Evaluation of the performance of high temperature conversion reactors for compound-specific oxygen stable isotope analysis

Abstract

In this study conversion conditions for oxygen gas chromatography high temperature conversion (HTC) isotope ratio mass spectrometry (IRMS) are characterised using qualitative mass spectrometry (IonTrap). It is shown that physical and chemical properties of a given reactor design impact HTC and thus the ability to accurately measure oxygen isotope ratios. Commercially available and custom-built tube-in-tube reactors were used to elucidate (i) by-product formation (carbon dioxide, water, small organic molecules), (ii) 2nd sources of oxygen (leakage, metal oxides, ceramic material), and (iii) required reactor conditions (conditioning, reduction, stability). The suitability of the available HTC approach for compound-specific isotope analysis of oxygen in volatile organic molecules like methyl tert-butyl ether is assessed. Main problems impeding accurate analysis are non-quantitative HTC and significant carbon dioxide by-product formation. An evaluation strategy combining mass spectrometric analysis of HTC products and IRMS O/O monitoring for future method development is proposed.

Helium‐flushed sheathed nickel tube reactor for continuous flow oxygen stable isotope compound‐specific analysis

“…Hitzfeld et al 17 note the presence of “small recombination products” or “organic by‐products” (C x H y in the range m/z 24–27) when an unlined ceramic reactor tube is used, but that these ions are mostly absent when a Pt liner is used, suggesting that the Pt liner offers some protection of the gas flow from the Al 2 O 3 ceramic. However, it is possible that the Pt might also be serving as a catalyst for more complete conversion of organic compounds.…”

“…For δ 18 O analysis, the 44 CO 2 / 28 CO ratio might similarly reflect the effective heating regime in the reactor, where the reaction CO 2 + C ↔ 2CO has an K eq value (at 1250°C and estimating C/CO = 20/1) of 1836 yielding 44 CO 2 / 28 CO of 3.10 −5 . However, with high extraneous oxygen, this ratio can run from 0.02 to as high as 0.2, 17 thus rendering it of limited value and likely a source of confusion. Only when non‐oxygenated compounds in the internal standard (in our case n‐alkanes) produce no CO might the 44 CO 2 / 28 CO ratio be considered as an indicator of reactor efficiency.…”

“…With the addition of a Pt wire catalyst, this configuration might become dual use simply by changing the trickle from hydrogen to oxygen. Further, when used for δ 18 O, the Pt might serve as a catalyst to remove small organic remnants/junk masses 17 if desired. However, the time to flush the previous gas dissolved in the Ni tube might prove excessive.…”

“…The first of these is lack of proper graphitization, easily demonstrated with additional graphitization 17 . The second is a simple chromatographic issue: the use of an open tube within a closed tube, leading to dead volumes or parallel flow paths 17 . Our sealed tube design eliminates this second issue.…”

“…However, the annular space between the ceramic tube and the Pt tube remains in communication with the primary flow path, resulting in a problematic parallel flow path or poorly flushed volume which may lead to peak broadening, tailing or fragmentation. Similarly, positioning the Ni wires and Pt tube in the heated zone has been suggested as a point of concern but there does not seem to be any firm basis for this 17 . Removal of the fused‐silica's polyimide coat is recommended to avoid polluting the flow stream, but leaves the capillary weakened and potentially more chemically active.…”