Due to aforementioned advantages, MSPME can be used to separate
trace elements from complex matrix samples without using special equipment.10 The principle of
MSPME is based on the retention of the analytes in the membrane. Therefore, the performance
of the membrane is one of the key aspects that determine the sensitivity and the selectivity of the
analytical method.10
Conventionally, optimization of analytical methodologies has been performed using
univariate technique, which means, monitoring one factor at time. The disadvantages of this
method are as follows; (i) it may lead to ambiguous results and misinterpretation because the
interactive effects among the variables are not examined. (ii) Univariate optimization increases
the number of experiments to be conducted. Therefore, this leads to an increase in analysis time
as well as an increase in the consumption of reagents and materials.12,13 Chemometric tools
(multivariate statistic techniques) have been commonly used to overcome the problems
connected to univariate techniques. In addition, multivariate statistic techniques allow the
simultaneous study of several experimental variables and the development of mathematical
models that permit the assessment of the relevance and statistical significance of factors being
241
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
studied.12,14 Moreover, these techniques facilitate the evaluation of the interaction effects
between factors.12
The aim of this work was to prepare a titania-alumina hollow fiber membrane for
simultaneous extraction and preconcentration of trace amounts of Co, Cr, Mo, Ni, Sb and V in
liquid fuel samples prior to their inductively coupled plasma mass spectrometry (ICP-MS)
determination. The optimization of the experimental parameters associated with extraction and
preconcentration of trace metal ions was performed by factorial and central composite designs.
12.2 EXPERIMENTAL
12.2.1 Instrumentation
A Perkin-Elmer Sciex ELAN 6000 (Perkin-Elmer SCIEX Instruments, Concord, Canada)
inductively coupled plasma mass spectrometer was used for all measurements. Argon of
99.996% purity (Afrox, South Africa) was used. The operating conditions are presented in Table
12.1. The Accurel S6/2 polypropylene hollow fiber membrane used here was obtained from
Membrana (Wuppertal, Germany). The wall thickness of the fiber was 450 μm, the inner
diameter was 1800 µm, and the pore size was 0.2 µm.
For comparative method, analyte metal ions (Co, Cr, Mo, Ni, Sb and V) were determined
using a Spectro Arcos 165 ICP OES (SPECTRO Analytical Instruments, GmbH, Germany)
equipped with Cetac ASX-520 autosampler. The operating conditions on the ICP OES
spectrometer during the measurements were as follows: forward power: 1400 W, plasma argon
flow rate: 13 L min-1, auxiliary argon flow rate: 2.00 L min-1, nebulizer argon flow rare: 0.95 L
min-1. The most prominent atomic and ionic analytical lines of metal ions were selected for
investigation, that is, Co 228.616 nm, Cr 267.716 nm, Mo 203.909, Ni 231.604 nm, Sb 217.581
nm and V 292.402 nm. Microwave assisted digestion was carried out in an Ethos D (Milestone,
Sorisole, Italy) with maximum pressure 1450 psi and maximum temperature 300°C.
242
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
Table 12.1. Operational ICP-MS parameters
RF power
Gas flow rates
Outer
Intermediate
Carrier
Resolution
Sweeps per reading
Dwell time
Readings per replicate
Replicates
Auto lens
Isotopes
a
1100
15 L min-1
1.2 L min-1
0.95 L min-1
0.7 a.m.u. (10% of the peak height)
1
25 ms
100
3
On
59
Co, 52Cr, 95Mo, 60Ni, 121Sb, 51V 45Sc (ISa),
103
Rh, 209Bi (IS)
IS = Internal standard
Morphological structure of the Al2O3, TiO2 and TiO2-Al2O3 were observed using scanning
electron microscope (SEM) (JSM-6360LVSEM, JEOL Co., Japan) after gold coating and the
diameter of the mixed metal oxide was measured by image processing software. The specific
surface area value was determined from adsorption isotherms by the Brunauer, Emmett and
Teller (BET) multipoint method using Surface Area and Porosity Analyzer (ASAP2020 V3.
00H, Micromeritics Instrument Corporation, Norcross, USA). All the gases used for analysis
were instrument grade. X-ray powder diffraction (XRD) measurements were carried out with a
Philips X-ray generator model PW 3710/31 a diffractometer with automatic sample changer
model PW 1775 (scintillation counter, Cu-target tube and Ni-filter at 40 kV and 30 mA).
12.2.2 Reagents and Solutions
All reagents were of analytical grade unless otherwise stated and double distilled deionised
water (Millipore, Bedford, MA, USA) was used throughout the experiments. Aluminum
isopropoxide and tetrabutyl titanate (Sigma-Aldrich, St. Loius, MO, USA) were used as a
precursor for the preparation on titania-alumina. Synthetic gasoline was prepared by mixing 91%
isooctane and 9% n-heptane (Sigma-Aldrich, St. Loius, MO, USA). Spectrascan stock solutions
(1000 mg L-1) of the target metal ions (Industrial Analytical (Pty) Ltd, Johannesburg, South
243
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
Africa) were used to prepare the working solutions for MSPME at concentrations of 10 µg L-1
for other metal ions. Working solutions (prepared in organic phase), as per the experimental
requirements, were freshly prepared from the stock solution for each experimental run. A
Spectrascan multi-element standard solution at concentration of 100 mg L-1 (Industrial
Analytical (Pty) Ltd, Johannesburg, South Africa) was used to prepare working standard
solutions for measurements of concentrations of analytes in model and sample solutions.
Solutions of nitric acid at different concentrations (used for the elution of the analytes from the
hollow fiber membrane) were prepared from ultrapure concentrated acid (65%, Sigma-Aldrich,
St. Loius, MO, USA). The pH adjustments were performed with glacial acetic acid (Merck,
Darmstadt, Germany) and ammonia (Sigma-Aldrich, St. Loius, MO, USA) solutions.
12.2.3 Synthesis Titania-Alumina Sol
The synthesis of titania-alumina sol was prepared according to Jung et al..15 To describe the
procedure briefly, proper amounts (1:1 ratio) of aluminum isopropoxide (Sigma-Aldrich, St.
Loius, MO, USA) and titanium butoxide (Sigma-Aldrich, St. Loius, MO, USA) were dissolved
in ethanol, and the solution was then diluted with double distilled deionised water. The pH of the
resulting solution was adjusted to 2 using 1.0 mol L-1 nitric and then it was stirred at 75°C for 24
h. Synthesis of nanometer-sized titania and alumina powders was according to Li et al.16 and
Rogojan et al.17.
12.2.4 Preparation of Titania-Alumina Hollow Fiber
The preparation of titania-alumina hollow was carried out according the methods reported by
Cui et al.10 and Huang and Hu11. Briefly, polypropylene hollow fibers were cut into equal
segments, ultrasonicated in acetone for 15 min and then removed and dried in air. For coating,
the dried polypropylene hollow fibers were immersed in the above prepared titania-alumina sol
for 2 h, followed by a drying procedure with careful temperature control at 80°C for 2 h. The
above immersion and drying processes were repeated several times, resulting in titania-alumina
coated-polypropylene hollow fibers. Finally, the coated hollow fibers were heated from room
temperature to 1000°C at 2°C min-1 and maintained for 3 h to remove the polypropylene
template and crystallize the titania-alumina hollow fiber membrane.
244
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
12.2.5 Preconcentration Method
The procedure for the preparation of gasoline–ethanol–water mixture was carried out
according to Ozcan and Akman.18 A 10 ml aliquot of synthetic gasoline sample was placed in a
100 ml polyethylene volumetric flask followed by the addition of 5 ml of concentrated HNO3
and 10 ml of water. The mixture was spiked with 1.0 mL of a 1.0 mg L-1 multi-element oil
standard solution and made up to the mark with ethanol to obtain 10 µg L-1 concentration of each
metal ion. The mixture was homogenized by shaking and a single phase solution was obtained. It
should be noted that the stability of the gasoline-ethanol- water mixture was not monitored. This
is because the resulting mixture was subjected the preconcentration system immediately after
homogenization. The extraction and preconcentration procedure for metal ions in synthetic
gasoline samples was carried out as follows: the fibers were placed in the sample solutions
containing 10 µg L-1 metal ions (50 mL) present in 100 ml plastic bottles. The bottles were
covered and shaken at 500 rpm. As the analytes from the sample solution diffuses through the
titania-alumina membrane, they get adsorbed onto the pores of the hollow fiber. At the end of the
preconcentration, the hollow fiber was taken out from the vial, rinsed with double distilled
deionised water and transferred into a polypropylene centrifuge vial containing 5.0 mL of
appropriate HNO3 concentration. The metal ions were desorbed from the fiber with ultrasonic
agitation for 5 min. The same procedure was applied to the blank solutions. In the case of real
sample analysis, an aliquot of 2.5 mL of gasoline sample was placed in a 250 mL polypropylene
volumetric flask followed by the addition of 1.25 mL of concentrated HNO3 and 25 mL double
distilled deionised water, respectively. The mixture was then diluted to the mark with ethanol.
12.2.6 Optimization Strategy
The optimization of the preconcentration system was carried out using a 2 4 full factorial and
central composite design. Four variables i.e. sample pH, eluent concentration (EC), extraction
time (ST) and eluent volume (EV) were regarded as important factors. Maximum, central point
and minimum levels in Table 12.2 for each factor were chosen according to the data from
previous experiments. All the experiments were carried out in random order. The experimental
245
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
data was processed by using the Minitab version 16 and Design Expect version 8 statistic
software programs.
Table 12.2. Factors and levels used in 24 factorial design for separation and preconcentration of
metal ions in fuel samples
Variable
pH
EC (mol L-1)
ET(min)
EV (mL)
Low level (-1)
5
1.5
10
5
Central point (0)
8
2.75
25
10
High level (+1)
11
4
40
15
12.2.7 Comparative Method
The Microwave acid digestion procedure was carried out according to Kowalewska et al.19
Briefly, 5.0 ml of the gasoline sample was placed in a Teflon vessel followed by 6.0 mL HNO3
(65%) and 2.0 mL H2O2 (30%). The vessels were inserted into a microwave unit and heated
according to the conditions recommended by the manufacturer. The digested content was left to
cool down to room temperature. After cooling, the vessels were opened and 2.0 ml of
concentrated HNO3 and 2.0 ml of hydrogen peroxide were added, and the heating program was
repeated. Finally, the Teflon vessel contents were cooled down to room temperature and
quantitatively transferred to a 50 ml calibration flask, 1.0 mL of concentrated nitric acid was
added and the samples were spiked with 20 µg L-1 of the target analytes. The flask was then
filled up to the mark using double distilled deionised water. The latter water was submitted to the
same procedure and used as a blank. The samples were then analyzed with ICP OES.
12.3 RESULTS AND DISCUSSION
12.3.1 Characterization of Titania-Alumina Hollow Fiber
The prepared titania-alumina hollow fiber was characterized by powder X-ray diffraction
(XRD),
scanning
electron
microscopy
adsorption/desorption measurements.
246
(SEM),
and
low-temperature
nitrogen
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
12.3.1.1 X-ray diffraction analysis
X-ray diffraction patterns (for 2θ diffraction angles from 10° to 80°) of the nanometer-sized
alumina, titania and titania-alumina hollow fiber calcined at 1000°C for 3 hours are presented in
Fig. 12.1. The XRD patterns showed well-crystallized structures when the calcinations metal
oxides and mixed metal oxide hollow fiber were at 1000°C. The XRD patterns for pure alumina
and titania powders were used as reference materials. It can be seen from Fig. 12.1C that the
peaks for titania and alumina were not overlapping. This shows that the mixed oxides were not
simply mixed phases of pure titania and alumina, but solid solutions with a single phase. 15
Fig. 12.1. XRD spectra of nanometer-sized alumina powder (A), nanometer-sized titania powder
(B) and titania-alumina hollow fiber (C) calcined at 1000°C for 3 hours. (Theta phases: α =
alpha-phase Al2O3, γ = gamma-phase Al2O3, R = rutile TiO2)
247
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
12.3.1.2 Pore structure parameters
The pore structure of the titania-alumina hollow fiber and the polypropylene hollow fiber
were investigated by nitrogen adsorption/ desorption experiments. The surface area for titania
hollow fiber membrane and the polypropylene hollow fiber were 116 and 26.6 m2 g-1,
respectively while the pore volume of titania hollow fiber and the polypropylene hollow fiber
were 0.04 and 0.09 mL g-1, respectively, with the pore size of 10.0 and 21.3 nm, respectively.
12.3.1.3 Scanning electron microscopy (SEM) analysis
Fig. 12.2 shows the SEM textural images of the titania-alumina (A) and polypropylene
hollow fiber (B). It can be seen from this figure that the textual image of titania-alumina hollow
fiber had different nanopores sizes which was different from that observed for polypropylene
hollow fiber. The latter showed fibrous like structures. It is worth mentioning that the nanopores
in the titania-alumina hollow fiber leads to an enhanced surface area and fast mass transfer for
the analyte during the preconcentration process.10,11
A
B
Fig. 12.2. SEM textural images of the titania-alumina (A) and polypropylene hollow fiber (B).
248
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
12.3.2 Screening Analysis of Membrane Solid Phase Microextraction (MSPME)
Preconcentration System
Two level (24) full factorial design was used as a screening method for optimization of
preconcentration system based on membrane solid phase microextraction using titania-alumina
hollow fiber. The significance and possibility of interactions between sample pH, eluent
concentration (EC), extraction time (ET) and eluent volume (EV) were evaluated. The effect of
factors on the MSPME preconcentration system was investigated by using analysis of variance
(ANOVA) taking into consideration the percentage recovery as the analytical response. Table
12.3 shows the experimental design matrix and the analytical results obtained in each run
expressed as average percentage recovery.
Table 12.3. List of experiments in the factorial design (actual values) for MSPME optimization
and the responses
Run pH
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
-1
1
-1
1
-1
1
-1
1
-1
1
-1
1
-1
1
-1
1
0
0
0
EC
-1
-1
1
1
-1
-1
1
1
-1
-1
1
1
-1
-1
1
1
0
0
0
ET
-1
-1
-1
-1
1
1
1
1
-1
-1
-1
-1
1
1
1
1
0
0
0
EV Co
-1
-1
-1
-1
-1
-1
-1
-1
1
1
1
1
1
1
1
1
0
0
0
70.73
63.81
71.01
60.11
83.14
75.22
84.12
74.81
71.11
62.91
71.35
61.40
82.55
74.38
81.38
73.82
93.83
93.67
93.74
Cr
Mo
61.84
58.77
63.14
57.81
77.87
71.87
78.93
70.72
64.45
58.56
65.01
57.98
79.14
67.57
77.67
62.39
81.93
82.03
81.88
Recovery (%)
75.15
66.00
71.37
60.73
76.44
67.24
70.53
59.17
85.00
78.75
80.32
67.46
86.13
79.35
79.17
68.03
74.66
64.24
68.94
56.18
75.26
65.62
69.58
57.34
86.09
80.15
73.73
70.65
83.41
81.96
72.18
71.11
95.63
88.12
95.45
87.79
95.57
87.81
249
Ni
Sb
V
72.50
68.91
70.69
69.17
90.33
83.79
90.92
80.83
71.68
69.55
71.59
68.56
90.21
79.83
92.12
80.79
96.33
96.40
96.22
72.76
65.55
73.94
66.15
95.91
88.62
96.76
90.03
70.80
64.97
73.66
65.44
97.81
90.78
98.98
91.55
97.72
97.54
97.98
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
The main effects and their interactions are presented in the Pareto charts shown in Figs. 3-5.
Bar lengths are proportional to the absolute value of the estimated effects, which helps in
comparing the significance of effects.
Cobalt: Fig 12.3A shows the Pareto chart of main effects produced from the ANOVA results
for cobalt. The interpretation of Pareto chart demonstrates that extraction time (12.19) and
sample pH (-8.68) are highly significant at 95% confidence level. It was observed that an
increase in extraction time increases the percentage recovery. Whereas increasing sample pH
decreases the analytical response. The interaction between sample pH and eluent concentration (0.88) was also statistically significant. However its significance was much lower compared to the
main effects (extraction time and sample pH). The eluent concentration and eluent volume were
not significant factors.
Chromium: According to Fig. 12.3B, sample pH (-7.46) and extraction time (11.99) were
statistically significant at the 95% confidence level. Moreover, interactions between sample pH
and extraction time (-2.19), extraction time and eluent volume (-2.35), as well as between sample
pH and eluent concentration (-1.51) were statistically significant. The results indicated that these
variables have a synergistic effect on percentage recovery. This means that, the use of extraction
time at high level and sample pH at the lower level and their combination also at low and high
levels may results to a better analytical response.
Fig. 12.3. Pareto charts of standardized effects for variables in the cobalt and chromium
preconcentration.
250
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
Molybdenum: Fig. 12.4A shows that sample pH (-7.05), extraction time (8.01) and eluent
volume (-2.52) as well as interactions between sample pH and extraction time (-1.78) were
statistically significant at 95% confidence level. The eluent concentration and eluent volume
were significant at 95% confidence level. However, the interaction between sample pH and
eluent volume was statistically significant. According to the negative effect (-1.72) of pH×EV
interaction, a higher analytical signal will be achieved by simultaneous decrease in sample pH
and eluent volume levels.
Nickel: Fig. 12.4B demonstrated that sample pH and extraction time as well as pH×ET and
ET×EV interactions were significant at the levels studied. The algebraic signs for the significant
main effects were similar those observed for other metal ions, that is, negative value (-9.08) for
sample pH and positive value (12.62) for extraction time. The negative effect of sample pH
implied that an increase of this factor decreases the nickel retention, thus decreasing its recovery.
Whereas increasing the extraction time leads to increased the nickel retention.
Fig. 12.4. Pareto charts of standardized effects for variables in the molybdenum and nickel
preconcentration.
Antimony: Fig. 12.5A shows the Pareto chart of standardized effects for variables in the
extraction preconcentration of antimony. It can be seen from Fig. 5A that all studied variables
and their interactions were not significant at 95% confidence level. However, considering the
251
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
positive sign of the main effect of extraction time (15.77), it can be concluded that the extraction
leads to a more pronounced enhancement of the analytical response.
Vanadium: The Pareto chart in Fig. 12.5B demonstrates that extraction time (24.47) provided
a more significant effect for extraction and preconcentration of vanadium. This means that an
increase in extraction time leads to higher percentage recovery. Sample pH (-6.87) and eluent
concentration (0.989) were also statistically significant. However, the effect of eluent
concentration on the analytical response was very small. The eluent volume was not significant
at 95% confidence level. However, the interaction between the extraction time and eluent
volume (1.09) was statistically significant.
Fig. 12.5.Pareto charts of standardized effects for variables in the preconcentration of antimony
and vanadium
By analysing the overall results in Figs. 12.3-12.5, it can be seen that extraction time and
sample pH were the most important variable for retention of the studied analytes. Based on the
effect estimate (negative values ranging from -6 to -9.1) for sample pH, the retention of all metal
ions decreased significantly with increasing sample pH. The properties of alumina and titania
surface strongly depend on pH and below points of zero charge 7.3 and 6.02 for alumina and
titania, respectively, the surface is positively charged.10,11,20 Therefore, the sample pH should be
above the points of zero charge. This is because above these points the surface of the titaniaalumina hollow fiber covered with OH groups is negatively charged. Therefore, it attracts the
252
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
analytes of interest and leads to an enhancement of the adsorption efficiency. Therefore, an
optimum sample pH value for the metal ion retention onto titania-alumina hollow fiber
membrane should occur in a range of 7.5-8.5.
The influence of extraction time factor in the proposed preconcentration method showed that
higher levels must be employed to optimize the retention process. The ANOVA results for all the
studied metal ions showed that the estimated value of pH×ET interaction had a negative value.
This implied that sample pH levels must be decreased while increasing the ET levels. Eluent
concentration and volume were not or less significant at 95% confidence level as compared to
the aforementioned variables. Therefore, eluent concentration was fixed at 2.75 mol L-1 while
eluent volume was fixed at 5 mL. The overall results obtained for the screening analysis using 24
full factorial experimental design indicated that sample pH and extraction time require a final
optimization. Therefore, the significant variables were further optimized using a central
composite design.
12.3.3 Optimization of MSPME Preconcentration System
After screening out the variables that did not have significant effect on the response, the
remaining two factors sample pH and extraction time (ET) were further optimized to provide the
maximum recovery. A central composite design containing a total of 14 experiments were
carried out to optimize these two variables. Table 12.4 shows the central composite design
matrix and the analytical results obtained in each run expressed as average percentage recovery
Fig. 12.6 shows the 3D surface responses of the quadratic models that were used to evaluate
the interactive relationships between independent variables (pH and extraction time) and
response. As mentioned before, the variables that were shown to be insignificant by full factorial
design were taken at fixed values, eluent concentration (2.75 mol L-1) and eluent volume (5 mL).
As it can be seen in Fig. 12.6, the maximum observed recoveries of all the studied metal ions
ranged from 95-100% at pH 8 and extraction time 25 min. Based on results obtained from 24 full
factorial and central composite designs, the optimum conditions that led to quantitative retention
and elution of metal ion were as follows: pH = 8.0, extraction time = 25 min, eluent
concentration = 2.75 mol L-1 and eluent volume = 5 mL.
253
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
Table 12.4. List of experiments in the central composite design (actual values) for MSPME
optimization and the responses
Run
1
2
3
4
5
6
7
8
9
10
11
12
13
14
pH
5
11
5
11
8
8
8
3.76
12.24
12.24
8
8
8
8
ET
10
10
40
40
25
25
25
25
25
3.79
46.21
25
25
25
Co
42.1
41.3
82.4
76.8
99.1
99.4
99.3
50.8
41.9
40.8
100.3
99.2
99.1
99.3
Cr
Mo
51.7
50.3
83.6
78.2
97.1
97.4
96.7
61.2
57.3
50.8
97.9
97.1
97.2
97.2
Recovery (%)
67.4
51.4
62.2
49.2
87.9
85.7
83.6
80.8
98.6
98.1
98.5
98.0
98.4
98.2
75.3
53.3
69.8
48.7
63.7
46.3
98.7
98.5
98.5
98.3
98.6
98.3
98.5
96.9
254
Ni
Sb
V
64.6
61.9
86.7
78.0
95.5
95.7
95.6
70.9
63.2
65.3
97.2
95.4
95.7
95.6
62.3
60.1
85.6
90.5
99.9
99.7
99.8
67.5
63.5
58.4
100.2
99.8
99.7
99.8
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
Fig. 12.6. Response surface for percentage recovery of Cr (A), Co (B), (C), Mo (D), Ni (E) and
V (F) as function of extraction time (ET), min.
12.3.4 Effect of Sample Volume
The effect of sample volume on the retention of the target analytes onto titania-alumina
hollow fiber membrane was investigated in the range of 50-300 mL, while keeping the metal ion
concentration fixed at 10 µg L-1. Fig. 12.7 presents the effect of the sample volume on the
255
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
recovery of the Co, Cr, Mo, Ni, Sb and V. It can be seen from this figure that the quantitative
recoveries for all analytes were obtained when the sample volume was 200 mL. Therefore, this
volume was used for further investigations.
Fig. 12.7. Effect of sample volume on the recoveries of metal ions
12.3.5 Adsorption Capacities and Regeneration of the Hollow Fiber
The investigation of adsorption capacities of an adsorbent is an important factor, because it
determines how much of sorbent is required to quantitatively concentrate the analytes from a
given solution.10 The adsorption capacity of the titania-alumina hollow fiber membrane was
studied and the experimental data were fitted into the general equation of the modified Langmuir
model presented in Eq. 1.21 The latter was used to calculate the maximum adsorption capacity.
Ce
1
1

Ce 
qe qmax
K L qmax
(1)
The results showed that adsorption capacity of the analytes probably differ due to their size,
degree of hydration and the value of their binding constant with titania-alumina hollow fiber
256
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
membrane. The maximum adsorption capacities were found to be 17.51, 18.74, 19.63, 15.39,
19.11 and 20.65 mg g-1 for Co, Cr, Mo, Ni, Sb and V, respectively.
The stability and regeneration possibility of the titania-alumina hollow fiber membrane were
investigated. The adsorbent can be reused after regeneration with 5.0 ml of a 2.75 mol L−1 HNO3
solution and 10 ml double distilled deionised water, respectively, and was relatively stable up to
60 runs without an obvious decrease in the recoveries for the studied ions.
12.3.6 Analytical Figure of Merit
The MSPME preconcentration method provided a linear dynamic range of calibration from
0.2 up to 250 μg L−1 for all studied analytes with satisfactory correlation coefficient ranging from
0.9983-0.9992. The limits of detection (LOD) and quantification (LOQ) of the proposed
preconcentration procedure were estimated under optimal experimental conditions and they were
1
1
calculated according to IUPAC recommendation from C LOD  3  SDm and C LOQ  10  SDm ,
where SD is the standard deviation of the blank (n=21) and m is the slope of the calibration
curve. For 200 mL sample volume, the sensitivity, LOD, LOQ and precision (in terms of relative
standard deviation) values are presented in Table 12.5.
Table 12.5. Analytical figure of merit of the MSPME system for preconcentration of metal ions
obtained under optimum conditions
Analyte
Co
Cr
Mo
Ni
Sb
V
Sensitivity
(cps L µg-1)
138.5
98.4
119.3
128.1
113.7
141.8
LOD (µg L-1)
LOQ (µg L-1)
0.07
0.09
0.11
0.09
0.08
0.06
0.23
0.29
0.35
0.29
0.27
0.21
257
Precision
(%RSD)
3.1
2.3
2.9
3.0
1.5
1.2
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
12.3.7 Validation, Application of MSPME Preconcentration System to Real Samples and
Comparison with a Standard Method
In order to assess the accuracy of the optimized MSPME methodology for preconcentration
of Co, Cr, Mo, Ni, Sb and V and their ICP-MS determination liquid fuels, diesel sample spiked
with inorganic and organic standard solutions of the target analytes (5 µg L-1) was analyzed
using the developed procedure. It is worth mentioning that certified reference materials (CRM)
in suitable matrixes such as diesel or gasoline sample at the working concentration ranges (trace
levels) were not available. The main objective of spiking the diesel sample with organic and
inorganic standard solutions was to evaluate the titania-alumina hollow fiber membrane sorption
efficiency to different metal species in liquid fuel samples. This is because trace element forms
in petroleum products are not fully known and different species may display different adsorption
behaviours.6 All analyses were performed in triplicate and the analytical results obtained are
given in Table 11.6. It can be seen from this table that the percentage recoveries range from 9599% for both aqueous and organic standards. The obtained results attest to the accuracy of the
proposed preconcentration procedure.
Table 12.6. Analytical results obtained in the analysis of spiked diesel sample. The concentration
and recovery values are expressed as the mean ± standard deviation of the three replicates
Analytes
Co
Cr
Mo
Ni
Sb
V
Added (µg L-1)
0
5
0
5
0
5
0
5
0
5
0
5
Inorganic standard
Found (µg L-1) R (%)
ND
4.8±0.7
96.2±1.1
9.1±1.5
14.0±1.5
97.3±2.1
105±3
111±4
98.8±1.8
496±3
501±4.
97.5±1.5
ND
4.9±0.5
98.1±1.2
6.9±0.7
11.9±0.9
99.2±1.3
258
Metallo-organic standard
Found (µg L-1) R (%)
ND
4.8±0.4
95.6±0.8
9.1±1.5
14.0±1.2
98.1±0.9
105±3
110±3
97.6±1.7
496.0±3
501±4
96.1±1.0
ND
4.9±1.1
97.8±2.0
6.9±0.7
11.8±1.3
97.5±2.2
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
The applicability of the proposed MSPME method was evaluated for extraction and
preconcentration of metal ions in gasoline and diesel samples. As there is no diesel, gasoline or
similar reference material available with certified values for Co, Cr, Mo, Ni, Sb and V, it was
essential to use an independent technique for validation of the analytical results obtained by the
proposed method. For this reason, the samples were digested using microwave-assisted digestion
(MAD) method and then the concentrations of the analytes were determined using ICP OES. It
should be noted that ICP OES determination after microwave-assisted digestion was taken as a
reference method. Table 12.7 summarizes the results obtained for the preconcentration and
determination of the target analytes in diesel samples by MSPME/ICP-MS and MAD/ICP OES.
It can be seen from Table 12.7 that cobalt and antimony were not quantified in diesel samples
as their concentrations were found to be below the LOD. Diesel (D1) and G1 samples had
relatively high concentrations of metal ion compared to D2 and G2 samples. The Ni
concentrations were relatively higher in D1, G1 and D2 samples than in G2. It is worth
mentioning that this element is quite abundant in the Earth's crust and also sample contamination
during the diesel and gasoline production process should not be disregarded. The concentration
of Mo was higher in diesel samples compared to gasoline sample. Molybdenum is normally used
as a catalyst in the desulfurisation of petroleum, petrochemicals and coal-derived liquids to
minimise sulfur dioxide emission from fuel combustion. Therefore, the relative high
concentration in diesel samples might be due to residues of Mo leached out during the
desulfurisation process. The concentration of other metal ions such as V, Cr (except in gasoline
samples) and Co were quite low (ranging from 2.2 to 11.6 µg L-1). The quantification of these
metal ions required an analytical technique with high detection capability, such as the one
reported in this study.
As stated before, the samples were also analysed by ICP OES after microwave-assisted
digestion. The results were compared with those obtained by the MSPME/ICP-MS method. In
the case of diesel samples, the two methods gave essentially similar results for quantification of
Mo and Ni. In the gasoline samples, the results were similar for determination of Cr, Mo, Ni and
Sb, in G1 samples, whereas Cr, Mo and Ni were similar for G2 samples. Statistically, these
results were not significantly different at 95% confidence level. This demonstrated the reliability
of the proposed method. When using the comparative method (MAD/ICP OES), the
259
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
concentration of Co, Co, Sb and V were not quantified in diesel samples as they were present in
trace levels (< 10 µg L-1). Furthermore, Cr and V were not quantified in G1 sample as their
concentrations were found to be below the LOD of the instrument. In the case of G2, Cr, Sb and
V were also not quantified. It should be noted that the samples were diluted ten times after acid
digestions. Therefore, the concentration of elements in diluted samples were less than or equal to
1.2 µg L-1 and the samples were spiked with 20 µg L-1 prior to ICP-OES determination. In
addition, the differences between the two methods might be attributed to incomplete
mineralization (especially in diesel samples).
The advantage of the proposed MSPME method is that it does not require rigorous acid
digestion
unlike
the
microwave-assisted
digestion
method.
In
addition,
MSPME
preconcentration method is advantageous because it minimizes the risks of incomplete
mineralization of the organic matrix and cross-contamination. Due to the relatively high
preconcentration factor, metal ions that are present in low levels (sub-ppb) were easily quantified
by the proposed method. The time required for preconcentration of 200 mL sample (25 min
extraction and 5 min elution) was about 30 min. It should be noted the thermostat shaker can
handle up to 64 samples at the same time. Therefore, the throughput sample was approximately
64 samples h-1. Although, the current method had relatively longer preconcentration time, The
MSPME method had a higher throuput compared to microwave-assisted digestion method (10
samples h-1).
260
Chapter twelve:
MSPME for preconcentration of trace metal ions in fuel
Table 12.7. Determination of Co, Cr, Mo, Ni, Sb and V (µg L-1) in commercial diesel (D1 and D2) and gasoline (G1 and G2) samples
by proposed MSPME and comparative method MAD/ICP OES (n = 3, at 95% confidence level).
Techniques
MSPME/ICP-MS
MAD/ICP OES
Analytes
Co
Cr
Mo
Ni
Sb
V
Co
Cr
Mo
Ni
Sb
V
D1
D2
ND
9.1±1.5
105±3
496±3
ND
6.9±0.7
ND
ND
105±3
495±4
ND
ND
Concentrations (µg L-1)
ND
11.6±1.4
2.2±0.7
90.9±2.5
73.2±2.7
47.5±1.4
122±3
367±6
ND
70.9±0.9
5.2±0.2
5.93±0.7
ND
ND
ND
90.4±3.1
72.9±2.1
48.1±1.8
122±4
363±6
ND
71.3±1.1
ND
ND
261
G1
G2
7.7±0.3
23.2±0.8
20.9±1.3
69.4±1.9
4.8±0.3
6.1±0.7
ND
22.8±0.7
21.2±1.2
69.1±2.3
ND
ND
Chapter twelve: MSPME for preconcentration of trace elements in liquid fuel
12.4 CONCLUSIONS
This study presents the preparation of titania-alumina hollow fiber membrane using
polypropylene hollow fiber as the template. The hollow fiber membrane was characterized
with XRD, SEM and BET. The prepared hollow fiber membrane was applied as a solid
phase material for the MSPME technique. The latter was applied to the separation and
preconcentration of Co, Cr, Mo, Ni, Sb and V in diesel and gasoline samples prior to ICPMS determination. The experimental parameters of the proposed method were achieved by
using chemometric methods namely 24 factorial and central composite designs. Under
optimized conditions, the quantitative retention and elution of metal ion was achieved
when sample pH, extraction time, eluent concentration and eluent volume were 8.0, 25
min, 2.75 mol L-1and 5 mL respectively. The optimized MSPME technique proved to be
suitable for simultaneous preconcentration of metal ions in diesel and gasoline samples.
The preconcentration step permitted the elimination of the organic matrix, thus, avoiding
the need for digestion of the samples before metal ion determination. The developed
method was applied for the determination of the target analytes in four liquid fuel (two
diesel and two gasoline) samples purchased from different filling stations. The developed
MSPME method can be considered as an alternative to other sample preparation
techniques such as microwave acid digestion because it displays relatively low LOD and
LOQ (0.07-0.11 and 0.21-0.35 µg L-1, respectively)
12.5 REFERENCES
1. Peiselt da Silva, K. M. & Pais da Silva, M. I. 2004. Copper sorption from diesel oil on
chitin and chitosan polymers. Colloids and Surfaces A: Physicochemical and
Engineering Aspects, 237, 15-21.
2. Trindade, J. M., Marques, A. L., Lopes, G. S., Marques, E. P. & Zhang, J. 2006. Arsenic
determination in gasoline by hydride generation atomic absorption spectroscopy
combined with a factorial experimental design approach. Fuel, 85, 2155-2161.
3. Chaves, E. S., De Loos-Vollebregt, M. T. C., Curtius, A. J. & Vanhaecke, F. 2011.
Determination of trace elements in biodiesel and vegetable oil by inductively coupled
plasma optical emission spectrometry following alcohol dilution. Spectrochimica Acta
Part B: Atomic Spectroscopy, 66, 733-739.
4. de Souza, R. M., Meliande, A. L. S., Da Silveira, C. L. P. & Aucélio, R. Q. 2006.
Determination of Mo, Zn, Cd, Ti, Ni, V, Fe, Mn, Cr and Co in crude oil using
262
Chapter twelve: MSPME for preconcentration of trace elements in liquid fuel
inductively coupled plasma optical emission spectrometry and sample introduction as
detergentless microemulsions. Microchemical Journal, 82, 137-141.
5. Sant’Ana, F. W., Santelli, R. E., Cassella, A. R. & Cassella, R. J. 2007. Optimization of
an open-focused microwave oven digestion procedure for determination of metals in
diesel oil by inductively coupled plasma optical emission spectrometry. Journal of
Hazardous Materials, 149, 67-74.
6. Santos, D. S. S., M.G. A. Korn, M. A. B. Guida, G. L. Dos Santos, V. A. Lemos And L.
S. G. Teixeira 2011. Determination of copper, iron, lead and zinc in gasoline by
sequential multi-element flame atomic absorption spectrometry after solid phase
extraction. Journal of Brazzilian Chemical Society, 22, 552-557
7. Nomngongo, P. N., Ngila, J. C., Kamau, J. N., Msagati, T. A. M. & Moodley, B. 2013a.
Preconcentration of molybdenum, antimony and vanadium in gasolsine samples using
Dowex 1-x8 resin and their determination with inductively coupled plasma–optical
emission spectrometry. Talanta, 110, 153-159.
8. Nomngongo, P. N., Ngila, J. C., Musyoka, S. M., Msagati, T. A. M. & Moodley, B.
2013. A solid phase extraction procedure based on the use of electrospun cellulose-goxolane-2,5-dione nanofibers for trace determination of Cd, Cu, Fe, Pb and Zn in
gasoline samples by ICP OES. Analytical Method 5, 3000-3008.
9. Korn, M. D. G. A., Dos Santos, D. S. S., Welz, B., Vale, M. G. R., Teixeira, A. P.,
Lima, D. D. C. & Ferreira, S. L. C. 2007. Atomic spectrometric methods for the
determination of metals and metalloids in automotive fuels - A review. Talanta, 73, 111.
10. Huang, C. & Hu, B. 2011. Synthesis and characterization of titania hollow fiber and its
application to the microextraction of trace metals. Analyst, 136, 1425-1432.
11. Cui, C., He, M. & Hu, B. 2011. Membrane solid phase microextraction with alumina
hollow fiber on line coupled with ICP OES for the determination of trace copper,
manganese and nickel in environmental water samples. Journal of Hazardous
Materials, 187, 379-385.
12. Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S. & Escaleira, L. A. 2008.
Response surface methodology (RSM) as a tool for optimization in analytical
chemistry. Talanta, 76, 965-977.
13. Lobo, F. A., Goveia, D., Oliveira, A. P. D., Pereira-Filho, E. R., Fraceto, L. F., Filho,
N. L. D. & Rosa, A. H. 2009. Comparison of the univariate and multivariate methods in
the optimization of experimental conditions for determining Cu, Pb, Ni and Cd in
biodiesel by GFAAS. Fuel, 88, 1907-1914.
14. Tarley, C. R. T., Silveira, G., Dos Santos, W. N. L., Matos, G. D., Da Silva, E. G. P.,
Bezerra, M. A., Miró, M. & Ferreira, S. L. C. 2009. Chemometric tools in
electroanalytical chemistry: Methods for optimization based on factorial design and
response surface methodology. Microchemical Journal, 92, 58-67.
263
Chapter twelve: MSPME for preconcentration of trace elements in liquid fuel
15. Jung, Y.-S., Kim, D.-W., Kim, Y.-S., Park, E.-K. & Baeck, S.-H. 2008. Synthesis of
alumina–titania solid solution by sol–gel method. Journal of Physics and Chemistry of
Solids, 69, 1464-1467.
16. Li, J., Qi, H.-Y. & Shi, Y.-P. 2009. Applications of titania and zirconia hollow fibers in
sorptive microextraction of N,N-dimethylacetamide from water sample. Analytica
Chimica Acta, 651, 182–187.
17. Rogojan, R., Andronescu, E., Ghitulica, C. & Vasile, B.S. Synthesis and
characterisation of Alumina nano-powder obtained by sol-gel method, U.P.B. Sci. Bull.
Series B 73 (2011) 65-76.
18. Ozcan, M. & Akman, S. 2005. Determination of Cu, Co and Pb in gasoline by
electrothermal atomic absorption spectrometry using aqueous standard addition in
gasoline–ethanol–water three-component system. Spectrochimica Acta Part B: Atomic
Spectroscopy, 60, 399-402.
19. Kowalewska, Z., Ruszczyńska, A. & Bulska, E. 2005. Cu determination in crude oil
distillation products by atomic absorption and inductively coupled plasma mass
spectrometry after analyte transfer to aqueous solution. Spectrochimica Acta Part B:
Atomic Spectroscopy, 60, 351-359.
20. Vassileva, E., Proinova, I. & Hadjiivanov, K. 1996. Solid-phase extraction of heavy
metal ions on a high surface area titanium dioxide (anatase). Analyst, 121, 607-612.
21. Qu, R., Sun, C., Ma, F., Cui, Z., Zhang, Y., Sun, X., Ji, C., Wang, C. & Yin, P. 2012.
Adsorption kinetics and equilibrium of copper from ethanol fuel on silica-gel
functionalized with amino-terminated dendrimer-like polyamidoamine polymers. Fuel,
92, 204-210.
264
CHAPTER THIRTEEN:
CHEMOMETRIC OPTIMIZATION OF HOLLOW FIBER-LIQUID PHASE
MICROEXTRACTION FOR PRECONCENTRATION OF TRACE ELEMENTS IN
DIESEL AND GASOLINE PRIOR TO THEIR ICP OES DETERMINATION
ABSTRACT
A hollow fiber-liquid phase microextraction (HF-LPME) method for the simultaneous
extraction and preconcentration of Ag, Al, As, Mn and Ti as ammonium pyrrolidine
dithiocarbamate (APDC) complexes in [C6MIM][PF6] ionic liquid, is reported.
Multivariate techniques such as 24 factorial and Box–Behnken designs were used for the
optimization of experimental parameters. Under optimized conditions, the limits of
detection and quantification ranged from 0.04-0.09 and 0.15-0.29 µg L-1, respectively. The
preconcentration factors of 150, 291, 112, 405, 367 for Ag, Al, As, Mn and Ti,
respectively, were achieved. The precision of the HF-LPME method estimated as the
relative standard deviation (RSD) for five replicate determinations of metal ions at a
concentration of 10 µg L-1 was less that 5%. The HF-LPME method was validated by
analysis of the target analytes in commercial diesel and gasoline samples. The accuracy of
the proposed method was confirmed by performing spike recovery experiments. The
recovery values (96-101%) indicated a satisfactory accuracy.
Keywords: Hollow fiber-liquid phase microextraction, ionic liquids, diesel, gasoline, trace
elements, chemometric optimization
13.1 INTRODUCTION
The determination of trace metal concentration in fuel samples is becoming
increasingly important in contamination monitoring and quality control studies. This is
because the presence of these elements in fuels can catalyze reactions responsible for
corrosion of engine parts, gum formation, fuel decomposition and catalyst poisoning.1,2
Therefore, depending on their concentration elements, poor engine performance and
increased levels of pollution can be observed.2 It is crucial, therefore, to accurately
determine trace element concentrations in fuels. These metals are normally present in trace
levels, therefore, their accurate quantification often requires sensitive and reliable
techniques. The latter includes inductively coupled plasma-optical emission spectrometry
265
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
(ICP OES), ICP-mass spectrometry (ICP-MS) and electrothermal atomic absorption
spectrometry (ETAAS). In spite of great improvements in the sensitivity of these
techniques, difficulties still lie in the analysis of trace elements because of their low
concentrations in the samples and the high complexity of the sample matrices.3,4 For
instance, inductively coupled plasma-based techniques have an advantage of multielement
detection capabilities. However, direct introduction of an organic matrix requires special
care, as the organic load may de-stabilize or extinguish the plasma.1 For these reasons,
sample introduction techniques (electrothermal vaporization) and pretreatment methods
(microwave digestion) have been developed and are reported in the literature.5,6 In
addition, separation and preconcentration of analytes prior to their ICP OES determination
has been reported.7 The benefit of using preconcentration techniques is that they combine
the advantages of separating the analyte from the complex matrix, by transferring it to an
aqueous phase and preconcentrating it at the same time.8
Hollow-fibre liquid phase microextraction (HF-LPME) is an attractive sample
preparation technique which allows extraction and preconcentration of analytes from
complex sample matrices with a high concentration factor. In addition, HF-LME is
qualified with rapid analysis time, simple setup and inexpensive.9,10 To achieve efficient
extraction, the selection of an appropriate extraction solvent is one of the important aspects
to be considered. According to Rasmussen and Pedersen-Bjergaard11 the solvent used
within the pores of the hollow fibre has to satisfy the following criteria: (i) it should be
immiscible with water to prevent leakage; (ii) it should be strongly immobilized in the
pores of the hollow fibre to prevent leakage and (iii) it should provide appropriate
extraction selectivity and high extraction recoveries.
Several
HF-LPME procedures involve the
use of organic solvents such
tetrachloromethane and toluene which are generally toxic and hazardous to organisms and
the environment. Recently, researchers focus have been on the replacement of traditional
organic solvents by alternative green ones such as room temperature ionic liquids
(RTILs).12,13 The RTILs are salts that are liquid over a wide temperature range including
room temperature and result from combination of organic cations with various anions.12,13
The unique physiochemical properties of RTILs such as insignificant vapor pressure, nonflammability as well as good extractability for various metal ions, make them very useful
for HF-LPME.13 Abulhassani and co-workers proposed the use of ionic liquid (1-hexyl-3methylimidazolium hexafluorophosphate, [C6MIM][PF6],) solvent for preconcentration of
266
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
lead and nickel from environmental and biological samples prior to determination by
electrothermal atomic absorption spectrometry.13
The use of chelating agents in HF-LMPE procedures allows various metal species to be
separated from other components in a sample. The extraction efficiency and selectivity of
the chelating agent can be affected by size of the chelate ring and type of its donor atoms,
oxidation state and size of the metal ion, and pH of the solvent system.12 Chelating agents
such as ammonium pyrrolidinecarbodithioate (APDC),10 8-hydroxyquinoline [14], 1phenyl-3-methyl-4-benzoyl-pyrazolone
(PMBP),15
benzoylacetone
(BZA),16
1-(2-
pyridylazo)-2-naphthol (PAN)17 and Kelex 100,18 among others, has been used for
complexation of metal ions in different analytical procedures.
Extraction and preconcentration of metal ions in organic matrices using HF-LPME is
difficult. This is because the organic phase in the lumen of the hollow fiber membrane is
miscible with donor phase (fuel sample). Therefore, sample pretreatment that will first
convert fuel sample to aqueous phase before it is subjected to HF-LPME system is
required. For this reason, diesel and gasoline samples were first digested using microwaveassisted digestion method. Thus, in this work, HF-LPME based room temperature ionic
liquid combined with APDC as a chelating agent was used for the extraction and
preconcentration of Ag, Al, As, Mn and Ti in diesel and gasoline samples.
In order to achieve the highest extraction of metal ions by HF-LPME system, the
optimization of several factors, such as sample solution pH, chelating agent concentration,
extraction time, and stripping solution concentration, among others, is required.
Optimizing these parameters using the conventional univariate procedure (one factor at a
time) is tedious and time consuming. In addition, this procedure requires quite a number of
experiments to be carried out in order to attain the best experimental condition.19,20
Therefore, multivariate procedures have been used to overcome the problems connected to
univariate techniques. The advantages of multivariate statistic techniques include reduction
in the number of required experiments, thus, resulting in lower reagent consumption and
significantly less laboratory work. Consequently, multivariate techniques are faster to
implement and more cost-effective than traditional univariate approaches. In addition,
multivariate statistic techniques allow the simultaneous study of several experimental
variables and the development of mathematical models that permit the assessment of the
relevance and statistical significance of factors being studied.19,20 Furthermore, these
techniques facilitate the evaluation of interaction effects between factors.19 Therefore, full
267
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
factorial and Box–Behnken designs were used for screening and optimization of factors
that influence preconcentration and stripping of the metal ions. These factors included
sample pH, concentration of the chelating agent, extraction time and stripping solution
concentration.
13.2 EXPERIMENTAL
13.2.1 Reagent and Standard Solutions
All reagents were of analytical grade unless otherwise stated and double distilled
deionised water (Millipore, Bedford, MA, USA) was used throughout the experiments.
Conostan custom made multi-element oil standard stock solutions (1.0 mg L-1) of Ag, Al,
As, Mn and Ti (SCP Science, Quebec, Canada) was used to prepare working solutions for
HF-LPME at concentrations of 10 µg L-1 for all target metal ions. Working solutions, as
per the experimental requirements, were freshly prepared from the stock solution for each
experimental run. The elemental standard solutions used for calibration were prepared by
diluting a Spectrascan multi-element standard stock (100 mg L-1) solution (Industrial
Analytical Pty Ltd, Johannesburg, South Africa). Ammonium pyrrolidinecarbodithioate
(Sigma-Aldrich, St. Loius, MO, USA) solution was prepared by dissolving the appropriate
amount of APDC in double distilled deionised water. Solutions of nitric acid were prepared
from ultrapure concentrated acid (65%, Sigma-Aldrich, St. Loius, MO, USA). The pH
adjustments were performed with nitric acid (Merck, Darmstadt, Germany) and sodium
hydroxide (Sigma-Aldrich, St. Loius, MO, USA) solutions. 1-butyl-3-methylimidazolium
hexafluorophosphate ([C6MIM][PF6]) ionic liquid was purchased from Sigma-Aldrich (St.
Loius, MO, USA). Synthetic gasoline was prepared by mixing 91% isooctane and 9% nheptane (Sigma Sigma-Aldrich, St. Loius, MO, USA).
13.2.2 Instrumentation
Analyte metal ions (Ag, Al, As, Mn and Ti) were determined using a Spectro Arcos
165 ICP OES (SPECTRO Analytical Instruments, GmbH, Germany) equipped with Cetac
ASX-520 autosampler. The operating conditions on the ICP OES spectrometer during the
measurements were as follows: forward power: 1400 W, plasma argon flow rate: 13 L min1
, auxiliary argon flow rate: 2.00 L min-1, nebulizer argon flow rate: 0.95 L min-1. The most
prominent atomic and ionic analytical lines of metal ions were selected for investigation,
268
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
that is, Ag 328.068 nm, Al 167.078 nm, As 193.759 nm, Cd 214.438, Mn 257.611 nm and
Ti 334.940. The polypropylene hollow fiber membrane used for HF-LPME was obtained
from Membrana (Wuppertal, Germany). The hollow fiber membrane had a wall thickness
of 200 µm, an inner diameter of 600 µm, and a pore size of 0.2 µm. A 25 µL Hamiltone
microsyringe (Bondaduz, Switzerland) was used to support the fiber and to introduce
organic solvent into the hollow fiber. Microwave assisted digestion was carried out in an
Ethos D (Milestone, Sorisole, Italy) with maximum pressure 1450 psi and maximum
temperature 300°C.
13.2.3 Preparation of the HF-LPME
The preparation of the HF-LPME procedure was adopted from Ghasemi et al.21
Describing the procedure briefly, the hollow fiber tube was cut into 8 cm portions and the
internal volume was about 20 µL. Each portion of fiber was used once for each treatment,
this was done in order to prevent the memory effect. The hollow fibers were sonicated for
5 min in acetone to remove any possible contamination and directly dried in air. One end
of the hollow fiber was sealed using tweezers, a hot plate and a soldering gun. The hollow
fiber membrane was submerged in the [C6MIM][PF6] ionic liquid for a few seconds to
impregnate the extractant onto the membrane pores of the hollow fiber wall. Then the 20
µL of [C6MIM][PF6] ionic liquid (inside the syringe) was injected into the hollow fiber in
order to fill up the inside of the tube completely. The fiber was then washed with water to
remove the excess ionic liquid from the surface of the hollow fiber membrane.
13.2.4 Extraction Procedure
The synthetic gasoline model samples were first mineralized using microwave assisted
digestion (MAD). The MAD procedure was carried out according to Kowalewska et al.22.
Briefly, 5.0 mL of the synthetic gasoline model sample containing 10 µg L-1 of the target
analytes, was placed into a Teflon vessel followed by 6 mL HNO3 (65%) and 2.0 mL H2O2
(30%). The vessels were inserted into a microwave unit and heated according to the
conditions recommended by the manufacturer. The digested content was left to cool down
to room temperature. After cooling, the vessels were opened and 2 mL of concentrated
HNO3 and 2 mL of hydrogen peroxide were added, and the heating program was repeated.
This step was done in order minimize incomplete mineralization of the organic matrix.
269
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
Finally, the Teflon vessel contents were cooled down to room temperature and
quantitatively transferred to a 100 ml calibration flask and made up to the mark using
double distilled deionised water. The same procedure was applied for the blank solutions
and real samples. It should be noted that for the real samples 1.0 mL was used instead of
5.0 mL.
The extraction procedure was carried out according to Xia et al.23 and Ghasemi et al.21
Aliquots of 20 mL of the digested samples containing different concentrations of APDC
(chelating agent) were placed into a 25 mL polypropylene sample bottles. It should be
noted that complexation is essential for the HF-LPME of trace elements to facilitate the
transfer of the target analytes to the acceptor phase.24 This implied that APDC was not only
used to form complexes with the target analytes, but also in promoting and facilitating the
HF-LPME process. The sample bottle was clamped in order to fix its position above a
magnetic stirrer. The prepared hollow fiber (Section 2.3) was immersed into the stirred
digested sample intended for analysis. Extraction of the target analytes from the sample
solution to the acceptor phase within the hollow fiber membrane was carried out over a
period of 10–60 min under a magnetic rotation speed of 900 rpm. After extracting for an
appropriate time, the magnetic stirrer was switched off and the microsyringe containing the
hollow fiber was then removed from the sample bottle. Finally, the acceptor phase was
withdrawn into the microsyringe and the hollow fiber was discarded. The target analytes
extracted in the ionic liquid-phase were then transferred to an aqueous phase by adding 5.0
mL of stripping solution (different nitric acid concentrations). The mixture was sonicated
for 10 min and the two phases were separated by centrifuging the mixture for 2 min at
1200 rpm. The upper phase (nitric acid) was collected to determine the metal concentration
using ICP OES.
13.2.5 Optimization Strategy
Chemometric optimization of the HF-LPME preconcentration system was carried out
considering four variables, namely sample pH, concentration of the chelating agent
([APDC]), extraction time (ET) and stripping solution concentration ([HNO3]). The
optimization was carried out by using the multivariate approach. Firstly, a screening of the
influential variables on the analytical response was tested by employing a two-level (24)
full factorial design with a central point. Table 13.1 lists the upper and lower values given
270
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
to each factor. The second step of the optimization strategy involved the application of a
RSM based on a Box–Behnken design. The latter was applied in the optimization of the
level of the variables that were considered as significant according to the results obtained
in 24 full factorial design. The design of experiments was performed using Minitab 16 and
Design Expert 8.0.7.1 Software programs.
Table 13.1. Factors and levels used in 24 factorial design for extraction and
preconcentration of metal ions in fuel samples
Variable
pH
[APDC] (%)
ET (min)
[HNO3] (mol L-1)
Low level (-1)
2.0
0.5
10
0.5
Central point (0)
5
3.25
35
1.75
High level (+1)
8
6.0
60
3.0
13.3. RESULTS AND DISCUSSION
13.3.1 Chemometric Optimization of HF-LPME
In order to achieve the highest extraction of metal ions by HF-LPME system, the
optimization of several factors, such as sample solution pH, chelating agent concentration,
extraction time, and stripping solution concentration, was carried out using factorial and
Box-Behnken designs.
13.3.1.1 Factorial Design
A two-level full factorial design, 24, involving 20 runs was used for evaluation of the
significance of sample pH, chelating agent concentration, extraction time and stripping
solution concentration on extraction of metal ions. Table 13.2 shows the experimental
design matrix and the analytical results obtained in each run expressed as average
percentage recovery. Analysis of variance (ANOVA) and p-values were used to investigate
the significance of the effects on the HF-LPME system. The Pareto chart of main effects
and their interactions produced from ANOVA results are shown in Figs. 1-3.
271
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
Table 13.2. List of experiments in the factorial design (actual values) for HF-LPME
optimization and the response values
Runs
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
pH
2
8
2
8
2
8
2
8
2
8
2
8
2
8
2
8
5
5
5
5
[APDC] ET
0.5
0.5
6
6
0.5
0.5
6
6
0.5
0.5
6
6
0.5
0.5
6
6
3.25
3.25
3.25
3.25
10
10
10
10
60
60
60
60
10
10
10
10
60
60
60
60
35
35
35
35
EC
Ag
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
3
3
3
3
3
3
3
3
1.75
1.75
1.75
1.75
53.8
48.9
71.3
63.9
82.2
70.9
92.1
78.9
54.1
47.8
77.6
64.9
83.3
72.1
93.9
80.2
95.9
96.1
95.8
95.7
Al
63.1
56.7
78.3
70.8
85.2
75.3
93.4
83.4
62.9
54.8
77.6
71.3
86.4
74.8
94.1
82.7
97.1
96.6
97.4
96.6
As
Mn
Recovery (%)
71.3
68.7
48.1
56.3
83.4
77.8
51.2
63.2
93.1
85.4
68.7
71.5
95.4
86.1
71.6
70.6
72.3
67.5
48.4
55.6
82.9
80.7
50.9
54.3
94.3
87.6
67.6
69.6
95.1
88.0
72.5
67.9
98.3
97.5
98.1
97.3
97.6
97.6
97.5
97.8
Ti
75.8
51.3
91.4
64.1
95.6
81.2
97.2
79.4
74.5
50.5
89.2
65.6
94.8
77.3
96.5
81.7
95.8
96.3
96.1
95.7
Silver and aluminium: The results of factorial design for Ag and Al are presented in
Fig 13.1A and B. The Pareto chart (Fig 13.1 A) for preconcentration of Ag shows that the
factors pH (-10.09), [APDC] (13.71), ET (21.41) and [HNO3] (1.48) as well as
pH×[APDC] (-1.66) and pH×ET (-2.26) interactions were significant at 95% confidence
level. For extraction and preconcentration of Al, pH (-8.90), [APDC] (11.50), ET (17.48)
as well as [APDC]×ET (-3.58) and pH×ET (-1.83) presented a probability (p) that was
lower than 0.05 and they were significant at 95% confidence level. Stripping solution
concentration and interactions (pH×[APDC], ET×[HNO3], [APDC] ×[HNO3] and
pH×[HNO3]) were not significant.
272
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
Fig. 13.1. Pareto charts of standardized effects for variables in the Ag and Al
preconcentration.
Arsenic and manganese: The main effects and their interactions for As and Mn are
shown in Fig. 13.2. It can be seen from this figure that sample pH, chelating agent
concentration and extraction time were statistically significant at the 95% confidence level
for both As and Mn preconcentration. In the case of As, the main effect values were 26.10, 4.90 and 18.72 for pH, [APDC] and ET, respectively. For Mn, the main effect
values were -16.60, 3.30 and 12.83 for pH, [APDC] and ET, respectively. In addition, for
preconcentration of Mn, the interaction between APDC concentration and extraction time
(-2.55) was also significant.
Fig. 13.2. Pareto charts of standardized effects for variables in the As and Mn
preconcentration.
273
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
Titanium: The ANOVA results obtained using the factorial design (Fig. 13.3) for Ti
demonstrated that pH (-20.49), [APDC] (8.01) and ET (17.66) as well as [APDC]×ET (6.54) and pH×ET (4.36) are significant at the 95% confidence level.
Fig. 13.3. Pareto charts of standardized effects for variables in the Ti preconcentration.
The overall results from Figs. 13.1-3 indicated that sample pH, chelating agent
concentration and extraction time were more significant at 95% for the studied metal ions.
It was observed that the algebraic sign for sample pH effect was negative while the effect
values for other variable were positive. The negative effect demonstrated that sample pH
lower than the maximum levels should be used. In addition, the positive effect indicated
that APDC concentrations and extraction times higher than the minimum levels must be
used for satisfactory extraction and preconcentration of metal ions. The stripping solution
concentration showed poor influence on the analytical response and was fixed at 1.75 mol
L−1. It should be noted that higher concentrations of nitric acid were not used due to the
possibility of PF6- decomposition. Taking into consideration the importance of factors pH,
[APDC] and ET within the experimental domain, response surface methodology based on
Box–Behnken design was used for final optimization.
274
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
13.3.1.2 Box–Behnken Design
The sample pH, APDC concentration and extraction time were simultaneously
optimized using the data obtained from 18 sets of experiments carried out according to the
Box–Behnken technique. The latter was chosen because of the reduced number of
experiments that need to be carried out. In addition, the results of this reduced number of
experiments provide a statistical model that is used to identify variables that lead to
quantitative retention of metal ions. Percentage recoveries of each analyte were
investigated as the response function of the Box–Behnken design model in order to
optimize the aforementioned variables. The levels (actual values) of the abbreviated
experimental variables and the respective response values for each analyte are presented in
Table 13.3. The experimental runs were randomized in order to minimize the effect of
uncontrolled factors.
Table 13.3. Experimental design using Box–Behnken design (CCD) and analytical
response values
Run no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
pH
2
8
2
8
2
8
2
8
5
5
5
5
5
5
5
5
5
5
[APDC] ET
0.5
0.5
6
6
3.25
3.25
3.25
3.25
0.5
6
0.5
6
3.25
3.25
3.25
3.25
3.25
3.25
35
35
35
35
10
10
60
60
10
10
60
60
35
35
35
35
35
35
Ag
Al
74.63
70.01
84.02
67.51
57.21
48.2
98.7
85.14
63.12
54.22
89.12
100.4
98.85
98.66
98.61
98.80
98.69
98.67
Recovery (%)
78.34
73.38 70.44
74.41
67.3
80.06
82.49
84.03 84.13
68.25
73.44 92.31
59.86
53.74 63.93
51.36
45.89 90.06
97.89
95.12 101.2
87.11
88.01 96.36
67.23
57.34 67.55
53.43
55.81 78.14
93.48
98.54 89.17
100.92
101.3 101.88
98.04
97.45 99.78
98.09
98.48 100.15
98.16
97.51 99.74
98.18
97.45 99.71
98.19
97.52 100.38
98.25
97.39 99.78
275
As
Mn
Ti
82.46
69.15
89.33
74.28
66.65
51.18
98.44
86.29
68.23
64.45
92.84
100.88
98.56
98.61
98.63
98.58
98.55
98.63
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
The analysis of variance (ANOVA) was used to evaluate the significance of the model
equation and related terms. The ANOVA results were not included for simplicity purposes.
The model equation and related terms were considered to be significant if F-values were
less than 0.05 (p-value at 95% confidence level). The overall ANOVA results for all the
metal ions demonstrated that the models, sample pH, extraction as well as their squares
were significant at 95% confidence level. Therefore, the experimental relationship between
each metal ion recovery and the three factors (pH, [APDC] and ET) in actual units
obtained by the application of response surface methodology was expressed in terms of
Eqs.1-5.
% Re cov ery( Ag )   2.10  16.04 A  8.37 B  1.92 C  0.36 AB
 1.52  10 2 AC  7.34  10  2 BC  1.62 A 2
(1)
 1.34 B 2  1.90  10  2 C 2
% Re cov ery( Al )   13.22  14.71 A  6.03 B  1.71C  0.31 AB
 7.60  10 3 AC  7.72  10  2 BC  1.50 A 2
2
 1.16 B  1.70  10 C
2
(2)
2
% Re cov ery( As )   4.43  16.07 A  7.63 B  2.09 C  0.14 AB
 2.47  10 3 AC  1.56  10 2 BC  1.70 A 2
(3)
 1.03B 2  1.86  10 2 C 2
% Re cov ery( Mn)   6.22  13.44 A  11.60 B  1.47 C  4.36  10 2 AB
 0.10 AC  7.71  10 3 BC  0.81A 2
(4)
 1.45 B 2  7.67  10 3 C 2
% Re cov ery(Ti)  26.24  11.76 A  5.44 B  1.57 C  5.27  10 2 AB 
1.11  10 2 AC  4.28  10 2 BC  1.43 A 2
2
 0.91 B  1.61 10 C
2
(5)
2
In Eqs (1)-(5), A, B and C correspond to independent variables of sample pH, APDC
concentration and extraction time respectively, while the terms AB, BC and AC
corresponds to the interactions of the variables. The algebraic sign of a coefficient (+ or -)
defines the direction of relationship between the related effect and the response. For
instance, the positive sign indicates that as the value of one effect changes, the value of the
response changes in the same direction too, whereas for the negative sign on the response
value, operates in the opposite direction. In addition, the absolute value of the coefficients
measures the strength of the relationship.25
276
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
Fig. 13.4 shows the three dimensional (3D) response surface plots for all the studied
metal ions. The 3D plots were used to establish the individual and collective effect of the
variables and the reciprocal interaction between them. It should be noted that the 3D
response surface plots were constructed as a function of two factors (namely pH and
extraction time). The APDC concentration was maintained at a fixed level (central point,
3.75%). This was done in order to understand the main effects and interaction effects of the
two factors (sample pH and extraction time). The reason why APDC concentration was
fixed at the central level is that it had a minor effect (-1.05 to 2.25) for almost all the
analytes (except Mn) compared to sample pH (-3.95 to -7.00) and extraction time (15.9921.25). It can be seen from Fig.13.4 that the results for each metal ion had maximum points
within the studied experimental domain. From these observations, it was possible to
establish that the target analytes can be preconcentrated simultaneously in one common
region.
The derivatization of the general equations (Eqs. 1-5) as (pH), ([APDC]) and (ET)
resulted in three new equations for each Eqs. 1-5 (the new equations are not shown). The
critical points (optimum conditions) for HF-LPME system in the response surface model
were estimated by solving the derivative equations; when
 ( R)
 ( R)
0,
 0,
 ([ APDC ])
 ( pH )
 ( R)
 0 . The critical points were calculated according to the methods reported by dos
 ( ET )
Santos et al.26 and Souza et al.27 The calculated values for the critical points for each metal
ion are tabulated in Table 13.4.
277
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
Fig. 13.4. Response surfaces obtained for (A) Ag, (B) Al, (C) As, (D) Mn and (E) Ti after
extraction and preconcentration by HF-LPME
Table 13.4. Calculated critical point values
Variables
pH
[APDC] (%)
ET (min)
Ag
4.26
4.09
55.81
Al
4.36
3.91
57.88
Analytes
As
4.50
3.84
58.00
Mn
4.40
4.09
64.21
Ti
4.27
4.24
55.50
In view of the results in Table 13.4, the optimum conditions recommended for
simultaneous extraction and preconcentration of target analytes were: sample pH 4.5,
APDC concentration 4.0 %, extraction time 55 min and stripping solution concentration
1.75 mol L-1. In order to investigate whether quantitative extraction and recovery of the
278
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
target analytes could be achieved when using the recommended optimum conditions, five
sets of experiments were carried out. The analytical results showed that quantitative
extraction and recovery of the analytes of interest can be achieved at these conditions.
13.3.2 Interference Studies
Metal ions such as alkali and alkaline earth elements as well as transition metals are
common constituents of liquid fuel samples. This might be due to the fact that they exist
naturally in the earth’s crust. It should be noted that the effect of alkali and alkaline earth
elements was not examined due to the non-selectivity of the chelating agent (APDC) used
towards the alkali and alkaline earth metals.19 This is because, trace elements present in the
liquid fuel samples can compete for complexation with APDC. This can interfere with the
target analytes thus leading to reduced extraction efficiency of the HF-LPME system.
Therefore, the effects of various metal ions on extraction and preconcentration of the target
analytes were evaluated under the optimized conditions. The tolerance limit was defined as
the interfering ions concentration that causes a relative error smaller than or equal to 5%
with respect to the preconcentration of analytes. Under the optimized conditions, the effect
of other metal ions on the recovery of Ag, Al, As, Mn and Ti (at a concentration of 10 µg
L-1 for each target metal ion), was investigated. The analytical results obtained showed that
the recoveries for the target analytes could remain above 95% even in the presence of 100
µg L-1 Fe, Zn, Cd, Co, Cu, Pb and Ni respectively. This demonstrates that the developed
HF-LPME method has a good tolerance to matrix interference.
13.3.3 Analytical Figure of Merit
Under the optimized conditions, the analytical figures of merit of HF-LPME described
in this study were investigated. The calibration curves were obtained after sequences of
standard solutions were subjected to the HF-LPME procedure and the resulting
concentration was determined by ICP OES. A relatively good linearity was obtained over a
concentration range of 0.040–50 µg L−1 with the correlation coefficient (R2) ranging from
0.9955-0.9978 for all the studied analytes. The preconcentration factors (PF) defined as the
ratio of ICP OES signals after and before extraction,21 sensitivity, limits of detection
(LOD), limits of quantification (LOQ) as well as precision estimated in terms of relative
279
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
standard deviation, are listed in Table 13.5. The instrument detection limits (IDL) for Cd,
Cu, Fe Pb, and Zn can seen in Chapters 6 and 8 (Table 6.4 and Section 8.3.4).
Table 13.5. Analytical figure of merits for the proposed HF-LPME method
Analytes
Sensitivity
(L µg-1)
PF
LOD (µg L-1)
218.3
228.9
201.6
257.6
237.8
150
291
112
405
367
0.08
0.06
0.09
0.04
0.05
Ag
Al
As
Mn
Ti
LOQ (µg L-1) Precision (%RSD)
0.27
0.18
0.29
0.15
0.17
4.5
3.2
4.1
1.5
1.3
13.3.4 Validation and Application of the Proposed Method
The accuracy of the HF-LPME method was assessed by performing a recovery tests.
The latter was achieved by adding 20 µg L-1 of each analyte to the digested diesel and
gasoline samples. The analytical results are presented in Table 13.6. As shown in this table,
satisfactory recoveries were obtained for all analytes and they ranged between 96 and
101%. The analytical results of recovery test confirmed the validity of the proposed HFLPME method.
Table 13.6. Analytical results obtained in the analysis of spiked diesel and gasoline
samples. The concentration and recovery values are expressed as the mean ± standard
deviation of the three replicates
Analyte Added (µg L1)
Ag
Al
As
Mn
Ti
0
20
0
20
0
20
0
20
0
20
Diesel
Found (µg L-1)
25.4±0.6
44.7±1.6
141±1
160±2
ND
19.4±0.9
28.3±0.4
48.0±1.1
ND
19.8±0.2
Recovery (%)
96.7±1.2
98.3±0.9
97.1±2.1
98.6±1.4
99.2±0.6
280
Gasoline
Found (µg L-1)
28.7±0.7
48.4±1.8
91.5±1.7
111±2
ND
19.3±1.2
25.6±0.8
45.5±1.4
ND
20.1±0.7
Recovery (%)
98.4±2.4
97.9±1.7
96.5±1.5
99.5±2.3
101±1
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
The optimized HF-LPME method was applied to the simultaneous extraction and
preconcentration of Ag, Al, As, Mn and Ti in diesel and gasoline samples. Based on the
analytical results obtained (Table 13.7), it was observed that elements such as As (all
sample) and Ti (in D2 and G2 samples) were undetectable (below LOD) while Ag, Al and
Mn, were present in all the samples. Aluminium was the most concentrated element in G1
followed by D1 and G2 had the lowest concentration. It is worth mentioning that Al is
quite abundant in the Earth's crust and also sample contamination during the diesel and
gasoline production process should not be ignored. The Ag and Mn concentrations ranged
from 25.4-97.6 µg L-1 and 25.6-88.6 µL-1, respectively. Titanium was found to be more
concentrated in D1 sample. Titanium is normally used as catalyst in the desulfurisation of
diesel oils to minimise sulfur dioxide emission from fuel combustion. Therefore, the
relative high concentration in D1 samples might be due to residues of Ti leached out
during desulfurisation process.
Table 13.7. The determination of Ag, Al, As, Mn and Ti in diesel and gasoline samples
using HF-LPME/ICP OES
D1
D2
G1
G2
-1
73.5±3.5
272±2
NDa
88.6±0.2
206±2
Ag
Al
As
Mn
Ti
a
Concentration (µg L )
25.4±0.6
97.6±1.7
141±1
973 ±4
ND
ND
28.3±0.4
85.0±0.9
ND
9.8±0.7
28.7±0.7
91.5±1.7
ND
25.6±0.8
ND
Not detectable
In order to test the accuracy of the proposed procedure, the results obtained were
compared with those obtained by ICP OES technique without HF-LPME pretreatment
(Table 13.8). According to the paired-t student test, the results obtained by the two
methodologies (optimized HF-LPME/ICP OES method and MAD/ICP OES method) were
not significantly different at 95% confidence level. These results confirmed the reliability
of the proposed HF-LPME/ICP OES method.
281
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
Table 13.8. The determination of Ag, Al, As, Mn and Ti in diesel and gasoline samples
using MAD/ICP OES
D1
D2
G1
G2
-1 a
Ag
Al
As
Mn
Ti
a
74.3±4.1
272±4
NDb
89.4±1.3
205±2
Concentration (µg L )
25.4±1.1
98.0 ±2.2
141.2±2.4
973±6
ND
ND
27.9±0.4
85.4±1.5
ND
10.2±1.3
29.3±1.8
91.3±2.1
ND
26.3±1.4
ND
Metal ion concentration obtained when applying standard addition method; b Not detectable
13.4 CONCLUSION
Multivariate optimization of the proposed method, based on HF-LPME, for
preconcentration of metal ions in diesel and gasoline samples, was achieved by factorial
and Box–Behnken designs. The analytical results obtained demonstrated that the HFLPME/ICP OES system for the preconcentration and determination of Ag, Al, As, Mn and
Ti exhibited a number of good features; these include relatively high enrichment factors,
sensitivity, simple operation, cost-effectiveness and non-consumption of organic toxic
solvents. The proposed method showed good precision (1.3–4.5%) as well as relatively
low LOD (0.04-0.09 µg L-1) and LOQ (0.15-0.29 µg L-1). Validation by ICP OES after
microwave digestion and spiked sample analysis confirmed the accuracy of the HFLPME/ICP OES method.
13.5 REFERENCES
1. Saint'pierre, T. D., Dias, L. F., Pozebon, D., Aucélio, R. Q., Curtius, A. J. & Welz, B.
2002. Determination of Cu, Mn, Ni and Sn in gasoline by electrothermal vaporization
inductively coupled plasma mass spectrometry, and emulsion sample introduction.
Spectrochimica Acta Part B: Atomic Spectroscopy, 57, 1991-2001.
2. Donati, G. L., Amais, R. S., Schiavo, D. & Nobrega, J. A. 2013. Determination of Cr,
Ni, Pb and V in gasoline and ethanol fuel by microwave plasma optical emission
spectrometry. Journal of Analytical Atomic Spectrometry, 28, 755-759.
3. Yin, J., Jiang, Z., Chang, G. & Hu, B. 2005. Simultaneous on-line preconcentration and
determination of trace metals in environmental samples by flow injection combined
with inductively coupled plasma mass spectrometry using a nanometer-sized alumina
packed micro-column. Analytica Chimica Acta, 540, 333-339.
282
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
5. Saint'pierre, T. D., Frescura, V. L. A. & Curtius, A. J. 2006. The development of a
method for the determination of trace elements in fuel alcohol by ETV-ICP-MS using
isotope dilution calibration. Talanta, 68, 957-962.
4. Yang, G., Fen, W., Lei, C., Xiao, W. & Sun, H. 2009. Study on solid phase extraction
and graphite furnace atomic absorption spectrometry for the determination of nickel,
silver, cobalt, copper, cadmium and lead with MCI GEL CHP 20Y as sorbent. Journal
of Hazardous Materials, 162, 44-49.
6. Pereira, J. S. F., Moraes, D. P., Antes, F. G., Diehl, L. O., Santos, M. F. P., Guimarães,
R. C. L., Fonseca, T. C. O., Dressler, V. L. & Flores, É. M. M. 2010. Determination of
metals and metalloids in light and heavy crude oil by ICP-MS after digestion by
microwave-induced combustion. Microchemical Journal, 96, 4-11.
7. Nomngongo, P. N., Ngila, J. C., Kamau, J. N., Msagati, T. A. M. & Moodley, B. 2013.
Preconcentration of molybdenum, antimony and vanadium in gasolsine samples using
Dowex 1-x8 resin and their determination with inductively coupled plasma–optical
emission spectrometry. Talanta, 110, 153-159.
8. Korn, M. D. G. A., Dos Santos, D. S. S., Welz, B., Vale, M. G. R., Teixeira, A. P.,
Lima, D. D. C. & Ferreira, S. L. C. 2007. Atomic spectrometric methods for the
determination of metals and metalloids in automotive fuels - A review. Talanta, 73, 111.
9. Pena-Pereira, F., Lavilla, I. & Bendicho, C. 2009. Miniaturized preconcentration
methods based on liquid-liquid extraction and their application in inorganic ultratrace
analysis and speciation: A review. Spectrochimica Acta Part B: Atomic Spectroscopy,
64, 1-15.
10. Li, L. & Hu, B. 2007. Hollow-fibre liquid phase microextraction for separation and
preconcentration of vanadium species in natural waters and their determination by
electrothermal vaporization-ICP OES. Talanta, 72, 472-479.
11. Rasmussen, K. E. & Pedersen-Bjergaard, S. 2004. Developments in hollow fibrebased, liquid-phase microextraction. TrAC Trends in Analytical Chemistry, 23, 1-10.
12. Lertlapwasin, R., Bhawawet, N., Imyim, A. & Fuangswasdi, S. 2010. Ionic liquid
extraction of heavy metal ions by 2-aminothiophenol in 1-butyl-3-methylimidazolium
hexafluorophosphate and their association constants. Separation and Purification
Technology, 72, 70-76.
13. Abulhassani, J., Manzoori, J. L. & Amjadi, M. 2010. Hollow fiber based-liquid phase
microextraction using ionic liquid solvent for preconcentration of lead and nickel from
environmental and biological samples prior to determination by electrothermal atomic
absorption spectrometry. Journal of Hazardous Materials, 176, 481-486.
14. Es’haghi, Z. & Azmoodeh, R. 2010. Hollow fiber supported liquid membrane
microextraction of Cu2+ followed by flame atomic absorption spectroscopy
determination. Arabian Journal of Chemistry, 3, 21-26.
283
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
15. Wu, Y., Hu, B., Peng, T., Liao, Z. & Jiang, Z. 2001. Electrothermal volatilization of
aluminum as 1-phenyl-3-methyl-4-benzoylpyrazolone[5] chelate for gaseous sample
introduction in ICP-AES. Talanta, 55, 841-845.
16. Xia, L., Hu, B., Jiang, Z., Wu, Y. & Liang, Y. 2004. Single-drop microextraction
combined with low-temperature electrothermal vaporization ICPMS for the
determination of trace Be, Co, Pd, and Cd in biological samples. Analytical Chemistry,
76, 2910-2915.
17. Xia, L., Li, X., Wu, Y., Hu, B. & Chen, R. 2008. Ionic liquids based single drop
microextraction combined with electrothermal vaporization inductively coupled plasma
mass spectrometry for determination of Co, Hg and Pb in biological and environmental
samples. Spectrochimica Acta Part B: Atomic Spectroscopy, 63, 1290-1296.
18. Bautista-Flores, A. N., Rodríguez De San Miguel, E., De Gyves, J. & Jönsson, J. Å.
2010. Optimization, evaluation, and characterization of a hollow fiber supported liquid
membrane for sampling and speciation of lead(II) from aqueous solutions. Journal of
Membrane Science, 363, 180-187.
19. Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S. & Escaleira, L. A. 2008.
Response surface methodology (RSM) as a tool for optimization in analytical
chemistry. Talanta, 76, 965-977.
20. Tarley, C. R. T., Silveira, G., Dos Santos, W. N. L., Matos, G. D., Da Silva, E. G. P.,
Bezerra, M. A., Miró, M. & Ferreira, S. L. C. 2009. Chemometric tools in
electroanalytical chemistry: Methods for optimization based on factorial design and
response surface methodology. Microchemical Journal, 92, 58-67.
21. Ghasemi, E., Najafi, N. M., Raofie, F. & Ghassempour, A. 2010. Simultaneous
speciation and preconcentration of ultra traces of inorganic tellurium and selenium in
environmental samples by hollow fiber liquid phase microextraction prior to
electrothermal atomic absorption spectroscopy determination. Journal of Hazardous
Materials, 181, 491-496.
22. Kowalewska, Z., Ruszczyńska, A. & Bulska, E. 2005. Cu determination in crude oil
distillation products by atomic absorption and inductively coupled plasma mass
spectrometry after analyte transfer to aqueous solution. Spectrochimica Acta Part B:
Atomic Spectroscopy, 60, 351-359.
23. Xia, L., Y. Wu And B. Hu 2007. Hollow-fiber liquid-phase microextraction prior to
low-temperature electrothermal vaporization ICP-MS for trace element analysis in
environmental and biological samples. Journal of Mass Spectrometry, 42, 803-810.
24. Molaakbari, E., Mostafavi, A. & Afzali, D. 2011. Ionic liquid ultrasound assisted
dispersive liquid–liquid microextraction method for preconcentration of trace amounts
of rhodium prior to flame atomic absorption spectrometry determination. Journal of
Hazardous Materials, 185, 647-652.
284
Chapter 13: HF-LPME for extraction and preconcentration of trace elements
25. Sereshti, H., Entezari Heravi, Y. & Samadi, S. 2012. Optimized ultrasound-assisted
emulsification microextraction for simultaneous trace multielement determination of
heavy metals in real water samples by ICP OES. Talanta, 97, 235-241.
26. dos Santos, W. L., Dos Santos, C. M. M., Costa, J. L. O., Andrade, H. M. C. &
Ferreira, S. L. C. 2004. Multivariate optimization and validation studies in on-line preconcentration system for lead determination in drinking water and saline waste from oil
refinery. Microchemical Journal, 77, 123-129.
27. Souza, A. S., dos Santos, W. N. L. & Ferreira, S. L. C. 2005. Application of Box–
Behnken design in the optimisation of an on-line pre-concentration system using
knotted reactor for cadmium determination by flame atomic absorption spectrometry.
Spectrochimica Acta Part B: Atomic Spectroscopy, 60, 737-742
285
CHAPTER FOURTEEN:
GENERAL CONCLUSIONS AND RECOMMENDATIONS
14.1 GENERAL CONCLUSION
The purpose of the study was to develop preconcentration methods for separation and
enrichment of trace Ag, Al, As, Cd, Co, Cr, Cu, Fe, Mn, Mo, Ni, Pb, Sb, Ti, V and Zn in
organic matrices prior to their ICP OES/MS determination. The benefit of using
preconcentration techniques prior to ICP OES/-MS is that they combine the advantages of
separating the analyte from the complex organic matrix by transferring it to an aqueous
phase, and preconcentrating it at the same time. Therefore, applicability of solid phase
extraction, solid phase microextraction and hollow fiber-liquid phase microextraction for
separation and preconcentration of trace elements in organic based samples was evaluated.
A summary of conclusions that support the main objectives of each investigation is given
below:
In solid phase extraction, performance of commercially ion exchange resins (Dowex
50W-x8, Chelex-100, Dowex 1-x8 and Dowex MAC-3), electrospun cellulose-g-oxolane2,5-dione nanofibers, nanometer-sized alumina for simultaneous preconcentration of trace
elements in short chain alcohols and liquid fuel, was evaluated. The kinetics and
equilibrium studies for the removal of metal ions using Dowex 50W-x8 cation exchange
resin was also performed. The results obtained for kinetics and equilibrium studies
indicated that Dowex 50W-x8 resin is suitable for removal of trace elements in ethanol
sample. In addition, it was concluded that ion exchange resins can be applied for solid
phase extraction of trace elements in organic matrices. In view of the information obtained
from the kinetics and equilibrium studies, Dowex 50W-x8, Chelex-100 and Dowex MAC3 were applied for the separation and preconcentration of Cd, Cr, Cu, Fe, Mn, Pb, Ti and
Zn in short chain alcohols. Under optimized experimental conditions, the analytical results
obtained showed that Dowex 50W-x8 resin was suitable for simultaneous separation and
preconcentration of eight metal ions in organic matrix. The optimized Dowex 50W-x8
solid phase extraction method displayed relatively good analytical performances. The
applicability of the method in real sample was also examined and the results obtained were
comparable with those obtained with an independent procedure.
Dowex 1-x8 (anion exchange) resin was used for preconcentration of Mo, Sb and V
anionic compound ions in gasoline. The Dowex 1-x8 SPE method was successful in pre286
Chapter 14:
General conclusions and recommendations
concentrating metal ions from large sample volume with a preconcentration factor of 120.
The positive features of the SPE method included relatively high selectivity, good
precision and accuracy as well as relatively low LOD and LOQ. The Dowex 1-x8 SPE
method was applied for simultaneous determination of trace amounts (µg L-1 range) of Mo,
Sb and V in gasoline samples. Taking into consideration the fact that trace elements in fuel
samples may occur in different forms, a dual-bed SPE column system was evaluated for
preconcentration of trace metals and metalloids in gasoline samples. A full factorial design
was used for the establishment of optimum conditions for separation and pre-concentration
of the analytes in gasoline sample. The dual-bed SPE method combined relatively low
LOD and LOQ values with higher sample throughput.
Synthetic adsorbents such as electrospun cellulose-g-oxolane-2,5-dione nanofibers and
nanometer-sized alumina were also evaluated for separation and preconcentration of metal
ions in gasoline samples. The solid phase extraction of Cd, Cu, Fe, Pb and Zn in gasoline
samples using cellulose-g-oxolane-2,5-dione performed efficiently, resulting in reasonably
analytical figures of merits such as relatively high preconcentration factor, low LOD,
LOQ, good precision and accuracy. A SPE method for simultaneous preconcentration of
trace metal ions in gasoline samples using nanometer-sized alumina prior to ICP-MS
determination was also investigated. Under optimized conditions, solid phase extraction of
metal ions occurred efficiently, resulting in a reasonably high preconcentration factor, low
LOD and LOQ. The developed nanometer-sized alumina SPE was found to be simple,
cheap, efficient, precise and accurate.
Solid phase microextraction was also used for extraction and enrichment of metal ions
in diesel samples using two approaches, namely, hollow fiber solid phase microextraction
and membrane solid phase microextraction. A hollow fiber solid phase microextraction
system based on fiber-supported sol-gel combined with cation exchange resin for
preconcentration of trace metal ions in liquid fuel samples prior to their ICP-MS
determination was developed. The two-level full factorial and central composite designs
were used for the optimization of experimental conditions. The accuracy of the proposed
method was confirmed by analysis of the spiked diesel and gasoline samples. The
measured concentrations were in good agreement at 95% confidence level with the added
values. The precision, expressed as relative standard deviation, was less than or equal to
3%. The optimized method was applied for simultaneous preconcentration of the target
analytes in commercial diesel and gasoline samples.
287
Chapter 14:
General conclusions and recommendations
Membrane solid phase microextraction (MSPME) based on titania-alumina hollow
fiber membrane was used for the simultaneous separation and preconcentration of Co, Cr,
Mo, Ni, Sb and V in diesel and gasoline samples prior to ICP-MS determination. The
preconcentration step permitted the elimination of the organic matrix, thus, avoiding the
need for digestion of the samples before metal ion determination. The developed method
was applied for the determination of the target analytes in liquid fuel samples. The
developed MSPME method displayed relatively low LOD and LOQ.
Hollow fiber solid phase microextraction (HF-LPME) based on ionic liquids was used
for extraction preconcentration of Ag, Al, As, Mn and Ti in diesel and gasoline samples.
The optimization of the HF-LPME system was achieved by factorial and Box–Behnken
designs. The advantages of the developed method included relatively high enrichment
factors, sensitivity, simple operation, cost-effective and non-consumption of organic toxic
solvents. The HF-LPME method showed improved precision as well as relatively low
LOD values.
The separation and preconcentration techniques developed in this work displayed some
advantages and disadvantages. For instance, the advantages included sample matrix
elimination, relatively low LOD and LOQ, high accuracy and precision and high sample
throughput, among others. The disadvantages included the use high sample volumes, long
analysis times (in the case of HF-LPME). For example, extraction and pre-concentration of
metal ions in organic matrices using HF-LPME is difficult. Therefore samples were first
digested before being subjected to the HF-LPME system. However, in view of the above
conclusions drawn from each preconcentration technique and also taking into consideration
some factors such as economy, sample throughput, easy implementation and operation,
sensitivity, precision, accuracy, and broad applicability, solid phase extraction was found
to be the best method for fuel analysis by either ICP-MS or ICP OES. This is because SPE
had many advantages over solid phase microextraction and HF-LPME. These included,
simplicity, flexibility, rapidity, higher enrichment factors, use of different solid materials
and absence of emulsion. Like any other analytical techniques SPE had some
disadvantages such as use of high volumes and contamination possibilities and these
challenges can be overcome by on-line SPE mode. The latter allows sample manipulation
between the pre-concentration and analysis steps so that analyte losses as well as the risk
of contamination are minimized.
288
Chapter 14:
General conclusions and recommendations
14.2 RECOMMENDATIONS
The preconcentration techniques developed in this study were suitable for separation
and enrichment of metal ions in organic matrices. In addition, the developed methods
presented similar or even better precision, LOD and LOQ when compared with other
methods reported in the literature. Furthermore, the current study provided information on
concentrations of trace elements in diesel and gasoline obtained from different filling
stations in Johannesburg (South Africa) which can be regarded as reference ranges.
Therefore, this study recommends these preconcentration techniques for routine and rapid
analysis of trace metals and metalloids in organic matrices. Apart from the aforementioned
features, there is a possibility of coupling these techniques with inductively coupled
plasma based techniques. On-line procedures allow sample manipulation between the
preconcentration and analysis steps, so that analyte losses as well as the risk of
contamination, are minimized.
The possible future work could include the following:
i) Designing an on-line separation, pre-concentration and determination of elements in fuel
samples, using solid phase extraction, hollow fibre-liquid phase microextraction and
hollow fibre-solid phase microextraction coupled to ICP-based techniques.
ii) Speciation analysis: since most metal ions are present in trace levels in fuel metals, they
are probably associated with ligands in different forms (organometallic and inorganic).
Therefore, speciation studies are therefore necessary to identify the distribution of trace
elements in fuel fractions.
ii) Sampling and profiling of different liquid fuel samples: a study based on profiling of the
levels of metal ions in liquid fuel samples obtained from different towns in Gauteng
province (RSA), is necessary. This will provide important information on safety and
quality of liquid fuels, thus helping in atmospheric pollution monitoring.
289