Determination of pressure baseline corrections for clumped-isotope signals with complex peak shapes

Räss, Stephan; Nyfeler, Peter; Wheeler, Paul; Price, Will; Leuenberger, Markus Christian

doi:https://doi.org/10.5194/egusphere-2024-3572

Preprints

https://doi.org/10.5194/egusphere-2024-3572

Preprints

06 Jan 2025

| 06 Jan 2025

Determination of pressure baseline corrections for clumped-isotope signals with complex peak shapes

Stephan Räss, Peter Nyfeler, Paul Wheeler, Will Price, and Markus Christian Leuenberger

Abstract. Pressure baseline corrections have been proposed to mitigate pressure-dependent background effects and reduce the apparent dependence of Δ₄₇ on δ₄₇ (non-linearity) observed in clumped-isotope studies of CO₂. In this work, we describe the determination of pressure baseline corrections for signals whose peak tops vary considerably across their width. Our study focuses on peaks with very small signal-to-baseline ratios (1.005 to 1.025) generated by the clumped isotopes ¹⁷O¹⁸O (linearly increasing peak top) and ¹⁸O¹⁸O (negatively curved peak top). The measurements were all performed in pure-oxygen gas using the compact, low-mass-resolution Elementar isoprime precisION Isotope Ratio Mass Spectrometer. We demonstrate that our corrections significantly reduce the influence of secondary electrons and that the adjusted clumped-isotope signals correctly increase with signal intensity. Furthermore, we extensively discuss correction procedures of varying complexity and explain why the best results were obtained by predicting multiple background values from the corresponding on-peak signals. Through this approach, we typically achieved standard deviations around 1 · 10^-9 (35/32), 0.2 ‰ (δ₃₅), 0.5 ‰ (Δ₃₅), 7 · 10^-9 (36/32), 0.1 ‰ (δ₃₆) and 0.1 ‰ (Δ₃₆) for at least 120 intervals (20 s integration). For the capital delta values, this corresponds to standard errors of the mean of less than 0.05 ‰, achieved with a total integration and analysis time of approximately 40 min and 6 h, respectively. We also show that the uncertainties of certain measurement parameters can be further reduced by optimising the measurement position (acceleration voltage) and applying additional drift corrections. For instance, for 35/32- and 36/32-related parameters we observed improvements of up to 1 order of magnitude and a factor of 7, respectively. Based on Monte Carlo simulations, we also show that the main uncertainties in our capital delta values are related to the on-peak signals, predicted backgrounds and the peak top curvature (only for Δ₃₆). Additionally, we present a brief study on the influence of pressure baseline corrections on major oxygen-isotope ratios and their delta values. While these corrections had an insignificant effect on their uncertainties, the absolute values of 33/32 and 34/32 changed markedly.

Received: 15 Nov 2024 – Discussion started: 06 Jan 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Stephan Räss, Peter Nyfeler, Paul Wheeler, Will Price, and Markus Christian Leuenberger

Status: final response (author comments only)

RC1:
'Comment on egusphere-2024-3572', Stefano Bernasconi, 10 Feb 2025

This manuscript describes a methodology for baseline corrections in an Elementar isoprime mass spectrometer used for the analysis of clumped isotopes of O2. The authors discuss the challenges posed by complex peak shapes and low signal-to-noise ratios in this low resolution instrument. The presented PBL correction procedure significantly improves measurement precision. Accuracy is not assessed due to the lack of standards. The method involves predicting multiple background values from on-peak signals, achieving sub-permil uncertainties in key isotopic ratios.
With Monte Carlo simulations uncertainty contributions from different factors are determined, and the major sources of error are discussed.
These corrections procedures can be useful for other similar instruments with low resolution and to improve the quality of data when the signal to noise ratio is small due to the low abundance of an isotopologue.
The paper is generally well written just some sections in the introduction could be shortened (see below). I suggest it can be accepted with minor revisions
Line 78 “By definition” instead of “by construction”
Lines 80-100 this paragraph should be shortened and updated. It is now clear that the negative backgrounds are the cause of the observed nonlinearities and PBL corrections are a must in clumped isotope analysis. HG and EG corrections are a wourkaround that does not correct the direct cause of the nonlinearity and are unnecessarily complicated.
Many papers have shown that PBL corrections eliminates the need for HG an EG corrections. He 2012 and Bernasconi 2013 were the start, but other papers have confirmed it, and all the results in the interlaboratory calibration exercise presented in Bernasconi et al. 2021 only utilize PBL corrected data.
Lines117-123 not really necessary,could be removed.
Line 124: it is not necessary to hit the center of the cup, the flat portion of the signal represent the width of the cup where the entire beam is collected in the cup. The signal can also be measured correctly on the side of the peak, as long as it’s in the flat part of the peak.
Bottom of page 9 : beginning of the sentence is missing.
Line 230-235 you could mention that such large changes in background shapes were also reported by Meckler et al. 2014 (DOI: 10.1002/rcm.6949) for a MAT253 mas spectrometer. As you mention, it is indeed very important to monitor backgrounds with scans as the shape and magnitude can strongly change. Just peak jumping to the side of the peak is not a sufficient way to monitor the change in background.
Line 437 what do you mean that the procedure for correcting is “ more involved”?
Line 580 this is only a problem on curved peak tops, it the peak top is flat than the exact position is not so important.
625 But now it is I-CDES (Bernasconi et al. 2021), the tying to the theoretical values is the first step, but the interlaboratory comparability can only be ensured by using traceable standards that can be measured in different laboratories.
The code and data should be made available on a repository.

Citation: https://doi.org/10.5194/egusphere-2024-3572-RC1
- AC2: 'Reply on RC1', Stephan Raess, 16 Apr 2025
  
  We would like to thank Prof. Dr. Bernasconi for reviewing our manuscript and providing constructive feedback. The comments were insightful and are greatly appreciated, as they helped us make meaningful improvements.
  Below, we repeat the posted comments along with our responses, which are provided in italics:
  Line 78 “By definition” instead of “by construction”
  Done.
  Lines 80-100 this paragraph should be shortened and updated. It is now clear that the negative backgrounds are the cause of the observed nonlinearities and PBL corrections are a must in clumped isotope analysis. HG and EG corrections are a wourkaround that does not correct the direct cause of the nonlinearity and are unnecessarily complicated.
  Many papers have shown that PBL corrections eliminates the need for HG an EG corrections. He 2012 and Bernasconi 2013 were the start, but other papers have confirmed it, and all the results in the interlaboratory calibration exercise presented in Bernasconi et al. 2021 only utilize PBL corrected data.
  The introduction was shortened and updated.
  Lines 117-123 not really necessary,could be removed.
  The corresponding lines were removed.
  Line 124: it is not necessary to hit the center of the cup, the flat portion of the signal represent the width of the cup where the entire beam is collected in the cup. The signal can also be measured correctly on the side of the peak, as long as it’s in the flat part of the peak.
  The following sentences were deleted: "To collect as many ions as possible, all ion beams of interest must hit the centres of the corresponding cups, which requires an ideal geometrical alignment of the cups. However, the smaller the mass resolution of the mass spectrometer, the more challenging it becomes to achieve this goal."
  Bottom of page 9 : beginning of the sentence is missing.
  Nothing was changed because the sentence starts on the previous page. This is confusing, as there are various plots in between.
  Line 230-235 you could mention that such large changes in background shapes were also reported by Meckler et al. 2014 (DOI: 10.1002/rcm.6949) for a MAT253 mas spectrometer. As you mention, it is indeed very important to monitor backgrounds with scans as the shape and magnitude can strongly change. Just peak jumping to the side of the peak is not a sufficient way to monitor the change in background.
  We added the sentence "Similar observations were made by Meckler (2014) using a Thermo Fisher MAT 253 mass spectrometer."
  Line 437 what do you mean that the procedure for correcting is “more involved”?
  The sentence "While our procedure for correcting m/z=35 signals is already more involved than the PBL corrections presented by He (2012) and Bernasconi (2013), m/z=36 peaks introduce an additional layer of complexity due to their curvature." was replaced by "While accounting for off-peak signals to the left and right of the m/z = 35 peaks already complicates the calculation of PBL corrections compared to the procedure outlined by He (2012) and Bernasconi (2013), the m/z = 36 peaks introduce an additional layer of complexity due to their curvature."
  Line 580 this is only a problem on curved peak tops, it the peak top is flat than the exact position is not so important.
  The sentences "Second, if the Faraday cups are not perfectly aligned, the centres of the peaks recorded on different cups are located at slightly different acceleration voltages. Therefore, it is advisable to choose a narrow cup for determining the measurement position." were replaced by "Second, if the Faraday cups are not perfectly aligned, the centres of the peaks recorded on different cups are located at slightly different acceleration voltages, which can pose a problem when the peak tops are not flat. In this case, it is advisable to choose a narrow cup for determining the measurement position."
  625 But now it is I-CDES (Bernasconi et al. 2021), the tying to the theoretical values is the first step, but the interlaboratory comparability can only be ensured by using traceable standards that can be measured in different laboratories.
  The sentence "Probably the most widely adopted approach is the Absolute Reference Frame introduced by Dennis (2011) directly tying the capital delta values to the theoretical equilibrium values calculated by Wang (2004)." was replaced by "Probably the most widely adopted approach nowadays is the Intercarb-Carbon Dioxide Equilibrium Scale (I-CDES) introduced by Bernasconi (2021), which ties the capital delta values to the theoretical equilibrium values calculated by Wang (2004) and ensures interlaboratory comparability through the use of traceable standards."
  The code and data should be made available on a repository.
  Once this paper is published, the basic code and its documentation will be made available in the Bern Open Repository and Information System (BORIS) as part of the dissertation "Measurement of oxygen clumped isotopes using mass spectrometry" by Stephan Räss.
  
  Citation: https://doi.org/10.5194/egusphere-2024-3572-AC2
RC2:
'Comment on egusphere-2024-3572', Amzad Laskar, 14 Feb 2025

The manuscript addresses an important issue related to pressure baseline changes for complex peaks upon the introduction of sample gases (O2 here) in a gas-source mass spectrometer (Elementar Isoprime). This is particularly relevant for the precise measurement of rare isotopologues, such as clumped isotopes. The authors proposed correction procedures and recommend scanning multiple background values corresponding to on-peak signals for improved baseline corrections. While the study is suitable for publication, some revisions are required.
The manuscript is too lengthy in its current form. It would benefit from improvements in language, as well as the removal of less important descriptions, elimination of repetitive content. Additionally, some figures (4-8), table (table 1) and a significant amount of texts could be moved to supplementary materials for better organization.
The manuscript mentions the mass spectrometer's ability to measure signal currents for masses 35 and 36 for O2, and associated error estimates are provided. Their statements at the beginning of introduction “we observed that our device is sensitive enough to measure the multiply-substituted oxygen isotopologues 17O18O and 18O18O in pure-oxygen gas, despite its low mass resolution and its use of 10¹¹ Ω resistors on the corresponding cups” is a bit misleading. The isobaric intereference to 18O18O from 36Ar cannot be avoided but resolved in the mass spectrometer (for that a resolving power of ~1170 or better is required). This is because a sample is very unlikely be 100% Ar free (even after passing through GC) and the amount of Ar present in a sample from the background (even for pure O2) would also interfere 18O18O measurement. Although the primary focus is on monitoring background shifts rather than measuring 18O18O, 17O18O, the observed errors could stem from variable contributions from isobaric interferences such as 36Ar and 35Cl for 18O18O and 17O18O, respectively. In section 3, authors mentioned that they cannot distinguish 17O17O from 16O18O due to low resolving power of the mass spectrometer used. The mass resolving power is not enough to separate 36Ar from 18O18O, let’s leave separating 17O17O from 16O18O. For separation of 17O17O from 16O18O, mass resolving power required is 8117. I would say separation of these two isotopologues is not possible even with a mass resolution of >8117 because of the intensity imbalance of the two isotopologues (16O18O signal is >27,000 times more than 17O17O, 10 % valley definition is practically inapplicable). Therefore, I suggest authors to make the statements a bit carefully. I Also suggest authors to discuss these somewhere in their manuscript.
Please shorten the conclusion section.
Detailed comments on the manuscript are provided in the annotated PDF.

Citation: https://doi.org/10.5194/egusphere-2024-3572-RC2
- AC1: 'Reply on RC2', Stephan Raess, 16 Apr 2025
  
  We would like to sincerely thank Prof. Dr. Laskar for taking the time to review our manuscript and we truly appreciate the thoughtful comments and valuable suggestions, which have helped us to improve the quality and clarity of the work.
  Below, we repeat the posted comment along with our responses, which are provided in italics:
  The manuscript is too lengthy in its current form. It would benefit from improvements in language, as well as the removal of less important descriptions, elimination of repetitive content. Additionally, some figures (4-8), table (table 1) and a significant amount of texts could be moved to supplementary materials for better organization.
  Section 3.1 and 3.2 were moved to the appendix. The key findings were summarized at the end of Section 3.
  The manuscript mentions the mass spectrometer's ability to measure signal currents for masses 35 and 36 for O2, and associated error estimates are provided. Their statements at the beginning of introduction “we observed that our device is sensitive enough to measure the multiply-substituted oxygen isotopologues 17O18O and 18O18O in pure-oxygen gas, despite its low mass resolution and its use of 1011 Ω resistors on the corresponding cups” is a bit misleading. The isobaric intereference to 18O18O from 36Ar cannot be avoided but resolved in the mass spectrometer (for that a resolving power of ~1170 or better is required). This is because a sample is very unlikely be 100% Ar free (even after passing through GC) and the amount of Ar present in a sample from the background (even for pure O2) would also interfere 18O18O measurement.
  The sentence "We observed that our device is sensitive enough to measure the multiply-substituted oxygen isotopologues 17O18O and 18O18O in pure-oxygen gas, despite its low mass resolution and its use of 10^11 Ω resistors on the corresponding cups" was reformulated as "We observed that our device is sensitive enough to measure the multiply-substituted oxygen isotopologue 17O18O in pure-oxygen gas. We also detected a peak at m/z = 36 u e^-1, which corresponds primarily to 18O18O, though minor impurities may contribute small amounts of 36Ar."
  Although the primary focus is on monitoring background shifts rather than measuring 18O18O, 17O18O, the observed errors could stem from variable contributions from isobaric interferences such as 36Ar and 35Cl for 18O18O and 17O18O, respectively. In section 3, authors mentioned that they cannot distinguish 17O17O from 16O18O due to low resolving power of the mass spectrometer used. The mass resolving power is not enough to separate 36Ar from 18O18O, let’s leave separating 17O17O from 16O18O. For separation of 17O17O from 16O18O, mass resolving power required is 8117. I would say separation of these two isotopologues is not possible even with a mass resolution of >8117 because of the intensity imbalance of the two isotopologues (16O18O signal is >27,000 times more than 17O17O, 10 % valley definition is practically inapplicable). Therefore, I suggest authors to make the statements a bit carefully. I Also suggest authors to discuss these somewhere in their manuscript.
  The sentence "Due to isobaric interferences of oxygen components, with our setup, it is not possible to distinguish the clumped isotope 17O17O (m/z = 34) from 16O18O." was replaced by "It should also be noted that, with our setup, it is not possible to detect 17O17O (m/z = 34). First, there is an isobaric interference with 16O18O; second, there is a large intensity difference between 17O17O and 16O18O, making the detection of the former isotopologue challenging, even for high mass-resolution instruments (Laskar, 2019)."
  Please shorten the conclusion section.
  Done.
  Detailed comments on the manuscript are provided in the annotated PDF.
  We clarified the statements on lines 13/14 and 399, and we have removed and/or shortened the other highlighted sections as suggested.
  
  Citation: https://doi.org/10.5194/egusphere-2024-3572-AC1
RC3:
'Comment on egusphere-2024-3572', Anonymous Referee #3, 18 Mar 2025

The manuscript provides a comprehensive account of how a low-resolution mass spectrometer, specifically the Elementar Isoprime precisION, can be utilised to measure clumped isotopes of oxygen. The measurements are conducted on pure oxygen. Due to secondary electrons the signal of the clumped isotopes is reduced and needs correction. Additionally, several other corrections are required to ensure reliable results. The study concludes that this low-resolution instrument can achieve results comparable to those obtained with more costly high-resolution alternatives. However, an absolute calibration of the system is currently lacking.
The manuscript is well written and highly detailed but is overly long and repetitive. The principles discussed are valuable and may have broader applicability to other measurements and instruments. Nevertheless, the manuscript would benefit considerably from substantial condensation and a more results-focused revision.
Abstract: The abstract should, in my view, begin by stating that minor isotopes are significantly affected by secondary electrons, which are dependent on bulk mass flow. It should then outline that the manuscript addresses the methodology for correcting this effect.
Line 10: The term on-peak signals is introduced without prior explanation. Clarification is needed at this stage.
Line 13: The phrase at least 120 intervals is ambiguous. How many intervals have actually been used for the calculation of standard deviations?
The introduction can be shortened significantly.
3.1 The conclusion of this section is that secondary electron suppression is not able to remove the secondary electrons fully and background corrections are still necessary. Please write this.
Figure 9: Please adjust the Y-scales.
The conclusions could be more concise. I recommend summarising the key successes of the approach and providing the achieved standard errors, ensuring the focus remains on the core findings.

Citation: https://doi.org/10.5194/egusphere-2024-3572-RC3
- AC3: 'Reply on RC3', Stephan Raess, 16 Apr 2025
  
  We are grateful to the reviewer for the thorough evaluation of our manuscript and for offering valuable suggestions for enhancing clarity and refining our work.
  Below, we repeat the posted comment along with our responses, which are provided in italics:
  The manuscript is well written and highly detailed but is overly long and repetitive. The principles discussed are valuable and may have broader applicability to other measurements and instruments. Nevertheless, the manuscript would benefit considerably from substantial condensation and a more results-focused revision.
  The introduction and conclusion were shortened. Furthermore, parts of Section 3 were moved to the appendix.
  Abstract: The abstract should, in my view, begin by stating that minor isotopes are significantly affected by secondary electrons, which are dependent on bulk mass flow. It should then outline that the manuscript addresses the methodology for correcting this effect.
  The abstract now starts as follows: "Secondary electrons emitted from the surfaces of Faraday cups can significantly influence the peak shapes and intensities of minor isotopes. As the amount of these electrons depends on the source pressure, pressure baseline corrections have been proposed to mitigate these pressure-dependent background effects and reduce the apparent dependence of Δ₄₇on δ₄₇(non-linearity) observed in clumped-isotope studies of CO₂."
  Line 10: The term on-peak signals is introduced without prior explanation. Clarification is needed at this stage.
  We replaced "on-peak signals" by "signals on the peak top (on-peak signals)".
  Line 13: The phrase at least 120 intervals is ambiguous. How many intervals have actually been used for the calculation of standard deviations?
  35/32: 1e-9: Table 4, 10 intervals
  
  δ₃₅: 0.2 %: Table 3, 120 intervals
  
  Δ₃₅: 0.5 %: Sect. 4.4.1, 120 intervals
  
  36/32: 7e-9: Table 6, 120 intervals
  
  δ₃₆: 0.1 %: Table 6, 230 intervals
  
  Δ₃₆: 0.1 %: Table 6, 230 intervals
  For simplicity, we dropped 'at least' because the standard deviation usually increases with the number of intervals. Furthermore, we already use 'typically' at the beginning of the sentence. The same adjustment was made in the conclusion.
  The introduction can be shortened significantly.
  Done.
  3.1 The conclusion of this section is that secondary electron suppression is not able to remove the secondary electrons fully and background corrections are still necessary. Please write this.
  The sentence "Most importantly, Fig. 2 illustrates that applying a negative potential to the Faraday cups is insufficient to achieve m/z = 36 signals that are higher than the collector zero value, resulting in negative background-corrected signals." was replaced by "Most importantly, Fig. A2 illustrates that the use of electron suppressors alone is insufficient to produce m/z = 36 signals above the collector zero value, resulting in negative background-corrected signals that necessitate further background correction."
  Figure 9: Please adjust the Y-scales.
  Done.
  The conclusions could be more concise. I recommend summarising the key successes of the approach and providing the achieved standard errors, ensuring the focus remains on the core findings.
  Done.
  
  Citation: https://doi.org/10.5194/egusphere-2024-3572-AC3

Stephan Räss, Peter Nyfeler, Paul Wheeler, Will Price, and Markus Christian Leuenberger

Viewed

Total article views: 184 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
120	54	10	184	8	7

HTML: 120
PDF: 54
XML: 10
Total: 184
BibTeX: 8
EndNote: 7

Views and downloads (calculated since 06 Jan 2025)

Month	HTML	PDF	XML	Total
Jan 2025	51	22	3	76
Feb 2025	36	13	3	52
Mar 2025	20	10	1	31
Apr 2025	13	9	3	25

Cumulative views and downloads (calculated since 06 Jan 2025)

Month	HTML	PDF	XML	Total
Jan 2025	51	22	3	76
Feb 2025	36	13	3	52
Mar 2025	20	10	1	31
Apr 2025	13	9	3	25

Viewed (geographical distribution)

Total article views: 174 (including HTML, PDF, and XML) Thereof 174 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 18 Apr 2025

Short summary

Clumped-isotope signals obtained through gas-source mass spectrometry are typically small and require pressure baseline corrections. While such corrections have been developed for square-shaped peaks, we present an approach for correcting peaks with complex shapes. Our method is demonstrated using oxygen clumped isotopes measured in pure oxygen, where the peak tops are linearly increasing and/or negatively curved.


Total:	0
HTML:	0
PDF:	0
XML:	0