Langmuir 2004, 20, 9889-9892
9889
Immobilization and Recovery of Au Nanoparticles from Anion Exchange Resin: Resin-Bound Nanoparticle Matrix as a Catalyst for the Reduction of 4-Nitrophenol Snigdhamayee Praharaj, Sudip Nath, Sujit Kumar Ghosh, Subrata Kundu, and Tarasankar Pal* Department of Chemistry, Indian Institute of Technology, Kharagpur-721302, India Received June 3, 2004. In Final Form: August 20, 2004 The immobilization of gold nanoparticles in anion exchange resin and their quantitative retrieval by means of a cationic surfactant, cetylpyridinium chloride, is studied. The resin-bound gold nanoparticles (R-Au) have been used successfully as a solid-phase catalyst for the reduction of 4-nitrophenol by sodium borohydride. At the end of the reaction, the solid matrix remains activated and separated from the product. The recycling of catalyst particles after the quantitative reduction of 4-nitrophenol and the recovery of gold nanoparticles with unaffected particle morphology from the resin-bound gold nanoparticle entity have been reported.
Introduction In recent years, the immobilization of metal nanoparticles in a solid matrix has fascinated scientists because of its immense importance in nanotechnology. Such investigation has been an integral part in fabricating practical heterogeneous catalysts.1,2 The storage of nanoparticles in a solid template and their recovery, whenever necessary, has been proved to be an important field to study. The reserved colloidal nanoparticles in solid matrixes can easily be exploited for diverse chemical reactions under different solvent systems, which eliminates the need for their phase transfer process. The most widely reported supports used to adsorb nanoparticles are inorganic solids such as charcoal,3 silica,4 alumina,5 and oxides such as mesoporous TiO26 or MgO7 and so forth. Grafting on the solid support by chemical bonds is another way to immobilize nanoparticles where polyacrylamide gel is used for this purpose.8,9 Latex10 and resins11 are also used for stabilization of nanoparticles as polymeric supports. High surface-to-volume ratio, exhibition of the quantum size effect, chemical reactivity, and other physical properties propelled the immobilization and regeneration of gold nanoparticles in their nanometric size regime. Among the known colloidal nanoparticles, gold is widely studied because of its characteristic optical, spectroscopic, and catalytic properties.12-14 Several methods have been * To whom correspondence should be addressed. E-mail: tpal@ chem.iitkgp.ernet.in. (1) Sun, L.; Crooks, M. Langmuir 2002, 18, 8231. (2) Reetz, M. T.; Koch, M. G. J. Am. Chem. Soc. 1992, 121, 7933. (3) Bo¨nnemann, H.; Korall, B. Angew. Chem., Int. Ed. Engl. 1992, 31, 1490. (4) Wang, Q.; Liu, H.; Wang, H. J. Colloid Interface Sci. 1997, 190, 380. (5) Reetz, M. T.; Quaiser, S. A.; Breinbauer, R.; Tesche, B. Angew. Chem., Int. Ed. Engl. 1995, 34, 2728. (6) Stathatos, E.; Lianos, P. Langmuir 2000, 16, 2398. (7) Rodriguez, J. A.; Perez, M.; Jirsak, T.; Evans, J.; Hrbek, J.; Gonzalez, L. Chem. Phys. Lett. 2003, 378, 526. (8) Hirai, H.; Ohtaki, M.; Komiyama, M. Chem. Lett. 1986, 269. (9) Ohtaki, M.; Toshima, N.; Komiyama, M.; Hirai, H. Bull. Chem. Soc. Jpn. 1990, 63, 1433. (10) Mayer, A. B. R.; Mark, J. E. J. Polymer. Sci., Part B: Polym. Phys. 1997, 35, 1207. (11) Nakao, Y.; Kaeriyama, K. J. Colloid Interface Sci. 1989, 131, 186. (12) Link, S.; El-Sayad, M. A. J. Phys. Chem. B 1999, 103, 4212. (13) Muvany, P. Langmuir 1996, 12, 788. (14) Pradhan, N.; Pal, A.; Pal, T. Langmuir 2001, 17, 1800.
reported to immobilize gold nanoparticles (NPs) in resins for use in different purposes. The construction of a covalently linked Au-NPs micropattern from the selfassembled multilayer film composed of notro-diazoresin and 4-mercaptophenol capping has been reported.15 Worden et al.16 employed polystyrene (PS) Wang resin for controlling the surface functionality of nanoparticles. Later on, Sung et al.17 described a versatile monofunctionalization procedure for gold NPs with L-lysine using PS Wang resin beads. However, the storage of gold nanoparticles in a solid matrix and their recovery with unaltered particle morphology is still a challenge to the researchers. In this paper, we have reported the immobilization of gold nanoparticles in Amberlite IRA-400 anion-exchange resin followed by their quantitative retrieval by means of a cationic surfactant, cetylpyridinium chloride. The recovery of nanoparticles from the resin bed in principle is unique in terms of exchangeable ions and in situ stabilization. The catalytic activity of the resin-bound gold nanoparticles (R-Au) has also been addressed for the reduction of 4-nitrophenol by sodium borohydride. The reduction of nitrophenols to their corresponding amino derivatives has been extensively studied in our laboratory using a number of coinage and platinum metal nanoparticles and their bimetallics as catalysts.14,18 The exploitation of resin-bound gold nanoparticles for the nitrophenol reduction is a definite departure from the usual track in terms of product separation. This is a novel and fruitful introduction of a solid-phase catalyst, which at the end of the reaction remains activated and separated from the product. The quantitative conversion of nitrophenol to the aminophenol and an effective recovery of the catalyst particles are important aspects of this report. To the best of our knowledge, there exists no other report of solidphase catalysis using the resin-bound gold nanoparticles under consideration. (15) Lu, C.; Wu, N.; Jiao, X.; Luo, C.; Cao, W. Chem. Commun. 2003, 1056. (16) Worden, J. G.; Shaffer, A. W.; Huo, Q. Chem. Commun. 2004, 518. (17) Sung, K.-M.; Mosley, D. W.; Pelle, B. R.; Zhang, S.; Jacobson, J. M. J. Am. Chem. Soc. 2004, 126, 5064. (18) Ghosh, S. K.; Mandal, M.; Kundu, S.; Nath, S.; Pal, T. Appl. Catal. A 2004, 268, 61.
10.1021/la0486281 CCC: $27.50 © 2004 American Chemical Society Published on Web 10/14/2004
9890
Langmuir, Vol. 20, No. 23, 2004
Letters
Experimental Section All the reagents used were of AR grade. Double-distilled water was used throughout the experiment. Amberlite IRA-400 resin was purchased from BDH Chemicals, U.K. Chloroauric acid (HAuCl4) and sodium citrate were used as received from Merck. Four surfactants, cetylpyridinium chloride (CPC), cetyltrimethylammonium bromide (CTAB), sodium dodecyl sulfate (SDS), and poly(oxyethylene)isooctyl phenyl ether (TX-100) were purchased from Merck. The reagent 4-nitrophenol (Aldrich) was used after its repeated crystallization from petroleum ether and ethyl acetate. Sodium borohydride (NaBH4) was received from Sigma, and aqueous solution was prepared freshly in ice-cold water when it was required. All the absorption spectra were monitored in a Shimadzu UV-160 digital spectrophotometer taking the solution in a 1-cm quartz cuvette. The nanoparticles were characterized by transmission electron microscopy (TEM) studies in a Hitachi H-9000 NAR Instrument at a magnification of 100K. The sample was prepared by placing a drop of solution on a carbon-coated copper grid. Gold nanoparticles were prepared by Frens’ method.19 A 50 mL aqueous solution of HAuCl4 (0.25 mmol dm-3) was heated to boiling, and trisodium citrate (650 µL, 1% by wt) was added to it with continuous stirring. Within 25 s of boiling, the solution turned faint blue. After 70 s, the blue color suddenly changed into red, indicating formation of gold nanoparticles. The reaction mixture was boiled for ∼30 min for completion of the reaction. Absorption measurement of the finally formed gold colloid showed an intense absorption band with a maximum at 531 nm. For the activation of anion-exchange resin, 5 g of resin was first treated with NaCl (2 mol dm-3) solution for 2 h. The excess NaCl was removed by washing with water, and finally the resin became activated in the chloride form. Then 15 mL of 0.25 mmol dm-3 gold was immobilized in an anion-exchange resin column (10 cm length and 1.5 cm diameter) loaded with 5 g of the resin, and the flow rate of the citrate-capped Au solution was 2 mL per min. A gradual change in color of the resin bead from yellow to black substantiates the binding process. The resin in the column was further washed with plenty of water to remove unbound gold nanoparticles between the beads. For the recovery of the metallic nanoparticles from the resin beads, the resin-bound gold NPs were transferred to a beaker and 30 mL of aqueous CPC (50 mmol dm-3) was added in portions of 5 mL at the respective stages. The solution above the resin became pink instantaneously and showed an absorption maximum at 533 nm indicating the regeneration of gold NPs. Catalytic activity of the above resin-bound gold nanoparticles was substantiated in the following way. In a conical flask, 4 mL of aqueous solution of 4-nitrophenol (0.1 mmol dm-3) was taken and a freshly prepared aqueous solution of NaBH4 (15 mmol dm-3) was introduced. Then, 0.2 g of the R-Au was added to the mixture and was stirred occasionally. After ∼1 h, the yellow color of the solution disappeared and again NaBH4 solution was added in portions (7.5 mmol dm-3) in four successive stages during the reaction period. The solution above the resin showed a continuous increase in absorption at ∼295 nm indicating the formation of 4-aminophenol.
Figure 1. (a) TEM images and (b) diameter histogram of citrate-capped gold nanoparticles.
Results and Discussion Anion-exchange resin is a polymer containing amine or quaternary ammonium groups as integral parts of the polymer lattice with an equivalent amount of anions such as chloride, hydroxyl, or sulfate.20 Amberlite IRA-400, a polystyrene quaternary ammonium resin supplied in chloride form, was activated by sodium chloride solution for efficient exchange of the incoming anions. Gold nanoparticles, as they bear a negative surface charge,21 cannot be immobilized on a cation-exchange resin bed, whereas strongly basic anion exchanger serves the purpose of immobilization. (19) Frens, G. Nat. Phys. Sci. 1973, 241, 20. (20) Vogel, I. Quantitative Inorganic Analysis, 3rd ed.; The English Language Book Society and Longman: London, 1969; p 704. (21) Mayya, K. S.; Sastry, M. Langmuir 1998, 14, 6344.
Figure 2. (a) TEM images and (b) diameter histogram of CPCstabilized gold nanoparticles after their recovery from resinbound gold nanoparticles.
The gold sol prepared by Frens’ method shows surface plasmon resonance at 531 nm. The position and shape of the plasmon absorption of the metallic nanoclusters are strongly dependent on the particle size, shape, dielectric constant of the medium, and surface-adsorbed species.22 The above sol remained immobilized by the resin and (22) Kreibig, U.; Vollmer, M. Optical Properties of Metal Clusters; Springer: Berlin, 1995.
Letters
Langmuir, Vol. 20, No. 23, 2004 9891
Scheme 1. Schematic Representation of the Immobilization and Extraction of Gold Nanoparticles from Resin
extracted by means of cationic surfactants. Two different surfactants were employed for the elution of Au NPs from the resin beds, and these surfactants subsequently stabilize the NPs from agglomeration. The extracted effluent shows a λmax at 533 and 534 nm for CPC and CTAB, respectively. TEM analysis (Figures 1 and 2) of the gold sol before and after extraction showed no significant change in the size and shape of the particles. Therefore, it is definite that the employment of the surfactant in the elution process results in a positional shift in the gold plasmon absorption band. Such immobilization and elution of gold nanoparticles in the resin has been schematically represented in Scheme 1. The preparation of the resin column is important for the efficient extraction of the gold particles from the resin. The gold nanoparticles could not be efficiently exchanged with the chloride form of the resin bed while the solutions were taken in a beaker. In that case, possible agglomeration of NPs might take place due to the electrolytic effect caused by the excess anions present in the solution. Once the resin bed binds with the Au NPs, they cannot be eluted from the resin using NaCl or HCl solutions. Moreover, these eluting agents facilitate subsequent aggregation of Au NPs in the effluent and hence were not used. Neither anionic nor nonionic surfactant is capable of extracting the metal particles quantitatively from the resin. Only the cationic surfactants CTAB and CPC happen to be efficient eluants to be used for this purpose. They not only elute the gold nanoparticles efficiently from the resin bed but also stabilize the regenerated sol. Hence, the cationic surfactants are the best suited eluting agent. These cationic surfactants possess a surface-active positive charge that is strongly adsorbed to the solid surfaces of the NPs which are already negatively charged. The surfactants have a long hydrocarbon chain containing headgroup -N+ and Cl- ions. In CPC, the pyridinium ring is adsorbed perpendicular to the surface of the NPs and the hydrocarbon chain is directed away from its surface.23 Presumably, due to the attachment of the positive end of the surfactant to the NPs surface, immobilized NPs are released and subsequently Cl- ions are attached to the resin beads. As the CPC-bound gold nanoparticles bear an overall positive surface charge, the further immobilization of the Au NPs in anion-exchange resin is inhibited. The above fact has been authenticated by a separate experiment, taking a sodium form of a cation exchange (Dowex-50) resin under consideration. The adsorption of CPC-bound gold in the cation-exchange resin substantiates the fact. The strong electrostatic interaction between water dipole and Au surface overcomes the hydrophobicity of the moiety. Such impact of electrostatic (23) Kreisig, S. M.; Tarazona, A.; Koglin, E.; Schwuger, M. J. Langmuir 1996, 12, 5279.
interaction is quite effective and leads to stable dispersion of nanoparticles in water. Among the two cationic surfactants tried, CPC has judiciously been chosen because of its better eluting capability than CTAB. The extent of elution with CTAB and CPC has been shown in Figure 3.
Figure 3. UV-visible spectra of gold nanoparticles after extraction from resin using CPC (50 mmol dm-3) and CTAB (50 mmol dm-3).
To study the catalytic activity of the resin-bound gold nanoparticles, the reduction of 4-nitrophenol with sodium borohydride was chosen as a model reaction. An aqueous solution of 4-nitrophenol shows a distinct spectral profile with an absorption maximum at 317 nm. Now addition of sodium borohydride solution results in the shifting of the peak position to 403 nm. This peak was due to the formation of 4-nitrophenolate ions in alkaline condition caused by the addition of NaBH4. The BH4- under the experimental condition (devoid of catalyst) is incapable of reducing 4-nitrophenol to the corresponding amino compound. Hence, the resin-bound gold entity has been employed as a catalyst for the effective reduction to occur. Addition of R-Au into the reaction mixture caused gradual fading of the yellow color of the solution. Such decolorization was reflected in the absorption spectrum which is shown in Figure 4. This is due to the steady exchange of phenolate ions with the chloride ion present in the resin. The ready exchange of Cl- ions was confirmed by AgNO3 test with the solution above the beads. The amount of BH4- (15 mmol dm-3) added was not sufficient to reduce nitrophenol due to immobilization of catalyst and substrate. Addition of excess BH4- (30 mmol dm-3) in four successive stages results in the reduction of 4-nitrophenol to 4-aminophenol, which has been authenticated spectrophotometrically. The gradual development of the peak at ∼295 nm confirms the increase in 4-aminophenol concentration in the reaction mixture (Figure 5). The R-Au helps to transfer electrons from BH4- to nitrophe-
9892
Langmuir, Vol. 20, No. 23, 2004
Letters Scheme 2. Schematic Representation of Reduction of 4-Nitrophenol
Figure 4. Absorption spectra of the successive adsorption of 4-nitrophenolate ions on the resin-bound gold at an interval of 10 min. Conditions: [4-nitrophenol] ) 0.1 mmol dm-3, [NaBH4] ) 15 mmol dm-3, R-Au ) 0.2 g.
from the reaction mixture. The catalyst was recovered, washed, and dried and could be recycled again for the reduction after eluting the product from the resin bed. Conclusion Figure 5. Gradual development of absorption spectra of 4-aminophenol at an interval of 1 h. Conditions: [4-nitrophenol] ) 0.1 mmol dm-3, [NaBH4] ) 30 mmol dm-3.
nolate and thus acts as a catalyst for the purpose. The adsorption of 4-nitrophenol on the R-Au surface and its reduction to 4-aminophenol are schematically presented in Scheme 2. The reduction of 4-nitrophenol was previously examined in our laboratory for the first time employing metallic hydrosols as catalysts. Esumi et al.24 used dendrimerencapsulated gold nanoparticles as catalysts for this purpose. The application of liquid suspensions of metal nanoparticles in catalysis is limited for its obvious drawbacks. With those systems, the separation of products and recycling of catalysts are not straightforward. This is the first fruitful employment of resin-bound gold nanoparticles as catalysts that overcomes the inherent difficulties. The resin-bound gold particles act as heterogeneous catalysts leading to complete conversion of 4-nitrophenol to the 4-aminophenol, which was isolated (24) Hayakawa, K.; Yoshimura, T.; Esumi, K. Langmuir 2003, 19, 5517.
We have reported the immobilization of gold nanoparticles in anion-exchange resin and their quantitative regeneration by means of efficient cationic surfactants. The inefficiency of anionic/nonionic surfactants and different electrolytes for the elution of the gold nanoparticles has also been discussed. It has been observed that the morphology of the gold particles remained unaltered after their elution from the resin. The resin-bound gold entity has effectively been employed as a catalyst for the reduction of 4-nirophenol to 4-aminophenol with complete selectivity for the product. The catalyst could be recycled a number of times for the reduction reaction. It is found that the R-Au acts as a heterogeneous catalyst in this case and the 4-aminophenol can be isolated from the aqueous solution without any contamination from colloidal particles. Acknowledgment. The authors are thankful to the University Grant Commission (UGC), the Council of Scientific and Industrial Research (CSIR), and the Department of Science and Technology (DST) New Delhi for financial assistance. LA0486281
