Proceedings of the International rilem conference Materials, Systems and Structures in Civil Engineering 2016

21

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

NANOSCALE SIMULATIONS OF CEMENT FORMATION AND 

STRUCTURAL EVOLUTION: A NEW KINETIC APPROACH 

 

Enrico Masoero 

(1)

, Igor Shvab 

(1)

 

 

(1) Newcastle University, Newcastle upon Tyne, United Kingdom 

 

 

 

 

 

 

 

Abstract 

The formation and degradation of cementitious materials are largely controlled by the 

nanostructural evolution of hydration products. Modelling such evolution across multiple 

length and time scales is a great challenge. Here we present a new kinetic approach to 

simulate the nucleation, dissolution, and aggregation of cement hydrates nanoparticles. The 

approach is based on a Kinetic Monte Carlo algorithm in which the rates of the transitions are 

obtained via a new coarse-graining procedure. The rates account for free energy changes due 

to both mechanical interactions and chemical reactions. The methodology is able to address 

the long timescale of cement formation and captures various possible mechanisms of 

nucleation and growth of the hydrates. By coupling chemistry, mechanics, and long 

timescales, this work is a first step towards simulating cement hydration and degradation. 

 

 

1. Introduction 

 

The hardened cement paste is the glue of concrete. It forms upon chemical reaction between 

dry cement powder and water [1]. This hydration process leads to the precipitation from ionic 

solution of several hydrated phases (HP). The HP progressively fills the space and induces the 

liquid-to-solid transition known as setting. In ordinary cement pastes, the main HP is calcium-

silicate-hydrate (C—S—H) and at least 50% of the total hydration reaction takes place during 

the first 24 hours after mixing dry cement with water. Setting typically occurs during this 

stage, which is known as “early hydration” [2]. 

 

There is a considerable scientific and technological interest in controlling the early hydration 

and setting of the cement paste. In the last decades, this led to a number of models and 

simulation approaches whose target is to predict the early hydration based on the chemistry 

and mix design proportions of the paste. These models consider length scales above the  m 

and use a combination of thermodynamics and chemical kinetics in order to reproduce the 

---

22

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

microstructural evolution of the paste [3-8]. Their results can fit well the rate of early 

hydration from isothermal calorimetry experiments. These models however have the 

limitation of considering the HP as homogenous domains, without information on the 

underlying texture at sub/micrometre length scales (except for user-defined parameters that 

sometimes are employed to mimic the anisotrpic formation of needles and foils [9]). Missing 

the sub-micrometre texture is a particularly limiting for the C—S—H hydration product, 

whose network of mesopores with size 1-50 nm dictates largely the macroscopic response to 

humidity cycles and creep [10-14]. 

 

Simulating the formation of HP at the sub-micrometre level is a challenging task. In the last 

decade, several authors have shown that aggregating nano-units of 1-10 nm lead to model 

structures of the cement HP that display many of the experimentally measured structural 

features and mechanical properties [15-23]. Some of these models have also attempted to 

describe the process of HP formation by precipitation from solution. However, the timescale 

of cement hydration is of the order of 24 hours while nanoparticle simulations are limited to 

dynamic processes in the timescale of nano-to-micro seconds or, on the other extreme, to 

equilibrium studies in the infinite-time limit. To overcome this limitation, the simulations to 

date have either introduced constraints on the mechanisms of particle aggregation, or applied 

ad-hoc nonlinear mapping between simulation steps and time [17, 19-21]. None of the 

simulations to date can predict the mechanisms of HP nanoparticle formation and aggregation 

directly from the chemistry of the aqueous solution. 

 

In this work, we propose a new approach to simulate the formation of cement HP at the 

mesoscale of 1-to-500 nm. Our approach is based on Kinetic Monte Carlo (KMC) simulations 

of nanoparticle insertion and deletion. The rates of insertion and deletion are calculated using 

a new coarse graining scheme that considers both Classical Nucleation Theory (CNT) and 

crystal growth theory. First results show that the rates can address the long timescale of 

cement hydration and consider the interplay between chemical driving force and mechanical 

interactions between the nanoparticles.   

 

 

2. Methodology 

 

The simulations of HP formation start with an orthogonal box that is empty except for a 30 

nm thick layer that represents the surface of a cement grain (see Figure 1). The cement layer 

is discretized using 10 nm spherical particles that are fixed, i.e. not displaced nor removed 

during the simulations. It is implicitly assumed that the box is filled with aqueous ionic 

solution, whose supersaturation   with respect to HP formation is known. 

 

After a certain number of KMC steps, a certain number of HP spherical particles will have 

formed in the box (see Figure 1). The next KMC step computes first the rates of all possible 

particle deletions and then the rates of all possible particle insertions. The insertion is tricky 

because there are infinite possible positions for a new particle. In order to manage this, we 

create many trial HP particles and allow them to move a bit in order to find a local minimum 

of interaction energy with the other existing particles (HP and layer; no interactions between 

two trial particles, because they do not exist yet; see Figure 1).  

---