The heterogeneous reaction of NO 2 with water on the surface of laboratory systems has been known for decades to generate HONO, a major source of OH that drives the formation of ozone and other air pollutants in urban areas and possibly in snowpacks. Previous studies have shown that the reaction is first order in NO 2 and in water vapor, and the formation of a complex between NO 2 and water at the air-water interface has been hypothesized as being the key step in the mechanism. We report data from long path FTIR studies in borosilicate glass reaction chambers of the loss of gaseous NO 2 and the formation of the products HONO, NO and N 2 O. Further FTIR studies were carried out to measure species generated on the surface during the reaction, including HNO 3 , N 2 O 4 and NO 2 + . We propose a new reaction mechanism in which we hypothesize that the symmetric form of the NO 2 dimer, N 2 O 4 , is taken up on the surface and isomerizes to the asymmetric form, ONONO 2 . The latter autoionizes to NO + NO 3 À , and it is this intermediate that reacts with water to generate HONO and surface-adsorbed HNO 3 . Nitric oxide is then generated by secondary reactions of HONO on the highly acidic surface. This new mechanism is discussed in the context of our experimental data and those of previous studies, as well as the chemistry of such intermediates as NO + and NO 2 + that is known to occur in solution. Implications for the formation of HONO both outdoors and indoors in real and simulated polluted atmospheres, as well as on airborne particles and in snowpacks, are discussed. A key aspect of this chemistry is that in the atmospheric boundary layer where human exposure occurs and many measurements of HONO and related atmospheric constituents such as ozone are made, a major substrate for this heterogeneous chemistry is the surface of buildings, roads, soils, vegetation and other materials. This area of reactions in thin films on surfaces (SURFACE ¼ Surfaces, Urban and Remote: Films As a Chemical Environment) has received relatively little attention compared to reactions in the gas and liquid phases, but in fact may be quite important in the chemistry of the boundary layer in urban areas.