The clay minerals are basically composed of repeating tetrahedral and octahedral planar sheets. The tetrahedral and octahedral sheets form a polymeric alumino-silicate layer that is mostly crystalline in structure. Impurities such as Fe2O3, TiO2, K2O and Na2O vary among clay samples from the different locations. Clay minerals share a basic set of structural and chemical characteristics, yet each clay mineral has its own unique set of properties that determine how it will interact with other chemical species  .
The application of clay materials is influenced by their physical and chemical properties which determine their suitability for specific usage. The geographical location for each clay materials confer unique physico-chemical properties which ultimately determines the type of material produced and its application  . Specific physical and chemical properties of clays such as kaolin are also dependent on the environment of deposition, geological origin, geographic source and the material method at the end of processing  . The basic physical and chemical properties of clay materials are widely used in both academic and industrial fields. These properties may vary under the influences of both natural processes and human disturbances  .
Clays are abundant and readily available. They also possess distinct adsorptive characteristics from one origin to another. The adsorption characteristics depend on the chemical and mineralogical composition as well as on textural, structural, and morphological properties  . Plasticity is also one of the most important clay properties that are related to different parameters. The most disaggregated clay minerals with major ionic exchange capability are more often plastic in nature. Montmorillonitic minerals are most plastic, than illitics, and least kaolinitics  . Clays and clay minerals such as montmorillonite, vermiculite, illite, kaolinite, and bentonite are widely used in process and petroleum industries, engineering and constructions, environmental remediation, ceramics and refractories, pharmaceuticals and agricultural sectors  .
Nigeria has large quantity of clay minerals that are widely spread across the country. Some of the studies conducted on Nigerian clays are reported in many literatures      . This study evaluates the physical and chemical properties of the clay deposits in selected locations in Plateau State, Central Nigeria using statistical analysis.
2. Materials and Methods
2.1. Study Site
The study sites referenced in this research are all located in Plateau State, Central Nigeria. The deposit sites include RarinSho (RC), Major Porter (MP), Kwi (KC), Wereng Camp (WC) and Naraguta (NC). The study areas are bounded between latitudes 8˚30 E and 9˚00 E and longitudes 9˚30 N and 10˚00 N as shown in Figure 1. The clays at the deposit sites were classified as kaolinite and ball clays by Nigerian Mining Corporation (NMC). Geographically, the areas are characterized by uneven topographic profile with many hills.
Figure 1. Map of the study sites [Adapted from Ministry of Land and Survey, Plateau State].
2.2. Sample Collection and Determination
The samples were obtained from underground local mines 10 - 20 cm beneath the soil surface. Ten different points within each deposit collected was by randomized sampling. In addition, 5 kg of fresh samples in lump form were obtained randomly from ten different points within each deposit from underground local mines pit at the depth of 10 - 20 cm. The samples were air-dried for several days and crushed using a set of Denver crushers by Denver Equipment Company England. Each crushed sample was thoroughly mixed, coned and quartered. Two opposite composite representatives were obtained and consequently milled and pulverized. These were packaged in small polyethene bags as representatives of the samples for the required tests  .
2.3. Methods of Analysis
2.3.1. Chemical Analysis
The chemical compositions of the pulverized samples were determined by Energy Dispersive X-ray Fluorescence (EDXRF) of the model PW4030 X-ray photometer that uses a rhodium anode tube. The sample film was placed firmly in a waxed and gold plated sample holder. The Energy Dispersive patterns were obtained with the help of a computer attached to the instrument and each compound recorded in percentage  .
2.3.2. Loss on Ignition
Loss on ignition was determined according to the Lechler and Desiletes method,  . The samples were oven-dried at 110˚C. Then 1 g of the sample of clay was placed in preheated, cooled and pre-weighed silica crucibles and heated to 1000˚C in a furnace for an hour. The crucibles and contents were removed, cooled in a desiccator to room temperature and weighed again. The loss in weight was calculated.
2.4. Physical Analysis
2.4.1. Plasticity Index of Clay Samples
The Atterberg plasticity method prescribed by  was used to calculate the plasticity index of the clay samples from their respective liquid and plastic limits determined.
2.4.2. Particle-Size Analysis
The particle size test was carried out by using the standard Hydrometer method  . 50 g of milled oven-dry clay was weighed into a 250 cm3 beakers and mixed with 100 cm3 calgon, then allowed to soak for 30 minutes. The mixed suspension was transferred into a sedimentation cylinder and filled up to mark point with distilled water.
The hydrometer was inserted into the mixture in the cylinder and the readings were taken after interval of 40 seconds twice. After 2 hours, another hydrometer reading and temperature were recorded. After 40 seconds and 2 hours, all the sand and silt particles would have settled and only clay will remain in suspension. The percentages of silt and clay were then calculated and the interpretation of result was made by using a textural triangle.
2.4.3. Drying and Firing Shrinkage of Clays
The clay drying and firing shrinkages were determined from brick bars prepared using a mechanical hydraulic press (model D-7064 Paul Weber) with its accessory moulds and the method adopted was  .
2.5. Statistical Analysis
The statistical tools employed to carry out the data analysis of this study were the Microsoft 2010 Excel Package and Statistical Package for Social Sciences (SPSS) version 17.0 software. The Microsoft Excel was used to compute the mean and standard deviation generated results while the SPSS was employed to carry out the Analysis of variance (ANOVA) by Post-hoc tambane multiple comparisons F-test and Kristal Wallis T-test all at 5% confidence level. This assisted in determining if significant differences exist in the average means of the concentration of the compounds from the different sample sites. The p-values that were equal to or less than 0.05 were considered significant.
Pearson correlation statistical tool was also used to establish if significant positive interrelationships exist between the chemical and physical parameters in each site of the clay samples. The strength of the variables on the scale of −1 (perfect inverse relation) through 0 (no relation) to +1 (perfect sympathetic relation) were determined at both confidence levels of 0.01 and 0.05.
3. Results and Discussion
3.1. Physicochemical Characteristics of the Clay Samples
The results of the chemical and physical analysis of RarinSho (RC), Major Porter (MP), Wereng Camp (WC), Kwi (KC) and Naraguta (NC) clay samples are summarized in Tables 1-3 respectively. The chemical properties of the samples were determined statistically by analysis of variance (ANOVA) at (F8, 36 = 52.40, p < 0.05) and Krystal Wallis one sample t-test (T8, 37.38, p < 0.01). The results of chemical analysis shows the different oxide components that indicate also the variations in the means of the concentrations of the compounds contained in the studied clay bodies.
The ANOVA in Table 1 showed that significant differences exist between the average means of the concentrations of the oxides of the different samples except for few minor constituents whose values are similar. The Krystal Wallis one sample t-test result of the randomized block design model used for major and trace oxides shown in Table 2 depicted that a greater degree of differences are observed from the p-values of silica (SiO2) and alumina (Al2O3). It shows that these oxides (SiO2 and Al2O3) are significantly more, followed by the p-values of
Table 1. Chemical composition of oxides in the clays.
a-d = Mean average values within a row with different superscripts are significantly different at 5% level and a-a = Mean average values within a row with the same small letter superscripts are not significantly different at 5% level.
Table 2. Kristal wallis one-sample t-test.
* = Significantly different and P = Probability value.
Table 3. Physical properties of the clay samples.
LL = Liquid Limit; PL = Plasticity Limit; PI = Plastic Index.
calcium oxide (CaO) and loss on ignition (LOI). The high amounts of SiO2, Al2O3 and LOI define the clay samples as hydrated alumino-silicate type of minerals  . This is attributed to the fact that kaolin clays are principally composed of SiO2, Al2O3 and water, which have a chemical composition of Al2Si2O5(OH)4, also represented as Al2O3∙2SiO2∙2H2O.
The results of the Pearson bivariate correlation coefficient determined between the chemical and physicalproperties of the clay samples in Tables 4-8 showed very significant, strong and positive correlations. Most of the correlation values fall within the range of 0.900 and above but less than unity in all the samples.
3.2. Correlation Associations between the Clay Parameters
Interrelationships in clay chemical and physical properties were assessed for each sample site by Pearson Bivariate correlation analysis. Multiple and very strong significant correlations were observed and selected. The correlations determined for RarinSho clay sample shown in Table 4 were highly significant. Very high significant interactions were observed between silt and PL (r = 0.953), K2O and CaO (r = 0.942), Quartz and pH (r = 0.925), and Al2O3 and PL (r =
Table 4. Pearson bivariate correlation for RC sample site.
*Correlation is significant at the 0.05 level (2-tailed). **Correlation is significant at the 0.01 level (2-tailed) R = Pearson correlation.
0.903) all at (p < 0.05). There was also a moderate level of significant correlation between TiO2 and Na2O (r = 0.896) at (p < 0.05). The chemical analysis of this sample showed that the value of SiO2 was much higher than normal, chiefly due to the quantity of silt and quartz present from its size analysis. The association also indicates that the content of silt and Al2O3 contribute positively to the increase of the plastic limit of the sample. The site also indicates that the increase in the concentration of K2O resulted to the slight rise of that of CaO.
Similarly, some interrelationships occurred between the physical and chemical properties of Major Porter clay sample. The most positive significant correlations amongst the oxides concentration was observed between Fe2O3 and TiO2 (r = 0.995) at (p < 0.01) shown in Table 5. Significant correlations were also observed between Fe2O3/TiO2 and clay (r = 0.977, 0.972), also at (p < 0.01), and K2O and Clay (r = 0.952) at (p < 0.05). Clay and PI (r = 0.950) also associated at (p< 0.05). There was also a strong double significant association between Fe2O3/TiO2 and the plastic index of the sample at (r = 0.958; p < 0.05). The sample in this site was notably observed to contain an appreciable amount of Fe2O3 and SiO2. The low LOI value of the sample is attributed to the amount of quartz obtained from its size analysis.
Table 6 shows a very strong significant interactions between Fe2O3 and clay (r = 0.967, p< 0.01), Al2O3 and TiO2 (r = 0.916), and SiO2 and TiO2 (r = 0.900) at (p < 0.05) in Wereng Camp clay deposit. Also observed was the significant
Table 5. Pearson bivariate correlation for MP sample site.
*Correlation is significant at the 0.05 level (2-tailed). **Correlation is significant at the 0.01 level (2-tailed). R = Pearson Correlation.
Table 6. Pearson bivariate correlation for WC sample site.
*Correlation is significant at the 0.05 level (2−tailed). **Correlation is significant at the 0.01 level (2−tailed). R = Pearson correlation.
Table 7. Pearson bivariate correlation for KC sample site.
*Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed). R = Pearson Correlation.
Table 8. Pearson bivariate correlation for NC sample site.
*Correlation is significant at the 0.05 level (2-tailed); **Correlation is significant at the 0.01 level (2−tailed); R = Pearson Correlation.
correlation between TiO2 and silt (r = 0.942; p < 0.05). The significant strong association of Fe2O3 and TiO2 can be explained from their appreciable quantities as impurities in the sample. The high concentration of TiO2 in the sample is due to its close association to the small quantity of the silt.
The most significant and very strong positive relationship amongst the properties in Kwi clay sample was between PL and pH (r = 0.991, p< 0.01). There was also a strong correlation between clay and quartz (r = 0.953, p < 0.05) and between quartz and PI (r = 0.944; p < 0.05) as shown in Table 7. The sample also indicated numerous negative correlations at different strengths but not considered. The interrelationship in this deposit has shown that the sample has the lowest pH in Table 3 contributing to the positive increase of its PL value.
Table 8 reveals many positive correlations in Naraguta clay deposit. The highest significant relationship was between clay and silt (r =0.978), Fe2O3 and TiO2 (r = 0.977), CaOand clay (r = 0.974) all at (p < 0.01). There was also a positive significant relationship between CaOand LL (r = 0.942). Other significant associations were found between CaO, clay and PI of the sample at (r = 0.933; p < 0.05). Also strongly significant was the correlation between LOI and LL (r = 0.926; p < 0.05). The high concentrations of Fe2O3 and TiO2 at 28% and 3% respectively indicate a very high level of impurities. The moderate values of LOI, LL, and PL observed could all be attributed to the very high amount of SiO2 and quartz.
The very strong and high positive correlations in the clay properties reveal the amount and level of the oxide concentrations which consequently influence their physical properties and thereby their quality and functions. The results of the chemical and physical properties show that all the samples are suitable for various industrial applications.