Size-Controlled Synthesis of Monodispersed Silver Nanoparticles

Sep 2, 2006 - NEt3 (58 mL, 400 mmol) was added by using a syringe, and then the reaction solution was stirred at 80 °C for 2 h. The surface of the wh...
0 downloads 8 Views 419KB Size
Download PDF
Langmuir 2006, 22, 8581-8586

8581

Size-Controlled Synthesis of Monodispersed Silver Nanoparticles Capped by Long-Chain Alkyl Carboxylates from Silver Carboxylate and Tertiary Amine Mari Yamamoto, Yukiyasu Kashiwagi, and Masami Nakamoto* Osaka Municipal Technical Research Institute, 6-50, 1-Chome, Morinomiya, Joto-ku, Osaka 536-8553, Japan ReceiVed January 3, 2006. In Final Form: August 2, 2006 Monodispersed silver nanoparticles capped by long-chain alkyl carboxylates were prepared by the reaction of silver carboxylate with tertiary amine at 80 °C for 2 h. This approach is a unique, size-controlled synthetic method for the large-scale preparation of silver nanoparticles. Long-chain alkyl carboxylate derived from a precursor acts as a stabilizer to avoid the aggregation of silver nanoparticles and to control particle size. In addition, amine plays an important role both as a reagent to form a thermally unstable, amine-coordinated intermediate, bis(amine)silver(I) carboxylate, and as a mild reducing agent for the intermediate to produce nanoparticles at a low temperature. The silver core and carboxylate-capping ligand of silver nanoparticles were characterized by various techniques such as transmission electron microscopy, optical absorption spectroscopy, powder X-ray diffraction, X-ray photoelectron spectroscopy, gas chromatograph mass spectroscopy, and thermogravimetric and differential thermal analysis. The diameter of the nanoparticles can be strongly influenced by the alkyl chain length and the structure of the carboxylate. The average diameters of the silver nanoparticles were controlled to less than 5 nm in the case of silver carboxylate with a single alkyl chain length of 13 or 17 carbon atoms. On the contrary, the average diameters of silver nanoparticles became large and polydisperse in the case of silver carboxylate with a chain length of 7 carbon atoms or a branched chain. In comparing triethylamine with trioctylamine, there was no obvious effect to regulate the size distribution of the nanoparticles because they could not function as a capping ligand of the nanoparticles due to their weak coordination to silver. In addition, the heat treatment of silver nanoparticles in solution rather than in the solid state was effective for the growth of particles while maintaining narrow size distributions.

Introduction The development of synthetic procedures for uniform nanometer-sized particles is essential for many advanced applications1 because the monodispersity of metal nanoparticles induces their precise size and shape-dependent optical,2 electronic,3 magnetic,4 and catalytic5 properties. From this point of view, a current research interest has been focused on the synthesis and properties of metal nanoparticles. The representative preparation method developed by Brust et al. is the two-phase reduction of HAuCl4 by a strong reducing agent, NaBH4, to produce alkanethiol-capped gold nanoparticles.6 Preparative methods * To whom correspondence should be addressed. Tel.: +81-6-69638089. Fax: +81-6-6963-8099. E-mail: [email protected]. (1) (a) Haes, A. J.; Van Duyne, R. P. J. Am. Chem. Soc. 2002, 124, 10596. (b) Malynych, S.; Chumanov, G. J. Am. Chem. Soc. 2003, 125, 2896. (c) Schmid, G.; Talapin, D. V.; Shevchenko, E. V. In Nanoparticles: From Theory to Applications; Schmid, G., Ed.; VCH: Weinheim, Germany, 2004; Chapter 4.2, pp 251-298. (2) (a) Templeton, A. C.; Pietron, J. J.; Murray, R. W.; Mulvaney, P. J. Phys. Chem. B 2000, 104, 564. (b) Jin, R.; Cao, Y.; Mirkin, C. A.; Kelley, K. L.; Schatz, G. C.; Zheng, J. G. Science 2001, 294, 1091. (c) Porel, S.; Singh, S.; Radhakrishnan, T. P. Chem. Commun. 2005, 2387. (3) (a) Chen, S.; Murray, R. W.; Feldberg, S. W. J. Phys. Chem. B 1998, 102, 9898. (b) Simon, U. In Nanoparticles: From Theory to Applications; Schmid, G., Ed.; VCH: Weinheim, Germany, 2004; Chapter 5.2, pp 322-362. (4) (a) Park, J., II; Cheon, J. J. Am. Chem. Soc. 2001, 123, 5743. (b) Hyeon, T.; Lee, S. S.; Park, J.; Chung, Y.; Na, H. B. J. Am. Chem. Soc. 2001, 1235, 12798. (c) Shevchenko, E. V.; Talapin, D. V.; Schnablegger, H.; Kornwski, A.; Festin, O.; Svedlindh, P.; Haase, M.; Weller, H. J. Am. Chem. Soc. 2003, 125, 9090. (d) Talapin, D. V.; Shevchenko, E. V.; Weller, H. In Nanoparticles: From Theory to Applications; Schmid, G., Ed.; VCH: Weinheim, Germany, 2004; Chapter 3.2.2, pp 199-230. (5) (a) Li, Y.; Boone, E.; El-Sayed, M. A. Langmuir, 2002, 18, 4921. (b) Narayanan, R.; El-Sayed, M. A. J. Am. Chem. Soc. 2003, 125, 8340. (c) Tamaru, M.; Fujihara, H. J. Am. Chem. Soc. 2003, 125, 15742. (d) Tsunoyama, H.; Sakurai, H.; Ichikuni, N.; Negishi, Y.; Tsukuda, T. Langmuir 2004, 20, 11293. (6) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801.

analogous to this route are now widely applied to synthesize silver nanoparticles. For example, silver nanoparticles were prepared by the chemical reduction of a homogeneous solution containing silver salt and a large amount of stabilizers7 such as reverse micelles,8 surfactants,9 dendrimers,10 alkanethiol,11 alkylamine,12 and carboxylic acid.13 Despite numerous reports, the synthetic methods based on chemical reduction still have difficulty in removing halides, byproducts, and excess stabilizers such as surfactants. In contrast to the salt reduction methods described above, we previously developed a controlled thermolysis of a specially designed metal complex with no use of solvent, stabilizer, or reducing agent.14-16 The ligands coordinated to the precursor complex can act as a stabilizer to avoid the aggregation of (7) Bradley, J. S.; Schmid, G. In Nanoparticles: From Theory to Applications; Schmid, G., Ed.; VCH: Weinheim, Germany, 2004; Chapter 3.2.1, pp 186-199. (8) Taleb, A.; Petit, C.; Pileni, M. P. Chem. Mater. 1997, 9, 950. (9) (a) Yi, K. C.; Ho´rvo¨lgyi, Z.; Fendler, J. H. J. Phys. Chem. B 1994, 98, 3872. (b) Liz-Marza´n, L. M.; Land-Tourin˜o, I. Langmuir 1996, 12, 3585. (10) (a) Esumi, K.; Suzuki, A.; Yamahira, A.; Torigoe, K. Langmuir 2000, 16, 2604. (b) Balogh, L.; Swanson, D. R.; Tomalia, D. A.; Hagnauer, G. L.; McManus, A. T. Nano Lett. 2001, 1, 18. (11) (a) Harfenist, S. A.; Wang, Z. L.; Alvarez, M. M.; Vezmar, I.; Whetten, R. L. J. Phys. Chem. B 1996, 100, 13904. (b) Sarathy, K. V.; Raina, G.; Yadav, R. T.; Kulkarni, G. U.; Rao, C. N. R. J. Phys. Chem. B 1997, 101, 9876. (c) Pileni, M. P. J. Phys. Chem. B 2001, 105, 3358. (12) Bunge, S. D.; Boyle, T. J.; Headley, T. J. Nano Lett. 2003, 3, 901. (13) (a) Wang, W.; Efrima, S.; Regev, O. Langmuir 1998, 14, 602. (b) Wang, W.; Chen, X.; Efrima, S. J. Phys. Chem. B 1999, 103, 7238. (c) Lin, X. Z.; Teng, X.; Yang, H. Langmuir 2003, 19, 10081. (14) (a) Nakamoto, M.; Yamamoto, M.; Fukusumi, M. Chem. Commun. 2002, 1622. (b) Nakamoto, M.; Kashiwagi, Y.; Yamamoto, M. Inorg. Chim. Acta 2005, 358, 4229. (15) Yamamoto, M.; Nakamoto, M. Chem. Lett. 2003, 32, 452. (16) Abe, K.; Hanada, T.; Yoshida, Y.; Tanigaki, N.; Takiguchi, H.; Nagasawa, H.; Nakamoto, M.; Yamaguchi, T.; Yase, K. Thin Solid Films 1998, 327-329, 524.

10.1021/la0600245 CCC: $33.50 © 2006 American Chemical Society Published on Web 09/02/2006

8582 Langmuir, Vol. 22, No. 20, 2006

nanoparticles through the thermolysis process. In practice, controlled thermolyses of the gold(I) complexes [C14H29N(CH3)3][Au(SC12H25)2]14 and [Au(C13H27COO)(PPh3)],15 under the conditions of 180 °C for 5 h, produced gold nanoparticles with an average diameter of 26 and 23 nm, respectively. In this way, the particle size was regulated by the stabilizing effect of the ligands derived from the precursors. Thermolysis of silver carboxylate, Ag(CnH2n+1COO), also produced carboxylatecapped silver nanoparticles with an average diameter of 5 nm under the conditions of 250 °C for 5 h.16 However, these methods require elevated temperatures (over 180 °C) because of the thermal stability of the precursor complexes. To produce silver nanoparticles under mild conditions, we have tried to produce a thermally unstable intermediate in situ, by using amine. It is also expected that amine acts as a mild reducing agent. As a result, we have succeeded in developing a preparative procedure for monodispersed silver nanoparticles capped by long-chain alkyl carboxylates. The silver nanoparticles were produced by the reaction of silver carboxylate with tertiary amine under the mild conditions of 80 °C for 2 h.17 It has recently been reported that silver benzoate18 and acetate19 were reduced by amine in an organic solvent such as toluene, which contains a stabilizer, and gave thiolate and amine-capped silver nanoparticles, respectively. On the other hand, our system has the great advantages of controlling the size and the size distribution of nanoparticles by long-chain alkyl carboxylates derived from a precursor, and producing monodispersed nanoparticles less than 5 nm in diameter. In addition, this method can apply to large-scale synthesis because the rapid aggregation of metal nanoparticles can be avoided by carboxylate ligands derived from a precursor, even under the conditions of high metal concentration. In this paper, we have described the influence of alkyl chain length and the structure of carboxylates and tertiary amines on particle size and the distribution of nanoparticles in order to elucidate the role of carboxylate and amine ligands. In addition, the heat treatment of long-chain alkyl carboxylatecapped silver nanoparticles in the solid state and in solution has been examined to control the particle size. Experimental Section Materials. All chemicals and solvents were reagent grade quality, obtained commercially, and used without further purification. Silver nitrate (AgNO3), n-octanoic acid (n-C7H15COOH), n-tetradecanoic acid (n-C13H27COOH), n-octadecanoic acid (n-C17H35COOH), 2-hexyldecanoic acid (n-C8H17CH(n-C6H13)COOH), cis-9-octadecanoic acid (n-C8H17CHdCH(CH2)7COOH), triethylamine (NEt3), trioctylamine (NOct3), acetone, toluene, and diethyl ether were obtained from Nakarai Tesque, Inc. Synthesis of Silver Carboxylates. Silver carboxylates Ag(nCnH2n+1COO) (n ) 7, 13, 17), Ag[n-C8H17CH(n-C6H13)COO], and Ag[n-C8H17CHdCH(CH2)7COO] were prepared as reported previously.16 Synthesis of Silver Nanoparticles. Silver tetradecanoate, Ag(n-C13H27COO) (6.70 g, 20 mmol), was placed in a 100 mL flask equipped with a magnetic stirrer. NEt3 (58 mL, 400 mmol) was added by using a syringe, and then the reaction solution was stirred at 80 °C for 2 h. The surface of the white silver tetradecanoate powder gradually turned to brown, and the insoluble precursor finally disappeared, producing a homogeneous solution of silver nanoparticles. The addition of acetone (20 mL) to the solution produced the precipitate, which was collected by filtration, washed with a small amount of acetone, and dried under vacuum. The silver nanoparticles (17) Yamamoto, M.; Nakamoto, M. J. Mater. Chem. 2003, 13, 2064. (18) Chaki, N. K.; Sudrik, S. G.; Sonawane, H. R.; Vijayamohanan, K. Chem. Commun. 2002, 76. (19) Hiramatsu, H.; Osterloh, F. E. Chem. Mater. 2004, 16, 2509.

Yamamoto et al. were isolated as a powder (2.45 g, yield 94% calculated from the metal content, 83%). The synthetic procedures of the other silver nanoparticles were the same as those described above. Synthesis of [Ag(n-C8H17NH2)2][n-C13H27COO]. n-Octylamine (3.3 mL, 20 mmol) was added to a suspension of silver tetradecanoate (335 mg, 1 mmol) in toluene (5 mL). The reaction mixture was stirred at room temperature until silver tetradecanoate was dissolved. The addition of diethyl ether (15 mL) produced white precipitates, which were collected by filtration and then washed several times with a small amount of diethyl ether and dried under vacuum to give a white powder (1.639 g, 59%). IR(KBr pellet): 1428 (ν˜ CdO), 1562 (ν˜ C-O) cm-1. Anal. Calcd for C30H65N2AgO2: C, 60.69; H, 11.04; N, 4.72. Found: C, 59.93; H, 11.33; N, 4.69. GC/MS: m/z 228 (n-C13H27CO2H), 129 (n-C8H17NH2). 1H NMR (300 MHz, CDCl3) δ 0.88 (t, 9H, J ) 7.0 Hz, CH3), 1.18-1.28 (m, 30H, CH2), 1.45-1.47 (m, 4H, CH2), 1.59-1.76 (m, 12H, CH2), 2.17-2.35 (m, 2H, CH2CO2), 2.72 (t, 4H, J ) 6.8 Hz, CH2NH2). Characterization. Elemental analysis was performed on a Carlo Erba Instruments EA 1110 CHNS-O analyzer. 1H NMR spectra were recorded on a JEOL JNM-AL300 instrument at 300 MHz using chloroform-d as the solvent and tetramethylsilane as the internal standard. Transmission electron microscopic images were obtained on a JEM-1200EX transmission electron microscope (TEM) operated at 100 kV. Samples for the TEM were dispersed in toluene by sonication and deposited on an amorphous carbon film-coated copper grid followed by natural evaporation at room temperature. The average diameter of nanoparticles, the standard deviation, and the particle size distribution were determined from more than 200 particles based on TEM images. The diameter of a nonspherical nanoparticle was determined by the average value of the longest and shortest diameter of each particle. UV-visible absorption spectra of nanoparticles redispersed in toluene were recorded on a Shimadzu UV-3150C spectrophotometer. The spectra were collected over the range of 300-800 nm. Powder X-ray diffractions (PXRDs) were measured using a Rigaku RINT 2500 diffractometer (Cu KR radiation) equipped with a monochromator operating at 40 kV and 50 mA. The gas chromatograph mass spectroscopy (GC/MS) spectra were taken using a pyrolizer PY-2020D for thermal extraction at 200 °C for 10 min and a Hewlett-Packard 6890 GC system equipped with an HP 5973 mass selective detector. X-ray photoelectron spectroscopy (XPS) measured on a Physical Electronics (PHI) model 5700 ESCA spectrometer with an Al monochromatic source (Al KR energy of 1486.6 eV) using BN as the internal standard and narrow scan photoelectron spectra were recorded for Ag(3d). The thermogravimetric and differential thermal analysis (TG/DTA) was carried out on a Seiko Instruments SSC/5200 thermal analyzer. Fourier transform infrared (FT-IR) spectra were measured with a PerkinElmer Spectrum GX I-RO spectrophotometer, using thin, transparent KBr pellets prepared by pressing a mechanically homogenized mixture of the dried sample with dehydrated KBr.

Results and Discussion Synthesis and Properties of Silver Nanoparticles 1(C13/ NEt3). Preparation of silver nanoparticles was carried out by a simple one-pot process. The reaction of silver tetradecanoate, Ag(n-C13H27COO), with NEt3 was carried out at 80 °C and maintained for 2 h. The surface of the white silver tetradecanoate powder gradually turned to dark-brown, and the amount of insoluble precursor decreased until it finally disappeared, resulting in a homogeneous solution of silver nanoparticles. At the end of the reaction, the nanoparticles were precipitated by adding acetone to the solution, collected by filtration, and dried under vacuum. The silver nanoparticles, abbreviated as 1(C13/NEt3), were isolated as a metallic deep blue powder in quantitative yield. Even after prolonged storage of the powder of the 1(C13/ NEt3) silver nanoparticles in air, 1(C13/NEt3) could be redispersed in toluene by ultrasonication. Furthermore, the redispersed solution showed no sign of aggregated precipitation for a week.

Synthesis of Carboxylate-Capped Ag Nanoparticles

Langmuir, Vol. 22, No. 20, 2006 8583

Figure 2. PXRD patterns of silver nanoparticles (a) 1(C13/NEt3) and (b) 2(C7/NEt3). (c) UV-vis absorption spectrum of 1(C13/NEt3).

Figure 1. TEM image (top panel) and the corresponding particle size distribution analysis (bottom panel) of 1(C13/NEt3) silver nanoparticles.

Figure 1 shows a typical TEM image of the 1(C13/NEt3) silver nanoparticles and the histogram of diameters. The TEM sample was made using a toluene suspension of the nanoparticles directly, without a size selection process. Only spherical nanoparticles were observed and assembled into hexagonally packed arrays, indicating their uniformity in size. The average diameter of the 1(C13/NEt3) silver nanoparticles was 4.4 ( 0.2 nm, with a narrow size distribution ranging from 3.8 to 5.0 nm. The crystal structure of the 1(C13/NEt3) Ag nanoparticles was examined by PXRD (Figure 2a). The diffraction peaks at 2θ ) 38.5, 45, 65, and 78° can be indexed to the (111), (200), (220), and (311) planes of face-centered cubic silver, respectively. The line broadening of the PXRD peaks (Figure 2a) compared with that of the 2(C7/ NEt3) Ag nanoparticles with a size of 15.4 ( 5.1 nm (Figure 2b, vide infra) was primarily due to the smaller particle size. The average particle size of 1(C13/NEt3) calculated by the Scherrer equation20 using the half-width of the intense (111) reflection was 3.1 nm, the value of which was relatively smaller than that obtained from the TEM image. Because the size obtained from PXRD indicates the average size of the single crystalline domain inside the nanoparticles, the silver nanoparticles may contain some crystalline lattice defects.4c Figure 2c shows UV-vis absorption spectrum of the redispersed silver nanoparticles, 1(C13/NEt3), in toluene. The peak maximum was centered at 430 nm, which corresponds to that of the as-prepared nanoparticles. However, the band appears to be broader with an asymmetric shape. This may be due to a partial, small aggregation of primary particles, because nanoparticles were precipitated out from the reaction mixture using acetone and redispersed into toluene by ultrasonication for the UV-vis measurement. Furthermore, the absorption maximum of this band shows a red-shift compared with the typical surface plasmon band of silver nanoparticles. For example, it is reported that cis-9-octadecanoate and trifluoroacetate-capped silver nanoparticles with a diameter of 7.3 nm have a typical plasmon band centered at 417 nm in hexane.13c It is noted that the absorption band strongly depends on the (20) (a) Scherrer, P. Go¨ttinger Nachrichten 2 1918, 98. (b) Klug, H. P.; Alexander, L. E. X-ray Diffraction Procedures, 2nd ed.; Wiley: New York, 1973.

density of free electrons, the scattering of oscillation electrons, and the dielectric properties.21 Thus, the position of the plasmon band is influenced by the refractive index of the surrounding solvent21d and by the donating/withdrawing properties of the capping ligand. The particle size also affects the plasmon band.21 For example, the alkanethiolate-capped silver nanoparticles with diameters of 3.8 and 2.3 nm show the peak maxima of the absorption bands at about 434 and 451 nm in hexane, respectively.21b Thus, the red-shift of the present silver nanoparticles may be attributed to differences in solvent, capping ligand, and particle size. In the XPS, the binding energies for the Ag 3d5/2 and Ag 3d3/2 peaks are 367.9 and 373.8 eV, respectively, which are in good agreement with the values of zerovalent silver (Ag 3d5/2: 367.9 eV; Ag 3d3/2: 373.9 eV).22 Thermogravimetry indicated that the 1(C13/NEt3) silver nanoparticles contain 83 wt % metal and 17 wt % capping ligand. The presence of the capping ligand was also evidenced by GC/ MS analysis, suggesting the existence of tetradecanoic acid (m/z 228), but NEt3 was not detected. Thus, only the tetradecanoate attaches to the surface of the silver core. The coverage ratio23 of capping ligands to silver atoms on the surface of 1(C13/NEt3) nanoparticles was estimated to be 34% (tetradecanoate/silver atom ) 1:2.9) from the results of TEM and TG, which agrees with that of alkanethiolate monolayers on planer hexagonally close-packed Au(111) surfaces (33%).24 On the basis of the TEM image, it was recorded that the nanoparticles formed a twodimensional superlattice with hexagonal packing, the average edge-to-edge distance (2.9 nm) was measured, and its value was approximately two times the length of tetradecanoate ligand (1.4 nm).16 This means that the silver core of a nanoparticle is surrounded by a monolayer of tetradecanoate. Reaction Route. We have already reported that the carboxylate-capped silver nanoparticles were obtained by the controlled thermolysis of silver carboxylate in the absence of amine,16 but at a temperature of 250 °C. On the contrary, the present study (21) (a) Alvarez, M. M.; Khoury, J. T.; Schaaff, T. G.; Shafigullin, M. N.; Vezmar, I.; Whetten, R. L. J. Phys. Chem. B 1997, 101, 3706. (b) Taleb, A.; Petit, C.; Pileni, M. P. Chem. Mater. 1997, 9, 950. (c) Petit, C.; Lixon, P.; Pileni, M. P. J. Phys. Chem. B 1993, 97, 12974. (d) Mulvaney, P. Langmuir 1996, 12, 788. (22) Wagner, C. D.; Riggs, W. M.; Davis, C. E.; Moulder, J. F. Handbook of X-ray Photoelectron Spectroscopy; Muilenberg, G. E., Ed.; Perkin-Elmer Corporation: Eden Praire, MN, 1979; p 122. (23) Terrill, R. H.; Postlethwaite, T. A.; Chen, C.; Poon, C.; Terzis, A.; Chen, A.; Hutchison, J. E.; Clark, M. R.; Wignall, G.; Londono, J. D.; Superfine, R.; Falvo, M.; Johnson, C. S., Jr.; Samulski, E. T.; Murray, R. W. J. Am. Chem. Soc. 1995, 117, 12537. (24) (a) Widrig, C. A.; Alves, C. A.; Porter, M. D. J. Am. Chem. Soc. 1991, 113, 2805. (b) Kim, Y.-T.; McCarley, R. L.; Bard, A. J. J. Phys. Chem. B 1992, 96, 7416.

8584 Langmuir, Vol. 22, No. 20, 2006

Yamamoto et al. Table 1. Analytical Data for Silver Nanoparticles

nanoparticles

silver carboxylate

amine

yield (%)

metal content (%)a

diameter (nm)b

1(C13/NEt3) 2(C7/NEt3) 3(C17/NEt3) 4(C13/NOct3) 5(C7/NOct3) 6(C17/NOct3) 7(br/NEt3) 8(ol/NEt3)

Ag(n-C13H27COO) Ag(n-C7H15COO) Ag(n-C17H35COO) Ag(n-C13H27COO) Ag(n-C7H15COO) Ag(n-C17H35COO) Ag[n-C8H17CH(n-C6H13)COO] Ag[n-C8H17CHdCH(CH2)7COO]

N(C2H5)3 N(C2H5)3 N(C2H5)3 N(n-C8H17)3 N(n-C8H17)3 N(n-C8H17)3 N(C2H5)3 N(C2H5)3

94 89 94 81 92 68 27 91

83 94 77 79 77 75 97 77

4.4 ( 0.2 15.4 ( 5.1 2.7 ( 0.3 3.6 ( 0.3 5.1 ( 1.7 4.4 ( 0.3 38.3 ( 7.1 3.7 ( 0.2

a

TG analysis showing total amount of metal. b TEM data.

has accomplished the preparation of silver nanoparticles at a considerably lower temperature of 80 °C through the controlled thermolysis of silver carboxylate in the presence of amine. Thus, amine plays an important role in the reaction process. Silver carboxylate does not dissolve in NEt3 at room temperature, but the reaction proceeded at 80 °C for 2 h, producing a homogeneous solution of silver nanoparticles. On the other hand, silver carboxylate was soluble in di-n-butylamine and n-octylamine at room temperature, and silver nanoparticles were produced at 100 °C (di-n-butylamine) and 130 °C (n-octylamine), respectively. These results suggest that intermediates involving both the silver carboxylate and the amine are formed, and their thermal stabilities become higher in the order of tertiary, secondary, and primary amine adducts. In addition, the reducing ability of amine is weaker in the order of tertiary, secondary, and primary amines. The isolation of the intermediate was tried by recrystallization from the reaction mixture of silver tetradecanoate and excess noctylamine in diethyl ether. As a result, bis(n-octylamine)silver(I) tetradecanoate ([Ag(n-C8H17NH2)2][C13H27CO2]; see Experimental Section) was isolated and confirmed by elemental analysis, GC/MS, FT-IR, and 1H NMR. It is noteworthy that the thermolysis of the isolated bis(n-octylamine)silver(I) tetradecanoate with n-octylamine at 130 °C for 4 h also produced silver nanoparticles (average diameter: 3.7 nm). Although tertiary amine adducts could not be isolated because of their thermal instability and the reducing power of NEt3, the corresponding intermediate may be formed in situ through the reaction of silver carboxylate with tertiary amine. Thus, silver carboxylate reacts with amine to form the intermediate, bis(amine)silver(I) carboxylate, followed by the thermolysis of the intermediate to produce silver nanoparticles. Particle Size Regulation by Carboxylate and Amine. To elucidate the role of carboxylate and amine in the formation of nanoparticles and their stabilization, the silver nanoparticles were systematically prepared using a series of silver carboxylates with different alkyl chain lengths and structures and different tertiary amines, such as NEt3 or NOct3 (Table 1). According to the preparative method of 1(C13/NEt3) silver nanoparticles, the silver nanoparticles 2(C7/NEt3), 3(C17/NEt3), 7(br/NEt3), and 8(ol/NEt3) were synthesized from Ag(n-CnH2n+1COO) (n ) 7, 17), Ag[n-C8H17CH(n-C6H13)COO], and Ag[nC8H17CHdCH(CH2)7COO] with NEt3, respectively. The silver nanoparticles, 4(C13/NOct3), 5(C7/NOct3), and 6(C17/NOct3) were also synthesized by the reaction of Ag(n-CnH2n+1COO) (n ) 13, 7, 17) with NOct3, respectively. The nanoparticles were obtained quantitatively, except for 7(br/NEt3). The low yield of 7(br/NEt3) may be attributed to the thermal stability of the silver carboxylate with a branched alkyl chain or its intermediate under these reaction conditions. Figure 3 shows TEM images and their corresponding particle size distributions of silver nanoparticles 1(C13/NEt3)-8(ol/NEt3). The average diameters of the nanoparticles were 4.4 ( 0.2 nm (1(C13/NEt3)), 15.4 ( 5.1 nm (2(C7/NEt3)), 2.7 ( 0.3 nm (3(C17/NEt3)), 3.6 ( 0.3 nm (4(C13/

NOct3)), 5.1 ( 1.7 nm (5(C7/NOct3)), 4.4 ( 0.3 nm (6(C17/ NOct3)), 38.3 ( 7.1 nm (7(br/NEt3)), and 3.7 ( 0.2 nm (8(ol/ NEt3)). Compared with the silver nanoparticles 1(C13/NEt3)-3(C17/ NEt3), the particle size of both 1(C13/NEt3) and 3(C17/NEt3) is smaller than 5 nm in diameter (Figure 3: 1(C13/NEt3) and 3(C17/NEt3)). On the other hand, the reaction of silver octanoate with NEt3 resulted in large and polydispersed nanoparticles (average diameter: 15.4 ( 5.1 nm; Figure 3: 2(C7/NEt3)). A similar tendency was observed in the case of the reaction with NOct3, as shown in Figure 3: 4(C13/NOct3)-6(C17/NOct3). Silver carboxylate with an unsaturated single alkyl chain, Ag[n-C8H17CHdCH(CH2)7COO], also produced small nanoparticles with a narrow distribution (average diameter: 3.7 ( 0.2 nm; Figure 3: 8(ol/NEt3)). Thus, it was confirmed that a short alkyl carboxylate (octanoate) showed insufficient size control to produce large and polydispersed nanoparticles (Figure 3: 2(C7/NEt3), 5(C7/NOct3)), while the long alkyl carboxylates (i.e., tetradecanoate, octadecanoate, and cis-9-octadecanoate) produced smaller nanoparticles with narrow size distributions. However, the average diameters of the nanoparticles with long alkyl carboxylates showed ambiguous size dependence on the difference in alkyl chain length (Figure 3: 1(C13/NEt3), 3(C17/ NEt3), 4(C13/NOct3), 6(C17/NOct3), and 8(ol/NEt3)). This result also corresponds to the previous results13b,16 indicating no correlation between the average diameter and the chain length of long saturated and unsaturated alkyl carboxylates (13, 15, 17, or 20 carbon atoms). The use of silver carboxylate with a branched alkyl chain, Ag[n-C8H17CH(n-C6H13)COO], as a precursor caused the aggregation of particles to generate large and polydispersed silver nanoparticles (average diameter: 38.3 ( 7.1 nm; Figure 3: 7(br/NEt3)). Thus, these observations suggest that a carboxylate ligand with a single long alkyl chain may be densely packed on the surface of nanoparticles through an alkane-alkane interaction25 and be able to control the particle size and the distribution more effectively than a short alkyl chain and a branched alkane. To elucidate the effect on the size regulation by tertiary amine, a series of silver nanoparticles, 1(C13/NEt3)-3(C17/NEt3), prepared with NEt3, and their corresponding nanoparticles 4(C13/NOct3)-6(C17/NOct3) using NOct3 were compared (Figure 3). In the case of silver carboxylate with an alkyl chain length of 13 or 17 carbon atoms, the amine with long alkyl chains, NOct3, had no apparent effect on the size reduction. However, comparing silver nanoparticles 2(C7/NEt3) (average diameter: 15.4 ( 5.1 nm) with 5(C7/NOct3) (average diameter: 5.1 ( 1.7 nm), the increase of the alkyl chain length of amine from C2 to C8 resulted in a smaller particle size. On the other hand, both of the values that make up the polydispersity, defined as the standard deviation/diameter ratio, were the same, 0.33. Therefore, (25) (a) Ulman, A. An Introduction to Ultrathin Organic Films; Academic Press: San Diego, CA, 1991. (b) Ulman, A. Chem. ReV. 1996, 96, 1533.

Synthesis of Carboxylate-Capped Ag Nanoparticles

Langmuir, Vol. 22, No. 20, 2006 8585

Figure 3. TEM images (top panel) and the corresponding particle size distribution analyses (bottom panel) of a series of silver nanoparticles: 1(C13/NEt3), 2(C7/NEt3), 3(C17/NEt3), 4(C13/NOct3), 5(C7/NOct3), 6(C17/NOct3), 7(br/NEt3), and 8(ol/NEt3).

NEt3 and NOct3 may act as a weak or noncoordinated solvent rather than a capping ligand in the growth process of silver nanoparticles. According to the reaction route, carboxylate-capped silver nanoparticles were produced by the controlled thermolysis of the intermediate formed in situ through the reaction of silver carboxylate with tertiary amine. Tertiary amine does not finally coordinate on the surface of silver nanoparticles because of its weak affinity and bulkiness. On the contrary, carboxylate acts as a ligand of a precursor and an intermediate, and functions as a capping ligand for final nanoparticles. Therefore, the particle size and the distribution could be strongly influenced by the alkyl chain length and the structure of the carboxylate. Particle Size Regulation by Heat-Treatment in the Solid State and in Solution. The diameters of nanoparticles could often be controlled by heat-treatment in the solid state26 and in solution27 to produce large nanoparticles with a narrow size distribution. With respect to the growth of the present silver nanoparticles in the solid state, the heat-treatment of 1(C13/NEt3)

silver nanoparticles with an average diameter of 4.4 ( 0.2 nm (Figure 4a) was conducted under atmospheric conditions in a furnace at 150 °C for 0.5 and 2 h. As a result, the average diameters of the heat-treated nanoparticles increased to 5.9 ( 0.6 nm and 5.7 ( 0.8 nm (Figure 4b), respectively. It is well-known that the heat-treatment process of alkanethiol-capped gold nanoparticles with tetraalkylammonium salt in the solid state proceeded through melting on the surface of smaller nanoparticles and coalescence.26 The growth of the present silver nanoparticles may also proceed the same way. However, the uniform growth of nanoparticles was not achieved in the solid state. To achieve the uniform growth of nanoparticles, the heattreatment of the nanoparticles was examined in solution.27 The diameter of the (C13/NEt3) silver nanoparticles produced by the (26) (a) Teranishi, T.; Hasegawa, S.; Shimizu, T.; Miyake, M. AdV. Mater. 2001, 13, 1699. (b) Shimizu, T.; Teranishi, T.; Hasegawa, S.; Miyake, M. J. Phys. Chem. B 2003, 107, 2719. (27) Maye, M. M.; Zheng, W.; Leibowitz, F. L.; Ly, N. K.; Zhong, C.-J. Langmuir 2000, 16, 490.

8586 Langmuir, Vol. 22, No. 20, 2006

Yamamoto et al.

the heat-treatment in solution is effective for the growth of nanoparticles while maintaining narrow size distribution.

Conclusions

Figure 4. TEM images of (a) 1(C13/NEt3) silver nanoparticles, (b) product obtained by the heat-treatment of 1(C13/NEt3) nanoparticles in the solid state at 150 °C for 2 h and (c) product obtained by the reaction of silver tetradecanoate (Ag(n-C13H27COO)) with NEt3 at 80 °C for the duration of 10 h.

reaction of silver tetradecanoate with NEt3 at 80 °C was measured by TEM. The particles grew to diameters of 4.4 ( 0.2, 5.8 ( 0.4, 5.7 ( 0.5, and 6.2 ( 0.6 nm (Figure 4c) for 2, 4, 6, and 10 h, respectively, while there was no apparent increase over 10 h up to 24 h. The growth of nanoparticles in solution proceeds through the desorption and readsorption of capping ligands and the coalescence of cores until the nanoparticle becomes thermodynamically stable.27 Thus, the present carboxylate-capped silver nanoparticles attained to the thermodynamically stable state by 10 h. Compared with the heat treatment in the solid state,

We have demonstrated the preparation of monodispersed silver nanoparticles capped by long-chain alkyl carboxylates through the reaction of silver carboxylate with tertiary amine. The longchain alkyl carboxylates derived from precursors act as capping ligands and control the size and size distribution of nanoparticles. In the synthetic procedure, tertiary amine reacts with silver carboxylate to form an intermediate, bis(amine)silver(I) carboxylate, and the subsequent thermolysis under mild conditions results in silver nanoparticles. The diameter of the nanoparticles can be strongly influenced by the alkyl chain length and the structure of the carboxylate because the carboxylate acts as a ligand of the precursor and the intermediate, and functions as a capping ligand for final nanoparticles. The average diameters of the silver nanoparticles were less than 5 nm in the case of a long alkyl chain length of 13 or 17 carbon atoms. On the contrary, silver carboxylate with a short alkyl chain (7 carbon atoms) or a branched one produces large and polydispersed nanoparticles. Compared with the carboxylate ligand, NEt3 and NOct3 have no influence on the size distribution of the nanoparticles because of their weak coordination to silver nanoparticles in the growth process. To achieve the uniform growth of nanoparticles, heat treatment in solution rather than in the solid state was effective. Compared with the usual procedures involving the reduction of silver salts by reducing agents in solutions, the present approach provides simple, convenient, size-controlled, and large-scale production of monodispersed silver nanoparticles, and is easily accessible to any other researcher. This approach could be applied to synthesize other precious metal nanoparticles. Finally, our investigations have suggested that monodispersed silver nanoparticles can also be applicable to self-assemble two-dimensional superlattices. LA0600245