Carbon capture and storage (CCS) is a technology that has been proposed to reduce what are perceived to be excessive concentrations of carbon dioxide (CO2) in the atmosphere. CCS will require the transportation of CO2 from the capture locations to the storage sites via pipelines. To ensure safety, an accurate prediction of CO2 decompression following pipeline fracture is crucial for the design and operation for these projects. A multi‐phase CO2 pipeline decompression model using computational fluid dynamics (CFD) techniques is presented in this paper. The GERG‐2008 equation of state (EOS) is employed to describe the properties of vaporand liquid phases. A phase change model using a mass transfer coefficient to control the inter‐phase mass transfer rateis implemented into the CFD code. By varying the mass transfer coefficient, the effect of delayed nucleation on the decompression wave speed can be investigated. The proposed multi‐phase CFD decompression modelis validated against the experimental data from a shock tube test. The performance of the proposed model is also compared with that of the Homogeneous Equilibrium Model (HEM). In addition, the influence of delayed nucleation on CO2 decompression characteristics is discussed and an optimum mass transfer coefficient for CO2 depressurization is obtained. © 2017 Society of Chemical Industry and John Wiley & Sons, Ltd.