Advertisement

Journal of Thermal Analysis and Calorimetry

, Volume 133, Issue 1, pp 399–404 | Cite as

Application of dilatometric analysis to the study of autoclaved calcium silicate materials

  • Dana Kubátová
  • Anežka Zezulová
  • Alexandra Rybová
  • Radovan Nečas
Article

Abstract

The process of shrinkage of calcium silicate hydrate was investigated by dilatometry up to 350 °C. The properties of this material are based on the formation of C–S–H phases during the reaction at temperatures between 180 and 205 °C and water vapor pressure lower than 16 bars. The main C–S–H phases are 11.3 Å tobermorite and xonotlite. 11.3 Å tobermorite converts to 9.3 Å tobermorite on air at temperatures around 300 °C. The hydrosilicate materials were prepared from quicklime and finely ground sand with different CaO/SiO2 ratios under different hydrothermal conditions. The reaction time was 24 h. Materials based on xonotlite and tobermorite were produced, and the calcium silicate phases were characterized by XRD and TG/DTA methods. Dilatometry measurements were used to study the effect of heating conditions on sample shrinkage. Dehydration of hydrated calcium silicate minerals occurred during heating. The results show that sample shrinkage is dependent on the type and amount of C–S–H phases, the amount of bound water and formation of 9.3 Å tobermorite. All samples showed shrinkage after heating up to 350 °C, but this change was not irreversible for all samples after cooling to room temperature.

Keywords

Calcium silicate hydrates Dilatometry Shrinkage Tobermorite Xonotlite 

Introduction

Autoclaved calcium silicate hydrate materials are widely used in the construction industry, nowadays. For autoclaved sand and lime materials, many properties are important, mainly mechanical properties, material compactness, high porosity, open pore size and sorption capacity. Mineralogical composition has a considerable influence on the properties of the prepared materials. Tobermorite Ca5Si6O16(OH)2·4(H2O) and xonotlite Ca6Si6O17(OH)2 are major constituents in autoclaved calcium silicate materials.

Tobermorite is stable in the range of composition form Ca/Si ratio between 0.8 and 1 and in a very narrow temperature range. Below 100 °C, it forms an amorphous gel, C–S–H (I). Well crystallized tobermorite is formed at 110 °C. Three polytypes of tobermorite exist, namely 14 Å, 11.3 Å and 9.3 Å tobermorite. The most common form is 11.3 Å tobermorite. At temperatures above 300 °C, it decomposes to 9 Å tobermorite Ca5Si6O16(OH)2·2(H2O). During heating up to 300 °C, the basal spacing of normal tobermorite decreased from 11.3 to 9.3 Å. Molecules of water are lost from the inter-layers, and the riversideite is formed. The 11.3 Å tobermorite also exists in its anomalous form, which does not convert to 9.3 Å tobermorite at test conditions [1, 2, 3]. Xonotlite structure is formed at higher temperatures than that of tobermorite, with the equilibrium phase boundary between the two minerals being at approximately 140 °C [4].

Crystallization of C–S–H phases depends on the C/S ratio of the components, the reaction temperature and the reaction time. These parameters affect the structure type, crystal morphology and crystallinity of the resulting product [5].

The reaction of the lime with water leads to the formation of portlandite, which reacts with quartz to form the C–S–H phase. Solubility of quartz and lime plays an important role in C–S–H formation and crystallization. The solubility of quartz increases, while portlandite solubility in water decreases with rising temperature [5]. The formation of 11.3 Å tobermorite is a two-step non-isokinetic process. The first step is controlled by partial dissolution of quartz and its reaction with portlandite. Than a layer of C–S–H phases is formed around quartz grains. The second step of the reaction is controlled by SiO2 diffusion through C–S–H phase layer. 11.3 Å tobermorite is formed by the reaction of quartz and C–S–H phases, and after all portlandite is depleted [6, 7].

Various studies on synthesis and properties of tobermorite and xonotlite have been carried out in previous years [5, 8, 9, 10]. The effects of various dopants, quartz particle size and water-to-solid ratio on hydrothermal synthesis of C–S–H phases were investigated by different methods [5, 8, 9]. The shrinkage of materials based on tobermorite was studied. Feldman has developed a model of the drying shrinkage of Portland cement paste. Chatterji [10] used aerated concrete to investigate the role of the interlayer spacing and water in the drying shrinkage of tobermorite-based materials. Autoclaved blocks were dried at 130 °C.

The objective of this work is to study the shrinkage of calcium silicate hydrate during drying, taking into account the different preparation conditions by means of dilatometry. The aim of the study is to find the optimal C/S ratio and temperature during autoclaving at which maximum mass shrinkage occurs. These parameters are important for optimization of hydrosilicate mass of production. Non-shrinking materials can cause technological problems.

Experimental

Materials and methods

Quicklime putty and sand were used for calcium silicate hydrate preparation. The quicklime was produced in a rotary kiln with purity higher than 97%. Slaked lime putties were prepared in the laboratory by mixing lime and deionized water under vigorous stirring at speed 300 RPM; the CaO/H2O ratio was 1:8. The lime putty was mixed with finely ground sand at CaO/SiO2 (C/S) ratios 0.82, 0.86, 0.9 and 0.95 under vigorous stirring. The sand contained 99% of SiO2, and a mean diameter of particles is 8 µm. This slurry was hydrothermally treated. Three sets of samples were prepared in autoclave at temperature (1) 180 °C, (2) 195 °C, (3) 205 °C in saturated steam for 24 h.

The final product was cut into 1 × 1 × 5 cm prisms for dilatometric analysis. The drying process was carried out in a dilatometer Netzsch 402 E/3 at 350 °C, with constant heating rate 6 °C min−1.

The TGA measurements were taken with a Mettler-Toledo TGA/DSC STRA system (Mettler-Toledo, Switzerland), where 10.00 (± 0.03) mg of powdered sample was placed in an open platinum crucibles under air atmosphere, and heated up to 1000 °C at a heating rate of 10 °C min−1.

The composition of hydrosilicate materials before and after drying was determined by X-ray powder diffraction. The X-ray diffraction analysis was conducted using Bruker D8 Advance apparatus with Cu anode (λKα = 1.54184 Å) and variable divergence slits at ΘΘ Bragg–Brentano reflective geometry. Quantitative phase analysis was carried out by Rietveld method. The amorphous phase content was determined using internal standard addition (20 wt% of fluorite—CaF2).

Results and discussion

Samples of the prepared materials were dried after autoclaving at 60 °C for 12 h and analyzed by X-ray powder diffraction; the xonotlite (PDF 29-0379) and orthorhombic 11.3 Å tobermorite (PDF 45-1480) were identified in all samples. Small amounts of quartz (PDF 46-1045) and amorphous phase were detected. It was found that the amount of xonotlite increases with increasing C/S ratio. This is related to the fact that the ratio C/S of the ideal xonotlite is equal to 1.0 [11]. The 2.98 Å (30.0° 2Θ) peak, which is unique to tobermorite, decreases and a 4.21 Å (20.8° 2Θ) peak, which is unique to xonotlite, increases. 1.8 Å (50.2° 2Θ) peak indicates the presence of Dreiketten chains [6]. It was confirmed that higher temperature during hydrothermal process promotes the formation of xonotlite. These results are in agreement with previous works [1, 7, 10, 11, 12, 13]. XRD pattern of sample with C/S 0.82 hydrothermally treated at 180 °C is shown in Fig. 5.

Dilatometric method is useful to characterize the behavior of hydrated calcium silicates during thermal processing. The course of thermal processing included heating to 350 °C followed by gradual cooling to room temperature. Total volume shrinkage after heating and cooling is the key parameter to characterize the stability of autoclaved samples. The change in length and rate of shrinkage are shown in Figs. 13, respectively. All samples were shrinking up to 350 °C, which could be associated with water loss and dehydroxylation of xonotlite and tobermorite. This final expansion after cooling to room temperature does not entirely compensate the total shrinkage of samples at 350 °C. The expansion of sample occurs during cooling and can be explained by relaxation of structure due to the evolvement of heat. Subsequent expansion after cooling to laboratory temperature could be attributed to polymorphic transformation of high-temperature calcium silicate hydrates to low-temperatures ones.
Fig. 1

Shrinkage of hydrosilicate samples with different C/S ratios (Set No. 1. 180 °C)

Fig. 2

Shrinkage of hydrosilicate samples with different C/S ratios (Set No. 2. 195 °C)

Fig. 3

Shrinkage of hydrosilicate samples with different C/S ratios (Set No. 3. 205 °C)

For samples with lower proportion of lime, the final shrinkage value reached higher values. This was confirmed in all sets. The relationship between the shrinkage and the C/S ratio was found.

All prepared samples, with different tobermorite and xonotlite content, were tested by TG-DSC analyses. DTG/TG/DSC characteristics of sample with C/S 0.82 hydrothermally treated at 180 °C are shown in Fig. 4, where three strong peaks at 130, 200 and 780 °C are observed at DTG. Three weak peaks can be also considered: below 100 °C characterizing loss of moisture, in temperature interval 260–360 °C and 450–600 °C. TG curve with total mass 12.06% presents multiple detectable dehydration steps corresponding to the number of peaks detected at DTG curve. The small peak below 100 °C corresponds to the loss of water physically bond or moisture. The loss of moisture corresponds to 1.5% mass loss at TG curve, but if there is a lot of moisture in the sample, the first peak can overlap with that characterizing the loss of molecular water from C–S–H dehydroxylation. Peak at 130 °C characterizes the thermal decomposition of different forms of calcium silicate hydrates including C–S–H (1) as precursor of tobermorite. Although the exact relationship between C–S–H (1) and tobermorite is not well established, they may differ only by degree of crystallinity. The mass loss corresponds to 3.04%.
Fig. 4

DTG/TG/DSC characteristics of sample with C/S 0.82 hydrothermally treated at 180 °C

In samples with higher tobermorite content, a strong endotherm occurs between 160 and 260 °C with endothermic peak at 190 °C; this peak is attributable to loss of molecular water from anomalous tobermorite [14] since normal tobermorite 11 Å dehydrates to the 9 Å form at approximately 300 °C [6]. The mass loss represents 3.40%.

Some of the weak peaks observed at 260–360 and 450–600 °C could result from decomposition of different forms of tobermorites or xonotlite with mass approx. 3.42%.

It was reported [6, 15] that dehydroxylation of xonotlite occurs at around 780 °C, but scawtite as a product of the carbonization that occurs frequently in hydrothermal curing conditions decomposes at the similar temperature range [16]. The processes were visible only on DTG curve of samples with higher tobermorite content. The mass loss at TG curve corresponds to 0.7% [17].

If the C/S ratio is lower than 0.95 (Set No. 3) or higher than 0.86 (Set No. 1), then there is no TG shift around 670 °C that would indicate tobermorite presence. All samples from Set No. 2 have recognizable tobermorite dehydroxylation, even C/S 0.95 has a small peak of this reaction. At around 800 °C, both tobermorite and xonotlite recrystallize to wollastonite. Dehydroxylation and recrystallization of xonotlite usually occur simultaneously [6]. However, the exothermic peaks of these reactions are not clearly noticeable in some samples. The exotherm at 800–850 °C is associated with crystallization of wollastonite [18]. Very weak, diffuse exotherm may be due to the low content of tobermorite with a low content of aluminum in its structure as confirmed by Mitchuda’s and Taylor’s [14] and Guo et al. [15] works. The aluminum-free tobermorite structure absorbed less energy in the process of transformation to wollastonite. The character of high-temperature exothermic peak appears to depend on the crystallinity of sample; thus, it may be weak or absent for highly crystalline starting materials [19].

Quantitative phase analysis of samples was carried out by Rietveld method after dilatometric analysis. 9.8 Å peak was identified in the sample set No. 1 with C/S ratio 0.82 only. After drying at 350 °C, the amount of amorphous phase is reduced and peak 1.8 Å is lost, in all sets. Peak 11.3 Å shifts to lower values for some samples. Effect of different hydrothermal treatments on basal spacing of 11 Å tobermorite is shown in Table 1. The fact that the basal spacings did not decreased from 11 to 9.3 Å indicates that samples of set No. 2 and No. 3 were of the anomalous variety. El-Hemaly et al. [20] introduced the term “normal” for tobermorites whose basal spacing decreased to 10.0 Å or less after heating to 300 °C and “anomalous” for those that did not shrink bellow 11.0 Å. The term “mixed tobermorite” is used for products with basal reflection around 10.5 Å at 300 °C. The significant changes during the drying process occur in samples where higher content of amorphous phase was identified. XRD pattern of sample after dilatometric analysis is shown in Fig. 5.
Table 1

Basal spacing of 11 Å tobermorite after heat treatment at 350 °C for 1 min

 

Initial C/S

0.82

0.86

0.90

0.95

Set No. 1, 180 °C

9.8

10.9

10.9

10.9

Set No. 2, 195 °C

10.9

11.0

10.9

11.0

Set No. 3, 205 °C

11.0

11.1

11.1

11.1

Fig. 5

XRD pattern of sample set No.1 with C/S ratio 0.82 before and after dilatometric analysis

The DTG/TG/DSC results of all samples indicate that heat treatment of anomalous tobermorite up to 350 °C leads to loss of molecular water from the samples. Although X-ray diffraction measurements of most samples indicate that the 9 Å tobermorite did not form, the dilatometry measurement reported above shows that shrinkage occurs after heating to 350 °C.

The next step was to heat the samples at 350 °C for 15 h. Basal spacing of 11 Å tobermorite is in the following Table 2. The influence of longer drying was demonstrated only in set No. 1 samples prepared at 180 °C. At 180 °C, xonotlite and 9.8 Å tobermorite were formed, while at 195 and 205 °C xonotlite and anomalous tobermorite were formed.
Table 2

Basal spacing of 11 Å tobermorite after heat treatment at 350 °C for 15 h

 

Initial C/S

0.82

0.86

0.90

0.95

Set No. 1, 180 °C

9.8

9.8

9.8

9.8

Set No. 2, 195 °C

10.9

10.9

10.9

10.9

Set No. 3, 205 °C

11.1

11.1

11.1

11.1

Comparison of the final shrinkage values after heating to 350 °C and cooling to room temperature is shown in Table 3. Longer drying time has mostly insignificant effect on the final shrinkage. Maximum shrinkage was found in the sample prepared at 180 °C with C/S ratio 0.82.
Table 3

Comparison of final shrinkage after heating to 350 °C with different temperature holds

Initial C/S

Shrinkage/%

0.82

0.86

0.9

0.95

Temperature hold

1 min

15 h

1 min

15 h

1 min

15 h

1 min

15 h

Set no. 1, 180 °C

− 0.6

− 0.7

− 0.2

− 0.2

− 0.2

− 0.2

0

0

Set no. 2, 195 °C

− 0.3

− 0.2

0

0

0

0

0

0

Set no. 3, 205 °C

− 0.2

− 0.1

0

0

0

0

0

0

It was found that mass shrinkage occurred during drying at 350 °C. This shrinkage is dependent on the input raw material ratio and the temperature of the hydrothermal process. A mixture of 9.8 Å tobermorite and xonotlite is formed in samples at 180 °C, while at higher temperatures of hydrothermal treatment, anomalous tobermorite is formed. Shrinkage of the samples can be attributed to the loss of water, dehydroxylation, and conversion of 11.3–9 Å tobermorite. Although the 9 Å tobermorite was not identified in the sets No. 2 and 3, a small shrinkage of − 0.2% occurred. This shrinkage could be attributed to the amount of amorphous phase that was identified by Rietveld analysis. The shrinkage of − 0.2% was recorded in samples where the amount of amorphous phase was calculated to be over 15%, while in cases where also 9 Å tobermorite was identified, shrinkage increased to − 0.6%. The average values of amorphous phase and corresponding shrinkage values at different temperatures are shown in Fig. 6.
Fig. 6

Relationship between shrinkage and amorphous content and C/S ratio in temperature range 180–205 °C

The presented results can be compared with El-Hemaly et al. [20]. The reaction of lime, water, and colloidal silica in unstirred suspension with 1.0 C/S ratio after 24 h of hydrothermal process resulted in mixed tobermorite, while suspension with 0.8 C/S ratio under the same conditions resulted in anomalous tobermorite.

The resulting product is influenced by a number of factors such C/S ratio, temperature, autoclaving time. The important fact is that the mixture was prepared from quicklime and crystalline quartz and was not stirred during autoclaving. In previous works [7, 20], the influence of mixing on formation of anomalous tobermorites was confirmed, as well as the effect of particle size on the reaction between SiO2 and CaO. Crystal growth is slower in the unstirred suspension, and thicker quartz particles can retard the reaction and so that the 11.3 Å tobermorite crystallizes more slowly.

Conclusions

The dilatometric measurement of autoclaved hydrosilicate matter with the C/S ratio from 0.82 to 0.95 was taken during heating up to 350 °C. It was proved that for all the sets of samples, shrinkage was obtained and this shrinkage should be associated with water loss and dehydroxylation of xonotlite and tobermorite. The expansion of sample occurs during cooling and could be associated with relaxation of structure due to the evolvement of heat or due to the polymorphic transformation of high-temperature calcium silicate hydrates to low-temperatures ones. This final expansion after cooling to room temperature does not entirely compensate the total shrinkage of samples at 350 °C. For samples prepared at 180 °C, greater final shrinkage was registered, which may be caused by dehydroxylation and 9 Å tobermorite content, which rises around 300 °C. Lower total shrinkage was registered in the C/S 0.82 samples in the set No. 2 and No. 3, which can be related to the amount of amorphous phase identified in the original sample by Rietveld method. Total shrinkage is influenced by the ratio between tobermorite and xonotlite, type of tobermorite and the amount of amorphous phase. To achieve maximum final shrinkage of the mass under current conditions, it is appropriate to set the C/S ratio around 0.82 and autoclave temperature around 180 °C.

Notes

Acknowledgements

This paper was elaborated with the institutional support for long-term development of research organizations by the Ministry of Industry and Trade of the Czech Republic.

References

  1. 1.
    Shaw S, Henderson CMB, Komanschek BU. Dehydration/recrystallization mechanism, energetics, and kinetics of hydrated calcium silicate minerals: an in situ TGA/DSC and synchrotron radiation SAXS/WAXS study. Chem Geol. 2000;167:141–59.CrossRefGoogle Scholar
  2. 2.
    Hara N, Inoue N. Thermal behaviour of 11 Å tobermorite and its lattice parameters. Cem Concr Res. 1980;10:53–60.CrossRefGoogle Scholar
  3. 3.
    Merlino S, Bonaccorsi E, Armbruster T. Tobermorites: their real structure and order-disorder (OD) character. Am Mineral. 1999;84:1613–21.CrossRefGoogle Scholar
  4. 4.
    Speakman K. The stability of tobermorite in the system CaO–SiO2–H2O at elevated temperatures and pressures. Mineral Mag. 1968;36:1090–103.Google Scholar
  5. 5.
    Hartmann A, Schulenberg D, Buhl JC. Synthesis and structural characterization of CSH-phases in the range of/S = 0.41–1.66 at temperatures of the tobermorite xonotlite crossover. J Mater Sci Chem Eng. 2015;3:39–55.Google Scholar
  6. 6.
    Shaw S, Clark SM, Henderson CMB. Hydrothermal formation of the calcium silicate hydrates tobermorite (Ca5Si6O16(OH)2·4H2O) and xonotlite (Ca6Si6O17(OH)2): an in situ synchrotron study. Chem Geol. 2000;167:129–40.CrossRefGoogle Scholar
  7. 7.
    Bernstein S. Determination of reaction kinetics and mechanism of 1.13 tobermorite in situ neutron diffraction. Prog Cryst Growth Charact Mater. 2012;58:84–9.CrossRefGoogle Scholar
  8. 8.
    Kikuma J, et al. Effect of quartz particle size and water-to-solid ratio on hydrothermal synthesis of tobermorite studied by in situ time-resolved X-ray diffraction. J Solid State Chem. 2011;184:2066–74.CrossRefGoogle Scholar
  9. 9.
    Nocuń-Wczelik W. Effect of Na and Al on the phase composition and morphology of autoclaved calcium silicate hydrates. Cement Concrete Res. 1999;29:1759–67.CrossRefGoogle Scholar
  10. 10.
    Chatterji S. The role of interlayer spacing and water in the drying shrinkage of tobermorite based materials. Cem Conc Res. 1975;5:83–7.CrossRefGoogle Scholar
  11. 11.
    Kalousek GL, Mitsuda T, Taylor HFW. Xonotlite: cell parameters, thermogravimetry and analytical electron microscopy. Cem Concr Res. 1977;7:305–12.CrossRefGoogle Scholar
  12. 12.
    Spudulis E, Šavareika V, Špokauskas A. Influence of hydrothermal synthesis condition on xonotlite crystal morphology. Mater Sci. 2013;19:190–6.Google Scholar
  13. 13.
    Hong SY, Glasser EP. Phase relations in the CaO–SiO2–H2O system to 200 °C at saturated steam pressure. Cem Concr Res. 2004;34:1529–34.CrossRefGoogle Scholar
  14. 14.
    Mitsuda T, Taylor HFW. Normal and anomalous tobermorite. Mineral Mag. 1978;42:229–35.CrossRefGoogle Scholar
  15. 15.
    Guo X, Meng F, Shi H. Microstructure and characterization of hydrothermal synthesis of Al-substituted tobermorite. Constr Build Mater. 2017;133:253–60.CrossRefGoogle Scholar
  16. 16.
    Kuzielova E, Zemlicka M, Masilko J, Palou MT. Pore structure development of blended G-oil well cement submitted to hydrothermal curing conditions. Geothermics. 2017;68:86–93.CrossRefGoogle Scholar
  17. 17.
    Yu P, Kirkpatrick RJ. Thermal dehydration of tobermorite and jennite. Concr Sci Eng. 1999;1:185–91.Google Scholar
  18. 18.
    Palou MT, Kuzielova E, Novotny R, Soukal F, Zemlicka M. Blended cements consisting of Portland cement–slag–silica fume–metakaolin system. J Therm Anal Calorim. 2016;125:1025–34.CrossRefGoogle Scholar
  19. 19.
    MacKenzie RC. Differential thermal analysis: fundamental aspects. 1st ed. London: Academic Press; 1970.Google Scholar
  20. 20.
    El-Hemaly SAS, Mitsuda T, Taylor HFW. Synthesis of normal and anomalous tobermorites. Cem Concr Res. 1977;7:429–38.CrossRefGoogle Scholar

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2017

Authors and Affiliations

  1. 1.Research Institute for Building MaterialsBrnoCzech Republic

Personalised recommendations