Abstract Carbon capture and storage has attracted attention as a feasible solution to global warming caused by anthropogenic emissions of greenhouse gases. The injection wells and well cement provide the wellbore integrity necessary for the long-term storage of carbon dioxide (CO 2 ). To date, ordinary Portland cement (OPC) has been used in injection wells, and its survival has been questioned as it is unstable in CO 2 -rich environments. Therefore, an experimental study was conducted to study geopolymer (G) as well cement and sandstone (S) as formation material. The sub- and super-critical CO 2 permeability of geopolymer, sandstone and G–S composites were studied using the high pressure triaxial set-up in the Department of Civil Engineering, Monash University. The undrained triaxial experiment was conducted at confining pressures from 14 to 26 MPa, and inlet pressures from 6 to 20 MPa to study the sub- and super-critical CO 2 permeability of wellbore materials. Based on the experimental results, the apparent CO 2 permeability of sandstone (0.8–30 μD) is approx. 1000 times higher than that of geopolymer (0.002–0.02 μD). The increase in pore pressure reduces the permeability of geopolymer, sandstone and G–S composite materials, and this is related to Klinkenberg's slip flow. In addition, the apparent permeability of CO 2 reduces due to pore volume shrinkage caused by the increase in confining pressure. The percentage permeability reduction (per 1 MPa increase in downstream pressure) of geopolymer, sandstone and G–S composite materials reduces with increase in pore pressure, and the reduction is significant from sub-critical to super-critical CO 2 pressure conditions. This observation shows the significance of super-critical CO 2 pressure conditions for effective and leak-free storage of CO 2 in deep underground reservoirs.