Elsevier

Cement and Concrete Research

Volume 48, June 2013, Pages 105-115
Cement and Concrete Research

Early age hydration of calcium sulfoaluminate (synthetic ye'elimite, C4A3S¯) in the presence of gypsum and varying amounts of calcium hydroxide

https://doi.org/10.1016/j.cemconres.2013.03.001Get rights and content

Abstract

Suspensions of synthetic ye'elimite (C4A3S¯) in a saturated gypsum (CS¯H2) and calcium hydroxide (CH) solution were examined in-situ in a wet cell by soft X-ray transmission microscopy and ex-situ by scanning electron microscopy. The most voluminous hydration product observed was ettringite. Ettringite commonly displayed acicular, filiform, reticulated, and stellate crystal habits. Additionally, pastes with C4A3S¯, 15% CS¯H2, and varying amounts of CH were prepared and examined with X-ray diffraction (XRD) and isothermal calorimetry. The XRD experiments showed that increasing CH content caused more solid solution (SO42 /OH) AFm phases to form at early ages (< 1 d) and more monosulfate to form at later ages (> 1 d). Calorimetry indicated that the increased production of solid solution AFm was accompanied with an increase in the initial (< 30 min) rate of heat evolution, and increasing CH generally reduced the time till the second maximum rate of heat evolution due to the formation of ettringite and monosulfate.

Introduction

Calcium sulfoaluminate (CSA) cements contain a significant fraction of synthetic ye'elimite (C4A3S¯1). C4A3S¯ is also referred to as “Klein's Compound” because it was originally identified and developed by Alexander Klein for use in expansive cements [1]. A large range of CSA cements were developed in China in the 1970s under the generic name of “TCS” cements, where TCS stands for “Third Cement Series,” the name being based on the idea that portland cements (PC) were the first series and high-alumina cements (HAC) were the second series in terms of historical development [2], [3]. In fact, C4A3S¯ is chemically equivalent to three units of monocalcium aluminate (CA) plus anhydrous calcium sulfate (CS¯), and in most applications C4A3S¯ plays roughly the same role in hydration as CA does in HAC. C4A3S¯ has the advantage, however, that it is more compatible with PC clinker phases at high temperatures and can thus be used to stabilize high-alumina clinker compositions in the presence of calcium silicates under normal clinkering conditions, thus permitting CSA clinkers to be manufactured in conventional rotary-kiln systems as used for PC clinkers, which significantly reduces the manufacturing costs relative to HAC.

Recently, there has been a renewal in interest in CSA cements from both researchers and industry because of their potential environmental benefits and performance advantages in special applications, and their rise in popularity has been accompanied with some state of the art reviews [4], [5]. CSA cement manufacture requires less limestone to be decarbonated per mass of clinker than for PC manufacture, and the maximum clinkering temperature is also about 200 °C lower. This results in very significant reductions in specific kiln fuel use and CO2 emissions. CSA clinkers also tend to be more friable than PC clinkers and thus require less electricity for grinding [6], [7]. Depending on the CSA clinker composition, particle size distribution, w/c, and the amount of gypsum added, the CSA cement can have similar rheology, workability, set time, dimensional stability, and strength gain to PC, or the CSA cement can be formulated to develop high early strength, or to be shrinkage-compensating or self stressing [4], [5], [8], [9], [10].

It is well known that the most voluminous early hydration product in most CSA cement formulations is ettringite (C6AS¯3H32). In the absence of excess calcium hydroxide (CH), crystalline ettringite is usually found together with smaller amounts of a largely amorphous hydrated alumina gel (AH3) [11]. Eqs. (1), (2) represent hydration reactions producing ettringite in CSA cements [12]. When CS¯H2 is depleted, the formation of monosulfate (C4AS¯H1218) becomes the dominant reaction, Eq. (3).C4A3S¯+2CS¯H2+34HC6AS¯3H32+2AH3C4A3S¯+8CS¯H2+6CH+74H3C6AS¯3H32C4A3S¯+18HC4AS¯H12+2AH3

Many CSA cements give very rapid strength development, and, since ettringite is usually the major hydration phase in the early-age microstructure, the question arises as to how this early strength is generated. The strength of PC systems is generally ascribed to bonding by the C–S–H phase, which like AH3, is largely amorphous. However, at early ages the AH3 gel appears very tenuous and seems unlikely to contribute significantly to strength; therefore, bonding and interlocking between ettringite crystals probably make the major contribution to the strength of the CSA paste. In fact, many quite strong hydraulic cementing systems exist based entirely on crystalline hydrates, for example: gypsum plasters, HAC, magnesium phosphate cements, Sorel cements, etc.

The subject of bonding in poly-crystalline hydraulic binders has been discussed in more detail with respect to the properties of gypsum-based binders. It was suggested that the gypsum crystals formed initially from individual nuclei may bond together in preferred orientations at very early ages when free rotation in the suspension can easily occur [13]. Similar preferred orientations could also in principle form from single nuclei by complex twinning processes at the early stages of crystal growth. Either way, these early preferred orientations should give rise to “domains” in the final hardened structure, the domains referring to contiguous groups of crystals oriented relatively in specific way to each other. Such domains are, however, very hard to observe by the conventional techniques of scanning electron microscopy applied to dense hardened cement pastes. This is because, due to the likely high porosity of such hypothetical domains, they may well interpenetrate to a significant degree with other adjacent domains, with completely random orientations between the interpenetrating domains, giving the impression of a completely random microstructure. Therefore, the best chance to observe the formation of such domains is probably at very early ages and/or in very dilute suspensions. In this work, we studied the hydration of C4A3S¯ in dilute suspensions under conditions as close as possible to those expected in CSA cements used for construction applications. Additionally, C4A3S¯ pastes with 15% CS¯H2 and varying amounts of CH were investigated with isothermal conduction calorimetry and X-ray diffraction (XRD) to better understand CH's effect on the hydration kinetics and hydration products formed.

Section snippets

Materials

A sample of C4A3S¯ was obtained from Construction Technology Laboratories, Inc., Skokie, IL, USA. It was synthesized by heating a stoichiometric blend of finely-ground reagent grade alumina, calcium carbonate and calcium sulfate in an electric furnace at 1000–1100 °C, followed by quenching in air. After cooling, it was ground in a ceramic mill to pass a 75 μm (#200) mesh sieve. Its particle size distribution was measured in an isopropyl alcohol suspension using a Horiba Partica LA-950® Laser

Soft X-ray microscopy

Fig. 2 shows in-situ soft X-ray images of hydrating C4A3S¯ particles suspended in S1. Each of the three columns in each figure corresponds to a single position in the cell viewed for almost 3 h. The elapsed time after mixing is indicated below each image, so that the evolution of each sample position can be followed by scanning down the columns. The scale bar in each image corresponds to 1 μm.

Fig. 2 shows that the majority of ettringite formed in this system occurred before the first images were

Conclusions

This work shows that addition of solid CH to a very dilute (50:1) suspension of C4A3S¯ in water pre-saturated with respect to both CS¯H2 and CH tends to reduce the initial rate of formation of well-crystallized ettringite needles. This reduction in rate seems to be associated with the stabilization of a layer of hydrates on the surface of the C4A3S¯ particles. This hydrate layer has a rather gelatinous appearance but it is also covered with fine “hairy” outgrowths which are probably fine

Acknowledgments

C.W. Hargis was supported by the Berkeley Fellowship for Graduate Study and the Carlson–Polivka Fellowship. A.P. Kirchheim acknowledges the financial support of CAPES (Fundação Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Ministério da Educação – Brasil) and CNPq (National Counsel of Technological and Scientific Development). Research at XM-1 is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under contract no.

References (37)

  • T. Matschei et al.

    The AFm phase in portland cement

    Cem. Concr. Res.

    (2007)
  • M.D. Cohen

    Theories of expansion in sulfoaluminate-type expansive cements: schools of thought

    Cem. Concr. Res.

    (1983)
  • G.W. Scherer

    Stress from crystallization of salt

    Cem. Concr. Res.

    (2004)
  • G.C. Bye

    Portland Cement

    (1999)
  • Y. Wang et al.

    Sulphoaluminate Cement

    (1999)
  • J. Diao et al.

    Production and Application of Sulfo-aluminate Cement in China

    (2006)
  • I. Odler

    Special Inorganic Cements

  • P.K. Mehta

    Investigation on energy-saving cements

    World Cem. Technol.

    (1980)
  • Cited by (0)

    View full text