Impact of physical and physicochemical properties of supplementary cementitious materials on structural build-up of cement-based pastes

Abstract

Understanding the structural build-up is relevant to new technologies that require workability control, such as 3D concrete printing. The effect of supplementary cementitious materials (SCM) particle size (i.e. specific surface area and particle density) and physicochemical behavior (i.e., chemical reactivity and surface potential) and their interactions with the primary mixture parameters (i.e., water to cement ratio, cement reactivity and SCM replacement) on the structural build up of cementitious paste were studied. The growth of static yield stress with time was used to characterize the structural build up of cementitious pastes. Calorimetric curves were also measured. Results showed that the effect of SCM on structural build up is governed by their particle density and surface potential. The structural build up can be raised by increasing the number of contact points (governed by particle density and w/c), the growth rate of C-S-H bridges (governed by surface potential), or the reactivity of cement.

Introduction

The workability of fresh concrete is a very important property that determines how well the material can be mixed, placed, consolidated, and finished. The cementitious paste and its interaction with the aggregates govern the workability and its evolution with time. As such, rheological measurements of the cementitious paste are reasonable indicators of concrete workability [1].

Fresh cement-based materials (CBM) [2], as well as other yield stress fluids [3,4], exhibit dynamic and static yield stresses. The dynamic yield stress (τ₀^D) is the minimum stress required for maintaining flow and is commonly obtained from the equilibrium flow curve. The static yield stress (τ₀^S) is the stress required to initiate flow [5]. Due to the structural states of these conditions, τ₀^S is expected to be higher than τ₀^D [6].

CBM can be defined as yield stress fluids [7]. The main characteristic of these materials is that they behave like solids when the applied stress is below a critical shear stress value, called yield stress, and like liquids if the applied stress is above the yield stress [8]. The most common method to measure the rheology of CBM is the equilibrium flow curve, plotted as the equilibrium shear stress versus the shear rate [6]. The yield stress in this method is equal to the shear stress at a shear rate of zero. Among the constitutive equations that have been proposed to represent the equilibrium flow curve of CBM, the Bingham model [9], the Herschel–Bulkley model [10], and the modified Bingham model [11] are the most common.

The dynamic yield stress, τ₀^D of CBM has been widely studied [6,[12], [13], [14]] and been related to measurements from common field tests, such as the slump flow test [15,16]. On the other hand, in the last decade, with the growing use of self-consolidating concrete (SCC) and development of new technologies that require better workability control [17,18], such as 3D printing of concrete, more attention has been paid to the static yield stress, τ₀^S.

It has been shown that τ₀^S of fresh CBM at rest increases over time [7,[19], [20], [21]]. Within a few minutes at rest, the cement particles flocculate and form a three-dimensional network due to electrostatic and van der Waals forces [22,23]. In addition, nucleation of C-S-H at the pseudo-contact points between cement and other particles during the dormant period [24,25] generates formation and growth of C-S-H bridges, making τ₀^S of the material to increase.

The term “thixotropy” is often used to describe the increase of τ₀^S of fresh CBM. One of the main properties of thixotropy is that it is a reversible process. However, for CBM at rest, both reversible (i.e., flocculation) and irreversible (i.e., cement hydration bonding, such as C-S-H bridges) processes happen simultaneously [25], and it is not simple to separate these two effects. As such, for CBM, structural build-up (A_thix), which involves both reversible and irreversible processes, is a more accurate term than thixotropy [26].

A_thix is an important property that affects the constructability and performance of concrete structures. Previous researchers [27,28] have found that for CBM, setting time is highly influenced by Athix. In the case of SCC, formwork pressure [21,29,30], concrete stability after casting [31,32], and bond strength in multi-layer casting [[33], [34], [35]] are governed by this property. It is also important to control A_thix in 3D-printed concrete in order to assure adequate bonding between the printed layers and adequate strength and stiffness for the layers to withstand the weight of the layers above [[36], [37], [38], [39]].

To accurately measure the A_thix of cementitious paste at rest, the sample should not be distressed during the test. Previous authors have used either small-amplitude oscillatory shear (SAOS) tests [25,40] or growth of static yield stress tests [24,27,[41], [42], [43]] to measure the development of A_thix. SAOS is considered a non-destructive test, which is less invasive and less prone to distressing the sample. On the other hand, the growth of static yield stress test is the most common technique, which according to Yuan et al. [26], provides appropriate testing parameters and gives comparable results with the SAOS test.

The A_thix of cementitious paste is affected by mixture parameters, such as the mixture composition, constituent properties, ambient conditions, and shear stress history. Previous studies have shown that increased water-to-cementitious materials volumetric ratio (w/c) reduces A_thix [25,44]. Cement properties, such as fineness [45], electrokinetic behavior [25], and chemical composition [46], also affect the change of τ₀^S over time. Temperature has been shown as an important factor as well [41,47,48], where a nonlinear increase in A_thix was observed in the range of 10 to 40 C° [41]. Additionally, Ma et al. [49] have shown that the A_thix of cement paste is highly affected by the pre-shear and rest condition of the sample.

In current construction technology, most of the concrete produced worldwide contains chemical admixtures, such as viscosity enhancing admixtures and high-range water-reducing admixtures, to improve the rheological properties of concrete and reduce the amount of cement. Several studies [44,49,50] have shown that the dosage and type of chemical admixtures have considerable effects on the rate of A_thix. However, the effectiveness of a chemical admixture depends on many factors, including the cement chemical composition, the mixture design and the mixing process [51]. Therefore, it is difficult to predict and control the effect of a chemical admixture on the rheological properties of concrete, which hinders concrete technologies that require better control of workability (such as 3D printing).

Supplementary cementitious materials (SCMs), also known as mineral admixtures (e.g., fly ash, silica fume), considerably affect the rheology of CBM as well. Most of the previous studies [1,2,[52], [53], [54], [55], [56], [57]] have investigated the effect of SCMs on τ₀^D and viscosity of cementitious materials. In comparison, only a few researchers have studied the effect of SCMs on A_thix [41,43,45,58,59]. Importantly, contradictory results have been found between different studies on this topic. For example, some researchers [59] have found that the use of fly ash and silica fume increases the Athix of SCC, while others [46] have found the opposite. A possible explanation of this discrepancy could be the packing density of particles, which depends on the range of the particle size distribution of the whole solid particles network [60]. To the best of the authors' knowledge, relationships for the physical and physicochemical properties of SCMs with the A_thix of cementitious paste have not been investigated.