S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... 705–713 EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING OF THE BELITE-SULFOALUMINATE CEMENT CLINKER USING X-RAY COMPUTED MICROTOMOGRAPHY RAZVOJ POR MED SINTRANJEM BELITNO-SULFOALUMINATNEGA CEMENTNEGA KLINKERJA Z RENTGENSKO RA^UNALNI[KO MIKROTOMOGRAFIJO Sabina Dolenec 1* , Lidija Korat 1 , Maru{a Bor{tnar 1 , Andrej Ipavec 2 , Lea @ibret 1 1 Slovenian National Building and Civil Engineering Institute, Dimi~eva Steet 12, 1000 Ljubljana, Slovenia 2 Salonit Anhovo, d.d., Anhovo 1, 5210 Deskle, Slovenia Prejem rokopisa – received: 2020-02-12; sprejem za objavo – accepted for publication: 2020-05-02 doi:10.17222/mit.2020.031 In this paper, the effects of the sintering temperature on the pore evolution of the belite-sulfoaluminate cement clinker were evaluated. Belite-sulfoaluminate cement clinker with a targeted composition of 65 w/% -belite, 20 w/% calcium sulfoaluminate and 10 w/% ferrite was sintered at three different temperatures: 1200 °C, 1250 °C and 1300 °C. To quantitatively evaluate the pore evolution during sintering, a 3D microstructure reconstruction by micro-CT was used. From the data, the pore volume frac- tion and pore number were extracted and compared, and the pore size distribution with the sintering temperature was obtained as well. Additionally, the pore shape and distribution were displayed in 3D based on actual microstructure data. Clinker samples were also characterized by Hg-intrusion porosimetry and gas sorption. The changes in the pore evolution occurred to a larger ex- tent when sintering at 1300 °C. Apart from a significant porosity decrease, pore coarsening was evident at this temperature, re- ducing their connectivity and shrinkage of the clinker. Simultaneously, the bulk and apparent densities increased with the tem- perature due to densification, while the BET surface area of the studied clinkers decreased, indicating the rounding of pores and particle coalescence with an increasing grain growth. Keywords: clinker, belite-sulfoaluminate, pore evolution, sintering, μ-CT Prispevek obravnava vpliv temperature sintranja na razvoj por pri belitno-sulfoaluminatnem cementnem klinkerju. Klinker z `eleno sestavo faz 65 w/% -belita, 20 w/% kalcijevega sulfoaluminata in 10 w/% ferita je bil sintran pri treh razli~nih temperaturah: 1200 °C, 1250 °C in1300 °C. Z ra~unalni{ko mikrotomografijo smo na podlagi 3D rekonstrukcije mikrostrukture kvantitativno ocenili razvoj por pri sintranju. Iz podatkov smo pridobili dele` volumna por in {tevilo por ter tudi porazdelitev por. Poleg tega je bila v 3D prikazana tudi oblika in porazdelitev por. Vzorci klinkerja so bili okarakterizirani tudi z `ivosrebrovo porozimetrijo in plinsko sorpcijo. Rezultati ka`ejo, da so se najve~je spremembe pri razvoju por zgodile pri sintranju na 1300 °C. Poleg ob~utnega zmanj{anja poroznosti je bilo pri tej temperaturi opaziti tudi ve~anje por, zmanj{evanje povezanosti med njimi in kr~enje klinkerja. Obenem sta se volumska in navidezna gostota s temperaturo pove~ali zaradi zgo{~evanja, medtem ko se je specifi~na povr{ina BET klinkerjev zmanj{ala, kar ka`e na zaobljevanje por in pove~evanje zrn. Klju~ne besede: klinker, belit-sulfoaluminat, razvoj por, sintranje, μ-CT 1 INTRODUCTION The mechanisms of clinker formation are very com- plex, especially due to the many constituents and phases present in the process, as well as the mineralogical and physical changes that occur simultaneously. 1 Clinkeri- zation comprises solid-state reactions, solid-liquid and/or liquid-state reactions, polymorphic transformation on heating, and polymorphic stabilisation and crystallization upon cooling. 2,3 Depending on the reaction conditions and variables such as the maximum heating temperature, the rate of temperature increase, retention time, the rate of cooling, the composition of the gaseous atmosphere and the chemical composition of kiln raw meal mixture, 4 a significant variety of phases, as well as the relative pro- portions can be formed, which will have an effect on the clinker and consequently also on the properties of the re- sulting cement and concrete. Furthermore, during sintering the density, shape, size and distribution of the pores change. 5 Pores play an important role in sintering densification, whereby interparticle pores in particulate materials are eliminated by atomic diffusion at high tem- peratures. 6,7 Porosity influences the sintered properties 8 and is also one of the factors, as well as other micro- structural factors (i.e., the crystal size, morphology, dis- tribution and content of individual clinker phases, or presence of trace elements) that affect the grindability of cement clinker and consequently determines the econ- omy of cement production. 9–11 Namely, the initial grindability was found to be determined by porosity 12 and it increased by hard burning and high melt content since they result in a clinker with a low porosity. 9 Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 705 UDK 620.1:662.613.12:621.762.3:621.78 ISSN 1580-2949 Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 54(5)705(2020) *Corresponding author's e-mail: sabina.dolenec@zag.si (Sabina Dolenec) While the clinkerization of conventional Ordinary Portland Cement clinkers has been extensively investi- gated over previous decades, there is only a limited num- ber of studies discussing the clinkerization processes of belite-sulfoaluminate cement clinkers in detail. 13–17 The cement based on those clinkers is considered to be low-CO 2 and low-energy cement due to the lower energy use, lower CO 2 emissions (20–30 % reduction compared to OPC) and consumption of by-products as raw materi- als. 13,15,18,19 The main mineral analogues in this clinker are larnite (belite, C 2 S) and sodalite-type ye’elemite (C 4 A 3 Œ), also called Klein’s salt tetracalcium aluminate sulphate or calcium sulfoaluminate. It may also contain other minor phases, such as brownmillerite-type ferrite (C 4 AF), mayenite (C 12 A 7 ), gehlenite (C 2 AS), perovskite (CT), periclase (M) and excess anhydrite (CŒ). 13,16,19–21 Such clinkers are generally produced by burning raw ma- terials like limestone, clay and bauxite in addition to cal- cium sulphate, within a temperature range of 1200 °C to 1350 °C. 18,22 However, several industrial by-products are considered as substitutes for natural raw materials, e.g., coal fly ash, slag, red mud, phosphogypsum, flue gas desulfurization (FGD) gypsum, etc. 15,23 Various approaches are possible in the investigation of porosity and a pore network of materials, but most of them do not give any information on the dimensional distribution of the pore structure components. Conven- tionally, the materials characterisation is performed in 2D, using scanning electron microscopy (SEM), while data acquired by Hg-intrusion porosimetry (MIP) only give quantitative information on the pore structure, and they are inadequate to know the dimensional distribution of the pores, and their connectivity. 24 In addition, MIP cannot detect closed porosity, and due to the ink-bottle effect it overestimates the smaller pore population. 25 Fur- thermore, the sample preparation processes in these test- ing methods such as polishing, drying and especially the high pressure employed by the MIP method can often damage the microstructure of the samples before or dur- ing the analysis. 26,27 3D pore-scale imaging methods, like high-resolution X-ray computed tomography are now widely used to image and reconstruct 3D images of ma- terials with a resolution on the micrometre-to-nanometre scale. 26,28–30 Clearly, X-ray computed tomography is a powerful non-destructive test method for porosity and defect analysis of various materials, providing basic in- formation on average porosity and pore size distribution, as well as information on pore connectivity or pore shapes through complex 3D analyses. 31 In this paper, belite-sulfoaluminate cement clinker with a targeted composition of 65 w/% -C 2 S, 20 w/% C 4 A 3 Œ,10w/% C 4 AF was sintered at three different tem- peratures: 1200 °C, 1250 °C and 1300 °C. 3D micro- structure images reconstructed by μ-CT together with the results of Hg-intrusion porosimetry and gas sorption are used to understand the evolution of pores and pore net- works. Pore fraction volume, pore number, pore size and distribution, and density were measured. The correla- tions of these parameters with the sintering process were discussed. 2 EXPERIMENTAL PART 2.1 Materials Cement clinker having the nominal phase composi- tion 65 w/% -C 2 S, 20 w/% C 4 A 3 Œ,10w/% C 4 AF was synthesised for the study. 32 The clinker was prepared with ratios of limestone, flysch, bauxite, white titanogypsum and bottom ash from a coal thermal power plant (smaller quantities of mill scale were used for cor- rection). The materials were proportioned by adapting the modified Bogue method. 19 All the raw materials were first ground to pass through a 200-μm sieve. The raw mixture (200 g) was then homogenized and ground for3hi n2 0 0m lo f isopropanol using a ball mill (CAPCO Test Equipment Ball Mill Model 9VS). Pressed pellets were prepared us- S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... 706 Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 Figure 1: Macroscopic behaviour of the cement clinker. Sample 0: un-sintered compact; Sample 1: sintering at 1200 °C; Sample 2: sintering at 1250 °C, Sample 3: sintering at 1300 °C Table 1: Mass change and dimensional behavior of the cement clinker during sintering at different temperatures Sintering (T) Weight (g) Diameter (mm) Height (mm) Mass change (%) Radial change (%) Axial change (%) Non-sintered sample 15.00 30.04 13.07 1200 °C 9.62 30.70 13.44 −35.86 +2.20 +2.83 1250 °C 9.83 24.14 11.69 −34.47 −19.64 −10.56 1300 °C 9.85 22.90 9.70 −34.33 −23.77 −25.78 ing a HPM 25/5 press at 10.6 kN. For each pellet, with a diameter of 30 mm, 15 g of material was used. The clinker mixtures were subjected to the following heating regimes: a) heating to 1200 °C, 1250 °C and 1300 °C, b) heating rate of 10 °C/min, c) holding time at the final temperature for 60 min, and d) natural cooling in a closed furnace. Figure 1 illustrates the macrographs of the cement clinker powder compacts before and after sintering. Table 1 provides mass and dimensional changes during sintering of the clinker compacts. 2.2 Methods X-ray micro-computed tomography (μ-CT) was used to study the structural characteristics of the investigated samples, using an "Xradia μCT-400" tomograph (Xradia, Concord, California, USA), set to 140 kV and 71 μA. The specimens had a diameter of 7 mm and they were imaged using a CCD camera equipped with a 10× mag- nification optical objective and a resolution of 2 μm. A high-precision rotating stage was used, so that 1600 pro- jection images were taken from different view-points with exposure times of 4 s per projection. In order to re- construct the pore structure of the cement clinkers, as well as to determine their overall porosity and pore size distribution, Avizo Fire 3D-image analysis software was used, following the process for pore segmentation and quantification as described by Korat et al. 29 The ROI (re- gion of interest) box was determined from the centre of the sample, which had a size of 4 mm. Image segmenta- tion was performed (based on the voxel intensity histo- gram of the grayscale image), to determine the difference between the solid matrix and the air. A watershed algo- rithm was applied to the binary images in order to sepa- rate the pores step by step with a distance map. Once the separate pores have been identified, they can be labelled and further measured. Using a multiple labelling process, each pore can be identified in the label image, and as- signed as unique index, mostly to get the volume, surface area, mean value etc. Such labelled images are displayed and quantified individually. Within this research, equiva- lent diameter and 3D volume (within Label Analysis module) were chosen. The pore system of the cement clinker samples sintered at different temperatures was investigated by means of mercury intrusion porosimetry (MIP). Small representative fragments, approximately 1 cm 3 in size, were dried in an oven for 24 h at 105 °C and then ana- lysed by Micromeritics®Autopore IV 9500 equipment (Micromeritics, Norcross, GA, USA). The samples were analysed within the range of 0–414 MPa using pen- etrometers for solid substrates. Two measurements were performed for each sintered clinker sample. Nitrogen adsorption measurements were performed at 77 K using a Micromeritics ASAP-2020 analyser (Micromeritics, Norcross, GA, USA). The cement clinker samples were crushed into small pieces, approxi- mately 0.5 cm 3 in size. The mass of the analysed samples was 2 g. The total specific surface area, the total pore volume and the pore-size distribution curves of the sub- strates were determined using the Brunauer–Emmet– Teller (BET) method, t-plot analyses and the Barrett–Joyner–Halenda (BJH) method, respectively. 33,34 The total pore volume and micropore volume of the sam- ples were calculated using t-plot analysis. The BJH method was used to obtain pore size distribution curves. 35 Two measurements were performed for each sintered clinker sample. 3 RESULTS AND DISCUSSION For the sintered cement clinker, the overall porosity, determined by μ-CT, slightly increased from 5.56 a/% 1200 °C to 6.32 a/% 1250 °C. Considering the sample sintered at 1300 °C, a decrease in total porosity to 3.28 % was observed. As seen from the MIP results in Table 2, the trend in porosity evolution during sintering is consistent with the μ-CT data, although with notice- able differences in absolute values, specifically because MIP covers a pore size range from 0.0055 μm (5.5 nm) to 360 μm and μ-CT a pore size range above 1 μm. Moreover, the MIP would detect open pores only, so that the results would not be representative for the determina- tion of closed pores. Thus, the open porosity of the clinker sintered at 1250 °C, determined by MIP, was increased to 39.0 % compared to 33.4 % of the 1200 °C sample. After that, the open porosity significantly decreased to 9.6 % at 1300 °C. In the early stages, the necks forming at the points of particle contacts have only small cross-sec- tional areas and particle rearrangements can take place during solid-state sintering, resulting in the local opening of large pores and the breaking of some contacts that are related to porosity growth. 36 The increase in porosity from 1200 °C to 1250 °C could also indicate that the pore formation was dominated by the thermal explosion reaction, 37 resulting in the deformation of the 1250 °C S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 707 Table 2: Porosity, average pore diameter, bulk density, and apparent density of the sintered cement clinker samples, as determined by Hg-porosimetry Porosity (%) Average pore diameter (μm) Median pore diameter by volume (μm) Bulk density (g/mL) Apparent density (g/mL) 1200 °C 33.4±0.7 5.36±0.13 5.06±0.08 1.00±0.01 1.50±0.04 1250 °C 39.0±0.1 3.49±0.14 4.05±0.03 1.82±0.02 2.98±0.03 1300 °C 9.6±0.4 0.02±0.01 0.42±0.55 2.72±0.04 3.01±0.02 clinker (Figure 1). Pore coarsening leads to a reduced gas pressure in the pores resulting in a porosity increase. 8 Namely, in the regime of high temperature ( 1200 °C), calcium sulfate is chemically unstable and provokes the release of SO 2 . 38 On the other hand, the reduction of the total accessible porosity of the sample sintered at 1300 °C is due to the densification of the microstructure leading to a drop in connectivity between the pore spaces, resulting in a reduction of open pores, i.e., chan- nel closure. 36 Reduction of porosity could also be ex- plained by the formation of a liquid phase, accompanied by the shrinkage of the compact due to the dissolving of particles, decreasing of distances between the centres of solid-phase particles, and melt penetrating into the interparticle pores. 39 At temperatures up to 1250 °C, solid-state reactions occur, while at higher temperatures, a liquid phase is formed. 3 Most of the belite phase as well as the ferrite phase formation occur below around 1250 °C via solid-solid interactions. 3 In general, belite is the first stable product to be formed at temperatures be- tween 1000 °C and 1200 °C. 40 Between 1100 °C and 1200 °C, the ferrite phase forms. 41 Calcium sulfoalumi- nate is formed at 900–1000 °C, which increases with the increasing temperature, peaking at 1300–1350 °C. 38 Moreover, the decrease in porosity could indicate an in- crease of the grain size, as an inverse relationship be- tween grain size and fractional porosity during sintering has been reported. 42 The reduction of pore volume results in shrinkage of the sintered clinkers (Figure 1, Table 1). While the mass does not change significantly with the sintering tempera- ture, the radial and axial shrinkage was obvious at sintering above 1200 °C. Furthermore, it was noted that the shrinkage increased sharply with increasing tempera- S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... 708 Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 Figure 2: Pore size distribution determined by the means of μ-CT for the samples, presented as: a) the volume of pores with regard to pore size diameter and b) as the number of pores with regard to pore size diameter Figure 3: 2D images (XZ view) of samples sintered at 1200 °C, 1250 °C and 1300 °C with enlarged bottom figures regarding FOV (field of view) ture, where a radial shrinkage of 23.77 % and an axial shrinkage of 25.78 % were reached at 1300 °C. With sintering temperature, differences also occur in pore-size distribution as indicated by μ-CT, MIP and gas sorption. The pore size distribution of the clinkers extracted from μ-CT data (Figure 2a) showed that the clinker sam- ple sintered at 1300 °C had the largest volume of larger pores (10–100 μm) as well as smaller pores (1–5 μm). On the other hand, the volume of pores in the range of 5–10 μm was smaller compared to the samples sintered at 1200 °C and 1250 °C. The volume of pores larger than 100 μm was almost negligible with all three sintering temperatures. The change in pore size distribution was smaller when comparing samples sintered at 1200 °C and 1250 °C than samples sintered 1250 °C and 1300 °C, similar to the observations for the dimensional and po- rosity changes described above. However, a continuous increase of larger pores in the range 10–100 μm with in- creasing temperature was obvious, which suggests the transition of the pores to isolated spherical pores. 43 Fur- thermore, as seen from Figure 2b, the number of pores, determined by μ-CT was significantly smaller for the sample sintered at 1300 °C. The pore size grows with sintering temperature, although a reduction in porosity occurs because of the elimination of pores from the sys- tem. 44 Consequently, the coalescence of small pores sig- nificantly impedes the densification process, 45 which can lead to a pore size distribution shift towards larger S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 709 Figure 4: 3D-pore distribution of cement clinkers sintered at different temperatures with respect to different pore ranges. The region of interest being approximately 1 mm in size (presenting overall porosity) and 200 μm in size (presenting connectivity/isolation of large pores on large scale), taken from the centre of the sample. sizes. 46 Pores collapse into a closed condition during sintering in what is termed the final stage, and due to pressure balances they grow as porosity decreases. 8 Apart from the larger pores, a shift to smaller pores with the enhanced temperature of sintering is obvious. At 1300 °C, small pores formed and showed a shrinkage ef- fect. 37 From the results presented in Figure 3, it is seen that when clinker samples are sintered, pore growth effects the result in altered pore morphology. For a better visu- alisation, Figure 4 shows the 3D pore shape and pore distribution of the clinker samples, the region of interest being approximately 1 mm in size, taken from the centre of the sample. It is evident that at lower sintering temper- atures, the pores are interconnected, while with contin- ued sintering, pores become isolated and grow. Namely, only late in the sintering process do the channels close and isolated pores are formed that may grow or, in those rare cases where full density is achieved, shrinking to zero. 36 A visual examination of CT data, prepared by Avizo software, has shown that open pores are typically thin, elongated and irregularly shaped, while the closed pores are typically more equiaxed (Figure 4). When the samples are sintered at higher temperatures, the large amount of small isolated pores disappear and some pores merge with others to form large pores. 7 As seen from Figure 4 (last row), pores are mostly isolated at 1300 °C. In regard to the results of the MIP (Figure 5), the pores of the samples sintered at 1200 °C and 1250 °C were unimodally distributed, with the largest intrusion being around 5 μm and 3 μm, respectively, with a shift toward smaller pores with a higher temperature. A bi- modal distribution of pores was characteristic of the sample sintered at 1300 °C, with an intrusion peak at around 0.01 μm and a larger intrusion peak at around 200 μm, yet again indicating a major change in the clinker microstructure that occurred during sintering at 1300 °C. Both the average and median pore diameters decreased with the sintering temperature, while the dif- ference between 1200 °C and 1250 °C was much smaller than between 1250 °C and 1300 °C (Table 2). Despite that the average pore diameter was smallest for the 1300 °C sample, from the pore size distribution it is clearly seen that the amount of larger pores increased in comparison to the samples sintered at lower tempera- tures. This could suggest pore shrinkage due to densification, simultaneous with pore coarsening. How- ever, the fact that MIP only covers open pores should be considered. Figure 6 shows that the amount of pores diminished with sintering temperature, supporting the results ob- tained by μ-CT. In terms of pore-size distribution, sintering resulted in a decrease or increase in the per- S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... 710 Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 Figure 6: Pore-size distribution of the clinker samples based on the Hg-porosimetry data Figure 5: Log differential intrusion versus pore size of the samples centage of pores in certain size ranges, which is espe- cially evident for the sample sintered at 1300 °C. As indicated by the nitrogen adsorption method sam- ples sintered at all three temperatures had a similar, con- tinuously increasing type of pore size distribution (Fig- ure 7a), in which the overall trend was that the finer the pore diameter, the greater the volume of pores devel- oped. Results showed a clear reduction in macropores (r > 50 nm) and mesopores (2 nm < r < 50 nm), since there was an increase in micropores (r < 2 nm) for the sample sintered at 1300 °C (Figure 8). The volume of pores accessible to the gas decreased with the sintering temperature (Table 3, Figure 8), due to the densification process and the formation of a liquid phase. The average pore diameter of the sintered clinkers was in the range of 5.56 nm to 6.43 nm (Table 3). All of the samples studied had a Type-II physisorption isotherm (Figure 7b), which is characteristic of non-porous and macroporous materi- als with diameters exceeding micropores. 47 Table 3: Results of the N 2 -adsorption measurements of the samples investigated Sintering T/°C BET surface area (m 2 /g) Total pore volume (cm 3 /g) Average pore diame- ter (nm) Micropore volume (cm 3 /g) 1200 °C 0.63±0.01 0.000871 5.56±0.08 −0.000005 1250 °C 0.37±0.04 0.000543 6.43±0.12 −0.000038 1300 °C 0.05±0.01 0.000085 5.84±0.10 −0.000014 Furthermore, both the bulk density and the apparent density show an upward tendency as the sintering tem- perature rises, which indicates the sample gradually un- dergoing densification (Table 2). The bulk density of the samples increased linearly with the sintering tempera- ture, from 1.00 g/mL at 1200 °C to 2.72 g/mL at 1300 °C (Figure 9). As the bulk density of the clinkers increased, individual grains began to grow – coarsening the grains. On the other hand, apparent density, correlated with in- ternal porosity, increased to the temperature of 1250 °C, indicating the internal porosity constantly eliminating, but at 1300 °C it remained unchanged (Figure 9), which can be explained by the internal pores being nearly non-existent. 48 Furthermore, pores require proximity to a grain boundary to shrink during sintering, while pores separated from grain boundaries remain stable and resist densification. Accordingly, grain growth is detrimental to the sintering since the grain boundaries become more widely spaced, have less proximity to pores, and often move faster than the pores. Rapid grain coarsening gen- erates the conditions where sintering prematurely termi- nates with a considerable residual porosity. 8 As the S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 711 Figure 7: The results of N 2 adsorption measurements: a) Pore-diame- ter logarithm differential distribution graph of clinker samples by ni- trogen adsorption, b) The N 2 -physisorption isotherms for the samples investigated Figure 8: Pore-size distribution of the clinker samples based on the nitrogen-adsorption data Figure 9: Porosity, bulk density and apparent density of clinker sam- ples sintered at different temperatures clinker cools, the main liquid phase crystallizes to form phases. Since higher temperatures inflate the balloons due to the sulfate exhalation, leading to clinker swelling, pores near free surfaces disappear, but internal pores un- dergo coarsening. 8 Consequently, the density increases during sintering with simultaneous pore growth instead of the expected pore shrinkage. On the other hand, the small pores that were interconnected or partly connected would develop into closed pores during the high-temper- ature sintering process, thus reducing the open porosity (ref. no. 37 ). The pores are assumed to be interconnected at low densities, cylindrical at intermediate densities, and spherical at high densities. 49,50 The values of the BET specific surface area de- creased with the sintering temperature (Table 3). In addi- tion to densification, this fact could also be due to two mechanisms, such as the rounding of pores (as evident from Figure 8) that decrease the surface area and parti- cle coalescence with increasing grain growth (ref. no. 5 ). Yet, many structural changes compatible with surface area reduction may occur, i.e., grain growth, pore shrink- age or growth, particle modification and variation in den- sity (ref. no. 5 ). 4 CONCLUSIONS Pore evolution of belite-sulfoaluminate cement clinker during sintering at different temperatures was evaluated using μCT, Hg-porosimetry and gas sorption. Larger changes in clinker microstructure were ob- served at1250 °C to 1300 °C than at 1200 °C to 1250 °C. The porosity of the clinkers increased when sintering from 1200 °C to 1250 °C and then diminished at 1300 °C, which is attributed to pore coarsening and densification. Pores are interconnected at lower sintering temperatures, while pore isolating and the volume of larger pores increased with the sintering temperature due to pore coalescence, which was evident in 3D analysis using μ-CT. Furthermore, with increasing temperature, the BET-specific surface was reduced, while the bulk and apparent densities increased, indicating densification of the clinker microstructure. The evolution of the pores during clinker sintering can be useful in understanding how sintering parameters like temperature might influence the evolution of microstructure and hence macroscopic properties. Acknowledgment Project No. C3330-17-529035 "Raziskovalci- 2.0-ZAG-529035" was granted by the Ministry of Educa- tion, Science and Sport of the Republic of Slovenia. The investment is co-financed by the Republic of Slovenia, Ministry of Education, Science and Sport and the Euro- pean Regional Development Fund. The authors acknowl- edge the financial support from the Slovenian Research Agency (research core funding No. P2-0273). The au- thors would also like to acknowledge Mr Andrej Kranjc for his help with μ-CT data processing. 5 REFERENCES 1 B. Hökfors, D. Boström, E: Viggh, R. Backman, On the phase chem- istry of Portland cement clinker, Adv. Cem. Res ., 27 (2015)1 , 50–60, doi:10.1680/adcr.13.00071 2 M. S. Idris, K. N. Ismail, S. B. Jamaludin, Comparative Character- ization of Clinkers microstructure at different temperature zone dur- ing cement production, Am. J. Appl. Sci, 4, (2007) 8, 543-546, doi: 10.3844/ajassp.2007.543.546 3 S. Telschow, F. Frandsen, K. Theisen, K. Dam-Johansen, Cement Formation—A Success Story in a Black Box: High Temperature Phase Formation of Portland Cement Clinker, Ind. Eng. Chem. Res., 51 (2012), 51, 10983-11004 doi:10.1021/ie300674j 4 W. Kurdowski, Cement and Concrete Chemistry, Springer, New York 2014, 700 str. 5 J. A. Varela, O. J. Whittemore, E. Longo, Pore size evolution during sintering of ceramic oxides, Ceram. Int., 16 (1990) 3, 177-189, doi:10.1016/0272-8842(90)90053-I 6 P. L. Chen, I. W. Chen, Sintering of Fine Oxide Powders, 1. Microstructural Evolution, J. Am. Ceram. Soc., 79 (1996)2 , 3129–41, doi:10.1111/j.1151-2916.1996.tb08087.x 7 Z. Hu, K. Lu, Evolution of Pores and Tortuosity During Sintering, J. Am. Ceram. Soc., 97 (2014) 8, 2383–2386, doi:10.1111/jace.13080 8 R. M. German, Coarsening in Sintering: Grain Shape Distribution, Grain Size Distribution, and Grain Growth Kinetics in Solid-Pore Systems, Crit. Rev. Solid. State, 35 (2010) 4, 263-305, doi:10.1080/10408436.2010.525197 9 G. C. Bye, Portland Cement: Composition, Production and Prop- erties, 1983, 149. str. 10 L. Hills, Clinker microstructure and grindability: updated literature review, SN2967, Portland cement assocation, Skokie, Illinois, USA, 2007, 15 str. 11 B. Felekoðlu, K. Tosun, B. Baradan, A. Altun, Relationship between clinker porosity and interstitial phase morphology, Adv. Cem. Res., 20 (2008), 3, 109–119 doi:10.1680/adcr.2008.20.3.109 12 L. M. Tavares, M. Cromwell Cerqueira, J. J. Less, Analysis of the grinding behaviour of various Portland cement clinkers, Materials science, 2009, https://www.zkg.de/en/artikel/zkg_Analysis_of_the_ grinding_behaviour_of_various_Portland_cement_clink- ers_106412.html 13 M. C. Martín-Sedeño, A. J. M. Cuberos, Á. G. De la Torre, G. Álvarez-Pinazo, L. M. Ordónez, M. Gateshki, M. A. G. Aranda, Alu- minum-rich belite sulfoaluminate cements: Clinkering and early age hydration, Cem. Concr. Res., 40 (2010) 3, 359-369, doi:10.1016/ j.cemconres.2009.11.003 14 B. Ma, X. Li, X. Shen, Y. Mao, H. Huang, Enhancing the addition of fly ash from thermal power plants in activated high belite sulfoaluminate cement, Constr. Build. Mater., 52 (2014), 261–266, doi:10.1016/j.conbuildmat.2013.10.099 15 F. Bullerjahn, D. Schmitt, M. Ben Haha, Effect of raw mix design and of clinkering process on the formation and mineralogical compo- sition of (ternesite) belite calcium sulphoaluminate ferrite clinker, Cem. Concr. Res., 59 (2014) 5, 87-95, doi:10.1016/j.cemconres. 2014.02.004 16 A. G. De la Torre, A. J. M. Cuberos, G. Alvarez-Pinazo, A. Cuesta, M. A. G. Aranda, In situ powder diffraction study of belite sulfo- aluminate clinkering, J. Synch. Rad., 18 (2011), 506–514, doi:10.1107/S0909049511005796 17 G. Q. Liu, Q. Yang, L. Jiang, P. Xue, X. L. Zhang, F. L. Han, Sintering characteristics of BCSAF cement clinker with added wastes from production of manganese and magnesium metals, Adv. Cem. Res., 29 (2017) 6, 227-235, doi:10.1680/jadcr.16.00035 S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... 712 Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 18 K. Quillin, Performance of belite–sulfoaluminate cements, Cem Conc Res, 31 (2001), 1341-1349, doi:10.1016/S0008-8846(01) 00543-9 19 I. A. Chen, M. C. G. Juenger, Synthesis and hydration of calcium sulfoaluminate belite cements with varied phase compositions, J. Mater. Sci., 46 (2011), 2568-2577, doi:10.1007/s10853-010-5109-9 20 P. Arjunan, M. R. Silsbee, D. M. Roy, Sulfoaluminate-belite cement from low-calcium fly ash and sulfur-rich and other industrial by-products, Cem. Concr. Res., 29 (1999) 8, 1305–131, doi:10.1016/S0008-8846(99)00072-1 21 G. Alvarez-Pinazo, I. Santacruz, L. León-Reina, M. A. G. Aranda, A. G. De la Torre, Hydration reactions and mechanical strength devel- opments of iron-rich sulfobelite eco-cements, Ind. Eng. Chem. Res., 52 (2013) 47, 16606-16614, doi:10.1021/ie402484e 22 B. Ma, X. Li, Y. Mao, X. D. Shen, Synthesis and characterization of high belitesulfoaluminate cement through rich alumina fly ash and desulfurization gypsum, Ceram. – Silik., 57 (2013) 1, 7-13 23 N. Ukrainczyk, N. Frankovi~ Mihelj, J. [ipu{i~, Calcium Sulfoaliminate Eco-Cement from Industrial Waste, Chem. Biochem. Eng.Q., 27 (2013) 1, 83–93, http://hrcak. srce.hr/99441. 24 R. Kumar, B. Bhattacharjee, Study on some factors affecting the re- sults in the use of MIP method in concrete research, Cem. Concr. Res., 33 (2003) 3, 417–424. doi:10.1016/S0008-8846(02)00974-2 25 S. Diamond, Mercury porosimetry: an in appropriate method for the measurement of pore size distributions in cement-based materials, Cem. Concr. Res., 30 (2000), 1517–1525, doi:10.1016/S0008- 8846(00)00370-7 26 L. Salvo, P. Cloetens , E. Maire, S. Zabler, J. Blandin, J.Buffie`#re, W. Ludwig, E. Boller, D. Bellet, C. Josserond, X-ray micro-tomogra- phy an attractive characterisation technique in materials science, Nucl. Instrum. Methods Phys. Res. Sect. B, 200 (2003), 273–286, doi:10.1016/S0168-583X(02)01689-0 27 Y. J. He, J. Mote, D. A. Lange, Characterization of microstructure evolution of cement paste by micro computed tomography, J. Cent. South Univ, 20 (2013) 4, 1115-1121. doi:10.1007/s11771- 013-1592-x 28 H. Tauda, T. R. Martinez-Angelesa, J. F. Parrotb, L. Hernandez- Escobedoa, Porosity estimation method by X-ray computed tomogra- phy, J. Petrol. Sci. Eng., 47 (2005) 3–4, 209-217, doi:10.1016/ j.petrol.2005.03.009 29 L. Korat, V. Ducman, A. Legat, B. Mirti~, Characterisation of the pore-forming process in lightweight aggregate based on silica sludge by means of X-ray micro-tomography (micro-CT) and mercury intru- sion porosimetry (MIP), Ceram. Int., 39 (2013) 6, 6997-7005, doi:10.1016/j.ceramint.2013.02.037 30 N. Bossa, P. Chaurand, J. Vicente, D. Borschneck, C. Levard, O. Aguerre-Chariol, J. Rose, Micro- and nano-X-ray computed-tomog- raphy: A step forward in the characterization of the pore network of a leached cement paste, Cem. Concr. Res., 67 (2015), 138-147, doi:10.1016/j.cemconres.2014.08.007 31 A. du Plessis, B. J. Olawuyi, W. P. Boshoff, S. G. le Roux, Simple and fast porosity analysis of concrete using X-ray computed tomog- raphy, Mater Struct 49, 553–562 (2016), doi:10.1617/s11527- 014-0519-9 32 L. @ibret, A. Ipavec, S. Kramar, Microstructure of belite sulfo- aluminate clinker and its influence on clinker reactivity, International Workshop on Calcium sulfoaluminate cements, Dübendorf 2018, 60 33 S. J. Gregg, K. S. W. Sing, Adsorption, surface area and porosity, 2 nd ed., Academic Press, London 1982, 303 str. 34 R. W. Adamson, A. P. Gast, Physical chemistry of surfaces, 6 th ed., Wiley, New York 1997, Chapter 17, 808 str. 35 E. P. Barrett, L. G. Joyner, P. P. Halenda, The determination of pore volume and area distributions in porous substances, I. Computations from nitrogen isotherms, J. Am. Chem. Soc., 73 (1951), 373–80, doi:10.1021/ja01151a046 36 H. E. Exner, G. Petzow, A Critical Assessment of Porosity Coarsening during Solid state Sintering, Adv. Sci. Tech., 45 (2006), 539-548, doi:10.4028/www.scientific.net/AST.45.539 37 X. Jiao, X. Wang, P. Feng et al., Microstructure Evolution and Pore Formation Mechanism of Porous TiAl3 Intermetallics via Reactive Sintering, Acta Metall. Sin. (Engl. Lett.) 31(2018), 440–448, doi:10.1007/s40195-017-0663-7 38 G. Huiling, X. Junlin, Thermodynamics and Kinetics of Calcium Sulphoaluminate, Journal of Wuhan University of Technology-Mater. Sci. Ed., 2011, 2383-2386, doi:10.1111/jace.13080 39 A. P. Savttskii, E. S. Kim, L. S. Martsunova, Compact shrinkage dur- ing liquid-phase sintering, Sov. Powder. Metall., 19 (1980), 593–596, doi:10.1007/BF00790545 40 Odler, I. (2000), Special Inorganic Cements, Taylor and Francis, London, 416 41 K. J. D. Mackenzie, R. A. Fletcher, The formation of portland ce- ment clinker under applied electric fields, I. Thermal reaction se- quence and kinetics of clinker phase formation, Thermochim. Acta, 28 (1979) 1, 161–173, doi:10.1016/0040-6031(79)87015-X 42 C. A. Bruch, Sintering kinetics for the high density alumina process, Ceram. Bull., 41 (1962), 799-806 43 T. Fang, H. Palmour, Evolution of pore morphology in sintering pow- der compacts, Ceram. Int., 16 (1990) 1, 1–10, doi:10.1016/0272- 8842(90)90056-L 44 D. Sen, T. Mahata, A. K. Patra, S. Mazumder, B. P. Sharma, Effect of Sintering Temperature on Pore Growth in ZrO2-8 mol % Y2O3 Ce- ramic Compact Prepared by Citric Acid Gel Route: A Small-Angle Neutron Scattering Investigation, J. Alloy. Compd., 364 (2004) 1, 304–10, doi:10.1016/S0925-8388(03)00613-3 45 Y. Xiong, J. F. Hu, Z. J. Shen, Dynamic Pore Coalescence in Nanoceramic Consolidated by Two-Step Sintering Procedure, J. Eur. Ceram. Soc., 33 (2013) 11, 2087–92, doi:10.1016/j.jeurceramsoc. 2013.03.015 46 A. K. Patra, S. Ramanathan, D. Sen, S. Mazumder, SANS Investiga- tion on Evolution of Pore Morphology for Varying Sintering Time in Porous Ceria, Pramana – j. Phys, 63 (2004), 327–31, doi:10.1007/ BF02704993 47 K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquérol, T. Siemieniewska, Reporting physisorption data for gas/solid systems, Pure Appl Chem (IUPAC), 57 (1985), 603–19, doi:10.1515/iupac.57.0007 48 H. Q. Ma et al., Influence of Sintering Temperature on the Microstructure and Property of Low-cost Ceramic Proppants Pre- pared by Adding Purple Sands, IOP Conf. Ser.: Mater. Sci. Eng., 230 (2017) 49 D. Veljovic, R. Jancic-Hajneman, I. Balac, B. Jokic, S. Putic, R. Petrovic, D. Janackovic, The Effect of the Shape and Size of the Pores on the Mechanical Properties of Porous Hap-Based Bio–ce- ramics, Ceram. Int., 37 (2011) 2, 471–9, doi:10.1016/j.msec. 2017.03.249 50 S. J. L. Kang, K. H. Kim, D. N. Yoon, Densification and Shrinkage During Liquid-Phase Sintering, J. Am. Ceram. Soc., 74 (1991)2 , 425–427, doi:10.1111/j.1151-2916.1991.tb06900.x S. DOLENEC et al.: EFFECTS OF THE TEMPERATURE ON THE PORE EVOLUTION DURING SINTERING ... Materiali in tehnologije / Materials and technology 54 (2020) 5, 705–713 713