Removal of fluoride from drinking water using highly

Journal of Alloys and Compounds 719 (2017) 460e469

Contents lists available at ScienceDirect

Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom

Removal of fluoride from drinking water using highly efficient nano-adsorbent, Al(III)-Fe(III)-La(III) trimetallic oxide prepared by chemical route Mrinal K. Adak a, Asmita Sen a, Arnab Mukherjee a, Shrabanee Sen b, Debasis Dhak a, * a b

Department of Chemistry, Sidho-Kanho-Birsha University, Purulia, 723104, WB, India Piezo-Ceramic Division, Central Glass and Ceramics Research Institute, Kolkata, WB, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 31 March 2017 Accepted 14 May 2017 Available online 15 May 2017

Al(III)-Fe(III)-La(III) trimetallic oxide was synthesized through chemical process. XRD revealed monoclinic phase when calcined at 800 C, 4 h. IR study confirmed the formation of desired compound. A high surface area of 22.14 m2/g was observed with a pore volume 1.5 102 cc/g. The porosity and the microstructure were studied by SEM and TEM. F adsorption capacity was studied varying the adsorbent dose in 100 ml: 0.05 g, 0.1 g, 0.2 g and 0.3 g and the contact time in minute: 15, 30, 45 and 60 taking 3 ppm, 5 ppm and 10 ppm fluoride solutions. Experimental observation revealed high fluoride adsorption capacity ~99%. © 2017 Elsevier B.V. All rights reserved.

Keywords: Drinking water Fluoride removal Nano-adsorbent Chemical synthesis

1. Introduction In recent years a variety of cost effective natural, modified and synthetic solid adsorbents were designed for the removal of fluoride from drinking water and waste water. Many different materials such as Fe-impregnated chitosan [1], alum impregnated activated alumina [2], MnO2 modified earthenware [3], Alumina modified iron oxide [4], Mg doped ferrihydrite [5] etc. Effective removal of fluoride can be made by using methods like Nalgonda process [6], Contact precipitation process [7], electrode deposition process [8], adsorption [9] etc. However, due to high cost, extreme pH and large dosage, it was not practicable for field application. Besides that most of the adsorbents exhibit the fluoride removal capacity not greater than 2 mg/L, therefore, they were not appropriate for the drinking water treatment purpose, especially as some of them could only work at an extreme pH value, such as activated carbon which was only effective for fluoride elimination at pH less than 3.0 [10]. Activated alumina was a potential adsorbent for removal of fluoride for years, which was dependent on the crystallinity, the activation process and the pH of the solution [11], was found to remove fluoride from fluorinated water up to 70% when 4 g/L activated alumina was used for the initial fluoride

* Corresponding author. E-mail address: [email protected] (D. Dhak). http://dx.doi.org/10.1016/j.jallcom.2017.05.149 0925-8388/© 2017 Elsevier B.V. All rights reserved.

concentration 13.2 mg/L at pH 7 [9]. The surface of activated alumina was modified by impregnation with alum and the fluoride absorption efficiency was increased up to 92.48% at pH 6.5 at an adsorbent dose 8 g/L for the initial fluoride concentration 25 mg/L for 3 h [2]. Farrah et al. [12] investigated the fluoride ion interaction with amorphous Al(OH)3, gibbsite or alumina(Al2O3) over a wide pH range 3e8 and fluoride concentrations 0.1e1.0 mM and the highest extent of fluoride removal was found to be 9 mol/kg at pH 5.5e6.5. Bimetallic oxides were found to increase the adsorption efficiency. Maliyekkal et al. [13] reported manganese-oxide-coated alumina (MOCA) and magnesia-amended activated alumina (MAAA) [14], the absorption efficiency were increased up to 95% at pH 4e6. Bansiwal et al. [15] reported copper oxide coated mesoporous alumina which enhances the fluoride removal capacity of water and the adsorption capacity was become 7.220 mg/g. The removal capacity was found to be three times higher than that of unmodified activated alumina (2.232 mg/g). Sulfate-doped Fe3O4/Al2O3 nanoparticles as reported by Chai et al. [16], used magnetic properties for the separation of fluoride removal from drinking water. The calculated adsorption capacity for fluoride of the sulfate-doped Fe3O4/Al2O3 was 70.4 mg/g at pH ¼ 7.0. Lv et al. [17] have shown that the alumina/chitosan composite displayed a highest fluoride removal capacity of 3809 mg/kg versus

M.K. Adak et al. / Journal of Alloys and Compounds 719 (2017) 460e469

alumina and chitosan alone (52 mg/kg). Rare earth elements containing materials are considered to be good absorber due to their strong affinity towards fluoride [18]. The metal ions having very high oxidation states and at the same time having small size were considered as hard acids and due to the fact that they had strong affinity for the hard base fluoride ions. Therefore, metals such as Al (III), Fe (III), Cu(II), Y (III), La (III), Ce (IV), and Zr (IV) as oxides, hydroxides and carbonates were of great importance for their fluoride capacities [18]. Wasay et al. [19] had enhanced the efficiency of activated alumina by impregnation of La (III)- and Y (III)-to alumina. Alumina impregnated with lanthanum hydroxide was found to enhance the adsorption capacity 0.350 mM/g while that of original alumina was in the range of 0.170e0.190 mM/g. The adsorption capability of this bimetallic oxide was considerably high from their individual oxide materials. This was due to the fact that a new chemical bond was formed between the two metals through the oxygen which increased the amount of hydroxyl groups on the adsorbent surface and adsorption capacity of the adsorbents. However, their adsorption efficiencies were still not good enough and regeneration of the adsorbents were found be frequently needed [20]. For this purpose an attempt was taken to synthesize trimetallic oxides. Wu et al. [18] have developed a Fe(II), Al(III) and Ce(IV) trimetallic oxide using co-precipitation technique having molar ratio 1:4:1 at high pH and attained a high adsorption capacity of 178 mg/ g was acquired at pH ¼ 7.0, which was highly competitive compared to other reported adsorbents [21]. Chen et al. [22] have synthesized Fe-Al-Ce trimetallic hydroxide adsorbent which was obtained by spraying the suspension of Fe-Al-Ce nano-adsorbent onto the glass beads with acrylic-styrene copolymer latex. MgeAleFe hydrotalcite showed the largest adsorption capacity of 14 mg/g at pH 6 with adsorbent dosage of 0.2 g/L [23]. In this communication we report synthesis, characterization and fluoride absorption efficiency of a new Al-Fe-La trimetallic (1:1:1) oxide sorbent. The combination of these three metallic oxides is very much thoughtful. Though aluminium was found to be an efficient element for fluoride adsorption it has a serious drawback due to the fact that the high dose of Al exposure causes either the developing or the speeding up of the Alzheimer syndrome in human beings [24]. Though the Fe(III) had high affinity towards fluoride ions and iron oxides were widely distributed in the environment some researches has shown that the removal efficiency of Fe-based materials was not ideal [25] and its Langmuir adsorption capacity is low at about 16.5 mg/g [26]. Considering the above two scenario, Al was partly substituted by the ferric ions in order to reduce the aluminium content in the oxide. The high electronegativity and small size of the rare earths lanthanum(III) made it suitable for the fabrication of fluoride selective sorbent and therefore it was employed for the combination with the Al(III) and Fe(III) keeping the adsorption efficiency as much as possible. There were several reasons for selecting Al and Fe for this purpose. Amorphous iron and aluminum oxides exhibit high specific surface areas and high reactive nature of their surface provided high capacity to retain heavy metal contaminants. The main advantage is that the adsorbents can be reused and easily separated from the solution by external magnet, thus Fe was used for this purpose [23]. Both the iron and aluminium oxides had affinity towards arsenate as well as fluoride made it useful for taking in our purpose. After thorough literature review and to the best of our knowledge no reports are found on the trimetallic Al-Fe-La oxide which had been used for the removal of fluoride from drinking water. In this study we report the synthesis of nanosized trimetallic Al-Fe-La oxide, its materials characterizations through XRD, FTIR, surface area measurements etc. followed by removal of fluoride by adsorption process varying time of adsorption, adsorbent dose, and

461

different fluoride concentration etc. The aim of this work is to increase the fluoride adsorption capacity with high adsorption isotherm. Reuse of this adsorbent is also a huge purpose of this work. 2. Experimental 2.1. Materials required All reagents were of taken analytical grade and for solution preparation a double distilled water was used in all experiments. The chemicals required were NaF (Merck Specialties Private Limited 99%), La2O3 (E. Marck India Ltd. 99.5%), Al(NO3)3 (E. Marck India Ltd. 99.5%) and Fe(NO3)3$9H2O (E. Merck India Ltd. 99.0%), HNO3 (65%, Merck, India; GR grade), and triethanolamine (abbreviated as TEA) (Merck Specialties Private Limited 97%). 2.2. Preparation of standard fluoride solution A stock solution of 1000 mg/L fluoride was prepared by dissolving appropriate quantity of sodium fluoride in double-distilled water in a 1000 ml volumetric flask. Now all the desired fluoride solutions required for the batch experiment were prepared from the freshly prepared stock solution. 2.3. Preparation of the fluoride solutions used for the analysis From the freshly prepared stock solution, mentioned in sec.2.2, 3 ppm, 5 ppm, 10 ppm fluoride solutions were prepared for a final volume 100 ml each. 2.4. Synthesis of Fe-Al-La (1:1:1) oxide nano-adsorbent material 2.4.1. Preparation of 0.05 M [La(NO3)3] solution Appropriate amount of lanthanum oxide (La2O3) was added pinch by pinch with continuous stirring to a hot 400 ml 6 N nitric acid solution so that the oxide would be completely converted to its nitrate. The clear solution was then cooled and transferred to a 1000 ml volumetric flask and the volume was filled by doubledistilled water. 2.4.2. Preparation of 0.05 M [Al(NO3)3] solution Appropriate amount of aluminium nitrate salt was dissolved in minimum quantity of double-distilled water in a 1000 ml volumetric flask and the volume was filled up by the double-distilled water. 2.4.3. Preparation of 0.05 M [Fe(NO3)3] solution Appropriate amount of ferric nitrate salt was dissolved in minimum quantity of double-distilled water in a 1000 ml volumetric flask and the volume was filled up by the double-distilled water. The nano sized Al-Fe-La oxide was synthesized by pyrolysis method at a constant pH. A solution containing metal nitrates of Al: Fe: La in 1:1:1 M ratio was prepared in a clean and dry 500 ml beaker. The precursor solution was prepared after adding TEA to the resultant mixture solution. TEA was added in excess (3 mol ratio with respect to total mole of metal ion taken) which acted as a chelating agent, and thus avoid any chances of precipitation. The pH of the solution mixture was maintained at 3e4 using HNO3. The clear solution of metal nitrates was evaporated to dryness on a hot plate at ~180 C. Continuous heating of the solution causes foaming and puffing. The evaporation of nitrate ions provides an in-situ oxidizing environment for the decomposition of the hydroxyl groups of TEA to carboxylic acids. After complete dehydration a voluminous, organic

462


Fig. 1. Flow chart for the synthesis of nanosized Al-Fe-La (1:1:1) oxide.

based, chocolate brown fluffy powder was left i.e. precursor powder. Well ground precursor powder was calcined at 800 C for 4 h. A gray colored powder was obtained after calcinations. The details stepwise synthesis process is given in the flow chart in Fig. 1. 2.5. Adsorption studies All the adsorption experiments were performed at room temperature (300 K). The fluoride solutions of concentrations 3 ppm, 5 ppm and 10 ppm are taken in 125 ml plastic bottles containing 100 ml fluoride solutions in each. Then the batch experiments were

carried out in four batches representing the time of adsorption taking 15 min, 30 min, 45 min and 60 min. In each batch 0.05, 0.1, 0.2, 0.3 g of synthesized nano adsorbents were taken in 100 ml of each fluoride solution. The total sorption process involved five steps as maintained in the flow chart diagram in Fig. 2. In step e I, the adsorbents having definite quantities mentioned above were mixed with the different concentration of fluoride solutions. The adsorbent containing fluoride solutions were then shaken in an orbital shaker (Paragon RPM-0249 TXT-7203, India) with 158 spm for the time mentioned for each batch as shown in Fig. 2. Then the solutions were then

Fig. 2. Experimental procedure for the sorption studies and reuse of adsorbent.


centrifuged at 2600 rpm for 10 min and the clear upper solutions were collected by decantation and filtration in plastic beaker (Fig. 2, step II) and stored in plastic bottles. The presence of F in filtered solution was tested by the indication of decolourization of yellow colored Zr-alizarin red S solution as mentioned in Fig. 2 step III. A double beam UVevisible spectrophotometer (Perkin Elmer, Lambda 35 UVeVis spectrophotometer) was used for spectrophotometric determination of fluoride at 427 nm wavelength using Zirconium- Alizarin S complex solution as shown in Fig. 2, step IV. The Zirconium- Alizarin S complex solution was prepared by mixing the dye (Alizarin S) with zirconyl nitrate mixed with H2SO4 and HCl. The fluoride ions reacted with the yellow colored complex solution forming ZrF2 which is more stable compared to 6 Zirconium-Alizarin S and also colourless as mentioned in Eq. (1) and (2). Thus by forming ZrF2 6 , the intensity of the yellow color of zirconium-Alizarin S decreases. The reaction involved is as follows:

Alizarin red S þ ZrðNO3 Þ4 /Zr Alizarin red S ðyellow colourÞ: (1) Zr Alizarin S þ F /ZrF2 6 ðcolourlessÞ þ Alizarin red S:

(2)

The color intensity was measured by measuring the absorbance of the solution spectrophotometrically. From the absorbance value the concentration of fluoride solution was determined. All the experiments were done at room temperature (30 C). The mechanism of adsorption of fluoride on the metal oxide is modelled as a twostep ligand-exchange reaction:

≡MOHðSÞ þ Hþ 4≡MOHþ 2

(3)

≡MOHþ 2 þ F 4MF þ H2 O:

(4)

463

≡MF þ H3 Oþ 4≡MOHþ 2 þ HF leaching out of F

(7)

≡MOHþ 2 þ OH 4≡MOHðsÞ þ H2 O ðNeutralizationÞ

(8)

where MF is the surface occupied site, MOHþ 2 represents the acidic surface site; MOH(s) represents a surface hydroxyl group which is neutral. 3. Results and discussions 3.1. Powder characterization The powder was characterized by X-ray diffraction (Model PW1710, Philips Research laboratories), Cu Ka radiation (wave length of 1.5406 Å) was used at a scanning speed of 2 2q/min, operated at 40 kV and 20 mA. The XRD data were indexed with standard JCPDS data files. Transmission Electron Micrograph (TEM) (Hitachi H-600) was used to study the topography of the powdered sample. Elemental composition was determined by Energy Dispersive X-ray (EDX) analysis in vacuum, in the specimen chamber of an EDX coupled with Scanning Electron Microscope (SEM) (Model JFM 5800 Jeol, Tokyo, Japan). Scanning electron micrographs were recorded using SEM. The surface morphology of the adsorbent was observed by scanning electron microscope. Fourier transform-infrared spectroscopy (FT-IR) was recorded in the range 4000e500 cm1 in KBr (Merck Spectra grade, Merck India Private Limited) phase containing ~1% sample using Perkin Elmer Spectrum (Model No. L1600300). The specific surface area (BET) and BarretteJoynereHalenda (BJH) pore size of the sample was determined by the N2 adsorption/desorption method at liquid N2 temperature of 77.350 K using Quantasorb using Quantachrome Instruments version 10.01(Nova Station B).

This, in combination gives

≡MOHðSÞ þ Hþ þ F 4MF þ H2 O:

(5)

This two step mechanism is suitable at acidic pH, at pH < 6 fluoride ions are mostly adsorbed by following mechanism,

≡MOHðSÞ þ F 4MF þ OH

(6)

3.1.1. XRD study Fig. 3 shows the diffraction pattern of the synthesized Al-Fe-La oxide calcined at 800 C for 4 h. The intense peaks are observed at 2q values of 21.983 (104), 22.975 (012), 25.652 (11-3), 27.883 (100), 29.618 (002), 32.491 (104), 35.815 (114), 40.029 (102), 40.624 (113), 46.574 (306), 46.921 (110), 48.705 (024), 52.969 (116), 54.952 (112,201), 57.778 (018), 58.472 (004), 61.844 (202),

where M-represents metal ion (Fe3þ or Al3þ or La3þ), MOH(s) represents a surface hydroxyl group and MF, a surface site occupied by a fluoride ion.

2.6. Reuse of adsorbent Effective reuse of adsorbent materials is highly required to minimize the cost and hence its sustainable utility in continuous batch adsorption processes. For the desorption purpose as mentioned in Fig. 2 step V, the fluoride adsorbed nano material, after the first adsorption process, was collected from the fluoride solution into a clean and dry plastic beaker and then it was treated with 0.1 M HCl solution maintaining the pH of the solution between 3 and 4 with continuous shaking for 30 min at 100 spm in a shaker, so that, the adsorbed fluoride could be effectively leached out by the acid in the form of HF. Then the adsorbent was repeatedly treated with double-distilled water and then the acidic adsorbent was neutralized by the 0.1 M NaOH solution. The neutralized adsorbent was then washed with distilled water and got dried at 180 C. A probable mechanism was shown below, the way in which the desorption occurs by this method stated above,

Fig. 3. The X-ray diffraction pattern of the Al-Fe-La (1:1:1) oxide.

464


68.487 (119) were found to be the characteristics of the Al-Fe-La (1:1:1) oxide [27,28]. The planes corresponding to the peaks were indexed with respect to the respective metal oxides are indicated by different color and shape as shown in Fig. 3. The X-ray structure revealed the amorphousness of the sample which was actually purposefully done to get the high surface area so that adsorption process was expected to be high. In order to minimize the growth of the crystallite, the heating rate and calcinations time were carefully monitored. The Al2O3, Fe2O3, and La2O3 crystallized in monoclinic phase with a crystal orientation at (104), (012), (110), and (004) plane. The intense peaks appeared at the 2q values 22.975 (012), 32.491 (104), 46.921 (110), and 48.705 (024) at which the three monoclinic Al2O3, Fe2O3, and La2O3 phases were present which was indicative of the formation of homogeneous phase at that planes. The crystallite size was calculated using Scherrer's formula [29]

D¼

K:l b$cosðqÞ

(9)

where D is the average crystallite size of the particle, l is the wavelength of the radiation, b is the full width at half maximum (FWHM) of the peak, q is the Bragg's angle. The calculated average crystallite size was obtained for the synthesized sample Al-Fe-La oxide was 13 nm. The lattice parameters were calculated from the following equation,

1 ¼ d

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi h2 k2 l2 þ þ a2 b2 c2

3.1.2.2. SEM & EDS study. Scanning electron microscope was used to show the surface morphology of trimetallic oxide before and after the fluoride adsorption at fluoride free condition shown in Fig. 5 (a) and 5 (b). The agglomeration of the particles was found throughout the dispersed grains. The surface of the oxide was very coarse and showed porous structure that was suitable for good adsorption capability of fluoride ions. The corresponding energy dispersive spectrum (EDS) in Fig. 5 (c) shows the signals generated from elements present at surface of the adsorbent material. The signals showed the presence of elements such as Al, Fe, La and O in the material. The approximate atomic weight percentage ratio for O:Al:Fe:La is 6:1:1:1. Here the extra oxygen atom is for carbonate impurity which can be confirmed from EDS study (also discussed in IR study). The EDS study of the adsorbent after adsorption experiments at fluoride free condition confirmed the absence of any fluoride. The average grain size of Al-Fe-La oxide was calculated using Image Tools software and was found to be 104 nm.

(10)

where h, k, l are the values of miller indices of a plane.

3.1.2. Microstructure 3.1.2.1. TEM study. Fig. 4(a) represents the bright field TEM of Al (III)-Fe (III)-La (III) trimetallic oxide calcined at 800 C for 4 h. The particles have mostly spherical morphology with uniform distribution of particle size shown in Fig. 4(b). The monoclinic structure was observable as the same also was detected from XRD study. In Fig. 4(a) the black spots indicate over focused particles and the white spots are under focused particle. The average particle diameter was determined using UTI image tool software (version 3.0) around 20 (±0.50) nm with 30% polydispersity.

3.1.3. Fourier transform infrared spectroscopy (FTIR) The FTIR spectra of Al-Fe-La oxide adsorbent before and after fluoride adsorption at pH 7 are given in Fig. 6. In Fig. 6(a) the FTIR spectra of the adsorbent before use showed the characteristic absorption bands at 568.74, 643.05, 1014.6, 1103.8, 1163.2, 1386.1, 1497.6, 1635, 2344.7, 2850, 2928, 3444.5 cm1. The stronger broadening band between 1000 cm1 and 500 cm1 was a characteristic absorption band of transition alumina [30]. The peak at 1014 cm1 corresponds to the Al-O stretching vibration [31]. In addition to the alumina bands there were bands due to chemisorbed and adsorbed species at the surface. The very large absorption band centred at 3444.5 cm1 resulted from the vibration bands of hydroxyl groups, isolated OH groups, and O-H stretching vibration of water molecules, due to presence of moisture in the sample [27]. The band at 1635 cm1 was due to bending mode of molecular water. The strong absorption band at 643.05 cm1 was attributed to the La-O-H bending vibration [32] and medium strong band positions in the range of 1430 cm1 to 1560 cm1 are possibly due to stretching vibrations of ions. The observed vibration band at 586.74 cm1 might be assigned due to the Fe-O-Fe stretching vibration [33]. The sharp band at 1386.1 cm1 might be assigned to the bending mode of O-H [34]. The narrow absorption peak around 1103.8 cm1 is ascribed to the CO2 stretching frequency [35] and a small band at 1497.6 cm1 is attributed to O]C]O stretching vi bration of chemically adsorbed impurities of CO2, CO2 3 , HCO3 that are very difficult to remove, even at temperature of 800 C [36]. The

Fig. 4. (a) Transmission electron micrograph and (b) particle size distribution of Al (III)-Fe (III)-La (III) trimetallic oxide powder calcined at 800 C for 4 h.


465

Fig. 5. SEM images of nano Aluminium-Iron-Lanthanum oxide (a) before and (b) after adsorption study at fluoride free condition and (c) EDS study of nano Aluminium-IronLanthanum oxide.

In Fig. 6 (b) the infrared spectra of Al-Fe-La oxide after fluoride adsorption was exhibited the characteristic bands which are visibly affected after fluoride removal. After fluoride adsorption the band at the 643.05 cm1 was shifted to higher frequency 646.6 cm1 as well as the downshift of the band at 568.74 cm1 to the 561.24 cm1 was indicative to the surface adsorption of fluoride.

3.1.4. BET study The surface area, pore diameter, and pore volume of the adsorbent synthesized was determined by the BET study using by a quantachrome novawin e data acquisition and reduction instrument quantachrome instruments version 10.01(Nova Station B). In this process nitrogen gas was adsorbed on the adsorbent surface at liquid N2 temperature of 77.350 K and the surface area of the adsorbent has calculated from the volume of the gas required to form unimolecular adsorbed layer on the solid. Prior to the adsorption measurements all samples were out gassed using nitrogen flow at 200 C for 18 h. The related Brunauer-Emmett-Teller equation is as follows, Fig. 6. (a) Infrared spectra of Al-Fa-La (1:1:1) oxide (before use) (b) Infrared spectra of Al-Fa-La (1:1:1) oxide (after use).

band between 2850 and 2928 cm1 corresponds to H-C stretching [30].

P WðP PÞ

¼

1 C1 P þ Wm C Wm C P

(11)

where W is the volume of the gas adsorbed under pressure P, P is the saturated vapor pressure at the same temperature, Wm is the

466


volume required for the monolayer coverage and C is a constant for ðE1 E2 Þ a given adsorbate such that C ¼ e RT , where E1 and E2 are the heat of adsorption in the first layer and heat of liquification of the gas. When W P1 1 was plotted against PP a liner BET plot was obtained ½ ðPÞ and volume of the gas required for monolayer coverage Wm was C1 and intercept 1 (not shown here). obtained from the slope W Wm C mC From the BET study it was found that the material was porous in nature which was purposefully done to attain the high surface area as well as high adsorption capability to remove maximum amount fluoride from fluoride containing drinking water (also discussed in sec. 3.1.1 and 3.1.2). The total surface area was found to be 22.14 m2/ g. 3.1.5. BJH desorption summary From the BarretteJoynereHalenda (BJH) study surface area, pore volume and pore diameter was determined. The samples were degassed at 200 C for 18 h to remove any adsorbed water or trapped gases. In this process surface area was measured using the nitrogen adsorption/desorption isotherm at the liquid nitrogen temperature of 77.350 K where nitrogen gas was used as an adsorbent. The material was found to be porous and the calculated total pore volume was 1.50 102 cc/g for pore diameter 3.324 nm. 3.2. Effect of contact time on the adsorption of fluoride One of the most important parameters for determination of the sorption potential of an adsorbent material is the variation of contact time using a fixed initial fluoride concentration of 3, 5 and 10 ppm and adsorbent dose (g/100 ml) within the time period for 15e60 min. As shown in Fig. 7(aed) the overall trend was an initial decrease in the % of fluoride removed and then a sudden increase in the adsorption of fluoride with the increase of the time period and attained a maximum above 99.0% fluoride removal within 60 min 120

120

0.05g/100 ml adsorbent dose

(a)

110

contact time at pH 7.0 when the adsorption dose was taken 0.3 g/ 100 ml. The observed behavior might be due to the high surface area and porous structure provided more number of adsorption sites with the increase of contact time. The decrease in the fluoride adsorption initially could be explained by the two factors, one of them was attributed to the fact that on adsorption, the active surface area of the adsorbed adsorbent decreased due to diffusion of adsorbed ions into the pores of the adsorbent and so the active surface area decreased [2] and a second factor raised when the F ions were adsorbed on the pore, it repelled the adsorbed fluoride on the surface and released the adsorbed fluorides [9]. Again porous structural features could also facilitate easy removal of fluoride as they favour the adsorption process [37]. The whole process is presented pictorially in Fig. 8. At initial time period, there is a significant amount of fluoride adsorption as represented in Fig. 8 stage I. At stage II, after a certain interval, dissolution starts due to repulsion among the F ions. In Fig. 7 (c) it was shown that the % of fluoride removed was decreased from the time period of 45 mine60 min. This fact was explained due to the fact that desorption was occurred after 45 min as represented in Fig. 8, stage - II. From the above study it was revealed that when the initial fluoride concentration was taken 3, 5, 10 ppm a very good fluoride adsorption was obtained at 15 min for 0.05, 0.1, 0.2, and 0.3 g/ 100 ml adsorbent doses removed fluoride up to ~99% for 0.3 g/ 100 ml dose for 3 ppm; ~99.2% for 0.3 g/100 ml dose for 5 ppm and ~99.6% for 0.3 g/100 ml dose for 10 ppm, but with the increase of the contact time up to 30 min adsorption was lowered for all the cases compared to the 15 min values. Thus the 30 min contact time was not good for fluoride removal in this study. Again an appreciable fluoride adsorption was occurred for 45 min maximum adsorption was ~99.1% for 0.2 g/100 ml dose for 3 ppm, ~99.1 for 0.2 g/100 ml dose for 5 ppm, ~98 for 0.2 g/100 ml adsorbent dose for 10 ppm. For 0.3 g/100 ml adsorption doses and contact time of

0.1g/100 ml adsorbent dose

100

90 80 70 60 50

3 ppm 5 ppm 10 ppm

40 30 10

20

30

40

50

% fluoride removed

% fluoride removed

(b)

110

100

90 80 70 60 50

3 ppm 5 ppm 10 ppm

40 30 10

60

20

Time (min) 120 110

0.2 g/100 ml adsorbent dose

110

50

60

0.3 g/100 ml adsorbent dose

(d)

100

90 80 70 60 50

3 ppm 5 ppm 10 ppm

40 30 10

20

30

40

Time (min)

50

60

% fluoride removed

100

% fluoride removed

40

Time (min) 120

(c)

30

90 80 70 60 50

3 ppm 5 ppm 10 ppm

40 30 10

20

30

40

50

60

Time (min)

Fig. 7. % of Fluoride removal vs. contact time taking the adsorbent dose (a) 0.05 g/100 ml, (b) 0.1 g/100 ml, (c) 0.2 g/100 ml, (d) 0.3 g/100 ml.


467

Fig. 8. Schematic diagram of Fluoride adsorption at different time interval.

45 min % fluoride removal decreased significantly. Thus, 0.3 g/ 100 ml adsorbent was not good enough for fluoride removal. For contact time of 60 min the extent of fluoride removal increases significantly which was implied that it was very good for 0.05, 0.1, 0.2, 0.3 g/100 ml adsorption doses. So, 60 min time of adsorption could be considered as the equilibrium condition where maximum adsorbed ion was observed as represented in Fig. 8 stage III. 3.3. Effect of adsorption dose on extent of fluoride removal The absorption efficiency of the adsorbent was extremely dependent upon the adsorption dose. The adsorption dose was varied within the range of 0.05e0.3 g/100 ml of fluoride solution

with initial concentration of 3, 5, and 10 ppm at neutral pH (¼7) at an ambient temperature of 300 K with a selected contact time period of 15, 30, 45, 60 min. Fig. 9(aed) illustrate that the % of fluoride removal was linearly increases with increasing adsorbent dose and attained maximum removal of fluoride about 99% at 0.3 g/100 ml adsorbent dose, pH 7.0. This could be explained due to the availability of large surface area and hence adsorption sites with increasing the adsorbent dose (g/100 ml). Availability of specific surface area and micro pore volume had played a vital role for surface adsorption process [38]. From the Table 1 it was observed that the % of fluoride removed became almost constant between the time period of 45 min and 60 min indicating the saturation of adsorption sites.

120

120 110

(a)

15 min

110

80 70 60 50

3 ppm 5 ppm 10 ppm

40 30 0.05

0.10

0.15

0.20

0.25

% fluoride removed

90

90 80 70 60 50

3 ppm 5 ppm 10 ppm

40 30

0.30

0.05

Adsorbent dose (g/ml)

(c)

0.15

0.20

0.25

0.30

120

45 min

110

100

(d)

60 min

100

90 80 70 60 50

3 ppm 5 ppm 10 ppm

40 30 0.05

0.10

0.15

0.20

0.25


0.30

% fluoride removed

% fluoride removed

0.10


120 110

30 min

100

100

% fluoride removed

(b)

90 80 70 60 50

3 ppm 5 ppm 10 ppm

40 30 0.05

0.10

0.15

0.20

0.25

0.30


Fig. 9. % of Fluoride removal vs. Adsorbent dose (g/100 ml) taking the contact times (a) 15 min, (b) 30 min, (c) 45 min and (d) 60 min.

468


Table 1 Effect of fluoride removal as a function of contact time and adsorption doses. Time (min)

3 ppm

15

30

45

60

5 ppm % of fluoride removal

Adsorbent dose (g)

% of fluoride removal

Adsorbent dose(g)

% of fluoride removal

0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3

84.4 80.1 94.1 99.0 41.3 78.3 90.0 98.9 68.5 92.1 98.6 98.8 76.1 94.3 99.2 99.9

0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3

70.9 60.9 85.4 99.2 31.1 39.7 73.8 91.6 42.7 46.8 99.1 99.1 72.2 99.7 99.4 99.3

0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3 0.05 0.1 0.2 0.3

99.0 97.2 99.9 99.6 72.8 93.7 96.0 99.2 73.8 96.7 98.4 99.1 99.9 99.8 97.8 99.8

The dissemination of active sites could be explained as due to overlapping of the active sites at higher dosage there is a decrease in the effective surface area resulting in the decrease in fluoride adsorption capacity [39] as occurred in the case of 45 min contact time at 0.3 g/100 ml adsorption dose shown in Fig. 9 (d). In Fig. 7 (d) it was observed that the fluoride removal became maximum of ~99% having 0.2 g/100 ml adsorbent dose with contact time of 45 min and was decreased with the further increase in the adsorbent dose up to 0.3 g/100 ml at pH 7.0. This fact also could be attributed by the fact stated above similar to that of the effect of contact time [37,39]. 4. Effect of pH The most important parameter that controls the adsorption of ions on the oxide surface is the pH of the aqueous solution. The probable mechanism given above in Eq. (1)e(4) [section 2.5]. Most of the work was done at pH 7.0 as was observed from the literature survey and it was the optimum pH value for the maximum fluoride removal [9,11,14e16,20]. So in this study all experiments were conducted at neutral pH value. The effect of pH on the percentage removal of Fluoride were studied on 3 ppm fluoride solution using the adsorbent dose 0.3 g/ 100 ml for 15 min contact time as this set was seemed to obtain highest fluoride removal capacity shown in Fig. 10. The pH of the

100 98

% fluoride removed

10 ppm

Adsorbent dose (g/100 ml)

96 94

Fluoride containing water was adjusted using 0.01 (M) NaOH and 0.01 (M) HCl solutions. The results showed that adsorption of fluoride increase with increasing the solution pH and a maximum adsorption was done at pH 6.7 i.e. near neutral pH range. At low pH the F ions form HF resulting availability of fluoride ions decrease for adsorption and at high pH (>7.0) a competition for adsorption by fluoride and hydroxyl ions due to similarity in charge and ionic radius [3,5]. 5. Conclusions In this study nano sized Al (III)-Fe(III)-La(III) oxide was used for the removal of fluoride from drinking water very effectively. Nanocrystalline Al-Fe-La oxide was prepared chemically and its monoclinic structure was established both from XRD and TEM studies. The crystallite size, grain size and the particle size were calculated from XRD, SEM and TEM studies and were found to be 13, 104 and 20 nm respectively. From the BET and BJH analysis, the surface area, pore volume and pore size distribution were calculated were found to be 22.14 m2/g, 0.015 cc/g, 3.324 nm respectively. It was observed that the maximum fluoride could be removed up to 99.8% at pH-7.0, contact time: 60 min, adsorbent dose: 0.3 g/100 ml implying the new adsorbent could be conventionally used for the removal of fluoride from drinking water sources. However 0.3 g/100 ml adsorbent dose for 15 min can also be used efficient with ~99% removal capacity at all level of fluoride concentration in drinking water. The % of fluoride removal was linearly increasing with increasing adsorbent dose and attained a maximum removal of fluoride about 99% at 0.3 g/100 ml adsorbent dose, pH 7.0. When the initial fluoride concentration increases keeping the adsorbent dose and the contact time fixed the % fluoride removal decreases. Further, the adsorbed fluoride could be effective removed by the use of 0.1 M NaOH or 0.1 M HCl, so that the adsorbent can sustainably be utilized for a number of cycles.

92

Acknowledgements

90 88 86 1

2

3

4

5

6

7

8

9

10

Authors would like to thank Department of Science and Technology, Government of West Bengal vide project sanction no. 674(Sanc)/ST/P/S&T/15G/5/2016 dated 09/11/2016 for financial support.

pH References Fig. 10. Variation of pH, conc. of fluoride: 3 ppm; time of contact: 60 min; adsorbent dose: 0.3 g/100 ml.

[1] J. Zhang, N. Chen, Z. Tang, Y. Yu, Q. Hu, C. Feng, A study of the mechanism of


[2]

[3]

[4]

[5]

[6] [7]

[8]

[9]

[10] [11]

[12] [13]

[14]

[15]

[16]

[17]

[18] [19]

[20]

fluoride adsorption from aqueous solutions onto Fe-impregnated chitosan, Phys. Chem. Chem. Phys. PCCP 17 (18) (2015) 12041e12050. Sushree Swarupa Tripathy, Jean-Luc Bersillon, Krishna Gopal, Removal of fluoride from drinking water by adsorption onto alum-impregnated activated alumina, Sep. Purif. Technol. 50 (3) (2006) 310e317. Venkataraman Sivasankar, Thiyagarajan Ramachandramoorthy, Darchen, Manganese dioxide improves the efficiency of earthenware in Andre fluoride removal from drinking water, Desalination 272 (1e3) (2011) 179e186. nchez, M. Solache-Ríos, V. Martínez-Miranda, C. Solís Morelos, J.J. García-Sa Removal of fluoride ions from drinking water and fluoride solutions by aluminum modified iron oxides in a column system, J. colloid interface Sci. 407 (2013) 410e415. M. Mohapatra, D. Hariprasad, L. Mohapatra, S. Anand, B.K. Mishra, Mg-doped nano ferrihydriteda new adsorbent for fluoride removal from aqueous solutions, Appl. Surf. Sci. 258 (10) (2012) 4228e4236. Meenakshi, R.C. Maheshwari, Fluoride in drinking water and its removal, J. Hazard. Mater. 137 (2006) 456e463. V. Ganvir, K. Das, Removal of fluoride from drinking water using aluminum hydroxide coated rice husk ash, J. Hazard. Mater. 185 (2e3) (2011) 1287e1294. C.Y. Hu, S.L. Lo, W.H. Kuan, Effects of co-existing anions on fluoride removal in electrocoagulation (EC) process using aluminum electrodes, Water Res. 37 (18) (2003) 4513e4523. Subhashini Ghorai, K.K. Pant, Equilibrium, kinetics and breakthrough studies for adsorption of fluoride on activated alumina, Sep. Purif. Technol. 42 (3) (2005) 265e271. R.H. McKee, W.S. Jhonston, Removal of fluoride from drinking water, Ind. Eng. Chem. 26 (1934) 849e850. R. Leyva Ramos, J. Ovalle-Turrubiartes, M.A. Sanchez-Castillo, Adsorption of fluoride from aqueous solution on aluminum-impregnated carbon, Carbon 37 (4) (1999) 609e617. H. Farrah, J. Slavek, W.F. Pickering, Fluoride interactions with hydrous aluminum oxides and alumina, Aust. J. Soil. Res. 25 (1987) 55e69. S.M. Maliyekkal, A.K. Sharma, L. Philip, Manganese-oxide-coated alumina: a promising sorbent for defluoridation of water, Water Res. 40 (19) (2006) 3497e3506. Shihabudheen M. Maliyekkal, Sanjay Shukla, Ligy Philip, Indumathi M. Nambi, Enhanced fluoride removal from drinking water by magnesia-amended activated alumina granules, Chem. Eng. J. 140 (1e3) (2008) 183e192. Amit Bansiwal, Pradnya Pillewan, Rajesh B. Biniwale, Sadhana S. Rayalu, Copper oxide incorporated mesoporous alumina for defluoridation of drinking water, Microporous Mesoporous Mater. 129 (1e2) (2010) 54e61. L. Chai, Y. Wang, N. Zhao, W. Yang, X. You, Sulfate-doped Fe3O4/Al2O3 nanoparticles as a novel adsorbent for fluoride removal from drinking water, Water Res. 47 (2013) 4040e4049. L. Lv, J. He, M. Wei, D.G. Evans, X. Duan, Factors influencing the removal of fluoride from aqueous solution by calcined MgeAleCO3 layered double hydroxides, J. Hazard. Mater. 133 (2006) 119e128. X. Wu, Y. Zhang, X. Dou, M. Yang, Fluoride removal performance of a novel FeAl-Ce trimetal oxide adsorbent, Chemosphere 69 (11) (2007) 1758e1764. Syed Abdul Wasay, Shuzo Tokunaga, Sang-Won Park, Removal of hazardous anions from aqueous solutions by La(lll)- and Y(lll)-impregnated alumina, Sep. Sci. Technol. 31 (10) (1996) 1501e1514. J. Das, B.S. Patra, N. Baliarsingh, K.M. Parida, Calcined MgeFeeCO3 LDH as an adsorbent for the removal of selenite, J. Colloid Interface Sci. 316 (2007)

469

216e223. [21] Bei Zhao, Yu Zhang, Xiaomin Dou, Xiaomei Wu, Min Yang, Granulation of FeeAleCe trimetal hydroxide as a fluoride adsorbent using the extrusion method, Chem. Eng. J. 185e186 (2012) 211e218. [22] L. Chen, T.eJ. Wang, H.eX. Wu, Y. Jin, Y. Zhang, X.eM. Dou, Optimization of a FeeAleCe nano-adsorbent granulation process that used spray coating in a fluidized bed for fluoride removal from drinking water, Powder Technol. 206 (2011) 291e296. [23] Wei Ma, Nannan Zhao, Gang Yang, Liyan Tian, Ren Wang, Removal of fluoride ions from aqueous solution by the calcination product of MgeAleFe hydrotalcite-like compound, Desalination 268 (1e3) (2011) 20e26. [24] J. Das, D. Das, G.P. Dash, D. Das, K. Parida, Studies on Mg/Fe hydrotalcite-like compound: removal of chromium (VI) from aqueous solution, Int. J. Environ. Stud. 61 (2004) 605e616. [25] F. Bruna, R. Celis, I. Pavlovic, C. Barriga, J. Cornejo, M.A. Ulibarri, Layered double hydroxides as adsorbents and carriers of the herbicide (4-chloro-2methylphenoxy)acetic acid (MCPA): systems MgeAl, MgeFe and MgeAleFe, J. Hazard. Mater. 168 (2e3) (2009) 1476e1481. [26] Soumen Dey, Saswati Goswami, Uday Chand Ghosh, Hydrous ferric oxide (HFO)da scavenger for fluoride from contaminated water, Water, Air, & Soil Pollut. 158 (1) (2004) 311e323. [27] M.R. Joya, J.B.-. Jaimez, J.B.- Ortega, Preparation and characterization of Fe2O3 nanoparticles, J. Phys. Conf. Ser. 466 (2013) 012004. [28] K.A. Matori, L.C. Wah, M. Hashim, I. Ismail, M.H. Zaid, Phase transformations of a-alumina made from waste aluminum via a precipitation technique, Int. J. Mol. Sci. 13 (12) (2012) 16812e16821. [29] A.L. Patterson, The scherrer formula for X-ray particle size determination, Phys. Rev. 56 (10) (1939) 978e982. [30] T.M.H. Costa, M.R. Gallas, E.V. Benvenutti, J.A.H. da Jornada, Study of nanocrystalline ç-Al2O3 produced by high-pressure compaction, J. Phys. Chem. B 103 (1999) 4278e4284. [31] K. Mach, B. Matyska, Infrared spectral analysis of aluminum triethyl, Collect. Czechoslov. Chem. Commun. 30 (8) (1965) 2841e2844. [32] G. Li, C. Li, Z. Xu, Z. Cheng, J. Lin, Facile synthesis, growth mechanism and luminescence properties of uniform La(OH)3 : Ho3þ/Yb3þ and La2O3 : Ho3þ/ Yb3þ nanorods, CrystEngComm 12 (2010) 4208e4216. [33] B. Zhao, Y. Wang, H. Guo, J. Wang, Y. He, Z. Jiao, M. Wu, Iron oxide (III) nanoparticles fabricated by electron beam irradiation method, Mater. Sci. Pol. 25 (2007) 1143e1148. [34] M. Aghazadeh, A.N. Golikand, M. Ghaemi, T. Yousefi, La2O3 nanoplates prepared by heat-treatment of electrochemically grown La(OH)3 nanocapsules from nitrate medium, J. Electrochem Soc. 158 (2011) E136eE141. [35] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Part B, sixth ed., Wiley, 2009, p. 153. ISBN 978-0-471-74493-1. [36] J.Y. Ying, J.B. Benziger, H. Gleiter, Photoacoustic infrared spectroscopy of nanoclustersAl2O3clusters and cluster-assembled solids, Phys. Rev. B 48 (3) (1993) 1830e1836. [37] Krishna Biswas, Kaushik Gupta, Arijit Goswami, Uday Chand Ghosh, Fluoride removal efficiency from aqueous solution by synthetic iron(III)ealuminum(III)echromium(III) ternary mixed oxide, Desalination 255 (1e3) (2010) 44e51. [38] V. Tomar, D. Kumar, A critical study on efficiency of different materials for fluoride removal from aqueous media, Chem. Cent. J. 7 (2013) 1e15. [39] S.K. Swain, Sulagna Mishra, Tanushree Patnaik, R.K. Patel, Usha Jha, R.K. Dey, Fluoride removal performance of a new hybrid sorbent of Zr(IV)eethylenediamine, Chem. Eng. J. 184 (2012) 72e81.