Effect of Different High-energy Grinding Technologies on Granulometric and Crystallographic Properties of Portland Cement

SECTION BUILDING MATERIALS & DIAGNOSTICS OF STRUCTURES VOLUME: 22 | NUMBER: 2 | 2022 | DECEMBER © 2022 TRANSACTIONS OF VSB - TECHNICAL UNIVERSITY OF OSTRAVA CIVIL ENGINEERING SERIES 74 EFFECT OF DIFFERENT HIGH-ENERGY GRINDING TECHNOLOGIES ON GRANULOMETRIC AND CRYSTALLOGRAPHIC PROPERTIES OF PORTLAND CEMENT Simona RAVASZOVÁ1, Karel DVOŘÁK1 1Brno University of Technology, Faculty of Civil Engineering, Veveří 331/95, 602 00, Brno Czech Republic [email protected].cz, [email protected]tbr.cz DOI: 10.35181/tces-2022-0019 Abstract. The present article deals with the effect of different high-energy grinding technologies through available laboratory mills on Portland cement's granulometric and crystallographic properties. Two types of commonly available laboratory mills were selected, namely the planetary mill and the vibratory mill. The type of laboratory mill used, the grinding environment, and the grinding time were monitored. The planetary laboratory mill with 10 mm in diameter grinding steel balls proved to be more efficient in the scope of material reduction, which effectively reduces the material, especially during wet grinding of the material. The agglomeration effect is significantly reduced, which is a big advantage of method of wet grinding. Keywords High-energy milling, laboratory mills, Portland cement. 1. Introduction In the current era of environmental measures, it is important to look for ways to save energy in production processes. A large part of the energy in the production of cement is consumed in the grinding of Portland clinker. For several decades, research has been carried out on the topic of making the grinding process more efficient. Although the topic of the effect of grinding on Portland clinker is relatively well researched from the point of view of macrostructure, it is also necessary to focus on the microstructure and monitor the effect of grinding on the crystalline lattice of the ground material 1-2. Grindability is one of the important properties of a material that is monitored in a wide range of industries. This is the ability of the material to disintegrate into smaller particles due to mechanical work. From an economic point of view, material that can be milled well is more suitable for the preparation of powder raw material 3-5. One promising technology in milling is high-speed milling technology. The concept of high-energy or high-speed grinding is not precisely defined anywhere in the literature. It has all the basic properties in common with grinding in the classical sense, such as refinement of grain size, increase in surface area, the opening of grains, etc. In contrast to classical grinding, high-energy grinding (and also high-speed grinding) produces certain phenomena (effects) that are not observed in ordinary grinding. And it is to these effects that a certain part of the expended energy is converted, which in normal grinding is converted to heat without benefit 6-10. Mechanochemical activation enables much more efficient use of energy, spent on the treatment, i.e., grinding of the substance. The effect of mechanochemical activation makes it possible, for example, to improve chemical reactions in the solid-state during grinding in high-speed grinding devices 14-15. This paper examines the effect of two different highenergy grinding technologies on the crystallographic properties of the main clinker mineral Portland cement. One of the mills used will be a vibratory mill. It belongs to the group of devices with free storage of grinding bodies, the movement of which is not only dependent on gravitational forces. This type of mill is used for both dry and wet environment and for very fine grinding to a particle size of up to 10 µm. The second type of laboratory mill will be the planetary mill. It uses the principle of centrifugal acceleration, not gravity, which is used in commonly used ball mills. The planetary mill produces a high mechanical activation, already after a relatively short period of grinding. The energy density in this mill is 1001000 times higher than that of commonly used mills [16]. The article also examines other influences such as the milling time and milling conditions on the granulometric properties and shape of the grains of the milled material. SECTION BUILDING MATERIALS & DIAGNOSTICS OF STRUCTURES VOLUME: 22 | NUMBER: 2 | 2022 | DECEMBER © 2022 TRANSACTIONS OF VSB - TECHNICAL UNIVERSITY OF OSTRAVA CIVIL ENGINEERING SERIES 75 2. Materials and methods 2.1. Characteristics of materials Portland clinker (Hranice a.s., Czech Republic) and energogypsum (Precheza a.s., Přerov, Czech Republic) were used as the main raw materials. The clinker was crushed in a jaw crusher to a grain size of 4 mm. Energogypsum was dried at a temperature of 40 °C for 30 min. The resulting Portland cement was composed of 95% Hranice Portland clinker and 5% energygypsum. Material was analyzed by X-ray powder diffraction (XRD) and Xray fluorescence (XRF). The XRD analysis was performed using a Panalytical Empyrean diffractometer (Malvern Panalytical Ltd., Almelo, Neetherland). The Panalytical HighScore 3 plus software was used to identify the individual phases. The ICSD (released in 2012) was used for a qualitative analysis of the diffraction patterns. Quantification was performed by the Rietveld method with a fundamental parameters approach. The XRF analysis was performed using a PANalytical Axios 2.4 kW sequential wave dispersive spectrometer with Rh anode (Malvern Panalytical Ltd., Almelo, Neetherland). The acquired data was evaluated using the SUPERQ V4.0 software. The resulting values represent the average of three independent measurements. 2.2. The effect of grinding on the granulometric properties of cement The cement was ground in the following two types of available laboratory mills. A laboratory planetary mill (PULVERISETTE 6, Fritsch, Idar-Oberstein, Germany) and a laboratory vibratory disc mill (RS 200, Retsch, Haan, Germany) were used. In a planetary mill, the cement was ground at 400 rpm for 80 min. Samples were taken for analysis at 1, 3, 5, 10, 20, 40, 60, and 80min. Grinding took place in steel grinding capsules with a volume of 500 ml using two different sizes of grinding steel balls: 10 mm in diameter (weight for one grinding was 300 g) and 20 mm in diameter (weight for one grinding was 100 g). Grinding in the vibratory mill took place again for 80 min and, as in the previous case, samples were taken at times 1, 3, 5, 10, 20, 40, 60 and 80 min. Grinding in the planetary and vibratory mill took place in both dry and wet environments. In the case of wet grinding, acetone was used. Samples of variously ground cement were subjected to analysis to determine the specific surface area by means of Blain's apparatus. Blaine-specific surface area was measured using a PC-Blaine-Star automatic device with a measurement cell capacity of 7.95 cubic centimetres. The measurement was performed three times to eliminate errors and the resultant value was the average of three readings. 2.3. Effect of grinding on the size of crystallites Calculation of the crystallite size is based on the measurement of the full width of half maximum (FWHM). The FWHM values were obtained in the program HIGHSCORE PLUS for selected diffraction lines. The crystallite size was evaluated only for selected peaks on suitably selected crystallographic planes having a major crystallographic direction. Peaks were selected that were sufficiently high, had little overlap with other peaks, and at the same time they covered the chosen crystallographic plane directions. The crystallite size was evaluated by the modified Scherrer equation: 𝐿  .  .     .  .    (1) L is crystallite size, K is the Scherrer constant (0.94), λ is wavelength of X-ray radiation (1.540598), θ is diffraction angle, B is FWHM, b is FWHM of used standard (LaB6). 2.4. Effect of grinding on grain shape The effect of milling on grain shape was evaluated through SEM analysis by the scanning electron microscope TESCAN MIRA3 XMU. Samples were prepared using a colloidal graphite solution in isopropyl alcohol and a rotary pump pumped by a carbon sputter. 3. Results and discussion 3.1. Characteristics of materials The chemical composition of the raw materials used, and the mineralogical composition of the resulting cement are shown in the following tables. Tab. 1: Chemical composition of Portland clinker Hranice [%]. clinker CaO SiO2 Al2O3 Fe2O3 MgO K2O Na2O other 65.3 20.3 5.2 5.0 1.1 1.1 0.1 1.9 Tab. 2: Chemical composition of energygypsum [%]. gypsum CaSO4.2H2O H2O CaSO3 TiO2 SiO2 Al2O3 Fe2O3 other 84.0 11.0 2.4 1.2 0.6 0.4 0.3 0.1 Tab. 3: Mineralogical composition of Portland clinker Hranice [%]. clinker C3S C2S CaO C4AF C3A* C3A** B Calcite G 71.2 8.7 0.0 8.7 3.0 1.7 1.2 1.8 2.8 *cubic, **orthorhombic, B-basanite, G-gypsum. According to the results of mineralogical analysis, Portland cement Hranice contains a high amount of alite. The mineral alite is found in the monoclinic modification, specifically M1. Therefore, crystallinity will be assessed SECTION BUILDING MATERIALS & DIAGNOSTICS OF STRUCTURES VOLUME: 22 | NUMBER: 2 | 2022 | DECEMBER © 2022 TRANSACTIONS OF VSB - TECHNICAL UNIVERSITY OF OSTRAVA CIVIL ENGINEERING SERIES 76 only on alite as the dominant phase, because other clinker minerals are represented in small amounts and their crystallinity would be very difficult to assess. 3.2. The effect of grinding on the granulometric properties of cement The input specific surface area of cement was 230 m2/kg. In Fig. 1, trends for grinding cement in individual types of laboratory mills in wet and dry environments are shown. Fig. 1: Dependence of specific surface area on grinding time for individual types of mills and environment. Dry milling After just 20 minutes of grinding, the value of the specific surface reached 660 m2/kg. When using 20 mm in diameter grinding balls, a value of 560 m2/kg was achieved. The lowest specific surface values of only 480 m2/kg were recorded after 20 minutes in the vibratory mill. As can be seen from the progress of grinding, a longer time no longer had a significant effect on the increase in the specific surface of the cement. The values remained approximately constant for further grinding time up to 80 minutes. Wet milling During wet grinding, the highest specific cement surface values were achieved again by grinding in a planetary mill with 10 mm in diameter grinding balls. After 80 minutes of grinding, a specific surface of 1470 m2/kg was achieved. With 20 mm in diameter balls, only 990 m2/kg was achieved after 80 minutes of grinding. For the vibratory mill, the value of the specific surface was determined to be 1090 m2/kg after 80 minutes of grinding. During wet grinding, the influence of agglomeration is significantly reduced, therefore the specific surfaces increased throughout the grinding period for both types of mills. The grinding limit was probably reached after 60 minutes by grinding in a planetary mill with 20 mm in diameter of grinding balls. In both methods of grinding in dry and wet environments, the highest fineness of cement was achieved in a planetary mill with 10 mm grinding balls. Based on the development of grinding fineness, it can be concluded that higher specific surfaces can be achieved with wet grinding than with dry grinding. During wet grinding, the agglomeration effect is significantly reduced, which is a big advantage of this method of grinding. However, high surface area values for wet milling can be misleading, as the accuracy of the Blaine apparatus decreases at high surface area values around 1000 m2/kg. On the other hand, during the dry grinding the material accumulates between the rings and thus dampens the grinding efficiency. During wet grinding the suspension spreads spontaneously over the entire grinding area and grinding is thus more efficient 3.3. Effect of grinding on the size of crystallites Due to the complexity of the lattice, it was difficult to select crystallographic planes that would clearly copy the major crystallographic direction, and which would also be reflected in the diffraction pattern by distinct isolated peaks. For this reason, distinct crystallographic planes determined by Miller indices „ hkl “41-1 were selected. The following Tab. 4 shows the crystallite size values calculated based on Equation 1. Tab. 4: Calculated crystallite size Å values for types of mills and milling environments. Time [min] Dry milling Wet milling PM 10mm PM 20mm VM PM 10mm PM 20mm VM 1 947.2 936.7 852.7 897.2 1038.8 916.5 3 916.5 1091.0 980.2 926.5 968.9 980.2 5 878.8 869.9 957.9 887.9 887.9 968.9 10 743.8 790.7 957.9 763.2 763.2 957.9 20 628.5 731.5 820.4 686.3 686.3 869.9 40 509.0 697.0 812.8 697.0 697.0 869.9 60 445.1 651.3 776.7 651.3 651.3 828.3 80 316.6 569.3 713.8 569.3 569.3 783.6 PM-planetary mill, VM-vibratory mill SECTION BUILDING MATERIALS & DIAGNOSTICS OF STRUCTURES VOLUME: 22 | NUMBER: 2 | 2022 | DECEMBER © 2022 TRANSACTIONS OF VSB - TECHNICAL UNIVERSITY OF OSTRAVA CIVIL ENGINEERING SERIES 77 Fig. 2: Graphic representation of the dependence of crystallite size on grinding time. 3.4. Effect of grinding on grain shape As part of the evaluation of the effect of grinding on grain shape, grinding in a planetary mill with 10 mm in diameter balls was compared with grinding in a vibratory mill. Dry and wet grinding were compared. The images taken were magnified 10000×. Fig. 3: SEM images of comparison of grinding in a planetary and vibratory mill in dry grinding after 10, 20, 40 and 80 minutes. Fig. 4: SEM images of comparison of grinding in a planetary and vibratory mill in wet grinding after 10, 20, 40 and 80 minutes. The difference between the influence of the planetary and vibratory mill on the shape of the grains is already observable after 10 minutes of grinding, where in the case of vibratory mill, the particles are larger than in the case of grinding in a planetary mill. Even at advanced times of grinding, a planetary mill grinds more intensively and creates more agglomerates than a vibratory mill. In the case of dry grinding, the difference between the used grinding technologies is already obvious after 10 minutes of grinding. In vibratory grinding, the ratio of larger grains in the order of units of micrometres prevailed over smaller particles, while in the case of grinding in a planetary mill, the ratio of smaller particles already prevailed. From the 20th minute onwards, agglomerates clump together for both technologies. At each grinding time when images were taken, the planetary grinding technology appears to be more efficient. In the case of wet grinding, the differences between the technologies used are not that significant. In both cases, however, a decrease in the formation of agglomerates can be observed, which is one of the characteristics of this grinding method. 4. Conclusions The main objective of this article was to assess the influence of different grinding technologies on the granulometric and crystallographic properties of Portland cement. Not only the type of laboratory mill used but also the grinding environment and grinding time were monitored. The planetary laboratory mill with 10 mm in diameter grinding steel balls appears to be the most efficient mill in the scope of material reduction, which effectively reduces the material especially during wet grinding of the material. During wet grinding, the agglomeration effect is significantly reduced, which is a big advantage of this method of grinding. Acknowledgements This research was funded by the project “Effects of mechano-chemical activation on the process of formation, structure and stability of selected clinker minerals” (2000676S) and by the project “A study of the crystallization kinetics of various polymorphs of tricalcium silicate” (FAST-J-21-7309). References 1 SAHOO, N., KUMAR, A., Review on energy conservation and emission reduction approaches for cement industry, Environmental Development, 44, 2022 [2] MUKHERJEE, S. Applied Mineralogy: Applications in Industry and Environment., Springer Science & Business Media, 2012, 562 p., ISBN: 978-94-0071162-4 [3] NEIKOV, O. D. Chapter 2 - Mechanical Crushing and Grinding, Elsevier, 2019, 65-90, ISBN 9780081005439 SECTION BUILDING MATERIALS & DIAGNOSTICS OF STRUCTURES VOLUME: 22 | NUMBER: 2 | 2022 | DECEMBER © 2022 TRANSACTIONS OF VSB - TECHNICAL UNIVERSITY OF OSTRAVA CIVIL ENGINEERING SERIES 78 [4] ZAWADA, J. Introduction to the Mechanics of Crushing Processes; Publishing and Printing House of the Institute for Sustainable Technologies: Radom, Poland, 1998. 5 BALÁŽ P.: Mechanochemistry in Nanoscience and Minerals Engineering, Chapter 2, High - Energy Milling, Springer, Hardcover, Netherland, 2008, ISBN: 978-3-540-74854-0 6 TKÁČOVÁ, K.: Mechanical Activation of Minerals. Amsterdam: Elsevier, 1989, Developments in mineral processing 7 MUCSI, G., RÁCZ, A., MÁDAI, V. Mechanical activation of cement in stirred media mill, Powder Technology, 235, 2013, 163-172, 8 ALTUN, O., BENZER, H., KARAHAN, E., ZENCIRCI, S., TOPRAK, A. The impacts of dry stirred milling application on quality and production rate of the cement grinding circuits, Minerals Engineering,155, 2020, 9 IVANOV, E., SURYANARAYANA, C. Materials, and process design through mechanochemical routes J. Mater Synth. Process., 8, 2000, 235-244 [10] SURYANARAYANA, C. Mechanical alloying and milling Prog. Mater. Sci., 46 (2001), 1-184, https://doi.org/10.1016/S0079-6425(99)00010-9 [11] HIGISHITANY, M., MASUDA, H., YOSHIDA, H. Powder technology handbook, Third edition, CRC Press, 2006, 920 p., ISBN 9781574447828 [12] DENIZ, V. The effects on the grinding parameters of chemical, morphological and mineralogical properties of three different calcites in a Hardgrove mill, Minerals Engineering, 176, 2022, https://doi.org/10.1016/j.mineng.2021.107348. [13] BALÁŽ, P., CHOI, W., FABIAN, M., DUTKOVÁ, E. Mechanochemistry in the Preparation of Advanced Materials, 11, 2006, 122-129 14 BOLDYREV, V.V., PAVLOV, S.V., GOLBERG, E.L. Interrelation between fine grinding and mechanical activation, Int. J. Miner. Process., 44–5, 1996, 181185 [15] HINT J.A. The basic problems of mechanical activation, in: Reports 5th Symp. Mechanochemistry Mech. Hard Mater., 1977, 14–32 [16] BOND F. C.: Crushing and grinding calculations Part I, British Chemical Engineering 6 1961.