The mechanisms of heat transfer for the controlled decompression of supercritical (SC) carbon dioxide (CO2) from an extraction vessel have been rarely studied. Previous work on heat and mass transfer during controlled decompression of extraction vessels observed that suggested correlations could be improved if studies were carried out on vessels with better-defined heat transfer areas and using materials with well-defined physical properties. This work studies heat and mass transfer during decompression of a 1-L extraction vessel vented through a partially opened needle valve using a newly adapted system, different flow coefficients, and model materials, including cylinders of sintered stainless steel. Results showed that, opposite to previous experiments, there is a difference between wall and bulk temperatures of the fluid, and data suggested that there might be clamping occurring on the valve when the flow coefficient is low. Correlations were proposed for both vented mass flow, for which adjustment improved as the flow coefficient decreased; and heat transfer through the steel wall. The correlation Nu = 0.086 x Da(-0.037) Ra (0.357) is proposed for the heat transfer through the steel wall for all material studied, which included the dimensionless Darcy number to take into account the properties and geometry of the packed bed, making it suitable for multiple systems. Mathematical simulation was used to verify calculations for the mean temperature of the steel wall of the vessel and results showed differences not higher than 1 degrees C between simulated and calculated data in almost all cases. These results can be used to validate those generated by mathematical simulation, which was shown in this work to be able to reproduce experimental results satisfactorily. A mathematical model can be later used to simulate decompressions of industrial vessels in order to determine the setup times required to replace exhausted by fresh substrate in industrial vessels in studies of minimization of production cost. (C) 2016 Elsevier B.V. All rights reserved.