The effect of some additives on the synthesis of cordierite via solid state reaction and on the thermal expansion behavior of cordierite

As the member of magnesium aluminum silicate minerals, cordierite (Mg2Al4Si5O18) can be found in nature but the naturally occuring form is very impure and usually contains iron. Because of the insufficient amount and insufficient purity of natural cordierite mineral, it is generally obtained by different synthetic routes for industrial applications. Cordierite has some characteristic properties like low thermal expansion coefficient, low dielectric constant, high chemical and mechanical stability. These key properties make cordierite a desirable material for many different industries. There are mainly three synthetic routes for synthesizing cordierite ceramics. The conventional routes for preparing cordierite ceramics are solid state sintering and the glass-ceramics method. There is also a sol-gel technique for obtaining cordierite ceramics. The solid state sintering is the most common method for industrial production. Cordierite can be synthesized using clay (Al2O3.2SiO2.2H2O), talc (3MgO.4SiO2.2H2O) and alumina as the raw materials. It can also be obtained by sintering fine oxide powders. Different starting materials and different processing methods for cordierite bodies can result in different properties of cordierite ceramics. For this reason, cordierite bodies have a wide range of values (0,7 x 10-6/oC – 2,2 x 10-6/oC) for thermal expansion coefficient in the literature. Cordierite is one of the ternary compound in MgO-Al2O3-SiO2 ternary system. The stochiometric composition of cordierite is located in the primary field of mullite (3Al2O3.2SiO2). So it melts incongruently at about 1460oC and produces mullite and a liquid phase. The primary field of cordierite is surrounded by six invariant points which are too close to each other. So, the unexpected temperature oscillations in the sintering furnace can result in the formation of non-desirable secondary phases. There are some limitations for solid state sintering of cordierite. The reaction between reactants (oxides or natural raw materials) is a diffusion controlled reaction at the beginning. However when the reaction continues, one or more species of reactants must be transported across an increasing barrier of product. Since the interposition of this barrier layer (interface layer) diminishes the effective contact between reactants, the diffusion of these reactant phases through this reaction interface is extremely slow. Therefore, the solid state cordierite formation is an interface controlled reaction and the reaction rate is too slow. Second limitation of cordierite formation via solid state sintering is that, the primary field of cordierite is surrounded by six invariant points (eutectic and peritectic points) which are too close to each other. So the small variations in the temperature can result in rapid melting or the formation of non-desired secondary phases. Therefore, it is easy to find cordierite with a secondary phase such as mullite, spinel, cristobalite or enstatite. The need for minimizing these problems about sintering cordierite materials is usually solved by using different sintering additives in the literature. These sintering additives not only enhance the rate of the reaction but also have an effect on narrow sintering range of cordierite. On the other hand, the sintering additives that are used for enhancing the sinterability of cordierite, degrade the thermal and electrical properties of cordierite. Many investigations have been made so far, to optimize the sintering behaviour and the properties of cordierite. In this study, the individual effects of 2-5 wt.% ZrO2, TiO2, CeO2 and MgF2 additives on the formation rate and thermal expansion behavior of cordierite obtained by the solid state reaction between the stoichiometric compositions of spinel (MgO.Al2O3) and silica (SiO2) powders, were investigated. In the experimental studies, the stoichiometric amounts of spinel-quartz mixtures in the same particle size were pressed at constant pressure (25 bar) and sintered at constant temperature at 1350oC. The individual effects of different additives were examined by changing the sintering time (10-50 hours). The sintered samples with or without additive were cooled in the furnace, after that the qualitative and quantitative X-ray diffraction analysis, dilatometric analysis and density measurements using Archimedes principle were carried out. Two general methods were used for qantitative X-Ray analysis studies for determining cordierite formation in the samples. These methods are Rietveld and the internal standard method. The results of the quantitative X-ray analysis indicated that the most effective additive for cordierite formation was CeO2 for this experimental conditions. When it is compared with additive-free samples, the reaction for cordierite formation takes place more slowly for samples containing ZrO2 additive. In other words, ZrO2 has a negative effect on cordierite formation. In the samples containing TiO2, magnesium titanate compound and magnesium alumino titanate solid solution were observed in the XRD phase analysis. The oscillation in the cordierite amount in samples containing TiO2, is thought to be the result of formation of compound and solid solution in the system. For the samples containing different amounts of MgF2, the amount of cordierite showed discontinuity. This situation is because of the melting of MgF2 at the sintering temperature and formation of amorphous glassy phase in the ceramic body. %4 and 5 wt.% addition of CeO2 and TiO2 which are known to have a positive effect on cordierite formation, did not show a significant change in the cordierite amount. In general, for all of the samples with different additives, the cordierite amounts for short sintering time were satisfactory for industrial applications. When Rietveld method and the internal standard method were compared, it can be said that the excessive overlapping in the cordierite system effects the cordierite amounts obtained by Rietveld refinement method. Especially on the higher cordierite amounts, the diffraction peaks of cordierite covers the diffraction peaks of spinel phase. Therefore, the difference in the cordierite amount of different quantitative X-Ray diffraction methods (Rietveld and the internal standard) is higher in the samples with CeO2 additive. Internal standard method is found to be more reliable method than Rietveld method for determining cordierite formation. The results of dilatometric analysis indicate that an increment in the cordierite content of sample resulted in a decrease in the coefficient of thermal expansion. The inverse relationship between cordierite amount and the thermal expansion coefficients was determined in all samples with or without additives. As an exception, the highest amount of cordierite was obtained in CeO2 addition but the lowest value of thermal expansion coefficient was obtained in TiO2 addition. This is because of the very low thermal expansion coefficient values of magnesium titanate compound and magnesium alumino titanate solid solution in the samples that have TiO2 additive. The samples' density measurements were carried out by using Archimedes principle. The density values in the reaction sintering vary with the sintering time, the type and amount of the phases in the ceramic body. Among all of reactants, additives and products, the lowest density values are cordierite and silica phases. Therefore, when the cordierite amount increased in the body, the density values are expected to decrease.