Capping of Gold Nanoparticles by the Amino Acid Lysine Renders

Langmuir 2003, 19, 3545-3549

Capping of Gold Nanoparticles by the Amino Acid Lysine Renders Them Water-Dispersible PR. Selvakannan, Saikat Mandal, Sumant Phadtare, Renu Pasricha, and Murali Sastry* Materials Chemistry Division, National Chemical Laboratory, Pune 411 008, India Received November 26, 2002. In Final Form: January 29, 2003

Introduction Following elucidation of the Brust protocol for the synthesis and capping of gold colloids with alkanethiols in nonpolar organic solvents,1 there have been a number of reports on the surface modification of gold colloids with alkanethiols2,3 and terminally functionalized alkanethiols4-6/arenethiols.7-10 The gold particles capped and stabilized with thiol-derivative monolayers may be obtained as a powder (by evaporation of the organic solvent) that is readily redispersible in nonpolar and weakly polar organic solvents. Murray and co-workers have shown that such monolayer-protected clusters (MPCs) may be viewed as novel chemical reagents wherein the gold nanoparticle core plays the role of a support for the reactive molecules constituting the shell.5,11 Development of synthesis protocols for realizing monolayer-protected water-dispersible gold nanoparticles, on the other hand, has received considerably less attention and would have immediate application in catalysis, sensors, molecular markers, and in particular, biological applications such as biolabeling and drug delivery. Recognizing that the solubility of gold, and in some instances silver, MPCs in water/strongly polar organic solvents would be dominated by the terminal functionality of the capping monolayer,5 a number of different thiol-coordinated functional groups such as tiopronin,12,13 glutathione,14 succinic acid,15 sulfonic acid,16 and ammonium ions17,18 have been shown to result in * To whom correspondence should be addressed. Ph: +91 20 5893044. Fax: +91 20 58939522/5893044. E-mail: sastry@ ems.ncl.res.in. (1) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801. (2) Porter, L. A.; Ji, D.; Westcott, S. L.; Graupe, M.; Czernuszewicz, R. S.; Halas, N. J.; Lee, T. R. Langmuir 1998, 14, 7378. (3) Logunov, S. L.; Ahmadi, T. S.; El-Sayed, M. A.; Khoury, J. T.; Whetten, R. L. J. Phys. Chem. B 1997, 101, 3713. (4) Weisbecker, C. S.; Merritt, M. V.; Whitesides, G. M. Langmuir 1996, 12, 3763. (5) Templeton, A. C.; Hostetler, M. J.; Kraft, C. T.; Murray, R. W. J. Am. Chem. Soc. 1998, 120, 1906. (6) Hostetler, M. J.; Green, S. J.; Stokes, J. J.; Murray, R. W. J. Am. Chem. Soc. 1996, 118, 4212. (7) Brust, M.; Fink, J.; Schiffrin, D. J.; Kiely, C. J. Chem. Soc., Chem. Commun. 1995, 1655. (8) Johnson, S. R.; Evans, S. D.; Brydson, R. Langmuir 1998, 14, 6639. (9) Chen, S.; Murray, R. W. Langmuir 1999, 15, 682. (10) Mayya, K. S.; Patil, V.; Sastry, M. Langmuir 1997, 13, 3944. (11) Templeton, A. C.; Hostetler, M. J.; Warmouth, E. K.; Chen, S.; Hartshorn, C. M.; Krishnamurthy, V. M.; Forbes, M. E. D.; Murray, R. W. J. Am. Chem. Soc. 1998, 120, 4845 and references therein. (12) Templeton, A. C.; Cliffel, D. E.; Murray, R. W. J. Am. Chem. Soc. 1999, 121, 7081. (13) Templeton, A. C.; Chen, S.; Gross, S. M.; Murray, R. W. Langmuir 1999, 15, 66. (14) Schaff, T. G.; Knight, G.; Shaffigulin, M. N.; Borkman, R. F.; Whetten, R. L. J. Phys. Chem. B 1998, 102, 10643. (15) Chen, S.; Kimura, K. Langmuir 1999, 15, 1075. (16) Shon, Y. S.; Wuelfing, W. P.; Murray, R. W. Langmuir 2000, 17, 1255.

3545

MPCs that are readily dispersible in water. A recent report describes the visible luminescence from such waterdispersible gold MPCs.19 However, gold nanoparticles capped with arylthiolates bearing OH-, COOH-, and NH2terminal functionalities were found to be soluble in polar organic solvents such as alcohols, acetone, and acetonitrile but not in water.7-9 Relatively more recent is the demonstration that amine derivatives complex with gold nanoparticles in a manner similar to that of thiol derivatives.20,21 Using amine chemistry for surface modification, we have shown that aqueous gold nanoparticles may be phase-transferred into nonpolar organic solvents by complexation with alkylamine molecules.22 This new approach opens up the exciting possibility of surface modification of gold nanoparticles with amino acids where binding with the gold nanoparticle surface may be accomplished through the amine functionality. In this paper, we report our finding that capping aqueous gold nanoparticles with the amino acid lysine stabilizes the particles in solution electrostatically and also renders them water-dispersible. The lysinecapped gold nanoparticles may be obtained in the form of a dry powder after evaporation of the aqueous component, this powder being extremely stable in air and readily redispersible in water. Development of protocols for the synthesis of water-dispersible nanoparticles has immense application in a variety of fields, but clearly more so in biorelated areas such as biolabeling and biosensing.23,24 The amino acid protected gold nanoparticles have been characterized by UV-vis spectroscopy, transmission electron microscopy (TEM), thermogravimetric analysis (TGA), and proton NMR spectroscopy. To the best of our knowledge, this is the first report on use of an amino acid for stabilizing and rendering gold nanoparticles waterdispersible. This study represents the first step in the possible application of amino acid derivatized gold nanoparticles as novel aqueous reagents in amide and esterification reactions in a manner similar to that demonstrated by Murray and co-workers for functionalized gold nanoparticles in an organic environment.5,11 Presented below are details of the investigation. Experimental Section Chemicals. Chloroauric acid (HAuCl4), sodium borohydride, and lysine were obtained from Aldrich Chemicals and used as received. Synthesis of Aqueous Lysine-Capped Gold Nanoparticles. One hundred milliliters of 10-4 M aqueous solution of chloroauric acid (HAuCl4) was reduced by 0.01 g of sodium borohydride (NaBH4) at room temperature to yield colloidal gold particles. This procedure results in a ruby-red solution containing gold nanoparticles of dimensions 65 ( 7 Å. The colloidal gold particles were capped by addition of 10 mL of an aqueous solution of 10-3 M lysine to 90 mL of the gold hydrosol. After addition of lysine and aging the gold colloidal solution for 12 h, this solution was subjected to ultracentrifugation to recover a pellet of lysinecapped gold nanoparticles. To further purify the gold nanoparticle solution by removal of uncoordinated lysine molecules, the pellet (17) Cliffel, D. E.; Zamborini, F. P.; Gross, S. M.; Murray, R. W. Langmuir 2000, 16, 9699. (18) Yonezawa, T.; Onoue, S.; Kimizuka, N. Langmuir 2000, 16, 5218. (19) Huang, T.; Murray, R. W. J. Phys. Chem. B 2001, 105, 12498. (20) Leff, D. V.; Brandt, L.; Heath, J. R. Langmuir 1996, 12, 4723. (21) Brown, L. O.; Hutchison, J. E. J. Phys. Chem. B 2001, 105, 8911. (22) Sastry, M.; Kumar, A.; Mukherjee, P. Colloids Surf., A 2001, 181, 255. (23) Cumberland, S. L.; Strouse, G. F. Langmuir 2002, 18, 269. (24) Gittins, D. J.; Caruso, F. ChemPhysChem 2002, 3, 110.

10.1021/la026906v CCC: $25.00 © 2003 American Chemical Society Published on Web 03/06/2003

3546

Langmuir, Vol. 19, No. 8, 2003

Figure 1. (A) UV-visible spectra of borohydride-reduced gold nanoparticles (curve 1), lysine-capped gold solution (curve 2), and redispersed lysine-capped gold nanoparticles in water (curve 3, pH 7). The spectra have been shifted vertically for clarity. (B) UV-visible spectra recorded from redispersed lysinecapped gold nanoparticles in water at pH 3 (curve 1), pH 7 (curve 2), and pH ) 10 (curve 3). Curve 4 corresponds to the spectrum shown as curve 3 after heating the solution at 90 °C for 10 min. was redispersed in deionized water and another cycle of centrifugation was carried out. The pellet was then redispersed in deionized water for UV-vis spectroscopy measurements and in D2O for 1H1 NMR analysis. The concentration of gold in the redispersed aqueous solution was 10-2 M. UV-Vis Spectroscopy Studies. The optical properties of the lysine-capped gold colloidal solution were monitored on a Hewlett-Packard diode array spectrophotometer (model HP-8452) operated at a resolution of 2 nm. UV-vis spectra of lysine-capped gold particles redispersed in water at different pH values were also recorded. TEM Measurements. Samples for TEM analysis were prepared by placing drops of the lysine-capped gold colloidal solutions on carbon-coated copper TEM grids. The films on the TEM grids were allowed to stand for 2 min, following which the extra solution was removed using a blotting paper and the grid was allowed to dry prior to measurement. TEM measurements were performed on a JEOL model 1200EX instrument operated at an accelerating voltage of 120 kV. 1H NMR Measurements. The purified powders of lysine 1 and lysine-capped gold were dispersed in D2O, and the proton (1H NMR) NMR spectra were recorded on a Bruker AC 200 MHz instrument and scanned in the range 0-15 ppm. Thermal Stability Measurements. TGA profiles of carefully weighed quantities of lysine and purified powders of lysine-capped gold nanoparticles were recorded on a Seiko Instruments model TG/DTA 32 instrument at a heating rate of 10 °C/min.

Results and Discussion Figure 1A shows the UV-vis spectra of the lysinecapped gold hydrosol at different stages of preparation. Curve 1 in the figure corresponds to the spectrum of gold colloidal solution obtained by borohydride reduction of aqueous chloroauric acid; curve 2 is the spectrum of gold colloidal solution after capping with lysine, and curve 3 is the spectrum of the redispersed purified powder of lysine-capped gold nanoparticles in water at pH 7. The spectra have been shifted vertically for clarity. A strong absorption in curve 1 at ca. 510 nm is observed that corresponds to excitation of surface plasmon vibrations in the gold nanoparticles. When the gold nanoparticles are capped with lysine, a broadening and red shift of the surface plasmon band are observed (curve 2) and indicate some aggregation of the gold nanoparticles consequent to surface modification. However, the lysine-capped gold

Notes

Figure 2. TGA data recorded from carefully weighed powders of purified lysine-capped gold nanoparticles (curve 1) and pure lysine (curve 2).

colloidal solution was stable for months with little evidence of further aggregation. What is interesting and particularly germane to this study is the fact that the spectrum recorded from the redispersed lysine-capped gold nanoparticle solution (curve 3) shows only slight broadening relative to that recorded from the as-capped gold colloidal solution (curve 2). This clearly shows that repeated centrifugation, washing, and finally drying of the lysinecapped solution and long-term storage of the powder thus obtained lead to a tolerable degree of aggregation of the gold nanoparticles. The above results show that it is indeed possible to stabilize gold nanoparticles in water by surface complexation with the amino acid lysine and also render them water-dispersible. While there are a few reports in the literature on the surface modification of gold nanoparticles with amino acids, notably that of Xiaoming and co-workers on surface-enhanced Raman spectroscopy (SERS) studies of lysine25 and that of Lieber et al. on the formation of gold nanowires on a polylysine template,26 the exact nature of interaction between amino acids and gold nanoparticles is not well understood. The strength of the interaction may conveniently be studied by thermogravimetric analysis of the lysine-gold nanoparticle conjugate powder. Figure 2 shows plots of TGA profiles recorded from carefully weighed purified powders of lysine-capped gold nanoparticles (curve 1) and pure lysine (curve 2). The lysine-capped gold nanoparticles (curve 1) display two weight losses in the temperature intervals 70-633 °C (25% weight loss) and 755-975 °C (an additional 30% weight loss), while pure lysine shows two sharp weight losses in the temperature intervals 255-365 °C (57% weight loss) and 478-589 °C (24% weight loss, curve 2). Assuming an area per lysine molecule of 10 Å2 and a gold particle diameter of 65 Å, the contribution of a monolayer of lysine on the gold nanoparticle surface to the overall weight of the lysine-gold nanoparticle conjugate material can be easily calculated to be 14%. That the overall weight loss in the TGA measurement of the lysine-gold nanoparticle material is much higher (curve 1, ca. 55%) than this number can be explained as follows. The monotonic weight loss (ca. 25%) in the temperature interval 70-633 °C is attributed to desorption of trapped water and decomposi(25) Xiaoming, D.; YoungMee, J.; Hiroshi, Y.; Shigeru, D.; Yukihiro, O. Appl. Spectrosc. 1999, 53, 133. (26) Yi, C.; Lauhon, L. J.; Gudiksen, M. S.; Jiangfang, W.; Lieber, C. M. Appl. Phys. Lett. 2001, 78, 2214.

Notes

Langmuir, Vol. 19, No. 8, 2003 3547

Scheme 1. Illustration of the Assembly of Redispersed Lysine-Capped Gold Nanoparticles in Water at Different pH Values

Figure 3. 1H NMR spectra recorded from lysine (curve 1) and lysine-capped gold nanoparticles redispersed in D2O (curve 2).

tion of lysine molecules hydrogen bonded with lysine molecules bound to the surface of gold nanoparticles (curve 1). The lysine-capped gold nanoparticle powder is extremely hygroscopic, and the samples were subjected to mild heating prior to TGA measurements. The complete decomposition of pure lysine occurs in the temperature interval 255-589 °C (curve 2) and is shifted to higher temperatures (755-975 °C) when the amino acid molecules are bound to the gold nanoparticle surface (curve 1). These results clearly indicate that the surface-bound lysine molecules are more stable than the amino acid molecules in the free form. Figure 3 shows the proton NMR spectra recorded from lysine (curve 1) and lysine-capped gold nanoparticles redispersed in D2O (curve 2). A number of peaks are seen in the spectra and are identified as follows. The peaks in the lysine NMR spectrum (curve 1) at 3.7 and 2.97 ppm are assigned to protons coordinated to the R-carbon of the amino acid and the carbon attached to the second amine group in the amino acid, respectively. The peaks at 2.17, 1.68, and 1.44 ppm are due to protons coordinated to the

carbons β, δ, and γ to the amino acid, respectively.27 A comparison of the 1H NMR spectra recorded from lysine (curve 1) and the lysine-capped gold solution (curve 2) shows significant differences in the region of the protons coordinated to the R-carbon of the amino acid and the carbon attached to the second amine group in the amino acid. It is observed that the peak at 3.7 ppm (protons coordinated to the R-carbon of the amino acid) is shifted downfield to 4 ppm due to binding with the gold nanoparticle surface while the peak at 2.97 ppm (protons coordinated to the carbon attached to the terminal amine group in the amino acid) is also shifted downfield to 3.55 pm and is accompanied by significant broadening. The broadening of the terminal NH2 group peak is most likely due to formation of hydrogen bonds with surface-bound lysine molecules of neighboring gold nanoparticles. Hydrogen bonds are known to broaden the signal since the lone pair electrons present in the oxygen and nitrogen atoms make the relaxation times much shorter. This result suggests that the binding of lysine to the gold nanoparticle surface occurs via the R-amine group in the amino acid while the terminal amine group forms hydrogen bonds with the carboxylic acid groups of surface-bound lysine molecules on neighboring gold nanoparticles. This is illustrated in Scheme 1 and agrees with the UV-vis measurements of the redispersed gold nanoparticle solution (Figure 1A) where evidence of cross-linking was inferred. The presence of metal centers in the sample can lead to significant broadening of the proton NMR spectra of groups in close proximity with the metal centers.20 What is interesting is that in earlier reports on 1H NMR studies of alkylamine-capped gold nanoparticles, it was observed that the protons coordinated to the R-carbon are broadened considerably and shifted upfield from 2.7 ppm to a broad range spanning 2-0.5 ppm.20 The behavior of lysine is (27) The Aldrich Library of C13 and H1 NMR Spectra, 1st ed.; Pouchert, C. J., Behnke, J., Eds.; Aldrich Chemical Co.: Milwaukee, WI, 1993; Vol. 1, p 890.

3548

Langmuir, Vol. 19, No. 8, 2003

Notes

Figure 4. Representative TEM micrographs of (A) sodium borohydride reduced gold nanoparticles and (B) lysine-capped gold nanoparticles synthesized as described in the text and (C) a histogram of particle size distribution from a measurement of the image shown in (B).

considerably different and indicates that the nature of binding of the amine groups in this molecule with the gold nanoparticle surface is rather different from that of alkylamines. We are currently addressing this issue by a systematic study of the proton NMR spectra of different amino acids bound to gold nanoparticles. The temperatures at which weight loss was observed in the TGA analysis of the lysine-capped gold nanoparticle powder were considerably higher than that observed in the alkylamine-capped gold nanoparticle powders20,22 and provide additional support for the NMR results presented above. Figure 1B shows the UV-vis spectra recorded from the redispersed lysine-capped gold nanoparticles as a function of pH of the hydrosol, curves 1, 2, and 3 in the figure corresponding to spectra recorded at pH 3, 7, and 10, respectively. In all the spectra, a strong absorption band at ca. 525 nm is observed that corresponds to excitation of surface plasmon vibrations in the lysine-capped gold nanoparticles. On comparison of curve 1 (pH 3) with curves 2 (pH 7) and 3 (pH 10), one observes a broadening of the surface plasmon resonance at the two higher pH values. The isoelectric point (pI) of pure lysine is 9.74.28 However, when lysine binds to gold nanoparticles through the R-amine group (as inferred from the NMR results presented earlier), its isoeIectric point is expected to change. Consequent to binding of one of the amine groups with the gold nanoparticle surface, the new isoelectric point will be the average of the pKa of the carboxylic acid group and the pKb of the terminal amine ( amine) group and, therefore, ca. 6.35. Below pH 6.35, the surface-bound lysine molecules exist in cationic form due to the formation of ammonium ions. The ammonium ions prevent formation of hydrogen bonds between neighboring gold nanoparticles as illustrated in Scheme 1. Above pH 6.35, surface-bound lysine molecules are negatively charged due to the formation of carboxylate ions which readily form hydrogen bonds with surface-bound amine groups of neighboring gold nanoparticles (Scheme 1). Hence, lysine-capped gold nanoparticles at pH 7 and 10 show broadening of the surface plasmon resonance in comparison with the gold nanoparticle solution at pH 3. That the broadening of the

surface plasmon resonance is due to hydrogen-bondinginduced aggregation of the gold nanoparticles is easily shown by heating the high-pH solution at 90 °C for 10 min (curve 4). It is observed that this spectrum is almost identical to that of the lysine-capped gold colloidal solution at pH 7, showing that particles may be peptized by this heating process as observed in our earlier work on cysteinecapped silver colloidal solutions.29 Parts A and B of Figure 4 show representative TEM images recorded from as-prepared borohydride-reduced gold nanoparticles and lysine-capped gold nanoparticles (redispersed in water at pH 7) deposited in the form of films onto carbon-coated TEM grids, respectively. Due to the fact that there is no stabilization of the gold nanoparticles in the borohydride-reduced nanoparticles, they aggregate to yield clusters in which the individual particles are difficult to distinguish (Figure 4A). On the other hand, capping of the gold nanoparticles with lysine stabilizes the particles and prevents their physical contact (Figure 4B). From Figure 4B, it is thus possible to get an idea of the particle size and monodispersity. Figure 4C shows a plot of the particle size distribution histogram for the image in Figure 4B. The solid line is a Gaussian fit to the histogram and yields an average particle size of 65 ( 7 Å. The gold nanoparticles are thus fairly monodisperse and adequately protected by the amino acid monolayer. Furthermore, it is observed that the lysine-capped gold nanoparticles assemble into a network with a very uniform separation between the nanoparticles. The average interparticle separation was estimated from Figure 4B to be 21 ( 4 Å, indicating the presence of a fairly thick coating of lysine molecules on the nanoparticle surface. Negligible sintering of the nanoparticles is observed, clearly indicating that the surface coating of lysine molecules stabilizes the particles in solution and in thin film form. It is clear that this layer of lysine also enables redispersion of the nanoparticles in water, which is not possible with uncapped borohydride-reduced gold nanoparticles. As mentioned earlier, UV-vis spectroscopy measurements of the lysine-capped gold nanoparticle solution indicated a pH-dependent association of the nanoparticles via hydrogen bond formation between amino acids on

(28) Neal, A. L. In Chemistry and Biochemistry: A Comprehensive Introduction; McGraw-Hill: New York, 1971; p 389.

(29) Mandal, S.; Gole, A.; Lala, N.; Gonnade, R.; Ganvir, V.; Sastry, M. Langmuir 2001, 17, 6262.

Notes

Langmuir, Vol. 19, No. 8, 2003 3549

Figure 5. Representative TEM micrographs of lysine-capped gold nanoparticles redispersed in water at pH 3 (A) and pH 10 (B).

neighboring particles (Figure 1B). Parts A and B of Figure 5 show representative TEM images recorded from lysinecapped gold nanoparticles redispersed in water at pH 3 and pH 10, respectively. At pH 10 (Figure 5B), it is clearly seen that the particles aggregate into large superstructures in which the individual gold particles are difficult to distinguish. On the other hand, the lysine-capped gold nanoparticles at pH 3 are well-separated from each other and form a two-dimensional network of hydrogen-bonded gold nanoparticles (Figure 5A. Thus, the TEM results of Figure 5 provide direct and unequivocal support to the UV-vis conclusions that the particles aggregate (reversibly) at pH 10 while they are fairly well-dispersed at pH 3 (Scheme 1). Conclusion To conclude, we have shown that capping gold nanoparticles with the amino acid lysine enables storage of the lysine-stabilized gold nanoparticles as a stable powder

that may be readily redispersed in water. The use of a biomolecule such as an amino acid for surface modification of gold nanoparticles and the resulting water-redispersible nanoparticles have important implications for the application of this methodology to the formation of other medically important bioconjugates and of novel reagents in reactions such as peptide bond formation, esterification, and so forth in aqueous media. Acknowledgment. PR.S. and S.M. thank the Council of Scientific and Industrial Research (CSIR) and the University Grants Commission (UGC), Government of India, respectively, for research fellowships. This work is partially funded by grants from the Indo-French Centre for the Promotion of Advanced Research (IFCPAR, New Delhi) and the Department of Science and Technology (DST), Government of India, and this funding is gratefully acknowledged. LA026906V