Chamber experiments and model calculations were performed in order to assess the protective effect of deposition of hydrogen fluoride (HF) on buildings and their contents. In a sealed exposure chamber (0.93 m3), plates or pieces of vinyl wall paper on plasterboard, plastic carpet of PVC or cotton textile were exposed to HF at initial concentration of 80 ppm. By measuring the concentration decrease with time, the transfer velocity and slope of adsorption isotherm were determined by fitting a amodel to the data. Model predictions for an initial concentration of 200-300 ppm were in good agreement when compared with validation experiments at the same initial concentration. Measured transfer velocities (1.3 x 10-3 - 4.6 x 10-3 m s-1) were larger than previously measured values for chlorine and ammonia and also larger than the expected maximum transfer velocity for dwellings (<10-3 m s-1). Measured values of the slope of the adsorption isotherm were similar to or larger than previously measured values for chlorine and ammonia, which means that there may be a tendency to higher desorption of HF than for chlorine and ammonia. Application of the validated model to predict the indoor concentration adn toxic load (c ii t) in dwellings following an outdoor HF release, shows that buildings give significant protection against accidental releases of HF. This is because of the deposition of HF on surfaces and material normally found in dwellings. Increased protection can be achieved by reducing the ventilation and/or increasing indoor turbulence, e.g. by fans. Increasing the amount of adsorbing material in the room, e.g. textile, vould also improve the protection. Increased ventilation by opening door and windows, when the HF cloud has passed outdoors, will further reduce the effect of the HF cloud.