Dear authors. Due to the highly-critical reviews and the large number of changes I had the article reviewed again. Only one of the previous reviewers was able to review again, so the other two reviewers are new but they have seen the previous reviews and your responses. There is another large set of critical comments that I hope you will be able to address. Best regards. Sergey Nizkorodov.

Dear authors. Thank you for responding to multiple comments from anonymous reviewers. This exchange with the reviewers has improved the manuscript but unfortunately it is still far from a publishable state.
My main issue with the manuscript is the very poor quality of SMPS measurements shown in Figures S1 and S2. Additionally, the large scatter in Figure 1 strongly suggests that something is wrong with your SMPS system - the data cannot be so noisy at these reasonably high mass concentrations of particles. I am hesitant to let a manuscript that focuses on SOA yields but relies on such noisy data to go through from ACPD to ACP.
The quality of the mass spectra shown in Figure 4 is also very poor, and generally below the threshold for the quality of publishable high-resolution mass spectra. For example, in the negative ion mode spectra, the peaks above m/z 230 look like noise to me (because they all have a similar peak abundance), and only a few peaks even appear in the high RH data set. The positive ion more spectra are of slightly higher quality, but it is quite unexpected that high RH peaks would completely disappear at lower m/z values. Finally, the peak abundance distribution shown in Figure 5a suggests that data are likely contaminated by a C10 impurity.
My second issue is that I no longer think that the explanation of the suppressed particle yield in the Hinks et al. (2018) study, which you discuss extensively, was correct. After reading more papers on the subject of wall losses in chamber experiments, I am now convinced that the apparent yield suppression in Hinks et al. (2018) was simply an effect of SVOC escaping to the walls under humid conditions. Your discussion on page 7 focuses on compounds that are not supposed to partition in SOA particles (such as glyoxal and acetone). It is the SVOCs that we need to worry about under these circumstances. Please take a look at the following paper [Huang, Y. L.; Zhao, R.; Charan, S. M.; Kenseth, C. M.; Zhang, X.; Seinfeld, J. H., Unified Theory of Vapor-Wall Mass Transport in Teflon-Walled Environmental Chambers. Environ. Sci. Technol. 2018, 52, 2134-2142] and see if it can change your thinking and/or of you can improve on the discussion on page 7. The large difference between the seed vs no seed yields is a direct proof that you are losing most of condensable material on wall the surface. I wish this Huang’s paper was out when we were writing up the Hinks et al. (2018) paper – we would have written a very different story.
Finally, the readability of the paper is still quite poor for the following reasons. a). Some of the paragraphs in the paper are very long and need to be split into smaller, logically coherent paragraphs. b). The flow of logic in your writing is not smooth, with frequent jumps from one idea to the next within the same paragraph or even sentence. c). There is a number of spelling and grammatical mistakes, some (but not all) of which are mentioned below.
I am afraid that I am prepared to reject the manuscript at this point based on my assessment above, and based on the recommendation of the reviewers, who rated the paper very poorly. I will select "Reconsider after major revisions" for now in order to not make this decision final, but I think a rejection is the correct course of action in this case. If you agree with me that that paper should not be published given the poor quality of your data, you do not have to do anything, and the submission will remain in the ACPD state. If you disagree with my decision, you are welcome to try to provide a rebuttal and a next revised version but I personally cannot see how you can achieve a publishable state with the data quality you currently have. It might be easier to start over at this point and submit a higher quality data set as a new paper somewhere else.
In case you elect to continue editing this manuscript, in addition to addressing my major concerns above, I would fix the following.

P2, L7: anthropogenic source -> anthropogenic sources
P2, L8: delete the 2nd instance of the "in the urban troposphere"
P2, L21: processed -> processes
P2, L26: condition that no -> condition, wherein no
P2, L34: this sentence does not make any to me, first it says it observed an RH effect but then is says it did not focus on it. Please rephrase.
P3, L5: I would split this sentence in two, one stating the limited RH range of the previous study and one discussing why GC may not be the optimal technique for analysis.
P3, L12: it is unclear to me what you mean by "Although the information of chemical compositions in SOA has been given". Compositions should be singular, composition.
P3, L21: Experiments of -> Experiments on
P3, L22: delete "So only a brief introduction is presented here" because the description that follows is not brief
P3, L25: "To remove the electric charge on the surface of the FEP reactor" - this is very interesting, do you have evidence that it helps? (Perhaps you can cite a previous study which shows that it helps.)
P4, L6: light -> lights
P4, L13: Was this determined by observing the total mass or total number of particles? Usually the particle wall constants are size dependent, so I would add statement that you approximate the wall loss as size independent.
P4, L34: As the SOA compositions are almost all water-soluble species -> The low-NOx xylene SOA is water soluble (provide a reference or explain why you think so)
P5: I would insert a statement explaining that your sampling instruments consumed X% of reactor volume during the experiment, and that you have not used make up air to dilute the bag (since this was brought up by one of the reviewers)
P6, L4: Delete "Thus, the LWC can be obtained by VGF." to avoid repetition
P6, L4: As shown in Fig. 1 -> In Fig. 1. This sentence would fit better at the beginning of the 2nd paragraph – it does not fit well in the flow of logic of the 1st paragraph.
P6, L7: Good references to cite here are:
(1) El-Sayed, M. M. H.; Amenumey, D.; Hennigan, C. J., Drying-Induced Evaporation of Secondary Organic Aerosol during Summer. Environ. Sci. Technol. 2016, 50, 3626-3633.
(2) El-Sayed, M. M. H.; Wang, Y.; Hennigan, C. J., Direct atmospheric evidence for the irreversible formation of aqueous secondary organic aerosol. Geophys. Res. Lett. 2015, 42, 5577-5586.
P6, L11: negligible -> neglected.
P7, L2: “the excess use of H2O2 can lead to an excess OH radicals” - this statement is incorrect. H2O2 is the primary reactant for OH under high H2O2 conditions, so OH steady state concentration is independent of H2O2 in this limit.
P7, 2nd paragraph: see comment above
P8: The paragraph on page 8 needs to be split.
P9, L2: Figure 2 -> Figure 3
P9, L1o: both two RH conditions -> both RH conditions.
P10, L3: deduction -> subtraction
P10, L3: blank-deducting -> blank-corrected
P10, L4: Figure3 -> Figure 4 (and fix numbering for the rest of the figures)
P10: use "peak abundance" instead of "intensity" when discussing heights of peaks in the mass spectrum
P10 and Table 3: what evidence do you have for the proposed epoxide structures? I am also quite concerned that the 2,6-dimethyl-1,4-benzoquinone compound listed in table 3 is too volatile, and should not be in the particle phase. Is it possible that you are looking at the results of fragmentation in the ion source?
P10, L29: Figure 4 -> Figure 5 (and adjust the rest of the figure numbers)
P11, L5: is proposed -> suggests
P11, L9: what method did you use to identify C2H2O as one of the most frequent mass differences?
P12, L7: differential -> different
P10, L13: "are much more oxygenated than those in the positive ion mode" - your signal-to-noise ratio is measuring O/C is not high enough to make these conclusions
P11-P12: the discussion of the mechanism of RH influence on the chemical composition is not supported by the presented data in my opinion
P20: the caption of Figure 1 should mention that SOA concentration was corrected for the amount of LWC in particles.
P23: Replace "Intensity" by "Abundance" and Nagetive by Negative in Figure 4



Supporting information
Figure S1 is not referred to in the text. Also, its caption says that it shows GC-MS data, which is obviously incorrect.
Figure S2: it is not clear how Figure S2 is different from S1. They seem to show slightly different data.
Figure S3: this figure is not at all discussed in the text. I understand the reviewers wanted to see MS/MS spectra, so those should be discussed properly.

Abstract: Please sentence stating that some of the reduction in the apparent yield may be due to the faster wall loss of semivolatile products of oxidation at higher RH. (This reflects discussion on page 7)
P6 L13: specify actual RH values (14% and 76%) instead of simply saying “low” and high”. I would do this throughout the text, especially at the beginning of new paragraphs. For example, Exp. 1 (low RH) -> Exp. 1 (14% RH). This way the readers will not have to go to Table 1 all the time.
P6 L18: nearly seven times larger -> nearly an order of magnitude larger
P7: Please remove acetone from this discussion. Contrary to what you state on line 4, acetone has not been shown to form SOA under any reasonable conditions. Discussion of its wall loss is therefore irrelevant.
Figure 1: image or caption – specify the actual RH values instead of saying “low” and high”. Sam goes for the rest of the figure captions.
Figures 1 and 2: I suggest starting the Y=axis at zero. The offset used in the current versions of the figures makes small values appear larger than they are.
All figures: the resolution of the images may need to be improved. The figures appear grainy at the moment (because of a grey background used in the figures?)