Gas-phase photochemical models do not account for the formation of a secondary altitude HNO3 maximum in the upper stratosphere at high latitudes during winter, suggesting that some processes are missing in the currently accepted chemistry of reactive nitrogen species <Kawa et al., 1995>. Heterogeneous chemistry on aerosol particles had been discounted as the cause because the aerosol surface area is expected to be very low at these altitudes. We have coupled a sulphate aerosol microphysical model to a chemical transport model to investigate this model deficiency in the Arctic during January 1992. The aerosol model predicts the formation of small sulphate particles at 1100 K. Comparisons with cryogenic limb array etalon spectrometer (CLAES) HNO3 and improved stratospheric and mesospheric sounder (ISAMS) N2O5 observations show that the heterogeneous conversion of N2O5 to HNO3 on the modeled small sulphate particles can account for some of the unexpected features seen in Upper Atmospheric Research Satellite (UARS) observations.¿ 1997 American Geophysical Union