Water transport in cements: A bottom-up approach
Concrete is an inherently low energy input material (600-800 MJ/tonne) comparable to wood (500 MJ/tonne). However, the enormous quantities used worldwide mean that it accounts for at least 5% of global CO2 production with demand for cement set to double / treble by 2050.
Water movement in concrete is a key factor influencing the long term performance and degradation of infrastructure by both physical and chemical means. Moreover, water is a key constituent of cement, the primary binder phase of concrete. However, remarkably, there is as yet no clear understanding of pore-water interactions in cements. Equally there is no good predictor of water transport in concrete. To gain this understanding will be to achieve a critical step towards predicting the long-term performance of concrete and the design of new cement with lower cement CO2 emissions.
Recent advances in nuclear magnetic resonance (NMR) relaxometry have opened an entirely new window to our understanding of pore water interactions and dynamics in cements at the nanoscale with identification of dynamics on timescales of 1 ns, 10 s and 5 ms. Equally, there have been impressive advances in numerical modelling of cement microstructure based on advances in other spectroscopies and microscopies. Coupling the two creates new opportunity to understand, and hence create predictive capability for water transport in cements from the atomic scale upwards.
A molecular dynamics simulation has been set up to model the two-dimensional cement pore. We have determined the NMR relaxation times from the simulations and we are seeking to discover how the relaxation times dependent on pore dimensions.
We plan to determine the diffusion coefficient of water which is surface bound and in bulk cash for the purpose of upscaling. A parallel computer simulation using the Monte Carlo technique is also under development.