the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Dec 2023
Technical note: Determining chemical composition of atmospheric single particles by a standard-free mass calibration algorithm
Abstract. The chemical composition of individual particles can be revealed by single particle mass spectrometers (SPMS). With higher accuracy in the ratio of mass to charge (m/z), more detailed chemical information could be obtained. In the SPMS, the conventional standard-based calibration methods (internal/external) are constrained by the inhomogeneity of ionization lasers and the finite focusing ability of the inlet system, etc., therefore, the mass accuracy is restricted. In this study, we obtained the detailed and trustable chemical composition of single particles utilizing a standard-free mass calibration algorithm. In the algorithm, the characteristic distributions of hundreds of ions were concluded and collected in a database denoted as prototype. Each single-particle mass spectrum was tentatively calibrated by a mass calibration function with specific coefficients. The range of coefficients is constrained by the magnitude of mass deviation to a finite vector space. To find the optimal coefficient vector, the conformity of each tentatively calibrated spectrum to the prototype dataset was assessed. With maximum conformity, the optimal calibrated spectrum was obtained. For more than 98 % ambient particles, a twentyfold improvement in mass accuracy, from ~10,000 ppm (integer) to ~500 ppm (2 decimal places) was achieved. The improved mass accuracy validated the determination of adjacent ions with m/z difference ~0.05 Th. Furthermore, atmospheric trace ions that were poorly studied before are successfully specified. The obtained detailed single-particle-level chemical information could help understand the source apportionment, reaction mechanism, and mixing state of atmospheric particles.
-
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
(1706 KB)
-
Supplement
(1006 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1706 KB) - Metadata XML
-
Supplement
(1006 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-2610', Anonymous Referee #1, 22 Dec 2023
The comment was uploaded in the form of a supplement.
-
AC1: 'Reply on RC1', Xin Yang, 13 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2610/egusphere-2023-2610-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Xin Yang, 13 Mar 2024
-
RC2: 'Comment on egusphere-2023-2610', Anonymous Referee #2, 13 Feb 2024
This manuscript describes a mass calibration algorithm applied to single-particle mass spectra obtained using the HP-SPAMS (Hexin Instrument Co.). This work is quite similar to the manuscript by Zhu et al. (2020, Atmos. Meas. Tech.), which the authors claim is “sufficient for calibration” (Lines 149-150). Yet, the current manuscript is lacking many details, including how their work differs from and improves upon Zhu et al., who also claim similar mass accuracy. Further, Zhu et al. include evidence of identifying nearby peaks (< 0.1 m/z), within the same mass spectra. Yet, in the current manuscript’s abstract, the authors state “a twentyfold improvement in mass accuracy, from ~10,000 ppm (integer) to ~500 ppm (2 decimal places) was achieved”, which seems misleading when comparing to Zhu et al. It is also unclear why the intended journal for the current manuscript is ACP, since the manuscript focuses on method development (new calibration algorithm) for a specific instrument rather than new atmospheric science results. Further, the assessment of the accuracy of the author’s approach is unclear without running lab standards to check their claims (only ambient particles were measured). There are also problems, described below, regarding misleading claims by the authors.
The authors state on Lines 208-209 that “the isotopic distribution and the accurate m/z values in Fig. 4 jointly imply the validity of our calibration.” However, Figure 4 focuses on metals with isotopes separated by ~1 amu, for which the “improved” mass accuracy is not necessary.
In Figure 5a, the authors zoom in on the averaged mass spectrum of particles that were determined to contain both CaOH+ and C3H5O+, but without viewing the remainder of the mass spectrum (to look for the co-presence of Ca+ and other oxygenated organics, for example), the reader is not able to evaluate whether the co-identification of these two different ions is accurate, as stated on Lines 221-222. It would be informative to view the calibrated individual particle mass spectrum, rather than an averaged mass spectrum of >1k particles.
Most of the text in Section 3.2 (“Determination of adjacent ions in single particles”) focuses on distinguishing V+ and C4H3+ (0.05 Th). However, in the text and Figure 5 parts b & c, the authors report these ions did not co-exist within the same particles. Therefore, the goal of separating 0.05 Th adjacent ions within mass spectra of single particles (i.e. within the same particle, i.e. the same mass spectrum) is not achieved by this example. Therefore, the concluding sentence about this particular analysis (quoted here) is inaccurate: “…the successful determination of adjacent ions also validates our calibration algorithm.” In addition, this makes the following statement at the end of introduction (Lines 72-73) also inaccurate: “For instance, two adjacent ions in tracing ship emissions, V+ (50.95 Th) and C4H3+ (51.00 Th), were isolated and analyzed for the first time at the single-particle level.” The authors are presumably also referring to this in the abstract with the misleading statement: “The improved mass accuracy validated the determination of adjacent ions with m/z difference ~0.05 Th.”
Further, it is stated that “distinguishing V+ and C4H3+ in spectra is crucial in shipping aerosol research”. Why is this, since organics have been previously shown (through single-particle measurements) to co-exist in shipping particles? Figure 5b also shows the presence of m/z 37 (unlabeled, but presumably C2H+) in the V-containing particles. Given the significant fragmentation of the original hydrocarbon compounds upon ionization, it is unclear what differentiating C4H3+ and C2H+ tells the reader, and this aspect is not discussed.
Lines 69 & 157 claim “successful” calibration, when I couldn’t find an explanation of how that was evaluated. This seems challenging for ambient particles with unknown composition. There needs to be discussion of how “successful calibration” was assessed, rather than simply a statement of “success”.
Further, for the authors to back up their claims it would be useful for the authors to make lab standards of mixtures of compounds producing ions that are close to one another (such as the hypothetical ion pairs in Table S1) to test experimentally whether they have sufficient resolution and accuracy to separate the ions within an individual mass spectrum.
The authors state, on Lines 112-114 (with a similar statement on Lines 147-149), “The coefficients of Eq. (4) are restricted within a specific range, partially because of the constrained initial position deviation of the particles and the size of the laser spot (Text S1, Fig. S1).” However, no information is provided in Text S1 or Fig S1 (general instrument schematic) about this “initial position deviation” or laser spot size, so the reader is not able to evaluate or understand this statement.
Section 2.1: Since the calibration of mass spectra from a specific single-particle mass spectrometer (the HP-SPAMS) is the focus of this manuscript, it is critical to include information about the instrument in the main text methods, rather than only in the SI text. Text S1 should be moved to the main text. Even the instrument name is missing here in the main text. Since the authors are presenting ambient particle data, it is also crucial that the authors provide additional sampling information, beyond only the location and month of sampling (e.g., dates, sampling inlet information, etc).
The Du et al (2022) reference corresponds to an EGUSphere pre-print that does not appear to have resulted in a publication. It is inappropriate to cite this manuscript.
Additional comments:
The authors state in the abstract (Lines 25-26) that “atmospheric trace ions that were poorly studied before are successfully specified”. This phrasing could easily confuse a reader, as “atmospheric ions” typically refers to gas-phase ions, rather than ions produced from particle ionization during measurement, which is what I believe the authors are referring to here.
Lines 208: The authors refer to the isotopic distributions of metal ions are showing “validity of our calibration”. However, while the spacing of the isotopes were assigned in the calibration, I did not find any report of isotopic ratios that would further aid in assessment of the accuracy of the assignments.
Line 243: It is stated that the mass spectra (shown in Figure b & c) are the result of “averaging millions of particles”. Please provide actual numerical values for each group of particles, and clarify if these millions were all obtained during the time period shown in Figure 5d.
Citation: https://doi.org/10.5194/egusphere-2023-2610-RC2 -
AC2: 'Reply on RC2', Xin Yang, 13 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2610/egusphere-2023-2610-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Xin Yang, 13 Mar 2024
-
EC1: 'Comment on egusphere-2023-2610', Dantong Liu, 21 Mar 2024
Dear authors,
Thanks for answering the comments from both reviewers. After reviewing all the comments, replies and the revised manuscript, though I appreciate the fairly novel method by the authors to improve the accuracy of m/z determination for the single particle mass spectra, I found a few places need more information and explanation, besides the current replies from authors.
The main argument of the comments from one reviewer is about whether the separation between V+ and C4H3+ at m/z51Th is real or can be applied to the ambient air.
1) The raw and calibrated peaks at 51Th of single particles should be given. A few examples of single particle mass spectra around 51Th in both raw and calibrated forms will help, rather than only showing the averaged mass spectra in Fig. 5. This will answer referee’s comments about whether both V and C4H3 may exist in the ambient air, and can also demonstrate how well your algorithm may work.
2) The authors should more explicitly explain why 50.95 should be assigned with V. The argument is about whether this has been proved in the laboratory. By tracking the literatures, I can see the original validation for V 50.95 is from the table 2. Du et al. has published the lab work where they generated particles with V mixing with other substances in single particle, and they found both peaks of 51 and 50.95 in single particle. I wouldn’t think the authors need to repeat this lab work, but a full explanation (with clear reference supporting your results) should be given to explanation what 51 and 50.95 should be assigned with. This should come with the single particle mass spectra above to demonstrate how both compositions may separate or mix with each other in the ambient air.
Citation: https://doi.org/10.5194/egusphere-2023-2610-EC1 -
AC3: 'Reply on EC1', Xin Yang, 29 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2610/egusphere-2023-2610-AC3-supplement.pdf
-
AC3: 'Reply on EC1', Xin Yang, 29 Mar 2024
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-2610', Anonymous Referee #1, 22 Dec 2023
The comment was uploaded in the form of a supplement.
-
AC1: 'Reply on RC1', Xin Yang, 13 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2610/egusphere-2023-2610-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Xin Yang, 13 Mar 2024
-
RC2: 'Comment on egusphere-2023-2610', Anonymous Referee #2, 13 Feb 2024
This manuscript describes a mass calibration algorithm applied to single-particle mass spectra obtained using the HP-SPAMS (Hexin Instrument Co.). This work is quite similar to the manuscript by Zhu et al. (2020, Atmos. Meas. Tech.), which the authors claim is “sufficient for calibration” (Lines 149-150). Yet, the current manuscript is lacking many details, including how their work differs from and improves upon Zhu et al., who also claim similar mass accuracy. Further, Zhu et al. include evidence of identifying nearby peaks (< 0.1 m/z), within the same mass spectra. Yet, in the current manuscript’s abstract, the authors state “a twentyfold improvement in mass accuracy, from ~10,000 ppm (integer) to ~500 ppm (2 decimal places) was achieved”, which seems misleading when comparing to Zhu et al. It is also unclear why the intended journal for the current manuscript is ACP, since the manuscript focuses on method development (new calibration algorithm) for a specific instrument rather than new atmospheric science results. Further, the assessment of the accuracy of the author’s approach is unclear without running lab standards to check their claims (only ambient particles were measured). There are also problems, described below, regarding misleading claims by the authors.
The authors state on Lines 208-209 that “the isotopic distribution and the accurate m/z values in Fig. 4 jointly imply the validity of our calibration.” However, Figure 4 focuses on metals with isotopes separated by ~1 amu, for which the “improved” mass accuracy is not necessary.
In Figure 5a, the authors zoom in on the averaged mass spectrum of particles that were determined to contain both CaOH+ and C3H5O+, but without viewing the remainder of the mass spectrum (to look for the co-presence of Ca+ and other oxygenated organics, for example), the reader is not able to evaluate whether the co-identification of these two different ions is accurate, as stated on Lines 221-222. It would be informative to view the calibrated individual particle mass spectrum, rather than an averaged mass spectrum of >1k particles.
Most of the text in Section 3.2 (“Determination of adjacent ions in single particles”) focuses on distinguishing V+ and C4H3+ (0.05 Th). However, in the text and Figure 5 parts b & c, the authors report these ions did not co-exist within the same particles. Therefore, the goal of separating 0.05 Th adjacent ions within mass spectra of single particles (i.e. within the same particle, i.e. the same mass spectrum) is not achieved by this example. Therefore, the concluding sentence about this particular analysis (quoted here) is inaccurate: “…the successful determination of adjacent ions also validates our calibration algorithm.” In addition, this makes the following statement at the end of introduction (Lines 72-73) also inaccurate: “For instance, two adjacent ions in tracing ship emissions, V+ (50.95 Th) and C4H3+ (51.00 Th), were isolated and analyzed for the first time at the single-particle level.” The authors are presumably also referring to this in the abstract with the misleading statement: “The improved mass accuracy validated the determination of adjacent ions with m/z difference ~0.05 Th.”
Further, it is stated that “distinguishing V+ and C4H3+ in spectra is crucial in shipping aerosol research”. Why is this, since organics have been previously shown (through single-particle measurements) to co-exist in shipping particles? Figure 5b also shows the presence of m/z 37 (unlabeled, but presumably C2H+) in the V-containing particles. Given the significant fragmentation of the original hydrocarbon compounds upon ionization, it is unclear what differentiating C4H3+ and C2H+ tells the reader, and this aspect is not discussed.
Lines 69 & 157 claim “successful” calibration, when I couldn’t find an explanation of how that was evaluated. This seems challenging for ambient particles with unknown composition. There needs to be discussion of how “successful calibration” was assessed, rather than simply a statement of “success”.
Further, for the authors to back up their claims it would be useful for the authors to make lab standards of mixtures of compounds producing ions that are close to one another (such as the hypothetical ion pairs in Table S1) to test experimentally whether they have sufficient resolution and accuracy to separate the ions within an individual mass spectrum.
The authors state, on Lines 112-114 (with a similar statement on Lines 147-149), “The coefficients of Eq. (4) are restricted within a specific range, partially because of the constrained initial position deviation of the particles and the size of the laser spot (Text S1, Fig. S1).” However, no information is provided in Text S1 or Fig S1 (general instrument schematic) about this “initial position deviation” or laser spot size, so the reader is not able to evaluate or understand this statement.
Section 2.1: Since the calibration of mass spectra from a specific single-particle mass spectrometer (the HP-SPAMS) is the focus of this manuscript, it is critical to include information about the instrument in the main text methods, rather than only in the SI text. Text S1 should be moved to the main text. Even the instrument name is missing here in the main text. Since the authors are presenting ambient particle data, it is also crucial that the authors provide additional sampling information, beyond only the location and month of sampling (e.g., dates, sampling inlet information, etc).
The Du et al (2022) reference corresponds to an EGUSphere pre-print that does not appear to have resulted in a publication. It is inappropriate to cite this manuscript.
Additional comments:
The authors state in the abstract (Lines 25-26) that “atmospheric trace ions that were poorly studied before are successfully specified”. This phrasing could easily confuse a reader, as “atmospheric ions” typically refers to gas-phase ions, rather than ions produced from particle ionization during measurement, which is what I believe the authors are referring to here.
Lines 208: The authors refer to the isotopic distributions of metal ions are showing “validity of our calibration”. However, while the spacing of the isotopes were assigned in the calibration, I did not find any report of isotopic ratios that would further aid in assessment of the accuracy of the assignments.
Line 243: It is stated that the mass spectra (shown in Figure b & c) are the result of “averaging millions of particles”. Please provide actual numerical values for each group of particles, and clarify if these millions were all obtained during the time period shown in Figure 5d.
Citation: https://doi.org/10.5194/egusphere-2023-2610-RC2 -
AC2: 'Reply on RC2', Xin Yang, 13 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2610/egusphere-2023-2610-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Xin Yang, 13 Mar 2024
-
EC1: 'Comment on egusphere-2023-2610', Dantong Liu, 21 Mar 2024
Dear authors,
Thanks for answering the comments from both reviewers. After reviewing all the comments, replies and the revised manuscript, though I appreciate the fairly novel method by the authors to improve the accuracy of m/z determination for the single particle mass spectra, I found a few places need more information and explanation, besides the current replies from authors.
The main argument of the comments from one reviewer is about whether the separation between V+ and C4H3+ at m/z51Th is real or can be applied to the ambient air.
1) The raw and calibrated peaks at 51Th of single particles should be given. A few examples of single particle mass spectra around 51Th in both raw and calibrated forms will help, rather than only showing the averaged mass spectra in Fig. 5. This will answer referee’s comments about whether both V and C4H3 may exist in the ambient air, and can also demonstrate how well your algorithm may work.
2) The authors should more explicitly explain why 50.95 should be assigned with V. The argument is about whether this has been proved in the laboratory. By tracking the literatures, I can see the original validation for V 50.95 is from the table 2. Du et al. has published the lab work where they generated particles with V mixing with other substances in single particle, and they found both peaks of 51 and 50.95 in single particle. I wouldn’t think the authors need to repeat this lab work, but a full explanation (with clear reference supporting your results) should be given to explanation what 51 and 50.95 should be assigned with. This should come with the single particle mass spectra above to demonstrate how both compositions may separate or mix with each other in the ambient air.
Citation: https://doi.org/10.5194/egusphere-2023-2610-EC1 -
AC3: 'Reply on EC1', Xin Yang, 29 Mar 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2610/egusphere-2023-2610-AC3-supplement.pdf
-
AC3: 'Reply on EC1', Xin Yang, 29 Mar 2024
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 396 | 109 | 39 | 544 | 54 | 22 | 18 |
- HTML: 396
- PDF: 109
- XML: 39
- Total: 544
- Supplement: 54
- BibTeX: 22
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Yechun Ruan
Yuanlong Huang
Xujian Chen
Antai Zhang
Jianhuai Ye
Guomao Zheng
Baohua Cai
Yaling Zeng
Yixiang Wang
Chunbo Xing
Yujie Zhang
Tzung-May Fu
Huizhong Shen
Chen Wang
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1706 KB) - Metadata XML
-
Supplement
(1006 KB) - BibTeX
- EndNote
- Final revised paper