the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Oct 2024
Product Ion Distributions using H3O+ PTR-ToF-MS: Mechanisms, Transmission Effects, and Instrument-to-Instrument Variability
Abstract. Proton-transfer-reaction mass spectrometry (PTR-MS) using hydronium ion (H3O+) ionization is widely used for the measurement of volatile organic compounds (VOCs) both indoors and outdoors. Unlike more energetic ionization methods (e.g., electron impact), H3O+ ionization can leave a target VOC molecule mostly intact and thus a VOC in a PTR-MS mass spectrum can be identified by its mass-to-charge ratio corresponding to the proton-transfer product (MH+). However, H3O+ ionization, and associated chemistry in the ion molecule reactor, is known to generate other product ions besides the proton-transfer product. The product ion distributions (PIDs) created during ionization include ions resulting from charge transfer reactions, water clustering, and fragmentation, all of which can create ambiguity when interpreting PTR-MS mass spectra. A standardized method of evaluating and quantifying the possible influence of PIDs on PTR-MS mass spectra is limited in part due to an incomplete understanding of the formation mechanisms and effects of instrument settings on measured PIDs, as well as the reasons for instrument-to-instrument variability.
We present a method, using gas-chromatography pre-separation, for quantifying PIDs from PTR-MS measurements of nearly 100 VOCs of different functional types including alcohols, ketones, aldehydes, acids, aromatics, halogens, and alkenes. Using this method we highlight major contributions of water cluster and fragment product ions to the PIDs of oxygenated VOCs. We characterize the influence of ion-molecule reactor conditions, ion transmission effects from quadrupole and ion optic tuning, and inlet capillary configuration on measured PIDs. We find that reactor conditions have the strongest impact on measured PIDs, but ion optic voltage differences and inlet capillary configuration can also affect PIDs.
Through an interlaboratory comparison of PIDs measured from calibration cylinders we characterize the variability of PID production from the same model of PTR-MS across seven participating laboratories. A subset of VOCs measured by the different laboratories had standard deviations (1 σ) associated with product ions that varied no more than 20 % thus providing a constraint for predicting PIDs across instruments operating under different conditions. We highlight the potential for misidentification of VOCs in PTR-MS mass spectra with a case study measurement of restroom air. We propose methods for identifying likely product ions and constraining the influence of PIDs on PTR-MS mass spectra. Finally, we present a library of H3O+ PIDs, from measurements acquired as part of this study, to be publicly available and updated periodically with user-provided data for the continued investigation into instrument-to-instrument variability of PIDs.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1346 KB)
-
Supplement
(316 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1346 KB) - Metadata XML
-
Supplement
(316 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-3132', Anonymous Referee #1, 02 Nov 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-3132/egusphere-2024-3132-RC1-supplement.pdf
- AC1: 'Reply on RC1', Michael Link, 06 Jan 2025
-
RC2: 'Comment on egusphere-2024-3132', Anonymous Referee #2, 10 Dec 2024
Link et al. describe a method to determine product ion distributions from PTR-TOF-MS measurements. They show the variability of such product ion distributions between various Vocus PTR-TOF-MS instruments from different labs, as well as a systematic analysis of how some instrument settings influence ion distributions. They also set up a library of product ion distributions to be used by the community. This is highly important work, since PTR-TOF-MS is widely used in atmospheric sciences, but fragmentation and water clustering often make the identification of single compounds, and even more so, of spectra from atmospheric observations, difficult.
The paper is well written and addresses an important concern of the PTR-MS community. The new product ion library along with the recommendations on how to use it will help users improve PTR-MS data quality.
I recommend its publication after the following comments have been addressed.
General comments:
I am curious if replicates were performed and if PIDs are presented as averages from multiple laboratory tests. Were any approaches taken to ensure these results are robust and statistically sound? Are the differences between measurements and between laboratories greater than the statistical noise? I encourage the authors to present a measure of the precision of the PIDs by collecting multiple measurements over a short time period for 1-2 instruments.
I am concerned with conflicting statements on the applicability of this method. A few specific examples are referenced below.
Specific comments:
Title: I was surprised by the simultaneous use of “H3O+” and “PTR” in the title. Is there any proton transfer reaction-MS technique that does not use H3O+?
Line 122: It should probably be mentioned that there is no good control of the ratio of H3O+ vs. NO+ or O2+ primary ions because of the BSQ-induced cutoff. (In traditional PTR-MS instruments, the H3O+ vs. O2+ or NO+ ratio could be optimized using the source valve which the Vocus does not have.)
Section 2.2.1: How old were the calibration cylinders from the different labs? Were they still certified or could aged, degraded calibration gas cause part of the observed differences? Did all labs apply dead-time correction / transmission efficiency in the same way?
Line 252 ff: Increasing the entire set of voltages is not what happens when users use the automatic tuning software “Thuner”, which individually changes BSQ or skimmer voltages to increase sensitivity. As a result, unwanted increases in fragmentation can be induced that the users do not even realize if they focus on using Thuner to increase sensitivity. So, at least a statement of caution is warranted. Did all the labs in the intercomparison use this approach of increasing the entire set of voltages simultaneously? If not, I wonder if some of the variability between labs (e.g., variability discussed in Line 411-414) could be a result of tuning these voltages. (Refer to Fig. S2 in Coggon et al. 2024.) If not all labs in the intercomparison used the Brophy & Farmer method for adapting skimmer and BSQ front/back voltages, please report which ones did, and how others decided on the deltaV1 and deltaV2s they used. Some Vocus instruments have two skimmer voltages, and the delta between both might impact the PID. For those instruments where two skimmer voltages exist, reporting the delta between both voltages would be helpful.
Line 379: Based on Fig. S2, the ratio of F1/parent ion changes between ~1 to ~1.75 depending on deltaV1. This seems like a significant difference. The way the PIDs are shown may dampen the impression of how strongly parent ion to fragment ratios change between Vocus settings. Notably, the ratio is what is used to correct ambient data later in the manuscript.
Line 387: I think the statement here (regarding ion optics) is not supported by enough evidence as it is based on just one VOC and one instrument. With the possibility of some Vocus users not using the Brophy & Farmer method, it has been shown that you can have intense fragmentation on ion optics voltages alone for certain functional groups (Coggon et al., 2024, Fig. S2).
Figure 3: Which BSQ voltage was used here? It would be helpful to indicate in the caption.
Line 326: Possible typo, change ‘the contributions fragment ions in the mass spectrum.’ to ‘the contributions of fragment ions in the mass spectrum.’
Line 332 ff: Is it possible that due to the close proximity of the sample inlet capillary to the ion source, the sample gas containing O2 and N2 also enters the ionization region through backdrift and causes impurities independent of a clean water supply?
Line 363: The BSQ is described as impacting ion transmission to the mass spectrometer and its use as a high-pass mass filter is noted in the text. How is it possible that an ion filter is causing ions with m/z 121.09 and above to be so greatly impacted by the BSQ voltage increasing (i.e, the blue line in figure 4a and 4c)? High BSQ voltages are described as causing more filtering, however, there is much higher ion signal from the water clusters at high BSQ voltages. Your results make it seem as though the changing BSQ is changing the ion chemistry in addition to just filtering. In figure 4c, the black line should also be relatively flat, but there is a pronounced ‘dip’ in the signal from BSQ 300 V to 400 V. This is unexpected for a pure mass filter. Do the authors suspect the BSQ is impacting ion chemistry or fragmentation in addition to serving as the high pass mass filter? Is it possible the pressures in the BSQ are high enough to allow collisions? The inconsistency between expected BSQ trends and what is observed should be discussed.
Line 387: possibly note that capillary distance can change sensitivity? (if this was observed in your study)
Line 419: Here, acetone is described as challenging to generalize. Later, it is used as the principal example for applying PID results to field data (e.g., line 687 says the method produces reasonable results).
Line 449 ff: The conclusion that higher O2+ and NO+ impurities are related to a higher inlet flow should be discussed further in terms of what this means physically. I think this shows that there is substantial drift of sample flow into the ionization region in the Vocus source.
Line 501 ff: Do the authors think that potentially another factor that influences the PIDs over time within one instrument could be ion source degradation/dirtiness? A dirty ion source could be related to a changing ion source voltage that may impact ion distributions. Were there any tests done to check the impact of the ion source state?
Line 520: for monoterpenes, there are more papers reporting product ion distributions that may be relevant here:
- Kari, E., Miettinen, P., Yli-Pirilä, P., Virtanen, A., and Faiola, C. L.: PTR-ToF-MS product ion distributions and humidity-dependence of biogenic volatile organic compounds, International Journal of Mass Spectrometry, 430, 87–97, https://doi.org/10.1016/j.ijms.2018.05.003, available at: http://www.sciencedirect.com/science/article/pii/S1387380617304943, 2018.
- Tani, A.: Fragmentation and Reaction Rate Constants of Terpenoids Determined by Proton Transfer Reaction-mass Spectrometry, Environmental Control in Biology, 51, 23–29, https://doi.org/10.2525/ecb.51.23, 2013.
Line 554: 6-MHO is usually 6-methyl-5-hepten-2-one (not 6-methyl-5-heptan-2-one as mentioned in the manuscript) when discussing skin oil oxidation products. Was the 6-MHO used in this study a different molecule or is this just a typo? If you did use a saturated ketone here, it should be mentioned explicitly that this isn’t the 6-MHO typically referred to in indoor air literature. If it is the typical 6-MHO, where 6-MHO and 2-octanone are directly compared, it should be noted that one is an unsaturated ketone which could have different ion chemistry available.
Figure 10: Consider changing the blue label in panel (b) to “acetone water cluster (calculated)” or something similar to increase clarity.
Line 798: An email address does not seem ideal for keeping up a library that is supposed to be accessible and added to in the future. Are there any plans to set it up as a website with a permanent DOI and a contact button that will be available even if at some point the email address is no longer active?
Citation: https://doi.org/10.5194/egusphere-2024-3132-RC2 - AC2: 'Reply on RC2', Michael Link, 06 Jan 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-3132', Anonymous Referee #1, 02 Nov 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-3132/egusphere-2024-3132-RC1-supplement.pdf
- AC1: 'Reply on RC1', Michael Link, 06 Jan 2025
-
RC2: 'Comment on egusphere-2024-3132', Anonymous Referee #2, 10 Dec 2024
Link et al. describe a method to determine product ion distributions from PTR-TOF-MS measurements. They show the variability of such product ion distributions between various Vocus PTR-TOF-MS instruments from different labs, as well as a systematic analysis of how some instrument settings influence ion distributions. They also set up a library of product ion distributions to be used by the community. This is highly important work, since PTR-TOF-MS is widely used in atmospheric sciences, but fragmentation and water clustering often make the identification of single compounds, and even more so, of spectra from atmospheric observations, difficult.
The paper is well written and addresses an important concern of the PTR-MS community. The new product ion library along with the recommendations on how to use it will help users improve PTR-MS data quality.
I recommend its publication after the following comments have been addressed.
General comments:
I am curious if replicates were performed and if PIDs are presented as averages from multiple laboratory tests. Were any approaches taken to ensure these results are robust and statistically sound? Are the differences between measurements and between laboratories greater than the statistical noise? I encourage the authors to present a measure of the precision of the PIDs by collecting multiple measurements over a short time period for 1-2 instruments.
I am concerned with conflicting statements on the applicability of this method. A few specific examples are referenced below.
Specific comments:
Title: I was surprised by the simultaneous use of “H3O+” and “PTR” in the title. Is there any proton transfer reaction-MS technique that does not use H3O+?
Line 122: It should probably be mentioned that there is no good control of the ratio of H3O+ vs. NO+ or O2+ primary ions because of the BSQ-induced cutoff. (In traditional PTR-MS instruments, the H3O+ vs. O2+ or NO+ ratio could be optimized using the source valve which the Vocus does not have.)
Section 2.2.1: How old were the calibration cylinders from the different labs? Were they still certified or could aged, degraded calibration gas cause part of the observed differences? Did all labs apply dead-time correction / transmission efficiency in the same way?
Line 252 ff: Increasing the entire set of voltages is not what happens when users use the automatic tuning software “Thuner”, which individually changes BSQ or skimmer voltages to increase sensitivity. As a result, unwanted increases in fragmentation can be induced that the users do not even realize if they focus on using Thuner to increase sensitivity. So, at least a statement of caution is warranted. Did all the labs in the intercomparison use this approach of increasing the entire set of voltages simultaneously? If not, I wonder if some of the variability between labs (e.g., variability discussed in Line 411-414) could be a result of tuning these voltages. (Refer to Fig. S2 in Coggon et al. 2024.) If not all labs in the intercomparison used the Brophy & Farmer method for adapting skimmer and BSQ front/back voltages, please report which ones did, and how others decided on the deltaV1 and deltaV2s they used. Some Vocus instruments have two skimmer voltages, and the delta between both might impact the PID. For those instruments where two skimmer voltages exist, reporting the delta between both voltages would be helpful.
Line 379: Based on Fig. S2, the ratio of F1/parent ion changes between ~1 to ~1.75 depending on deltaV1. This seems like a significant difference. The way the PIDs are shown may dampen the impression of how strongly parent ion to fragment ratios change between Vocus settings. Notably, the ratio is what is used to correct ambient data later in the manuscript.
Line 387: I think the statement here (regarding ion optics) is not supported by enough evidence as it is based on just one VOC and one instrument. With the possibility of some Vocus users not using the Brophy & Farmer method, it has been shown that you can have intense fragmentation on ion optics voltages alone for certain functional groups (Coggon et al., 2024, Fig. S2).
Figure 3: Which BSQ voltage was used here? It would be helpful to indicate in the caption.
Line 326: Possible typo, change ‘the contributions fragment ions in the mass spectrum.’ to ‘the contributions of fragment ions in the mass spectrum.’
Line 332 ff: Is it possible that due to the close proximity of the sample inlet capillary to the ion source, the sample gas containing O2 and N2 also enters the ionization region through backdrift and causes impurities independent of a clean water supply?
Line 363: The BSQ is described as impacting ion transmission to the mass spectrometer and its use as a high-pass mass filter is noted in the text. How is it possible that an ion filter is causing ions with m/z 121.09 and above to be so greatly impacted by the BSQ voltage increasing (i.e, the blue line in figure 4a and 4c)? High BSQ voltages are described as causing more filtering, however, there is much higher ion signal from the water clusters at high BSQ voltages. Your results make it seem as though the changing BSQ is changing the ion chemistry in addition to just filtering. In figure 4c, the black line should also be relatively flat, but there is a pronounced ‘dip’ in the signal from BSQ 300 V to 400 V. This is unexpected for a pure mass filter. Do the authors suspect the BSQ is impacting ion chemistry or fragmentation in addition to serving as the high pass mass filter? Is it possible the pressures in the BSQ are high enough to allow collisions? The inconsistency between expected BSQ trends and what is observed should be discussed.
Line 387: possibly note that capillary distance can change sensitivity? (if this was observed in your study)
Line 419: Here, acetone is described as challenging to generalize. Later, it is used as the principal example for applying PID results to field data (e.g., line 687 says the method produces reasonable results).
Line 449 ff: The conclusion that higher O2+ and NO+ impurities are related to a higher inlet flow should be discussed further in terms of what this means physically. I think this shows that there is substantial drift of sample flow into the ionization region in the Vocus source.
Line 501 ff: Do the authors think that potentially another factor that influences the PIDs over time within one instrument could be ion source degradation/dirtiness? A dirty ion source could be related to a changing ion source voltage that may impact ion distributions. Were there any tests done to check the impact of the ion source state?
Line 520: for monoterpenes, there are more papers reporting product ion distributions that may be relevant here:
- Kari, E., Miettinen, P., Yli-Pirilä, P., Virtanen, A., and Faiola, C. L.: PTR-ToF-MS product ion distributions and humidity-dependence of biogenic volatile organic compounds, International Journal of Mass Spectrometry, 430, 87–97, https://doi.org/10.1016/j.ijms.2018.05.003, available at: http://www.sciencedirect.com/science/article/pii/S1387380617304943, 2018.
- Tani, A.: Fragmentation and Reaction Rate Constants of Terpenoids Determined by Proton Transfer Reaction-mass Spectrometry, Environmental Control in Biology, 51, 23–29, https://doi.org/10.2525/ecb.51.23, 2013.
Line 554: 6-MHO is usually 6-methyl-5-hepten-2-one (not 6-methyl-5-heptan-2-one as mentioned in the manuscript) when discussing skin oil oxidation products. Was the 6-MHO used in this study a different molecule or is this just a typo? If you did use a saturated ketone here, it should be mentioned explicitly that this isn’t the 6-MHO typically referred to in indoor air literature. If it is the typical 6-MHO, where 6-MHO and 2-octanone are directly compared, it should be noted that one is an unsaturated ketone which could have different ion chemistry available.
Figure 10: Consider changing the blue label in panel (b) to “acetone water cluster (calculated)” or something similar to increase clarity.
Line 798: An email address does not seem ideal for keeping up a library that is supposed to be accessible and added to in the future. Are there any plans to set it up as a website with a permanent DOI and a contact button that will be available even if at some point the email address is no longer active?
Citation: https://doi.org/10.5194/egusphere-2024-3132-RC2 - AC2: 'Reply on RC2', Michael Link, 06 Jan 2025
Peer review completion
Post-review adjustments
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 471 | 241 | 129 | 841 | 24 | 13 | 8 |
- HTML: 471
- PDF: 241
- XML: 129
- Total: 841
- Supplement: 24
- BibTeX: 13
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
Megan S. Claflin
Christina E. Cecelski
Ayomide A. Akande
Delaney Kilgour
Paul A. Heine
Matthew Coggon
Chelsea E. Stockwell
Andrew Jensen
Jie Yu
Han N. Huynh
Jenna C. Ditto
Carsten Warneke
William Dresser
Keighan Gemmell
Spiro Jorga
Rileigh L. Robertson
Joost de Gouw
Timothy Bertram
Jonathan P. D. Abbatt
Nadine Borduas-Dedekind
Dustin Poppendieck
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1346 KB) - Metadata XML
-
Supplement
(316 KB) - BibTeX
- EndNote
- Final revised paper