Autogenous deformations of cement pastes: Part II. W/C effects, micro–macro correlations, and threshold values

https://doi.org/10.1016/j.cemconres.2004.10.020Get rights and content

Abstract

A broad experimental study has been performed, from the end of mixing up to 2 years, on a set of plain cement pastes prepared with the same type I ordinary portland cement (OPC) and various water-to-cement ratios (W/C), and cured at various constant temperatures. Several parameters have been measured on the hydrating materials, such as chemical shrinkage, volumetric and one-dimensional autogenous deformations, degree of hydration of the cement, Ca(OH)2 content and Vicat setting times. Drying shrinkage has also been measured on the mature materials. In this part II of the paper, the effects of W/C within the range 0.25–0.60 have in particular been analysed in relation to the microstructural characteristics of the materials. This micro–macro analysis has highlighted a W/C threshold value (located around 0.40) both at the macro-level (on autogenous, but also on drying deformations and durability-related properties) and at the micro-level (characteristics of the hydration products, MIP porosity and pore size distribution, etc.).

In addition, volumetric and one-dimensional autogenous shrinkage deformations have been compared in the case of W/C=0.25 and T=20 °C. Finally, a critical twofold (chemical and structural) effect of calcium hydroxide has been found. When significant structural effects, generated by the formation and the growth of large-size Ca(OH)2 crystals, take place, swelling can become prominent, as observed for one-dimensional autogenous deformations in the case of medium and high W/C, and deviations are recorded on linear relationships.

Introduction

Nowadays, engineers have to take into account the required service lifetime, when selecting a concrete mixture and designing a structure. Hence, in recent years, the durability of reinforced or prestressed concrete structures has becoming a main concern. However, the most reliable, scientific-based, and accurate methodology used in the design stage to insure the required service lifetime for the structure will remain quite inefficient, if the early hours of the material life are not treated with the same accuracy in terms of execution quality and curing. The long-term properties of the material and even the whole life of the structure itself will indeed be determined by the early-age behaviour of concrete. For example, autogenous shrinkage as well as thermal dilation, which occur at early age, may yield, if restrained, the development of prejudicial cracking, particularly in high-performance concrete structures. In addition to mechanical problems, species aggressive to concrete or to reinforcement can more easily be carried from the surface through these (micro)cracks and then reduce the durability of the structure.

Various physical and chemical mechanisms are at the origin of autogenous volume changes and cracking of cementitious materials. The precise mechanisms are not very well understood, in particular because complex moisture and temperature combined effects take place. This makes a reliable predictive approach to the behaviour of concrete structures impossible.

As regards the test methods, autogenous deformations of cement pastes can be measured in two different ways: measurement of volumetric deformations (see for example Refs. [1], [2], [3], [4], [5]), and measurement of one-dimensional deformations (e.g. Refs. [6], [7], [8]). The first type of measurement is particularly suitable to obtain shrinkage data at (very) early age, whereas the other is more often performed for long-term investigations.

Within the framework of the study of cementitious materials shrinkage, in addition to the fact that a lot of research work has already been done regarding the long-term period, it seems particularly relevant from both the scientific and practical points of view, as previously mentioned, to focus on the early age, in particular the first 24 h. The question of the selection of a relevant initial (reference) time t0 (<24 h) for the deformation measurements has, for example, not been solved at the present time. Besides, along with the parameters related to the hydration process, it is well known that mix parameters, and in particular the water-to-cement ratio (W/C), influence autogenous deformations (see for example [6], [8], [9], [10], [11]). As far as chemical shrinkage is concerned, Justnes et al. [3] did not find any significant effect of W/C (ranging from 0.30 to 0.50) on the final chemical shrinkage value of cement pastes at 20 °C, from 0 to 48 h after mixing. Boivin et al. [12] confirmed these results and pointed out a purely kinetic influence of W/C on chemical shrinkage. Nevertheless, a broader range of investigations is still required, in order to accurately characterize the influence of W/C.

The purpose of this paper is to improve the knowledge of cementitious materials and to contribute to the understanding of the mechanisms related to autogenous deformations. This step is of primary importance in order to enable the prediction of both early-age (autogenous) and delayed (drying, carbonation, etc.) deformations. Within this framework, the monitoring of the hydration process (through degree of hydration of the cement, Ca(OH)2 content, Vicat setting times, etc.) and the characterisation of the cement paste microstructure have been carried out here as tools to help understand and interpret deformation results. The paper is separated in two parts. Part I was devoted to the analysis of curing temperature effects on early-age (≤24 h) volume changes and to the investigation of the relevance of the maturity concept. In addition, micro–macro relationships have been pointed out in part I between volume changes and chemical and physical parameters. In this part II, the effects of W/C within the range 0.25–0.60 will be investigated on various properties (chemical shrinkage, autogenous deformations, drying shrinkage, etc.), in relation to the microstructural characteristics of the materials. Moreover, a comparison between volumetric and one-dimensional autogenous shrinkage will be carried out in the case of W/C=0.25 and T=20 °C.

Section snippets

Experimental program and materials

A broad experimental study has been performed from the end of mixing up to 2 years on a set of plain cement pastes prepared with the same type I ordinary portland cement (OPC) (CEM I—52.5, according to the EN 197-1 European standard). W/C ranged from 0.25 up to 0.60 and the mixtures were submitted to an isothermal curing process at 10, 20, 30, 40, or 50 °C (±0.1 °C). The cement contained 70.2% C3S, 7.8% C2S, 3.8% C3A, and 6.0% C4AF (Bogue calculation). Its loss on ignition was 2.3% and its

Calcium hydroxide content

The calcium hydroxide content of the cement pastes with W/C ranging from 0.25 up to 0.60 and cured at T=20 °C is plotted in Fig. 2a vs. the degree of hydration of the cement (from 3 h up to 2 years). The degree of hydration of the cement vs. the age is displayed in Fig. 3 (see Ref. [13] for results obtained at other temperatures).

The results show that for 0.25≤W/C≤0.40, both at early age and in the long term, the amount of Ca(OH)2 produced at a given α is not influenced by W/C (note that a

Interpretation of the swelling observed on length changes with medium and high W/C

Autogenous swelling within the first days has also been reported in the literature by other authors, for cementitious materials with medium or high W/C (e.g. Refs. [6], [10]). Nevertheless, until now any deep insight into its associated mechanism has not really been performed.

The swelling observed on the experimental results presented here cannot be attributed to thermal effects due to the exothermic nature of chemical reactions. These effects disappeared indeed quickly as a result of the small

Micro–macro analysis: existence of a W/C threshold value

In part I of the paper, it has been pointed out that the early-age (≤24 h) volumetric autogenous shrinkage of cement pastes with W/C=0.25 was linearly related to the degree of hydration (α) of the cement or to the Ca(OH)2 content, from α=7% and whatever the temperature. In the present part II, it is possible to analyse further the role of these two parameters, in particular in the long-term range, and as a function of W/C. Fig. 14a and b show that one-dimensional autogenous deformations at T=20

Comparison between volumetric and one-dimensional autogenous shrinkage for W/C=0.25

A comparison between volumetric and one-dimensional autogenous shrinkage measurements has been performed at T=20 °C for the cement paste with W/C=0.25, after converting the volumetric deformations into corresponding 1-D deformations (assumed as 1/3 of the volumetric ones) in μm/m. The results are plotted in Fig. 15a when the reference is taken at about 2.4 h, which approximately corresponds to the initial Vicat setting time. A factor of about 4 is found between the two types of measurement at

Summary and concluding remarks

Experimental tools for assessing key-parameters have been presented in this paper. These tools can be used in view of designing and optimising mixtures, within the framework of the prediction of early-age deformations and of the prevention of early-age cracking. Nevertheless, as illustrated for example by the comparison between volumetric and one-dimensional autogenous shrinkage measurements, further investigations are still required with respect to both the techniques and the test procedures,

Acknowledgments

The authors are grateful to Roger Coué, from IUT de Saint-Nazaire, for his assistance in setting up the chemical shrinkage device and to Ali Kheirbek, from LCPC, for his contribution as regards the length-change measurements.

References (40)

  • H. Justnes et al.

    Total and external chemical shrinkage of low W/C ratio cement pastes

    Adv. Cem. Res.

    (1996)
  • S. Garcia-Boivin, Retrait au jeune âge du béton—Développement d'une méthode expérimentale et contribution à l'analyse...
  • H. Mitani et al.

    Autogenous shrinkage of different types of cement and different types of admixtures at early ages

  • E. Tazawa et al.

    Autogenous shrinkage caused by self-desiccation in cementitious material

  • O.M. Jensen et al.

    A dilatometer for measuring autogenous deformation in hardening Portland cement paste

    Mat. Struc.

    (1995)
  • V. Baroghel-Bouny et al.

    Effect of mix-parameters on autogenous deformations of cement pastes—microstructural interpretations

    Concr. Sci. Eng.

    (2001)
  • O.M. Jensen et al.

    Autogenous deformation and change of the relative humidity in silica-fume-modified cement paste

    ACI Mater. J.

    (1996)
  • P. Schiessl et al.

    Influence of binders and admixtures on autogenous shrinkage of high performance concrete

  • H. Mitani, Variations volumiques des matrices cimentaires aux très jeunes âges: approche expérimentale des aspects...
  • S. Boivin et al.

    Experimental assessment of chemical shrinkage of hydrating cement pastes

  • Cited by (174)

    View all citing articles on Scopus
    View full text