Modeling of the chemistry in oxidation flow reactors with high initial NO

Peng, Zhe; Jimenez, Jose L.

doi:https://doi.org/10.5194/acp-17-11991-2017

Articles | Volume 17, issue 19

https://doi.org/10.5194/acp-17-11991-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/acp-17-11991-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 17, issue 19

Research article

|

10 Oct 2017

Research article |

| 10 Oct 2017

Modeling of the chemistry in oxidation flow reactors with high initial NO

Zhe Peng and Jose L. Jimenez

Download

Final revised paper (published on 10 Oct 2017)
Supplement to the final revised paper
Preprint (discussion started on 10 Apr 2017)
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: 'Review on acp-2017-266', Anonymous Referee #1, 08 Jun 2017
RC2: 'Robust gas phase numerical study, if the OFR operates in the absence of particles...', Anonymous Referee #2, 25 Jun 2017
AC1: 'Final response', Zhe Peng, 15 Jul 2017

Peer-review completion

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

AR by Zhe Peng on behalf of the Authors (16 Jul 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (17 Jul 2017) by Dwayne Heard

RR by Gordon McFiggans (21 Aug 2017)

Suggestions for revision or reasons for rejection

The authors have largely addressed my concerns in their rebuttal. I am happy that the paper can form the basis of ongoing discussions in the field of OFR deployment and interpretation. I am particularly satisfied with the RTD analysis and its explanation of the deviation from plug flow. I am still slightly concerned by the continued separation of the gaseous and condensed phase processes. I am in full agreement with the authors that "OFR modeling is a subfield in itself, and our group cannot be expected to address every single possible topic". However, where a process can have a substantial influence on the processes that are the subject of a manuscript, then this possibility should be acknowledged.

It can be argued that the two statements in point 2.2 of the authors response:

i) "The presence of aerosols has typically negligible impacts on the gas-phase chemistry" and
ii) ..."gas-phase species have only limited impacts on OA"

are not demonstrably correct for all conditions in OFRs.

To rebut i), consider the typical concentrations in diesel emissions. Concentrations of NOx in raw diesel exhaust are typically between 50 and 1000 ppm depending on running conditions and technology (and can be very much higher during transients and below 17 degrees C when EGR is not mandated). Clearly this is the sort of NOx target regime of the current manuscript. Whilst PM emissions do not respond in the same way as NOx to engine technologies (e.g. EGR generally increases PM whilst decreasing NOx, and only DPF fitted vehicles have significantly reduced PM) or load-speed conditions, typical concentrations from a modern light-duty (EURO5) diesel generally range from between 1 and 30 mg/m^3 in raw exhaust. Assuming 80 nm modal diameter, 1 mg/m^3 will provide a mass transfer rate ("condensation sink") of about 4 s^-1 (using an uptake coefficient of unity); so a lifetime of 0.25 s for such a condensing gas (and 30 mg/m^3 would give a lifetime of less than 0.01 s). A lower uptake coefficient would obviously lead to a longer lifetime (e.g. 1 s for 0.01 at 30 mg/m^3).

Lines 88 to 92 explicitly include OFR conditions where there is a substantial likelihood of such high primary PM mass (an urban tunnel, "where NOx was high enough to be a major OH reactant"... and ... emissions of vehicles, biomass burning, and other combustion sources, "where NO can often be hundreds of ppm"). Looking at the Karjalainen et al., 2016 case presented in Figures 7 and 8, the authors are carrying out calculations under raw, 12 x and 100 x dilution conditions for gasoline engine emissions. Figure 7 in Karjalainen reported average primary PM values of 0.45 mg/m^3 for parts of the test cycle (assumed raw), rising to more than 10 mg/m^3 including the SOA from a gasoline engine. Similarly, the Link et al., 2016 study of diesel emissions at 45 - 110 dilution employed no primary particle removal technology to emissions from a turbocharged, intercooled, heavy-duty, off-road diesel engine likely to emit massively more than the light-duty levels stated above (in excess of 100 mg/m^3 is readily possible in raw exhaust from such engines). In both these cases the mass transfer of potentially condensing closed shell and radical species to PM could clearly provide very significant sinks of gaseous components that should be considered in a model of OFRs.

To address author response 2.2 ii), clearly gas-phase species have a strong impact on OA, being 100% responsible for all SOA. Mass transfer of semi-volatile and low volatility gas-phase species (in the case of exhaust experiments, almost completely due to condensation on existing primary PM) has a determinant effect on PM mass. Gas phase oxidants may have a limited impact on OA chemistry, but gas phase species have a profound effect on OA. Given the paper title relates to modelling the chemistry in OFRs (not modelling the oxidants), it is not solely transfer of radical species between phases that is of concern.

In the context of the above discussion, I do not understand the final paragraph of the authors suggested added text in point 2.2. I think this requires further explanation before inclusion in the paper.

Whilst I do not expect the paper to explicitly address coupling of the gaseous and particulate processes, I would expect the current manuscript to at least acknowledge the interaction between the gas phase chemistry and gaseous losses associated with condensation and the resultant increase in PM mass. The authors should state that their study is completely relevant for low PM loadings in OFRs, but care should be taken when applying it to high ambient PM concentration or direct emission studies (both raw and diluted). Clearly the authors are aware of the necessity to include coupled multiphase processes and should be commended in their work with Jeff Pierce's group on this.

Hide

ED: Reconsider after minor revisions (Editor review) (23 Aug 2017) by Dwayne Heard

AR by Zhe Peng on behalf of the Authors (28 Aug 2017) Author's response Manuscript

ED: Publish as is (04 Sep 2017) by Dwayne Heard

AR by Zhe Peng on behalf of the Authors (06 Sep 2017) Manuscript

Download

Article (5390 KB)
Full-text XML

Short summary

Oxidation flow reactors (OFRs) have been increasingly used to study atmospheric chemistry at high NO. We show that it is very difficult to obtain high-NO chemistry (in terms of RO₂ fate) in OFRs by initial NO injection. Past OFR studies with combustion sources generally had too-high precursor and NO_x concentrations that caused several types of experimental artifacts. A strong dilution (× 100 or larger) may be needed for such experiments to avoid undesired chemistry.