Due to their mobility and toxicity, crude oil volatile organic compounds (VOCs) are representative components for oil pipeline contaminated sites detection. Therefore, contaminated location risk assessment, with airborne light detection and ranging (LIDAR) survey, in particular, requires ground-based determinative methods for oil VOCs, the interaction between oil VOCs and soil, and information on how they diffuse from underground into atmosphere. First, we developed a method for determination of crude oil VOC binary mixtures (take n-pentane and n-hexane as examples), taking synergistic effects of VOC mixtures on polydimethylsiloxane (PDMS) solid-phase microextraction (SPME) fibers into consideration. Using this method, we further aim to extract VOCs from small volumes, for example, from soil pores, using a custom-made sampling device for nondestructive SPME fiber intrusion, and to study VOC transport through heterogeneous porous media. Second, specific surface Brunauer-Emmett-Teller (BET) analysis was conducted and used for estimation of VOC isotherm parameters in soil. Finally, two models were fitted for VOC emission prediction, and the results were compared to the experimental emission results. It was found that free diffusion mode worked well, and an empirical correction factor seems to be needed for the other model to adapt to our condition for single and binary systems.