Cement is the most produced material in the world. A major player in greenhouse gas emissions, it is the main binding agent in concrete, providing a cohesive strength that rapidly increases during setting. Understanding how such cohesion emerges is a major obstacle to advances in cement science and technology. Here, we combine computational statistical mechanics and theory to demonstrate how cement cohesion arises from the organization of interlocked ions and water, progressively confined in nanoslits between charged surfaces of calcium-silicate-hydrates. Because of the water/ions interlocking, dielectric screening is drastically reduced and ionic correlations are proven notably stronger than previously thought, dictating the evolution of nanoscale interactions during cement hydration. By developing a quantitative analytical prediction of cement cohesion based on Coulombic forces, we reconcile a fundamental understanding of cement hydration with the fully atomistic description of the solid cement paste and open new paths for scientific design of construction materials.

Monodisperse suspensions of Brownian colloidal spheres crystallize at high densities, and ordering under shear has been observed at densities below the crystallization threshold. We perform large-scale simulations of a model suspension containing over [Formula: see text] particles to quantitatively study the ordering under shear and to investigate its link to the rheological properties of the suspension. We find that at high rates, for [Formula: see text], the shear flow induces an ordering transition that significantly decreases the measured viscosity. This ordering is analyzed in terms of the development of layering and planar order, and we determine that particles are packed into hexagonal crystal layers (with numerous defects) that slide past each other. By computing local [Formula: see text] and [Formula: see text] order parameters, we determine that the defects correspond to chains of particles in a squarelike lattice. We compute the individual particle contributions to the stress tensor and discover that the largest contributors to the shear stress are primarily located in these lower density, defect regions. The defect structure enables the formation of compressed chains of particles to resist the shear, but these chains are transient and short-lived. The inclusion of a contact friction force allows the stress-bearing structures to grow into a system-spanning network, thereby disrupting the order and drastically increasing the suspension viscosity.

Like-charge attraction, driven by ionic correlations, challenges our understanding of electrostatics both in soft and hard matter. For two charged planar surfaces confining counterions and water, we prove that, even at relatively low correlation strength, the relevant physics is the ground-state one, oblivious of fluctuations. Based on this, we derive a simple and accurate interaction pressure that fulfills known exact requirements and can be used as an effective potential. We test this equation against implicit-solvent Monte Carlo simulations and against explicit-solvent simulations of cement and several types of clays. We argue that water destructuring under nanometric confinement drastically reduces dielectric screening, enhancing ionic correlations. Our equation of state at reduced permittivity therefore explains the exotic attractive regime reported for these materials, even in the absence of multivalent counterions.

During cement hydration, C−S−H nanoparticles precipitate and form a porous and heterogeneous gel that glues together the hardened product. C−S−H nucleation and growth are driven by dissolution of the cement grains, posing the question of how cement grain surfaces induce spatial heterogeneities in the formation of C−S−H and affect the overall microstructure of the final gel. We develop a model to examine the link between these spatial gradients in C−S−H density and the time-evolving effective interactions between the nanoparticles. Using a combination of molecular dynamics and Monte Carlo simulations, we generate the 3D microstructure of the C−S−H gel. The gel network is analyzed in terms of percolation, internal stresses, and anisotropy, and we find that all of these are affected by the heterogeneous C−S−H growth. Further analysis of the pore structure encompassed by the C−S−H networks shows that the pore size distributions and the tortuosity of the pore space show spatial gradients and anisotropy induced by the cement grain surfaces. Specific features in the effective interactions that emerge during hydration are, however, observed to limit the anisotropies in the structure. Finally, the scattering intensity and specific surface area are computed from the simulations in order to connect to the experimental methods of probing the cement microstructure.