General comments

The authors described the relatively low-cost analytical method for measurement of water-soluble and total Fe and Cu in aerosols and analyzed the aerosol samples over Atlanta. Such a high-frequency monitoring measurement is extremely useful for validation of the models and environmental assessment. I can recommend this paper for publication in Atmospheric Measurement Techniques and have minor comments to improve the paper.

Specific comments

Title

The reader might expect measurement of various chemical forms of iron and copper mentioned in introduction. Please consider rephrasing it by total and water-soluble, etc.

p.7, l.191: Please discuss feasibility of in-situ measurements of ambient aerosols to investigate the speciation of WS Fe and WS Cu.

p.10, l.289: Please specify the chemical composition for the mass fraction of 5.63%

Conclusion

Please discuss feasibility of high-frequency monitoring of Fe and Cu in size-resolved aerosol and rainwater samples.

Data availability

How did you estimate the solubilities when the total and WS concentrations were blank?

Please comment on the data such as 2017/3/20, which showed higher concentration through ultrafiltration than a 0.45 μm filter.

Please comment on the data such as 2017/4/10 and 2017/12/18, which showed higher solubilities than 100%.

p.16, l.510:  Please correct the unit.

Supplement

p.2, l.32: Please correct “colloidal??”.

p.2, l.53: Please correct “Figure S8”.

The manuscript needs a major revision. I think a high-throughput technique to measure Cu and Fe in PM 2.5 will definitely contribute to the atmospheric research community. However, I do have a few concerns regarding the methodology part of this manuscript, for which I suggest a major revision before acceptance.

PM2.5 is very heterogeneous in terms of elemental composition. You must consider the effect of other co-occuring elements when you are using a colorimetric method to measure a specific component (e.g. Cu or Fe). When multiple metal elements (such as Fe, Cu, Ni, Mn , Al  and Co) are present at similarly high concentrations) simultaneously, is your colorimetric method still valid to measure Fe or Cu alone without being affected by other co-occuring elements?   I saw you measured Mn, but your Mn concentration is very low, which would not interfere your results. But what is the threshold concentration of Mn and other transitional elements in air that makes your colorimeteic method invalid? In addition, it is commonly observed that PM2.5 can contain a high level of both Cu and Fe. If we are interested in measuring both Cu and Fe, is your method able to separate Cu and Fe in concentration measurement? Does a high Cu or Fe concentartion affect the measurement of the other? What will be the detection limit when you have them co-occuring?

I did not see enough negative control and positive control groups. I suggest you include following controls, to make your work useful for the whole atmosphere research community.

1) Positive controls should be done to validate the colorimetric method for Cu and Fe. For example, if you have 100 ppb of Fe (II) with 100 ppb of WS Cu, 100 ppb WS Mn , 100 ppb WS Ni and 100 ppb WS Co, will your ferrozine exclude other co-occuring these other elements and accurately measure e (II)?

2) Similarly you need negative control groups as well. For example, if you have 0 ppb of Fe (II) with 100 ppb of WS Cu, 100 ppb WS Mn , 100 ppb WS Ni and 100 ppb WS Co, will your ferrozine method exclude other co-occuring these other elements and show the absence of Fe (II) ? 

You need these control groups for ferrozine-Fe method and bathocuproine-Cu method. The discussion of artifacts due to effects of other elements is very important, because it informs other air researchers when your method is applicable.