(MnFe2O4) Nanoparticles - ACS Publications - American Chemical ...

Langmuir 2005, 21, 11173-11179

11173

Fast Removal and Recovery of Cr(VI) Using Surface-Modified Jacobsite (MnFe2O4) Nanoparticles Jing Hu,† Irene M.C. Lo,‡ and Guohua Chen*,§ Environmental Engineering Program, School of Engineering, Department of Civil Engineering, and Department of Chemical Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong Received April 22, 2005. In Final Form: September 1, 2005 In this work, the effectiveness of surface-modified jacobsite (MnFe2O4) nanoparticles was investigated for the removal and recovery of Cr(VI) from synthetic wastewater. Ten nanometer modified MnFe2O4 nanoparticles were produced to be a new adsorbent using a co-precipitation method followed by a surface redox reaction. The equilibrium time for Cr(VI) adsorption onto modified MnFe2O4 nanoparticles was as short as 5 min, and the adsorption data fit the Langmuir model well. The maximum uptake of 31.5 mg of Cr(VI)/g of modified MnFe2O4 was obtained at pH 2, which was comparable with other common adsorbents such as activated carbon and sawdust. The effects of ligands (EDTA, SO42-, NH4+) and ionic strength were studied in a pH range of 2-10. EDTA and SO42- inhibited the adsorption of Cr(VI) over the entire pH range studied, whereas NH4+ enhanced the uptake of Cr(VI) at pH greater than 6.5. The mechanisms leading to Cr(VI) adsorption by modified MnFe2O4 nanoparticles were determined by X-ray diffraction and X-ray photoelectron spectroscopy to be a combination of electrostatic interaction and ion exchange. Regeneration studies indicated the potential reuse of the modified MnFe2O4 nanoparticles without sacrificing adsorption capacity and the possible recycling of Cr(VI) without changing the valence.

Introduction The removal of heavy metals from both natural water supplies and industrial wastewater streams is becoming increasingly important as awareness of the environmental impact of such pollutants is fully realized. In particular, environmental contamination by chromium is a major problem in industrialized areas, especially those that have large electroplating and metal finishing industries. Other important industrial sources of chromium are wood treatment and tannery facilities as well as chromium mining and milling operations. Every year, in the United States these industries produce billions of gallons of wastewater that contains chromium.1 In general, chromium-containing waste has been disposed by discharge into surface impoundments or lagoons. Leakage from these lagoons into the groundwater supply has been relatively common. Waste solutions of metal ions are commonly found at levels that are in excess of acceptable disposal limits. It is well known that Cr(VI) is toxic to animals and plants, whereas Cr(III) is considered to be less toxic. Hexavalent chromium species are strong oxidants that act as carcinogens, mutagens, and teratogens in biological systems.2 Adverse effects of the hexavalent form on the skin may include ulcerations, dermatitis, and allergic skin reactions. Inhalation of hexavalent chromium compounds can result in ulceration and perforation of the mucous membranes of the nasal septum, irritation of the pharynx and larynx, asthmatic bronchitis, bronchospasms, and edema. Respiratory symptoms may include coughing and wheezing, shortness of breath, and nasal itch.3 Dissolved chromate is toxic to many organisms at low aqueous * Corresponding author. E-mail: [email protected]. Phone: (852)23587138. Fax: (852)23580054. † Environmental Engineering Program, School of Engineering. ‡ Department of Civil Engineering. § Department of Chemical Engineering. (1) Brown, P. A.; Gill, S. A.; Allen, S. J. Water Res. 2002, 34, 3907. (2) Lee, M. G.; Cheon, J. K.; Kam, S. K. J. Ind. Eng. Chem. 2003, 9, 174.

concentrations and is a regulated constituent in domestic water supplies. Bioaccumulation of chromium in flora and fauna creates ecological problems. In light of chromium’s toxicity and environmental hazards, the release of chromium-containing substances into the environment should be controlled to minimize the effects on aquatic life and downstream users. The treatment of aqueous wastes containing soluble Cr(VI) requires the removal of heavy metals followed by recovery or safe disposal. Various techniques have been employed for treating Cr(VI)-contaminated water, such as chemical redox followed by precipitation, ion exchange, and reverse osmosis. But a major drawback related to precipitation is sludge production. Ion exchange and reverse osmosis are not economically attractive because of their high operating costs. Adsorption has emerged as a cost-effective technique for removing metals from wastewater and has been widely studied during recent decades. Activated carbon has been tested and used as an adsorbent for heavy metals, in addition to successful applications of activated carbon in removing dissolved hydrophobic organic compounds.4,5 Sharma and Forster6 discovered that Spagnum moss peat is an effective adsorbent for hexavalent chromium, although chromium recovery was only 50%. Begum7 found that sawdust is effective in the removal of Cr(VI) and other heavy metals. Recently, Ozdemir and co-workers8 reported on the selective adsorption of Cr(VI) by activated sludge from wastewater that contained chromium, cadmium, and copper ions. Kobya9 showed the effectiveness of hazelnut-shell activated carbon in the removal of Cr(VI). Despite these promising advances in the development of (3) Nriagu, J. O.; Nieboer, E. Chromium in the Natural and Human Environments; Wiley: New York, 1988. (4) Nagasaki, Y. Jpn. Kokai 1974, 74, 477. (5) Kim, J. I. Diss. Abstr. Int., B 1977, 37, 3566. (6) Sharma, D. C.; Forster, C. F. Water Res. 1993, 22, 1201. (7) Begum, S. Sci. Int. 1992, 4, 67. (8) Ozdemir, G.; Ozturk, T.; Ceyhan, N.; Isler, R.; Cosar, T. Bioresour. Technol. 2003, 90, 71. (9) Kobya, M. Bioresour. Technol. 2004, 91, 317.

10.1021/la051076h CCC: $30.25 © 2005 American Chemical Society Published on Web 10/13/2005

11174

Langmuir, Vol. 21, No. 24, 2005

adsorbents, the recovery of metals from these adsorbents and the regeneration of spent adsorbents have not been covered in the literature except for our recent studies.10,11 For the last several years, we have been developing a series of nanoscale magnetic materials to remove and regenerate Cr(VI) from aqueous effluents. The application of magnetic particle technology to resolve environmental problems is one of several new and innovative methods that have received considerable attention in recent years.12-14 Compared with conventional separation, the advantages of magnetic separation come from its speed, accuracy, and simplicity. Magnetic particles can be used to adsorb contaminants from aqueous effluents, and after the adsorption is completed, the adsorbent can be separated from the solution by a simple magnetic process. Furthermore, the adsorbed metals can be concentrated into a small volume by stripping metals off of the adsorbent’s surface using an appropriate stripping agent for reuse of the metals as well as the magnetic particles. However, most of the materials used in such processes have the disadvantage of a small adsorption capacity or slow adsorption rates due to their small surface areas or their porous properties, respectively. These disadvantages limit the applications of magnetic particles.15 Nanosized magnetic particles overcome these problems and can produce larger specific surface areas and thus may result in a higher adsorption capacity for metal removal.16 In our earlier efforts, nanoscale magnetite was produced and applied successfully to the removal of Cr(VI) from electroplating wastewater, and synthesized maghemite nanoparticles were shown to be effective in the selective removal of heavy metals.10,11 In addition to studies on the removal and recovery of heavy metals using some common magnetic nanoparticles, attempts have been made to extend the application of other magnetic nanoparticles to the field of industrial wastewater treatment. Recently, a comparative study on the adsorption/desorption of Cr(VI) using various synthesized, nanoscale ferrites was conducted, and MnFe2O4 nanoparticles had the highest Cr(VI) adsorption efficiency and the shortest adsorption time.11 Thus, we focus on this material in this study. Because of a possible chemical reaction between MnFe2O4 and Cr(VI) that may affect the recovery of the adsorbent and adsorbate, our MnFe2O4 nanoparticles were chemically modified to realize the efficient recovery of the used adsorbent as well as the Cr(VI) ions. Generally, coexisting ions and ligands, variable ionic strengths, and background electrolytes in industrial wastewater unavoidably affect the adsorption of targeted metals. Consequently, our objectives in this study were to (1) investigate the removal and recovery of Cr(VI) using surface-modified MnFe2O4 nanoparticles in a “ligand-free” environment; (2) examine the effects of organic and inorganic ligands, varying ionic strength, and background electrolytes; and (3) explore the mechanisms of Cr(VI) uptake onto modified MnFe2O4. (10) Hu, J.; Lo, I. M. C.; Chen, G. H. Water Sci. Technol. 2004, 50, 139. (11) Hu, J.; Lo, I. M. C.; Chen, G. H. International Symposium on Nanotechnology in Environmental Protection and Pollution; Asia Pacific Nanotechnology Forum (APNF): Bangkok, 2005. (12) Kaminski, M. D.; Nunez, L. J. Magn. Magn. Mater. 1999, 194, 31. (13) Navratil, J. D.; Tsair, M. T. S. Water Sci. Technol. 2002, 47, 29. (14) Oliveira, L. C. A.; Rios, R. V. R. A.; Fabris, J. D.; Sapag, K.; Garg, V. K.; Lago, R. M. Appl. Clay Sci. 2003, 22, 169. (15) Schwarz, J. A.; Contescu, C. I. Surfaces of Nanoparticles and Porous Materials; Marcel Dekker: New York, 1999. (16) Peng, Z. G.; Hidajat, K.; Uddin, M. S. J. Colloid Interface Sci. 2000, 271, 277.

Hu et al.

Materials and Methods In this study, all chemical stock solutions were prepared from reagent-grade chemicals using Milli-Q ultrapure water. NaNO3, Na2SO4, NaOH, and NH4OH were obtained from Nacalai Tesque (Japan). HNO3, Fe(NO3)3‚6H2O, and Mn(NO3)2‚4H2O were obtained from Fisher. EDTA and K2CrO4 were obtained from Riedel deHae¨n (Germany). Preparation of Modified MnFe2O4 Nanoparticles. Jacobsite (MnFe2O4) nanoparticles were prepared in the laboratory by combining the appropriate amounts of Mn(II) salt and Fe(III) salt in an alkaline solution that caused the spinel ferrite, MnFe2O4, to precipitate from the solution following eq 1:

Mn(NO3)2 + 2Fe(NO3)3 + 8NaOH ) MnFe2O4(s) + 8NaNO3 + 4H2O (1) First, 200 mL of purified, deoxygenated water was bubbled with nitrogen gas for 30 min, and 5 g of Mn(NO3)2‚4H2O and 14 g of Fe(NO3)3‚6H2O were dissolved in ultrapure water with vigorous mechanical stirring. Then, a 2.0 M NaOH solution was added dropwise to the above mixture to reach pH 11 while stirring. The mixture was then heated to 100 °C and kept at this temperature for 2 h. Black precipitate was collected under an external magnetic field and washed with ultrapure water. This washing was repeated three times to remove the impurities (e.g., OH-, NO3-, Na+) associated with the procedures. Finally, pure MnFe2O4 nanoparticles were obtained after freeze drying. Depending on the preparation conditions, such as temperature and pH, jacobsite (MnFe2O4) may contain Mn(III) or Fe(II) ions in addition to Mn(II) and Fe(III). Specifically, the MnFe2O4 particles produced at a lower temperature (100°C) are composed of solely Mn(II) and Fe(III).17 The Mn(II) from MnFe2O4 will be easily oxidized by adsorbed Cr(VI), and hence the regeneration of MnFe2O4 using NaOH will become ineffective. To avoid a possible chemical reaction between Cr(VI) and the adsorbent particles, the surface of pure MnFe2O4 should be modified using an oxidizing agent until external Mn(II) oxidizes to its highest valence. Mn(II) can be fully oxidized to Mn(IV) by heating Mn(II) in a strong basic solution with oxygen added continuously. Consequently, the collected MnFe2O4 particles were added to 200 mL of a 2 M NaOH solution while bubbling condensed air and stirring. The mixture was then heated in a 100 °C water bath for another 2 h. The surface-modified MnFe2O4 nanoparticles were separated via a magnetic field and washed with ultrapure water. Finally, the modified MnFe2O4 nanoparticles were collected after freeze drying. This modifying reaction follows eq 2:

MnFe2O4 + 2NaOH + 2O2 ) MnO2 + Fe2O3 + Na2Fe2O4 + H2O (2) Batch Experiments. A chromium sample was prepared by dissolving a known quantity of potassium chromate (K2CrO4) in ultrapure water; it was then used as a stock solution. Adsorption kinetic studies were carried out by mixing 0.2 g of modified MnFe2O4 nanoparticles with 40 mL of K2CrO4 solution of varying concentration in a rotary shaker with 125 mL stopper conical flasks. All of the adsorption experiments were conducted at room temperature (25 °C) and pH 2 unless otherwise noted. Most of the Cr(VI) solutions had a matrix of 0.1 M NaNO3 to keep the ionic strength relatively constant. One set of experiments used a 0.01 M NaNO3 solution to determine the possible effect of ionic strength on Cr(VI) adsorption. The conditions affecting Cr(VI) adsorption onto modified MnFe2O4 nanoparticles were studied by systematically varying the total Cr(VI) concentration, pH, ionic strength, and various ligands. The pH of the suspensions was controlled using 0.1 M HNO3 and 0.1 M NaOH solutions. To explore the competitive effect of various coexisting ligands (EDTA, NH4+, and SO42-) in industrial wastewater on the removal of Cr(VI), binary solutions containing Cr(VI) and each of the these ions were prepared. Forty milliliters of a 100 mg/L Cr(VI) solution containing the same molar ratio of each ligand as Cr(VI) and 0.2 g of jacobsite nanoparticles were shaken under pH ranging from (17) Bonsdorf, G.; Schafer, K.; Teske, K.; Langbein, H.; Ullmann, H. Solid State Ionics 1998, 110, 73.

Fast Removal and Recovery of Cr(VI)

Langmuir, Vol. 21, No. 24, 2005 11175

2 to 10 until equilibrium conditions were reached. Adsorption equilibrium studies were conducted at solution pH values of 2, 4, 7, and 10 by varying the initial concentration of Cr(VI) from 10 to 200 mg/L. Commonly, adsorption isotherm equations are derived on the basis of the relation between the amount of metal ion in the solution and the amount that is adsorbed onto the adsorbent particles when the two phases are in equilibrium at a given temperature. An adsorption isotherm was used to characterize the maximum adsorption capacity of the given adsorbent. Because data on adsorption from a liquid-phase fit the Langmuir equation is easily found, the adsorption isotherm was tested to validate the Cr(VI) uptake behavior of the modified MnFe2O4 nanoparticles. The Langmuir model assumes that the adsorption of metal ions occurs on a homogeneous surface by monolayer adsorption without any interaction between the adsorbed ions and is expressed as

Figure 1. XRD image of modified MnFe2O4 nanoparticles.

Ce Ce 1 ) + qe qm bqm

(3)

where Ce is the equilibrium concentration in mg/L, qe is the amount adsorbed at equilibrium in mg/g, and qm and b are the Langmuir constants that are related to the adsorption capacity and apparent heat change, respectively. In a wastewater treatment that uses adsorption, regeneration of the adsorbent is crucially important, and the desorption efficiency directly determines the potential reuse of the adsorbent. Desorption studies were conducted by mixing 0.1 g of Cr(VI)loaded modified MnFe2O4 nanoparticles and 5 mL of 0.01 M NaOH and shaking for 30 min. The desorption efficiency was calculated according to the following equation

CedVd desorption efficiency ) × 100% qeM

external area of the particles and aggregates. Thus, the BET surface area was considered to be a lower estimate of the surface area of ultrasmall particles. The zeta potential of the jacobsite nanoparticles obtained at different pH values was determined using a zeta potential analyzer (ZETA PLUS, Brookhaven Instruments Corporation). The concentrations of chromium, manganese, and iron were measured by an inductively coupled plasma-atomic emission spectrometer (ICP-AES, Optima 3000XL, Perkin Elmer) in accordance with the standard method.18 The concentrations of Cr(VI) and anions were measured using an ion chromatograph (IC, Dionex) according to the diphenyl carbazide method.19

Results and Discussion (4)

where Ced is the equilibrium concentration in mg/L, Vd is the volume of the eluant, qe is the adsorption capacity, and M is the mass of the adsorbent nanoparticles. Six successive cycles of adsorption-desorption, using the same adsorbent nanoparticles, were monitored in the most potent eluant, 0.01 M NaOH, found previously10 to assess the regeneration ability of these nanoparticles. To realize the complete desorption of all of the adsorbed Cr(VI), the surface-modified MnFe2O4 nanoparticles were separated using a magnet after each cycle and then added to another 2 mL of same concentration of NaOH and shaken for 30 min. The adsorbent nanoparticles were thoroughly washed with ultrapure water to neutrality and reconditioned for adsorption in the succeeding cycle. After the adsorption/desorption had reached equilibrium, the adsorbent nanoparticles were separated via an external magnetic field with the supernatant collected for metal analysis. Their capacity for the readsorption of Cr(VI) ions in repeated cycles was analyzed and compared. To ensure the accuracy, reliability, and reproducibility of the collected data, all of the experiments were conducted in duplicate, and the average values of the two data sets were presented. Morphology and Metal Analysis Methods. The dimensions of the synthesized materials were examined by transmission electron microscopy (TEM) (JEOL-2010). Each sample was prepared by placing a very dilute particle suspension onto 400 mesh carbon grids coated with Formvar film. The elemental information and structure of the synthesized materials were determined by an X-ray diffractometer (XRD) (PW-1830 Philips) at ambient temperature. The instrument was equipped with a copper anode generating Cu KR radiation (λ ) 1.5406 Å). The elemental composition and chemical oxidation state of the surface and near-surface species were investigated by X-ray photoelectron spectroscopy (XPS) (PHI-5600). XPS measurements were made with an Al KR X-ray source at a constant retard ratio of 40. The magnetic behavior was analyzed by a vibrating sample magnetometer (VSM) (7037/9509-P). Hysteresis measurements were performed at 300 K with applied magnetic fields up to 1 T. The surface area of the particles was measured by a Brunauer, Emmett, and Teller (BET) surface area analyzer (SA-3100, Coulter) for nitrogen adsorption. The BET method was carried out under relatively high vacuum and measured primarily the

Characterization of Modified MnFe2O4 Nanoparticles. A typical TEM micrograph of the surfacemodified MnFe2O4 particles indicated multidispersed particles with a mean diameter of 10 nm. The electron diffraction pattern indicated that the modified MnFe2O4 particles were highly crystallized. The XRD image of synthesized particles (Figure 1) shows the appearance of R-Fe2O3 and β-MnO2 peaks in addition to the crystalline MnFe2O4. The surface area of the freeze-dried particles, measured using the BET method, was 208 m2/g. The point of zero charge, pHpzc, of the modified MnFe2O4 was approximately 6.5. The superparamagnetic properties of the magnetic particles were verified by the magnetization curve measured by VSM, where no reduced remanence and coercivity were observed. From the hysteresis loop, often used to characterize the magnetic performance of the obtained sample, the maximum saturation magnetization (σmax) of the modified MnFe2O4 was found to be 3.6 emu/g at 300 K. This result could be explained by the fact that magnetic particles smaller than 30 nm will exhibit superparamagnetism.20 The saturation magnetization of modified MnFe2O4 (3.6 emu/g), which is slightly lower than the reported magnetization of 4.3 emu/g for pure MnFe2O4, is expected because nonmagnetic R-Fe2O3 and β-MnO2 covered the surface of the particles, lowering the overall magnetization of the original material.21 However, this large saturation magnetization of magnetic particles can still make them very susceptible to magnetic fields without any permanent magnetization and therefore render an easy solid- and liquid-phase separation. (18) APHA; APWA; WPCF. Standard Methods for the Examination of Water and Wastewater, 18th ed.; American Public Health Association: Washington, DC, 1992. (19) Muthukumaran, K.; Balasubramanian, N.; Ramakrishna, T. V. Met. Finish. 1995, 4, 46. (20) Watson, J. H. P.; Cressey, B. A. J. Magn. Magn. Mater. 2000, 214, 13. (21) Balaji, G.; Gajbhiye, N. S.; Wilde, G.; Weissmuller, J. J. Magn. Magn. Mater. 2002, 242, 617.

11176


Figure 2. Kinetic studies of Cr(VI) adsorption onto modified MnFe2O4 nanoparticles.

Adsorption Kinetic Studies. The effect of contact time on the adsorption of Cr(VI) is shown in Figure 2. It can be seen that the Cr(VI) uptake was finished within 5 min with 95% of the Cr(VI) removed during the first minute of the reaction. The equilibrium time is independent of the initial concentration of Cr(VI). In the results of our previous study,11 the equilibrium time of Cr(VI) adsorption by unmodified MnFe2O4 was also 5 min, which indicates that the surface modification did not affect the adsorption time. The rapid adsorption of Cr(VI) by nanoscale MnFe2O4 is perhaps caused by external surface adsorption, which is different from the microporous adsorption process. Because almost all of the adsorption sites of MnFe2O4 nanoparticles exist in the exterior of the adsorbent compared with their existance in the interior of porous adsorbents, it is easy for the adsorbate to access these active sites, thus resulting in a rapid approach to equilibrium. At equilibrium, the adsorption capacities of Cr(VI) at initial concentrations of 50 and 100 mg/L were found to be 9.99 and 19.36 mg/g, respectively. The similar shape of the curves at each initial concentration indicates that the percentage adsorption of Cr(VI) decreased with an increase in the initial Cr(VI) concentration. It should be noted that after treatment with 5 g/L of modified MnFe2O4 the initial 50 mg/L of Cr(VI) was reduced to 0.03 mg/L, which is below the discharge limit for Cr(VI) released into inland surface waters.22 This result is very promising for practical applications because there is no need for a further treatment unit following the adsorption process because the concentration of Cr(VI) in the effluent undergoing adsorption can meet the discharge requirement. Furthermore, the modified MnFe2O4 nanoparticles were advantageous to the adsorption equilibrium time when compared with other adsorbents such as activated carbon and clay.23,24 Effects of pH, Ionic Strength, and Ligands. Ionic strength is a parameter worth considering because a number of metal-bearing waste streams (e.g., electroplating waste) contain large amounts of total dissolved salts. In Figure 3, the Cr pH-adsorption curves for modified MnFe2O4 nanoparticles at ionic strengths of 0.01 and 0.1 M NaNO3 are presented. We found that metal removal decreased slightly as the ionic strength increased. McBride25 suggested that adsorbate ions forming outer(22) Environmental Protection Agency (EPA). Environmental Pollution Control Alternatives EPA/625/5-90/025, EPA/625/4-89/023; Environmental Protection Agency: Cincinnati, OH, 1990. (23) Aoyama, M.; Tsuda, M. Wood Sci. Technol. 2001, 35, 425. (24) Low, K. S.; Lee, C. K.; Low, C. H. J. Appl. Polym. Sci. 2001, 82, 2128. (25) McBride, M. B. Clays Clay Miner. 1997, 45, 598.

Hu et al.

Figure 3. Effect of ionic strength on the adsorption of Cr(VI) by modified MnFe2O4 nanoparticles.

Figure 4. Adsorption of Cr(VI) by modified MnFe2O4 nanoparticles in the presence of SO42-, NH4+, and EDTA.

sphere complexation with adsorbent surfaces show decreasing adsorption with increasing solution ionic strength. Changes in the ionic strength will alter the metal’s aqueous chemistry and the electric double layer surrounding the adsorbent surface. At higher ionic strengths, the surface is swamped by the electrolyte, and access to the surface is made more difficult than at lower ionic strengths. The adsorption of Cr(VI) showed a dependence on ionic strength, suggesting that Cr(VI) formed an outersphere adsorption complex at the jacobsite-water interface. The evaluation of the effect of the concentrations of background electrolytes on adsorption behavior can be considered to be a macroscopic method of inferring adsorption mechanisms. The presence of organic/inorganic ligands and competing adsorbates can alter metal removal from that observed in the metal-only system. Factors determining the effect that ligands have on metal adsorption include type and concentration of the ligand and metal, adsorbent type, and solution pH. In systems with more than one adsorbate, competition between the adsorbates for surface sites may occur. Generally, the degree of competition is dependent on the type and concentration of the competing ions, the number of surface sites, and the affinity of the surface for the adsorbate. The experimental data on Cr(VI) uptake in the absence and presence of organic and inorganic ligands at various pH values are presented in Figure 4. The general shape of the pH-adsorption curves in complex systems remained relatively unchanged compared with that of the ligand-free system. In such cases, the Cr(VI) adsorption capacity decreased sharply from around 18.9 to 2.6 mg/g when the solution pH was increased from 2 to 10. Obviously, at lower pH values, H+ ions are adsorbed on the surface so that the net charge is positive. This


Langmuir, Vol. 21, No. 24, 2005 11177 Table 1. Langmuir Constants for Cr(VI) Adsorption onto Modified MnFe2O4 at Different pH Values Langmuir constants pH

qm (mg/g)

b (L/mg)

R2

2.0 4.0 7.0 10.0

31.55 22.68 18.02 5.14

1.326 0.926 0.518 0.022

0.999 0.999 0.997 0.966

Table 2. Summary of Cr(VI) Adsorption Capacities of Various Adsorbents at Room Temperature (25 °C) type of adsorbents

Figure 5. Langmuir isotherms of Cr(VI) adsorption onto modified MnFe2O4 nanoparticles at different pH values.

qm (mg/g)

equilibrium time (h)

optimum pH

references

coconut tree sawdust lignin distillery sludge blast-furnace slag diatomite aluminum oxide anatase activated carbon beech sawdust hazelnut shell spent grain carbon slurry larch bark modified jacobsite

3.46

3

3.0

30

5.64 5.7 7.5 11.55 11.7 14.56 15.47 16.13 17.7 18.94 24.05 31.25 31.55

24 1.75 6 2 8 24 3 1.33 5 8 1 48 0.08

2.5 3.0 1.0 3.0 4.0 2.5 4.0 1.0 2.0 2.0 2.5 3.0 2.0

waste tyre

58.48

8

2.0

31 32 33 34 35 36 37 38 39 24 40 23 present study 28

enhances the adsorption of the negatively charged Cr(VI) anions. The surface charge is neutral at the point of zero charge, pHpzc, which is 6.5 for modified MnFe2O4 nanoparticles. As the pH increases, the OH- concentration increases on the surface, and Cr(VI) adsorption decreases dramatically. EDTA, sulfate, and ammonia are common coexisting ligands in some chromium plating wastewater. The binding of Cr(VI) by modified MnFe2O4 nanoparticles was thus measured separately in the presence of EDTA, sulfate, and ammonia ligands to investigate the possible enhanced or inhibited effects of ligand adsorption over the 0.1 M NaNO3 electrolyte. As observed from Figure 4, the coexisting EDTA and SO42- ligands individually inhibited the adsorption of Cr(VI) over the entire pH range studied. Several factors could contribute to the reduced capacity of the modified MnFe2O4 to adsorb Cr(VI) in the presence of EDTA and SO42-. EDTA and SO42- may consume surface sites of the adsorbent and hence reduce the available adsorption surface sites for Cr(VI) ions. They would also decrease the surface charge and thereby increase the electrostatic repulsion between the surface and the Cr(VI) anions. Moreover, aqueous EDTA and SO42formed strong complexes, which would have low affinities for the surface sites.26 Overall, these two ligands competed with Cr(VI) for active surface sites. Compared with SO42-, however, EDTA showed weaker competition at pH less than 6.5 and stronger competition at pH greater than 6.5. We attribute these results to the distribution of different EDTA species at various pH values. At lower pH, EDTA is more negatively charged than SO42-, whereas at higher pH, it is less negative than SO42-. More negatively charged ligands occupied more surface sites, leading to fewer sites for CrO42- anions. As far as the ammonium ligand is concerned, the presence of ammonium induced little change in the percentage adsorption of Cr(VI) at pH less than 6.5 but an increase in the adsorption of Cr(VI) at pH greater than 6.5. The small shift of the Cr(VI) adsorption curve in 0.1 M NaNO3 after the addition of NH4+ was consistent with an electrostatic effect induced by cation adsorption. NH4+ adsorption reduced the net negative charge in the double layer over that in 0.1 M NaNO3 alone and thus electrostatically encouraged additional anion retention. Equilibrium Studies. The data on Cr(VI) adsorption onto the modified MnFe2O4 nanoparticles as a function of dissolved concentration were modeled and fitted to the Langmuir isotherm equation (Figure 5). A linear relation is observed among the parameters in the plot at different quantities of adsorbent used, which indicates the ap-

plicability of the Langmuir equation. Therefore, only monolayer adsorption occurred on the surface, indicative of the homogeneity of the modified surface. The qm and b values and the correlation coefficients (R2) for the adsorption of Cr(VI) by modified MnFe2O4 nanoparticles at different pH values are listed in Table 1. The values of qm and b decreased with an increase in pH, which indicates that the lower pH favors adsorption. The adsorption capacity and equilibrium time of modified MnFe2O4 nanoparticles were compared with those of other adsorbents examined for the removal of Cr(VI) under similar conditions reported in the literature (Table 2). The surface-modified MnFe2O4 nanoparticle gives a much shorter adsorption equilibrium time. Most adsorbents listed in this Table are highly porous materials, providing adequate surface area for adsorption. However, the existence of intraparticle diffusion possibly leads to the decrease in the adsorption rate and available capacity, especially for macromolecules. Also, it has a relatively higher adsorption capacity than all of the stated adsorbents, except for waste tires. However, it should be noticed that the equilibrium time for Cr(VI) adsorption onto our adsorbent is only 5 min but is as long as 8 h for Cr(VI) adsorption onto waste tires.27 This result is very encouraging because the equilibrium time is an important parameter for an economical wastewater treatment plant. Adsorption Mechanisms. To explore any possible mechanisms of Cr(VI) adsorption onto the surface-modified MnFe2O4 nanoparticles, more investigations combining spectroscopic studies and experimental data were performed. For modified MnFe2O4, the concentration of Cr(VI) in the solution was always equal to that of total Cr; consequently, the chromium valence never changed during the adsorption/desorption processes. This suggests that no redox reaction occurred in the solution. To investigate the changes in elements on structure and valence, an XPS study was conducted. Cr(2p3/2) and Cr(2p1/2) spectra

(26) Dimitri, M.; Vladimir, G.; Abraham, W. Ion Exchange; Marcel Dekker: New York, 2000.

(27) Hamadi, N. K.; Chen, X. D.; Farid, M. M.; Lu, M. G. Q. Chem. Eng. J. 2001, 84, 95.

11178


Hu et al.

Figure 6. XPS spectra of modified MnFe2O4 nanoparticles (a) before and (b) after the adsorption of Cr(VI).

collected from a modified MnFe2O4 surface after adsorption at pH 2 are shown in Figure 6. The peak for Cr(2p3/2) was centered at 579.8 eV and the peak for Cr(2p1/2) was centered at 587.2 eV, and both were entirely from Cr(VI).28 The Fe(2p3/2) spectrum registered at 711.5 eV, and the Fe(2p1/2) spectrum registered at 724.3 eV in the XPS spectra, which indicated fully oxidized iron on the surface.29 Mn(2p3/2) and Mn(2p1/2) peaks centered at 642.8 and 653.9 eV, respectively, came from either Mn(II) or Mn(VI) because both are hardly distinguished only from their spectra.17 However, the valence of the adsorbed chromium did not change as discussed above; the valence of manganese should not vary because iron is fully oxidative. The valence of Mn is thus solely tetravalent. Furthermore, a comparison of Figure 6a and b shows that the status of iron and chromium was not affected by the adsorption reaction. The iron and manganese charges on the surface complexes of chromium and the modified jacobsite are the same as those on the pristine substrate, indicating that the adsorption of chromium on modified MnFe2O4 does not involve a noticeable electron transfer between the surface and the adsorbate, suggesting that these anions form weak bonds with the modified MnFe2O4 substrate. Thus far, it can be deduced that there is not any chemical redox reaction that occurs both on the adsorbent surface and in the Cr(VI) solution. Our results show that the uptake of Cr(VI) by modified MnFe2O4 is through physical adsorption at lower pH, which is mainly due to electrostatic attraction. However, (28) Ding, M.; Jong, B. H. W. S.; Roosendaal, S. J.; Vredenberg, A. Geochim. Cosmochim. Acta 2000, 64, 1209. (29) Qiu, S. R.; Lai, H. F.; Roberson, M. J.; Hunt, M. L.; Amrhein, C.; Giancarlo, L. C.; Flynn, G. W.; Yarmoff, J. A. Langmuir 2000, 16, 2230.

at pH greater than pHzpc, no Cr(VI) should be adsorbed on the modified MnFe2O4 nanoparticles if electrostatic repulsion is the only mechanism in this pH range. However, a certain number of Cr(VI) ions were still adsorbed on the modified MnFe2O4 at pH greater pHzpc (Figure 3), and thus there must be other mechanisms involved in this chemical process. Because the affinity of chromate anions with metal oxide is higher than that of hydroxide with metal oxide, the chromate can replace the hydroxide from the surface of the hydrolyzed metal oxides.26 We propose that ion exchange is involved along with electrostatic repulsion in this range. The adsorption of Cr(VI) on modified MnFe2O4 is through a combination of an electrostatic interaction with an ion exchange. Moreover, the surface of the particles can be verified to be completely covered by R-Fe2O3 and β-MnO2, deduced from both the XRD and XPS results. At pH below pHpzc, the surface is protonated and expressed as MeOH2+ (Me represents Fe and Mn), and the positively charged surface attracts chromate through Coulombic interaction as shown in eqs 5 and 6; at pH above pHpzc, the surface is hydroxylated and expressed as MeO-, and the negatively charged surface can exchange chromate anions as shown in eqs 7 and 8:

≡MeOH2+ + HCrO4- ) MeOH2+-HCrO4-

(5)

≡MeOH2+ + CrO42- ) MeOH2+-CrO42-

(6)

≡MeO- + HCrO4- + H2O ) MeOH-HCrO4- + OH(7) ≡MeO- + CrO42- + H2O ) MeOH-CrO42- + OH(8)


Adsorbent Life Test. In many applications, reuse of the adsorbent through regeneration of its adsorption properties is an economic necessity. With the rising prices of raw materials and wastewater treatment processes, the attractiveness of product recovery processes has increased significantly. Adsorption is an unsteady process. It is therefore desirable to achieve the desorption of Cr(VI) with a proper adsorbent. Magnetic particles are superparamagnetic; that is, they do not become permanently magnetized after aggregate formation. The particles may therefore be reused without sacrificing adsorption capacity. In this study, the adsorption of Cr(VI) onto the modified MnFe2O4 nanoparticles was highly dependent on pH. The desorption of Cr(VI) can be achieved by increasing the solution pH. We treated 0.1 g of Cr(VI)adsorbed modified MnFe2O4 with 5 mL of 0.01 M NaOH. The concentration of Cr(VI) in NaOH solution was 620 mg/L. According to eq 4, the desorption efficiency was calculated to be 98.9%; that is, most of the Cr(VI) ions were desorbed. By comparison, Sharma and Forster6 reported that exhausted peat is hardly recovered because of the strong peat-Cr(VI) bond and thus requires disposal through incineration. Aoyama and Tsuda23 showed that only 17.5% of adsorbed Cr(VI) could be desorbed using 0.1 M HNO3 but it was completely recovered by burning Crladen bark. Selvaraj et al.32 discovered that a maximum Cr(VI) desorption efficiency of 82% was achieved using 0.014 M NaOH. It is clear that the regeneration of adsorbents for reuse purposes has not been widely studied, especially when considering economical and technical issues. In this study, to achieve the complete desorption of the adsorbed Cr(VI) from the adsorbent, another 2 mL of the same concentration of NaOH was used for continuous desorption until equilibrium was reached when the nanoparticles were thoroughly washed with ultrapure water and freeze dried for XPS investigation. We observed that no chromium peak was present in the XPS spectra, indicating that the remaining chromium was entirely removed from the nanoparticles. Studies pertaining to the regeneration of modified MnFe2O4 nanoparticles and readsorption of Cr(VI) were conducted in six consecutive cycles. The results are shown in Figure 7. As can be seen, in the first cycle, the adsorption capacity is 31.4 mg of (30) Selvi, K.; Pattabi, S.; Kadirvedu, K. Biores. Technol. 2001, 80, 87. (31) Lalvani, S. B.; Hubner, A.; Wiltowski, T. S. Energy Sources 2000, 22, 45. (32) Selvaraj, K.; Manonmani, S.; Pattabhi, S. Bioresour. Technol. 2003, 89, 207. (33) Srivastava, S. K.; Gupta, V. K.; Mohan, D. J. Environ. Eng. 1997, 123, 461. (34) Dantas, T. N. D.; Neto, A. A. D.; Moura, M. C. P. Water Res. 2001, 35, 2219. (35) Gupta, V. K.; Morhan, D.; Sharma, S.; Park, K. T. The Environmentalist 1999, 19, 129. (36) Weng, C. H.; Wang, J. H.; Huang, C. P. Water Sci. Technol. 1997, 35, 55. (37) Sandhya, B.; Tonni, A. K. Chemosphere 2004, 54, 951. (38) Acar, F. N.; Malkoc, E. Bioresour. Technol. 2004, 94, 13. (39) Cimino, G.; Passerini, A.; Toscano, G. Water Res. 2000, 34, 2955. (40) Singh, V. K.; Tiwati, P. N. J. Chem. Technol. Biotechnol. 1997, 69, 376.

Langmuir, Vol. 21, No. 24, 2005 11179

Figure 7. Regeneration studies of modified MnFe2O4 nanoparticles after six cycles.

Cr/g of nanoparticles. In the sixth cycle, the adsorption capacity is 31.3 mg/g. By comparison, the adsorption capacity remained almost unchanged during the six cycles, indicating that there were no irreversible sites on the surface of the adsorbent and suggesting a high recovery capacity of the adsorbent. In addition to chromium, iron and manganese were measured using ICP after each adsorption/desorption process, and the concentrations of these two metals were found to be nearly zero, indicating that the dissolution of the nanoparticles under the stated experimental conditions was not a concern. Therefore, the stability and durability of the adsorbent during the adsorption and desorption processes were verified. Conclusions In this study, an innovative method combining nanoparticle adsorption and magnetic separation was developed for the removal and recovery of Cr(VI) from wastewater. The surface-modified MnFe2O4 nanoparticles were synthesized, characterized, and examined as adsorbents of Cr(VI). Results show that surface-modified MnFe2O4 nanoparticles are efficient adsorbents for the fast removal of Cr(VI) from aqueous solutions. The adsorption process followed by magnetic separation leads to the rapid and inexpensive removal of metal ions; furthermore, the recovery of these materials in a highly concentrated material form is possible. This system has the potential to overcome the separation difficulties associated with adsorption-based treatment techniques with its advantages of high performance, high capacity, and low space requirements. Further experiments to validate the efficiency of this material with respect to the removal of chromium in a continuous upscale system may be of value in developing sound remediation strategies for water contaminated with toxic heavy metals. Supporting Information Available: Magnetization curve of the nanoparticles. This material is available free of charge via the Internet at http://pubs.acs.org. LA051076H