Deconvolution of FIGAERO–CIMS thermal desorption profiles using positive matrix factorisation to identify chemical and physical processes during particle evaporation

Buchholz, Angela; Ylisirniö, Arttu; Huang, Wei; Mohr, Claudia; Canagaratna, Manjula; Worsnop, Douglas R.; Schobesberger, Siegfried; Virtanen, Annele

doi:https://doi.org/10.5194/acp-20-7693-2020

Articles | Volume 20, issue 13

https://doi.org/10.5194/acp-20-7693-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-20-7693-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 20, issue 13

Research article

|

02 Jul 2020

Research article |

| 02 Jul 2020

Deconvolution of FIGAERO–CIMS thermal desorption profiles using positive matrix factorisation to identify chemical and physical processes during particle evaporation

Angela Buchholz, Arttu Ylisirniö, Wei Huang, Claudia Mohr, Manjula Canagaratna, Douglas R. Worsnop, Siegfried Schobesberger, and Annele Virtanen

Download

Final revised paper (published on 02 Jul 2020)
Supplement to the final revised paper
Preprint (discussion started on 20 Nov 2019)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Comments on “Deconvolution of FIGAERO-CIMS thermal desorption profiles using positive matrix factorisation to identify chemical and physical processes during particle evaporation” by Angela Buchholz et al.', Anonymous Referee #1, 14 Jan 2020
- AC1: 'Authors' response to Reviewer 1', Angela Buchholz, 15 Mar 2020
RC2: 'Review of Bucholz et al. ACPD 2019, “Deconvolution of FIGAERO-CIMS thermal desorption profiles using positive matrix factorisation to identify chemical and physical processes during particle evaporation”:', Anonymous Referee #2, 22 Jan 2020
- AC2: 'Authors' response to Reviewer 2', Angela Buchholz, 15 Mar 2020
RC3: 'Review of Buchholtz et al., 2019', Anonymous Referee #3, 01 Feb 2020
- AC3: 'Authors' response to Reviewer 3', Angela Buchholz, 15 Mar 2020

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Angela Buchholz on behalf of the Authors (15 Mar 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (18 Mar 2020) by Joachim Curtius

RR by Anonymous Referee #3 (23 Mar 2020)

RR by Anonymous Referee #1 (24 Mar 2020)

RR by Anonymous Referee #2 (02 Apr 2020)

Suggestions for revision or reasons for rejection

I thank the authors for addressing my comments. However, I need to follow up on my general comment number 2, as I don't think their response was adequate. Regarding the first part of my comment, I was suggesting that since Tmax changes across the various highOC cases, that this suggests 'blending' of factors, and that their chosen PMF solution isn't doing a great job separating factors properly. The authors replied that indeed the fact that Tmax changes is strong evidence that aqueous chemistry is occurring, and gave some physical reasons why Tmax would change. They state that only subsets of compounds within each factor are changing due to aqueous chemistry (or some subsets changing more than others). I agree totally with their reasoning of why Tmax is changing, and that aqueous chemistry is likely the cause, but they have missed my main point: in a proper PMF solution, these various subsets would be separated into different factors, such that T_max doesn't change for any factor, only concentrations of each factor change. It would of course be a lot of extra work to sort through the many more PMF factors you'd need to get to the most useful solution, but I think the authors should at least acknowledge that, while changing factors is strong evidence of aqueous chemistry, their solution isn't doing the best job of describing the factors, and suggest that as future work for the next PMFers.
Regarding the second part of my comment 2, I have to very strongly disagree with the authors characterization of my comment as a "misconception"! And your reply including Figure R2_1 shows exactly what I was referring to, and more strongly makes my case, that the HB1 factor is roughly constant across all temperatures for the wet,4h case (and blank) but dips down close to zero for the other three. You have agreed with me that this is further evidence of blending of factors. However, you have made no effort to adjust your PMF solutions, or even to acknowledge that your factors appear blended. At least you must acknowledge it. Better still would be to find better PMF solutions, because it is my view that this evidence of blending suggests the PMF solution isn't easily interpreted, and may not be reproducible in other lab or field measurements, and so the knowledge gained regarding the chemical and physical processes pertaining to aqueous chemistry may be liimited. But again, you could get by with suggesting this as future research, as of course this is a complicated project.

Hide

ED: Publish subject to minor revisions (review by editor) (05 Apr 2020) by Joachim Curtius

AR by Angela Buchholz on behalf of the Authors (15 Apr 2020) Author's response Manuscript

ED: Publish subject to technical corrections (01 May 2020) by Joachim Curtius

AR by Angela Buchholz on behalf of the Authors (04 May 2020) Manuscript

Download

Article (5808 KB)
Full-text XML

Short summary

To understand the role of aerosol particles in the atmosphere, it is necessary to know their detailed chemical composition and physical properties, especially volatility. The thermal desorption data from FIGAERO–CIMS provides both but are difficult to analyse. With positive matrix factorisation, we can separate instrument background from the real signal. Compounds can be classified by their apparent volatility, and the contribution of thermal decomposition in the instrument can be identified.