Wei-Shan Chiang

Multiscale structure of calcium- and magnesium-silicate-hydrate gels

Concrete is the worlds most widely used building material. However, the production of CaO-based cements generates large amounts of anthropogenic emissions of CO2. Among different strategies to reduce CO2 emissions, newly developed MgO-based cements, though currently suffering from inferior mechanical properties, are some of the most promising and attractive options. By combining wide- and small-angle X-ray scattering and electron microscopy, we identified differences in the multiscale structure of the two main binding phases: the calcium-silicate-hydrate (C-S-H) gel for CaO-based cements and the magnesium-silicate-hydrate (M-S-H) gel for MgO-based cements. We found the primary unit at the nanoscale level of C-S-H to be a multilayer disk-like globule, whereas for M-S-H it is a spherical globule. These prominent differences result in diverse microstructures, leading to disparities in mechanical properties and durability for the associated cements. Modulating the M-S-H structure and enhancing the compatibility between C-S-H and M-S-H will be the key to improve the robustness of eco-friendly MgO-based binders.

Microstructural changes of globules in calcium-silicate-hydrate gels with and without additives determined by small-angle neutron and X-ray scattering

The microstructure of calcium-silicate-hydrate (C-S-H) gel, a major hydrated phase of Ordinary Portland Cement, with and without polycarboxylic ether (PCE) additives is investigated by combined analyses of small-angle X-ray scattering (SAXS) and small-angle neutron scattering (SANS) data. The results show that these comb-shaped polymers tend to increase the size of the disk-like globules but have little influence on the thickness of the water and calcium silicate layers within the globules. As a result, the fractal packing of the globules becomes more open in the range of a few hundred nanometers, in the sense that the mass fractal dimension diminishes, since the PCE adsorption on the globules increases the repulsive force between and polydispersity of the C-S-H units. Moreover, scanning electron microscope (SEM) study of the synthesized C-S-H gels in the micrometer range shows that the PCEs depress the formation of fibrils while enhancing the foil-like morphology.

Boson Peak in Deeply Cooled Confined Water: A Possible Way to Explore the Existence of the Liquid-to-Liquid Transition in Water

In their Letter, Wang et al. [1] report on an inelastic neutron scattering (INS) experiment where they describe the pressure evolution of a low energy (E ∼ 6 meV) excitation, emerging in confined protonated water only below 230 K at an exchanged momentum Q 1⁄4 2.0 Å−1. Water confinement was used to overcome the unavoidable crystallization occurring below ∼250 K in bulk water. The authors report that a similar finding was also obtained in both bulk (numerical simulations [2]) and confined water at ambient pressure. They refer to this low temperature excitation as a boson peak (BP) [3], and relate its occurrence to the Widom line, concluding that the observed pressure behavior of the BP reveals the signature of the high-density liquid (HDL) to the low-density liquid (LDL) transition proposed [4], though severely questioned [5], for bulk water. We believe these claims to be unconvincing for the following reasons. Comparison with corresponding findings in liquid water.—The authors not only overlook commenting on the actual density of confined liquid water [6–9], but they also neglect to establish any physical relationship with the well known excitations of coherent or incoherent origin occurring at similar energies in bulk liquid water. Since the seminal Raman scattering room temperature studies by Bolla [10], a mode in the ∼5–7 meV range has indeed been regularly observed with optical [11–14], numerical [15], and inelastic x-ray scattering [16–18] and INS techniques [19,20] over a wide thermodynamic range (0–2 kbar, 250–450 K) in both H2O and D2O (see Fig. 4 in Ref. [21]). The microscopic nature of such a mode, underdamped and still well defined at Q 1⁄4 2.0 Å−1, is the subject of controversial single-particle [12,22,23] or collective [21,24] interpretations. Irrespective of its incoherent or coherent nature, this evidence is unquestionable and cannot be ignored. This mode is not easily detectable in high temperature neutron spectra from H2O because of the overwhelming quasielastic contribution. However, its presence always emerges in calculating the hydrogen vibrational density of states, as was done in Ref. [22] at T 1⁄4 256 K, and in bulk or confined H2O from 300 K down to 242 K [25], but not mentioned in Ref. [1]. This mode, but not the BP, was also observed when investigating the vibrational dynamics in amorphous ices [26,27]. Moreover, a bulklike excitation not dependent on temperature was observed down to 205 K in an INS measurement on slightly salty liquid water [28]. Data analysis and treatment.—(i) INS probes at the same time the coherent and incoherent properties of matter with a weight given by their respective neutron cross section and dynamic structure factor. H2O is considered as an incoherent scatterer by reason of the high σinc=σcoh ratio. Yet, this approximation cannot be uncritically adopted as was done in Ref. [1] and a proper estimation of the related ratio SincðQ;ωÞ=ScohðQ;ωÞ at the thermodynamic (P, T) and kinetic (Q;ω) investigated point should be addressed. (ii) An arbitrary interpolating metric is adopted to determine the locus of the BP appearance: the TB parameter is a clumsy, large-error quantity inherent in the slowing down of the thermal diffusion. The peak associated with the low energy—and virtually temperature independent (see Fig. 4 of Ref. [1])—excitation is enhanced by the narrowing of the quasielastic signal upon lowering the temperature. (iii) The exact internal pressure existing in such tiny pores is not directly related to the He applied pressure and is therefore unknown [29]. As a consequence, the confined water phase diagram and properties cannot be unconditionally assigned to those of bulk water. (iv) In order to support the authors’ claims at a less speculative level, the correct BP shape should be determined by calculating the vibrational density of states in excess of that of the corresponding crystalline phase. In conclusion, the whole large body of numeric and experimental investigations on the single particle and collective properties of liquid water report the presence of a weakly dispersing excitation in the 5–7 meV range. We believe that, in order to use the BP as a marker of the HDL or LDL bulk water phases, the authors should perform a more complete data treatment and establish a relation, if any, between the supposed BP peak they observe in a confined environment and the well established bulk mode present in a wide portion of the phase diagram at the same energy.

Dynamic crossover in hydration water of curing cement paste: the effect of superplasticizer.

The influence of a new comb-shaped polycarboxylate-based superplasticizer (CSSP) on the hydration kinetics and transport properties of aged cement pastes has been investigated by high-resolution quasi-elastic neutron scattering (QENS) and low temperature differential scanning calorimetry (LT-DSC). A new method of analysis of QENS spectra is proposed. By applying the refined method we were able to access to four independent physical parameters including the self-diffusion coefficient of the hydration water confined in the cement paste. Mean squared displacement (MSD) of the hydrogen atom for mobile water molecules displays a dynamic crossover temperature in agreement with DSC data. The experimental results indicate that CSSP polymer added into cement paste moderates the hydration process and decreases the dynamic crossover temperature of the hydration water.