Do NAD and NAT form in liquid stratospheric aerosols by pseudoheterogeneous nucleation?

Laboratory data of the freezing of nitric acid hydrates (NAD, NAT) from HNO(3)/H(2)O and HNO(3)/H(2)SO(4)/H(2)O solution droplets have been evaluated with respect to a "pseudoheterogeneous" (surface-induced) nucleation mechanism of NAD and NAT, which has been argued to possibly lead to the formation of polar stratospheric clouds (PSCs). In addition, a parametrization of pseudoheterogeneous nucleation of NAD and NAT suggested recently (Tabazadeh et al. J. Phys. Chem. A 2002, 106, 10238-10246) has been analyzed, showing that this parametrization should not be used in stratospheric modeling studies. The analysis of several laboratory data sets yields an upper limit of the pseudoheterogeneous nucleation rate coefficient of NAD of 2.2 x 10(-5) cm(-2) s(-1). In contrast, the upper limit of the pseudoheterogeneous nucleation rate coefficient of NAT could not be constrained satisfactorily, since formation of NAT has not been observed at stratospheric conditions in laboratory experiments applying small droplets. Maximum NAD production rates of 9.6 x 10(-9) cm(-3) (air) h(-1) in the stratosphere have been estimated assuming a pseudoheterogeneous nucleation mechanism that is constrained by the experimental observations. If maximum NAD supersaturation persisted for 4 weeks in the polar stratosphere the corresponding NAD particle number densities are estimated to be about 6 x 10(-6) cm(-3). These particle number densities are 3 orders of magnitude lower than particle number densities recently observed in the stratosphere. In conclusion, on the basis of laboratory data it is found that a pseudoheterogeneous nucleation mechanism is not sufficient to explain recent observations of large nitric acid containing particles in the polar stratosphere.