I am trying to analyze some seawater and sea ice core samples for hydrogen and oxygen isotopic composition using a high temperature conversion — isotope ratio mass spectrometry method. I've seen groups mention that the salt in those samples can make analysis challenging, but I can't find specifics. What should I expect and what extra steps should I take for samples like these?
Any suggestions or methods/papers would be appreciated.
Presence of halogen ions from dissolved salts such a NaCl, KBr, KCl etc. make 2H/18O of sea water or brines challeging. Halogen gases formed during thermolysis (in a reducing environment) such as Cl2 are aggressive (corrosive) and may damage your instrument. On the other hand, formation of HCl or HBr cannot be completely ruled out. Even if these are formed in very small amounts, it still would mean formation of H2 is not quantitative which will lead to mass discriminatory effects and thus isotopic fractionation.
If your only option is submitting your samples for analysis on a standard HTC/EA-IRMS system, I would suggest you demineralize your water by reverse osmosis. Reverse osmosis is known not to suffer from mass discriminatory effects as far as water is concerned.
If you are in control ("own") the HTC/EA-IRMS system, you would have to modify the packing of the glassy carbon tube. Instead of packing the tube with glassy carbon chips only, you would have to pack the tube with a mixture of glassy carbon chips and Chromium (>99%) granules of approx. 3 mm in size. Please, note when analysing liquid samples by injection, the glassy carbon tube should be packed almost all the way to the top to avoid a dead volume (which would results in very broad H2 and CO peaks that may results in unreliable isotope ratio measurements). I also recommend replacing the graphite insert (that sits on top of the glassy carbon tube) with a stainless steel insert.
Alternatively, you could have your samples analysed on a CRDS instrument (LosGatos or Picarro). On these instruments, the "only" drawback of running salt waters is the need for cleaning the instrument (esp. the mirrors and inlet capillary) on a regular basis.
I agree with the answer of my friend Wolfram Meier-Augenstein I also found that the injection syringe does not deliver the same amount of water. A poor reproducibility. Thus the formation of H2 is not quantitative, which will lead to discriminating effects on the mass and therefore the isotopic fractionation.
Issam Moussa's cautionary tale prompts me to add the following. It is possible to analyse freshwater, ice / snow melt and precipitation by injection using a syringe (barrel volume 1 - 1.5 uL) fitted with a needle of 26 gauge and 70 mm in length. Surprisingly (or perhaps not), it is easier to inject water (injection volume 100 - 120 nL) into an HTC/EA manually rather than by using an autosampler. Injecting manually, the analyst can feel if friction would cause the plunger not to hit home in a smooth continued fashion. Rather than to proceed (and damage the syringe in the process as an autosampler would) one aborts the run and either replaces the plunger or uses a spare syringe. We recently published the results of a survey of 127 freshwater lakes. Water samples were analysed for d2H and d18O values by liquid injection on to a HTC/EA-IRMS system. Samples and reference waters VSMOW, GISP and SLAP were injected in replicates of 5 with results from the 1st injection discarded in case of any memory (carry over). Typical error of measurement for average d2H and d18O values of 4 injections was ±0.64 and ±0.19 ‰ respectively.
https://doi.org/10.1080/10256016.2019.1609958
- Badges
- Science topic
- Similar topics
- Chemistry
- Sulfites
- Inorganic Chemistry
- Inorganic Chemicals
- Salts