Effects of mechanical and thermal activation on pozzolanic activity of kaolin containing mica

Applied Clay Science 123 (2016) 173–181

Contents lists available at ScienceDirect

Applied Clay Science journal homepage: www.elsevier.com/locate/clay

Research paper

Effects of mechanical and thermal activation on pozzolanic activity of kaolin containing mica Biljana Ilić a,⁎, Vlastimir Radonjanin b, Mirjana Malešev b, Miodrag Zdujić c, Aleksandra Mitrović a a b c

Institute for Testing of Materials, Bulevar vojvode Mišića 43, 11000, Belgrade, Serbia Faculty of Technical Sciences, University of Novi Sad, Department of Civil Engineering, Trg Dositeja Obradovića 6, 21000 Novi Sad, Serbia Institute of Technical Sciences of the Serbian Academy of Sciences and Arts, Knez Mihailova 35, 11000, Belgrade, Serbia

a r t i c l e

i n f o

Article history: Received 6 November 2015 Received in revised form 15 January 2016 Accepted 19 January 2016 Available online 4 February 2016 Keywords: Thermal activation Mechanical activation Kaolin Pozzolanic activity

a b s t r a c t Kaolin from a Serbian deposit, characterized by high content of impurities, such as mica and quartz, disordered kaolinite and high specific surface area, was subjected to thermal and mechanical activation on a large scale. Activations were carried out with a goal to investigate the effects of applied methods on the chemical, structural and morphological changes that occurred in the kaolin, and their influence on the pozzolanic activity. The changes were monitored using XRD, FTIR, TG/DTA, PSD, BET and SEM analyses. Pozzolanic activity was evaluated by determination of the 7-day compressive strength conducted on the lime mortars. The results showed that mechanical activation by milling the kaolin for 20 h in the conventional horizontal ball mill was competitive for obtaining pozzolana as thermal activation. Pozzolanic activity of thermally activated kaolin was mainly influenced by formation of metakaolin. Combined effects (such as significant specific surface area increase, as well as partial kaolinite and complete mica phase amorphization) were dominant in pozzolanic activity of mechanically activated kaolin. © 2016 Elsevier B.V. All rights reserved.

1. Introduction In recent decades, environmental considerations and energy efficiency requirements have motivated researchers to focus on finding new and optimized pozzolanic material, specially intended for use in cement-based systems (Sabir et al., 2001; Schneider et al., 2011). Among the new generation of such materials, both thermally and mechanically activated kaolin, have been a subject of many studies (Vizcayno et al., 2010; Rashad, 2013; Fabbri et al., 2013; Alujas et al., 2015; Fitos et al., 2015; Hamzaoui et al., 2015; Souri et al., 2015) due to a significant variation in the composition of kaolin, leading to the increased diversity of obtained products and their technical benefits (Siddique and Klaus, 2009). Metakaolin is produced by thermal activation (calcination) of kaolin (Sabir et al., 2001; Siddique and Klaus, 2009). Thermal activation at appropriate temperature transforms a crystalline kaolinite phase into a disordered phase (metakaolinite) through a crystal lattice failure. During the thermal activation of kaolin, two hydroxyl groups from the surface of kaolinite, join to form a water molecule leaving the remaining oxygen as a superoxide anion (Maiti and Freund, 1981; Brindley and Lemaitre, 1987; Frost and Vassallo, 1996). The instabilities caused by the anion imbalance result in crystal structure collapse (rearrangement of Si and Al atoms leading to a decrease in octahedral Al and the ⁎ Corresponding author at: Bulevar vojvode Mišića 43, 11000 Belgrade, Serbia. E-mail address: [email protected] (B. Ilić).

http://dx.doi.org/10.1016/j.clay.2016.01.029 0169-1317/© 2016 Elsevier B.V. All rights reserved.

appearance of penta- and tetra-coordinated Al (Rocha and Klinowski, 1990)), thus forming metastable, anhydrous aluminum silicates. Dehydroxylation and disordering of the kaolinite lead to a formation of reactive material with pozzolanic activity, i.e. the ability of materials to react with calcium hydroxide (CH) in presence of water and produce calcium silicate hydrate, stratlingite, tetracalcium aluminate hydrate and other compounds having binding properties (Oriol and Pera, 1995; Wild and Khatib, 1997; Cabrera and Rojas, 2001). Pozzolanic activity of the metakaolin depends on several factors, most important being the nature and abundance of clay minerals in the raw material, thermal activation conditions (He et al., 1994; Kostuch et al., 1996; Badogiannis et al., 2005) crystallinity of the original kaolinite (Kakali et al., 2001), and the average grain size of the metakaolin (Oliveira et al., 2005; Fabbri et al., 2013). During the last years, a significant number of studies (Badogiannis et al., 2005; Potgieter-Vermaak and Potgieter, 2006; Gutierrez et al., 2008; Ilić et al., 2010; Fernandez et al., 2011; Said-Mansour et al., 2011; Tironi et al., 2012; Fabbri et al., 2013; Alujas et al., 2015) have been focused on optimization of thermal activation process and the influence of the process parameters (temperature and heating time) on the pozzolanic activity of obtained metakaolin. The optimum temperature and heating time necessary to obtain material with a high pozzolanic activity varies in literature (Rashad, 2013) as a result of huge differences in the kaolin composition and structure. Although the capability of metakaolin as pozzolanic material to improve mechanical properties and durability of concrete when used as

174

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181

a partial replacement of Portland cement is well documented in literature (Sabir et al., 2001; Siddique and Klaus, 2009; Siddique and Iqbal Khan, 2011; Aiswarya et al., 2013), its utilization in building industry is quite limited, mainly due to its high price. Continuous efforts are being made to develop processes that enable the production of materials with better technical characteristics at a lower price and with a lower impact on the environment. Among them, mechanical activation attracted special attention because of the advantages it has over the traditional technological procedures (Boldyrev, 2006; Balaž, 2008). Mechanical activation of kaolin has been often studied in recent years (Horvath et al., 2003; Vizcayno et al., 2005; Vizcayno et al., 2010; Valášková et al., 2011; Dellisanti and Valdre, 2012; Mitrović and Zdujić, 2014; Hamzaoui et al., 2015). It has been found that mechanical activation of kaolin leads to structural changes (breaking the kaolinite bonds, losing of the kaolinite surface hydroxyls and replacing them with water both coordinated and adsorbed on the surface of the octahedral sheet). As a result, a new material with significantly altered surface of kaolinite, different cation-exchange capacity, porosity and water absorption capacity, reduced crystal size and a higher specific surface area is formed (Kristof et al., 1993; Frost et al., 2001; Horvath et al., 2003). Mechanically activated kaolin is different from thermally activated, due to the presence of coordinated water in amorphized kaolinite, formed as a result of the mechanical dehydroxylation of kaolinite. Using a high energy ball mill Hamzaoui et al. (2015) established that mechanically activated kaolinite has a similar X-ray diffraction pattern as thermally activated one. Most studies of mechanical activation of kaolin focused on structural and morphological changes occurring during the milling. The pozzolanic activity of mechanically activated kaolinite, particularly important for its use in cement-based systems, has been investigated only in few papers (Vizcayno et al., 2010; Mitrović and Zdujić, 2014; Fitos et al., 2015; Souri et al., 2015). Reported data show that pozzolanic activity of the mechanically activated kaolinite is a function of the mineralogical composition of the raw kaolin and the milling time (Vizcayno et al., 2010; Mitrović and Zdujić, 2014) as well as the particle size (Mitrović and Zdujić, 2014) and specific surface area (Souri et al., 2015). Comparison of thermal and mechanical activation on the large-scale milling of kaolin and its influence on the pozzolanic activity has not been reported so far. Hence, the main goal of this research is to compare structural and morphological changes that occurred in kaolin after thermal and mechanical activation and the effect they have on the pozzolanic activity of materials. In this work, kaolin from a Serbian deposit, having a significant content of impurities, such as mica and quartz, disordered kaolinite and high specific surface area, was subjected to thermal and mechanical activation on a large scale, in order to obtain materials with similar pozzolanic activity. The changes were monitored using X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), thermogravimetric/differential thermal analysis (TG/DTA), particle size distribution (PSD), specific surface area (BET) and scanning electron microscopy (SEM) analyses. Pozzolanic activity of the activated samples was evaluated by determination of the 7-day compressive strength on the lime mortars. 2. Materials and methods 2.1. Properties of starting kaolin Kaolin used in this work was collected from deposit Vrbica, Arandjelovac basin, Serbia. The kaolin (~ 150 kg) was dried at 105 °C until constant mass, milled 10 min in a ball mill, (capacity of 50 kg) and homogenized. Chemical composition was determined using X-ray fluorescence spectroscopy on an Oxford ED 2000 instrument. Reactive SiO2 content of the starting and activated kaolins was determined in accordance with SRPS EN 197-1 and SRPS EN 196-2 standards. Chemical composition

of starting kaolin is given in Table 1. The major oxides of starting kaolin were silica (SiO2) and alumina (Al2O3) in total content of 80.09% (mass percent) and the kaolin also contained a small amount of impurities (K2O, Fe2O3, and TiO2). The semi-quantitative mineralogical estimation of starting kaolin, based on the characteristic XRD peaks of each mineral, in combination with the bulk chemical analysis, showed following phase composition (in mass percent): 61% kaolinite, 16% mica, 14% quartz and 9% other phases. According to the literature (Aras et al., 2007) starting kaolin was a medium-quality raw material. 2.2. Kaolin activation procedures 2.2.1. Thermal activation Thermal activation of kaolin samples weighting 500 g was carried out in a laboratory furnace using a heating rate of 8 °C/min. Temperatures and times were chosen based on the results obtained in previous experiments with the kaolin from the same basin (Ilić, 2010; Ilić et al., 2010). Thermal activation was carried out at 700 °C and 750 °C with the activation time ranging from 30 to 180 min. After thermal activation, the calcined material was left in the furnace to slowly cool down to the room temperature. 2.2.2. Mechanical activation Mechanical activation was carried out in a conventional horizontal ball mill, under the same milling conditions as previously applied for another Serbian kaolin (Mitrović and Zdujić, 2014). A cylindrical steel vial of inner diameter 360 mm and height 340 mm (volume 0.0346 m3) filled with hardened steel balls of 20–60 mm diameter was used as the milling media. The total mass of the balls was 50.9 kg and the balls-to-powder mass ratio was about 10. The angular velocity of a vial was 4.8 s−1 (46 rpm). In each milling run, 5 kg of kaolin was milled from 30 min to 20 h. 2.3. Characterization 2.3.1. XRD The XRD data were collected on a Philips PW 1710 diffractometer using Cu-Kα graphite-monochromatized radiation (λ = 1.5418 Å) in the 2θ range 4–90° (step size: 0.02° 2θ, time per step: 0.8 s). The working conditions were 40 kV and 30 mA. 2.3.2. FTIR spectroscopy FTIR measurements were performed on a Nicolet 6700 Thermo Scientific spectrometer on the solid sample prepared using KBr presseddisk technique over the wave number range of 4000–400 cm−1. Spectral manipulation such as baseline adjustment, smoothing, and normalization was performed using the software package OMNIC. Table 1 Chemical composition of starting kaolin. Starting kaolin, mass % Loss of ignition SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O SO3 P2O5 TiO2 SUM Reactive SiO2

11.90 49.66 30.43 3.78 0.65 0.48 0.10 1.90 b0.01 0.18 0.89 99.97 19.52

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181

175

2.3.3. TG/DTA Thermal behavior was investigated from the room temperature up to 1100 °C using a SDT Q600 simultaneous TG/DTA instrument (TA Instruments) with a heating rate of 20 °C min− 1 under a dynamic (100 cm3 min−1) N2 atmosphere. 2.3.4. PSD PSD was measured by laser particle size analyzer (PSA) on a Mastersizer 2000 (Malvern Instruments Ltd., UK), which covers the particle size range of 0.02–2000 μm. 2.3.5. Specific surface area analysis Specific surface area of the samples was determined in a Micromeritics ASAP 2020 instrument. Samples were degassed at 150 °C for 10 h under reduced pressure. The specific surface area of samples was calculated according to the Brunauer–Emmett–Teller (BET) method from the linear part of the nitrogen adsorption isotherms. The total pore volume, Vtotal was given at p/p0 = 0.998. The volume of the mesopore, Vmeso was calculated according to the Barrett–Joyner–Halenda method from the desorption branch of isotherm. The volume of micropores, Vmicro, was calculated from alpha-S plot. The maximal pore diameter, Dmax, and the average pore diameter, Dsr, were also calculated. 2.3.6. SEM analysis The morphology of the particles was observed by scanning electron microscopy (SEM) using a JEOL JSM-6610LV electron microscope. The samples were previously subjected to gold deposited coating carried out on LEICA SCD005 equipment. 2.4. Pozzolanic activity test Pozzolanic activity was determined according to the Serbian standard method (SRPS B.C1.018, 2001). The mortars were prepared by mechanical mixing of hydrated lime, activated kaolin, standard sand and water in mass ratio 1:2:9:1.8. The mortars were compacted in moulds. The specimens (40 mm × 40 mm × 160 mm) were stored for 24 h ± 15 min in atmosphere with a relative humidity of 90% and temperature 20 ± 2 °C. After 24 h, moulds were placed in a thermostat at 55 ± 2 °C for the period of 5 days ± 20 h. Before testing, remolded specimens are stored in atmosphere with a relative humidity of at least 90% and temperature 20 ± 2 °C. Compressive strengths after 7 days were determined according to the standard method (SRPS EN 196-1, 2008). 3. Results and discussion 3.1. XRD XRD pattern of starting kaolin (Fig. 1) indicates the presence of kaolinite phase (JCPDS card No. 89–6538), the low temperature α-quartz (JCPDS card No. 89–8934)), as well as of some minerals from the mica group (JCPDS card No. 88–0971). In the starting kaolin, kaolinite reflections in the range 19–23° 2θ (020, 110 and 111), overlap with the reflections of quartz 100 and mica 111, which makes it impossible to determine the degree of crystallinity applying either Aparicio–Galán–Ferrell or Hinckley method (Chmielova and Weiss, 2002). Full width at half maximum of the kaolinite reflections (001 and 002) were used to estimate crystallinity of kaolinite phase, i.e. to determine FWHM (001) and FWHM (002) indexes (Tironi et al., 2014, Aparicio and Galan, 1999). Fitting of kaolinite reflections by Lorentz function, values of FWHM (001) = 0.4 and FWHM (002) = 0.5 were obtained, hence kaolinite was disordered (index values N0.4: disordered, index values b0.3: ordered kaolinite).

Fig. 1. XRD patterns of starting kaolin, mechanically and thermally activated; K — kaolinite, Q — quartz, M — mica.

Mechanical activation causes a gradual decrease in intensity and a slight broadening of basal kaolinite reflections (Fig. 1). Unlike other studies (Vizcayno et al., 2010; Mitrović and Zdujić, 2013; Mitrović and Zdujić, 2014), where the kaolinite reflections gradually reduced with milling time and finally vanished, after 20 h of milling, kaolinite reflections were still visible. Such observation suggests that complete kaolinite amorphization was not achieved. Intensity of mica M(002) reflection decreased, while the width increased, and after 20 h of milling, the reflection disappeared, which indicates complete amorphization of mica phase. The intensities of the two strongest quartz reflections (100 and 101) were slightly reduced after 6 h of milling, but were not further altered with prolonged milling, which is in accordance with previous research (Frost et al., 2002; Vizcayno et al., 2010). For the thermally activated sample, a broad hump in the range 15–35° 2θ indicates occurrence of amorphous material (Fig. 1). The absence of kaolinite reflections (001 and 002), evident in all thermally activated samples, but independent of temperature and activation time, indicates the complete transformation of kaolinite to amorphous metakaolinite. The mica reflection (002) intensity slightly decreased, with the exception of the sample thermally activated for 90 min at 700 °C (not presented in Fig. 1), which exhibits a significant decrease and broadening of the reflection and suggests partial mica amorphization. The XRD pattern indicates that the quartz phase has not been altered by thermal activation.

3.2. Thermal behavior In order to study thermal behavior of starting and activated kaolins, TG and DTA analysis were conducted and shown in Fig. 2. DTA analysis of starting kaolin shows that adsorbed and surface water are released in two stages, at about 68 °C and 132 °C. Within the temperature range of 200–350 °C mass loss is attributed to the

176

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181

is not yet finished. These observations are similar to the results reported previously (Frost et al., 2003; Horvath et al., 2003). As milling time increases, a mass loss in the range of 30–350 °C resulting from adsorbed and coordinated water also increases, while the mass loss resulting from dehydroxylation decreases, due to the reduction in the concentration of hydroxyl groups. A part of structure has already been disrupted and hydroxyl groups are released at lower temperatures (Vizcayno et al., 2005). A slight increase in the total mass loss from 12.85% for the starting kaolin to 13.06% for the sample mechanically activated for 20 h, indicates that the water molecules from the atmosphere bond to the activated surface (Frost et al., 2001). On the DTA curves of thermally activated samples (Fig. 2), characteristic endothermic peak cannot be resolved, and it can be concluded that complete kaolinite dehydroxylation was achieved by thermal activation. Endothermic heat effect at ~ 573 °C representing α- to β-quartz transformation as well as the exothermic effect at around 980 °C corresponding to the metakaolinite to spinel (Kristof et al., 1993; Cheng et al., 2012) reaction can be detected on all sample curves. A slight mass loss of about 1.5 to 2.3% for thermally activated samples can be attributed to the release of residual water. A mass loss taking place in the range 450–850 °C suggests that residual water originates mostly from hydroxyl group of mica (Gridi-Bennadji et al., 2008), which are not completely dehydroxylated, as well as from possibly remaining kaolinite phase.

3.3. FTIR FTIR spectroscopy (Fig. 3) was used to follow structural changes after mechanical and thermal activation.

Fig. 2. DTA and TG curves of starting, mechanically activated and thermally activated kaolin.

process of pre-dehydroxylation, resulting from reorganization in the octahedral sheet (DTA peak at 238 °C) (Kakali et al., 2001). Dehydroxylation of the kaolinite starts at about 350 °C, ends at about 750 °C, with the endothermic peak at 514 °C. Disordered kaolinites dehydroxylate below 530 °C and exhibit a more pronounced endothermal reaction between 80 °C and 150 °C (Emmerich, 2011). Exothermic peak at ~987 °C indicates recrystallization of metakaolinite and formation of spinel structure (Cheng et al., 2012; Chakraborty, 2014). At a shoulder of the principal endothermic peak hardly noticeable peak at ~ 573 °C, corresponds to the transformation of quartz from α- to β-form. DTA analysis of mechanically activated samples (in Fig. 2 only curve for the sample milled 20 h is presented) shows two major changes. With a milling time increase, endothermic peak, assigned to the dehydration of adsorbed water, becomes broader compared to the starting kaolin. This water originates from the released hydroxyl groups produced by mechanical dehydroxylation of kaolinite, subsequently connected to the freshly formed active surfaces (Dellisanti and Valdre, 2008). Endothermic peak corresponding to dehydroxylation of kaolinite (Horvath et al., 2003) shifts to lower temperature (from 514 °C to 498 °C for the sample before and after milling for 20 h, respectively), becomes wider and its symmetry is reduced, due to an increase of defects in octahedral structure (Dellisanti and Valdre, 2008; Vaculíková et al., 2011). Decrease of the dehydroxylation temperatures has been also attributed to the particle size reduction and the increase of structural disorder (Franco et al., 2003). The presence of dehydroxylation peak for the sample milled 20 h indicates that crystalline phase of kaolinite is still present, as well as that the mechanically induced amorphization

Fig. 3. FTIR spectra of starting, mechanically and thermally activated kaolin.

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181

FTIR spectrum of the starting kaolin shows characteristic bands of kaolinite at the 3696, 3648 and 3621 cm−1 (Van der Marel and Beutelspacher, 1976; Russell and Fraser, 1994; Saikia et al., 2003). Bands at 3696 and 3648 cm−1 are assigned to OH stretching modes of two outer surface hydroxyl groups (OuOH), while the band at 3621 cm−1 is assigned to the stretching mode of an inner hydroxyl group (InOH) (Frost et al., 2001; Vaculíková et al., 2011; Dellisanti and Valdre, 2012; Hamzaoui et al., 2015). Weak intensity bands at 3474 cm−1 and 1646 cm−1 result from OH stretching and HOH bending vibrations of the adsorbed water are detected. In the Si–O region stretching vibrations bands at 1105 cm− 1 (Si bonds with apical oxygen in the same plane with OH), 1033 cm−1 and 1009 cm−1 (Si with the O in the plane) are noticed. The presence of quartz is confirmed by bands at 793 cm− 1, 754 cm−1 and 695 cm−1. Bands associated with two types of the Si–O bond include Si with basal oxygen (793 cm−1) and Si with apical oxygen (754 cm−1 and 695 cm−1). In a region corresponding to the Al–OH bending vibration (from 908 to 940 cm−1) the band at 914 cm−1 is assigned to bending of the inner hydroxyl groups (InOH) linked to Al. The band at 470 cm−1 is assigned to Si–O–Si bending deformation, band at 538 cm−1 represents Si–O– AlVI bending vibration, while the 431 cm− 1 band results from Si–O bending vibrations. Degree of crystallinity of the kaolinite may be estimated based on the relative intensity of bands in the region of OH stretching and bending vibrations. According to classification (Vaculíková et al., 2011), it can be concluded that the starting kaolin is poorly ordered, because only one band at 3648 cm−1 could be identified. Additionally, P0 and P2 indexes, where P0 is the ratio between the intensities of the OH stretching bands at 3620 cm−1 and 3700 cm−1 and P2 is the ratio between the intensities of bands at 3670 cm−1 and 3650 cm−1, were also used to estimate crystallinity of kaolinite phase (Bich et al., 2009). The two-point baseline method was used to obtain the intensity of the OH bands. According to the obtained values of indexes, P0 = 0.8 and P2 = 0 (because the band at 3670 cm− 1 was not observed), kaolinite was classified as disordered structures (Ambroise et al., 1992; Russel, 1987). These results confirm XRD and DTA results concerning the structure order of kaolinite. On the FTIR spectra of mechanically activated samples milled for 6 h and 10 h (not presented in Fig. 3), band at 3648 cm− 1 gradually decrease and completely vanishes after 20 h of milling, while the intensities of OH bands at 3696 cm− 1, 3621 cm− 1 and 914 cm− 1 decrease. Milling destroys hydrogen bonds between the kaolinite layers (Vizcayno et al., 2005), which causes a decrease in the Si–O and Al–OH bond strengths and the release of inner surface hydroxyl groups. The existence of bands arising from OH stretching and bending vibrations is the evidence that after 20 h of milling, mechanical dehydroxylation is not completed and that kaolinite phase is still present, which is in accordance with the results of thermal and XRD analyses. As the milling continues, band at 3444 cm−1, assigned to HOH bending vibration becomes considerably wider and the intensity of the band at 1646 cm−1, assigned to the water connected to the surface, formed by the reaction of the released hydroxyl groups, increases (Mako et al., 2001; Vizcayno et al., 2010; Valášková et al., 2011; Dellisanti and Valdre, 2012). These water molecules hydrate the activated surface of kaolinite (Frost et al., 2001) as confirmed by the results of thermal analysis. Mechanical activation leads to the breaking of Si–O bond as may be concluded by disappearance of Si–O stretching vibration at 1105 cm− 1, as well as by the decrease in intensity of other bands in the Si–O region at 1033 cm− 1, 1009 cm− 1 and bands at 793 cm− 1, 754 cm− 1 and 694 cm− 1. The band intensity at 539 cm− 1 assigned to Si–O–AlVI bending vibration also decreases. Its existence suggests that the octahedral structure of Al layer in the mechanically activated sample is preserved.

177

After thermal activation of kaolin the following changes in the FTIR spectra (Fig. 3) can be determined: • The absence of characteristic bands of kaolinite at 3696 cm− 1, 3648 cm−1, 3621 cm−1 and 914 cm−1, implying completion of the dehydroxylation process. • The appearance of a wide band at ~3446 cm−1, arising from the HOH deformation vibration, assigned to the adsorbed water. • Bands at 1105, 1033, and 1009 cm−1 disappeared and a new band at ~1069 cm−1, assigned to amorphous silica, appeared (Chakraborty, 2014). • Bands at 793 cm−1, 754 cm−1 and 695 cm−1 disappeared and a new wide band at ~ 800 cm−1 corresponding to Al4+–O vibrations was occurred. Therefore, the coordination number of Al changed from six in kaolinite to four and five in metakaolinite. Also, the disappearance of the band at 538 cm−1 suggests that the octahedral structure of Al became tetrahedral and pentahedral in metakaolinite (Chakraborty, 2014). • The new band at ~550 cm−1 assigned to free amorphous aluminum oxide vibrations (Chakraborty, 2014) appeared.

3.4. Particle size distribution Particle size distribution of the starting kaolin, and two characteristic samples, namely, milled for 20 h, and sample thermally activated for 60 min at 700 °C, are shown in Fig. 4. Particle size distribution of starting kaolin shows a bimodal distribution; one fraction is in the size range of 0.2–130 μm and the other in the range of 130–382 μm, with two maxima at about 6 μm and 207 μm, respectively, and the volume mean diameter D [4,3] of 32 μm. Mechanical activation leads to a significant reduction of particle size, hence, volume

Fig. 4. Particle size distribution of starting kaolin, mechanically activated for 20 h and thermally activated for 60 min at 700 °C.

178

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181

Table 2 Specific surface area, Sp, and porosity Vtotal Vmeso Vmicro Dsr Dmax of starting, mechanically (MA) and thermally activated (TA) kaolin. Sample

Sp, m2/g

Vtotal, cm3/g

Vmeso, cm3/g

Vmicro, cm3/g

Dsr, nm

Dmax, nm

Starting kaolin MA 6 h MA 10 h MA 20 h TA 700 °C 30 min TA 700 °C 60 min TA 700 °C 90 min TA 700 °C 120 min TA 750 °C 30 min

32.5 38.4 38.0 49.8 25.8 26.6 25.4 25.4 22.4

0.115 0.109 0.118 0.145 0.110 0.119 0.104 0.115 0.112

0.113 0.106 0.116 0.142 0.108 0.118 0.102 0.114 0.110

0.010 0.012 0.014 0.016 0.008 0.008 0.008 0.008 0.008

13.5 10.1 13.2 10.8 13.8 14.2 15.0 14.9 16.4

3.7 3.8 4.0 3.8 3.7 3.7 3.8 3.8 3.6

fraction between 6 and 120 μm, becomes much smaller. The mean particle size (d50) of the starting kaolin is reduced from 8 to 6 μm after 6 h of milling (not presented in Fig. 4), and with prolonged milling, further changes are insignificant. Also, this prolonged milling causes gradual agglomeration of the particles and an increase in particle size fractions of coarse powder in comparison to the starting kaolin, which is consistent with reported results (Sánchez-Soto et al., 2000; Mako et al., 2001). Similar trend was previously observed for another Serbian kaolin milled under the same conditions (Mitrović and Zdujić, 2014). Particle size distribution of thermally activated sample is shifted to coarser particles, thus indicating that thermal activation caused partial aggregation of smallest particles. Mean particle size (d50) of thermally activated samples increased to a maximum of 20 μm for the sample thermally activated for 90 min at 700 °C (not presented). The lowest value of d50 = 16 μm is observed for the sample thermally activated for 60 min at 700 °C. These results are consistent with findings (Fabbri et al., 2013) that thermal activation causes the process of particle aggregation and an increase in mean particle size (d50). 3.5. Specific surface area and porosity Specific surface area, Sp, of starting kaolin of 32.5 m2/g is very high, compared to the kaolin used in our previous research, 2 m2/g (Ilić, 2010) and 14.19 m2/g (Mitrović and Zdujić, 2014) and the kaolin used for the production of commercial metakaolin – 4.13 m2/g (Kovárík et al., 2015). Mechanical activation reduces the particle size and increases the specific surface area. Throughout the milling, highly porous particles with very small pores are formed, so that the specific surface area, corresponding to the reactive surface, becomes very high. As can be seen, specific surface area of the sample milled for 6 h increased 18% compared to the starting kaolin, but remained almost the same as milling continued to 10 h. However, when the milling was prolonged to 20 h, specific surface area increased an additional 31%. This is in contrast with already reported findings, where in the first period of milling reduction of kaolin particles and an increase in surface area occurred, while prolonged milling caused stacking of the particles, and the decrease in surface area (Sánchez-Soto et al., 2000; Mako et al., 2001; Perrin-Sarazin et al., 2009; Mitrović and Zdujić, 2014). Changes in the particle morphology and size during the mechanical activation directly cause an increase in the mesopore volume (Table 2). On the other hand, the effect of mechanical activation leads to a distortion of the crystal structure of octahedral layers allowing an increase in the volume of micropores, and therefore an increase in total porosity (Cristóbal et al., 2009). Since the thermal activation promoted to some extent particle coarsening, decrease in specific surface area could be expected. However, disagreement with reported results can be found in literature. Several studies have reported gradual decrease in specific surface area with thermal activation (Cristóbal et al., 2009; Bich et al., 2009), while others reported an increase (Varga and Trník, 2006; Štubna et al., 2006). Our findings indicate that in the sample thermally activated for 30 min at 700 °C, specific surface area decreased approximately 20% compared

Fig. 5. SEM microphotographs of the (a) starting kaolin (b) mechanically activated for 20 h and (c) thermally activated for 60 min at 700 °C.

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181

to the starting kaolin, while no significant changes occurred with the longer heating time. As the temperature increased to 750 °C, specific surface area further decreased 32% in comparison to the starting kaolin. The reason for this decrease is likely due to the particle aggregation and sintering (Fabbri et al., 2013). It is possible that the applied degassing (at 150 °C for 10 h under reduced pressure) can change the specific surface area and porosity of studied samples. 3.6. Morphology The morphology of the powder particles of starting kaolin, mechanically activated for 20 h and thermally activated for 60 min at 700 °C is presented in Fig. 5. The particles of the starting kaolin have a platelike pseudohexagonal shape of the original kaolinite. Mechanical activation changes the plate-like morphology of the kaolinite to round-edged particles with a ragged surface, accompanied with a decrease in particle size. The lamellar structure of kaolinite is still present after 20 h of milling. The morphology of the starting kaolin is partially preserved after thermal activation as particles in the form of broken plates, but agglomeration of particles is more apparent. These features are similar to those found in earlier reports (Vizcayno et al., 2010). 3.7. Pozzolanic activity Pozzolanic activity of mechanically and thermally activated kaolin is given in Fig. 6. As can be seen in Fig. 6a, pozzolanic activity continuously increases with milling time, from 4.6 MPa to 11 MPa, in the first 10 h of milling. Continued milling causes additional pozzolanic activity increase to 13.7 MPa for kaolin milled 20 h. However, structural changes, detected by XRD and FTIR analyses, were also quite different when the milling period was restricted to 10 h compared to those obtained when milling was prolonged to 20 h. While partial kaolinite and mica amorphization is achieved in the first 10 h, additional milling to 20 h only caused complete amorphization of the mica phase. Also, the specific surface area in the first 10 h of milling increased 17% compared to the starting kaolin,

179

while milling prolonged to 20 h led to an additional increase of 31%. These findings reveal that is possible to obtain high pozzolanic activity by mechanical activation of kaolin with high content of mica, in spite that kaolinite is only partially amorphized (about 51%, as determined by an increase in reactive silica content after 20 h of milling) and dehydroxylated. It is generally accepted that the key factor in producing pozzolan is to achieve as near to complete dehydroxylation (Ramezanianpour, 2014). However, in this case the partial kaolinite amorphization and dehydroxylation is compensated by a significant increase in specific surface area and complete mica phase amorphization. As there are limited data regarding pozzolanic activity of the mechanically activated kaolins and also of the influence of mineralogical composition on pozzolanic activity, findings from earlier study (Mitrović and Zdujić, 2014) were compared with present results (Fig. 6a). In the previous study, mechanical activation was performed in the same milling equipment, under the same conditions, on the kaolin having significantly different mineralogical composition, namely with a higher content of impurities (40% of quartz) and lower content of kaolinite. As can be seen in Fig. 6a, pozzolanic activity in both cases increases with milling time, whereas corresponding values of pozzolanic activity are quite close regardless of the starting kaolin composition. In contrast to the present study, where partial kaolinite amorphization was achieved after 20 h of milling, the previous study revealed almost complete kaolinite amorphization. The present results clearly show that increased pozzolanic activity of the mechanically activated kaolin with a high content of mica, is a consequence of both partial kaolinite and complete mica phase amorphization and the significant specific surface area increase. In the previous study, the presence of quartz, which acted as an additional milling media, facilitated the amorphization of kaolinite, and thus contributed to an increase in pozzolanic activity. The same effect of quartz was reported by other authors (Benezet and Benhassaine, 1999; Mako et al., 2001; Vizcayno et al., 2010). Both studies indicate that is possible to obtain material of high pozzolanic activity, even in the low-energy ball mill. The influence of temperature and time on pozzolanic activity during thermal activation is given in Fig. 6b. The highest pozzolanic activity is

Fig. 6. Pozzolanic activity as a function of (a) milling time and (b) temperature and heating time. Pozzolanic activity of another Serbian kaolin milled under same conditions is also presented for comparison (Mitrović and Zdujić, 2014).

180

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181

achieved for the sample activated for 60 min at 700 °C, and as the heating time extended, the pozzolanic activity decreased. As the temperature increased to 750 °C, the highest pozzolanic activity was obtained after 30 min, while longer heating times also led to a significant decrease in pozzolanic activity. As XRD, TGA and FTIR results showed, the complete kaolinite amorphization is achieved for all investigated temperatures and times, so it could be assumed that a decrease in pozzolanic activity is the consequence of particle agglomeration or sintering (Fabbri et al., 2013). The results indicate that the complete amorphization of kaolinite phase is the main factor affecting pozzolanic activity of thermally activated samples. Complete amorphization of kaolinite phase is also confirmed by increase in the reactive silica content by 97% (for the sample 700 °C, 60 min). The summary of structural and morphological changes achieved by mechanical and thermal activation is given in Table 3.

4. Conclusions Mechanical and thermal activation applied on the same kaolin caused different chemical, structural and morphological changes. In spite of significant differences in the structure and physico-chemical properties accomplished by these two different activation methods, pozzolanic activity of kaolin mechanically activated for 20 h is competitive to the kaolin thermally activated. Therefore, the mechanical activation of kaolin in the conventional horizontal ball mill is equally suitable for obtaining pozzolana as thermal activation, despite differences in the level of amorphization and dehydroxylation of kaolinite obtained by these two methods. As shown by XRD, DTA, FTIR and reactive silica content, the pozzolanic activity of thermally activated kaolin is mainly affected by kaolinite amorphization, i.e. formation of metakaolin. Combined effects of partial kaolinite and complete mica amorphization, followed by a significant increase in the specific surface area, were the main cause of the increase in pozzolanic activity of mechanically activated kaolin. In addition, results of mechanical activation, from our present and the previous study, performed on a large scale in conventional horizontal ball mill, suggested that the pozzolanic activity was also affected by the presence of the impurities in the starting kaolin, and that different impurities, such as quartz and mica, contributed in different ways to the pozzolanic activity. While mica contributed through its complete amorphization, quartz acted as an additional milling media leading to kaolinite amorphization.

Table 3 Summary of results achieved by mechanical and thermal activation of kaolin. Methods

Mechanical activation (milled 20 h)

Thermal activation (700 °C, 60 min)

XRD

Partial amorphization of kaolinite and complete amorphization of mica Decreasing of characteristic kaolinite bands intensity and increasing of bands assigned to the adsorbed/coordinated water on the surface Partial dehydroxylation of kaolinite 49.8 m2/g (increased 53%) 6 μm (decreased 31%) 29.52% (increased 51%)

Complete amorphization of kaolinite

13.7 MPa (increased 298%)

15.8 MPa (increased 343%)

FTIR

DTA Specific surface area Particle size, d50 Reactive silica content Pozzolanic activity

The absence of characteristic kaolinite bands and appearance of new bands corresponding to metakaolin Complete dehydroxylation of kaolinite 26.6 m2/g (decreased 18%) 16 μm (increased 98%) 38.40% (increased 97%)

Acknowledgments This work was financially supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia (Grants Nos. TR 36017 and 45001). References Aiswarya, S., Arulraj, G.P., Dilip, C., 2013. A review on use of metakaolin in concrete. Eng. Sci. Tech. Int. J. 3, 592–597. Alujas, A., Fernández, R., Quintan, R., Scrivener, K.L., Martirena, F., 2015. Pozzolanic reactivity of lowgrade kaolinitic clays: influence of calcination temperature and impact of calcination products on OPC hydration. Appl. Clay Sci. 108, 94–101. Ambroise, J., Martin-Calle, S., Péra, J., 1992. Pozzolanic behaviour of thermally activated kaolin. Proc. 4th Int. Conf. Fly Ash, Silica Fume, Slag and Naturals Pozzolans in Concrete 1. ACI, Detroit, USA, pp. 731–748. Aparicio, P., Galan, E., 1999. Mineralogical interference on kaolinite crystallinity index measurements. Clay Clay Miner. 47, 12–27. Aras, A., Albayrak, M., Arikan, M., Sobolev, K., 2007. Evaluation of selected kaolins as a raw material for the Turkish cement and concrete industry. Clay Miner. 42, 233–244. Badogiannis, E., Kakali, G., Tsivilis, S., 2005. Metakaolin as supplementary cementitious material — optimization of kaolin to metakaolin conversion. J. Therm. Anal. Calorim. 81, 457–462. Balaž, P., 2008. Mechanochemistry in Nanoscience and Minerals Engineering. Springer Verlag, Berlin. Benezet, J.C., Benhassaine, A., 1999. The influence of particle size on the pozzolanic reactivity of quartz powder. Powder Technol. 103, 26–29. Bich, C., Ambroise, J., Péra, J., 2009. Influence of degree of dehydroxylation on the pozzolanic activity of metakaolin. Appl. Clay Sci. 44, 194–200. Boldyrev, V.V., 2006. Mechanochemistry and mechanical activation of solids. Russ. Chem. Rev. 75, 177–189. Brindley, G.W., Lemaitre, J., 1987. Thermal oxidation and reduction of clay minerals. In: Newman, A.C.D. (Ed.), Chemistry of Clay and Clay Minerals. Longman, Essex, pp. 319–370. Cabrera, J., Rojas, M.F., 2001. Mechanism of hydration of the metakaolin-lime-water system. Cem. Concr. Res. 31, 177–182. Chakraborty, A.K., 2014. Phase Transformation of Kaolinite Clay Springer India. Cheng, H., Liu, Q., Yang, J., Ma, S., Frost, R.L., 2012. The thermal behavior of kaolinite intercalation complexes — a review. Thermochim. Acta 545, 1–13. Chmielova, M., Weiss, Z., 2002. Determination of structural disorder degree using an XRD profile fitting procedure. Application to Czech kaolins. Appl. Clay Sci. 22, 65–74. Cristóbal, A.G.S., Castelló, R., Luengo, M.A.M., Vizcayno, C., 2009. Acid activation of mechanically and thermally modified kaolins. Mater. Res. Bull. 44, 2103–2111. Dellisanti, F., Valdre, G., 2008. Linear relationship between thermo-dehydroxylation and induced-strain by mechanical processing in vacuum: the case of industrial kaolinite, talc and montmorillonite. Int. J. Miner. Process. 88, 94–99. Dellisanti, F., Valdre, G., 2012. The role of microstrain on the thermostructural behaviour of industrial kaolin deformed by ball milling at low mechanical load. Int. J. Miner. Process. 102/103, 69–77. Emmerich, K., 2010. Thermal analysis in the characterization and processing of industrial minerals. In: Christidis, G. (Ed.), Advances in the Characterization of Industrial Minerals. The Mineralogical Society of Great Britain and Ireland, Twickenham, pp. 129–170. Fabbri, B., Gualtieri, S., Leonardi, C., 2013. Modifications induced by the thermal treatment of kaolin and determination of reactivity of metakaolin. Appl. Clay Sci. 73, 2–10. Fernandez, R., Martirena, F., Scrivener, K.L., 2011. The origin of the pozzolanic activity of calcined clay minerals: a comparison between kaolinite, illite and montmorillonite. Cem. Concr. Res. 41, 113–122. Fitos, M., Badogiannis, E., Tsivilis, S., Perraki, M., 2015. Pozzolanic activity of thermally and mechanically treated kaolins of hydrothermal origin. Appl. Clay Sci. 116/117, 182–192. Franco, F., Pérez Maqueda, L.A., Pérez-Rodríguez, J.L., 2003. The influence of ultrasound on the thermal behaviour of well ordered kaolinite. Thermochim. Acta 404, 71–79. Frost, R.L., Vassallo, A.M., 1996. The dehydroxylation of the kaolinite clay minerals using infrared emission spectroscopy. Clay Clay Miner. 44, 635–651. Frost, R.L., Horvath, E., Mako, E., Kristof, J., Tamás, C., 2003. The effect of mechanical activation upon the intercalation of a high-defect kaolinite with formamide. J. Colloid Interface Sci. 265, 386–395. Frost, R.L., Mako, E., Kristof, J., Horvath, E., Kloprogge, J.T., 2001. Mechanical treatment of kaolinite. J. Colloid Interface Sci. 239, 458–466. Frost, R.L., Mako, E., Kristof, J., Kloprogge, J.T., 2002. Modification of kaolinite surfaces through mechanical treatment — a mid-IR and near-IR spectroscopic study. Spectrochim. Acta A 58, 2849–2859. Gridi-Bennadji, F., Beneu, B., Laval, J.P., Blanchart, P., 2008. Structural transformations of muscovite at high temperature by X-ray and neutron diffraction. Appl. Clay Sci. 38, 259–267. Gutierrez, M.R., Torres, J., Vizcayno, C., Castelló, R., 2008. Influence of the calcination temperature of kaolin on the mechanical properties of mortars and concretes containing metakaolin. Clay Miner. 43, 177–183. http://dx.doi.org/10.1180/claymin.2008.043.2.02. Hamzaoui, R., Muslim, F., Guessasma, S., Bennabi, A., Guillin, J., 2015. Structural and thermal behavior of proclay kaolinite using high energy ball milling process. Powder Technol. 271, 228–237. He, C., Makovicky, E., Osbaeck, B., 1994. Thermal stability and pozzolanic activity of calcined kaolin. Appl. Clay Sci. 9, 165–187.

B. Ilić et al. / Applied Clay Science 123 (2016) 173–181 Horvath, E., Frost, R.L., Mako, E., Kristof, J., Cseh, T., 2003. Thermal treatment of mechanically activated kaolinite. Thermochim. Acta 404, 227–234. Ilić, B., 2010. The Study of Metakaolin Synthesis Process and the Research of Characteristics of Composite Portland-Cement – Metakaolin (MSc thesis) University of Belgrade (In Serbian). Ilić, B., Mitrović, A., Miličić, L., 2010. Thermal treatment of kaolin to obtain metakaolin. Hem. Ind. 64, 351–356. Kakali, G., Perraki, T., Tsivilis, S., Badogiannis, E., 2001. Thermal treatment of kaolin: the effect of mineralogy on the pozzolanic activity. Appl. Clay Sci. 20, 73–80. Kostuch, J.A., Walters, V., Jones, T.R., 1996. High performance concretes incorporating metakaolin: a review. In: Dhir, R.K., Jones, M.R. (Eds.), Concrete 2000: Economic and Durable Construction through Excellence. E&FN Spon, London, pp. 1799–1811. Kovárík, T., Belsky, B., Novotny, P., Riha, J., Savková, J., Medlín, R., et al., 2015. Structural and physical changes of re-calcined metakaolin regarding its reactivity. Constr. Build. Mater. 80, 98–104. Kristof, E., Juhasz, A.Z., Vassanyi, I., 1993. The effect of mechanical treatment on the crystal structure and thermal behavior of kaolinite. Clay Clay Miner. 41, 608–612. Maiti, G.C., Freund, F., 1981. Dehydration-related proton conductivity in kaolinite. Clay Miner. 16, 395–413. Mako, E., Frost, R.L., Kristof, J., Horvath, E., 2001. The effect of quartz content on the mechanical activation of kaolinite. J. Colloid Interface Sci. 244, 359–364. Mitrović, A., Zdujić, M., 2013. Mechanical treatment of Serbian kaolin to obtain high reactive pozzolana. J. Serb. Chem. Soc. 78, 579–590. Mitrović, A., Zdujić, M., 2014. Preparation of pozzolanic addition by mechanical treatment of kaolin. Int. J. Min. Process. 132, 59–66. Oliveira, L.A.P., Jalali, S., Fernandes, J.M., Torres, E., 2005. The use of metakaolin to produce eco-efficient concrete. Mater. Struct. 38, 403–410. Oriol, M., Pera, J., 1995. Pozzolanic activity of metakaolin under microwave treatment. Cem. Concr. Res. 25, 265–270. Perrin-Sarazin, F., Sepehr, M., Bouaricha, S., Denault, J., 2009. Potential of ball milling to improve clay dispersion in nanocomposites. Polym. Eng. Sci. 49, 651–665. Potgieter-Vermaak, S.S., Potgieter, J.H., 2006. Metakaolin as an extender in south African cement. J. Mater. Civ. Eng. 18, 619–623. Ramezanianpour, A.A., 2014. Cement Replacement Materials. Properties, Durability, Sustainability. Springer, Berlin Heidelberg, pp. 225–255. Rashad, M.A., 2013. Metakaolin as cementitious material: history, scours, production and composition — a comprehensive overview. Constr. Build. Mater. 41, 303–318. Rocha, J., Klinowski, J., 1990. 29Si and 27Al magic-angle-spinning NMR studies of the thermal transformation of kaolinite. Phys. Chem. Miner. 17, 179–186. Russel, J.D., 1987. Infrared Spectroscopy of Inorganic Compounds. Laboratory Methods in Infrared Spectroscopy. Wiley, New York. Russell, J.D., Fraser, A.R., 1994. In: Wilson, M.J. (Ed.), Clay Mineralogy: Spectroscopic and Chemical Determinative Methods. Chapman & Hall, London, UK, pp. 11–67.

181

Sabir, B.B., Wild, S., Bai, J., 2001. Metakaolin and calcined clays as pozzolans for concrete: a review. Cem. Concr. Compos. 23, 441–454. Said-Mansour, M., Kadri, E.H., Kenai, S., Ghrici, M., Bennace, R., 2011. Influence of calcined kaolin on mortar properties. Constr. Build. Mater. 25, 2275–2282. Saikia, N.J., Bharali, D.J., Sengupta, P., Bordoloi, D., Goswamee, R.L., Saikia, P.C., Borthakur, P.C., 2003. Characterization, benefication and utilization of a kaolinite clay from Assam India. Appl. Clay Sci. 24, 93–103. Sánchez-Soto, P.J., Jiménez de Haro, M.C., Pérez-Maqueda, L.A., Varona, I., PérezRodríguez, J.L., 2000. Effects of dry grinding on the structural changes of kaolinite powders. J. Am. Ceram. Soc. 83, 1649–1657. Schneider, M., Romer, M., Tschudin, M., Bolio, H., 2011. Sustainable cement production—present and future. Cem. Concr. Res. 41, 642–650. Siddique, R., Iqbal Khan, M., 2011. Supplementary Cementing Materials. Springer-Verlag, Berlin Heidelberg, pp. 175–230. Siddique, R., Klaus, J., 2009. Influence of metakaolin on the properties of mortar and concrete: a review. Appl. Clay Sci. 43, 392–400. Souri, A., Kazemi-Kamyab, H., Snellings, R., Naghizadeh, R., Golestani-Fard, F., Scrivener, K., 2015. Pozzolanic activity of mechanically and thermally activated kaolins in cement. Cem. Concr. Res. 77, 47–59. Štubna, I., Varga, G., Trnik, A., 2006. Investigation of kaolinite dehydroxylation is still interesting. Építoanyag 58, 6–9. Tironi, A., Trezza, M.A., Scian, A.N., Irassar, E.F., 2012. Kaolinitic calcined clays: factors affecting its performance as pozzolans. Constr. Build. Mater. 28, 276–281. Tironi, A., Trezza, M.A., Scian, A.N., Irassar, E.F., 2014. Potential use of Argentine kaolinitic clays as pozzolanic material. Appl. Clay Sci. 101, 468–476. Vaculíková, L., Plevová, E., Vallová, S., Koutník, I., 2011. Characterization and differentiation of kaolinites from selected Czech deposits using infrared spectroscopy and differential thermal analysis. Acta Geodyn. Geomater. 8, 59–67. Valášková, M., Barabaszová, K., Hundáková, M., Ritz, M., Plevová, E., 2011. Effects of brief milling and acid treatment on two ordered and disordered kaolinite structures. Appl. Clay Sci. 54, 70–76. Van der Marel, H.W., Beutelspacher, H., 1976. Atlas of Infrared Spectroscopy of Clay Minerals and Their Admixtures. Elsevier, Amsterdam. Varga, G., Trník, A., 2006. Struktúra kaolinitu a metakaolinitu. Silikátnik 3, 11–19. Vizcayno, C., Castello, R., Ranz, I., Calvo, B., 2005. Some physico-chemical alterations caused by mechanical treatments in kaolinites of different structural order. Thermochim. Acta 428, 173–183. Vizcayno, C., Gutiérrez, R.M., Castello, R., Rodriguez, E., Guerrero, C.E., 2010. Pozzolan obtained by mechanical and thermal treatments of kaolin. Appl. Clay Sci. 49, 405–413. Wild, S., Khatib, J.M., 1997. Portlantide consumption in metakaolin cement pastes and mortars. Cem. Concr. Res. 27, 137–146.