In this study, the heterogeneous uptake of SO 2 on R-Al 2 O 3 and MgO particles at 296 K was investigated. Transmission FT-IR experiments reveal that under dry conditions SO 2 reacts with R-Al 2 O 3 and MgO to form strongly adsorbed bisulfite, HSO 3 -(aq), and sulfite, SO 3 2-(aq), on R-Al 2 O 3 and sulfite on MgO. To quantify the uptake of SO 2 under dry conditions, heterogeneous uptake coefficients were measured with a Knudsen cell reactor. The Knudsen cell data were modeled to account for gas diffusion in to the underlying layers and surface saturation. The initial uptake coefficient, γ o , is calculated using the BET surface area of the powdered samples as diffusion of SO 2 into the powder is found to readily occur. Values of γ o were determined to be 9.5 ( 0.3 × 10 -5 and 2.6 ( 0.2 × 10 -4 for SO 2 uptake on R-Al 2 O 3 and MgO, respectively, at 296 K under dry conditions at a gas concentration near 1 × 10 11 molecules/cm 3 . Additionally, water uptake on sulfitecoated R-Al 2 O 3 and sulfite-coated MgO particles was investigated with FT-IR spectroscopy. In the presence of adsorbed water, some of the adsorbed sulfite converted to sulfate on MgO but not on R-Al 2 O 3 . Atmospheric implications of these studies on SO 2 uptake on mineral dust aerosol are discussed.

One of the most important applications of the Knudsen cell reactor is in determining heterogeneous reaction kinetics of potentially important atmospheric reactions. Knudsen cell measurements involving gas reactions on atmospherically relevant particle surfaces, including salt, carbon black, soot, and mineral dust, are often obtained using powdered samples. In this study, we have investigated the importance of gas diffusion into the underlying layers of powdered samples when determining kinetic parameters from Knudsen cell experiments. In particular, we show that the use of the geometric surface area of the sample holder is, in general, not justified in determining initial uptake coefficients or reaction probabilities because the interrogation or probe depth of gas-phase molecules into the bulk powder can be anywhere from tens to thousands of layers deep. One problem encountered by current models used to account for gas diffusion into the underlying layers is that the diffusion constant of the gas through the powdered sample must be known. Typically, diffusion constants for gases into powdered samples are unknown and are difficult to measure or accurately calculate. One way to circumvent this problem is to use thin samples such that the thickness of the sample is less than the interrogation depth of the gas-phase molecules. Under these conditions, the observed initial uptake coefficient is directly proportional to the surface area of the entire sample. This region is termed the linear mass-dependent regime and can be experimentally accessed for many, but not all, heterogeneous reactions. Several examples discussed here include heterogeneous reaction of NO 2 on γ-and R-Al 2 O 3 , R-Fe 2 O 3 , carbon black; HNO 3 on CaCO 3 ; and acetone on TiO 2 .