Search

• A cement hydration model to simulate the hydration of a CNT-reinforced cement paste. • The CNT dispersion algorithm was proposed to generate the pre-hydration model. • Multiphase voxels were proposed to consider the CNT in the CEMHYD3D. • Dissolution rates were proposed to consider accelerated or delayed hydration. • Simulation results are consistent with experimental results. The study proposed a cement hydration model based on multiphase voxels and a novel dissolution rate to simulate the hydration of a carbon nanotube (CNT)-reinforced cement paste. The proposed multiphase voxels and dissolution rates were validated by considering accelerated or delayed hydration owing to CNTs and superplasticizers. The model can simulate the formation of low-density and high-density calcium silicate hydrates (CSHs) using multiphase voxels and was validated using previous experimental data for cement pastes containing 0.05–0.25 wt% CNTs. The simulation results, including those on the heat flow and heat of hydration, are consistent with the experimental results. [ABSTRACT FROM AUTHOR]

• The proposed random distribution method was much saving-time than the correlation function method. • The surface area of silicate phases generated by the RD_13 and the CF methods were quite close. • There were in good agreement between the simulation results using the RD_13 method and experimental results in the OPC with w/c = 0.4 and 0.55. This study investigated the influences of cement particle distribution on the hydration process of cement paste using three-dimensional computer simulation. The pre-hydrated models of cement paste were generated by using both the correlation function (CF) method based on SEM images and the random distribution (RD) method. The hydration process of cement paste was simulated using the model based on the CEMHYD3D. The RD method was simpler and required less time in generating a pre-hydrated model than the CF method. In addition, the simulation results of the RD method were similar to those of the CF method in the properties of cement paste using the cement 133 with water-to-cement ratio of 0.3. Although the simulation results of the RD method was in good agreement with those of the CF method, authors highly recommended that the cement particles larger than a critical diameter (13 µm) in the simulation contain multiple cement crystalline phases whereas the other ones smaller than the critical diameter should contain a single phase. The proposed RD method was validated by using the previously reported experimental results using ordinary Portland cement as well. [ABSTRACT FROM AUTHOR]