Al2O3 nanoparticles synthesized using various

Journal of Science: Advanced Materials and Devices xxx (2017) 1e10

Contents lists available at ScienceDirect

Journal of Science: Advanced Materials and Devices journal homepage: www.elsevier.com/locate/jsamd

Original Article

Al2O3 nanoparticles synthesized using various oxidizing agents: Defluoridation performance M. Changmai*, J.P. Priyesh, M.K. Purkait Department of Chemical Engineering, Indian Institute of Technology Guwahati, Guwahati, 781039, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 May 2017 Received in revised form 16 August 2017 Accepted 6 September 2017 Available online xxx

This study concerns the removal of fluoride using aluminium oxide nanoparticles synthesized in the presence of oxidizing agents H2SO4, KMnO4 and K2Cr2O7. The obtained nanoparticles were characterized using TGA, FESEM, EDX and XRD analysis. The almost constant weight loss was observed from the TGA data for the temperature range from 400 to 650 C. XRD analysis with and without oxidizing agents indicated the crystalline behaviour increased with increasing the temperature. Prepared Al2O3 nanoparticles exhibited a considerable potential for fluoride adsorption from aqueous medium in the concentration range of 2e8 mg/l. In this case, around 92% fluoride was adsorbed at pH ¼ 4.7. The equilibrium data were well fitted with Freundlich adsorption isotherm, whereas the adsorption kinetic data followed the pseudo second order model. © 2017 The Authors. Publishing services by Elsevier B.V. on behalf of Vietnam National University, Hanoi. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Keywords: Al2O3 Nanoparticles Characterization Fluoride Adsorption Regeneration

1. Introduction Fluoride is one of the major water polluting components, which is formed due to both natural and man-made reasons. It is highly reactive and is found naturally as CaF2. Fluoride ions infiltrate surface water and ground water mainly from soil leeching, precipitation, weathering of fluoride bearing rocks and human emissions. Although the minute quantity of fluoride is required for the formation of dental enamel and normal bone mineralization in the human body, the excessive intake leads to slow, progressive crippling scourge known as fluorosis. It was reported that fluorosis is prevalent in 17 states of India [1]. India is one of 21 nations with serious health problems due to consumption of fluoridecontaminated drinking water. The safe limit of fluoride in drinking water should be less than 1 mg/l (according to the Bureau of Indian Standards (BIS) [2]) or between 0.6 and 0.9 ppm (according to US standard [3]). However, there, the fluoride concentrations in drinking water generally varies from 1.5 to 39 ppm [4]. It is also estimated that more than 200 million people worldwide are

* Corresponding author. Fax: þ91 361 2582291. E-mail address: [email protected] (M. Changmai). Peer review under responsibility of Vietnam National University, Hanoi.

dependent on drinking water with the fluoride content exceeding WHO guideline. The concentration and the duration of continuous intake determine whether the impact of fluoride in drinking water can be beneficial or detrimental to mankind. Fluoride generally gets deposited in the joints of pelvic, knee, neck and shoulder bones and makes it difficult for a person to move or walk. It may even lead to a rare bone cancer, spondylitis or arthritis osteo-sarcoma and finally spine, major joints, muscles and nervous system may get damaged [5]. Hence, it has become a necessity to reduce the fluoride content to the safe limit. Various techniques have been introduced to solve this proplem, including both chemical and physical methods. Chemical methods include electro-coagulation processes and coagulation precipitation, whereas the physical methods include membrane separation and adsorption technique, mainly related to nano-filtration and reverse osmosis. The adsorption using a nanoadsorbent such as schwertmannite with adsorption capacity of 17.24 mg/g is one of such technique, which can be used for defluoridation [6,7]. Recent work on defluoridation using an iron oxide hydroxide suggested a considerable potential for fluoride removal of 11.3 mg/g from aqueous medium [8]. Water remediation using efficient nano-adsorbent is getting greater importance since recently. A wide variety of nano-adsorbents have been produced till date for the removal of fluoride from water [9e11]. Alumina supported metal oxide nanoparticles are the most extensively used

http://dx.doi.org/10.1016/j.jsamd.2017.09.001 2468-2179/© 2017 The Authors. Publishing services by Elsevier B.V. on behalf of Vietnam National University, Hanoi. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Please cite this article in press as: M. Changmai, et al., Al2O3 nanoparticles synthesized using various oxidizing agents: Defluoridation performance, Journal of Science: Advanced Materials and Devices (2017), http://dx.doi.org/10.1016/j.jsamd.2017.09.001

2

M. Changmai et al. / Journal of Science: Advanced Materials and Devices xxx (2017) 1e10

nano-adsorbent for the defluoridation process [12e14]. Metal oxide nanoparticles are quite promising in the fields of adsorption for their large surface area and porous structure along with short diffusion route [15]. Because of comparatively large surface areas, it is likely that nanosized adsorbents with strong affinity towards fluoride can be a useful tool in enhancing the adsorption capacity in drinking water treatment. However, due to their small particle size, the isolation of nanosized adsorbents from matrices is difficult for practical applications. Al2O3 nanoparticles were found to have a high affinity towards fluoride ions. Moreover, because of their costefficiency for large scale defluoridation processes, they have much more preferable advantages over other adsorbents [16e18]. In the present work, a nano-alumina structure is prepared by the chemical treatment on the surface of an aluminium foil in the range of 10 nme150 nm. The formed nanoparticles exhibited a considerable prospective to be utilized in fluoride removal. To improve the oxidation efficiency, oxidation agents such as H2SO4, KMnO4 and K2Cr2O7 in mild concentration (10 mM) were treated on the surface before annealing. It was found that each oxidizing agent resulted in nanoparticles with different morphologies. With H2SO4 as oxidizing agent, flower like structures with nano-chains were observed. K2Cr2O7 formed plate like structures and KMnO4 formed uniform nanoparticles on the foil surface without any chain formation. Fluoride adsorption efficiency was determined in a batch mode. The adsorption studies were carried out considering various parameters such as contact time, initial fluoride concentration, adsorbent mass, pH, stirring speed and the effect of other ions.

2.3. Adsorption experiment Batch adsorption was carried out to remove excess fluoride from the water using Al2O3 nanostructures. Al2O3 nanostructures prepared without oxidizing agents at 450 C and with oxidizing agents H2SO4, K2Cr2O7 and KMnO4 at 550 C were used as adsorbent. The initial concentrations of fluoride used were 2 mg/l, 4 mg/l and 8 mg/l with an adsorbent dose of 4 g/l. The removal efficiency R was calculated by

R ¼ ðCo Ce Þ=Co 100%

(1)

2.4. Characterization techniques Thermo gravimetric analysis (TGA) (Mettler Toledo) with an N2 gas flow rate of 40 ml/min and a purge gas flow rate of 20 ml/min was carried out to determine the oxidation properties of the Aluminium foil. Field emission scanning electron microscope (FESEM by LEO 1430 vp at 3.00e5.00 KV) was used to examine the morphological structure and to measure the average particle size. The FESEM analysis of the aluminium foil treated without any oxidizing agents revealed the formation of nanostructures. A wide angle X-Ray diffractometer (Bruker D8) was used to study the crystallite structure of the nanoparticles. Energy-dispersive X-ray spectroscopy (EDX) technique was applied for the elemental analysis or chemical characterization of the sample and the data have confirmed the formation of Al2O3 nanoparticles. It is found that sulphur impurities remained in the sample treated with H2SO4 at 550 C.

2. Materials and methods 3. Results and discussion

2.1. Materials Aluminium foils of 2 mm thickness were obtained from Hindalco Ltd., Sodium fluoride (NaF) from Titan biotech Ltd., India, HCl and NaOH from Merck, India. Other chemicals, such as acetone, ethanol, H2SO4, KMnO4 (99% purity) and K2Cr2O7 (99.9% purity) were obtained from Merck, India. All the chemicals were of analytical grade and used without further purification. 2.2. Synthesis and characterization of Al2O3 nanostructure Aluminium foils of 2.5 cm 2.5 cm were washed with acetone and ethanol to remove the organic impurities from the surface. The foils were then treated with 1 M HCl solution to eliminate the surface oxidation layer formed already on the surface by the reaction with air and then finally washed with deionized water to remove HCl from the surface. The foil was annealed in air for 3 h to produce Al2O3 nanostructures (see Fig. 1). The experiment was repeated by changing the annealing temperature as 400 C, 450 C, 500 C and 550 C. To improve the oxidation efficiency, oxidation agents, such as H2SO4, KMnO4 and K2Cr2O7 in mild concentration were treated on the surface of the foils before annealing [19e22].

Aluminium Foil (2.5×2.5×10-4 m2)

Washed with acetone and ethanol to remove organic impurities

3.1. Characterization of Al2O3 nanostructures To determine the oxidizing characteristics of the Al foil, thermogravimetric analysis (TGA) was carried out. Oxidation of the pure aluminium foil at elevated temperatures in air was done in the temperature range of 28e650 C. The highest temperature was restricted to 650 C, because the melting point of Al foil is 665 C. The heating rate was set at 20 C/min. The percentage weight loss in the temperature range from 28 to 120 C was of 8e9% due to the moisture removal. The weight loss then linearly increases to reach further 16% when the temperature was increased to 400 C. From 400 to 650 C the weight of the sample remained almost unchanged (see Fig. 2). For the preparation of the nanostructures, optimum conditions are required. If the oxidation rate is very high, the foil will oxidize completely and a bulk material will form. On the other hand, if the oxidation rates are very low, only surface oxidation will occur and an oxidation layer will form on the surface. From the data it was observed that in the temperature range of 400e650 C, the weight loss almost remains constant confirming the attainment of thermal stability. So, the experimental temperature range was set to be 400e650 C.

Treated with HCl to remove surface oxidation layer

Treated with oxidizing agents to improve oxidation

Annealed at different temperature

Fig. 1. Flow chart for the preparation of Al2O3 nanoparticles.



105 Weight loss due to moisture removal +burning of volatile impurities

100

Mass (%)

95

Reduced weight loss due to surface oxidation

90 85

Weight loss is constant

80 75 70

0

100

200

300

400

500

600

700

o

Temperature ( C) Fig. 2. Thermogravimetric analysis (TGA) plot of Al2O3 foil.

The morphology and the size of Al2O3 nanostructures after annealing at 400 C were examined by FESEM which shows only surface oxidation without the formation of nanostructures (see Fig. 3a). The surface was entirely covered by an oxidation layer

3

which might be considered as onset of the initialization of the nanoparticle formation with the symmetry of the oxidation layer being round structures joined together. The Al foil treated without oxidizing agent at 450 C (shown in Fig. 3b), clearly indicates the formation of nanostructures. The nanoparticle size ranges about 10e80 nm and the average chain length is as large as about 1200 nm. At 500 C (Fig. 3c) and 550 C (Fig. 3d), not only the surface but the entire aluminium foil was oxidized. At 500 C, along with the bulk structures, nano-chain like structures are also visible with an average particle size of about 100 nm. The oxidation rate was adjusted by adding oxidizing agents H2SO4, KMnO4 and K2Cr2O7. FESEM results show the formation of various types of nanoparticles when the Al foil was treated with different oxidizing agents (H2SO4, KMnO4 and K2Cr2O7) and at different annealing temperatures (450 and 550 C). Based on standard potential values of 0.45 V, 1.33 V and 1.51 V for H2SO4, K2Cr2O7 and KMnO4, respectively, KMnO4 is considered as the best oxidizing agent. Hence, the stronger the oxidizing agents, the smaller the nanoparticles formed. For the aluminium foil annealed with H2SO4 at 450 C (see Fig. 4a) the structures formed are large and the arrangement is a flower like structure along with nanochain formation. The nanoparticle's size ranges from 50 to 150 nm and the average chain length is observed of about 1800 nm. Nano-rods were also formed on the surface with nano-chains which

Fig. 3. FESEM images of the Al foil after annealing without any oxidizing agent. (a) at 400 C (b) at 450 C (c) at 500 C (d) at 550 C.


4


Fig. 4. FESEM images of the Al foil after annealing at 450 C and 500 C, respectively, using various oxidizing agents, such as H2SO4 (a) and (b), K2Cr2O7 (c) and (d), KMnO4 (e) and (f).

were spread all over the surface. The width of the nano-chain is about 20 nm and length about 200 nm. At 550 C (see Fig. 4b), the Al foil was oxidized showing traces of nanoparticles and chains. For the Al foil treated with K2Cr2O7 at 450 C (see Fig. 4c), rod like nanoparticles were formed with size ranging from 10 to 120 nm and

an average length of 600 nm. At 550 C (Fig. 4d), the average size of the nanoparticles is in the range from 15 to 180 nm, whereas plate like chains show an average length in the order of 2000 nm. Al foils were also treated with KMnO4 at 450 C and 550 C, respectively. Annealing at 450 C (see Fig. 4e) resulted in nanoparticles with sizes



5

Table 1 Atomic and weight percentage values from EDX data. Sample

Element

Weight %

Atomic %

Al foil after annealing at 550 C without any oxidizing agent

Oxygen Aluminium Oxygen Aluminium Sulphur Oxygen Aluminium Oxygen Aluminium

33.60 66.40 41.94 27.40 30.36 42.61 57.39 52.68 47.32

46.05 53.95 57.05 22.34 20.61 55.60 44.40 65.25 34.75

Al foil after annealing at 550 C with H2SO4

Al foil after annealing at 550 C with K2Cr2O7 Al foil after annealing at 550 C with KMnO4

ranging from 25 to 130 nm and instead of chains, the nanoparticles were formed uniformly throughout the surface as a layer. At 550 C (Fig. 4f), KMnO4 being a stronger oxidizing agent than K2Cr2O7 resulted in the formation of smaller nanoparticles with sizes in the range of 10e110 nm and an average chain length of 600 nm. In this case also nanoparticles were formed throughout the surface and the structure was almost round everywhere. Energy-dispersive X-ray spectroscopy (EDX) was taken for the samples treated at 550 C to examine the Al2O3 formation and to check for impurities present in the sample. As seen from Table 1, the EDX data have confirmed the presence of the elements Al and O in various stoichiometric ratios in all samples, whereas the sulphure impurity occurs in the one treated with H2SO4 only. It can also be seen that along with the increasing chemical potentials of the oxidizing agents, the weight percentage of Al decreased and that of O increased in the fabricated samples. X-Ray diffraction analysis (XRD) was carried out for the samples prepared at 450 C and 500 C with and without oxidizing agents,

respectively. It was found that either in the presence or absence of oxidizing agents, the crystalline behaviour increased with increasing annealing temperature due to the development of large crystallites and a higher degree of crystallization [38]. Diffraction peaks corresponding to the Al are indexed with (111), (200), (220) and (311) and those corresponding to Al2O3 are identified as (311), (222), (400), (511) and (444). The intensity of the peaks increased in the presence of oxidizing agents and with increased reaction temperature. Three phases a, g and q are found present in the prepared samples as reported in the literature [14,20,21]. 3.2. Adsorption studies 3.2.1. Effect of the contact time and the initial fluoride concentration Al2O3 nanostructures prepared without oxidizing agent at 450 C and with oxidizing agents H2SO4, K2Cr2O7 and KMnO4, respectively, at 550 C were used as adsorbents. The initial flouride

Fig. 5. Effect of contact time and initial fluoride concentration using adsorbent dose: 4 g/l (a) nanoparticles without oxidizing agent (b) nanoparticles with H2SO4 as oxidizing agent (c) nanoparticles with K2Cr2O7 as oxidizing agent (d) nanoparticles with KMnO4 as oxidizing agents.



Residual fluoride concentration (mg/l)

6

9

1 g/l, the residual fluoride concentration decreased sharply beyond which it was almost constant for doses of 2 mg/l and 4 mg/l. This sharp decrease in fluoride concentration is attributed due to the greater surface area and, thus, the availability of more adsorption sites. The number of active sites and the bulk fluoride concentration decreased and reached equilibrium with the increasing time. Therefore, the amount of residual fluoride concentration will be almost the same with the further increase of the adsorbent dose.

Temperature = 25 oC, Stirring speed = 200 rpm Adsorbent dose (g/l) 2 4 8

8 7 6 5 4 3 2 1 0

0

1

2

3

4

5

Adsorbed dose (g/l) Fig. 6. Variation of the residual fluoride concentration with Al2O3 adsorbent without oxidizing agent.

concentrations of 2 mg/l, 4 mg/l and 8 mg/l were used with an adsorbent dose of 4 g/l. The rapid adsorption of fluoride took place within 20 min, after which the adsorption process became slow and almost reached equilibrium within 90 min. Further increase in contact time for 24 h increased the fluoride removal by less than 1% only. Nanostructures created without an oxidizing agent at 450 C with the initial fluoride concentration of 2 mg/l offered more than 90% removal, whereas 74.87% removal was obtained with the highest fluoride concentration of 8 mg/l. The percentage fluoride removal through the adsorption the nanoparticles prepared using oxidizing agents H2SO4, K2Cr2O7 and KMnO4 at 550 C at the initial floride concentrations of 2 g/l as 88.5%, 91% and 92%, respectively, is presented as a function of the contacting time in Fig. 5. With the initial concentration 8 g/l, the fluoride removals are found as 82.75%, 87.87% and 88.12%, respectively. It was also observed that the adsorption was fast for lower initial fluoride concentrations. 3.2.2. Effect of the adsorbent mass With the increase in the Al2O3 nanoparticle dose from 1 g/l to 4 g/l, at 500 C, without any oxidizing agent, the residual fluoride concentration decreased and the permissible limit (1 mg/l) was achieved. As it is seen, 4 g/l of alumina was required to maintain the permissible limit for 2 mg/l, 4 mg/l and almost for 8 mg/l initial fluoride concentration (Fig. 6). Up to an adsorbent dose of

3.2.3. Effect of pH The equilibrium sorption of fluoride with the initial fluoride concentration of 8 mg/l was investigated over a pH range of 2.67e11.28 and at the operating temperature of 25 C to determine the optimum pH level for the maximum removal of fluoride. Results are shown in Fig. 7 (inset). From the figure it is clear that the fluoride adsorption on alumina strongly depends on the pH valure. The maximum fluoride removal of 88.12% took place at pH ¼ 4.7. The fluoride adsorption initially increased with pH, reached the maximum at pH ¼ 4.7 and then decreased slowly when the pH values rise up to pH ¼ 9.31. Beyond pH ¼ 9.31, the percentage removal decreased sharply (see Fig. 7). As pH is increasing the amount of OH ions in the solution would increase as well. Since F is the ion concerned for removal, hence as pH is increasing the repulsion between the F and OH ions would increase, which results in a decrease in the fluoride removal. Al2O3 remains stable at pH ¼ 11 whereas it starts to dissolve at a pH lower than 3 [10e14]. 3.2.4. Effect of the stirring speed Stirring is an important parameter in adsorption phenomena, because it provides and promotes proper contact between the adsorbent and the solution. It helps to distribute the solute in bulk solution and also aids the formation of an external boundary film. We utilized the shaking incubator from Lab tech to study the effect of the stirring speed in our samples. Stirring speed of 100, 200 and 300 rpm were used with contact time of 90 min on KMnO4 treated nanoparticles was used with an adsorbent dosage of 4 g/l at 25 C. With the increase in stirring speed from 100 to 300 rpm the percentage fluoride removal changed from 85.87 to 87%, respectively. The increase in the fluoride removal with the increase stirring speed is explained by the fact that the increase in stirring speed reduced the film boundary layer surrounding the adsorbent, thus increasing the external film transfer coefficient and hence stimulating better fluoride adsorption.

0.93

100

y =-159.45x + 1.407 R2 = 0.988

0.92

log ((qe/Ce) × m)

80 14

60

12 10

Final pH

Fluoride removal (%)

Initial fluoride concentration = 8 mg/l, Ads. dose = 4 g/l Temperature = 25 oC

40

8 6

0.90 0.89 0.88

4

20

0.87

2 2

0

0.91

4

6

8

10

12

14

Initial pH

0

2

4

6

0.0031

8

10

Initial pH of fluoride solution

12

Fig. 7. Effect of pH on the fluoride adsorption, Inset: pHzpc of Al2O3 adsorbent.

0.0032

0.0033

0.0034

1/T (1/oC) Fig. 8. Arrhenius plot for the adsorption of fluoride by Al2O3 adsorbent prepared with KMnO4 as oxidizing agent.



7

Table 2 Thermodynamic parameters for adsorption of fluorine. Temperature ( C)

DH (KJ/mol)

DS (KJ/mol)

S* (KJ/mol)

Ea (KJ/mol)

DG (KJ/mol)

25 35 45 55

3.053

26.94

0.039

2.707

8.025 8.294 8.563 8.833

temperatures 25 C, 35 C, 45 C and 55 C. The experiment was conducted while maintaining an initial fluoride concentration of 8 mg/l and an adsorbent dose of 4 g/l of Al2O3 prepared with KMnO4. It was observed that, with the increase of the operating temperature from 25 to 55 C, the percentage removal increased from 88.12 to 89.2% which indicates the endothermic behaviour of the adsorption. The thermodynamic parameters, such as the change in the standard free energy, the enthalpy, and the entropy were calculated to study the spontaneous nature and the thermodynamic feasibility of the process [23e26]. These parameters can be calculated by using the following equations: DG ¼ DH T DS . The enthalpy of adsorption (DH) and the entropy ðDS Þ can be calculated from the slope and the intercept, respectively, of the logðqe m=Ce Þ versus 1/T plot where, m is the adsorbent dose (in g/ l), qe is the amount of arsenic adsorbed per unit mass of adsorbent (in mg/g), Ce is the equilibrium concentration (in mg/L), T is the temperature in Kelvin, and then the ratio of qe/Ce is called the adsorption affinity (see Fig. 8). The thermodynamic parameters at the initial fluoride concentration of 8 mg/l are summarized in Table 2. The positive enthalpy of adsorption ðDH Þ and the negative Gibbs free energy change ðDG Þ suggest the process to be endothermic and spontaneous in

3.2.5. Effect of other ions Natural fluoride contaminated water contains several other ions, such as nitrate (KNO3), chloride (NaCl), sulphate (Na2SO4), carbonate (Na2CO3) and bicarbonate (NaHCO3), which can affect the fluoride ion adsorption process. To study the effects of Na ions, 300 mg/l of NaCl solution for each ion was added separately to the fluoride solution. The adsorption experiments were performed in an 8 mg/l fluoride solution at 25 C with 4 g/l of adsorbent prepared with KMnO4 as oxidizing agent. It was observed that the effect of the chloride, nitrate and sulphate ions were negligible and the related equilibrium fluoride removal was 87.62, 86 and 85.32%, respectively. The decrease in the fluoride removal is due to the competition of these anions for the adsorption with fluoride. The percentage fluoride removal decreased to 32.5 and 26.12% by the effect of bicarbonate and carbonate ions, respectively, and it was due to the significant increase of the pH values in the solution. With the increasing pH, the number of OH ions would increase resulting in an enhanced repulsion of F ions, thereby reducing the percentage fluoride removal. 3.3. Thermodynamical considerations To study the effect of temperature on the percentage fluoride removal, experiments were conducted with the operating 250

250 (b)

Fluoride concentration (mg/l) 2 4 8


200

t/qt (g min mg-1)

t/qt (gminmg-1)

200

(a)

150

150

100

100

50

0 0

20

40

60

80

0

100

0

Time (min) 240

40

60

80

100

80

100

200 (d)



160

t/qt (g min mg-1)

t/qt (g min mg-1)

20

Time (min)

(c)

200

50

160

120

120 80 40 0

0

20

40

60

Time (min)

80

100

80

40

0

0

20

40

60

Time (min)

Fig. 9. Pseudo second order kinetic model fitting. (a) Adsorbent without oxidizing agent. (b) Adsorbent with H2SO4. (c) Adsorbent with K2Cr2O7. (d) Adsorbent with KMNO4.


8


Table 3 Kinetic model parameters with adsorbent treated without oxidizing agent and with oxidizing agents H2SO4, K2Cr2O7 and KMnO4, respectively. Oxidizing agent

Kinetics

Equation

None

Pseudo first order

ln(qe qt) ¼ lnqe k1t

None

Pseudo second order

t/qt ¼ 1/k2qe þ t/qe

H2SO4

Pseudo first order


H2SO4

Pseudo second order


K2Cr2O7

Pseudo first order


K2Cr2O7

Pseudo second order


KMnO4

Pseudo first order


KMnO4

Pseudo second order


3.4. Kinetic aspects The kinetics of the fluoride adsorption on the surface of the Al2O3 nanostructures were studied using different models, such as pseudo-first-order and pseudo-second-order ones [23e26]. Four different Al2O3 nano-adsorbents were prepared without oxidizing agent at 450 C, with oxidizing agents H2SO4, K2Cr2O7 and KMnO4, respectively, at 550 C. The pseudo-first order model assumes that the changing rate of the solute uptake with time is directly proportional to the amount of solutes adsorbed with time and the difference in equilibrium concentration. The first-order adsorption rate constant is given by dqt =dt ¼ K 1 ðqe qt Þ; where qt and qe are the fluoride amount of adsorbed (mg/g) during the contact time t

qe,expt (mg/g) qe,cal (mg/g) k1 (min1) R2 qe,cal (mg/g) k2 (min1) R2 qe,expt (mg/g) qe,cal (mg/g) k1 (min1) R2 qe,cal (mg/g) k2 (min1) R2 qe,expt (mg/g) qe,cal (mg/g) K1 (min1) R2 qe,cal (mg/g) k2 (min1) R2 qe,expt (mg/g) qe,cal (mg/g) k1 (min1) R2 qe,cal (mg/g) K2 (min1) R2

2

4

8

0.41 0.15 0.07 0.89 0.41 2.79 0.99 0.44 0.11 0.06 0.99 0.44 3.84 0.99 0.45 0.10 0.07 0.79 0.45 5.12 0.99 0.46 0.13 0.07 0.82 0.46 3.69 0.99

0.79 0.34 0.06 0.90 0.80 1.02 0.99 0.86 0.26 0.07 0.99 0.86 1.90 0.99 0.89 0.19 0.06 0.75 0.89 2.70 0.99 0.90 0.30 0.08 0.88 0.90 1.85 0.99

1.49 0.81 0.06 0.94 1.52 0.37 0.99 1.65 0.73 0.07 0.99 1.66 0.59 0.99 1.75 0.67 0.07 0.89 1.76 0.64 0.99 1.76 0.65 0.05 0.83 1.77 0.46 0.99

(min) and at equilibrium; k1 is the pseudo-first-order rate constant (g/mg-min). The adsorption rate constant k1 and equilibrium adsorption capacity qe can be calculated from the plot of lnðqe qt Þ versus t. From the calculated values it is clear that the kinetics of fluoride adsorption on the Al2O3 nanoparticles does not follow the pseudo first-order kinetics and, hence, neither the diffusion controlled phenomena. The sorption kinetics may be represented by the pseudo-second-order model given by: dqt =dt ¼ k2 ðqe qt Þ, where, k2 is the equilibrium rate constant for the pseudo-second order sorption (g/mg min). The values of qe and k2 were calculated from the slope and the intercept of t/qt versus t plot. Fig. 9 shows the plots of t/qt versus t for the different adsorbents. The calculated values of k2, qe and R2 are presented in Table 3. The values of the regression coefficient R2 is nearly unity (0.99) for all

2.5

2.0

qe (mg/g)

nature. The low value of ðDS Þ implies that no remarkable change in entropy occurred during the fluoride adsorption process. The positive value of ðDS Þ reflects an increase in randomness at the solidsolution interface during the adsorption. The low value of enthalpy of adsorption ðDH Þ indicates that the physical adsorption process is dominating over the fluoride adsorption one. In order to further confirm the assertion that the physical adsorption is the predominant mechanism in the fluoride adsorption process, the values of activation energy (Ea) and sticking coefficient (S*) were determined from the experimental data by using a modified Arrhenius type equation related to surface coverage (q) as the following: S ¼ ð1 qÞeEa =RT . The Sticking coefficient S* is a function of the adsorbateeadsorbent system under investigation, its value lies in the range 0 < S* < 1, and is completely dependent on temperature of the system. As the name specifies, the parameter S* indicates the potentiality of an adsorbate to remain on the adsorbent system. The value of q can be calculated by using the following equation: q ¼ 1 Ce/Co. The activation energy (Ea) and the sticking coefficient (S*) were estimated from the plot of (1 q) versus 1/T and the result is shown in Table 2. The lower value of (Ea) & (S*) confirms the physsorption behaviour of the present adsorption system [27,34,35e37].

Initial fluoride concentration (mg/l)

Parameters

Adsorbent dose = 4 g/l Experimental Langmuir Freundlich

1.5

1.0

0.5

0.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Ce (mg/l) Fig. 10. Adsorption isotherm plot for the adsorption of fluoride by Al2O3 adsorbent prepared with KMnO4 as oxidizing agent.



9

Table 4 Langmuir and Freundlich isotherm constants for fluoride adsorption. Isotherm value Langmuir 3.82 0.87 0.99 Freundlich 1.75 0.71 0.99

Equation 1 qe

¼q

1

m bC e

þ q1

m

1=n

qe ¼ K F C e

Plot

Parameters

A plot of 1=qe versus 1=C e indicated a straight line with a slope of 1=bqm and intercept of 1=qm .

qm (mg/g) b (L/mg) R2

The values of K F and 1=n were obtained from the slope and intercept of the linear plot of lnqe versus lnC e .

KF (mg/g) n (L/mg) R2

Table 5 Comparison of fluoride adsorption capacity of acidic alumina with a few reported adsorbents. Adsorbent

Adsorbent capacity (mg/l)

Reference

Plaster of paris Granular red mud Pyrophyllite Hydrated cement Activated alumina Manganese oxide coated alumina Mg/Al layered double hydroxide Iron oxide-hydroxide nanoparticle Alumina (KMnO4 oxidized at 550 C)

0.37 0.85 2.2 2.68 2.41 2.85 0.99 1.66 3.82

Gopal et al., 2007 [27] Tor et al., 2009 [28] Goswami et al., 2011 [29] Kagne et al., 2008 [30] Ghorai et al., 2005 [31] Maliyekkal et al., 2006 [32] Elhalil et al., 2016 [33] Raul et al., 2012 [34] Present study

the adsorbents at all the initial fluoride concentrations. These values reveal that the kinetics of fluoride adsorption is of a pseudosecond order process nature. Hence, the adsorption process is favouredly converned by the pseudo-second order kinetic model.

3.5. Equilibrium parameters It is essential to establish the most appropriate correlation for the equilibrium curve to optimize the design of an adsorption system. The commonly used and most useful equilibrium isotherms are the Langmuir and the Freundlich isotherms [23–26]. Langmuir isotherm is based on the assumption that there is a finite number of active sites which are homogenously distributed on the adsorbent surface. These binding sites on the surface of the adsorbent have the same affinity for the adsorption of a single molecular layer, and there is no interaction between the adsorbed molecules. The equation of Langmuir isotherm is: qe ¼ qm bC e =ð1 þ bC e Þ, where qe is the adsorbed amount at equilibrium (in mg/g); Ce is the equilibrium arsenic concentration (in mg/l); the constant b is related to the energy of adsorption (in l/mg); and qm is the Langmuir monolayer adsorption capacity (in mg/g). The parameters can be determined from the plot of 1/ qe versus 1/Ce (see Fig. 10). To check the feasibility of the isotherm, the dimensionless equilibrium parameter RL was determined by RL ¼ 1=ð1 þ bC 0 Þ, where b (in l/mg), is the Langmuir constant and C0 (in mg/l) is the initial concentration in the liquid phase. The value of RL indicates the shape of the isotherm to be either unfavourable (RL > 1), linear (RL ¼ 1), favourable (0 < RL < 1) or irreversible (RL ¼ 0). For the present study, RL values obtained are in the range of 0.36e0.10 for the initial fluoride concentration in the range of 2e10 mg/l. The RL value indicates that the fluoride adsorption is more favourable at higher initial fluoride concentrations than at the lower ones. The Freundlich isotherm model is based on the multilayer adsorption of an adsorbate onto the heterogeneous surface of an adsorbent. 1=n The expression for Freundlich isotherm is given as qe ¼ K F C e (in Fig. 10), where KF is the Freundlich constant is related to the bonding energy, and 1/n is a measure of intensity of the adsorption. The higher the 1/n value, the more favourable is the

adsorption. From the Table 4, it is clear that both the models provide a high regression correlation coefficient (R2 ¼ 0.99 & 0.99), so both the Langmuir and Freundlich models can be used for describing the adsorption equilibrium of fluoride. From the Langmuir and Freundlich adsorption isotherm data, it is clear that both the simulated isotherms are comparable with experimental values [27,29e31,34e37] (see Table 5). 4. Regeneration The regeneration study was performed to develop an efficient adsorbent that can be reused thereby making it cost effective. In this case, KMnO4 treated nanoparticles were used as the adsorbent; 8 mg/l fluoride was adsorbed on 4 g/l of adsorbent. Then, the adsorbent was transferred into 100 ml water and the pH was adjusted to be acidic at pH ¼ 4.7. In the acidic pH range, it is rather hard to leach any fluoride. As the pH was increased above 4.7, more than 75% of fluoride was desorbed in about 90 min. The as-desorbed adsorbent was then washed with 0.1 M HCl to make it acidic in nature and to activate it for further reuse in adsorption. 5. Conclusion Al2O3 nanostructures were synthesized on the surface of aluminium foils in the range of 10e80 nm. In presence of the oxidizing agents H2SO4, K2Cr2O7 and KMnO4, and an increasing temperature, the size of the nanoparticles increased. Oxidizing agents help to form particles with sizes of around 150, 130 and 150 nm, respectively. The suitable temperature of 450 C was found for the preparation of nanostructures without any oxidizing agent and with H2SO4, whereas it was 550 C for the preparation with the oxidizing agents K2Cr2O7 and KMnO4. Furthermore, the nanoparticle density was high in the materials prepared with KMnO4 as the oxidizing agent at 550 C. The current study has highlighted that the Al2O3 nanoparticles, which were acidic in nature possesses a considerable potential for the fluoride adsorption from an aqueous medium. A maximum of 92% fluoride can be adsorbed by Al2O3 nanoparticles at pH ¼ 4.7.


10


References [1] A.K. Susheela, Fluorosis management programme in India, Curr. Sci. 77 (1999) pp.1250e1256. [2] WHO, Guidelines for Drinking-water Quality, Health Criteria and Supporting Information, 2, World Health Organization, Geneva, Switzerland, 1984. [3] US Public Health Service Drinking Water Standards, US Government Printing Office, Department of Health Education and Welfare, Washington, DC, 1962. [4] A.K. Susheela, G. Ghosh, Fluorosis management in India: the impact due to networking between health and rural drinking water supply agencies, in: Interdisciplinary Perspectives on Drinking Water Risk Assessment and Management Proceedings of the Santiago (Chile) Symposium, 1998, 2000. Publ. no. 260. [5] S.L. Choubisa, K. Sompura, Dental fluorosis in tribal villages of Dungarpur district (Rajasthan), Poll. Res. 15 (1996) 45e47. [6] A. Goswami, M.K. Purkait, Removal of fluoride from drinking water using nano magnetite aggregated schwertmannite, J. Water Process Eng. 1 (2014) 91e100. [7] A. Dey, R. Singh, M.K. Purkait, Cobalt ferrite nanoparticles aggregated schwertmannite: a novel adsorbent for the efficient removal of arsenic, J. Water Process Eng. 3 (2014) 1e9. [8] A. Goswami, M.K. Purkait, Defluoridation of water by schwertmannite, World Acad. Sci. Eng. Technol. 73 (2013) 1156e1161. [9] S. Ishihara, H. Suematsu, T. Nakayama, S. Tsuneo, K. Niihara, Synthesis of nanosized alumina powders by pulsed wire discharge in air flow atmosphere, Ceram. Int. 38 (2012) 4477e4484. [10] V. Piriyawong, V. Thongpool, A. Piyapong, P. Limsuwan, Preparation and Characterization of Alumina Nanoparticles in Deionized Water Using Laser Ablation Technique, Applied Nanotechnology Laboratory (ANT Lab), Department of Physics, 2011. [11] A.I.Y. Tok, F.Y.C. Boey, X.L. Zhao, Novel synthesis of Al2O3 nano-particles by flame spray pyrolysis, J. Mat. Process. Technol. 178 (2006) 270e273. [12] K.M.S. Khalil, Formation of mesoporous alumina via hydrolysis of modified aluminium isopropoxide in presence of CTAB cationic surfactant, Appl. Surf. Sci. 255 (2008) 2874e2878. [13] H.S. Kim, N.K. Park, T.J. Lee, H.M. Myeong, M. Kang, Preparation of nanosized a-Al2O3 particles using a microwave pretreatment at mild temperature, Adv. Mat. Sci. Eng. (2012) 1e6. I.D. 920105. [14] A. Janbey, R.K. Pati, P. Pramanik, A new chemical route for the synthesis of nanocrystalline a-Al2O3 powder, J. Eur. Ceram. Soc. 21 (2001) 2285e2289. [15] X. Zhao, Y.L. Shi, Y.Q. Cai, S.F. Mou, Cetyltrimethylammonium bromide-coated magnetic nanoparticles for the pre-concentration of phenolic compounds from environmental water samples, J. Environ. Sci. Technol. 1139 (2008) 178e184. [16] A.K. Chaturvedi, K.P. Yadava, K.C. Yadava, K.C. Pathak, V.N. Singh, Defluoridation of water by adsorption on fly ash, Water Air Soil Pollut. 49 (1990) 51e61. [17] G. Singh, B. Kumar, P.K. Sen, J. Majumdar, Removal of fluoride from spent potline leachate using ion exchange, Water Environ. Res. 71 (1999) 36e42. [18] S. Saha, Treatment of aqueous effluent for fluoride removal, Water Resour. 27 (1993) 1347e1350. [19] D. Clifford, J. Matson, R. Kennedy, Activated alumina: rediscovered adsorbent for fluoride, humic acids and silica, Ind. Water Eng. 15 (1978) 6e12.

[20] V.S. Giri, R. Sarathi, S.R. Chakravarthy, C. Venkataseshaiah, Studies on production and characterization of nano-Al2O3 powder using wire explosion technique, Mat. Lett. 58 (2004) 1047e1050. [21] J. Li, Y. Pana, C. Xianga, Q. Gea, J. Guoa, Low temperature synthesis of ultrafine a-Al2O3 powder by a simple aqueous solegel process, Ceram. Int. 32 (2006) 587e591. [22] S. Anandan, X. Wen, S. Yang, Room temperature growth of CuO nanorod arrays on copper and their application as a cathode in dye-sensitized solar cells, Mat. Chem. Phys. 93 (2005) 35e40. [23] C.C. Tang, S.S. Fan, P. Li, M. Lamy, H.Y. Dang, In situ catalytic growth of Al2O3 and Si nanowires, J. Cryst. Growth 224 (2001) 117e121. [24] X. Zhichuan, B. Zhuanfang, S. Chengmin, A facile fabrication of Cu2O nanowire arrays on Cu substrates, Cent. Eur. J. Eng. 2 (3) (2012) 364e368. [25] M.G. Sujana, H.K. Pradhan, S. Anand, Studies on sorption of some geomaterials for fluoride removal from aqueous solutions, J. Hazard. Mat. 161 (2009) 120e125. [26] M.G. Sujana, S. Anand, Fluoride removal studies from contaminated ground water by using bauxite, Desalination 267 (2011) 222e227. [27] V. Gopal, K.P. Elango, Equilibrium, kinetic and thermodynamic studies of adsorption of fluoride onto plaster of Paris, J. Hazard. Mat. 141 (1) (2007) 98e105. [28] A. Tor, N. Danaoglu, G. Arslan, Y. Cengeloglu, Removal of fluoride from water by using granular red mud: batch and column studies, J. Hazard. Mat. 164 (1) (2009) 271e278. [29] A. Goswami, M.K. Purkait, Kinetic and equilibrium study for the fluoride adsorption using pyrophyllite, Separ. Sci. Technol. 46 (2011) 1797e1807. [30] S. Kagne, S. Jagtap, P. Dhawade, S.P. Kamble, S. Devotta, S.S. Rayalu, Hydrated cement: a promising adsorbent for the removal of fluoride from aqueous solution, J. Hazard. Mat. 154 (2008) 88e95. [31] S. Ghorai, K.K. Pant, Equilibrium, kinetics and breakthrough studies for adsorption of fluoride on activated alumina, Sep. Purif. Technol. 42 (2005) 265e271. [32] S.M. Maliyekkal, A.K. Sharma, L. Philip, Manganese-oxide-coated alumina: a promising sorbent for defluoridation of water, Water Res. 40 (2006) 3497e3506. [33] A. Elhalil, S. Qourzal, F.Z. Mahjoubi, R. Elmoubarki, M. Farnane, H. Tounsadi, M. Sadiq, M. Abdennouri, N. Barka, Defluoridation of groundwater by calcined Mg/Al layered double hydroxide, Emerg. Contam. 2 (2016) 42e48. [34] P.K. Raul, R.R. Devi, I.M. Umlong, S. Banerjee, L. Singh, M. Purkait, Removal of fluoride from water using iron oxide-hydroxide nanoparticles, J. Nanosci. Nanotechnol. 12 (2012) 3922. [35] M. Changmai, M.K. Purkait, Kinetics, equilibrium and thermodynamic study of phenol adsorption using NiFe2O4 nanoparticles aggregated on PAC, J. Water Proc. Eng. 16 (2017) 90e97. [36] S. Karaca, A. Gurses, M. Ejder, M. Acikyildiz, Adsorptive removal of phosphate from aqueous solutions using raw and calcinated dolomite, J. Hazard. Mater. 128 (2006) 273e279. http://www.sciencedirect.com/science/journal/03043894. [37] A. Goswami, M.K. Purkait, The defluoridation of water by acidic alumina, Chem. Eng. Res. Des. 90 (2012) 2316e2324. [38] M. Changmai, M.K. Purkait, Interaction of fatty acid chain length with NiFe2O4 nanoparticles, Surf. Interfaces 8 (2017) 45e50.
