The accurate prediction of heat transfer behavior in packed bed is of great importance to the design and operation of this kind of facility. In this paper, heat transfer between packed bed and adjacent wall in heat recovery processing is investigated theoretically and experimentally. A bed-to-wall heat transfer model is presented with consideration of two essential components including particle-to-wall contact heat transfer and heat conduction through the thermal penetration layer. Details of model parameter determination are provided. Through comprehensive sensitivity analysis of those parameters, it is shown that particle temperature, particle residence time and particle diameter are critical parameters to heat transfer coefficient. Besides, surface roughness, solid thermal conductivity and environmental gas species are all influencing factors of the overall heat transfer coefficient. In particular, a dimensionless particle residence time is introduced to characterize the combining effect of residence time, effective thermal conductivity and wall contact heat transfer. Meanwhile, a revised simplified formula for calculating particle contact heat transfer coefficient is correlated as well, which provides better prediction than the one widely adopted in previous studies. Furthermore, experimental heat transfer test on an industrial packed bed char cooler is performed for validating the model predicting capability in heat recovery process. The systematical test on a bench scale packed bed heat exchanger has also been carried out. Parametric analysis of the experimental data is consistent with the conclusions drawn from the model analysis section. The model provides good predicting accuracy on the overall heat transfer coefficient from the bench scale facility as well as data from the field test by correlating 95% of the measured data within 20%. The implementation of the proposed model shows good prediction of heat transfer coefficient in packed bed for heat recovery purpose. It is recommended that the model can be applied in industrial design. (C) 2013 Elsevier B.V. All rights reserved.