|The introduction cites literature that has nothing to do with atmospheric aerosols and are based on photochemical studies of aquatic chromophoric dissolved organic matter. This is not stated here, and hence the references used here are inappropriate and misleading.|
Unfortunately the introduction did not improve and does not make correctly use of the primary literature. A lot of photochemical studies on aerosols are still missing and wrong citations have been used. Grammar remains distracting and some of the descriptions are confusing as a result of this.
Overall, the manuscript improved slightly but remains poorly written and misused the literature. The overall novelty is still not clear to me, but perhaps the evaluation of the ROS is of merit. I am also concerned to combine optical data that were collected in different solvents because of the known substantial matrix effects associated with different solvent matrices. I still cannot recommend publication of this manuscript in its current state.
Line 10: HULIS needs to be defined.
Line 38: Murphey et al 2018 did not study aerosols. This study was undertaken with dissolved organic matter from different aquatic systems and cannot be directly compared to aerosols.
Line 42: A lot of the cited references are not specific to aerosols and have been used for aquatic CDOM. This is misleading and is a false statement.
Line 62: super oxide and not super-oxygen and hydroxyl radicals and not hydroxyl
Line 91-92: At 500 C, the combustion of organics is supposed to be complete, especially by introducing fresh air, I remain confused why this was done. Also, how do you get POA after achieving complete combustion as it was stated in the text?
Line99-108: The description of the reactor system is highly confusing and even the photos do not help to clarify the setup.
Line 125-130: How was organic carbon quantified here? It is not clear at all.
Line 138: Optical properties are highly dependent on the matrix and it is not expected to be able to directly compare EEMS obtained in methanol versus the one collected in water. It is even more problematic when the dataset is combined to create a PARAFAC model.
Line 146L It still does not state here what triplet state species is referred to.
Figure 4: Absorbance should be normalized to pathlength. Please see Helms, J. R.; Stubbins, A.; Ritchie, J. D.; Minor, E. C.; Kieber, D. J.; Mopper, K., Absorption spectral slopes and slope ratios as indicators of molecular weight, source, and photobleaching of chromophoric dissolved organic matter. Limnology and Oceanography 2008, 53, (3), 955-969. And equation (2) within.