The performance prediction of C-S-H gel is critical to the theoretical research of cement-based materials. In the light of recent computational material technology, modeling from nano-scale to micro-scale to predict mechanical properties of structure has become research hotspots. This paper aims to find the inter-linkages between the monolithic "glouble" C-S-H at nano-scale and the low/high density C-S-H at the micro-scale by step to step method, and to find a reliable experimental verification method. Above all, the basic structure of tobermorite and the "glouble" C-S-H model at nano-scale are discussed. At this scale, a "glouble" C-S-H structure of about 5.5 nm3 was established based on the 11Å tobermorite crystal, and the elastic modulus of the isotropic "glouble" is obtained by simulation. Besides, by considering the effect of porosity on the low/high density of the gel morphology, the C-S-H phase at micro-scale can be reversely characterized by the "glouble". By setting different porosities and using Self-Consistent and Mori-Tanaka schemes, elastic moduli of the low density and high density C-S-H from that of "glouble" are predicted, which are used to compare with the experimental values of the outer and inner C-S-H. Moreover, the nanoindentation simulation is carried out, where the simulated P-h curve is in good agreement with the accurate experimental curve in nanoindentation experiment by the regional indentation technique(RET), thus the rationality of the "glouble" structure modeled is verified and the feasibility of Jennings model is proved. Finally, the studies from the obtained ideal "glouble" model to the C-S-H phase performance has realized the mechanical properties prediction of the C-S-H structure from nano-scale to micro-scale, which has great theoretical significance for the C-S-H structural strengthening research.
Mechanical Properties Prediction of Gray Cast Iron Considering Trace Elements Based on Deep Learning