Multi-decade, multi-scale modeling of aging basic creep of concrete

This study presents a multi-scale model for predicting multi-decade basic creep of concrete. Aging of cement is modeled through hydration, densification, and polymerization of the calcium-silicate-hydrate (C-S-H) phases. The model accounts for the separate mechanisms of viscoelastic compliance and aging viscous flow of the C-S-H, and for the dissolution-precipitation of elastic and viscoelastic phases during hydration that causes apparent creep in the composite. Upscaling is performed in the time-domain simultaneously for all loading ages. The results show that short-term viscoelastic compliance observed from nanoindentation tests dominates short-term creep, but cannot explain long-term creep rates observed in macroscopic concrete creep tests. Such observations can only be replicated by considering viscous flow that develops over time scales unobservable by minutes-long tests on the microscale. Dissolution creep may explain some irreversible basic creep at very early ages but rapidly diminishes in relevance as the concrete continues to age.