Cluster activation studies with a diffusive condensation particle counter: Effect of chemical composition

Abstract

We use KanomaxFMT's Fast Condensation Particle Counter (CPC) to study the growth (activation) of purified singly charged cluster ions of several salts by condensation of n-butanol vapor. Salts investigated include 1-ethyl-3-methylimidazolium tris(pentafluoroethyl)trifluorophosphate (EMI-FAP), and four tetra-alkylammonium bromides (CpH2p+1)4N+-Br-, with p = 4, 6, 7, 12. The clusters studied contained up to 20 salt molecules, which were formed via a bipolar electrospray source, and were size-selected with a differential mobility analyzer. The CPC is operated at variable condenser and fixed saturator temperatures, Tc, Ts. It uses 75% sheath gas, such that most aerosol-containing fluid streamlines are near the channel center, and experience very similar maximal saturation ratios. This results in a steep initial dependence of the activation probability on Ts for selected particles. Therefore, a well-defined activation onset temperature Ts∗ is determined as a function of cluster composition, polarity (±), and mobility diameter (dp). Notwithstanding a modest material dependence found for the inferred Tc∗(dp) curves (±0.25 nm), the steepness of the activation curves measured permits determining particle size distributions for aerosols of known composition. There is some ambiguity in the precise temperature profile in the CPC channel. Nevertheless, using our best estimate, the onset condition for all but one of the seed particles studied agrees well with predictions from classical heterogeneous nucleation theory (Fletcher, 1962) when the pre-exponential term K in the nucleation rate is chosen appropriately. This enables the incorporation of materials properties into the sizing process in terms of only this single parameter K. This finding also provides a method to determine K experimentally, suggesting that the material dependence of the activation probability is due to the different interaction potentials between the clusters and the vapor. Tetraheptylammonium bromide is peculiar in that its cluster anions fit classical theory very much as all other materials studied, but its cluster cations do not.