A comprehensive study on predicting the elastic modulus and Poisson ratio of hardened cement pastes via micro-scale cement hydration simulations

Abstract

A comprehensive study was conducted to predict the elastic modulus and Poisson ratio of hardened cement pastes (HCPs), including numerical work, comparison of two modelling approaches, and experimental validation. Modelling approach 1 (M1) applied a continuum micromechanics model (CMM) whereas modelling approach 2 (M2) used finite element analysis (FEA). This study presents a reliable method for predicting the mechanical properties of HCPs by firstly employing the microstructures of HCPs, obtained by CEMHYD3D analysis with random distribution of cement particles as input parameters for both M1 and M2 approaches. The elastic modulus and Poisson ratio predicted by M1 and M2 approaches were verified through the experimental data obtained from compressive tests combined with digital image correlation analysis. The elastic moduli of HCPs were significantly affected by the input C-S-H modulus, with 90% of reported C-S-H elastic moduli falling within the range of 20.25 GPa and 30.4 GPa. The predicted elastic moduli of HCPs were well-matched with experimental data when assigning the elastic modulus, bulk modulus, and shear modulus, of C-S-H, as 20.25, 12.98, and 8.17 GPa, respectively. A sensitivity analysis demonstrated a linear increase in the elastic modulus of HCPs with increase of C-S-H modulus, with correlation coefficients R² of 0.9985 and 0.9996 for HCPs with water-to-cement ratios (w/c) of 0.3 and 0.5, respectively. Additionally, the predicted results showed greater consistency at later age (after 7 d) compared to earlier stages. The integral absolute error values of the elastic moduli obtained via M1 and M2 for HCPs with w/c of 0.3 were 6.32% and 6.23%, respectively while those for HCPs with w/c of 0.5 were 4.8% and 5.67%, respectively. M1 approach is more recommended since it is straightforward with time and cost efficiency compared to M2 approach in addition to the suitable prediction accuracy.