Abstract
A growing demand of concrete in the construction industry has resulted in an immense
increase of global cement production. As a consequence, the world sees a high amount
of energy consumption and carbon footprint generated from cement industry. To defeat
this environmental crisis, the behavior of water in cement at the nanoscale is imperative.
The properties of cement at the nanoscopic level is closely linked to the water dynamics.
A powerful method known as 1H Nuclear Magnetic Resonance (NMR) can be used to
probe the nanodynamics of spins in fluids contained in cement. 1H NMR measures the
spin-lattice or spin-spin relaxation rates due to the relative motion of pairs of spins in
cement nanopores. However, the interpretation of NMR dispersion curves is difficult due
to the complexity of fluid movement in cement.
The work in this thesis presents Molecular Dynamic (MD) simulations as a method to
understand the water and ion transport properties atomistically. MD simulations assist
the development of new NMR theoretical models to interpret the NMR dispersion curve.
Starting with the investigation of pure water system, MD simulations validates the probability density function of 1H–1H water vector and the L´evy rotor model is used to estimate
the intramolecular contribution to the longitudinal relaxation rate. Accordingly, the model
from pure water is applied to the ion-1H vector and validated by MD simulations confirming the resemblance of Brownian motion for aquoions. Model fitting to NMR experimental
data from aquoions provide a good understanding of the dynamical processes in the relaxation rate response. Last but not least, cement hydration is explored by quantifying the
key dynamics of water under the influence of calcium ions through the planar pore model
simulations. The role of calcium in slowing down the water dynamics in sub–3 nm cement
nanopores, which may have a secondary effect to the relaxation rate, is also observed.