Sorption rate parameters were determined for three organophosphorus (OP) compounds [dimethyl methylphosphonate (DMMP), diethyl ethylphosphonate (DEEP), and triethyl phosphate (TEP)] as surrogates for the G-type nerve agents sarin (GB), soman (GD), and tabun (GA). OP surrogates were injected and vaporized with additional volatile organic compounds into a 50 m3 chamber finished with painted wallboard. Experiments were conducted at two furnishing levels: (i) chamber containing only hard surfaces including a desk, a bookcase, tables, and chairs and (ii) with the addition of plush materials including carpet with cushion, draperies, and upholstered furniture. Each furnishing level was studied with aged and new painted wallboard. Gas-phase concentrations were measured during sealed chamber adsorb and desorb phases and then fit to three mathematical variations of a previously proposed sorption model having a surface sink and allowing for an embedded sink. A four-parameter model allowing unequal transport rates between surface and embedded sinks provided excellent fits for all conditions. To evaluate the potential effect of sorption, this model was incorporated into an indoor air quality simulation model to predict indoor concentrations of a G-type agent and a nonsorbing agent for hypothetical outdoor releases with shelter-in-place (SIP) response. Sorption was simulated using a range of parameters obtained experimentally. Simulations considered outdoor Gaussian plumes of 1- and 5-h duration and infiltration rates of 0.1, 0.3, and 0.9 h(-1). Indoor toxic loads (TL) for a 10-h SIP were calculated as integral C2 dt for a G-type agent. For the 5-h plume, sheltering reduced TLs for the nonsorbing agent to approximately 10-65% of outdoor levels. Analogous TLs for a G-type agent were 2-31% or 0.3-12% of outdoor levels assuming slow or moderate sorption. The relative effect of sorption was more pronounced for the longer plume and higher infiltration rates.