Review of manuscript “Gas-Equilibrium Membrane Inlet Mass Spectrometry (Ge-MIMS) for water at high pressure”

Journal: Geoscientific Instrumentation Methods and Data Systems

Date: August 2023

Overview

In this manuscript, the authors present the results of a new technique for dissolved gas measurements in subsurface water samples at high pressures. The technique is based on gas-liquid equilibration (GE-MIMS device), and measurement with a residual gas analyzer (the mini RUEDI). The mini RUEDI coupled to GE-MIMS was shown to perform well for the determination of dissolved gas, but usually uses a LIMM membrane that breaks at water pressures above approximatively 3 bar. In this paper, the authors describe tests made with a SOMM membrane that can withstand water pressures up to 100 bar. The SOMM membrane was developed for this application. As highlighted by the authors, this work is of great relevance for the field as measurement of dissolved gas in high pressure fluids (e.g. deep subsurface) is up to now still a technical challenge.

The paper is well structured, and the results give proof of evidence for the authors’ statements. The paper should be accepted, with minor technical precisions. Please find below some questions and minor comments for the improvement of the manuscript.

General remarks

1. I would appreciate in the discussion an opening on challenges that could occur at higher pressures (i.e. closer to the maximum that can withstand). Does the equilibration time with the SOMM membrane (and thus the required maximum duty cycle) vary with the water pressure?
2. Could suspended solids be a challenge for the use of such membranes (LIMM and SOMM)? In the different experiments was the water filtered to avoid such issue?
3. What are the current methods used for measurement of dissolved gases in the subsurface (other than GE-MIMS coupled to mini RUEDI)? What are the advantages of the mini RUEDI compared to these methods? This could be a nice addition to the introduction/discussion.

Specific comments

1. Line 53: typo, However
2. Lines 93-94: please give information on the local aquifer in term of location, turbidity, suspended solids (see general remark 3)
3. Lines 93-94: please precise the parameters of the mini RUEDI used (i.e. flushing time, number of repetition on the detectors)
4. Line 139: “SOMM can be as low as” change to “SOMM should be as low as”
5. Line 147: “slight trend”, please precise (i.e. Increase? Decrease? Significative?)
6. Lines 153-163: should go to material and methods (as a 2.5 Field test subchapter)
7. Line 189: is the GE-MIMS protocol only about changing the duty cycle, or were other parameters of the mini RUEDI changed as well?
8. Line 192: “there is therefore no need for …”, what do you mean by no need? The conclusion should be that with SOMM the temporal resolution cannot be as high as with LIMM. Using the term "no need" sounds like it is a default of LIMM that we can sample more often.
9. Line 193: “allows”, same comment as 8: this is not really a positive aspect, should be replaced by something like “the very low duty cycle is also required to prevent gas depletion in the gas phase, and the resulting offset of the partial pressures measured with the mini RUEDI”
10. Line 194: “avoids”, same comment as a 8 and 9: this is not really a positive aspect, should be replaced by “is necessary to”
11. Lines 209-211: it would be nice to add references of studies where dissolved gas measurement was a limitation/was hindered

Comments on figures

1. Figure 3: which LIMM data did you use for each SOMM data for the calculation of RSD, average of SOMM data/average of LIMM data?

This manuscript describes a useful extension for the on-site analysis of gases dissolved in water, in that higher pressure water can be analysed.  The manuscript should be accepted for publication after a few clarifications are made, and some further practical details included in respect to the analysis parameters of the miniRUEDI, so that an interested researcher could reproduce the results.

Ultimately, this study would have benefitted by passing large volumes of air saturated water (or some other standardised water) through both the SOMM and the LIMM devices to demonstrate the accuracy of each - presumably this was deemed infeasible due to the large volumes of water required.  The reader is left to trust in the excellent published work of the authors (especially MB and RK) that the LIMM-based GE-MIMS/miniRUEDI system provides accurate results within stated error margins, and therefore, because the LIMM results are consistent with the SOMM results then the SOMM system is also accurate.  Some mention of the absolute accuracy of the method should be made, why a water standard was not measured, and some reference made to the logical sequence reliant on LIMM providing accurate gas partial pressures.

Specific comments follow:

Line 9,18,134: Henry’s Law should have an apostrophe

Line 38: It is not immediately clear whether gas transfer across a membrane would fractionate a gas mixture, or if any such fractionation could have a dependence on the dissolved gas composition - this is doubly relevant for the present manuscript due to there being two different membrane materials being investigated.  Please expand on whether gas species fractionation across a membrane is inherently obviated due to the design or how these GE-MIMS systems are used (e.g., adequate dead volume present; time allowed for differential equilibration across the membrane to become unimportant, etc.), and/or provide more specific citations for how this aspect has been demonstrated in peer-reviewed literature (Mächler 2012 is a PhD thesis, and might not qualify as peer-reviewed, and Brennwald et al. 2016 is similarly general as to the reasons no fractionation is present).  

Line 53: However is misspelled

Line 77: Is it possible to provide a link to this patent application, should a researcher require further details of the design and fabrication materials?  If the SOMM is commercially available at present, please confirm this in response to reviews - if it is not available at present, state this and when it will be available in the manuscript.

Line 93: Can the authors make a statement about the concentrations of the dissolved gases in this “local aquifer” (e.g., has it been measured by other techniques, such as gas chromatography or mass spectrometry)?   If these concentrations are unknown and/or unnecessary to be stated for the hypothesis being tested this should be clearly noted, but it would still be better to give the reader at least a qualitative idea of the water being used in this test.

Line 118: Was the data here (and in the rest of the manuscript) collected on a Faraday Cup, and using atmospheric pressure air as a normalising standard?  Is the pressure at the inlet of the miniRUEDI monitored/recorded/used in the calculation for sample and standard?  I recommend being specific on this in the text or in a supporting document, including whether the normalising gas was ambient air, and what considerations one might make about using ambient air in an underground lab as a standard.

Line 154: Define CS

Line 163: I think “autonomous” (adjective) should be “autonomously” (adverb).

Line 167: I am not sure the authors have correctly defined “(ASW, see p_atm values in Fig. 4)”.  The N2/Ar ratio in air is 84; the ratio in Figure 4 is also 84.  In contrast, the N2/Ar ratio in air saturated water is 38 (at sea level, 20 degrees C).  I suspect what the authors actually mean is that Patm is the partial pressure in the membrane equilibrator headspace when air saturated water is passed through the equilibrator, producing an exsolved equilibrium gas mixture which is equivalent to air (unless there are any artefacts to the membrane system, which the authors imply are negligible).  I feel it would be helpful for the authors to make this somewhat subtle but ultimately confusing point more clear in the text and the Figure legend. 

Lines 171-172: The phrase “The N2 and Ar partial pressures show a slight excess relative to ASW” also suffers from the logical discontinuity between dissolved gas concentrations in ASW and those measured in ASW passed through a membrane equilibrator and measured on a miniRUEDI.  It would be helpful to be more specific, to avoid confusion.

Figure 4: The notation on the He panel is confusing.  For all panels "Patm = xx hPa" is shown, except for the He panel which appears to be alphaPa, which I presume to be a typo of some kind.