Chemical Engineering Journal 217 (2013) 1–9
Contents lists available at SciVerse ScienceDirect
Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej
Synthesis and characterization of mesoporous alumina and their performances for removing arsenic(V) Caiyun Han, Hongying Li, Hongping Pu, Hongli Yu, Lian Deng, Si Huang, Yongming Luo ⇑ Faculty of Environmental Science and Engineering, Kunming University of Science and Technology, Kunming 650500, PR China
h i g h l i g h t s " Arsenic is a toxic and carcinogenic element, and adsorption has attracted much attention. " Mesoporous alumina was used as an effective adsorbent for removing arsenic(V). " Arsenic species, pH and adsorption mechanism were investigated and illustrated. " Calcination temperature plays a crucial role in determining the performance of adsorbent.
3
" The presence of SiO4 , PO4
a r t i c l e
and F- caused the sharp reduce on arsenic(V) removal.
i n f o
Article history: Received 13 August 2012 Received in revised form 16 November 2012 Accepted 17 November 2012 Available online 29 November 2012 Keywords: Mesoporous alumina Removal arsenic Adsorption kinetics Thermodynamics
a b s t r a c t Mesoporous alumina, synthesized by combining the three-block copolymer Pluronic P123 as a template with aluminum hydroxide sol at room temperature followed by isolation and calcining, was employed as an effective adsorbent for removing arsenic(V) in the pH region of 2.5–7.0. Arsenic adsorption data of MA(400) are well fitted by the Langmuir isotherm model and the maximum adsorption capacity is 36.6 mg/g at near neutral (pH = 6.6 ± 0.1). Calcination temperature plays an important role in determining the performance of MA(400), MA(600) and MA(800) for removing arsenic(V), and the corresponding As(V) removal is in the order of MA(400) >> MA(600) > MA(800). The kinetics data were well fitted to pseudo-second-order, which implies that ‘‘surface reaction’’ might be the rate limiting step. Thermodynamic parameters illustrated that As(V) adsorption over MA(400) was a spontaneous and endothermic process. Adsorption energy (2.61 kJ/mol) is less than 8 kJ/mol, indicating the adsorption process may be dominated by physisorption. The influence of coexisting anions on As(V) removal demonstrated that the 2 3 removal was slightly affected by the presence of NO 3 and SO4 , while the presence of SiO4 , PO4 and F- caused a sharp fall in removal effectiveness, especially when the ratio of coexisting anion concentration to As(V) was larger than 1.12. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Human activities and natural phenomena have caused releases of arsenic (As) into groundwater and surface water, creating potentially serious environmental problems for humans and other living organisms. It is well-documented that arsenic contamination has led to serious health risks such as cancer in lung, bladder, kidney and skin, as well as other skin changes for many countries all over the world [1]. Current estimates are that, e.g., 35–50 million people in the West Bengal and Bangladesh area, more than 10 million people in Vietnam, and over 20 million people in China are exposed to harmful arsenic intake through potable water consumption [2,3]. It is well-known that the inorganic arsenic species (arsenate and ⇑ Corresponding author. Tel./fax: +86 871 5103845. E-mail address:
[email protected] (Y. Luo). 1385-8947/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2012.11.087
arsenite) are the predominant species in most environments and the corresponding toxicities of inorganic arsenic species are much higher than those of organic arsenic species [4–6]. Therefore, there is an urgent demand for developing highly effective, reliable, and economical techniques to remove inorganic arsenic species from contaminated groundwater and surface water. Compared with other conventional techniques including oxidation, coagulation/ precipitation, filtration, ion exchange, membrane/reverse osmosis and biological treatment, adsorption has attracted much attention due to the following advantages: (i) it usually does not need a large volume and additional chemicals, (ii) it is easier to set up as a POE/ POU (point of entry/point of use) arsenic removal process [7], (iii) it does not produce harmful byproducts [8,9] and can be more costeffective [10]. In general, the adsorbent plays a key role in determining the performance of an adsorption system. Activated alumina is one
2
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9
of the most common adsorbents for removing arsenic from contaminated water [11–14]. However, the adsorption capacity and adsorption rate of conventional activated alumina (CAA) for removing arsenic is relatively low and slow, which has been attributed to its ill-defined pore structure together with small surface area. A perfect example can be found in Lin and coworker’s report that the maximum As(V) adsorption capacity of granular activated alumina with a surface area of 118 m2/g (Macherey–Nagel, Germany) is 15.9 mg/g [12]. Mesoporous materials have attracted considerable attention in the region of chemical engineering, optics, electromagnetics, biomedicine, industrial catalysis and adsorption, environmental protection, and fabrication of novel nano-object materials due to high surface area, well-defined and adjustable pore diameter of 2–50 nm [15,16]. Therefore, it could be deduced that mesoporous alumina (MA) should be an ideal adsorbent for removing arsenic. Of late, limited efforts have been made to synthesize appropriate mesoporous alumina for removing As(V), which included high-temperature crystallization [17] or the use of expensive aluminum alkoxide (such as aluminum tri-sec-butoxide) and organic solvent (2-butanol) [18]. Despite significant progress in the synthesis of mesoporous alumina, further studies in this area, especially in relation to low-temperature synthesis route by using low-cost, nontoxic and biodegradable templates are desirable. In this contribution, mesoporous alumina has been synthesized by using non-ionic triblock copolymer EO20PO70EO20 (Pluronic P123) and aluminum tri-isopropoxide (AIP) as a structure-directing agent and an aluminum source, respectively, and the aforementioned materials were characterized by N2 adsorption–desorption, XRD, SEM, TEM and FT-IR. The performance of mesoporous alumina for removing arsenate from contaminated water was evaluated and compared with commercial activated alumina (CAA) and other conventional adsorbents. Various affecting parameters such as the calcination temperature in the preparation of the adsorbent, initial arsenic concentration, pH, time, coexisting anion, thermodynamics, adsorption kinetic and energy were investigated in detail. The Mineql+ program was used to define the species of arsenate in different pH range, and the adsorption mechanism was also discussed.
cined at 400 °C, 600 °C and 800 °C for 5 h, respectively. Mesoporous alumina samples synthesized via this route are designated as MA(x), where ‘‘x’’ is calcination temperature. For comparison purposes, commercial activated alumina was also calcined at 400 °C in air for 5 h before being used to adsorb As(V), and is nominated as CAA(400).
2. Experimental
Qt% ¼
2.1. Materials Nonionic triblock copolymer EO20PO70EO20 (Pluronic P123) was purchased from Sigma–Aldrich. Aluminum tri-isopropoxide (AIP), concentrated hydrochloric acid (37 wt.%), nitric acid (65 wt.%), sodium nitrate, sodium phosphate, sodium fluoride and sodium sulfate were obtained from the Shanghai Chemical Regent Company of China. 2.2. Synthesis of mesoporous alumina Mesoporous alumina was synthesized by using P123 and AIP as a structure-directing agent and an aluminum source, respectively. In the typical synthesis batch, 20.4 g of AIP and a small amount of nitric acid (65 wt.%) were added into 160 mL of hot denionized water. Meantime, 7.5 g of P123 and 0.24 mol of HCl were added into 150 mL of denionized water under continuous stirring under room temperature (RT). After P123 completely dissolved, the aluminum hydroxide sol was added dropwise into the P123 solution, and the mixture was stirred at 40 °C for 12–24 h. Then, the pH value of the mixture was adjusted to 7.0 by using sodium hydroxide solution. The resultant mixture was kept at RT under static conditions for 2–3 days, and the reaction products were filtered, washed and dried at 105 °C for 24–48 h. Finally, the dried sample was cal-
2.3. Characterization of mesoprous alumina BET surface area and N2 adsorption–desorption isotherms were determined on an ASAP 2020 apparatus at 196 °C. All samples were degassed at 250 °C for 2 h prior to analysis. The BET specific surface area was calculated from adsorption data in the relative pressure range from 0.05 to 0.25. X-ray diffraction (XRD) patterns were performed on a Rigaku D/max 2550PC diffractometer. Scanning electronic microscopy (SEM) and transmission electron microscopy (TEM) images were obtained on a Philips XL-30 environmental scanning electron microscope and a JEM-2010 HR transmission electron microscope, respectively. Fourier transform infrared (FT-IR) spectra of the samples in the form of KBr pellets were recorded by using a Nicolet 560 IR spectrometer.
2.4. Arsenic(V) adsorption The batch experiments of arsenic adsorption were carried out with mixed adsorbent and arsenic solution in a series of 250 mL conical flasks under magnetic stirring conditions, and the resulting mixture was centrifuged after adsorption. pH of the solution was adjusted by adding HCl or NaOH. The arsenic concentrations before and after adsorption were measured by an atomic fluorescence spectrometer (AFS-2301). The adsorption capacity and removal percentage of As(V) were calculated by using the following equations:
qt ¼
ðC 0 C t Þ V m ðC 0 C t Þ 100 C0
ð1Þ
ð2Þ
where qt and Qt denote the As(V) uptake capacity (mg/g) and removal percentage at time t (min), respectively. C0 and Ct are the initial and equilibrium concentration (mg/L) of As(V) solution at time t, respectively. V is the volume of adsorption solution (L), m is the weight of the adsorbent (g). Adsorption kinetic experiments were carried out by fixing adsorbent dosage (0.05 g) with the contact times in the range of 0.5–12 h, and two groups of experiments were performed: (i) effect of initial arsenic(V) concentration: the adsorbent was added in 50 mL of arsenic solution with various initial concentrations (11.175, 22.352, 31.806, 44.703 and 69.005 mg/L) under RT; (ii) effect of pH: 50 mL of solution with several pH values (3.0, 4.5, 6.6 and 8.5) and arsenic initial concentration of 44.703 mg/L was measured under RT. The thermodynamic studies were carried out by varying the adsorption temperature at 298, 313 and 333 K. The other operating parameters were as follows: 0.1 g adsorbent, 50 mL of adsorption solution with arsenic initial concentration of 44.703 mg/L, the pH was kept at 6.6 ± 0.1. The spent MA samples were added into deionized water or sodium hydroxide solution (concentration ranging from 0.001 to 0.1 M) under stirring for 12–24 h with the ratio of solid to liquid at 1.0 g/L, and the mixture was filtered and dried at 110 °C for 12–24 h.
3
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9
3. Results and discussion
Table 1 BET surface area of MA(400), MA(600), MA(800) and CAA.
3.1. Characterization
Adsorbent 2
BET surface area (m /g)
Fig. 1 illustrates the N2 adsorption–desorption isotherm and Barrett–Joyner–Halenda (BJH) pore size distribution curve of MA(400). According to the classification made by IUPAC [19], the N2 adsorption–desorption isotherm of MA(400) is found to be of type IV. Two well distinguished regions in the adsorption isotherm are evident: (i) monolayer–multilayer adsorption and (ii) capillary condensation. The first increase in adsorption at relative pressure P/P0 < 0.1 is due to multilayer adsorption on the surface while the second increase at P/P0 = 0.60–0.85 arises from capillary condensation in the mesopores with nitrogen multilayers adsorbed on the inner surface [20]. Moreover, a clear type-H1 hysteresis loop is observed for MA(400). As can be seen from Fig. 1(inset), a relatively narrow pore size distribution is detected, and the average BJH pore diameter calculated by using the adsorption branch data of the N2 adsorption–desorption isotherms is about 5.8 nm. These above results indicate that MA(400) displays the typical behavior of a mesoporous structure. Table 1 gives the BET surface area of MA(400) and compares it with MA(600), MA(800) and CAA(400). The BET surface area decreases as calcination temperature increases from 400 to 800 °C, and BET surface area of MA(400) is also much higher than that of CAA(400). To further demonstrate the presence of a mesoporous structure, MA(110) and MA(400) were characterized by small angle X-ray diffraction (in the range of 2h = 0.5°–8.0°), and the corresponding patterns are presented in Fig. S1. The presence of only a single diffraction peak in 2h region below 2° for both MA(110) and MA(400) is indicative of a mesostructure [21,22]. Compared with MA(110), the diffraction peak of MA(400) shifts slightly to larger 2h value, which might be attributed to the constriction of the unit cell during calcination. Analogous phenomena were observed for mesoporous silica such as SBA-15 [23,24], SBA-16 [25] and other oxides [26]. The morphologies of MA(400) were characterized by scanning electronic microscopy (SEM), and the corresponding images are displayed in Fig. S2. In low magnification image (Fig. S2A), the particles with various irregular morphologies are observed for MA(400). However, it is noted that these particles are composed of many porous subunits with high surface area, as further evidenced by high magnification SEM image (Fig. S2B). Fig. 2 provides TEM images of MA(400). As shown in Fig. 2, the pore structures of MA(400) reflect the wormlike motif of the
a
MA(400)
MA(600)
MA(800)
CAAa
312
209
169
160
Commercial activated alumina.
Fig. 2. TEM images of MA(400) obtained on a JEM-2010 HR transmission electron microscope. (A) Low magnification image and (B) high magnification image.
Fig. 1. N2 adsorption–desorption isotherm and BJH pore size distribution curve (inset) of MA(400).
micellar structure (Fig. 2A), which is in good agreement with the small angle XRD characterization result that only one peak is observed in the small angle XRD pattern of MA(400). The pore structures of MA(400) are in marked contrast to the channels with long-range hexagonal arrangement observed for SBA-15. Although long-range packing order is not discernible, the channels in the network have a regular diameter, originating from the random packing of alumina nanorods (Fig. 2B), which is in excellent agreement with the narrow pore size distribution of the sample (Fig. 1 inset). Analogous phenomena have been observed for mesoporous SiO2 [27] and mesoporous alumina synthesized with a cation template such as cetyltrimethylammonium bromide [22].
4
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9
3.2. Comparison of adsorbents It is well-known that the adsorbent is one of the most important factors that affects the behaviors of the adsorption system. Fig. 3 illustrates the performances of stilbite, active carbon, clinoptilolite, CCA(400) and MA(400) for removing arsenic(V). The maximum arsenic(V) removal of MA(400) is 94.7%, which is far higher than CCA(400) and other absorbents. Moreover, it is noted that the adsorption equilibrium time of MA(400) is about 2 h, while the corresponding time of CCA(400) is more than 9 h, thus indicating that adsorption kinetics of MA(400) is much faster than that of CCA(400). The difference between MA(400) and CCA(400) in adsorption kinetics is likely because MA(400) with pore diameter of 5.8 nm is less susceptible to pore blockage, which is favorable for the diffusion, transportation of arsenic species [15,25]. It is well-demonstrated that many properties such as BET surface area, pore size, hydroxyl groups and crystal structure of alumina would be distinctly changed with increasing calcination temperature, which should play a crucial role in determining the behavior of adsorbent. The influence of calcination temperature on the removal of arsenic(V) over alumina samples calcined at different temperatures was investigated, and the corresponding results are presented in Fig. 4. It is noted that the arsenic(V) removal of MA(400), MA(600) and MA(800) under the same conditions is in the order of MA(400) >> MA(600) > MA(800), which is in good agreement with the change of their BET surface area. In order to further explain their difference, MA(400), MA(600) and MA(800) were characterized by XRD and FT-IR, and the corresponding results are presented in Figs. 5 and S3, respectively. Six diffraction peaks (centered at 2h = 32.2°, 36.9°, 39.3°, 45.8°, 60.6° and 66.8°) are well-expressed in the patterns of MA(400) and MA(600), which is an indication of c-Al2O3 [28–30], thus suggesting that the crystal structures of them are the same. The bands centered at 3436 and 1631 cm1 are also observed for both MA(400) and MA(600) (Fig. S3), which are the characteristic vibrations of hydroxyl groups and adsorbed water [31,32], respectively. However, the intensities of these two bands of MA(400) are stronger than those of MA(600), implying that the number of surface hydroxyl groups of MA(400) is more than MA(600). In addition, the BET surface area of MA(400) is much larger than that of MA(600). The difference in As(V) adsorption between MA(400) and MA(600) could be accounted for by high BET surface area and more surface hydroxyl groups of MA(400). Further increas-
Fig. 4. Influence of calcination temperature on the removal of arsenate for (a) MA(400), (b) MA(600) and (c) MA(800). Experimental conditions: adsorbent = 0.1 g, and initial arsenic concentration = 44.703 mg/L, adsorption temperature = 25 °C, pH = 6.6 ± 0.1, solution volume = 50 mL.
Fig. 5. Large angle XRD patterns of: (a) MA(400), (b) MA(600) and (c) MA(800).
ing calcination temperature, some new diffraction peaks are detected in the larger XRD pattern of MA(800), which indicates other crystal structures were formed within MA(800). Moreover, the number of surface hydroxyl groups of MA(800) is less than that of MA(600). In addition, the BET surface area of MA(800) is smaller than that of MA(600). Therefore, the further lowering of arsenic removal efficiency as calcination temperature is increased from 600 °C to 800 °C might be attributed to the synergistic effect of smaller surface area and fewer surface hydroxyl groups accompanying crystal structure change. On the basis of the above comparison and characterization, it can be concluded that mesoporous alumina is a considerably more efficient adsorbent for arsenic than the other adsorbents shown, and that increases of calcination temperature above 400° in its preparation reduce its efficiency. 3.3. Effect of adsorption conditions Fig. 3. Comparison of various adsorbents. (a) Stilbite, (b) active carbon, (c) clinoptilolite, (d) CCA(400) and (e) MA(400). Experimental conditions: adsorbent = 0.2 g (a–c) and 0.1 g (d and e), and initial arsenic concentration = 44.703 mg/L, adsorption temperature = 25 °C, pH = 6.6 ± 0.1, solution volume = 50 mL.
3.3.1. Effect of initial concentration and contact time Fig. 6 illustrates the influence of initial concentration and contact time on MA(400) for removing As(V). The removal efficiency
5
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9
investigate the adsorption characteristics of MA(400). The Langmuir isotherm model is defined and linearized by using the following equations, respectively:
qe ¼
qmax b C e 1 þ b Ce
ð3Þ
Ce 1 Ce ¼ þ qe qmax b qmax
ð4Þ
Here b and qmax are Langmuir adsorption constants. The former is related to the affinity between adsorbate and adsorbent, and the later is the maximum adsorption capacity; Ce and qe are the equilibrium concentration (mg/L) and equilibrium adsorption capacity (mg/g), respectively. The Freundlich isotherm model is depicted and linearized by using the following equations: Fig. 6. Influence of initial arsenic concentration on the removal of arsenic with MA(400). (a) 11.175 mg/L (b) 22.352 mg/L, (c) 44.703 mg/L, (d) 89.415 mg/L and (e)178.816 mg/L. Experimental conditions: MA(400) = 0.1 g, initial pH = 6.6 ± 0.1, adsorption temperature = 25 °C, solution volume = 50 mL.
increases with the decrease in initial arsenic concentration, and the removal percentage of As(V) is near 100% when initial concentration is lower than 22.352 mg/L, thus indicating MA(400) is an excellent absorbent for treating the water containing a low concentration of As(V) such as drinking water, as further confirmed by Fig. 7. With the initial As(V) concentration increasing from 22.352 mg/L to 44.703 mg/L, the As(V) removal percentage of MA(400) decreases from 99.9% to 94.7%. On further increasing initial As(V) concentration, the corresponding removal percentage of MA(400) rapidly decreases. As can be seen from Fig. 6, the equilibrium time increases with initial As(V) concentration. A perfect example can be founded that the equilibrium times for 11.175 mg/L and 44.703 mg/L are 0.5 h and 3 h, respectively. The adsorption isotherm is usually used to describe the relationship between adsorbent and adsorbate under equilibrium conditions, and the corresponding experimental data of MA(400) are provided in Fig. 7. It is noticeable that the adsorption capacity of MA(400) is about 17 mg/g under As(V) equilibrium concentration of 0.85 mg/L. In general, Langmuir and Freundlich isotherms are useful for assessing the potential use of an adsorbent for particular applications. Herein, Langmuir and Freundlich isotherms are used to
qe ¼ K f C e1=n lnðqe Þ ¼ lnðK f Þ þ
ð5Þ 1 lnðC e Þ n
ð6Þ
Here 1/n and Kf are empirical constants incorporating all factors affecting the adsorption process; Ce and qe are the equilibrium concentration (mg/L) and equilibrium adsorption capacity (mg/g), respectively. Linearized isotherms of the Langmuir and Freundlich models are depicted in Fig. S4, and the calculated constants of them are tabulated in Table 2. The experimental data of MA(400) for As(V) adsorption are better fitted by the Langmuir isotherm than Freundlich isotherm, which indicates that monolayer adsorption occurs on the MA(400) surface for arsenic removal. The maximum As(V) adsorption capacity (qmax) of MA(400) at near neutral (pH = 6.6 ± 0.1) rather than at optimal pH value is 36.6 mg/g, which is higher than or compatible with the results reported in previously. A perfect example can be found in Lin and coworkers’ report that the maximum As(V) adsorption capacity of granular activated alumina (Macherey–Nagel, Germany) is 15.9 mg/g [12]. 3.3.2. Effect of pH It was well demonstrated that the solution pH plays an important role in determining the behaviors of adsorbent. The effect of pH on MA(400)’s performance for removing As(V) was investigated. As shown in Fig. 8, the As(V) removal of MA(400) is above 90% in the pH region between 2.5 and 7.0 and the maximum As(V) removal of MA(400) is about 98.7%. The effect of pH value on MA(400) is markedly different from CCA whose optimum pH for removing As(V) is in a narrow range of 5.5–6.0 with the removal of As(V) sharply decreasing beyond the range [31]. 3.4. Kinetic study In order to investigate the adsorption mechanism of As(V) on MA(400), the experimental data were analyzed by using pseudofirst-order, pseudo-second-order and intra-particle diffusion kinetic models. As for pseudo-first-order, the Lagergren rate equation is depicted as the following equation:
logðqe qt Þ ¼ log qe
K1 t 2:303
ð7Þ
Table 2 Parameters of the isotherm equations.
Fig. 7. Arsenic(V) adsorption isotherm of MA(400). Experimental conditions: MA(400) = 0.1 g, initial pH = 6.6 ± 0.1, adsorption temperature = 25 °C, solution volume = 50 mL, initial arsenate concentration range 11.175–178.816 mg/L.
Isotherm
Parameters
R2
Langmuir Freundlich
qmax = 36.6 mg/g, b = 0.50 L/mg n = 5.18, Kf = 16.26
1.0 0.95
6
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9
Fig. 8. The effect of pH on arsenate removal using MA(400). Experimental conditions: MA(400) = 0.1 g, initial arsenate concentration 44.703 mg/L, adsorption temperature = 25 °C, solution volume = 50 mL, pH range 2.0–10.
in which, qt and qe are the adsorption capacity (mg/g) of adsorbent at anytime t (min) and under equilibrium, respectively. K1 (min1) is the related rate constant of the pseudo-first-order model, which can be calculated from the slope by the use of the plot of log(qe–qt) vs. time (t) (Fig. S5). Compared with pseudo-first-order, the pseudo-second-order model can predict the behaviors over the whole range, which is defined by using the following equation:
t 1 1 ¼ þ t qt K 2 q2e qe
ð8Þ
where qt and qe are the adsorption capacity (mg/g) of adsorbent at anytime t (min) and under equilibrium, K2 (g/mg min) is the rate constant of pseudo-second-order adsorption. The qe and K2 are obtained from the slope and intercept of the linear plot of t/qt vs. time (t), respectively, as shown in Fig. 9. In addition, the adsorption data were further analyzed by intraparticle diffusion model, which is described by the following equation:
qt ¼ K 3 t1=2 þ C
ð9Þ
where qt is the adsorption capacity (mg/g) of adsorbent at anytime t (min), and K3 (mg/g min1/2) is the intraparticle diffusion rate constant and C (mg/g) is the constant. K3 and C are determined from the slope and intercept of the plot of qt vs. square root of time (t1/2), as shown in Fig. S6. The corresponding parameters of the above three kinetic models under various initial As(V) concentrations and under different pH values are summarized in Tables 3 and 4, respectively. In general, the validity of the kinetic model is determined by the magnitude of the regression coefficient (R2) together with the match between experimental adsorption capacity qe(exp) and calculated adsorption capacity qe(cal)[33]. As can be seen by comparing Tables 3 and 4, the regression coefficient (R2) of pseudo-second-order model is more than 0.99 under all experimental conditions, while the corresponding regression coefficients (R2) of pseudo-first order model and intraparticle diffusion models are less than 0.98. More important, it is noticeable that the calculated adsorption capacities qe(cal) obtained from the pseudo-second kinetic model, are in excellent agreement with experimental adsorption capacities qe(exp). Therefore, it can be concluded that the As(V) adsorption over MA(400) abided by a pseudo-second-order model, thus indicating that the ‘‘surface reaction’’ was a predominant factor and controlling stage for As(V) removal[13].
Fig. 9. Plot of pseudo-second-order kinetic model for the adsorption of As(V) on mesoporous alumina at (A) different initial concentration and (B) different pH value.
3.5. Thermodynamic study The thermodynamic parameters for the adsorption process, including Gibbs free energy DGo (kJ/mol), the changes in enthalpy DHo (kJ/mol) and entropy DSo (kJ/mol) were estimated by using the following equations:
DGo ¼ RT ln K a
ð10Þ
DGo ¼ DHo T DSo
ð11Þ
ln K a ¼
DSo DH o 1 T R R
ð12Þ
where T is adsorption temperature in Kelvin (K), R is the ideal gas constant (8.314 J/mol K). DG0, DH0 and DS0 is the change in free energy (kJ/mol), the change in enthalpy (kJ/mol) and the change in entropy (kJ/mol K), respectively. Ka is the solid–liquid distribution coefficient and is defined as the following equation:
Ka ¼
C ads Ce
ð13Þ
where Cads and Ce is the concentration of As(V) within the adsorbent (mg/L) and in the solution under equilibrium, respectively. The thermodynamic parameters DHo and DSo can be determined from the slope and intercept respectively of the linear plot of ln Ka vs. 1/T.
7
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9 Table 3 Pseudo-first order, pseudo-second order and intraparticle diffusion kinetic parameters for As(V) adsorption on mesoporous alumina at different concentration. C0 (mg/L)
qe(exp)
11.175 22.352 31.806 44.703 69.005
Pseudo-first order
10.891 21.234 24.740 28.539 31.910
Pseudo-second order
K1
qe
0.0055 0.0021 0.0074 0.0055 0.0023
0.689 1.132 1.578 5.674 5.658
2
(cal)
Intraparticle diffusion 2
R
K2
qe(cal)
R
0.9318 0.2929 0.4077 0.7648 0.9230
0.0219 0.0107 0.0131 0.0034 0.0018
10.942 21.233 24.885 28.492 32.057
0.9999 0.9997 0.9999 0.9993 0.9976
K3
C
R2
0.0315 0.0611 0.1403 0.2351 0.2491
10.1289 19.6115 21.7083 22.7259 24.9386
0.8907 0.6937 0.2922 0.7766 0.9540
Table 4 Pseudo-first order, pseudo-second order and intraparticle diffusion kinetic parameters for As(V) adsorption on mesoporous alumina at different pH value. pH
3.0 4.5 6.6 8.5
qe(exp)
Pseudo-first order
32.971 35.967 28.539 18.315
Pseudo-second order
Intraparticle diffusion
K1
qe(cal)
R2
K2
qe(cal)
R2
K3
C
R2
0.0037 0.0099 0.0055 0.0025
4.782 12.943 5.677 5.585
0.9800 0.8757 0.7648 0.9396
0.0026 0.0018 0.0034 0.0016
33.227 36.766 28.495 18.527
0.9995 0.9994 0.9993 0.9926
0.2110 0.3371 0.2351 0.2411
27.5143 27.8833 22.7259 11.5754
0.9776 0.8736 0.7766 0.9644
Table 5 Thermodynamic parameter. T (K)
Ka
4Go (kJ/mol)
4Ho (kJ/mol)
4So (kJ/mol K)
298 313 333
30.97 39.51 65.56
9.01 10.36 12.16
17.81
0.09
ion-exchange; (iii) when E is in the range of 20–40 kJ/mol, chemisorption is predominant in the adsorption procedure. Adsorption energy calculated from the isotherm and Eq. (16) is 2.61 kJ/mol, thus suggesting the mechanism which dominated the adsorption procedure is physisorption at pH = 6.6 ± 0.1. 3.7. Effect of coexisting anions
The values of the thermodynamic parameters are tabulated in Table 5. The negative DGo at all temperature indicates that As(V) adsorption over MA(400) is spontaneous and feasible. Moreover, it noted that the absolute value of DGo increased with adsorption temperature, which has been ascribed to the fact that high temperature is in favor of the increase of adsorption impetus, thus promoting As(V) adsorption. The positive DHo confirms that the adsorption process is endothermic, and the positive DSo is an indication of randomness increase at the solid/liquid interface during As(V) adsorption. 3.6. Adsorption energy In order to further investigate the procedure of As(V) adsorption over MA(400), a Dubinin–Radushkevich (D–R) isotherm model was applied for analyzing the adsorption energy. The linearized D–R isotherm model is defined as the following equation:
ln qe ¼ ln qm K DR e2
ð14Þ
where KDR is the constant related to the mean free energy of adsorption; qe and qm is the equilibrium and theoretical saturation capacity, respectively; e is the Polanyi potential expressed by using the following equation:
e2 ¼ RT lnð1 þ ð1=C e ÞÞ
ð15Þ
where parameters R, T and Ce are the ideal gas constant, adsorption temperature in Kelvin (K) and equilibrium concentration, respectively. KDR and qm were obtained from the slope and intercept of the plot for ln qe against e2 (Fig. S7). Then, the adsorption energy (E) can be calculated by using the following equation:
E ¼ ð2K DR Þ1=2
Anions including silicate, phosphate, fluoride, nitrate and sulfate are always present in natural and/or polluted water with different levels, which will be closely associated with the removal of arsenic(V) [38]. The effects of those anions (such as SiO4 , PO3 4 , 2 F-, NO and SO ) on arsenic(V) removal over MA are presented (400) 3 4 in Fig. 10. The experimental condition for the initial arsenic(V) concentration was kept at 44.703 mg/L and the initial concentration of coexisting anions was varying from 0 mg/L to 200 mg/L. The arsenic(V) removal was slightly affected by the presence of NO 3 3 and SO2 and F- caused a 4 . However, the presence of SiO4 , PO4 sharp reduction in the effectiveness of the adsorbent in removing arsenic(V), especially when the ratio of coexisting anion concentration to arsenic(V) is larger than 1.12. This might due to the fact that the binding affinity of silicate, phosphate and fluoride with MA(400) was much stronger than that of arsenate, which leading to competition between arsenate and these interfering anions for the same sites of MA(400) [14,39,40].
ð16Þ
It was well demonstrated that adsorption energy is closely associated with the adsorption process [34–37]: (i) when E is less than 8 kJ/mol, physical adsorption is the major process; (ii) when E is in the range of 8–16 kJ/mol, the adsorption behavior is dominated by
3.8. Adsorption mechanism The arsenic(V) species were defined by using a well-known computer program (Mineql+, 4.6), and the corresponding patterns are displayed in Fig. S8. According to the diagram of arsenic(V) spe2 cies, high percentages of H2 AsO 4 and HAsO4 species are present in the pH range of 2.5–7.0. Therefore, it can be concluded that these species are adsorbed on the surface of MA(400) during the adsorption process. On the basis of these analyses and the variation of the solution pH value in the whole adsorption process (Fig. S9), the corresponding arsenic(V) adsorption mechanism for MA(400) at pH = 6.6 is described by the following equations:
AlðOHÞ3 þ nHþ ! fAlðOHÞ3 nHgnþ
ð17Þ
fAlðOHÞ3 Hgþ þ H2 AsO4 ! AlðOHÞ2 H2 AsO4 þ H2 O ðn ¼ 1Þ ð18Þ
8
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9
4. Conclusion
Fig. 10. Influence of coexist anion on the removal of arsenic, anion concentration: (a) 0 mg/L, (b) 10 mg/L, (c) 50 mg/L and (d) 200 mg/L. Experimental condition: adsorbent = 0.1 g, and initial arsenic concentration = 44.703 mg/L, adsorption temperature = 25 °C, pH = 6.6 ± 0.1, solution volume = 50 mL.
fAlðOHÞ3 2Hg2þ þ HAsO2 ðn ¼ 2Þ 4 ! AlðOHÞ HAsO4 þ 2H2 O ð19Þ +
This mechanism can be briefly explained as follows: H of arsenic(V) solution was first adsorbed on the surface of MA(400) via surface hydroxyl groups to form positive aluminum hydroxide species. After that, negative arsenate species were adsorbed on the surface of MA(400) via electrostatic interactions between positive aluminum hydroxide species and negative arsenate species, thus removing arsenic from water. The pH influence on MA(400) performance can be conceivably expounded as follows: when pH is too low (pH < 2.0), though the positive aluminum hydroxide species was formed, arsenic(V) was predominantly presented as nonionic H3AsO4, thus indicating the absence of sufficiently strong interaction between H3AsO4 species and positive aluminum hydroxide species. On the other hand, when pH is too high (pH > 8.0), arse3 nate species existed predominantly as HAsO2 4 and AsO4 anions, while the number of H+, arising from the arsenic(V) solution, is inadequate to yield the positive aluminum hydroxide species, which also leads to the lack of sufficiently strong interaction between arsenate species and aluminum hydroxide species. The significant differences between MA(400) and CAA(400) in the optimum pH ranges and the maximum adsorption capacity might be attributed to the differences of BET surface areas and amounts of surface hydroxyl groups, which will be of great benefit to form positive aluminum hydroxide species. 3.9. Desorption study To evaluate the possibility of MA(400) reuse, desorption and regeneration processes of MA(400) for arsenic removal were investigated. The percentages of arsenic removal for MA(400) regenerated by using deionized water, 0.001, 0.01, 0.05 and 0.1 M NaOH were founded to be 32%, 45%, 78%, 82%, 84%, respectively. Though the percent of MA(400) regenerated with 0.1 M NaOH is slightly higher than that of the sample regenerated with 0.05 M, the faster dissolution of aluminum for the former caused severe degradation in the pore structure for MA(400). Therefore, it can be concluded that the solution of 0.05 M NaOH was the most suitable desorption agent for MA(400) desorption and regeneration. It also noted that the amount of arsenic adsorption was decreased slightly with the cycle number. However, more than 80% of initial uptake capacity was maintained after five adsorption/desorption/regeneration cycles.
In summary, mesoporous alumina has been synthesized under low-temperature by using P123 and AIP as a template and an aluminum source, respectively. The As(V) removal and the corresponding adsorption kinetics of MA(400) is far higher and faster than CCA(400), respectively, which can be explained as follows: the mesoporous structure of MA(400) is in favor for the diffusion and transportation of arsenate species; on the other hand, high surface area and more hydroxyl groups will be of great benefit to form positive aluminum hydroxide species. MA(400) shows high performance for removing 2 H2 AsO in the pH region of 2.5–7.0, and the As(V) 4 and HAsO4 adsorption data are well fitted by the Langmuir isotherm and the maximum adsorption capacity at near neutral pH (pH = 6.6 ± 0.1) is 36.6 mg/g. The As(V) removal of MA(400), MA(600) and MA(800) is in the order of MA(400) >> MA(600) > MA(800). The difference between MA(400) and MA(600) is likely because high BET surface area and more surface hydroxyl groups of MA(400) should be in favor of As(V) adsorption. As(V) removal of MA(800) is lower than MA(600), which might be ascribed to synergistic effects of small surface area, fewer surface hydroxyl groups and crystal structure change. Kinetics of As(V) adsorption was evaluated as a function of initial concentration and pH, and the data exhibited that surface reaction was a predominant factor and controlling stage for the sorption process. From the results of thermodynamic study, the negative DGo values and the positive value of DHo indicated that adsorption process was spontaneous and endothermic. In addition, adsorption energy (2.61 kJ/ mol) is less than 8 kJ/mol, which indicates that the adsorption process may be dominated by physisorption. Acknowledgements This work was supported by the Natural Science Foundation of China (Grant No. 20867003 and 51068010), China Postdoctoral Science Foundation funded project (Grant No. 20100471686), and Young Academic and Technical Leader Raising Foundation of Yunnan Province (Grant No. 2008py010).
Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2012.11.087. References [1] A. Gomez-Caminero, P. Howe, M. Hughes, E. Kenyon, D.R. Lewis, M. Moore, J. Ng, A. Aito, G. Becking, Arsenic and Arsenic Compounds, Environmental Health Criteria 224, World Health Organization, Geneva, Switzerland, 2001. [2] D. Stuben, Z. Berner, D. Chandrasekharam, J. Karmakar, Arsenic enrichment in groundwater of West Bengal, India: geochemical evidence for mobilization of As under reducing conditions, Appl. Geochem. 18 (2003) 1417–1434. [3] M. Berg, S. Luzi, P.T.K. Trang, P.H. Viet, W. Giger, D. Stben, Arsenic removal from groundwater by household sand filters: comparative field study, model calculations, and health benefits, Environ. Sci. Technol. 40 (2006) 5567–5573. [4] A.A. Duker, E.J.M. Carranza, M. Hale, Arsenic Geochem. Health, Environ. Int. 31 (2005) 631–641. [5] S. Chen, S.R. Dzeng, M. Yang, K. Chiu, G. Shieh, C.M. Wai, Arsenic species in groundwaters of the blackfoot disease area, Taiwan, Environ. Sci. Technol. 28 (1994) 877–881. [6] T. Balaji, T. Yokoyama, H. Matsunaga, Adsorption and removal of As(V) and As(III) using Zr-loaded lysine diacetic acid chelating resin, Chemosphere 59 (2005) 1169–1174. [7] M. Jang, W.F. Chen, F.S. Cannon, Preloading hydrous ferric oxide into granular activated carbon for arsenic removal, Environ. Sci. Technol. 42 (2008) 3369– 3374. [8] F.H. Genc, J.C. Tjell, D. McConchie, Adsorption of arsenic from water using activated neutralized red mud, Environ. Sci. Technol. 38 (2004) 2428–2434. [9] Y. Zhang, M. Yang, X.M. Dou, H. He, D.S. Wang, Arsenate adsorption on an Fe– Ce bimetal oxide adsorbent: role of surface properties, Environ. Sci. Technol. 39 (2005) 7246–7253.
C. Han et al. / Chemical Engineering Journal 217 (2013) 1–9 [10] G.S. Zhang, J. H Qu, H.J. Liu, R.P. Liu, G.T. Li, Removal mechanism of As(III) by a novel Fe–Mn binary oxide adsorbent: oxidation and sorption, Environ. Sci. Technol. 41 (2007) 4613–4619. [11] M.A. Anderson, J.F. Ferguson, J. Gavis, Arsenate adsorption on amorphous aluminum hydroxide, J. Colloid. Interface Sci. 54 (1976) 391–399. [12] T.F. Lin, J.K. Wu, Adsorption of arsenite and arsenate within activated alumina grains: equilibrium and kinetics, Water Res. 35 (2001) 2049–2057. [13] S.S. Tripathy, A.M. Raichur, Enhanced adsorption capacity of activated alumina by impregnation with alum for removal of As(V) from water, Chem. Eng. J. 138 (2008) 179–186. [14] T.S. Singh, K.K. Pant, Kinetics and mass transfer studies on the adsorption of arsenic onto activated alumina and iron oxide impregnated activated alumina, Water Qual. Res. J. Can. 41 (2006) 147–156. [15] J.Y. Ying, C.P. Mehnert, M.S. Wong, Synthesis and applications of supramolecular-templated mesoporous materials, Angew. Chem. Int. Ed. 38 (1999) 56–77. [16] D.E. De Vos, M. Dams, B.F. Sels, P.A. Jacobs, Ordered mesoporous and microporous molecular sieves functionalized with transition metal complexes as catalysts for selective organic transformations, Chem. Rev. 102 (2002) 3615–3640. [17] Astam K. Patra, Arghya Dutta, Asim Bhaumik, Self-assembled mesoporous cAl2O3 spherical nanoparticles and their efficiency for the removal of arsenic from water, J. Hazard. Mater. 201–202 (2012) 170–177. [18] M.J. Yu, X. Li, W.S. Ahn, Adsorptive removal of arsenate and orthophosphate anions by mesoporous alumina, Micropor. Mesopor. Mater. 113 (2008) 197–203. [19] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R. A Pierotti, J. Rouquerol, T. Siemieniewska, Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity (Recommendations 1984), Pure Appl. Chem. 57 (1985) 603–619. [20] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, Academic Press, New York, 1982. [21] P.T. Tanev, T.J. Pinnavaia, A Neutral Templating Route to Mesoporous Molecular Sieves, Science 267 (1995) 865–867. [22] S. Cabrera, J.El. Haskouri, J. Alamo, A. Beltrán, D. Beltrán, S. Mendioroz, M.D. Marcos, P. Amorós, Surfactant-assisted synthesis of mesoporous alumina showing continuously adjustable pore sizes, Adv. Mater. 11 (1999) 379–381. [23] D.Y. Zhao, J.L. Feng, Q.S. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Triblock copolymer syntheses of mesoporous silica with periodic 50– 300 angstrom pores, Science 279 (1998) 548–552. [24] Y.M. Luo, Z.Y. Hou, R.T. Li, X.M. Zheng, Rapid synthesis of ordered mesoporous silica with the aid of heteropoly acids, Micropor. Mesopor. Mater. 109 (2008) 585–590. [25] H. Wang, C.Y. Han, H.P. Pu, L.Y. Zhang, Z.H. Zou, Y.M. Luo, Rapid synthesis of mesoporous SBA-16 silica in weakly acidic media, Bull. Chem. Soc. Jpn. 83 (2010) 852–854.
9
[26] P.D. Yang, D.Y. Zhao, D.I. Margolese, B.F. Chmelka, G.D. Stucky, Generalized syntheses of large-pore mesoporous metal oxides with semicrystalline frameworks, Nature 396 (1998) 152–155. [27] S.A. Bagshaw, E. Prouzet, T.J. Pinnavaia, Templating of mesoporous molecular sieves by nonionic polyethylene oxide surfactants, Science 269 (1995) 1242– 1244. [28] Y.M. Luo, J.M. Hao, Z.Y. Hou, L.X. Fu, R.T. Li, P. Ning, X.M. Zheng, Influence of preparation methods on selective catalytic reduction of nitric oxides by propene over silver–alumina catalyst, Catal Today 93–95 (2004) 797– 803. [29] M.H. Zahir, K. Sato, H. Mori, Y. Iwamoto, M. Nomura, S. Nakao, Preparation and properties of hydrothermally stable c-alumina-based composite mesoporous membranes, J. Am. Ceram. Soc. 89 (2006) 2874–2880. [30] Z.X. Sun, T.T. Zheng, Q.B. Bo, D. Vaughan, M. Warren, Effects of alkali metal ions on the formation of mesoporous alumina, J. Mater. Chem. 18 (2008) 5941– 5947. [31] Y. Zhang, Min Yang, Xia Huang, Arsenic(V) removal with a Ce(IV)-doped iron oxide adsorbent, Chemosphere 51 (2003) 945–952. [32] F. Babou, G. Coudurier, J.C. Vedrine, Acidic properties of sulfated zirconia: an infrared spectroscopic study, J. Catal. 152 (1995) 341–349. [33] S.K. Swain, Tanushree Patnaik, V.K. Singh, Usha Jha, R.K. Patel, R.K. Dey, Kinetics equilibrium and thermodynamic aspects of removal of fluoride from drinking water using meso-structured zirconium phosphate, Chem. Eng. J. 171 (2011) 218–1226. [34] J. Saikia, B. Saha, G. Das, Efficient removal of chromate and arsenate from individual and mixed system by malachite nanoparticles, J. Hazard. Mater. 186 (2011) 575–582. [35] S. Vasiliu, I. Bunia, S. Racovita, V. Neagu, Adsorption of cefotaxime sodium salt on polymer coated ion exchange resin microparticles: kinetics, equilibrium and thermodynamic studies, Carbohyd. Polym. 85 (2011) 376– 387. [36] A.M. El-Kamash, A.A. Zaki, M. Abed El Geleel, Modeling batch kinetics and thermodynamics of zinc and cadmium ions removal from waste solutions using synthetic zeolite A, J. Hazard. Mater. 127 (2005) 211–220. [37] S.S. Tahir, N. Rauf, Removal of cationic dye from aqueous solutions by adsorption onto bentonite clay, Chemosphere 63 (2006) 1842–1848. [38] X.G. Meng, Sunbaek Bang, George P. Korfiatis, Effects of silicate sulfate and carbonate on arsenic removal by ferric chloride, Water Res. 34 (2000) 1255– 1261. [39] W.J. Shao, X.M. Li, Q.L. Cao, F. Luo, J.M. Li, Y.Y. Du, Adsorption of arsenate and arsenite anions from aqueous medium by using metal(III)-loaded amberlite resins, Hydrometallurgy 91 (2008) 138–143. [40] G.S. Zhang, J.H. Qu, H.J. Liu, R.P. Liu, R.C. Wu, Preparation and evaluation of a novel Fe–Mn binary oxide adsorbent for effective arsenite removal, Water Res. 41 (2007) 1921–1928.