KINETIC AND THERMODYNAMIC STUDIES OF TRIVALENT ARSENIC REMOVAL BY INDIUM-DOPED ZINC OXIDE NANOPOWDER
L. KHEZAMIa, KAMAL K. TAHAa,b*
1. Introduction
, A. MODWIa,c
aDepartment of Chemistry, College of Sciences, Al-Imam Muhammad Ibn Saud University(IMSIU), ,Riyadh 11623, Saudi Arabia.
bDept. Chem.and Ind. Chem., College of Applied & Industrial Sci., University of Bahri, Sudan
cDepartment of Chemistry, College of Science, Omdurman Islamic University, Omdurman, Sudan
The uptake of trivalent arsenic from aqueous solutions onto indium-doped ZnO (IZO) nanopowder was studied. The nanopowder was fabricated using a modified sol–gel method under supercritical drying conditions of ethanol. X-ray diffraction and nitrogen adsorption techniques were employed to characterize the IZO nanopowder. The N2 adsorption isotherms reveal that pure and doped ZnO nanopowders were mesoporous materials. Removal of arsenite ions from the solutions by IZO nanopowder was conducted in a batch-mode reactor. The impact of initial concentration, indium-doping dose, temperature and pH were considered. Arsenite removal was pH sensitive scoring its maximum elimination capacity at pH 3 and 7. The adsorption equilibrium was well delineated by Langmuir isotherm. Data were further found to comply with the pseudo- second-order kinetic law, and the process was spontaneous endothermic physisorption.
(Received October 16, 2016; Accepted December 28, 2016)
Keywords:Nanostructures; IZO; trivalent arsenic; Adsorption; kinetics.
Heavy metals ions are disposed into the environment due to human anthropogenic activities. As they are toxic, they are hazardous to living organisms and environment, especially when tolerance levels are exceeded. Arsenic is a lethal pollutant that results from industrial and other anthropogenic activities as well as its natural abundance in ground water. Wastewater effluents from the pharmaceutical, pesticidal, chemical, metallurgical, mining, and leather tanning industries have significant contribution to arsenic pollution. Arsenic is present in the trivalent As(III) and pentavalent As(V) form stermed as arsenites and arsenates, respectively, with the former considered highly perilous than the latter. Sicknesses such as dermal, lungs and bladder cancer, as well as gastroenteritis [1 - 4] were related to the contact with arsenic. The permissible limit for arsenic in industrial wastewaters in most developed countries such as Japan is about 10 µg L-1[5].
Methods of heavy metals ions elimination like precipitation, filtration and ion exchange are commonly employed. Such techniques are less adopted as they are expensive, unsuitable to daily life activities. Adsorption is alternative versatile technique extensively applied for water systems purification from pollutants. Alumina, silica, metal hydroxides [6, 7] activated carbon [8, 9], and zeolites [8], or natural products such as clays [10] and red soil [11] were utilized for toxic metal ions elimination. In previous studies, IZO nanopowder has been used as an adsorbent for heavy metals [12 – 14].
This study aimed to assess the adsorption efficiency of IZO nanopowder for As(III) ion selimination from aqueous solutions. The effects of indium-doping percentage, pH and initial As(III) concentration were evaluated through batch adsorption experiments. First- and second-
*Corresponding author: [email protected]
order rate representations were used to delineate the adsorption kinetics at different initial arsenitei on concentrations, pH 3 and 7 at 25 °C. In addition, adsorption capacities of As(III) on IZO adsorbent were correlated by Langmuir, Freundlich, and Dubinin–Radushkevich (D-R) isotherms at the two different pH values. Constants derived from the adsorption thermodynamic studies were evaluated.
2. Experimental
2.1. IZO nanopowder synthesis
IZO nanopowder was synthesized by thesol–gelprocedure following EL MIR et al. [16, 17]. IZO nanopowder were produced by mixing zinc acetate dehydrate and a sufficient volume of methyl alcohol. The mixture was magnetically stirred for 10 minutes, then indium chloride equivalent to different ratio of [In]/([Zn]+[In]) was added to the mixture. After extra stirring for 15 minutes, the mixture was placed in an autoclave and dried under supercritical conditions of ethanol.
2.2. IZO Characterization
The crystallite pattern of the obtained IZO nanopowders at different doping concentrations was identified using X-ray diffraction (XRD). While its porosity, was characterized by nitrogen adsorption using ASAP 2020 Micromeritics analyzer. Prior to each analysis, ZnO and ZnO: Inx%
samples were out gassed at 250 °C for 6 h in vacuum. The nanopowder’s surface area and pore size and diameter were calculated from the BET equation and Lippens and de Boer [18] t-plot method.
2.3. Solutions Preparation:
An stock solution of 1000 mg L−1 was prepared from a commercial pure As[III] standard.
Solutions required to carry out the experiments were prepared by diluting the stock solution to intended concentrations. The solutions pH was fixed using HNO3and NaOH. All reagents used were of high purity and were used as received.
2.4. Batch-mode adsorption
Batch mode experiments were performed by mixing 13 mg of IZO and 25 ml of a known As(III) molarity and pH in 50 ml Erlenmeyer flask. Adsorption studies were conducted at different pH values (2.3–9.5) and initial As(III) concentration (5–30 mg L-1) to obtain equilibrium isotherms. Several flasks were placed on a magnetic stirrer and stirred at 550 rpm. About 15 ml of suspension was sampled from each flask after 12 h, centrifuged with Hettich Zentrifugen EBA 20), and then passed through 0.25 µm cellulose acetate filter. The As(III) content in filtrate was estimated using atomic emission spectroscopy (Genius, ICP-EOS, Germany). All experiments were carried in triplicate and their mean was used.
The mass of metal ions adsorbed (qe) per gram of IZO (mg g-1), and the percentage of removal values were obtained using the relationships:
( )
m V qe = C0 −Ce ∗
(1)
( )
100 C *
Re
0 0
C moval C
% = − e
(2)
Here C0 and Ce are the As(III) ion concentrations (mg L-1) at t =0 and equilibrium respectively, whereas V (L) and m(g) are the solution volume and dry adsorbent weight.
3. Results and Discussion 3.1. IZO nanopowder characterization 3.1.1. XRD
Fig. 1 portrays the XRD structures of IZO recorded in 2θ degree. The obtained peaks were identical to the hexagonal wurtzite structure ZnO. For all indium-doping percentages, no peak corresponding to indium or indium oxide phases were observed, which denoted that In ions have perforated into the lattices of ZnO. Thus there is no peak corresponding to a secondary phase of indium. At the highest In doping ratio (5 %), the peak 101 has been enhanced indicating a preferential growth orientation in the (101) plane [19]. A similar behavior has been observed for aluminum doped ZnO[20]. Nanoparticle size was estimated using XRD data and applying Scherer equation:
(3)
here λ is the X-ray wavelength (λ = 1.78901 Å for Co Kα), β the full width at half maximum (FWHM) of the diffraction peak θβ(in rad). The mean grain size of nanoparticles was almost equal to 53 nm. The values of crystallite size obtained in this work is comparable with the 65 nm of undoped ZnO and 58 nm doped sample reported by Hjiri et al. [21] and Ben Ayadi et al.[22].
Fig. 1 XRD patterns of indium doped ZnO powder with different indium concentrations
3.1.2. Nitrogen adsorption
Isotherm shape provides information on pore size, which is usually categorized as micropore, mesopore or macropore. For instance, Fig.2 depicts the evolution of N2 adsorption–
desorption isotherms for pure ZnO and IZO nanopowders. These isotherms were obviously type II, with a hysteresis of H4 type, according to the IUPAC or Brunauer’s categorization of sorption isotherms [23]. At elevated relative pressure P/P0, the hysteresis of H4 type was due to the filling up of meso pores by capillary condensation, indicating a shape of pores that was flatter instead of cylindrical. Particle size, pore characteristics, and BET surface area are recorded in Table 1. BET surface area was found to increase with increased indium-doping concentration.
G= 0.9
COS β λ
β θ
Table 1 Main characteristics of IZO nanopowders
Nanoparticles Grain Size (nm)
Pore diameter (nm)
Pore volume (cm3g-1)x102
Surface BET (m2g-1)
ZnO 62.00 46.5 7.1 8.25
ZnO:In 1% 61.00 14.3 5.0 8.18
ZnO:In 2% 60.71 18.4 9.5 13.57
ZnO:In 3% 61.70 24.2 14.5 20.07
ZnO:In4% 62.62 12.2 6.3 17.92
ZnO:In 5% 63.11 18.3 11.5 19.51
The surface area significantly increased from 8 to 20 m2 g-1 with increased doping ratio from 1% to 3%, and then slightly decreased for nanopowders of 4% and 5% doping ratio. Results showed increased mean diameter of pores (from 14.3 to 24.1 nm) and total volume (5×10-2to 14×10-2 cm3g-1) for samples with doping ratio ranging from 1% to 3%, followed by a gradual decrease for samples with 4% and 5% doping doses. This finding was in agreement with XRD results (Table 1 and Fig. 1).
0.0 0.2 0.4 0.6 0.8 1.0
0 20 40 60 80 100
Quantity adsorbed, cm3 /g STP
Relative pressure P/P0 ZnO
ZnO:In1%
ZnO:In2%
ZnO:In3%
ZnO:In4%
ZnO:In5%
Fig.2 Adsorption-desorption isotherms of N2 at 77 K of pure and Indium doped zinc oxide nanopowders
Al Dahoudi et al.[24] have reported a very similar trend of BET surface area for crystalline IZO nanopowder synthesized by hydrothermal treatment. They observed that the incorporation of indium ions into ZnO particles obviously affected both the growth of ZnO lattice and nanostructure. They further demonstrated an evolution of nanoparticle shape, i.e., rod-like pure ZnO progressively transformed to spherical, thereby leading to decreased nanoparticle size with increased indium-doping concentration.
The BET surface area was reported to enlarge from 6.4 to 20 m2g-1 when ZnO was doped with Al [25]. Identical findings were reported by Wang et al.[26]who found that the BET surface area of boron doped titanium oxide increased when the ratio was increased from 0.11 to 0.57 (17 to 34 m2g-1). Whereas, when the ratio was raised to 1.14 the surface area dropped drastically to 13 m2g-1.
3.2. Arsenic adsorption
3.2.1. Effect of indium-doping concentration on arsenic removal
In this study, 10 mg of IZO adsorbent with different indium-doping concentrations (1% to 5%) were added to 25 ml of 15 mg L-1 metal ion solution. Fig.3 illustrates the impact of indium- doping percentage on the adsorption of As(III) ions by IZO.
Fig.3 Influence of Indium doping concentration on the amount of adsorption and specific surface area
The amount of adsorbed metal ions qe sharply increased with increased doping dose, reaching a maximum of 26.34 mgg-1 at 3% doping dose. Afterwards, this optimum qe declined with increased doping dose. The maximum amount of adsorption was obtained at 3% doping dose, indicating that the powder was highly electrical conducting and that adsorption may be controlled by Van der Waals forces. Fig. 3 also shows a perfect coherence between the values of surface area of IZO nanopowder and its capacity of adsorption at various indium-doping doses. Thus, adsorption capacity may be enhanced by BET surface area increase, similar to the report of Bhattacharya et al. [5]. They showed that As(III) adsorption on different uncalcined and calcined aluminum hydroxide powders mostly depended on surface area. During the removal of arsenic by Mn-doped iron oxide Garcia et al. [27], observed an enhancement in its binding when the Mn ratio was 50 % and a decrease as the ratio was increased to 75 %. They attributed the first case to the substitution of Mn in the oxide lattice, while at higher ratio a second phase of Mn was formed leading to the suppression of arsenic binding.
3.2.2. Effect of the pH on arsenic removal
Working solution pH is considered one of the key factors controlling heavy-metal removal [28]. At pH 0–9, As(III) is stable neutral H3AsO3, whereas H2AsO−3, HAsO2−3, and AsO3−3 exist as stable entity at pH 10–12, 13, and 14, respectively. Numerous investigations conducted within this range have indicated maximum arsenic ion removal using different adsorbents [10 -15, 17]. Thus, the pH influence on As(III) adsorption was investigated within pH 2.3–9.5 at 25 °C, with the initial As(III) concentration fixed at 20 mgL-1 and 10 mg of IZO nanoparticles. The results of the pH effect on adsorption performance are illustrated in Fig.4. The removal of As(III) increased with increased initial solution pH, reached the first maximum at pH 3.0 (48.8 mgg-1), and then declined at pH 6.0. The maximum of adsorption quantity increased with increased pH to reach again a second maximum (40.7 mgg-1), followed by a drastic decrease at pH 9.5.
Fig.4 The effect of pH on the adsorption of As (III)onto indium doped ZnO at 25oC.
0 5 10 15 20 25 30
0 5 10 15 20 25 30
0 1 2 3 4 5 6
SBET, m2g-1
qe, mgg-1
Indium doping concentration, % SBET
qe
26.0 48.8
31.7 32.4
20.5 40.7
23.4
8.5 0.0
10.0 20.0 30.0 40.0 50.0 60.0
0 1 2 3 4 5 6 7 8 9 10
qe, mgg-1
Initial pH
Indeed, arsenite ions removal by IZO is highly pH dependent, as has been reported by several investigators[ 6 – 11], who showed enhanced As(III) adsorption by various adsorbents at pH 0–9. Kelly and Tarek [8] showed that the adsorption of arsenates and arsenites on iron-treated activated carbon and zeolites was greater at pH 7.0–11.0. Yunhai et al. [9] studied the removal of chromium and arsenic by activated carbon and found that As(III) adsorption reached the upper limit at pH 7.0. Pravin et al. [11] announced that the removal As(III) by red soil was independent of pH within the range of 4–10, with increased adsorption at the acidic pH range. Upon investigating the adsorption of As(III) on Jang et al. [29] showed that the optimum binding of the arsenic ions to iron oxide nanomaterials was at pH 6 to 9 and a sharp decrease was at pH 10. Thi et al.[30]reported an optimum adsorption of As (III) on iron magnetic particles at pH 7.0 and less at pH < 5 due to instability of the adsorbent at low pH values. An utmost adsorption was achieved at pH 3 by Garcia et al. [27] for As(III) onto mixed ferrite and hausmannite nanomaterials. An increase in arsenic removal in a pH range of 2.8 – 3.8 was attributed to an increase of de protonated arsenic species [31].
3.2.3 Kinetic study
The pseudo-first- and pseudo-second-order kinetic representations were employed to speculate the adsorption data of As(III) relation to time. The firs-order law is illustrated by the equation [32]:
(
q q)
ln( )
q k .tln e− t = e − 1 (4)
Here qe and qt (mgg-1) are the masses of As(III) adsorbed by a gram of IZO at equilibrium and time t (min), respectively, and k1 (min-1) is the rate constant. The intercept and slope of the ln(qe – qt) against twill give qe and k1values respectively. The second rate law is given by the formula [33]:
e 2 e 2
t q
t .q k q
t = 1 +
(5)
k2 stands for rate constant (g(mgmin)-1) value. The constants qe and k2 are obtained from t/qt
against t. Alternatively, k2 is computed using t1/2 (half-adsorption time) expressed in equation (6).
The time for half of the maximal mass of metal ion adsorbed is indicated as t1/2.
e 2 2
1 k q
t = 1 (6)
Values of t1/2, as well as two models variables, are reported in Table 2.
Table 2 The adsorption kinetics parameters at different concentrations and pH
pH Initial conc.
(mgL-1)
t1/2 (min)
qe(exp)a (mgg-1)
First-order Second-order
k1x 102 (min-1)
qe(cal)b (mgg-1)
r2 k2 x 104 (gmg-1min-1)
qe(cal)b (mgg-1)
r2
10 84.28 50 1.54 42.77 0.958 2.46 48.31 0.987
3 20 71.92 81 1.04 64.72 0.9991 1.68 82.64 0.9967
30 50.95 117 0.85 91.30 0.9553 1.6 119.05 0.9954
10 129.92 43 0.45 38.70 0.9782 1.85 41.67 0.9951
7 20 96.77 74 0.56 65.01 0.9742 1.41 73.53 0.9913
30 80.75 99.05 0.64 77.39 0.9553 1.20 101.01 0.9933
a: experimental data
b: calculated data from models
The sorption data time relation for As(III) on IZO is displayed in Figure 5 for three initial concentrations and the optimum pH of 3 and 7 for all experiments. In all experiments, results showed starting speedy rise in adsorption capacity qt, that becomes steady after about 180 minutes.
Fig.5 Effect of contact time on the adsorption of As(III) on IZO nanoparticles with different initial concentrations of As (III) ions pH 3.0 and 7.0 at 25 oC.
The relevance of kinetics data was negotiated by the value of the coefficient r2 tabulated in Table 2. Remarkably, the r2 values for the pseudo-first-order law is always <0.98, (Fig.6(a)), which indicates poor correlation. The data exhibits a significant difference between the theoretical qe value and that obtained practically.
In contrast, using the pseudo-second-order representation resulted in better regression coefficients values i.e. all >0.99 (Fig.6(b)). Moreover, the qe value obtained experimentally matches that derived mathematically, as reported in Table 2; therefore, it can be concluded that the As(III) adsorption follows the second-order law. In literature, many studies have indicated similar finding [26, 34, 35].
Fig.6 Pseudo-first order (a) and pseudo-second order (b) kinetic equation for adsorptions of As(III) on indium doped zinc oxide at different initial concentrations pH 3.0 and 7.0 at 25 oC.
3.2.4 Mechanism of adsorption
The adsorbed species may also be transported from the solution bulk to the solid phase by intra-particle diffusion/transport operation. The intra-particular diffusion is a controlling step of a number of adsorption processes. The likelihood of intra-particular diffusion is investigated by the Weber and Morris diffusion mode [36 -38]:
C .t k
qt = dif 1/2 + (7)
0 20 40 60 80 100
0 50 100 150 200 250 300 350
% Removal
t, min
pH 3: 10 mg/L pH 3: 20 mg/L pH 3: 30 mg/L pH 7: 10 mg/L pH 7: 20 mg/L pH 7: 30 mg/L
0.0 2.0 4.0 6.0 8.0 10.0 12.0
0 50 100 150 200 250 300 350
t/qt(min.g/mg )
t, min (b)
pH 3: 10 mg/L pH 3: 20 mg/L pH 3: 30 mg/L pH 7: 10 mg/L pH 7: 20 mg/L pH 7: 30 mg/L
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
0 50 100 150 200 250
Ln (qe-qt)
t, min (a)
pH 3: 10 mg/L pH 3: 20 mg/L pH 3: 30 mg/L pH 7: 10 mg/l pH 7: 20 mg/L pH 7: 30 mg/L
The values of C and the intra-particles speed constant kdif, are respectively, derived from the slope and intercept of qt against t1/2 graph. The value of the constant C stipulates the solution boundary layer thickness. The kdif values for the arsenic ions adsorption at two pH values are computed and reported (Fig. 7 and Table 3). The uptake of As(III) at the surface of the adsorbent may be governed by the intra-particle diffusion kinetic formula, since, qt and t1/2 hold a linear correlation.
Fig. 7 qt versus t1/2 plot for the intraparticle diffusion
Besides, the regression coefficient values are more than 0.94 denoting the relevance of the data the model. The intra-particle diffusion diagrams are shown in Fig. 7, where the main parameters of this model are determined and gathered in Table 3. The thickness of boundary layer is strongly correlated to the intercept values. The larger intercept of the graph (C value, Table 3) signifies a greater boundary layer effect.
Table 3 The intraparticle diffusion model constants for the As(III) ions adsorption onto IZO.
pH kdif1, mg/g.min1/2 C r2 3 6.5391 10.806 0.9359 7 3.5771 25.984 0.9755
As can be seen from Fig. 7, the graph is linear, with a graph not passing through the origin. This observation can be attributed to some level of boundary layer control. Such behavior is an indication that the intra-particle diffusion is not the sole rate controlling step, as other kinetic processes may influence the adsorption rate of adsorption. In other words, all of these operations are operating concurrently[39].
3.2.5 Equilibrium study
As(III) adsorption data were analogized with the Langmuir, Freundlich, and D-R models:
.b C Q
Q q C
0 e 0 e
e = 1 + 1 linear form of Langmuir equation (8)
k log nlogC
ogq
l e =1 e+
linear form of Freundlich equation (9)
βε2
−
= max
e logq
q
log linear form of D-R equation (10)
qt(pH 3)= 6.5391 t1/2 + 10.806 r² = 0.9359
qt(pH 7) = 3.5771 t1/2+ 25.984 r² = 0.9755
0 15 30 45 60 75 90 105 120
0 5 10 15 20
qt(mgg-1)
t1/2(min1/2) pH 3: 30 mg/l
pH 7: 30 mg/l
The mass of As(III) adsorbed by one gram of IZO is qe(mg g-1), the solute concentration Ce is expressed in (mgL-1), Q0 is the concentration of solid-phase required to form a monolayer on adsorption sites [28], and b is a coefficient associated with the free energy of adsorption(∆Gads).
The terms k and n of Freundlich equation are connected with the strength and distribution of the adsorptive bond [32]. The coefficient β is related to (∆G) of adsorption (mol2kJ-1), andε is the Polanyi potential (kJmol-1) expressed as:
C ) log(
RT
e
1+ 1
ε = (11)
Langmuir model suggests no dependence of (∆Gads)on surface coverage. This model also anticipates solid surface saturation (Q0=qe) due to adsorbate monolayer coverage at large Ce
values, while at small Ce linear adsorption takes place. The constants Q0 and bare derived from the slope and intercept of (Eq. 8) graph. The Langmuir representation is also indicated by a dimensionless separation coefficient RL that delineates the category of isotherm:
0
L b.C
R = + 1
1 (12)
where, the term C0represents the solute concentration at t = 0. The value of RL confirms the viability of adsorption. The process is irreversible, favorable, linear or unfavorable, when RL
values are = 0, <1, = 1 and >1 respectively.
The coefficients k and ncan be derived from the slope and intercept of log qe against log Ce
graph respectively (Eq. 9). The term k represents the amount of solute adsorbed when Ce= 1, whereas, the 1/n value quantifies sorption strength and surface heterogeneity [40]. When n = 1, the two phases separation concentration independent. The case of n >1 is the most ordinary and correlates with the L-type Langmuir model[41], while n < 1 is indicatory of collective sorption [42] bringing in a strong adsorbate molecules interactions. For the D-R model, the slope of the log qe versus ε2 (Eq. 10) gives the value of β. Knowledge of the value of this coefficient permits calculation of the mean value of activation energy of adsorption E, which is indicative of the physical or chemical nature of adsorption [43]. The value of E can be determined from the following equation [44]:
β
2= 1
E (13)
From the equilibrium study, isotherm data were obtained by testing the two optimum values of pH, i.e., 3 and 7, and then pH effect was determined at As(III) concentrations ranging from 5 to 55 mgL-1, 25 °C and 4 h contact time. The three isotherms are graphically plotted in Fig.
8 the two pH values. The equilibrium constants calculated by linearized Langmuir, Freundlich, and D-R equations are given in Table 4.
Table 4 Langmuir and Freundlichand Dubinin-Radushkevich models parameters, RL and regression coefficients.
pH
Langmuir constants Freundlich constants Dubinin-Radishkevich constants qe(mgg-
1)
b (Lmg-
1)
r2 RL n K r2
qmax (mgg-
1)
β(mol2kJ-
1)
E (kJmol-
1)
r2
3 104.17 0.3902 0.9931 0.0457 2.63 56.86 0.922 68.35 0.1628 1.75 0.9115 7 83.34 0.3221 0.9998 0.0454 2.51 25.57 0.9142 55.92 0.1927 1.61 09179
Finding out the model that gives the best fitting to the adsorption of As(III) on the nano- particles powder is based on the value of the correlation coefficient (r2).Having a look at Table 3, the (r2) values for the Langmuir model is the highest (≈ 1.0),denoting its best fit to adsorption data at both pH values. Similarly, Zeng [7] investigated the adsorption of arsenic on Fe(III)-Si binary oxide adsorbent and found a good matching of his data with the Langmuir isotherm. Meanwhile, Kelly and Tarek [8] showed that isotherm data of the adsorption of arsenite onto iron-treated activated carbon and zeolites were better fitted by the Freundlich equation. Table 3 indicates that the Langmuir, Freundlich, and D-R parameters Q0, k, and qmax, respectively, were higher at pH 3.0 than at pH 7.0. This finding confirmed that at this acidic pH, As(III) capacity adsorption was larger. Moreover, maximum adsorption capacities estimated from the Langmuir model were 104.17 and 83.34 mg of arsenite per gram of nanoparticles at pH 3 and 7, respectively. Table 3 also shows that the very low values of the separation factor, RL, attested to the favorable sorption power of IZO nanopowder. At both pH values, the n values greater than 1 are indicative of a Langmuir process and of very weak interactions between molecules of solute[32]. However, the influence of the experimental conditions upon the parameter n is not obvious. According to the coefficient RL values, the present adsorption systems revealed all favorable with IZO. Moreover, the very low values of free energy 1.75 and 1.61 kJmol-1, estimated from D-R model, for pH 3 and 7 respectively, are indicative of a physisorption process [40 – 45].
Fig 8 Langmuir (a) and Freundlich (b)D-R (c) linear isotherms at 25 °C for the adsorption of As(III) at : (♦) pH =3.0 and(▲) pH = 7.0.
3.2.5 Effect of temperature in the adsorption process
Temperature is a detrimental factor in adsorption, therefore its effect was studied at 298, 313 and 328 K. The As(III) removal increases with the rise in temperature as displayed by the Langmuir graph (Fig. 9). The compliance of the adsorption data with Langmuir model for the chosen temperatures is evidenced by the correlation coefficient (r2) values shown in Table 5.
0.00 0.06 0.12 0.18 0.24 0.30 0.36
0 5 10 15 20 25 30
Ce/qe, (g/l )
Ce, mg/l (a)
1.0 1.2 1.4 1.6 1.8 2.0
-0.35 -0.15 0.05 0.25 0.45 0.65 0.85 1.05 1.25 1.45
Log qe
Log Ce (b)
2 2.5 3 3.5 4 4.5
0 2 4 6 8 10
Log qe
ε2(kJ2.mol-2) (c)
Fig. 9 Langmuir isotherms for adsorption of As (III) onto doped indium ZnO nanopowder at different temperatures
Table 5 Equilibrium constants for the removal of heavy metal ions at different temperatures.
T(K) Langmuir constants Freundlich constants
qm(mgg-1) b(Lmg-1) RL r2 n kf r2
298 104.2 0.390 0.9097 0.9931 2.630 56.8 0.9220
313 125.0 0.534 0.9241 0.9930 2.645 114.2 0.9260
328 142.9 0.875 0.9360 0.9920 2.670 118.4 0.9590
The Freundlich constant n value is more than 1.0 for the three chosen temperatures. The larger deviation of n from unity indicates greater distribution of surface bond energies of arsenic adsorption on indium doped zinc oxide nanopowder [41]. The values of the Freundlich constant n are greater than the unit. Therefore, nanoparticles ZnOIn 3% exhibts a wider As(III)-surface bond energies distribution as a result of temperature rise[41].
The thermodynamic functions: enthalpy change (∆H0), free energy change (ΔG0) and entropy change (ΔS0) for the adsorption of As(III) by the adsorbent are obtained using the relations:
a 0 R.T.lnK
ΔG =− (14)
T ΔG
ΔS0 = ΔH0− 0 (15)
The Ka value is computed by the formula Ka = Q0·b. The ΔH0value the slope of ln(Ka)against T-1 graph (Figure 10), while ΔG0 and ΔS0 are obtained using Eq. 14 and Eq. 15.
Fig. 10 Plot of ln Ka versus the reciprocal temperature of arsenic ion adsorption by ZnOIn 3% nanopowder
0 30 60 90 120
0 7 14 21 28
qe, mg/g
Ce, mg/L
T = 298K T = 313K T = 328K
Thermodynamic parameters are gathered in Table 6. Positive ∆H0 indicates the endothermic character of the adsorption process. This is clearly evidenced by the enhancement of arsenic ions removal at elevated temperatures. Also the ∆H0 value suggests an electrostatic attraction between As(III) ions and ZnOIn in a physical adsorption process. Moreover the positive value of ∆H0 may be attributed to the eviction of hydration water molecules from the metal ions as well as solid–solution interface [46]. Thus the energy released when the ions are attached to the adsorbent surface will be used up by this dehydration process resulting in an endothermic sorption process [47] confirming the temperature study results. Moreover an increase in chaos at the interface between the solid and solution is revealed by the positive ∆S0 values.
Table 6 Thermodynamic parameters for As(III) adsorption Temperature
(K)
K ΔG0
(kJmol-1)
ΔS0 (kJmol-1K-1)
ΔH0 (kJmol-1)
r2
298 1.1578 -9.637 0.1327
30.355 0.991
313 1.3469 -11.649 0.1319
328 1.5934 -13.663 0.1327
The data shown in Table 6, shows that ∆G0 is negative with further decrease in its values as the temperature is increased. This designates the spontaneous character of the adsorption process, as well as its being favorable at elevated temperature.
4. Comparative study
To ability of IZO nanopowder to eliminate As(III) is contrasted with that of other adsorption materials reported in literatures as recorded in Table 7. The juxtaposition between the findings obtained in this study with other adsorbents, manifests that IZO is highly effective in arsenic removal from aqueous solutions.
Table 7: A Comparison of IZO adsorption capacity with some nano-adsorbents for As(III)
Adsorbent qe (mg·g−1) Temp. (K) pH Ref.
Fe(III)-Si Binary Oxide 21.54 298 7 [7]
Activated carbon - 298 7 [9]
Hazelnut shell 0.714 314 4-9 [11]
Char carbon 89 298 2-3 [48]
Activated carbon 29.9 298 6.4-7.5 [48]
TiO2 59.93 298 7 [49]
TiO2 (Hombikat UV 1000) 22.7 295 4 [50]
MnO2-leaded resin 53 295 7-8.5 [51]
Zr resin 79.42 298 8 [52]
IZO 104 298 3 Present work
- 125 313 3 -
- 142 328 3 -
- 84 298 7 -
5. Conclusions
This work described the elimination of As(III) ions from aqueous solutions using IZO nanopowder prepared by modified sol–gel method. The effects the initial arsenite concentration,
pH, temperature and indium-doping ratio were evaluated through batch-mode experiments. Results showed that the removal efficiency of As(III) was strongly pH dependent and optimum at pH 3 and 7. The adsorption equilibrium was found to comply with the Langmuir adsorption isotherm, with an utmost adsorption capacity of about 104.17 and 83.34 mgg-1, at pH 3 and 7, respectively.
IZO nanopowder was effective for the As(III) removal, and data were found to comply with the pseudo-second-order kinetic model. The process of arsenite ions adsorption was an spontaneous endothermic physisorption.
Acknowledgements
This work was supported by the National Plan, for Sciences, Technology and innovation, at Al-Imam Mohammed Ibn Saud Islamic University, college of Sciences, Kingdom of Saudi Arabia.
References
[1] F.W. Pontius, K.G. Brown, C.J. Chen, Am. Water Works Assoc. 86, 52 (1994).
[2] M.R.Karagas, T.A.Shukel, J.S. Morris, T.D.Tosteson, J.E. Weiss, S.K. Spencer, E.R.Greenberg, Am. J. Epidemiol. 153, 559 (2001).
[3] M.R.Karagas, T.D.Tosteson, J.Blum, Environ. Health Perspect. 106, 1047 (1998).
[4] Y. M.Kuo, C. W. Liu, K. H.Lin,Water Res. 38, 148 (2004).
[5] H.Takanashi, A. Tanaka, T. Nakajima, A.Ohki, Water Science and Technology 50(8), 23 (2004).
[6] I.N. Bhattacharya, P.K.Gochhayat, S. Paul, P.K.Mitra, The European Journal of Mineral Processing and Environmental Protection 5(1), 1303 (2005), pp. 48 (2005).
[7] L.Zeng, Water Qual. Res. J. Canada, 39(3), 267 (2004).
[8] B. P. Kelly, M. A.Tarek, Journal of Environmental Science and Health, 40,723 (2005).
[9] W.Yunhai, M.Xuhong, F. Maowu, L. Minmin, Journal of Hazardous Materials 159, 380(2008).
[10] A. M. Bruce, S. Goldberg. Clays and Cloy Minerals, 44(5), 609 (1996).
[11] D. N.Pravin, A. M. Kadam, H. S. Shankar, Journal of environmental biology, 30(4), 499 (2009).
[12] J. Benny, P. K. Manoj, V. K.Vaidyan, Bulletin of Materials Science, 28(5), 487 (2005).
[13] S.Girjesh, S. B.Shrivastava, J.Deepti, S.Pandya, T. Shripathi, V.Ganesan, Bulletin of Materials Science 33(5), 581 (2010).
[14] M. R.Rezaee, M.Behdani, H. Arabshahi, N.Hosseini, International Review of PHYSICS, 12(3), 103 (2009).
[15] Y.W. Chen, Y. C. Liu, S. X Lu, C.S.Xu, C. L. Shao, C Wang, J.Y. Zhang, Y.M. Lu, D. Z. Shen, X.W.Fan, J Chem Phys., 123(13), 134701-(1-5) (2005).
[16] L. El Mir, I. El Ghoul, S.Alaya, M. Ben Salem, Barthou C. abd Von Bardeleben H. J.
J. Physica B, 403, 1770 (2008).
[17] L. El Mir, Z. Ben Ayadi, M.Saadoun, H.J. Von Bardeleben, K. Djessas, A.Zeinert, Phys. Stat. Sol. (a), 204, 3266(2007).
[18] B.C. Lippens, J.H. De Boer, J. Catalysis, 4, 319 (1965).
[19] R.Biswal, A. Maldonado, J. Vega-Pérez, D. R. Acosta, M.De La Luz Olvera, Starting from Zinc Acetylacetonate and Indium Chloride. Materials, 7, 5038 (2014).
[20] W. Tang, D.C.Cameron, Thin Solid Films. 238(1), 83 (1994).
[21] M.Hjiri, R.Dhahri, K.Omri, L.ElMir, S.G.Leonardi, N.Donato, G.Neri, Materials Sciencein Semiconductor Processing 27,319 (2014).
[22] Z. Ben Ayadi, L.ElMir, K. Djessas, S.Alaya,Thin SolidFilms, 517, 6305 (2009).
[23] F.Rouquerol, J.Rouquerol, K.Sing, Adsorption by Powder and Porous Solids: Principles, Methodology and Applications, Academic Press. 1999.
[24] N. Al Dahoudi, A.AlKahlout, S.Heusing, P.Herbeck-Engel, R. Karos, P.Oliveira,
J. Sol-Gel SciTechnol, 67, 556 (2013).
[25] A.Alkahlout, N. Al Dahoudi, I.Grobelsek, M. Jilavi, P. W.de Oliveira, Journal of Materials, 2014, Article ID 235638, 8 pages. 2014.
[26]
5, 4138 (2015).
[27] S. Garcia, S.Sardar, S. Maldonado, V. Garcia, C.Tamez, J.G. Parsons, Microchemical Journal 117, 52 (2014).
[28] C.P. Huang, E.H.Smith, Chemistry in water reuse, Ann Arbor Science Publishers, Ann Arbor, Michigan, 2. 1981.
[29] M. Jang, S.H. Min, T.H. Kim, J.K.Park Environ. Sci. Technol. 40, 1636 (2006).
[30] T. M.Thi, N. T. H. Trang, N. T. V.Anh, Applied Surface Science 340, 166 (2015).
[31] D.Morillo, G. Pérez, M. M.Valiente, Journal of Colloid and Interface Science 453, 132 (2015).
[32] V.K. Gupta, S. Sharma, I.S. Yadau, M.Dinesh, Journal Chemical Technology and Bio-thechnol. 71, 180 (1889).
[33] Y.S. Ho, G. Mckay, Chemical Engineering Journal, 70, 115(1998).
[34] K.J. Reddy, K.J. McDonald, H.King, Journal of Colloid and Interface Science 397, 96 (2013).
[35] S. Kong, Y. Wang, Q. Hu, A. K.Olusegun, Colloids and Surfaces A: Physicochemical and Engineering Aspect, 457, 220 (2014).
[36] W. J. Jr. Weber, J. C.Morris, J. Sanitary Eng. Div. Proceed. Am. Soc. Civil Eng.
89, 31 (1963).
[37] G.McKay,J. Chem. Technol. Biotechnol. A 33, 196 (1983).
[38] G. McKay, M.S. Otterburn, A.G.Sweeney, Water Res. 14, 15 (1980).
[39] M.Arami, N. Y Limaee, N. M. Mahmoodi, Chem. Eng. J. 139(10), 2 (2008).
[40] M.M.Benjamin, Effects of competing metals and complexing ligands on trace metal adsorption at the oxide/solution interface, PhD Thesis, Standford University 1878.
[41] B.E. Reed, M.R.Matsumoto, Separation Science and Technology, 28(13 & 14), 97 (1993).
[42] F. Haghesresht, G.Lu, Energy Fuels 12, 1100 (1998).
[43] V.B.H. Dang, H.D. Doan, V. T. Dang, A.Lohi,Biores. Technol. 100, 211 (2009).
[44] MHorsfall, A. I. Spiff, A.A.Abia, Korean Chemical Society, 25(7), 969 2004.
[45] J. Monika, V Garg, K.Kadirvelu, J. Hazardous materials 162, 365 (2009).
[46] M. Ricci, P. Spijker, K. Voıtchovsky, Water-induced correlation between single ions imaged at the solid–liquid interface. Nat. Commun. 5, 4400 doi: 10.1038/ncomms5400 (2014).
[47] E.Tipping, Interactions of organic acids with inorganic and organic surfaces, in: E.M. Perdue, E.T. Gjessing (Eds.), Organic Acids in Aquatic Ecosystems. Life Sciences Research Reports, Wiley, New York, 209–221. 1990.
[48] J.Pattanayak, K.Mondal, S. Mathew, S.B.Lalvani, Carbon 38, 589 (2000).
[49] M.E. Pena, G.P.Korfiatis, M. Patel, L. Lippincott, X.Meng, Water Res. 11, 2327 (2005).
[50] P.K.Dutta, A.K. Ray, V.K. Sharma, F.J.Millero, J. Colloid Interf. Sci.278, 270 (2004).
[51] V.Lenoble, C.Chabroullet, R. Al Shukry, B.Serpaud, D. Véronique, J. C.Bollinger J. Colloid Interf. Sci.280 (1), 62 (2004).
[52] T.M. Suzuki, J.O.Bomani, H. Matsunaga, T.Yokoyama, React. Funct. Polym., 43(1–2),165 (2000).
