Growing evidence links exposure of particulate matter smaller than 2.5 micro-meters (PM2.5) to significant adverse health effects. Design of effective control strategies, however, awaits improved scientific understanding of the routes of exposure and mechanisms of injury. Although most PM2.5 sources originate in outdoor air, exposure studies find that personal PM2.5 exposures are more tightly correlated with concentrations indoors than outdoors. Here, we describe a project to develop a semi-mechanistic model that captures the relevant physio-chemical processes affecting indoor PM2.5 concentrations of outdoor origin. A discussion of experimental work underpinning the model focuses on development of instrumentation for and measurements conducted at a house in San Joaquin Valley, CA. Results from this work show that indoor concentrations of PM2.5 were significantly lower that outdoor concentrations, in large part, because volatile particulate species vaporized upon entry into the building. Initial results from the application of a size dependent model of transient non-volatile particles confirms the need for additional particle loss mechanisms. By incorporating additional physio-chemical processes and the effects of building occupancy into the model we expect to produce a testable probabilistic model that may eventually be used to help design PM2.5 control strategies for regional housing stocks.