Lung functional magnetic resonance imaging (MRI) has become a reality using different inert hyperpolarized gases, such as 3He and 129Xe, which have provided an extraordinary boost in lung imaging and has also attracted interest to other chemically inert gaseous contrast agents. In this context, we have recently demonstrated the first diffusion-weighted images using thermally polarized inhaled sulfur hexafluoride (SF6) in small animals. The aim of this study was to evaluate whether or not the diffusion coefficient of this fluorinated gas is sensitive to pulmonary structure, gas concentration and air pressure in the airways. Diffusion coefficients of SF6 (both pure and in air mixtures) measured in vitro at different pressures and 20 degrees C showed an excellent agreement with theoretical values. Measurements of diffusion coefficients were also performed in vivo and post-mortem on healthy rats, achieving satisfactory signal-to-noise ratios (SNRs), and SF6 gas was found to be in an almost completely restricted diffusion regime in the lung, i.e., the transport by molecular diffusion is delayed by collisions with barriers such as the alveolar septa. This observed low diffusivity means that this gas will be less sensitive to structural changes in the lungs than other magnetic resonance sensitive gas such as 3He, particularly at human scale. However, it is still possible that SF6 plays a role since it opens a new structural window. Thus, the interest of researchers in delimiting the important limiting technical factors that makes this process very challenging is obvious. Among them, T2 relaxation is very fast, so gradient systems with very fast switching rate and probably large radiofrequency (RF) power and high field systems will be needed for hexafluoride to be used in human studies.