This paper investigated the heterogeneous N2O5 uptake and ClNO2 production on the saline mineral dust through laboratory experiments, and evaluated the impacts of this heterogeneous process on tropospheric ClNO2 using a 3-D model. The results showed substantial formation of ClNO2 from the heterogeneous process on different saline mineral dust samples, and the ClNO2 yield varied with the mass fraction of particulate chloride and RH. The model simulation also showed significant impacts of this heterogeneous process on ClNO2 production and even O3 formation during a severe dust event in China. This study provides valuable information on the heterogeneous process of N2O5 and ClNO2 on saline mineral dust particles, the information of which has been very limited. The results will be useful to better understand the impacts of this heterogeneous process in different environments, and also will be helpful to improve the air quality model performance. Overall, the manuscript is well written, and thus I suggest that the manuscript can be published after addressing the following comments.

1. Line 152-153, please clarify what does the ‘initial N2O5 concentrations’ mean. Does the author mean the N2O5 generated from the reaction chamber or before passing the sample filters?
2. Line 162, although the dust particle loading method has been introduced in previous studies, a brief description will be useful and should be included here.
3. Line 193-196 and Line 373-375. In addition to the uptake and yield on dust particles, the parameters used for non-dust particles also should be explicit. Some information needs to be briefly provided in the main text or supplementary.
4. Line 201, the detection limit for these species should be provided in the experimental section.
5. Table 2. Considering the errors given by the standard deviation, the author should avoid using excessive significant digits. This also needs to be checked thoroughly for the whole manuscript.
6. Figure 3. Please clarify the meaning of particulate water, and definition of mw/m0.
7. Line 285-290. It’s interesting to see that the Ca and Mg amount may affect the ClNO2 yield. Can any figures or plots better depict the dependence of ClNO2 yields on Ca and Mg concentration or fraction in the saline mineral dust samples?
8. Line 331-332, as the author stated later, the assumption that all choloride is soluble may lead to overestimated ClNO2 yield. What would be a more reasonable assumption here, any semi-quantitative information on the water-soluble Cl fraction/content can be inferred? Is there previous data that can be used to compare the [Cl]/[H2O] ratio on the dust samples with the normal ambient particles? I think this will be very useful for further modeling simulation works.
9. Line 365, can the author explain more the rationale for choosing 0.1 as the fixed ClNO2 yield in the model simulation?
10. Line 397-399. The ‘short’ night in summer may still be enough to accumulate ClNO2 with plenty of NOx, O3, and particles. The statements here need further revision and improvement.

Nitryl chloride (ClNO₂) is an important precursor of atmospheric chloride radical, which influences the atmospheric oxidation and regulates the fate of air pollutants. This work conducted a comprehensive lab study of ClNO2 formation from N₂O₅ uptake on eight kinds of saline mineral dust samples collected from different regions in China. The result shows that the ClNO₂ yield largely impacted by the chloride contents in the saline mineral dust, but the relative humidity seems have no consistent rule in influencing the yield, indicating a complicated relationship between RH and the yield. Further simulation by GEOS-CHEM model demonstrates that the heterogeneous uptake of N2O5 on saline mineral dusts acted as an important source for the atmospheric ClNO₂ during the dust event over China. Overall, this topic is interesting and within the scope of ACP, the data analysis is sound and the manuscript is well written. It can be considered to accept after addressing the following several minor comments.

1. Line 40, suggest adding a phrase such like “in addition” before the sentence “Assuming a uniform φ(ClNO2)…”, to make clear that the subsequent contents have no relationship with the previous sentence “We further found that current parameterizations significantly overestimated φ(ClNO2)…”.
2. Line 214, in Fig. 2, the RH values are not completely consistent with those listed in Table 2, pleased confirm them.
3. Line 244, in Fig. 3(b), the value of mw (wet particle mass) to m0 (dry particle mass) under different humidity should be usually in the range from 1 to more than 1 due to the hygroscopic effect. The current values can be deltam/m0 (the ratio of mass difference to the dry mass). Suggest correcting it if there is any mistake. Same comment is put forward for Fig. 4 and 5.
4. Line 310, Figure 6 shows that when the mass ratio of Cl to total less than 0.1, the increase in ClNO₂ yield with respect to the increasing Cl content seems more significant at high RH condition (56% and 75%), is it possible that high RH promote the dissolution of chloride into the aerosol liquid water?
5. Line 348, here the inconsistent results between measurement and calculation may be due to the overestimated [Cl-]/[H2O(aq)], but another possibility is that compounds suppressed the formation of ClNO2 or compete with Cl- to react with NO2+. I encourage the authors to do some discussion.
6. Line 355-357, the sentence is not very clear. Nonhomogeneous chloride distribution across road salt aerosol particles during the field observation resulted in higher ClNO2 yield than the theory prediction, right?