Abstract: In order to assess the safety performance of Kerjian bentonite as one of the engineering barrier materials, a process-based TOUGHREACT modeling approach to predict reactive transport of Pu was employed in this paper, following the previous topic on geochemical simulation of bentonite evolution. This approach, which combines hydrodynamic modeling and geochemical modeling, has a style of dynamic coupling of multiple processes, such as fluid flow, molecular diffusion, mineral dissolution and precipitation, sorption via surface complexation, and radioactive decay. Specifically, reactive transport modeling was performed using TOUGHREACT and other simulators to quantify the complex interplay between Pu transport and reaction processes over a long period of time. The surface complexation model of Pu sorption on bentonite was constructed based on the diffusion layer model (DLM) and literature data, from which the equivalent distribution coefficient (K_d) was subsequently derived. Eventually, the migration of Pu in the Kerjian bentonite layer was simulated in terms of aqueous Pu species, surface complexation model, sorption and retardation, and transport patterns. The results show that in the context of the pH evolution range (i.e. 8.1-10.3) in groundwater, Pu(Ⅳ) is the dominant oxidation state in aqueous phase, which is generally considered to be readily sorbed on solid surfaces and thus relatively insoluble and immobile in the environment. Meanwhile, Pu(OH)₄(aq) species dominates in groundwater due to the formation of stable aqueous complexes, suggesting more sophisticated Pu transport mechanisms. According to the simulated K_d-pH curve of Pu in bentonite, the equivalent K_d can be as high as 16 500 mL/g and 14 500 mL/g, corresponding to the pH values of 8 and 9.5 respectively, indicating that the bentonite has a strong effect on retarding Pu migration. Due to the low permeability and strong sorption properties of bentonite in the normal scenario, the diffusive migration of Pu is less than 1 m in distance over long timescales (e.g. 500 000 years). However, in the conservative scenario, where the strong retardation caused by sorption is ignored and diffusion with hydrodynamic seepage becomes the dominant mechanism for Pu migration, Pu tends to penetrate the bentonite layer in a relatively short period of time (e.g. less than 10 000 years), implying a loss of barrier performance over time. In view of the inevitable uncertainties for longer timescales, it is suggested that more attention should be paid to the investigation of other processes and scenarios dominated by both water flow and molecular diffusion mechanisms. In these cases, the topics include bentonite/clay erosion, colloid mobility, colloid-facilitated radionuclide transport, and multi-field coupled processes, etc. In summary, the above achievements provide important insights into Pu transport behavior with the geochemical evolution of bentonite and groundwater properties.