Application note: 050 δ18O analyses of various sample matrices with the vario PYRO cube Introduction We have used a low-cost, high–precision and rapid method of oxygen isotope analysis applied to various oxygen bearing matrices, organic and inorganic (sulfates, nitrates and phosphates), previously measured for δ18O. It was first successfully applied to δ18O analyses of natural and synthetic phosphate samples. The technique utilises a newly developed pyrolysis method based upon high-temperature EA–Pyrolysis combined with patented “purge & trap” chromatography interfaced to an IsoPrime100 IRMS. Using this pyrolysis method we have been able to generate a single calibration line for samples showing pyrolysis eﬃciencies independent on the type of matrix pyrolysed. This could be a very important step forward as silver phosphate is a very stable material, it is weakly hygroscopic and easily synthesised with predictable δ18O values. It could then be considered as a good candidate to become a reference material for the determination of δ18O by EA-pyrolysis-IRMS. Technical Setup The configuration follows standard high temperature pyrolysis methods using a glassy carbon tube filled with glassy carbon chips and carbon black held inside a ceramic furnace tube heated to 1450°C. However, the vario PYRO cube also oﬀers a number of advantages over other EA systems. Autosampler ‘Column Flush’ He Flow Analytical He Flow Analytical He Vent Sheath He Vent TCD Dilutor He Flow Sample Dilutor IsoPrime100 Reference He Flow P Sheath He Flow CO Column CO Reference Gas Injector Figure 1. During column flushing the source of the He carrier gas is switched from ‘Analytical He’ to ‘Flush He’ when CO is desorbed from the ‘purge and trap’ column. Purge & trap chromatography At the heart of the vario PYRO cube EA is a patented ‘purge & trap’ separation device which is used in place of the traditional packed GC column. This system traps CO from pyrolysis at 40°C and rapidly releases the sample CO by heating to 150°C. The TCD output determines when the trap is heated. This results in excellent chromatography which has obvious benefits for the integration of the sample peak Patented column backflushing to improve separation of N2 from CO when organic materials are pyrolised During sample pyrolysis the carrier He flow is set to “straight” mode. N2 is not trapped on the column and flows to the IRMS whereas CO is trapped. Once pyrolysis is complete, the He flow is set to “column flush” mode to divert slowly decomposing N2 away from the IRMS and to use a clean He flow to move the desorbing CO into the IRMS. The ‘purge & trap’ column is then heated to 150°C, the sample CO is desorbed and enters the Thermal Conductivity Detector (TCD) where the signal is detected as a focused peak. This patented column flushing technique oﬀers excellent advantages for this complex technique (See Technical Note 005). Two independent He flows through the furnace tubes The analytical He travels in a standard way, through the centre of the reaction tube carrying sample CO through the EA to the IRMS. However, a second independent He flow passes between the inner glassy carbon tube and the outer ceramic tube (furnace purge helium). This flow purges any ambient air which has diﬀused through the ceramic tube at high temperature which can lead to high “blank” peaks in other systems. Removable ash finger made from glassy carbon For simple maintenance, an ash finger made of glassy carbon is used which can be removed easily. Heated sample tray The sample tray can be heated at a user definable temperature to prevent any interference on the results by atmospheric moisture. Isoprime Ltd Isoprime House Earl Road Cheadle Hulme SK8 6PT, U.K. www.isoprime.co.uk [email protected] A member of elementargroup Application note: 050 Results Chromatography 4 2 0.15 CO Ref. Peak 6 CO Sample Peak N2 Ref. Peak 8 N2 Sample Peak 10 A significant advantage of the vario PYRO cube system is the quality of the CO peak shapes compared with conventional systems using packed GC separation column. This has a direct influence on the quality of the isotopic data integration. Also, there is no deterioration with time of the peak quality or the N2 - CO separation as has been shown with packed GC systems. 0.20 Mass 28 Mass 29 Mass 30 0.10 0.05 Signal Minor (nA) Signal Major (nA) 12 0.00 0 0 100 200 300 400 500 600 700 Time (Seconds) Results The instrumental setup was derived from Sieper et al. (2010) and details are included in Technical Note 005. First of all we investigated the treatment of the background signal; we conclude that it is necessary to use a blank subtraction routine in order to correct the data, considering that in any conventional system using GC column separation, any blank entering the system will be taken into account as a constant background dealt with during peak integration. Standard (Measured) SD n 40 (Ref. Values) NBS120c 21.70 0.10 5 21.70 Sucrose 35.55 0.21 4 36.40 Saccharose 27.96 0.04 5 27.90 9.01 0.22 4 AgPO4 30 Sucrose 25 Vanilin Vanilin y = 1.0022x + 0.1397 R² = 0.9977 35 9.50 20 15 BaSO4 10 AgNO3 BaSO4 11.14 0.42 4 12.00 AgNO 3 20.34 0.28 4 19.60 EP02 3.29 0.24 4 3.36 EP05 8.83 0.29 6 8.91 Saccharose 5 0 0 10 20 EP011 15.79 0.23 6 30 ! " 40 # 15.84 Diﬀerent pyrolysis and desorption temperatures have been investigated to optimize the analysis of silver phosphate samples. The best results were obtained by using a pyrolysis temperature of 1450°C and a CO desorption temperature of 150°C. Using these parameters we conducted an experiment mixing various sample matrices within the same batch using calibrated material measured with conventional techniques. Sucrose, vanillin, BaSO4, AgNO3 were then pyrolysed alongside our silver phosphate samples (Table 1; Figure 3). Once again, the quality of the best fit line (R2 = 0.997) obtained between measured values and reference values (3‰ < δ18O < 35‰) emphasises the performance of this pyrolysis method. A key parameter is the slope of the regression line (1.0022) which is very close to unity and an advantage for this technique compared to conventional EA-pyrolysis-IRMS. Conclusions The data obtained through this study confirm that the EA–pyrolysis method has the potential to be reliably used for the high precision determination of δ18O not only of phosphates, but also various oxygen bearing organic and inorganic compounds. This new procedure, using patented ‘purge & trap’ technology on the vario PYRO cube elemental analyser, is able to generate robust data in terms of accuracy and precision, taking into account that the various caveats related to the chromatographic separation of gases, as described by Farquhar et al (1997) are avoided. The lack of universal calibration material has been a major drawback to the development of oxygen pyrolysis techniques. Silver phosphate is a chemical compound very stable with time, weakly hygroscopic and thus could be considered as a top candidate to become a reference material for the determination of δ18O by high–temperature pyrolysis. Isoprime Ltd IsoPrime100 Applications Isoprime House Earl Road Cheadle Hulme SK8 6PT, U.K. Phone: +44 161 488 3660 Fax: +44 161 488 3699 Email: [email protected] A member of elementargroup