MENVSC / Department of Ecology and Resource Management / Some boreholes in South Africa which serve as a source of drinking water for rural communities are reported to have high fluoride concentration, much above the WHO guideline of 1.5 mg/L. This study aimed at activating aluminosilicate clay soil mechanochemically, modifying aluminosilicate clay soil with Al-oxide and fabricating porous ceramic granules using Al-oxide modified mechanochemically activated aluminosilicate clay soil/ mechanochemically activated clay soil/ corn starch and evaluating their performances in defluoridation of groundwater. The raw clay materials were mechanochemically activated for 5, 10, 15 and 30 minutes for physicochemical transformation of the solid aggregate. The morphology of the samples showed the honeycomb structure. The surface area analyses of samples using Brunauer–Emmett–Teller (BET) gave the highest surface area of 50.5228 m2/g at 30 min activation time. Hence, the optimum activation time was 30 min. The Fourier Transform Infrared (FT-IR) analysis showed increase in the absorbance of FT-IR by Si-O-H groups at 510 cm-1 with increasing milling time. This is evidence that more surface Si-O-H groups were available at higher particle surface area that would be necessary to interact with fluoride. X-ray diffraction (XRD) analyses revealed that, at 30 minutes milling time, the peak broadening is intensified whereas the reflection peak intensities decreased. The X-ray fluorescence spectrometry (XRF) results for 30 minutes milling time showed that silica and alumina were the highest components in the clay soil. Using the activated clay in batch defluoridation of fluoride-spiked water, a maximum fluoride removal of 41% was achieved at a pHe of 2.41. The initial fluoride concentration was 9 mg/L while the sorbent dosage was 0.6 g/100 mL and the contact time being 30 minutes. The adsorption data fitted to both Langmuir and Freundlich isotherms. The adsorption data fitted only the pseudo-second-order kinetic, showing chemisorption. Optimization of Al3+ concentration for modification was carried out by modifying the mechanochemical activated aluminosilicate clay soil with different concentrations of Al3+ from which the optimum modification was achieved with 1.5 M. Characterisation studies on the Al-oxide modified mechanochemically activated aluminosilicate clay soil by SEM, BET, FT-IR, XRD and XRF, analyses were carried out to determine the resultant changes in physicochemical properties of the adsorbent owing to modification. The SEM image of Al-oxide modified mechanochemically activated clay soil showed many small pores and honey-comb structure on the surface of different images. The BET surface area and the BDH adsorption cumulative area of the Al-oxide modified mechanochemically activated v aluminosilicate clay soil were more than double those for the raw clay soil. There was also an increase in pore volume of the Al-oxide modified mechanochemically activated aluminosilicate clay soil. The FT-IR spectra showed that there was increase in the absorbance by the Si-OH, H-O-H, Al-O-H and Si-O-Al. The equilibrium pH of solution was higher than the point-of-zero charge (pHpzc) implying that fluoride removal occurred at solution pH > pHpzc where the net surface charge of the mechanochemically activated clay aluminosilicate soil was negative.The efficiency of 1.5 M Al-oxide modified aluminosilicate clay soil to remove fluoride from water was studied and found to be 96.5 % at pHe 6.86, contact time of 30 minutes and dosage of 0.3 g/100 mL for 10 mg/L fluoride solution at 200 rpm shaking speed. The result shows that Al-oxide modified mechanochemically activated aluminosilicate clay soil is effective for defluoridation. The adsorption data fitted to both Langmuir and Freundlich isotherms. The adsorption data fitted only the pseudo-second-order kinetic, showing chemisorption. Al-oxide modified mechanochemically activated aluminosilicate clay soil was tested for fluoride removal on field water and the percentage fluoride removal was 96.5 % at the dosage of 0.6 g/100 mL with the pHe of 6.48. The optimum Al-oxide modified mechanochemically activated aluminosilicate clay soil/ mechanochemically activated clay soil/ corn starch mixing ratio for fabrication of porous ceramic granules was determined by varying ratios and temperature. The optimum ratio found was 20:5:1.The porous ceramic granules were characterised using SEM, BET, FT-IR, XRD and XRF. SEM analysis showed that the porous ceramic granules have the porous structure of the organic foam template. The porous ceramic granule showed an increase in pore surface area and volume as compared to mechanochemically activated aluminosilicate clay soil. The FT-IR showed the presence of a strong broad bending and stretching vibrations band at about 993 cm-1 which shows the presence of Si–O–Si bonds. Mineralogical characterisation showed the presence of quartz, albite, horneblende and microcline as the main minerals of the calcined porous ceramic granules. The major oxides of the porous ceramic granules as shown by XRF analysis were SiO2, Al2O3, MnO and Na2O. The porous ceramic granules reduced the concentrations of fluoride in the water from 10 to 3.31 mg/L. The optimum adsorption capacity was 0.6648 mg/g at a pHe of 6.32 and the percentage fluoride removal was 66.9 % at an adsorbent dosage of 1.0063 g/100 mL and a temperature of 600 ⁰C. The porous ceramic granules were tested for fluoride removal on field water and the percentage fluoride removal was 45.4 % at the dosage of 1.0009 g/100 mL with the pHe of 7.87. Mechanochemically activated aluminosilicate clay soil showed higher adsorption capacity at acidic pH, therefore it is recommended that future work should focus on improving their adsorption capacity at wider range of pH. The porous ceramic granules can also be evaluated in column dynamic flow experiments.

MENVSC / Department of Ecology and Resource Management / Fluoride is one of the anionic contaminants which is found in excess in groundwater because of geochemical reaction or anthropogenic activities such as the disposal of industrial wastewaters. Among various methods used for defluoridation of water such as precipitation, ion-exchange processes, membrane processes, the adsorptions process is widely used. It offers satisfactory results and seems to be a more attractive method for the removal of fluoride in terms of cost, simplicity of design and operation. In this work, the preparation of clay polymer nanocomposites (CPNCs) used in defluoridation began by modifying the original natural Mukondeni clay to render the layered silicate miscible with the chosen polymer, microcrystalline cellulose. Clay polymer nanocomposites (CPNCs) were synthesized using the melt intercalation method. Mukondeni black clay with microcrystalline cellulose as polymers was melt mixed at 220 °C for 10 minutes in an extruder for exfoliation of the resulting composite. Physicochemical characteristics and mineralogical characteristics of the CPNC was determined using XRD, XRF, BET, FTIR and SEM. Batch adsorption experiments were conducted to determine the efficiency of CPNCs in defluoridation of groundwater. The pH, EC, TDS and fluoride concentration of field water was determined using the CRISON MM40 multimeter probe and the Orion versastar fluoride selective electrode for fluoride concentration. Elemental analysis revealed that CPNC 1:1 is mainly characterized of cellulose, Quartz and Albatite as the major minerals with traces of Montmorillonite, Ednite and Magnesium as minor minerals constituting CPNC 1:1. The structure of 1:4 CPNC was partially crystalline and partially amorphous showing increased cellulose quantity (1:4 clay to cellulose) as compared 1:1 CPNC, 1:2 CPNC and 1:3 CPNC. Maximum adsorption of fluoride was attained in 10 minutes using 0.5g of 1:4 CPNC removed 22.3% of fluoride. The initial fluoride concentration for the collected field groundwater was 5.4 mg/L, EC 436 μS/cm, and TDS 282 mg/L. The regeneration potential of CPNCs was evaluated through 3 successive adsorption desorption cycles. Fluoride removal decreased after the first cycle for all ratios of CPNCs, a continued decreased can be observed following the second cycle. CPNC 1:2 decreased from 9.32 % at the 1st cycle to 2.84 % and 0.56 % on the 2nd and 3rd cycle respectively. CPNC 1:4 decreased from 8.22 % at the 1st cycle to 4.80 % and 0.72 % on the 2nd and 3rd cycle respectively. The fluoride-rich Siloam groundwater had a slightly alkaline pH of 9.6. iv The low adsorptive characteristic displayed by all 4 CPNCs can be deduced from the BET analysis that revealed low surface area, pore volume, and pore size, it is evident from the BET analysis that less fluoride will be absorb as adsorption sites will be limited. Based on the findings of this study, recommendations are designing of correct preparation techniques to obtain nanocomposites with desirable properties, polymer melting points and evaporation point of the binder should be taken into consideration. / NRF