Dear authors

you have taken most comments of the reviewers into account. I have a few additional suggestions for modification. Concerning the first point raised by Reviewer 1: I agree with the authors that the study of regional background sites is important to put e.g. urban measurements into perspective. One may indicate that this is important to do even at more sites and covering all seasons. In Europe, a comprehensive study was done mostly at regional background sites in different seasons using aerosol mass spectrometric measurements (Crippa et al., 2014). In the future, long-term measurements using aerosol chemical speciation monitors would be very helpful to cover all seasons at multiple sites.
I suggest you cite the study of Crippa et al., 2014 and the need of more extended measurements at these and additional background sites in China.

Crippa, M., F. Canonaco, V. A. Lanz, M. Aijala, J. D. Allan, S. Carbone, G. Capes, D. Ceburnis, M. Dall'Osto, D. A. Day, P. F. DeCarlo, M. Ehn, A. Eriksson, E. Freney, L. H. Ruiz, R. Hillamo, J. L. Jimenez, H. Junninen, A. Kiendler-Scharr, A. M. Kortelainen, M. Kulmala, A. Laaksonen, A. Mensah, C. Mohr, E. Nemitz, C. O'Dowd, J. Ovadnevaite, S. N. Pandis, T. Petaja, L. Poulain, S. Saarikoski, K. Sellegri, E. Swietlicki, P. Tiitta, D. R. Worsnop, U. Baltensperger, and A. S. H. Prevot (2014) Organic aerosol components derived from 25 AMS data sets across Europe using a consistent ME-2 based source apportionment approach, Atmospheric Chemistry and Physics, 14(12), 6159-6176.

Concerning point 4 of reviewer 1: the detection limit to measure more refractory elements by the open-closed difference is not very good. So I would not say that interferences of metal nitrates can be excluded. One may however say that one assumes and expects low contributions of metal nitrate in the submicron range (which would provide similar responses as organic nitrate) in contrast to the coarse mode where e.g. NaNO3 can be very abundant (you may find a reference for this).
Concerning point 5 of reviewer 1: Also in Hayes et al. (2013), they describe different results in the Los Angeles Basin that are due to mixing of some primary and secondary components. So I would tone it down even a bit more by saying .. "while the Van Krevelen diagram may still be useful for constraining..."
and above I would mention explicitely the mixing of different airmasses and components/sources as possible reasons for variations in the slope.

Details:
Change summarization to summary throughout the text.
page1 : space between PM1 and components
Page 2: before the year of the citations, sometimes spaces are missing. Please check throughout the manuscript.
Figure 3: make sure that size of Figures a, b are exactly the same including the text. add more space for the x and y-axes text. dM/dlogd .. nomenclature different in a, b versus c-h please make it the same. ug/m-3 is very hard to read on my printout.
Figure 4: check y-axis and x-axis text .. do them the same for all figures with the right distance to the axes numbers..
Figures 6c, 6d: add the x-axes at zero to the upper graphs (ticks and axes text not necessary)