Acta Polytechnica CTU Proceedings https://doi.org/10.14311/APP.2024.47.0113 Acta Polytechnica CTU Proceedings 47:113–116, 2024 © 2024 The Author(s). Licensed under a CC-BY 4.0 licence Published by the Czech Technical University in Prague EFFECT OF PREPARATION PROCESS ON PURITY OF TRICALCIUM ALUMINATE Simona Ravaszová∗, Karel Dvořák, Andrea Jančíková Brno University of Technology, Faculty of Civil Engineering, Veveří 331/95, 602 00 Brno, Czech Republic ∗ corresponding author: ravaszova.s@fce.vutbr.cz Abstract. In the realm of cement production, traditional fuels are steadily giving way to alternative sources. While this shift yields economic and ecological advantages, it introduces notable technological challenges. Specifically, there is an escalation in certain chemical elements, leading to variability in the mineralogical composition. This variability is subsequently manifested in the characteristics of Portland clinker. It becomes imperative, therefore, to synthesize pure clinker minerals. This synthesis is essential for both understanding their behavior in industrially manufactured cement and creating analytical standards for mineral identification through XRD. The primary focus of the research is on the production of pure tricalcium aluminate, a key clinker mineral, within laboratory conditions. A direct laboratory methodology involving solid-phase synthesis is used. The investigation meticulously tracks the impact of homogenizing the raw material mixture on the resultant mineral’s purity. To achieve this, a high-speed Pulverisette 6 planetary mill is utilized along with two types of grinding bodies. A corundum grinding capsule with corundum grinding bodies is employed to prevent contamination by foreign ions. Simultaneously, a highly efficient steel grinding capsule with steel grinding bodies is utilized. The raw material mixture is then subjected to sintering at six different temperatures ranging from 1 200 to 1 450 °C, and the purity of the produced mineral is quantified using Rietveld analysis. Laser granulometry results indicate a strikingly similar comminution of the raw material in both grinding cases. Iron contamination, as revealed by XRF analysis, is minimal, accounting for only 0.21 %. The purity of 97.6 % C3A, is achieved at 1 450 °C using a steel grinding capsule, with 2.4 % unreacted lime Conversely, when using a corundum capsule, a purity of 93.6 % is achieved, accompanied by residues of 2.2 % mayenite and 3.8 % lime. This discrepancy may be attributed to an insufficient degree of raw material comminution in the corundum capsule. Intriguingly, a slight iron contamination during the grinding of the raw material mixture exerts a positive influence on purity. In this case, iron acts as a flux, fostering a more favorable reaction of the mineral mayenite. This multifaceted exploration enhances our understanding of clinker mineral synthesis, offering insights into optimizing purity based on grinding methods, sintering temperatures, and the impact of minor contaminants. Keywords: Tricalcium aluminate, synthesis, contamination, Rietveld analysis. 1. Introduction Tricalcium aluminate C3A is one of the main compo- nents of Portland clinker. C3A acts as a melt during clinker production and is part of the liquid phase dur- ing firing. In the prepared clinker, it becomes part of the interfacial substance. Together with C4AF, it forms most of the 22 to 28 % of the clinker melt that is essential for the formation of other clinker minerals such as alite C3S and belite C2S, which are the main minerals that bring the cement parameters. Ordi- nary Portland clinker contains 8 to 11 % of C3A. In cement, the role of C3A is related to the kinetics of cement hydration. These minerals determine a very high proportion of the final physical properties of the produced clinker. In practice, they are never found in pure form but are contaminated with other chemicals. A better understanding of all the mechanisms involved in C3A formation and hydration requires the ability to prepare pure C3A. [1] There are several different procedures for the prepa- ration of tricalcium aluminate by the solid-state syn- thesis, but they all involve the following steps: • Preparation of raw materials: raw materials such as limestone and corundum are usually first ground into small particles. • Mixing the raw materials: the ground materials are then mixed in the correct proportions to ensure that the final product contains the correct amount of calcium and aluminium. • Processing of the mixture. • Drying and firing: the resulting mass is then dried and fired at high temperature to become a solid tricalcium aluminate [2]. The synthesis of tricalcium aluminate by the solid- state synthesis means that tricalcium aluminate is produced directly from solid materials without the use of liquid solvents. This method is often used to produce large quantities of tricalcium aluminate and to achieve high purity and quality of the final prod- uct. Solid synthesis is carried out using a mixture of 113 https://doi.org/10.14311/APP.2024.47.0113 https://creativecommons.org/licenses/by/4.0/ https://www.cvut.cz/en S. Ravaszová, K. Dvořák, A. Jančíková Acta Polytechnica CTU Proceedings calcium oxide (CaO) and aluminium oxide (Al2O3) in a molar ratio of 3:1, which is then calcined at a tem- perature of around 1 100–1 200 °C for 2–3 hours. This process leads to the formation of tricalcium aluminate in crystalline form [3, 4]. Coprecipitation is a process in which two or more elements or compounds are precipitated as a solid crys- tal under specific conditions such as temperature and pH. In addition, C3A and cement mixtures contain- ing C3A can be modified by the addition of various additives such as calcium oxide (CaO), silicates or carbonates. These additives can alter the properties of the cement, such as hydration (i.e. the reaction of the cement with water), setting, strength and resis- tance to chemicals. Modifications to C3A and cement mixtures can also be made by various processes, such as changing the temperature or pressure during man- ufacture [5]. This topic is very relevant lately because developments in cement production are increasingly influenced by environmental and economic factors. For this reason, it is very important to know in detail all the mineralogical processes of transformation of each polymorphic modification of each clinker mineral. 2. Material and methods To produce cubic tricalcium aluminate, calcium car- bonate (p.a. purity, 99.7 %) and aluminium oxide (p.a. purity 99.5 %) were used in an amount correspond- ing to the molar ratio of C3A or 3 moles of CaO and 1 mole of Al2O3, see Table 1 (Chemical composition of the raw material mixture). The homogenisation of the raw materials was car- ried out by high-speed milling in a laboratory plan- etary mill Frisch Pulverisette 6. Distilled water was added to the raw material mixture at a powder/water ratio of 0.8. Two distinct grinding capsules made of different materials were chosen. First half of the raw material powder was homogenized in a steel-bodied capsule with 25 steel milling bodies for 15 minutes at 500 rpm. The second half of the raw material powder was ho- mogenized in a corundum capsule with 25 corundum milling bodies for 30 minutes at 350 rpm. The reason for the choice of two different materials is possible iron contamination during abrasion of the grinding capsule and the ground material. However, it was also checked if grinding in a corundum capsule achieve the desired degree of comminution as a steel capsule. Possible iron contamination will be verified by XRF analysis. The degree of comminution of the raw meal will be verified by laser granulometry by Malvern Paralyti- cal Mastersizer 2000. After milling, the raw material powder was left to dry into nodules 10–15 mm in di- ameter were formed spontaneously in a laboratory dryer at 105 °C for 24 hours. In the next step, a series of two-phase sintering’s were carried out from both prepared raw material flours. In the first stage of sintering, calcination occurs at 900 °C with soaking time of 60 minutes. In the second sintering stage, the actual solid phase synthesis takes place. The sinter- ing temperatures were chosen in the range in range between 1 200–1 450 °C. The purity of the prepared products was verified by X-ray analysis. CaCO3 A2O3 H2O C3A cubic 111.124 37.736 120.000 Ca/Al=1.500 H2O/C3A=0.8 Table 1. Chemical composition of the raw material mixture in grams and relevant ratios. 3. Results and discussion XRF analysis of the raw material powders was car- ried out due to possible iron contamination during milling in the steel capsule. According to the results, it can be seen that the iron contamination is minimal, specifically 0.21 %. The complete results are shown in Table 2. The results of the laser granulometry, which was performed after the XRF evaluation see Figure 1, show that the particle size distribution of the powder milled in the steel capsule for 15 minutes and in the corundum capsule for 30 minutes is almost identical. However, the steel capsule ground meal has a higher particle size distribution in the range of 10 to 100 µm. Figure 1. Particle size distribution (PSD) of raw material mixture after milling in Pulverisette 6. The evaluation of the samples by XRD analysis, which were milling in a steel and corundum capsule and underwent a two-phase sintering first at 900 °C with a soaking time of 60 min. The samples were then subjected to a second sintering at 1 200, 1 250, 1 300, 1 350, 1 400 and 1 450 °C. From the diffractograms, increasing the firing temperature influences the for- mation of C3A. As shown in the detail images, with increasing temperature the intensity at 18.1° 2θ an- gle decreases indicating the decrease of the mineral mayenite Ca12Al14O33 (M) and the intensity increase of the peak at 17.4° 2θ indicating an increase in the amount of C3A mineral. 114 vol. 47/2024 Effect of preparation process on purity of tricalcium aluminate SiO2 Al2O3 Fe2O3 CaO MgO Na2O3 K2O SO3 TiO2 MnO Sum. steel 0.0 36.522 0.21 63.268 0.0 0.0 0.0 0.0 0.0 0.0 100.0 corundum 0.0 37.109 0.0 62.891 0.0 0.0 0.0 0.0 0.0 0.0 100.0 Table 2. XRF analysis: chemical composition of raw material mixture after milling. Figure 2. XRD analysis of C3A (steel capsule). The X-ray diffraction patterns show that the purest C3A was formed at 1 450 °C, however, the lime CaO and the mineral mayenite are still present in the sam- ple; the inability of these components to react is prob- ably due to insufficient comminution of the corundum component of the raw material. The mineral mayen- ite is more evident on samples milled in a corundum grinding capsule. Quantification by the Rietveld method was per- formed in HighScore Plus version 3.0.5. Table 3 and Table 4 show the results of the first batch of sintered samples. The Rietveld method confirmed that C3A reaches the highest purity at 1 450 °C, however, as already known from Figure 2 and Figure 3 of the XRD anal- ysis, mayenite and free lime are also present in the samples. These partially reacted minerals are found in greater quantities in the samples that were milled in the corundum capsule. This is likely due to the slight iron contamination proven above by XRF anal- ysis, this iron contamination in this case acting as a melting medium that helped to react the remaining mayenite. Mmanaged to demonstrate that the difference in results when grinding in a corundum and steel capsule is minimal, as iron contamination is only 0.21 %. Furthermore, it was found that during the prepara- tion of cubic C3A, the purity of this mineral increases with the firing temperature, the highest purity was achieved by the mineral C3A at 1 450 °C in the sample C3A Ca12Al14O33 CaO Al2O3 1 200 °C 55.4 24.2 19.3 1.1 1 250 °C 66.5 19.3 13.2 1.0 1 300 °C 78.3 13.9 7.3 0.5 1 350 °C 95.5 1.5 2.6 0.4 1 400 °C 94.4 2.4 3.1 0.0 1 450 °C 97.6 0.0 2.4 0.0 Table 3. Rietveld analysis: mineral composition (steel capsule). C3A Ca12Al14O33 CaO Al2O3 1 200 °C 49.2 31.4 18 1.4 1 250 °C 59.3 26.6 12.6 1.5 1 300 °C 71.3 20.4 7.9 0.4 1 350 °C 94.8 3.6 1.2 0.4 1 400 °C 94.5 3.4 1.6 0.5 1 450 °C 93.6 2.2 3.8 0.4 Table 4. Rietveld analysis: mineral composition (corundum capsule). that was ground in a steel capsule. Here, iron as a flux most likely helped to achieve a C3A purity of 97.6 % according to the Rietveld method. 115 S. Ravaszová, K. Dvořák, A. Jančíková Acta Polytechnica CTU Proceedings Figure 3. XRD analysis of C3A (corundum capsule). 4. Conclusion • Laser granulometry results showed similar comminu- tion of the raw material in both cases of grinding. • Iron contamination based on XRF analysis was minimal at only 0.21 %. • The highest purity of 97.6 % C3A was achieved at a temperature of 1 450 °C when grinding in a steel capsule, with 2.4 % representing unreacted lime. • Grinding in a corundum capsule resulted in a purity of 93.6 %, with a residue of the mineral mayenite of 2.2 % and lime of 3.8 %. This may be attributed to insufficient comminution in the corundum capsule. • Slight iron contamination during grinding had a positive effect on purity, as iron acted as a flux, improving the reaction of the mineral mayenite. Acknowledgements We acknowledge the financial support to project (FAST-S- 23-8246) Study of the formation and stability of various tricalcium aluminate polymorphs. References [1] J. Moon, S. Yoon, R. M. Wentzcovitch, et al. Elastic properties of tricalcium aluminate from high-pressure experiments and first-principles calculations. Journal of the American Ceramic Society 95(9):2972–2978, 2012. https: //doi.org/10.1111/j.1551-2916.2012.05301.x [2] J. R. Salasin, C. Rawn. Structure property relationships and cationic doping in [Ca24Al28O64]4+ framework: A review. Crystals 7(5), 2017. https://doi.org/10.3390/cryst7050143 [3] G. Shin, F. Glasser. Interdependence of sodium and potassium substitution in tricalcium aluminate. Cement and Concrete Research 13(1):135–140, 1983. https://doi.org/10.1016/0008-8846(83)90137-0 [4] P. Mondal, J. W. Jeffery. The crystal structure of tricalcium aluminate, Ca3Al2O6. Acta Crystallographica Section B 31(3):689–697, 1975. https://doi.org/10.1107/S0567740875003639 [5] F. Lee, H. Banda, F. Glasser. Substitution of na, fe and si in tricalcium aluminate and the polymorphism of solid solutions. Cement and Concrete Research 12(2):237–246, 1982. https://doi.org/10.1016/0008-8846(82)90010-2 116 https://doi.org/10.1111/j.1551-2916.2012.05301.x https://doi.org/10.1111/j.1551-2916.2012.05301.x https://doi.org/10.3390/cryst7050143 https://doi.org/10.1016/0008-8846(83)90137-0 https://doi.org/10.1107/S0567740875003639 https://doi.org/10.1016/0008-8846(82)90010-2 Acta Polytechnica CTU Proceedings 47:113–116, 2024 1 Introduction 2 Material and methods 3 Results and discussion 4 Conclusion Acknowledgements References