Applied Clay Science 149 (2017) 8–12 Contents lists available at ScienceDirect Applied Clay Science journal homepage: www.elsevier.com/locate/clay Research paper Comparison of dehydration in kaolin and illite using DC conductivity measurements MARK Marian Kublihaa, Viera Trnovcováb, Ján Ondruškab,⁎, Igor Štubňab, Ondrej Bošáka, Tiit Kaljuveec a b c Institute of Materials, Faculty of Materials Science and Technology, Slovak University of Technology, J. Bottu 25, 917 24 Trnava, Slovakia Department of Physics, Faculty of Natural Sciences, Constantine the Philosopher University in Nitra, A. Hlinku 1, 949 74 Nitra, Slovakia Laboratory of Inorganic Materials, Faculty of Chemical and Materials Science, Tallinn Technical University, Ehitajate 5, 190 86 Tallinn, Estonia A R T I C L E I N F O A B S T R A C T Keywords: Illite and kaolinite DC conductivity DTG DSC EGA Dehydration Temperature dependencies of the DC conductivity of kaolin and illite are measured from 20 °C up to 450 °C using “as received” samples and samples after heating to 400 °C and 450 °C. Release of physically bound water (PBW) from green kaolin shows two maxima, at 55 °C and 298 °C. Release of the PBW from green illite takes place in 3 steps at 72 °C, 186 °C, and 298 °C. Up to 200 °C, the dominant charge carriers are H+ and OH– ions in both materials. At higher temperatures, alkali ions (both Na+ and K+ in kaolin, prevalently K+ in illite) are the dominant charge carriers with the conduction activation energy of 1.19 ± 0.02 eV in kaolin and of 1.12 ± 0.02 eV in illite. Above 400 °C, dehydroxylation runs in illite. Due to this process, the DC conductivity of illite increases; in partly dehydroxylated illite, higher conduction activation energy (1.22 eV) indicates that alkali metal ions are moving in a more disordered structure. In kaolin, dehydroxylation starts only above 450 °C. 1. Introduction Kaolinite and illite are common plastic components in traditional ceramics. They determine various processes which occur during heating of green ceramic bodies (Ferrari and Gualtieri, 2006). The structure of kaolinite consists of a repetition of tetrahedron-octahedron (T-O) sheets in one layer. The layers are tightly laid on top of each other, and no space for alkali metal cations or H2O molecules exists between them (Grim, 1962) (Fig. 1a). The illite structure consists of a repetition of tetrahedron-octahedron-tetrahedron (T-O-T) sheets in one layer. The interlayer space is occupied by K+ ions (Grim, 1962) and likely also with water molecules (Carroll et al., 2005; Drits and McCarty, 2007; Ferrari et al., 2006; Gualtieri et al., 2008; Viczian, 1997) (Fig. 1b). During heating up to 300 °C, kaolin and illitic clay lose physically bound water. Above 450 °C, dehydroxylation starts and kaolinite and illite lose chemically bound water. An addition of illite to kaolin increases the content of the glassy phase due to a fluxing action of K2O and decreasing the firing temperature (Ferrari and Gualtieri, 2006; Wattanasiriwech and Wattanasiriwech, 2011). Many methods of thermal analyses are used for a study of the processes that run in kaolin and illite during their heating. Measurement of temperature dependencies of their electrical conductivity (so called thermoconductometry) is a convenient method to detect the release of both physically (PBW) and chemically (ChBW) bound water (Kriaa ⁎ et al., 2014). Electrical properties of kaolin (MacKenzie, 1969; Maiti and Freund, 1981; Ondruška et al., 2015b; Podoba et al., 2014; Štubňa and Kozík, 1996; Trnovcová et al., 2012) and illite (Kriaa et al., 2014; Kubliha et al., 2016; Ondruška et al., 2015a, 2015b) have been experimentally studied using various experimental arrangements, making any comparisons of these materials difficult. In this paper, we use identical measurement arrangements for both materials to improve the existing comparisons of their dehydration processes. Illitic clay and kaolin were chosen because of their abundant use in ceramics due to their favorable influence on firing process (Ferrari and Gualtieri, 2006; Wattanasiriwech and Wattanasiriwech, 2011). Thermal analyses (e.g. DTA, TGA, DSC, TDA) do not allow to observe motion of the electrical charge carriers. Therefore, the electrical methods can be useful supplements to traditional thermal analyses. The factors that influence the ionic DC conductivity in solids are the concentration of charge carriers, temperature, the availability of vacantaccessible sites which is controlled by the density of defects and/or disordering of the structure, and the facility with which an ion can jump into a neighboring site. This facility is controlled by the activation energy which indicates the free energy barrier an ion must overcome for a successful jump between the sites. When the charge of the ions is large they find high conduction activation energy (CAE) barrier for their jumps between sites. Therefore, monovalent ions of small or medium diameters (e.g. Na+ or K+) when they are present in Corresponding author. E-mail address: jondruska@ukf.sk (J. Ondruška). http://dx.doi.org/10.1016/j.clay.2017.08.012 Received 18 January 2017; Received in revised form 28 June 2017; Accepted 12 August 2017 Available online 07 September 2017 0169-1317/ © 2017 Published by Elsevier B.V. Applied Clay Science 149 (2017) 8–12 M. Kubliha et al. Fig. 1. Structure of a) kaolinite and b) illite. Modified from Grim (1962). with a diameter of 11 mm were obtained from this plastic mass using a laboratory extruder. After a 5-day free drying, samples for thermal analyses (DTG, EGA, DSC) and electrical measurements were cut. The samples for measurements of the DC conductivity had dimensions of ∅11 × 2 mm. They were coated with conductive graphite layers on both contact surfaces (so called “sandwich” arrangement). For contact surfaces of electrodes, Pt foils were used. The electrical current was measured with picoammeter Keithley 6517B at a constant voltage of 10 V, in N2, using Novocontrol Concept 90, from 20 °C up to 400 °C or 450 °C. The temperature was measured with sensitivity of ± 1 °C using a Pt-PtRh10 thermocouple. The DC conductivity of green samples was measured repeatedly in three runs. The 1st run was measured from the room temperature up to 400 °C, the 2nd and 3rd runs were measured from the room temperature to 450 °C. All runs were done at an increasing temperature with a heating rate of 5 °C/min. Thermogravimetry (DTG) and DSC were measured using MettlerToledo TGA/SDTA 851 and Mettler-Toledo DSC 822, respectively, with the heating rate of 5 °C/min. Specific surface areas of the samples (330 mg) were measured using Sorptometer Kelvin 1042 with adsorptive gas N2 and carrier gas He. The samples for DTG and electrical measurements were cut from the same blank; 48 mg compact samples are used for DTG measurements. The EGA was measured using Setaram LabSys 1600 thermal analyzer coupled with a Pfeiffer Omnistar Mass Spectrometer in an atmosphere of 79% of Ar + 21% of O2 with a flow rate of 60 ml/min. Table 1 The chemical composition (in mass %) of illite and kaolin. Illite Kaolin SiO2 Al2O3 Fe2O3 TiO2 CaO MgO K2 O Na2O L.O.I. 58.0 45.80 24.0 37.31 0.6 0.98 0.05 0.17 0.38 0.58 1.70 0.46 7.85 1.17 0.10 0.58 7.3 12.95 sufficiently high concentrations (Table 1) are the most probable dominant charge carriers in silicates. The conduction activation energies can be determined using the Arrhenius relation σ = σ0 exp(−Eσ kT ) (1) where σ is the conductivity at absolute temperature T, k is Boltzmann constant, Eσ is the conduction activation energy, and σ0 is the pre-exponential factor. The ionic conductivity relates to the heterodiffusion coefficient D of dominant charge carrying ions by Nernst-Einstein equation σ D = nq2 kT (2) where n is the concentration of dominant charge carrying ions and q is their charge. Electrical methods allow only estimation of the product of concentration and mobility of dominant charge carriers. The used electrical fields are low and do not influence the structure. The charge carriers, i.e. alkali ions or H+, H3O+, and OH− ions from a decomposition of water, are released by the heat and move in the electrical field; what results in the electrical current. Thus, measurements of temperature dependencies of the DC current can give a supplementary insight into the dehydration processes in both materials. The aim of this paper is to demonstrate the potential of DC conductivity measurements as supplementary methods for detection of temperature regions of PBW and ChBW release (Kriaa et al., 2014; Kubliha et al., 2016). This potential is demonstrated in this study of dehydration mechanisms in illite and kaolin. It is also compared the release of the physically bound water from illitic clay (86% of illite + 6% of montmorillonite) and kaolin (92% of kaolinite + 6% of illite/ muscovite) using the DC conductometry as to obtain a better understanding of differences of dehydration mechanisms in both materials. 3. Results and discussion In raw kaolin samples, the measured specific surface area gives the pore area equal to 10.33 m2. The initial mass of the physically bound water in the sample 4.7 × 10− 3 g (according to TG) gives the area of the monomolecular water layer of 9.64 m2, which is close to the pore area in the sample. In illite samples, the measured specific surface area gives the pore area equal to 13.62 m2. The initial mass of the physically bound water is 12.54 × 10− 3 g, which can create a monomolecular water layer of 25.71 m2. It is 1.9 times larger than the real pore area. These results indicate that, in raw samples, water layers continuously cover the crystallites of kaolin or illite at room temperature. In these water coatings, self-ionization of water (2H2O → H3O+ + OH−, H3O+ → H2O + H+) occurs. The created H+ ions, most likely, move using the hopping Grotthuss mechanism (Agmon, 1995), in which very mobile H+ ions jump between H2O molecules creating short living H3O+ ions. The short increase of the conductivity between room temperature and 70 °C (Figs. 2 and 5) despite the evaporation of PBW demonstrates an intensive creation of H+ and OH– ions. After this, the decrease of the conductivity is visible up to 150 °C because of the evaporation of PBW which is the source of H+ and OH– ions. Temperature dependencies of the DC conductivity of kaolin samples, measured 3 times successively, are presented in Fig. 2. The first run shows 2 extremes, at 55 °C and 298 °C. The temperature of the first 2. Experimental Füzérradvány illitic clay from Tokai region in Hungary is composed of illite (86 mass %), montmorillonite (6 mass %), quartz (4 mass %) and orthoclase (4 mass %). Sedlec kaolin from Czech Republic contains kaolinite (92 mass %), illite/muscovite (6 mass %) and quartz (2 mass %). Their chemical composition is summarized in Table 1. Raw illite and raw kaolin were crushed and sieved with a mesh size of 100 μm. The powder was mixed with distilled water to obtain a plastic mass with a water content of ~20 mass %. Cylindrical blanks 9 Applied Clay Science 149 (2017) 8–12 M. Kubliha et al. σdc [S.m-1] 500 400 300 200 of our investigations. The mobility of H+ ions increases with an increasing temperature. Therefore, an increase of the electrical current with an increasing temperature is observed at the beginning of heating (Fig. 2). At the same time, released water molecules evaporate from the sample, and the number of H+ and OH– charge-carriers decreases. Therefore, at the first run, a sharp decrease of the DC conductivity takes place from 60 °C up to 150 °C. In the low-temperature region, 20 °C–200 °C, the main charge carriers are H+ and OH−. After the evaporation of the PBW, monovalent Na+, and K+ ions gradually become dominant charge carriers. In the temperature range of 210 °C–300 °C, the slope of the relationship σ(1/T) is almost constant which gives a conduction activation energy of 0.76 eV. It indicates a cooperative movement of mobile ionic species (Na+, K+, and H+). At 298 °C, an anomaly in the temperature dependence of the DC conductivity indicates that the last fraction of PBW is hydrolyzed to H+ and OH– ions. The OH– ions associate with mobile alkali ions into neutral complexes that do not contribute to the DC conductivity. Thus, the slope of the temperature dependence decreases (Trnovcová et al., 2007). This association process does not influence the DTG plots. It is not observed in EGA as water molecules are not emitted. Above 350 °C, the effect of the PBW on the DC conductivity disappears. The conductivities during the first, second, and third runs of measurements are identical. Their conduction activation energies are the same, equal to 1.19 ± 0.02 eV. This value is common for the motion of alkali metal ions in oxide ceramics and glasses (Majková et al., 1977; Trnovcová et al., 2007). In the 2nd run, the temperature dependence of the DC conductivity of kaolin samples is approximately composed of 2 parts: 1) the lowtemperature part, where only tightly bound H+ and OH– ions of an almost constant concentration are the dominant charge carriers, with conduction activation energy of approximately 0.32 eV, and 2) the high-temperature part with Na+ and K+ charge carriers, with conduction activation energy of 1.19 ± 0.02 eV. In the 3rd run, contribution of H+ and OH– ions to the DC conductivity is almost totally suppressed, and the low-temperature part has a conduction activation energy of 1.02 ± 0.02 eV. The conduction activation energy of ~0.3 eV is characteristic for proton conduction in solids, the conduction activation energy of ~ 1 eV is common for a movement of K+ and Na+ ions in oxide glasses and ceramics (Majková et al., 1977; Trnovcová et al., 2007). It is evident that, in kaolin, there is no phase transformation up to 450 °C because the temperature dependencies of the DC conductivity above 350 °C for all three runs are identical. Thus, our results imply that in kaolin, dehydroxylation starts at a temperature higher than 450 °C. Temperature dependencies of the DC conductivity of raw illite samples, measured 3 times successively, are presented in Fig. 5. The first run shows 3 maxima, at 72 °C, 186 °C, and 298 °C, resulting from the PBW release. These temperatures coincide with DTG plots (Fig. 6), which show 3 minimums at 59 °C, 166 °C, and 266 °C. The EGA indicates only two peaks of emission of H2O with maximums at 56 and 263 °C (Fig. 7) which are almost the same temperatures as those in DTG. The discrepancy of the temperatures between different experimental methods results from the increasing mobility of H+ ions with an increasing temperature. This increase partly compensates for the decrease of the proton concentration and shifts the DC conductivity maxima to higher temperatures. Up to 220 °C, the dominant charge carriers in illite are H+ and OH– ions moving in the water cover of the illite crystals. An increase in the conduction activation energy at higher temperatures indicates that alkali metal ions, especially K+ ions present in substantial amounts in illite, contribute to the DC conductivity due to their rapidly increasing mobility with an increasing temperature. At 298 °C, in a narrow temperature range, an association reaction between alkali metal ions and released OH– ions into neutral complexes slows down the increase of the DC conductivity with an increasing temperature. In the 2nd run, after heating at 400 °C, the temperature dependence T [°C] 100 1x10-6 1x10-8 1x10-10 10-12 10-14 1.5 2.0 2.5 3.0 3.5 1000/T [K-1] 0.0 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1.0 0.0 -2.0x10-5 -4.0x10-5 DTG [s-1] DSC [W.g-1] Fig. 2. Temperature dependencies of the DC conductivity of kaolin samples: 1st run (□), 2nd run (○), and 3rd run (∆). -6.0x10-5 -8.0x10-5 -1.0x10-4 DSC DTG 0 100 -1.2x10-4 200 300 400 500 600 T [°C] H2O ionic current [10-12 A] CO2 ionic current [10-14 A] Fig. 3. DSC and DTG curves of kaolin. 90 H2O ionic current CO2 ionic current 80 70 60 50 40 30 20 10 0 0 200 400 600 800 1000 T [°C] Fig. 4. EGA plots of kaolin; H2O release (○), CO2 release (□). maximum coincides with the DTG plot (Fig. 3), which shows a minimum at 65 °C. The DTG plots of kaolin give only one minimum for the release of the PBW. The results of mass spectrometric measurements indicate also only one peak of emission of H2O connected with the release of the PBW with a maximum at 106 °C (Fig. 4). The released mass of the PBW represents a ~1.5% loss of the initial mass of the sample. Fig. 3 and Fig. 4 also show DTG, DSC and EGA results over 400 °C, i.e. show the influence of dehydroxylation which is not a topic 10 Applied Clay Science 149 (2017) 8–12 M. Kubliha et al. σdc [S.m-1] 500 400 300 200 activation energy of 1.22 eV, is observed. It results from the onset of dehydroxylation during the preceding heating to 450 °C. Upon partial dehydroxylation, the DC conductivity of illite increases due to the enhanced concentration of defects; a slightly higher conduction activation energy indicates that alkali metal ions are moving in a more disordered structure (Freund, 1967). T [°C] 100 1x10-6 1x10-8 4. Conclusions 1x10-10 The comparison of results of the DC conductivity measurements of kaolin and illite leads to the following conclusions: The loss of physically bound water in “as received” kaolin samples takes place in two steps, visible in the temperature dependencies of the DC conductivity at 55 °C and 298 °C. Up to 220 °C, the dominant charge carriers are H+ and OH– ions. At higher temperatures, K+ and Na+ ions contribute to the DC conductivity. The loss of physically bound water in “as received” illite samples takes place in 3 steps, which are visible in the temperature dependencies of the DC conductivity at 72 °C, 186 °C, and 298 °C. Up to 220 °C, the dominant charge carriers are H+ and OH– ions. At higher temperatures, K+ ions contribute to the DC conductivity. After heating at 400 °C, in the low-temperature part of the temperature dependence of the DC conductivity H+ and OH– ions are the dominant charge carriers, with conduction activation energy of 0.31 eV in both materials. At higher temperatures, Na+ and K+ ions are dominant charge carriers with a conduction activation energy of 1.19 ± 0.02 eV in kaolin; K+ ions are most important charge carriers with a conduction activation energy of 1.12 ± 0.02 eV in illite. Above 400 °C, the dehydroxylation starts in illite samples and the DC conductivity of illite increases. The slightly higher conduction activation energy (1.22 eV) indicates that alkali metal ions are moving in a more disordered structure. In contrary to illite, kaolin has only 2 stages of PBW release and its dehydroxylation starts above 450 °C. At the 2nd run, when dehydroxylation does not start, temperature dependencies of the DC conductivity of both materials are very close. 10-12 10-14 1.5 2.0 2.5 3.0 3.5 1000/T [K-1] 0.0 0.05 0.00 -1.0x10-5 -0.05 -0.10 DTG [s-1] DSC [W.g-1] Fig. 5. Temperature dependencies of the DC conductivity of illite samples: 1st (▽), 2nd (◇), and 3rd (◁) run. -2.0x10-5 -0.15 -3.0x10-5 -0.20 DSC DTG -0.25 -4.0x10-5 -0.30 0 100 200 300 400 500 600 T [°C] Fig. 6. DSC and DTG curves of illitic clay. H2O ionic current [10-11 A] CO2 ionic current [10-14 A] Acknowledgement 45 H2O ionic current CO2 ionic current 40 35 This work has been supported by the Ministry of Education, Science, Research, and Sport of the Slovak Republic [grant number VEGA 1/ 0162/15]. The authors thank ceramic plant PPC Čab for providing Sedlec kaolin. The authors are also indebted to Mr. J. Biber (Inter-ILI Engineering Office, Hungary) for providing illitic clay. 30 25 20 References 15 Agmon, N., 1995. The Grotthuss mechanism. Chem. Phys. Lett. 244, 456–462. http://dx. doi.org/10.1016/0009-2614(95)00905-J. Carroll, D.L., Kemp, T.F., Bastow, T.J., Smith, M.E., 2005. Solid-state NMR characterisation of the thermal transformation of a Hungarian white illite. Solid State Nucl. Magn. Reson. 28, 31–43. http://dx.doi.org/10.1016/j.ssnmr.2005.04.001. Drits, V.A., McCarty, D.K., 2007. The nature of structure-bonded H2O in illite and leucophyllite from dehydration and dehydroxylation experiments. Clay Clay Miner. 55, 45–58. http://dx.doi.org/10.1346/CCMN.2007.0550104. Ferrari, S., Gualtieri, A.F., 2006. The use of illitic clays in the production of stoneware tile ceramics. Appl. Clay Sci. 32, 73–81. http://dx.doi.org/10.1016/j.clay.2005.10.001. Ferrari, S., Gualtieri, A.F., Grathoff, G.H., Leoni, M., 2006. Model of Structure Disorder of Illite: Preliminary Results. Z. Kristallogr. Suppl. 23, 493–498. Freund, F., 1967. Kaolinite-metakaolinite, a model of a solid with extremely high lattice defect concentrations. Ber. Dtsch. Keram. Ges. 44, 5–13. Grim, R.E., 1962. Applied Clay Mineralogy. McGraw Hill, New York. Gualtieri, A.F., Ferrari, S., Leoni, M., Grathoff, G., Hugo, R., Shatnawi, M., Paglia, G., Billinge, S., 2008. Structural Characterization of the Clay Mineral Illite-1M 41. pp. 402–415. http://dx.doi.org/10.1107/S0021889808004202. Kriaa, A., Hajji, M., Jamoussi, F., Hamzaoui, A.H., 2014. Electrical conductivity of 1:1 and 2:1 clay minerals. Surf. Eng. Appl. Electrochem. 50, 84–94. http://dx.doi.org/10. 3103/S1068375514010104. Kubliha, M., Trnovcová, V., Ondruška, J., Štubňa, I., Bošák, O., Kaljuvee, T., Bačík, P., 2016. DC conductivity of illitic clay after various firing. J. Therm. Anal. Calorim. 124, 81–86. http://dx.doi.org/10.1007/s10973-015-5129-4. 10 5 0 0 100 200 300 400 500 600 700 800 T [°C] Fig. 7. EGA plots of illite; H2O release (□), CO2 release (○). of the DC conductivity of the illite sample is composed of 2 parts (Fig. 5): 1) the low-temperature part, where H+ and OH– ions are the dominant charge carriers, with conduction activation energy of approximately 0.31 eV, and 2) the high-temperature part with dominant K+ charge carriers, with conduction activation energy of 1.12 ± 0.02 eV. At higher temperatures, the DC conductivity approaches that of the first run. The third run consists of the same low-temperature part as the 2nd run. However, above 130 °C, higher conductivity, with conduction 11 Applied Clay Science 149 (2017) 8–12 M. Kubliha et al. electrical conductivity of kaolin. J. Therm. Anal. Calorim. 118, 597–601. http://dx. doi.org/10.1007/s10973-013-3533-1. Štubňa, I., Kozík, T., 1996. Electric conductivity of kaolin in the temperature range 150–560 °C. J. Therm. Anal. 46, 607–610. http://dx.doi.org/10.1007/BF02135040. Trnovcová, V., Furár, I., Hanic, F., 2007. Influence of technological texture on electrical properties of industrial ceramics. J. Phys. Chem. Solids 68, 1135–1139. http://dx.doi. org/10.1016/j.jpcs.2007.03.004. Trnovcová, V., Podoba, R., Štubňa, I., 2012. DC conductivity of kaolin-based ceramics in the temperature range 20–600 °C. Építőanyag 64, 52–55. http://dx.doi.org/10. 14382/epitoanyag-jsbcm.2012.8. Viczian, I., 1997. Hungarian Investigations on the “Zempleni” Illite. vol. 45. pp. 114–115. Wattanasiriwech, D., Wattanasiriwech, S., 2011. Fluxing action of illite and microcline in a triaxial porcelain body. J. Eur. Ceram. Soc. 31, 1371–1376. http://dx.doi.org/10. 1016/j.jeurceramsoc.2011.01.025. MacKenzie, K.J.D., 1969. Development of electric potentials in clay minerals and alumina—silica systems. Nature 222, 469–471. http://dx.doi.org/10.1038/222469a0. Maiti, G.C., Freund, F., 1981. Dehydration-related proton conductivity in kaolinite. Clay Miner. 16, 395–413. Majková, E., Trnovcová, V., Bohun, A., 1977. Charge and mass transport in “alkali-free” alumino-phosphate glasses. Part 2. Diffusion of monovalent and bivalent cations. Phys. Chem. Glasses 18, 83–86. Ondruška, J., Štubňa, I., Trnovcová, V., Húlan, T., Vozár, L., 2015a. DC conductivity of illite with fly-ash between 20–1050 °C. Adv. Mater. Res. 1126, 123–128. http://dx. doi.org/10.4028/www.scientific.net/AMR.1126.123. Ondruška, J., Trnovcová, V., Štubna, I., Podoba, R., 2015b. DC conductivity of ceramics with calcite waste in the temperature range 20–1050 °C. Ceram. Silikaty 59, 176–180. Podoba, R., Štubňa, I., Trnovcová, V., Trník, A., 2014. Temperature dependence of DC 12