my review is attached. 

While most researchers nowadays focus on light absorption by brown carbon when trying to address the climate effect of aerosols,  Shen et al. focused on the overlooked aerosols scattering, which is simplified in a lot of studies, that may induce large uncertainty in radiative forcing estimation. The topic fits well with the scope of ACP. The manuscript is generally well written and adequately organized. I have a few comments before it can be accepted for further consideration.

• In the abstract and lines 471-474. The authors found an increasing trend of real refractive index with O/C or aging and stated this finding conflicts with most lab results. For example, He et al. (2018) observed a declining trend of real refractive index when OA aged from LO-OOA to MO-OOA for photooxidation of b-pinene and p-xylene. However, He et al (DOI: 10.1021/acs.est.1c07328) and Li et al. (DOI: 10.5194/acp-19-139-2019) also observed increasing refractive index with aging for naphthalene+NOx and biomass burning aerosols. Lab experiments did not conclude on this aspect, as it seems to depend on what kind of aerosol was studied. I would say such of statement exaggerates the significance of the finding in this study.
• The results obtained from this study are from a single observation site near the emission source. How could the results and conclusions from this unique location be applied to a larger scale or different locations with potentially different emission sources and experiencing different atmospheric processes?
• I am not clear how one could use PMF to do source apportionment for PNSD. The authors also mentioned in their manuscript that LOOA and MOOA are generally not externally mixed and are likely to be prone to internal mixing (lines 463-465). But in the discussion of lines 440-446, the authors stated that POA and SOA tend to be optically independent at the single-particle level. The authors did observe SOA at the observation site. According to our knowledge of new particle formation and condensation growth, SOA and POA would not be completely externally mixed. Condensation of secondary products on existing particles (e.g., POA) would change the size of the particles, but does not change the number. How could PMF deal with this situation?
• Figure 1b, for me, it looks like the mr1064,400/mr1064,235 changed suddenly at mr1064,400=1.53. What is the mechanism behind this?
• Line 386-387, would the correlation coefficient of R=0.25 and R=-0.24 be significantly different?
• The authors used the empirical method proposed by Li et al. (2023). I would say this might not be so relevant for this paper. Li’s method was developed based on refractive index data for pure compounds without N element. However, in urban locations, aerosols already have nitrogen-containing species that affect the refractive index of the aerosols. The explanations in lines 474-478 are not well supported, as the aerosols themselves are different. It is not clear how much the formation pathway matters.