This work investigates the use of polymers during CO 2 WAG using a physical model. Coreflooding experiments were conducted to compare polymer assisted WAG, conventional WAG, continuous CO 2 flooding, and polymer flooding. The effect of miscibility in the CO 2 -oil interface was evaluated from coreflooding at two different pressures, above and below the MMP of 1563 psig The MMP was experimentally determined using a fast version of the slim tube procedure with a shorter column. The influence of reservoir heterogeneity was observed by conducting corefloodings in homogeneous and heterogeneous rocks with permeability ranging from 13 to 1300 md. The core plugs were selected using computed tomography to ensure the presence or absence of heterogeneous features.Our results show that under miscible displacement in a homogeneous rock, continuous CO 2 injection can reduce residual oil saturation (S or ) to less than 10%, even for a permeability as low as 13 md. The implementation of conventional WAG and polymer assisted WAG in this case was detrimental for the process as water saturation shielded a portion of the oil preventing CO 2 contact. The departure from the ideal miscible displacement resulted in a S or as high as 32 %. For the heterogeneous rock, at 285 psig above MMP, the WAG CO 2 was able to reduce S or to as low as 11%, whereas at 260 psig below MMP a S or of 23% was reached. As the level of heterogeneity increases, the oil recovery of polymer assisted WAG relative to conventional WAG increased suggesting that a certain degree of heterogeneity is needed for the polymer assisted WAG process to be beneficial. A throughout discussion on the recovery mechanisms and the interactive / combined role of miscibility, heterogeneity, permeability and viscosity is presented.This work adds to the understanding of the CO 2 flooding process implemented in the North Burbank Unit (NBU) and the potential to use WAG and polymer assisted WAG to improve mobility control. Sweep efficiency is the most important challenge from a reservoir engineering standpoint for CO 2 flooding, but it is particularly critical in the NBU for two reasons. The high vertical heterogeneity in the field, with a permeability variation of three orders of magnitude within 30 ft of reservoir thickness, which greatly exacerbates viscous fingering, and the use of anthropogenic CO 2 which makes the economics of the project more sensitive.