Microsoft word - sajidu et al pdf.doc

International Journal of Physical Sciences Vol. 3 (1), pp. 001-011, January, 2008 Available online at http://www.academicjournals.org/IJPS ISSN 1992 - 1950 2008 Academic Journals
Ful Length Research Paper
Groundwater fluoride levels in villages of Southern
Malawi and removal studies using bauxite
Sajidu, S. M. I.1, Masamba, W. R. L.1,2*, Thole, B.3 and Mwatseteza, J. F.1
1Chemistry Department, Chancel or Col ege, University of Malawi, P.O. Box 280, Zomba, Malawi. 2Harry Oppenheimer Okavango Research Centre, University of Botswana, P/Bag 285, Maun, Botswana. 3The Polytechnic, University of Malawi, P/ Bag 303, Chichiri, Blantyre 3, Malawi Accepted 16 January 2008 Dental and skeletal fluorosis are known to be caused by excessive fluoride ingestion particularly from
drinking water sources. Dental fluorosis is common in some parts of Malawi but studies on fluoride
levels of drinking water sources have not been adequately done. This paper presents our findings in
fluoride levels of some drinking water sources in selected rural areas of Southern Malawi and studies
on the potential of locally sourced bauxite in water defluoridation at laboratory scale. The work has
revealed high levels of fluorides (>1.5 mg/L) in some parts of the study area. A positive correlation was
observed between the pH of the water and fluoride concentrations. No correlation existed between
fluoride concentration and electroconductivity. Experiments on water defluoridation with bauxite
showed that the raw bauxite has a capacity of 93.8 % at a dose of 2.5g/200 ml of 8 mg/L F- solution.
Powder X-ray diffraction characterization of the raw bauxite showed gibbsite (Al(OH)3) and kaolinite
(Al2Si2O5(OH)4) as the major components. The high defluoridation capacity of the bauxite is thus
attributable to gibbsite and kaolinite minerals. Precipitation of AlF3 is predicted to be the fluoride
removal mechanism with the gibbsite while exchange of OH- groups in the gibbsite layers of the
kaolinite with F- ions occurs in the kaolinite component of the defluoridation material. Evaluated by the
residual fluoride concentration in solution the fluoride uptake reaction kinetics of the system was found
to be consistent with pseudo-first-order kinetics.

Key words: Fluoride, bauxite, Southern Malawi, gibbsite, kaolinite.

INTRODUCTION
Groundwater contributes only 0.6% of the total water intensively cultivated fields, disposal of hazardous resources of the earth (Meenakshi and Mehshwari, wastes, liquid and solid wastes from industries, sewage 2006); however, it is the major and most preferred source disposal and surface impoundments (Anwar, 2003; Oren of drinking water in rural as wel as urban areas in deve- et al., 2004; Amina et al., 2004; Kass et al., 2005). Fluo- loping countries because it does not require treatment. rides are one such type of contaminants that leach from Efforts in addressing water issues in Malawi concentrate rocks and soils into ground water. The high fluoride on bringing water to the people with little attention to ground waters are general y of the sodium-chloride, water quality (NEP, 1996; SEP, 1998). In 1993, 24% of sodium-calcium-chloride or sodium-bicarbonate-chloride rural population depended on boreholes for domestic types characterized by a moderately high pH value in the water supply, 22% relied on tap water, 6% on shal ow range 8.2 to 9.4 pH units (Kohut and Hodge, 1985). wel s and 48% had no access to safe water (GoM, 1993). Fluoride is toxic at concentrations greater than 1.5 Various natural and anthropogenic ecological factors pol- mg/L and is associated with dental fluorosis (Harper et al, lute the groundwater because of deep percolation from 1979). At fluoride levels between 3.0 to 6.0 mg/L in drink- ing water skeletal fluorosis may be observed and when a concentration of 10 mg/L is exceeded crippling fluorosis can ensue. Recent investigations have shown that non- *Corresponding author. E-mail: [email protected]. skeletal fluorosis whereby soft tissues are affected can 002 Int. J. Phys. Sci. also occur due to prolonged intake of fluoride in high con- tration membranes (Hu and Dickson, 2006) have also centrations (Mjengera and Mkongo, 2003). In Malawi, cases of dental fluorosis are very common in localized In Malawi, free fluoride levels of up to 8.6 mg/L in bor- parts of the country. At Mtubwi Primary School in Mac- ehole water have been reported at Ulongwe in Machinga hinga District, for example, mild to severe dental fluorosis District (Sibale et al., 1998), 7 mg/L at Mazengera and cases in Standard 3 and 4 pupils has been determined to Nathenje in Lilongwe, 9.6 mg/L in Nkhotakota, 8.0 mg/L be 50 to 80% (Sajidu et al., 2005). in Karonga, 5.8 mg/L in Nsanje and 3.4 mg/L in Mwanza Some of the solutions to fluoride and fluorosis problems (Carter and Bennet, 1973; Msonda, 2003). This cal s for are; change to alternative sources of drinking water, water defluoridation technologies in Malawi since the improvement in nutritional status of the people at risk and levels are way above the WHO guideline value of 1.5 defluoridation which is the removal of excess fluoride in mg/L. Earlier studies had shown that clays can effectively the water. In most rural areas where alternative sources be used for water defluoridation in the country (Msonda, of water are unavailable, defluoridation may offer a prac- 2003). Our group has shown high defluoridation capacity tical solution to the problem. Several methods of excess- of local y sourced gypsum (Masamba et al., 2005). A fluoride removal from portable water are used in many dose of 10 g gypsum calcined at 400oC in 200 ml of 8 mg/L fluoride solution resulted in about 70% fluoride The Nalgonda technology where aluminium sulphate is removal within equilibration time of 30 min. added to precipitate fluorides in raw water to which lime This work was therefore aimed at providing more data may be added to obtain the appropriate alkalinity (Sushe- on water fluoride levels in the country (particularly in the ela, 1992) is one such defluoridation technique. The Southern region). The work also investigated the poten- major concerns with this technique are that the do-ses of tial of bauxite in water defluoridation at laboratory scale. alum and lime are determined after assessing the fluoride levels and alkalinity of the water changes with season, and, if the alum dose is not adhered to there is a possi- MATERIALS AND METHODS
bility of excess aluminium contamination in the water Chemicals and reagents
(Susheela, 1992). Clays and soils have been studied for water de- Analytical reagent grade chemicals for analyses were purchased fluoridation after the work of Bower and Hatcher (1967). from Technilab Company (Ltd), Blantyre, Malawi. Deionised water Their work showed positive results in fluoride removal was used in al solution preparations and analyses. particularly in the case of aluminium hydroxide (gibbsite). Fol owing this study, many soils and clay minerals have Water fluoride levels in selected rural areas of Southern Malawi
been investigated. Il inois soils in the USA (Omueti and Jones, 1977), sodic soils in India (Chhabra et al., 1980), The Southern Region of Malawi consists of 12 districts of the total 27 districts in the country (Figure 1). It covers an area of 31, 753 activated alumina (Schoeman and MacLeod, 1987), km2 and it is the most densely populated region with a population of electro-activated alumina (Lounici et al., 2004), combi- 5,345,045. Due to lack of data on fluorosis cases in health authority nation of aluminium and zirconium oxide with specific offices, selection of sampling sites was based on dental Donnan dialysis (Garmes et al., 2002), electrocoagulation technicians' perception of areas thought to be fluorosis endemic. and floatation processes using monopolar aluminium The sites included six boreholes in Nsanje district, four boreholes electrodes (Emamjomeh and Muttucumaru, 2006), clay and one river in Chikwawa district, eleven boreholes in Zomba, eleven boreholes in Machinga district and sixteen boreholes in pottery (Chaturvedi et al., 1988), Ando soils in Kenya Mangochi district. Samples were col ected in triplicate at each (Zevenbergen et al., 1996), fired clay chips in Ethiopia sampling point into thoroughly rinsed one-litre polythene bottles. (Moges et al., 1996), fly ash (Chaturvedi et al., 1990), Concentration of fluoride was determined using a fluoride ion kaolinite (Kau et al., 1997), il ite-goethite soils in China specific electrode (Orion number 9409) with Sargent Welch (Wang and Reardon, 2001), South African clays and bau- pH/activity meter model PAX 900. The sample or the standard xite (Coetzee et al., 2003; Das et al., 2005) have been solution (25 ml) was initial y mixed with 25 ml of total ionic strength adjusting buffer (TISAB) which had been prepared according to recommended procedure in order to provide stable analytical Charred bone at 550oC has also been employed in the conditions by adjusting pH and complexing interfering ions. The bone charcoal defluoridation technology (PHE 2001). The samples were also tested for pH (on site at each water source) and bone charring process if not properly carried out results in electrical conductivity (EC) . a product of low defluoridation capacity that leaves the treated water with a rotten meat taste and smel . The Use of bauxite for water defluoridation
whole process from bone preparation to charring and application may take over 24 h. Other carbonaceous Raw bauxite for defluoridation work was obtained from the materials in addition to the borne char have also been Geological Survey of Malawi in Zomba but was original y acquired investigated for defluoridation (Abe et al, 2004). Methods from Mulanje mountain in Mulanje district. The bauxite was ground based on use of anion exchange resins under elect-rodia- into powder form. It was then calcined in air using a muffle furnace at temperatures of 200, 300, 400, and 500oC to acquire different lysis conditions (Zen et al., 2005; Tor, 2006) and nanofil- phases of bauxite for experimentation so as to determine the phase Sajidu et al. 003
Figure 1. Map of Southern Region of Malawi showing al districts including those under study.
with the highest defluoridation capacity. The samples were held at 12 h. Fluoride levels in water samples were determined as des- each reaction temperature for 2 h and quench-cooled to room tem- cribed in section 2.2 above. To determine the effect of bauxite on perature. Defluoridation capacity determinations were carried out by water quality pH, aluminium, silicon, sulphate, carbonate and phos- mixing 200 ml of 8 mg/L fluoride solution with 2.5 g of defluoridating phate concentrations in water were determined. Aluminium and material (bauxite or a calcined phase of bauxite) and shaking for 30 silicon were analysed using Atomic Absorption spectrophotometry. min. Fluoride concentration in the solution was monitored hourly for Sulphates were determined by a turbidimetric method on Jenway 004 Int. J. Phys. Sci. 6405 UV-Visible spectrophotometer, phosphates were determined ±0.01) was observed at Mtubwi borehole in Machinga through a vanadomolybdophosphoric acid colorimetric method and District. There was a general positive correlation between carbonates were determined titrimetrical y as described in APHA (1985). The percent defluoridation capacity (%E) was calculated as pH and fluoride concentrations in the water samples (Table 1a and b) while no correlation was observed between the fluoride concentrations and electrical (C − C x100 conductivity (EC) values. EC is a valuable indicator of the amount of material dissolved in water. Its values fluctuated widely at the different sampled sites. The where CO and C are the initial and final (after defluoridation) concentrations respectively of the fluorides in solution. The raw recommended value of EC for portable water is 2500 bauxite was characterized by powder X-ray diffraction (PXRD) µS/cm (WHO, 1988). The high EC values in some water using Shimadzu 600 X-Ray Diffractometer. The PXRD patterns samples (such as 6800 µS/cm at Chigonele in Mangochi were col ected in continuous scan mode with monochromatic CuKα district) show that they are unfit for human consumption. (λ=1.5418 Å) radiation that was selected using a curved germanium The high levels of fluo-ride can largely be attributable to (111) monochromator. X-rays which were produced at the X-ray source (copper radiation) were reflected at the germanium hydrogeochemical origin since anthropogenic factors monochromator (111 planes) giving pure CuK including disposal of hazardous sewage, liquid and solid α radiation which was diffracted by the sample. The data were col ected in the range from wastes from Industries are very insignificant in the 10.0 to 80.0o (2θ) by a linear PSD which was set at a step size of sampled rural areas. Therefore, the fluoride concentration 0.5o (2θ) and counting time of 60s per step. Compound identification depends on the geological, chemical and physical was made using a search-match computer program supported by characteristics of the aquifer, the porosity and acidity of the Joint Committee on Powder Diffraction Standards database the soil and rocks, temperature and the depth of the JCPDS (1997) together with diffraction profiles obtained from standards of bauxite and clay minerals as reported in literature wel s. These are some of the factors that may have to be (Mineral database, 2002). In order to determine the dehydration further investigated in those high fluoride areas if perhaps temperature and also the amount of water in the bauxite occurring they can assist in locating alternative sites for planting both as surface physisorbed water and water coordinated to the boreholes or consider defluoridation techniques viable in bauxite, thermogravimetric analysis (TGA) was approximated by heating the bauxite sample in air in a muffle furnace at increasing specified temperatures and measuring the weight loss of the material using an analytical balance. Use of bauxite for water defluoridation

RESULTS AND DISCUSSION
Defluoridating material characteristics
Water fluoride levels in selected rural areas of
The mineralogical characteristics of the bauxite (and its Southern Malawi
calcined phases) used in this study are shown by the PXRD patterns in Figure 2a and b. The raw and 200oC Tables 1a and b give the groundwater fluoride levels in calcined phase profiles qualitatively indicate presence of the sampled areas of Southern Malawi. The fluoride data gibbsite (Al(OH)3) as shown by intense reflection at 2θ at twenty-one of the forty-nine sampled locations was angle of 17.93° (d= 4.94 Å), and doublet at 2θ angle of found to be above the WHO maximum limit of 1.50 mg/L. 20.51° probably combining the literature reflections at d= Five out of the six (83%) sampled boreholes in Nsanje 4.34 Å and 4.30 Å of a crystal ine gibbsite. Intense showed fluoride levels above 1.5 mg/L ranging from 1.65 kaolinite (Al2Si2O5(OH)4) reflections are also shown at 2θ to 7.50 mg/L. Nsanje is one of the districts in the country angles of 11.58° and 24.80°. There are no observable where sight of dental fluorosis amongst the locals is structural changes between the raw and the 200oC common. This could therefore be attributable to high phase. At 400oC gibbsite has started to transform to levels of fluorides in the ground water. The other notable boehmite (AlOOH) as can be seen by the loss of its district is Machinga particularly in Liwonde neighbour- reflections in Figure 2b. However kaolinite shows its high hood which encompasses the eleven sampled boreholes resistance to heat and its reflections are very intense at in the district. Of the eleven sites studied in the Liwonde 400oC indicating that it becomes more crystal ine. Upon neighbourhood, eight (73%) indicated high levels of calcinations to 500oC even the kaolinite structure is fluoride, higher than 2 mg/L in al cases. The results destroyed resulting in highly amorphous phase that could extend the database of fluoride levels in this area since be attributed to different activated aluminas such as χ the earlier study by Sibale et al. (1998) only looked at one and γ-Al2O3. Thermogravimetric results (Figure 3) of the borehole (Mtubwi B/H). Only one borehole in Zomba bauxite indicated a very steep mass loss up to 200oC district at Mbando vil age which is close to Lake Chilwa fol owed by very smal and slow mass loss as the indicated high fluoride levels (6.51 ± 0.01 mg/L). High temperature was increased. The first mass loss could be levels of fluorides were also sporadical y spotted in attributable to physisorbed water on the material while Mangochi district with the highest value recorded at the second one could arise from slow dehydroxylation Nsauya-1 vil age (3.64 ± 0.01 mg/L). Maximum pH (9.50 process of the gibbsite as it transforms to different forms Sajidu et al. 005 Table 1a. Drinking water fluoride concentrations, pH and electroconductivity in suspected fluorosis endemic areas in Southern
Malawi (Nsanje, Chikwawa and Mangochi districts) District
Fluoride conc. (mg/ L)
Kaleso vlge/Bangula 1.65± 0.06
4.91± 0.03
1.89± 0.00
2.45± 0.01
7.25± 0.01
Nyamphota/Nkombezi Tomali Trading Centre 1.91± 0.00
1.93± 0.01
St. Michael's Sec. Sch. 1.85± 0.01
Chimbenda vge/Malindi Mlangalanga vge/Malindi 2.60± 0.00
Mtakataka turn off Mangochi Hospital, 2.45± 0.01
Chigonele (Nsauya) 1.85 ±0.02
Nsangazi F.P. Sch. 3.64 ±0.01
Monkeybay Trad. Center
Figure 2a. PXRD profile of the raw bauxite and its 200oC calcined
phase used for defluoridation in the study. Gibbsite and kaolinite reflections are indicated within the profile Figure 2b. PXRD of 400oC and 500oC calcined phases of bauxite.
006 Int. J. Phys. Sci. Table 1b. Drinking water fluoride concentrations, pH and electroconductivity in suspected fluorosis endemic areas in
Southern Malawi (Zomba, and Mangochi districts). District
Fluoride conc.
Mchilima F.P. Sch 2.76 ± 0.01
5.06 ± 0.00
Evangelical Bap. Church 5.08 ±0.02
Machinga D.Hospital 2.04± 0.01
4.88± 0.00
phases of the bauxite. The highest defluoridation capacity of 95.3% was obtained in defluoridation with the 200oC calcine and this was fol owed closely by that obtained with the raw bauxite at 94.8%. These two results did not differ significantly at 5% level of significance; as such the calcination is not necessary. The lowest defluoridation capacity of 87.1% was obtained in defluoridation with the 500oC bauxite calcine. The slight increase in defluori- dation from raw bauxite to the 2000C calcine could be attributed to absence of physisorbed water that was lost during heating as shown by the thermogravimetric data. The loss of water meant that there was an increase in concentration of the reactive gibbsite and kaolinite for the removal of fluorides. The defluoridation capacities de- creased steadily from 300oC through 400oC to the 500oC calcine. This could be a result of phase changes occur- Figure 3. Thermogravimetric plot of the raw bauxite.
ring during the dehydroxylation processes. Gibbsite is known to undergo the dehydroxylation sequence gibbsite, boehmite (AlOOH) and χ-Al2O3 (about 300oC), γ-Al2O3 of alumina (Al2O3). (about 450oC), δ-Al2O3 (about 800oC), θ-Al2O3 (about 900oC) and final y the crystal ine α-Al Defluoridation capacity and reaction rates
known as corundum (above 1100oC) (Sajidu, 2001). γ- alumina (γ-Al2O3, the phase at 500oC) which can have Table 2 presents defluoridation capacities of the different surface area up to 400 m2/g is an important industrial ca- Sajidu et al. 007 Table 2. Fluoride concentrations and defluoridation capacities of bauxite and its calcined phase at 2.5 g/200 ml solution of 8
Bauxite phase
Initial Fluoride
Equil. Fluoride
Fluoride removed
Defluoridation capacity
Conc. (mg/L)
conc. (mg/L)
Table 3. Correlation coefficients between increase in concentrations and decrease in fluoride concentrations in defluoridation
Correlation coefficient catalyst support needed, for example, in automobile ex- [F ]= [F ] kt haust catalytic converters. The decrease in fluoride up- where [F-] is the residual take as calcination temperature increases therefore sug- gests that the uptake mechanism is not simple adsorption concentration at time t and [F]o is the initial fluoride con- onto the material surfaces since calcination to 500oC centration which is 8 mg/L and k is a reaction constant. increases the surface area through formation of amor- The fol owing reaction rate equation was applied (Atkins, phous alumina phases which would have increased the 1982): for the first order reaction kinetics. A plot of In[F] uptake if adsorption was the only uptake mechanism. It versus t (Figure 4b) gave a straight line with k = 0.1568 was observed that the pH of the solution increased after hr-1 (R2 = 0.9772) indicating that the reaction is wel defluoridation indicating release of OH- ions. This there- described by the first order reaction kinetics. The fore suggests a precipitation mechanism where Al3+ ions concentration change of the fluoride in this reaction was from the material react with the F- ion to form stable AlF3 assumed to be a direct indication of the reaction progress that precipitate out releasing OH- in the process. since the con-centration change of Al3+ in both gibbsite The removal of fluoride in this raw bauxite is not only and kaolinite was difficult to monitor (due to lack of due to gibbsite but also kaolinite. The kaolinite structure quantitative characterization of this heterogeneous mate- includes highly accessible hydroxyl groups positioned on rial). It was also assumed that the bauxite concentration the gibbsite layers which al ow for anion exchange was significantly larger than that of fluoride. The ob- modeled in the fol owing equation (Kau et al., 1997): served kinetics can thus be treated as pseudo-first order The insignificant reduction in fluoride removal capacity by the material (95.3% for 200oC calcine to 87.1% for 500oC In order to determine the maximum dose of the bauxite calcine) could also be explained by the fact that kaolinite in a 200 ml solution of 8 mg/L F-, different masses of the is much less prone to desiccation and therefore it is likely bauxite were mixed with the solution and the equilibrium that it returns its structure at high temperature and residual fluoride concentration was determined. Figure 5 continues to remove the fluoride by almost a constant shows that equilibrium concentrations decreased quickly capacity up to about 400oC of calcinations. At 500oC the with increase in mass of bauxite used in defluoridation up bauxite is almost amorphous (without crystal ine kaolinite to about 10g dose. After this point the equilibrium con- or gibbsite phases). Fluoride uptake mechanism by the centrations were within 0.05 to 0.2 mg/L. This implied that 500oC phase can therefore be attributed to adsorption on increasing the mass of bauxite per volume of water would activated aluminas formed during the calcinations. only increase defluoridation efficiency up to a dose of 10 The changes in residual fluoride concentrations of the g/200ml (1 g/20ml) solution mixture (2.5 g of 200oC per 200 ml of 8 mg/L F-) as a function of time at room temperature are as shown in Figure 4a. It can be seen from the Figure that fluoride Effects of various ions on defluoridation
concentration decreases exponential y with time til equilibrium concentration is obtained after 24 hrs of Table 3 presents correlation coefficients between initial reaction. Therefore we correlated the residual fluoride concentrations of various ions in solution and concentra- concentration with time as fol ows: tions of fluoride that were attained in solution after deflori- 008 Int. J. Phys. Sci. Table 4. Effect of the bauxite on water quality after defluoridation (2.5g of 200oC calcine 200 ml of 8mg/L F-)
Parameter
Initial conc.
Conc. after defluoridation
Change in conc.
WHO limit
bdl: below detectable limits; n.a: not available. defluoridation. Positive correlation shows that more fluo- for precipitation with the Al3+ of the gibbsite or adsorption rides remained in solution when the concentration of the into the kaolinite layer structure through exchange with particular ion was increased. Negative correlation shows OH-. At high pH, the lower efficiency could be due to both greater fluoride removal as concentration of the particular Le Chatelier's principle and competition. OH- and F- are ion increases. Carbonates and chlorides showed high isoelectronic with same charge and ionic radi . The higher direct positive correlation indicating that higher initial con- the concentration of OH-, the more difficult it is for OH- centrations of carbonates and chlorides resulted in more already attached to Al3+ to be replaced by F- and go into fluorides remaining in solution (high concentration of solution. Similar observation has been made by earlier residual fluoride in water). Carbonates and chloride researchers who noted optimum pH of defluoridation by hindered defluoridation with bauxite. The negative effect clay minerals at pH 5.8 of the F- solution (Moges, 1996). of carbonates on fluoride sorption could be explained on the basis of comparative solubility of carbonates and
Effects of raw bauxite on water quality
fluorides. Carbonates being general y less soluble in water than fluorides interacted more strongly with Al3+ Table 4 shows effects of the raw bauxite on water quality ions than the fluorides. Chloride hindrance to defluori- after the defluoridation process. The pH changed from dation was a result of similarity of chemistry between 6.90 ± 0.01 to 7.40 ± 0.07 representing a pH increase by fluorides and chlorides both ions being halides. General selectivity trends for sorption also place chloride before There was insignificant increase in aluminium and fluoride ion (NAS, 1998): silicon in the water. The water quality changes however Sulphate >iodide >nitrate >bromide >chloride >fluoride were within recommended limits (WHO, 2004). The Calcium ions showed a high negative correlation indi- bauxite coloured the water to earth brown making the wa- cating a decrease in residual fluoride concentration with ter unaesthetic for drinking. The colour could easily be increase in calcium ion concentration. Calcium thus clarified by using activated carbons or Moringa oleifera enhanced defluoridation and this could be attributable to seed powder. M. oleifera has demonstrated to clarify formation of CaF2. Sulphate ions had a moderate enhan- turbid river water from 500 NTU to 5 NTU (Folkard et al., cement on fluoride sorption a result that contradicts 1993; Henry et al., 2004). general selectivity trends above. From such a trend sul- phate would be expected to interfere with de-fluoridation yet sulphate enhances defluoridation with bauxite. This is Defluoridation results with actual ground water using
in part explained by specific resinion interaction imply-ing the 200oC calcine
that fluoride interact more with bauxite compared to sul- Figure 7 gives defluoridation results on water samples phate ion. There exist resin/ion combinations that wil not col ected from high fluoritic sites (Bangula, Level cross- adhere to general selectivity trends (Coulson and ing, Mlangalanga, Zimba, Machinga District Hospital and Richardson, 1997). Mtubwi). Batch defluoridation experiments were done on the actual groundwater samples as described for the Effects of pH on defluoridation
synthetic fluoridated water above using the 200oC calcined phase of bauxite. The results are consistent with Figure 6 shows defluoridation capacities (2.5g of 200oC the observations made on the synthetic water. In al calcine in 200 ml of 8mg/L F-) at different pH levels. The cases fluoride concentrations were reduced to below 1.5 defluoridation was much lower at pH 2 and pH 10 and optimum pH was obtained at pH 4. The low fluoride up- take at pH 2 could be explained by the greater tendency Conclusion and Recommendations
of fluoride to form aqueous protonated fluoride; thus, reducing the concentration of free fluoride available for In expanding the database of fluoride levels in Malawi, Sajidu et al. 009
Figure 4a. Residual fluoride concentration against time.

Figure 4b. Plot of In[F] against time.
the work has revealed high levels of fluorides (>1.5 component of the defluoridation material. Field trials in a mg/L)in most vil ages in Nsanje district and in locations high fluoritic water source such as Mtubwi B/H need to be around Liwonde in Machinga district. Laboratory experi- carried out to test the applicability and efficiency of the ments have shown the potential of raw bauxite in defluo- method in real life settings. ridation up to 93.8% at a dose of 2.5 g/200ml of 8 mg/L F. This suggests that raw bauxite can be used for defluori-
ACKNOWLEDGEMENTS
dation to treat water in high fluoride areas. The high defluoridation capacity of the bauxite is attributable to We are very grateful to the International Program in gibbsite and kaolinite minerals. Precipitation of AlF3 is Chemical Sciences (IPICS) through project code, MAW: thus predicted to be the fluoride removal mechanism with 02 for supporting the ongoing work on water quality in the gibbsite while exchange of OH- groups in the gibbsite Malawi and for supporting Benard Thole to undertake his layers of the kaolinite with F- ions occurs in the kaolinite MSc study on water defluoridation with in the project. 010 Int. J. Phys. Sci.
Figure 5. Effects of different amounts of defluoridating bauxite on the fluoride.

Figure 6. Plot of % fluoride removed versus pH (2.5 g of 200oC calcined phase
of bauxite in 200 ml of 8 mg/L F-).
Figure 7. Defluoridation results on water samples.
Sajidu et al. 011 Thanks also go to District Dental Technicians of Chik- clay. J. of Contaminant Hydrol. 28:267-288. wawa, Machinga and Mangochi for advising the team on Kohut AP, hodge W (1985). Ground water resource of British Columbia fluorosis susceptible areas in their districts, Messrs J.F – Gulf islands. Lounici H, Belhocine D, Grib H, Drouiche M, Pauss A, Mameri N (2004). Kamanula and F.F.F Masumbu for carrying out the labo- Fluoride removal with electro-activated alumina. Desalination ratory analyses. The Geological Survey of Malawi is also thanked for providing the raw bauxite samples for our de- Meenakshi , Mahenshwari RC (2006). Fluoride in drinking water and its removal. J. of Hazard. Mater. B137:456-463 Masamba WRL, Sajidu SMI, Thole B, Mwatseteza JF (2005). Water defluoridation using Malawi's local y sourced gypsum. Physics and Chemistry of the Earth, Parts A/B/C 30 (11-16):846 - 849 REFERENCES
Mineral database (2002). http://webmineral.com/data last update: Abe I, Iwasaki S, Tokimoto T, Kawasaki N, Nakamura T, Tanada S Mjengera H, Mkongo G (2003). Appropriate defluoridation technology (2004). Adsorption of fluoride ions onto carbonaceous materials. J. of for use in fluoritic areas in Tanzania. Physics and Chemistry of the Col oid Interface Sci., 275:35-39. Earth 28:1097-1104. Amina C, Lhadi LK, Younsi A, Murdy J (2004). Environmental impact of Moges G, Zrwge F, Socher M (1996). Preliminary investigations on the an urban landfil on a coastal aquifer. J. of Afr. Earth Sci., 39 (3- defluoridation of water using fired clay chips. J. of Afr. Earth Sci., 21 Anwar F (2003). Assessment and analysis of industrial liquid waste and Msonda WK (2003). A study of groundwater quality, water defluoridation sludge disposal at unlined landfil sites in arid climate. Waste and impact of dental fluorosis on children in Nathenje Lilongwe manage. 23(9):817-824. Malawi. MSc Thesis, University of Malawi: Malawi. APHA (1985), American Public Health Association, American Water NAS (Nuffield Advanced Science) A (1998). Ion Exchange: Special Works Association and Water Pol ution Control Federation. Standard Study. Penguin Books publication. methods of the examination of water/wastewater. 16th Edn. APHA, NEP (National Environmental Policy), Malawi Government (1996). AWWA, and WPCF, New York. Ministry of Natural Resources and Environmental Affairs, Lilongwe, Atkins PW (1982). Physical Chemistry. 2nd Edn: Oxford University Press, Oxford; United Kingdom. Omueti JAI, Jones RL (1977). Fluoride adsorption by Il inois soils. J. Bower CA, Hatcher JT (1967). Adsorption of fluoride by soils and Soil Sci. 28:546-572. minerals. Soil Sci. 103:151-154. Oren O, Yechieli Y, Bohlke JK, Dody A (2004). Contamination of Carter GS, Bennet JD (1973). The Geology and Mineral Resources of groundwater under cultivated fields in an arid environment, Central Malawi. Bul etin No 6, 2nd Ed., Government print, Zomba: Malawi. Arava Val ey, Israel. J. Hydrol. 290 (3/4):312-328. Chaturvedi AK, Phathak KC, Singh VH (1988). Fluoride removal from PHE (2001). Protection of the Human Environment (PHE). Fluoride in water by adsorption on China Clay. Applied Clay Sci. 3:337-346. Chaturvedi AK, Yadava KP, Phathak KC, Singh VH (1990). Defluoridation of water by adsorption on fly-ash. Water Air and Soil Sajidu SMI (2001). Thermal Studies of Layered Oxides and Hydroxides. Pol ution 49:51-61. MPhil Thesis, University of Cambridge: U. K. Chhabra R, Singh A, Abrol IP (1980). Fluorine in sodic soils. Soil Sci. Sajidu SM, Masumbu FFF, Fabiano E (2005). Drinking water quality Soc. of Am. J., 44:33-36. and mapping of fluoritic areas in Liwonde, Malawi. National Research Coetzee PP, Coetzee LL, Puka R, Mubenga S (2003). Characterization Council of Malawi Annual Research Conference, Kalikuti Hotel, of selected South African Clays for defluoridation of natural water. Lilongwe, Malawi, 16-17 June, 2005. Water SA. 29 (3):331-338. Schoeman JJ, MacLeod H (1987). The effect of particle size and Coulson JM, Richardson JF (1997). Chemical engineering 3rd Edn, interfering ions on fluoride removal by activated alumina. Water SA. Butterworth-Heineman: Oxford. Das N, Pattaik P, Das R (2005). Defluoridation of drinking water using SEP (State of Environment Report for Malawi), Malawi Government activated titanium rich bauxite. J. of Col oid. Interface Sci., 292:1-10. (1998). Ministry of Natural Resources and Environmental Affairs, Emamjomeh MM, Muttucumaru S (2006). An empirical model for Lilongwe, Malawi defluoridation by batch monopolar electrocoagulation/floatation (ECF) Sibale FK, Chidothe C, Tsakala M (1998). Oral health status among 12 process. J. of Hazard. Mater., B131:118-125. and 15 year old school pupils in Machinga District, Malawi. Folkard GK., Sutherland JP, Grant WD (1993). Natural coagulants at Susheela AK (1992). Defluoridation of drinking water – merits of pilot scale. 18th WEDC Conference, Kathmandu, Nepal. alternative technologies. A Nutrition Foundation of India paper. Loughborough University Press, pp 51-54. Garmes H, Persin F, Sandeaux J, Pourcel y G, Mountadar M (2002). Tor A (2006). Removal of fluoride from water using anion exchange Defluoridation of groundwater by hybrid process combining membrane under Donnan dialysis conditions. J. of Hazard. Mater., adsorption and Donnan dialysis. Desalination 145:287-291. GoM (Government of Malawi) (1993). Malawi situation analysis of Wang Y, Reardon EJ (2001). Activation and regeneration of soil sorbent poverty. UN. Lilongwe, Malawi. for defluoridation of drinking water. Applied Geochem., 16:531-539. Harper HA, Rodwel VW, Mayes PA, (1979). Review of physiological WHO (1988). Assessment of fresh water quality. Global environmental chemistry. Lange Medical Publications, California. monitoring system (GEMS) report on the related environmental Henry E, Dembo EG, Sajidu SM, Zimba G (2004). A preliminary monitoring. World Health Organization; Geneva. investigation on use of Moringa Oleifera polyelectrolytes for WHO (2004). Guidelines for drinking water quality. 3rd Edn, Vol. 1, wastewater treatment in Malawi. Proceedings of WAMDEC 2004 28- Recommendations, World Health Organization: Geneva. 30 July, Elephant Hil s, Victoria Fal s, Zimbabwe. Zevenbergen C, Van Reeuwijk LP, Frapporti G, Louws RJ, Shuiling RD Hu K, Dickson JM (2006). Nanofiltration membrane on fluoride removal (1996). A simple method for defluoridation of drinking water at vil age from water. J. of Membr. Sci., 279:529-538. level by adsorption on Ando soils in Kenya. Science of the Total JCPDS (1997). In Powder Diffraction File. JCPDS International Centre Environment 188:225-232. for Diffraction Data. Swarthmore, Pennsylvania. Kass A, Yechiele G, Vengosh A, Starinky A (2005). The impact of freshwater and wastewater on the chemistry of shal ow groundwater: a case study from the Israel Coastal aquifer. J. of Hydrol., 300 (1- Kau PMH, Smith DW, Binning P (1997). Fluoride retention by Kaloin

Source: http://ithuteng.ub.bw/bitstream/handle/10311/1091/Sajidu_IJPS_2008.pdf?sequence=5