Langmuir 2008, 24, 10961-10967
10961
Enhanced Adsorption of Arsenate on the Aminated Fibers: Sorption Behavior and Uptake Mechanism Shubo Deng,*,†,‡ Gang Yu,†,‡ Sihuang Xie,† Qiang Yu,†,‡ Jun Huang,†,‡ Yasuyuki Kuwaki,§ and Masahiro Iseki§ Department of EnVironmental Science and Engineering, and POPs Research Center, Tsinghua UniVersity, Beijing 100084, People’s Republic of China, and ECO Technology Research Center, Sanyo Electric Company, Ltd. 1-1-1, Sakata Oizumi-Machi, Ora-Gun, Gunma 370-0596, Japan ReceiVed July 19, 2008 Novel aminated polyacrylonitrile fibers (APANFs) were prepared through the reaction of polyacrylonitrile fibers (PANFs) with four multinitrogen-containing aminating reagents, and the best adsorbent was obtained after the optimization of preparation experiments. The APANFs were effective for arsenate removal from aqueous solution, and the sorption behaviors including kinetics, isotherms, effect of pH, and competitive anions were investigated. Experimental results show that the equilibrium of arsenate sorption on the fibers was achieved within 1 h, and Langmuir equation described the sorption isotherms well with a high sorption capacity of 256.1 mg/g obtained. The thermodynamic parameters calculated show that the sorption was spontaneous and exothermic under the condition applied. The zero point of ζ potential of the APANFs was at about pH ) 8.2, in contrast with that of the PANFs at pH ) 3.6. Fourier transform infrared (FTIR) and X-ray photoelectron spectroscopy (XPS) for the APANFs before and after arsenate adsorption revealed that the amine groups on the fiber surface played an important role in the removal of arsenate from water, attributed to the electrostatic interaction between the positive protonated amine groups and negative arsenate ions.
Introduction Arsenic is a worldwide pollutant in ground and surface waters with serious health effects upon long-term intake of even low concentrations through portable water consumption.1 Arsenic contamination in drinking water has become a major concern in many regions around the world, which may have affected more than 100 million people worldwide.1,2 Recently, the acceptable value of arsenic in drinking water is limited to 10 µg/L according to the U.S. Environmental Protection Agency and World Health Organization, but many developing countries still comply with a limit of 50 µg/L.1,3 The more stringent arsenic drinking water standard requires the installation of new water treatment systems and the upgrading of existing ones. Many countries are becoming concerned about increasing levels of arsenic found in drinking water and put forth efforts to develop cost-effective technologies to remove arsenic from drinking water. Some water treatment technologies including coagulation, coprecipitation, ion exchange, adsorption, reverse osmosis, and electrodialysis are available for removing arsenic from water.4-6 Among them, adsorption is one of the commonly used techniques to remove arsenic from water, especially for individual homes and small community systems in rural areas in low income regions because the system is simple to operate and cost-effective. * To whom correspondence should be addressed. Phone: 86-10-62792165. Fax: 86-10-62794006. E-mail:
[email protected]. † Department of Environmental Science and Engineering, Tsinghua University. ‡ POPs Research Center, Tsinghua University. § Sanyo Electric Co. (1) Berg, M.; Luzi, S.; Trang, P. T. K.; Viet, P. H.; Giger, W.; Stuben, D. EnViron. Sci. Technol. 2006, 40, 5567–5573. (2) Deng, S. B.; Ting, Y. P. Water Sci. Technol. 2007, 55, 177–185. (3) Gu, Z. M.; Fang, J.; Deng, B. L. EnViron. Sci. Technol. 2005, 39, 3833– 3843. (4) Hering, J. G.; Chen, P. Y.; Wilkie, J. A.; Elimelech, M. J. EnViron. Eng. 1997, 123, 800–807. (5) Singh, T. S.; Pant, K. K. Sep. Purif. Technol. 2004, 36, 139–147. (6) Uddin, M. T.; Mozumder, M. S. I.; Islam, M. A.; Deowan, S. A.; Hoinkis, J. Chem. Eng. Technol. 2007, 30, 1248–1254.
Many adsorbents including activated alumina, iron oxide, mixed metal oxides, resin, and modified materials were reported to be effective for arsenic removal.7-10 Activated alumina has been the most often used adsorbent for arsenic removal, but its disadvantages include relatively low sorption capacity and the need for pH adjustment due to low working pH range.5,9 Recently, more effective Fe(III)-bearing materials such as goethite, ferrihydrite, granular ferric hydroxide (GFH), granular ferric oxide (GFO), iron-oxide-coated activated carbon, and Fe(III)-loaded cellulose have been developed for arsenic removal because of the Fe(III) affinity toward inorganic arsenic species and consequent selectivity of the adsorption process.3,11-14 More recently, the nanocrystalline TiO2 was found to be a promising material for arsenic removal.15,16 Experimental results demonstrated that the nanocrystalline TiO2 adsorbent was effective for the removal of both arsenate and arsenite in groundwater with higher sorption capacities than other adsorbents. Although these adsorbents are effective for arsenic removal, they are normally in the form of powders and have to be granulated or loaded into porous materials for fixed-bed adsorption in actual application.3,17 In contrast, fibrous adsorbents can be used directly and easily separated from aqueous solution after adsorption due (7) Dixit, S.; Hering, J. G. EnViron. Sci. Technol. 2003, 37, 4182–4189. (8) Loukidou, M. X.; Matis, K. A.; Zouboulis, A. I.; Liakopoulou-Kyriakidou, M. Water Res. 2003, 37, 4544–4552. (9) Lin, T. F.; Wu, J. K. Water Res. 2001, 35, 2049–2057. (10) Zhang, J. S.; Stanforth, R. Langmuir 2005, 21, 2895–2901. (11) Sherman, D. M.; Randall, S. R. Geochim. Cosmochim. Acta 2003, 67, 4223–4230. (12) Munoz, J. A.; Gonzalo, A.; Valiente, M. EnViron. Sci. Technol. 2002, 36, 3405–3411. (13) Zhang, G. S.; Qu, J. H.; Liu, H. J.; Liu, R. P.; Wu, R. C. Water Res. 2007, 411921-1928. (14) Driehaus, W.; Jekel, M.; Hildebrandt, U. Aqua 1998, 47, 30–35. (15) Dutta, P. K.; Ray, A. K.; Sharma, V. K.; Millero, F. J. J. Colloid Interface Sci. 2004, 278, 270–275. (16) Pena, M.; Meng, X. G.; Korfiatis, G. P.; Jing, C. Y. EnViron. Sci. Technol. 2006, 40, 1257–1262. (17) Theis, T. L.; Iyer, R.; Ellis, S. K. J. Am. Water Works Assoc. 1992, 84, 101–105.
10.1021/la8023138 CCC: $40.75 2008 American Chemical Society Published on Web 09/05/2008
10962 Langmuir, Vol. 24, No. 19, 2008
Deng et al.
to the long size.18 In a fixed-bed, they can be easily compressed or loosened according to the requirements for operation and washing. Some researchers have prepared some modified fibers and used them to remove humic acid, fluoride, and phosphate,19,20 but few studies were conducted to remove arsenate.19 The objective of this study is to prepare cost-effective fibrous adsorbents with high sorption capacity for arsenate. The multinitrogen-containing molecules were grafted onto the surface of polyacrylonitrile fibers through a one-step reaction, and the preparation conditions were optimized. The obtained aminated polyacrylonitrile fibers were effective for anionic arsenate removal, and the sorption behavior and mechanism were investigated in detail.
Experimental Section Materials. The polyacrylonitrile fibers (PANFs) were kindly provided by Daqing Polyacrylonitrile Factory, China. The PANFs contain 90% polyacrylonitrile and 10% vinyl acetate and have an average diameter of about 25 µm and a specific gravity of about 1.18 g/cm3. The aminating agents including ethylenediamine, diethylenetriamine, triethylenetetramine, and tetraethylenepentamine were purchased from Sigma-Aldrich Co. Others chemicals were of reagent grade. Synthesis of the APANFs. The APANFs were prepared in a 100 mL reaction vessel with 50 mL of ethylenediamine (diethylenetriamine, triethylenetetramine, or tetraethylenepentamine). A 2 g amount of the PANFs was added into the mixture with stirring, and the reaction in the vessel was allowed to proceed at 100-160 °C for a predetermined time for the amination reaction. After the reaction, the modified fibers were removed from the solution, thoroughly rinsed with deionized water till neutral pH, and then dried in an oven at 60 °C until constant weight. In the optimization experiments for the adsorbent preparation, the sorption experiments with 0.05 g of fibers in 100 mL of 100 mg/L arsenate solution at pH 6 and 25 °C for 12 h were conducted to evaluate the prepared adsorbents. The sorption amount was calculated and used to evaluate the APANFs obtained under different conditions. Batch Sorption Experiments. Batch sorption experiments were conducted to examine the adsorption isotherm, the kinetics, and the effect of solution pH and competitive anions on the adsorption behaviors. The sorption experiments were carried out in 250 mL flasks, each of which contained 100 mL of As(V) solution prepared with Na3AsO4. A 0.01 g amount of adsorbents was added to a flask and shaken at 125 rpm in a thermostatic shaker at 25 °C for 12 h. In the sorption isotherm experiments, the As(V) concentrations varied from 0.1 to 150 mg/L, and the solution pH was adjusted to 7 and controlled constant throughout the sorption experiment. The adsorption kinetic experiments were carried out at an initial As(V) concentration of 5 or 20 mg/L, and their equilibrium pH values were 6.1 or 7.5, respectively. In the experiments for the effect of solution pH, the As(V) concentration was 5 mg/L, and the solution pH values were controlled or uncontrolled in the range of 3.7-10. The effects of coexisting anions on the arsenate sorption were conducted in 5 mg/L arsenate solution. After the addition of competing anions into arsenate solution, the solution pH was adjusted to 7 and kept constant during the sorption process, and the equilibrium concentrations of competing anions were measured after the sorption. After the sorption, the fibers were separated from the solutions by filtration, briefly rinsed with deionized water to remove residual solution trapped among the fibers, and then prepared for other analyses. The initial and final As(V) concentrations in the solutions in each of the flasks were determined with inductively coupled plasma optical emission spectrometry (ICP-OES, IRIS Interpid II XSP). (18) Zhang, A. Y.; Asakura, T.; Uchiyama, G. React. Funct. Polym. 2003, 57, 67–76. (19) Liu, R. X.; Guo, J. L.; Tang, H. X. J. Colloid Interface Sci. 2002, 248, 268–274. (20) Deng, S. B.; Bai, R. B. EnViron. Sci. Technol. 2003, 37, 5799–5805.
Figure 1. Effects of different aminating agents and reaction time on the sorption of arsenate on the APANFs with 0.05 g of fibers in 100 mL of 100 mg/L arsenate solution at pH 6.
Figure 2. FTIR spectra of the (a) PANFs and (b) APANFs.
When the arsenic concentration was below 1 mg/L, the graphite furnace atomic absorption spectroscopy was adopted. The sorption amount of the APANFs for As(V) was calculated according to the concentration difference before and after the sorption. FTIR Spectroscopy. The samples of the PANFs, APANFs, and As-adsorbed APANFs were cut into about 1 mm pieces blended with KBr and then pressed into a disk for FTIR analysis. The spectra were recorded on a FTIR spectrophotometer in the wavenumber range of 400-4000 cm-1 under ambient conditions. ζ Potential Measurement. A 0.1 g amount of the PANFs, APANFs, or As-adsorbed APANFs was cut into small pieces and placed into 100 mL of deionized water. The mixture was stirred for 12 h, and then the pH of the solution was adjusted with 0.1 M NaOH or 0.1 M HCl solution to a desired value. After 1 h stabilization, the solution pH was recorded, and the supernatant with small fiber fragments was then decanted and used to conduct ζ potential measurements with a Zeta-Plus4 instrument (Brookhaven Corp.). All data were determined five times, and the average value was adopted. XPS Analysis. XPS analyses of the APANFs before and after the adsorption of arsenate were carried out on an AEM PHI 5300X spectrometer with an Al KR X-ray source (1486.71 eV of photons) to determine the C, N, O, and arsenate on the surface of the APANFs. The X-ray source was run at a reduced power of 150 W, and the pressure in the analysis chamber was maintained at less than 10-8 Torr during each measurement. All binding energies were referenced to the neutral C1s peak at 284.6 eV to compensate for the surface charging effects. The software package XPSpeak 4.1 was used to fit the XPS spectra peaks, and the full width at half-maximum was maintained at 1.2 for all components in a particular spectrum.
Results and Discussion APANFs Preparation. The effects of different aminating agents and reaction time on the sorption capacity of the aminated
Enhanced Adsorption of Arsenate on the Aminated Fibers
Langmuir, Vol. 24, No. 19, 2008 10963
Scheme 1. Conversion of the PANFs to the APANFs
fibers are shown in Figure 1. The fibers reacted with four aminating agents including ethylenediamine, diethylenetriamine, triethylenetetramine, and tetraethylenepentamine at 120 °C for different reaction time. It can be found that the sorption amount for arsenate increased with increasing reaction time for each aminating agent used to prepare the aminated fibers. As the fibers can dissolve in the four aminating agents after certain reaction time, the reaction had to be stopped before their resolving. It was found that the time for the fibers beginning to resolve increased with increasing molecular weight of the aminating agents, and they were 1.5 h for ethylenediamine, 2 h for diethylenetriamine, 15 h for triethylenetetramine, and 25 h for tetraethylenepentamine. Sorption experiments show that the sorption capacities for arsenate using the APANFs reacted with ethylenediamine, diethylenetriamine, triethylenetetramine, and tetraethylenepentamine are 68.3, 70.9, 90.8, and 124.5 mg/g, respectively. It is reasonable that the sorption capacities increased with the increase of amine numbers in one molecule of the four aminating agents, but the enhanced amount did not reach what we expected. As there are 2, 3, 4, and 5 amine groups in one molecule of ethylenediamine, diethylenetriamine, triethylenetetramine, and tetraethylenepentamine, respectively, their sorption capacities should be directly proportional to the number of amine groups if the same number of the aminating agents are grafted on the fiber surface. The possible reasons include that less molecules are grafted on the surface for the aminating agents with higher molecular weight, or several amine groups in one molecule of aminating agent react with several active sites on the fiber surface, making the grafted amine groups unfavorable for the sorption of arsenate. The effects of reaction temperature and aminating agent concentration on the sorption capacity of the APANFs for arsenate were also investigated. The sorption capacity increased significantly with increasing reaction temperature in the range of 100-140 °C and became gradual after 140 °C. The concentration of aminating agents influenced the resolving velocity of the fibers and the velocity of amination reaction. The higher is the concentration of aminating agent, the higher is the reaction velocity. To shorten the reaction time, the pure aminating agents were used in the preparation. Theoretically, higher sorption capacity of the fibers can be achieved if the aminating agent with more amine groups in one molecule is adopted to modify the fibers. However, some polyamine (e.g., polyethylenimine) chemicals are very expensive. As the four aminating agents and the PANFs used in this study are widely used in industry and very cheap in price, the APANFs are cost-effective and have a promising application in arsenate removal from water. Some researchers used ethylenediamine and diethylenetriamine to modify the PANFs,20,21 but triethylenetetramine and tetraethylenepentamine were never used. The fibers modified with tetraethylenepentamine not only have the highest sorption capacity of 124.5 mg/g for arsenate, but also possess the strongest mechanical strength. Therefore, they were used in the following experiments. Surface Reaction. The FTIR spectra of the PANFs and APANFs modified with tetraethylenepentamine are shown in (21) Ko, Y. G.; Shin, D. H.; Choi, U. S. Macromol. Rapid Commun. 2004, 25, 1324–1329.
Figure 2. The characteristic peaks in the spectrum of the PANFs can be assigned as follows: 3438 cm-1 (OH stretching), 2941 cm-1 (CH stretching in CH, CH2, and CH3 groups), 2244 cm-1 (CtN stretching), 1735 cm-1 (CdO stretching), 1454 cm-1 (CH blending), 1361 cm-1 (symmetric blending of CH3 in CCH3), 1073 cm-1 (C-O stretching in acetate ester), and 538 cm-1 (CdO twisting).20 After the reaction with tetraethylenepentamine, the spectrum of the APANFs shows some significant changes. The peak at 2244 cm-1 disappeared completely after the amination reaction, which suggests that all nitrile groups on the surface of the PANFs were converted during the reaction. The new bands at peaks of 1659, 1566, and 1111 cm-1 for the APANFs can be assigned to the CdO group in amide, the N-H group in amine, and the C-N group in amide on the APANFs, respectively.22 These results suggest that the amide and amine groups were introduced on the surface of the APANFs. In addition, the broadband with the peak at 3435 cm-1 for the APANFs appeared. This strong broadband ranging from 3100 to 3700 cm-1 usually corresponds to the combination of the stretching vibration bands of both OH and NH groups, suggesting that OH and NH groups were introduced or formed on the surface of the APANFs. The OH group may be produced on the surface of the APANFs by the hydrolysis of the ester group on the PANFs, which can be supported by the disappearance of the bands at 2941, 1735, 1073, and 538 cm-1. According to the FTIR spectra, the location where the chemical reaction took place during the preparation of the APANFs may be proposed as in Scheme 1. Sorption Kinetics. Figure 3 shows the sorption of arsenate on the APANFs at different solution pH and arsenate concentration. It can be found that the sorption process was very fast and the sorption equilibrium was achieved within 60 min. To evaluate the sorption rates of arsenate on the APANFs, the pseudo-secondorder equation was adopted to model the experimental data. The pseudo-second-order expression was used to describe chemisorption and has been widely applied to the adsorption of pollutants from aqueous solutions in recent years.23 As shown in Figure 3, the pseudo-second-order kinetic model fitted the data well, and the corresponding parameters are also calculated and shown in Figure 3. When the initial arsenate concentration was 20 mg/L, the initial sorption rate was up to 36.1 mg/(g · min), and the sorption equilibrium was almost achieved after 30 min sorption. The higher arsenate concentrations facilitated the fast transport of arsenate to the fiber surface, and more amine groups were protonated at lower solution pH. As the APANFs are nonporous materials, the intrasorbent diffusion is negligible, and thus the arsenate sorption is only dominated by external mass transport. The fast sorption of arsenate on the APANFs was due to the property of weakly basic anion exchangers. Sorption Isotherm. Adsorption isotherm has commonly been used to evaluate the adsorption capacity of an adsorbent for an adsorbate, and Langmuir equation has been successfully applied to model many sorption processes.24,25 The Langmuir equation (22) Shriner, R. L.; Hermann, C. K. F.; Morrill, T. C.; Curtin, D. Y.; Fuson, R. C. The Systematic Identification of Organic Compounds, 7th ed.; John Wiley & Sons: New York, 1998. (23) Ho, Y. S. J. Hazard. Mater. B 2006, 136, 681–689. (24) Deng, S. B.; Bai, R. B. J. Colloid Interface Sci. 2004, 280, 36–43. (25) Deng, S. B.; Ting, Y. P. Langmuir 2005, 21, 5940–5948.
10964 Langmuir, Vol. 24, No. 19, 2008
Deng et al. Table 1. Calculated Equilibrium Constants and Thermodynamic Parameters for the Sorption of Arsenate on the APANFs Langmuir constantsa T (°C)
b (L/mg)
qm (mg/g)
R2
∆G° (kJ/mol)
∆H° (kJ/mol)
∆S° (kJ/(mol · K))
25 40
0.0601 0.0391
256.1 249.7
0.993 0.994
-20.8 -20.7
-22.8
-0.0067
a
Figure 3. Sorption kinetics of arsenate on the APANFs at equilibrium pH ) 7.5, C0 ) 5 mg/L (a), pH ) 6.1, C0 ) 5 mg/L (b), and pH ) 7.5, C0 ) 20 mg/L (c). Symbols: experimental data. Curve: modeled results using the pseudosecond-order equation (t/qt ) 1/υ0 + t/qe, υ0 represents the initial sorption rate, and qe is the adsorption capacity at equilibrium).
Figure 4. Sorption isotherm of arsenate on the APANFs at 25 and 40 °C with 0.01 g of the APANFs in 100 mL arsenate solution at equilibrium pH 7 at high (a) and low (b) equilibrium arsenate concentration.
assumes monolayer coverage of adsorbate over a homogeneous adsorbent surface, and the sorption of each molecule onto the surface has equal sorption activation energy. Figure 4a shows the adsorption isotherms of arsenate conducted at 25 and 40 °C, and the Langmuir equation was used to fit the experimental data; the corresponding parameters for the plot are given in Table 1. It can be found that Langmuir model described the experimental data well according to their correlation coefficients (R2), indicating the possible monolayer sorption of arsenate onto the fiber surface. The Langmuir constant (b) decreased from 0.0601 to 0.0391 L/mg when the sorption temperature increased from 25 to 40 °C, which suggests strong affinity of the APANFs for arsenate at lower temperature.
thermodynamic parametersb
qe ) qmCe/(1/b + Ce).2
b
∆G° ) -RT ln b, ∆G° ) ∆H° - T∆S°.25
The maximum sorption capacity (qm) of the APANFs for arsenate at 25 °C reached 256.1 mg/g according to Langmuir modeling, which is higher than that of most adsorbents used to remove arsenate in the literature.14,26 Normally, the sorption capacity of average adsorbents for arsenate is less than 100 mg/g except for a few specific adsorbents.13,27 However, the maximum sorption capacities of these adsorbents were achieved at high equilibrium arsenate concentration. For drinking water treatment, the sorption amount of the adsorbents for arsenate at low equilibrium concentration is critical as the guideline for arsenic in drinking water set by the World Health Organization is 10 µg/L. Therefore, the sorption amount of the APANFs for arsenate at low concentration was investigated and shown in Figure 4b. It can be found that the sorption amount was about 1.7 mg/g at the equilibrium arsenate concentration of 10 µg/L. As most researchers investigated the sorption isotherm of the adsorbents at equilibrium arsenate concentration at mg/L level, it is difficult to find their sorption amount at 10 µg/L of equilibrium arsenate concentration.13,27,28 However, the APANFs had a higher sorption amount at 10 µg/L arsenate concentration than did some conventional adsorbents. Jeong et al. reported that the sorption capacities of iron oxide and aluminum oxide for arsenate were about 0.6 and 0.1 mg/g at an arsenate equilibrium concentration of 100 µg/L.29 Of course, the sorption capacity of the APANFs for arsenate at low concentration will be further verified in column experiments in our future study. The thermodynamic parameters calculated are shown in Table 1. The negative values of ∆G° indicate that the sorption of arsenate on the APANFs was spontaneous under the experimental conditions. ∆S° and ∆H° are calculated to be -0.0067 kJ/(mol · K) and -22.8 kJ/mol, respectively. The negative value of the enthalpy change (∆H°) indicates that the arsenate sorption process on the APANFs is an exothermic reaction, while the negative value of the entropy change (∆S°) suggests a decreased randomness at the fiber/solution interface. Effect of Solution pH. Figure 5 shows the effect of solution pH on arsenate sorption onto the APANFs with or without pH control during the sorption process. The sorption amount decreased with increasing solution pH, and almost no arsenate was adsorbed at pH 10. In the case of no pH control, it is observed that the final solution pH was higher than the initial pH values at acidic solution, which verified the protonation of the amine groups on the fiber surface occurred in the sorption process. In case of pH control experiments, acid was added at initial pH below 8 to keep the constant solution pH. As more amine groups were protonated, the sorption amount of the APANFs for arsenate with pH control in the sorption process was higher than that of the sorption without pH adjustment at pH below 8. The sorption (26) Deschamps, E.; Ciminelli, V. S.; Holl, W. H. Water Res. 2005, 39, 5212– 5220. (27) Kim, Y. H.; Kim, C. M.; Choi, I. H.; Rengaraj, S.; Yi, J. H. EnViron. Sci. Technol. 2004, 38, 924–931. (28) Bang, S.; Patel, M.; Lippicott, L.; Meng, X. G. Chemosphere 2005, 60, 389–397. (29) Jeong, Y.; Fan, M.; Singh, S.; Chuang, C. L.; Saha, B.; van Leeuwen, H. Chem. Eng. Process. 2007, 46, 1030–1039.
Enhanced Adsorption of Arsenate on the Aminated Fibers
Figure 5. Effect of pH on the sorption of arsenate on the APANFs with or without pH control during the sorption process with 0.01 g of fibers in 100 mL of 5 mg/L As(V) solution.
Figure 6. Effects of anions on the sorption of arsenate on the APANFs at equilibrium pH 7 with 0.01 g of fibers in 100 mL of 5 mg/L As(V) solution.
amount for arsenate reached 46.8 mg/g (93.6% removal percent) when the solution was controlled at pH 6, much higher than 35.2 mg/g obtained without pH control. Solution pH not only affects the charge property of the APANFs surface through the protonation of amine groups, but also influences arsenate speciation in solution. Arsenate exists as H3AsO4, H2AsO4-, HAsO42-, and AsO43- species in aqueous solution. H2AsO4- is the main species in solution at pH from 3 to 6, while HAsO42- and AsO43- become major species at pH above 8. Therefore, arsenate species are mainly anions in the pH range studied, and electrostatic attraction should play an important role in the sorption process. With the increase of solution pH, the number of protonated amine groups decreased, while the number of negative charge of arsenic species increased. As a consequence, the sorption amount of the APANFs for arsenate decreased. Competitive Sorption. Some anions might exist in groundwater and compete with anionic arsenate for the available adsorption sites in the sorption process. To evaluate the efficiency of the APANFs for arsenate removal at the presence of anions, the arsenate solutions were spiked with phosphate, sulfate, bicarbonate, nitrate, chloride, and fluoride at different concentrations. The effects of the final equilibrium concentration of different competitive anions on arsenate uptake at pH 7 are illustrated in Figure 6. It can be found that all anions interfered with the sorption of arsenate on the APANFs, especially at high concentrations of the competitive anions. Sulfate caused the greatest decrease in arsenate sorption among the anions, while fluoride had little effect on the sorption of arsenate. In general, the effects of divalent anions on arsenate sorption are more obvious than those of the monovalent anions. Their effects on arsenate sorption show the
Langmuir, Vol. 24, No. 19, 2008 10965
Figure 7. Zeta potentials of the PANFs, APANFs, and As-adsorbed APANFs as a function of solution pH.
following decreasing order: SO42- > HPO42- > HCO3- > NO3> Cl- > F-. Actually, the competing sorption is complicated, and the binary sorbate sorption equilibrium should be represented clearly by a three-dimensional sorption isotherm, showing the sorption amount with the final equilibrium concentrations of both sorbates.30,31 The resulting evaluation usually requires the modeling approach as seen in the literature.30,32 Many studies showed that phosphate ions significantly interfered with arsenate sorption using the iron adsorbents as the molecular structure and chemical characteristics of phosphate ions were very similar to those of arsenic ion and could form an inner sphere complex with iron hydroxide.12,33 In this study, sulfate inhibited the arsenate sorption more significantly than did phosphate. Similarly, Awual et al. also reported that sulfate significantly interfered with uptake of monovalent H2AsO4- by a weak-base anion exchange fiber, and the uptake of arsenate decreased greatly when the popular quaternary ammonium-type anion exchange resin was used.34 In fact, the APANFs are a kind of anion exchangers, and some anions may exchange with the hydroxyl or chloride ions around the aminated groups on the APANFs surface. Zeta Potential Measurement. To investigate the effect of surface charge on the fiber surface on the sorption, the ζ potentials of the PANFs, APANFs, and As-adsorbed APANFs were measured and shown in Figure 7. The ζ potentials of the APANFs are positive at pH < 8.2, while the ζ potentials of the PANFs are positive at pH < 3.6, indicating that the zero point of ζ potential for the APANFs was enhanced after the surface amination reaction. The much higher zero point of ζ potentials may be attributed to the protonated amine groups on the APANFs surface because the pKa of the secondary amine groups in tetraethylenepentamine molecule is above 10. Therefore, the positive APANFs can be expected to provide better sorption amount for negative arsenate ions via anionic exchange. It also can be found that the zero point of ζ potentials of the APANFs after arsenic sorption was decreased to 6.5, suggesting that the negative arsenate adsorbed on the protonated amine groups and neutralized some positive charges on the adsorbent surface. FTIR Analysis. Figure 8 shows the FTIR spectra of the APANFs before and after arsenate sorption at pH 6.1 and 7.5, (30) Volesky, B. Sorption and Biosorption; BV Sorbex: Montreal-St. Lambert, Canada, 2003; pp 121-164. (31) Figueira, M. M.; Volesky, B.; Ciminelli, V. S. T. Biotechnol. Bioeng. 1997, 54, 344–350. (32) Chong, K. H.; Volesky, B. Biotechnol. Bioeng. 1995, 47, 451–460. (33) Meng, X. G.; Korfiatis, G. P.; Bang, S.; Bang, K. W. Toxicol. Lett. 2002, 133, 103–111. (34) Awual, M. R.; Urata, S.; Jyo, A.; Tamada, M. Water Res. 2008, 42, 689–696.
10966 Langmuir, Vol. 24, No. 19, 2008
Deng et al.
Figure 8. FTIR spectra of the APANFs (a), As-adsorbed APANFs at pH 6.1 (b), and As-adsorbed APANFs at pH 7.5 (c).
respectively. It can be found that the spectra of the APANFs showed some changes after the sorption of arsenate. The broad peak at 3435 cm-1 attributed to the overlapping of the stretching vibration of amine and hydroxyl groups shifted to 3442 and 3433 cm-1 after arsenate sorption at pH 6.1 and 7.5, respectively, which suggests the possible involvement of hydroxyl or amine groups in the sorption process. The band at peak of 1566 cm-1 for the APANFs assigned to the N-H group in amine shifted to 1549 and 1550 cm-1 after arsenate sorption at pH 6.1 and 7.5, respectively, and the peak intensity also reduced obviously. It also can be found that the new peaks at 802 and 798 cm-1 appeared in Figure 8b and c, which correspond to the stretching vibration of As-O.16 All results indicate that amine groups were involved in the sorption of arsenate. XPS Analysis. X-ray photoelectron spectroscopy (XPS) has been used to analyze the interactions between metal species and membranes or between adsorbates and adsorbents in adsorption.20,35-39 To investigate the interactions between arsenate and the APANFs in the sorption process, XPS studies of the APANFs before and after arsenate adsorption at pH 6.1 and 7.5 were conducted. The wide scan spectra show that the percent of nitrogen atom was increased from 2.51% to 4.71% after the fiber modification, and some arsenic atoms were detected on the APANFs surface after arsenate sorption, indicating the successful amination reaction and the occurrence of arsenic adsorption on the APANFs surface. To further verify the amine groups responsible for arsenate adsorption, the XPS N1s core-level spectra of the PANFs, APANFs before and after arsenate sorption were analyzed. Figure 9a shows the spectrum of the PANFs, and the binding energy at 398.5 eV is attributed to the N in the CtN group. After the amination reaction, the binding energies at 398.5 (peak I), 399.4 (peak II), and 400.6 eV (peak III) shown in Figure 9b can be assigned to the N in the CtN, NH2 (NH), and NH3+ (-NH2+-) groups on the surface of the APANFs, respectively,40,41 which verified that the enhanced nitrogen content on the fiber surface was from amine groups in the aminating agent. When arsenate (35) Tombacz, E.; Dobos, A.; Szekeres, M.; Narres, H. D.; Klumpp, E.; Dekany, I. Colloid Polym. Sci. 2000, 278, 337–345. (36) Ariza, M. J.; Benavente, J.; Rodriguez-Castellon, E.; Palacio, L. J. Colloid Interface Sci. 2002, 247, 149–158. (37) Cairns, D. B.; Armes, S. P.; Chehimi, M. M.; Perruchot, C.; Delamar, M. Langmuir 1999, 15, 8059–8066. (38) Deng, S. B.; Ting, Y. P. EnViron. Sci. Technol. 2005, 39, 8490–8496. (39) Rojas, O. J.; Ernstsson, M.; Neuman, R. D.; Claesson, P. M. Langmuir 2002, 18, 1604–1612. (40) Pels, J. R.; Kapteijn, F.; Moulijn, J. A.; Zhu, Q.; Thomas, K. M. Carbon 1995, 33, 1641–1653. (41) Li, Z.; Beachner, R.; McManama, Z.; Hanlic, H. Microporous Mesoporous Mater. 2007, 105, 291–297.
Figure 9. XPS N1s core-level spectra of the PANFs, APANFs before and after As sorption at equilibrium pH 6.1 or pH 7.5.
was adsorbed on the APANFs at pH 6.1 and pH 7.5, an obvious higher binding energy peak at 400.6 eV appeared (see Figure 9c and d), attributed to the protonated nitrogen atom (N+). It can be seen that the proportion of the positively charged nitrogen atoms (peak III) to the uncharged nitrogen atoms in NH2 or NH (peak II) on the surface of the APANFs increased after the arsenate sorption. The area ratio of peaks III/II of the APANFs before the sorption is only 0.08, while it increased to 0.75 at pH 6.1 and 0.41 at pH 7.5 after arsenate sorption, which indicates that about 43% and 29% of the amine groups (NH2 and NH) were protonated, respectively. The amine groups were first protonated at acidic solution during the sorption of arsenate on the APANFs, and then adsorbed the negative arsenate. These results indicate that the protonated amine groups are responsible for arsenate sorption. The more protonated amine groups found at lower solution pH are consistent with the higher positive zeta potential in Figure 7 and higher sorption amount for arsenate in Figure 5. Additionally, no As(III) was found in the XPS As3d core-level spectra, indicating that arsenate was not reduced during the sorption. Previous studies have shown that the conventional adsorbents including activated alumina and iron oxides adsorb arsenate through the formation of inner-sphere surface complexes of bidentate binuclear,11 and the hydroxide groups on the surface played an important role. Some researchers found that anion exchange was attributed to the sorption of arsenate onto the surfactant-modified zeolite and biomass.8,41 According to the above zeta potential, FTIR, and XPS analysis in this study, it can be concluded that the protonated amine groups on the fiber surface are responsible for the sorption of arsenate on the APANFs; the electrostatic attraction between protonated amine groups and negative arsenate ions plays an important role.
Conclusions Adsorption is one of the commonly used techniques to remove anionic arsenate from water in drinking water treatment, and the adsorbents used play a vital role in the process. In this study, we successfully prepared the novel aminated fibers through the modification of the fiber surface using the multinitrogencontaining aminating reagents, and the adsorbent with highest sorption capacity was obtained when tetraethylenepentamine was used, due to the more amine groups in one molecule of tetraethylenepentamine than other aminating reagents. Batch
Enhanced Adsorption of Arsenate on the Aminated Fibers
sorption experimental results show that the sorption of arsenate on the PANFs was fast and pH-dependent. The APANFs had a maximum sorption capacity of 256.1 mg/g for arsenate at high equilibrium arsenate concentration according to Langmuir fitting, suggesting the favorable application in the treatment of industry wastewater containing high concentration arsenate. The sorption amount of the APANFs for arsenate reached about 1.7 mg/g when the equilibrium arsenate concentration was 10 µg/L, higher than that of the conventional iron and aluminum oxide adsorbents. Because of the protonation of amine groups on the fiber surface, the APANFs had a higher zero point of ζ potential at pH 8.2. The decrease of zeta potential after arsenate sorption indicated the involvement of electrostatic neutralization in the sorption.
Langmuir, Vol. 24, No. 19, 2008 10967
FTIR and XPS analysis revealed that the amine groups on the fiber surface were responsible for arsenate removal through the electrostatic interaction between the positive protonated amine groups and negative arsenate. The APANFs prepared in this study have a promising application in water and wastewater treatment for arsenate removal. Acknowledgment. We thank the Sanyo Electric Co., Ltd., for financial support, and this research was also supported by the Program for New Century Excellent Talents in University. The analytical work was supported by the Laboratory Fund of Tsinghua University. LA8023138
