Poly(allylamine)-Stabilized Colloidal Copper Nanoparticles: Synthesis

Feb 11, 2010 - For a more comprehensive list of citations to this article, users are .... Physica E: Low-dimensional Systems and Nanostructures 2016 7...
2 downloads 0 Views 2MB Size
Download PDF
pubs.acs.org/Langmuir © 2010 American Chemical Society

Poly(allylamine)-Stabilized Colloidal Copper Nanoparticles: Synthesis, Morphology, and Their Surface-Enhanced Raman Scattering Properties Yanfei Wang†,‡ and Tewodros Asefa*,†,‡ †

Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, 610 Taylor Road, Piscataway New Jersey 08854, and ‡Department of Chemical Engineering and Biochemical Engineering, Rutgers, The State University of New Jersey, 98 Brett Road, Piscataway, New Jersey 08854 Received November 4, 2009. Revised Manuscript Received January 24, 2010 Poly(allylamine)-stabilized spherical- and rod-shaped copper nanoparticles are synthesized by a simple chemical reaction. The synthesis is performed by the reduction of copper(II) salt with hydrazine in aqueous solution under atmospheric air in the presence of poly(allylamine) (PAAm) capping agent. Noteworthy of the advantages of the synthetic method includes its production of water dispersible copper nanoparticles at room temperature under no inert atmosphere, making the synthesis more environmentally friendly. The resulting copper nanoparticles are investigated by UV-vis spectroscopy and transmission electron microscopy (TEM). The results demonstrate that the amount of NaOH used is important for the formation of the copper nanoparticles while the reaction time and concentration of PAAm play key roles in controlling the size and shape of the nanoparticles, respectively. The resulting colloidal copper nanoparticles exhibit large surface-enhanced Raman scattering (SERS) signals.

1. Introduction Noble metal nanoparticles have attracted considerable attention owing to their potential applications in such fields as catalysis, biology, electronics, and information technology.1-4 In particular, copper nanoparticles are widely used as a catalyst for various chemical transformations including water-gas shift5 and gas detoxification6 reactions, and as an electrocatalyst in solid-oxide fuel cells.7 Another important application for copper nanomaterials is their potential use for surface-enhanced Raman scattering (SERS), whose enhancement factor is dependent on the strength of the local electric field and the nature of the material.8 Indeed, it has already been reported that noble metals such as Au, Ag, and Cu nanostructures generate large SERS. However, the investigation of Cu nanoparticles synthesized in solution phase as SERS substrate has remained largely unexplored. This is mainly because copper nanostructures are prone to fast oxidation and, consequently, are hard to synthesize chemically. This has inspired us to develop synthetic methods for stable colloidal Cu nanoparticles in aqueous solution and explore their SERS properties. *To whom correspondence should be addressed. Telephone: 732-445-2970. Fax: 732-445-5312. E-mail: [email protected]. (1) Xia, Y.; Xiong, Y.; Lim, B.; Skrabalak, S. E. Angew. Chem., Int. Ed. 2008, 47, 2. (2) Tao, A. R.; Habas, S.; Yang, P. Small 2008, 4, 310. (3) Daniel, M. C.; Astruc, D. Chem. Rev. 2004, 104, 293. (4) Rosi, N. L.; Mirkin, C. A. Chem. Rev. 2005, 105, 1547. (5) Ressler, T. B.; Kniep, L.; Kasatkin, I.; Schl€ogl, R. Angew. Chem., Int. Ed. 2005, 44, 4704. (6) Vukojevic, S.; Trapp, O.; Grunwaldt, J.; Kiener, C.; Schth, F. Angew. Chem., Int. Ed. 2005, 44, 7978. (7) Park, S.; Gorte, R. J.; Vohs, J. M. Appl. Catal., A 2000, 200, 55. (8) Bozzini, B.; D’Urzo, L.; Re, M.; De Riccardis, F. J. Appl. Electrochem. 2008, 38, 1561. (9) Wu, S.; Chen, D. J. Colloid Interface Sci. 2004, 273, 165. (10) Chang, Y.; Lye, M. L.; Zeng, H. C. Langmuir 2005, 21, 3746. (11) Huang, H.; Yan, F.; Kek, Y.; Chew, C.; Xu, G.; Ji, W.; Oh, P.; Tang, S. Langmuir 1997, 13, 172. (12) (a) Zhou, G.; Lu, M.; Yang, Z. Langmuir 2006, 22, 5900. (b) Liu, Z.; Yang, Y.; Liang, J.; Hu, Z.; Li, S.; Peng, S.; Qian, Y. J. Phys. Chem. B 2003, 107, 12658. (13) Gole, A.; Murphy, C. J. Chem. Mater. 2004, 16, 3633. (14) Sau, T. K.; Murphy, C. J. J. Am. Chem. Soc. 2004, 126, 8648. (15) Pileni, M. P. Nat. Mater. 2003, 2, 145.

Langmuir 2010, 26(10), 7469–7474

Recently, a few different types of synthetic methods for making shaped copper nanoparticles including rods or cubes have been reported.9-19 The first and most commonly used method involves the chemical reduction of copper ions in aqueous solution in the presence of various capping agents. For instance, using surfactant capping agents such as cetyltrimethylammonium bromide (CTAB),8,9 ethylenediamine (EDA),10 polyvinylpyrrolidone (PVP),11 and sodium dodecylbenzene sulphonate (DBS),12 various shaped copper nanoparticles were produced by reduction of copper(II) ions. As in the synthesis of many other metal nanomaterials, adjustment of the synthetic parameters, in particular the type of the surfactant, the preparation of the seed solution,13 temperatures, and the ratios of additives in the growth solution14 render obvious shape evolution of the copper nanoparticles. Pileni and her research group demonstrated the size and shape controlled synthesis of copper nanoparticles by using mixed reverse micelles as a template.15 Furthermore, copper nanoparticles, nanorods, and nanodisks with different size and shape were synthesized by varying the water content, the concentration of reducing agent, or the salt concentrations in AOT-water-oil reverse micellar systems.16-19 In a second commonly used method, copper nanoparticles were prepared in organic solvents by decomposing or reducing various copper complexes in the presence of capping agents. For instance, copper nanorods and nanocubes were synthesized by treating Cu(acac)2 in an octyl ether solvent in the presence of oleic acid and oleyl amine capping agents.20 In this case, the shapes of the nanoparticles were tuned by changing the reaction temperature. In the third synthetic method known as the polyol reduction process, ethylene glycol was used both as a solvent and as a reductant for copper ions while PVP was used as a capping agent for (16) Lisiecki, I.; Pileni, M. P. J. Am. Chem. Soc. 1993, 115, 3887. (17) Salzemann, C.; Urban, J.; Lisiecki, I.; Pileni, M. P. Adv. Funct. Mater. 2005, 15, 1277. (18) Cason, J. P.; Roberts, C. B. J. Phys. Chem. B 2000, 104, 1217. (19) Lisieki, I.; Pileni, M. P. J. Phys. Chem. 1995, 99, 5077. (20) Galkowski, M. J.; Wang, L.; Luo, J.; Zhong, C. Langmuir 2007, 23, 5740.

Published on Web 02/11/2010

DOI: 10.1021/la904199f

7469

Article

Wang and Asefa

making copper nanoparticles.21 Finally, physical vapor deposition,22,23 chemical vapor deposition,24 γ-irradiation,25 UV-light irradiation,26 sonochemical method,27 and electrochemical approach28-30 have been reported as synthetic methods for the preparation of copper nanostructures. In general, most of the methods reported so far have rarely resulted in copper nanoparticles having uniform sizes below 50 nm or in high yield. Furthermore, the procedures used in previously reported methods are rather complicated, usually involving a number of steps or requiring an inert gas such as Ar or N2 in order to avoid oxidation of the Cu nanoparticles. Herein, we report on a one-step synthetic method to colloidal Cu nanoparticles by reducing copper ions with hydrazine using poly(allylamine) (PAAm) as a stabilizing agent. The advantages of the method include the use of atmospheric air (or no use of inert atmosphere) for the synthesis of the nanoparticles and its production of copper nanoparticles with reasonably uniform size and relatively higher yield. While changing the amount of NaOH used in the synthesis was found to help us to produce pure copper nanoparticles as opposed to copper oxide, varying the reaction time and concentration of PAAm enabled us to control the size and shape of the nanoparticles. Investigation of the SERS property of the resulting colloidal copper nanoparticles showed that the nanoparticles gave a 104 times SERS enhancement for 4-mercaptopyridine (4-Mpy) adsorbed on the nanoparticles compared to bulk 4-Mpy.

2. Experimental Section 2.1. Materials and Reagents. Copper(II) sulfate, hydrazine hydrate solution (78-82%), poly(allylamine) solution (PAAm) (average Mw ∼ 17 000, 20 wt % in H2O), and 4-mercaptopyridine (4-Mpy) (95%) were obtained from Sigma-Aldrich. Sodium hydroxide (98.8%) was purchased from Fisher Scientific. All chemicals were used as received without further treatment. 2.2. Synthesis of Copper Nanoparticles. In a typical synthesis of copper nanoparticles, CuSO4 (0.05 mmol) and various amounts (0.2-0.3 mL) of PAAm were completely dissolved in Millipore H2O (10 mL) under vigorous stirring at 60 °C for 20 min, forming a transparent light-blue solution. Then, 0.6-0.8 mL of NaOH (0.5 M) was added dropwise into the above solution. After stirring for 20 min, 1.0 mmol N2H4 3 H2O solution was dropped into the above solution under constant stirring. The reactor was kept in a water bath at 60 °C for 40-90 min. It should be pointed out that N2H4 3 H2O can also increase the pH of the solution. The reaction was monitored by UV-vis spectroscopy until no change of the absorbance spectrum was observed. The metallic copper nanoparticles in this work were obtained from the redox reaction between Cu2þ and hydrazine in basic solution in the presence of PAAm capping agent. 2.3. Characterization. The UV-vis absorbance spectra were measured with a Lambda 950 spectrophotometer (PerkinElmer). Transmission electron microscopy (TEM) samples were prepared by casting a drop of the as-prepared copper (21) Cha, S. I.; Mo, C. B.; Kim, K. T.; Jeong, Y. J.; Hong, S. H. J. Mater. Res. 2006, 21, 2371. (22) Wang, J.; Huang, H.; Kesapragada, S. V.; Gall, D. Nano Lett. 2005, 5, 2505. (23) Vitulli, G.; Bernini, M.; Bertozzi, S.; Pitzalis, E.; Salvadori, P.; Coluccia, S.; Martra, G. Chem. Mater. 2002, 14, 1183. (24) (a) Wang, J.; Yang, T.; Wu, W.; Chen, L.; Chen, C.; Chu, C. Nanotechnology 2006, 17, 719. (b) Kim, C.; Gu, W.; Briceno, M.; Robertson, I. M.; Choi, H.; Kim, K. Adv. Mater. 2008, 20, 1859. (c) Choi, H.; Park, S. H. J. Am. Chem. Soc. 2004, 126, 6248. (25) Henglein, A. J. Phys. Chem. B 2000, 104, 1206. (26) Kapoor, S.; Palit, D. K.; Mukherjee, T. Chem. Phys. Lett. 2002, 355, 383. (27) Vijaya Kumar, R.; Mastai, Y.; Diamanta, Y.; Gedanken, A. J. Mater. Chem. 2001, 11, 1209. (28) Bozzini, B.; D’Urzo, L.; Mele, C. Trans. IMF 2008, 86, 267. (29) Huang, L.; Lee, E.; Kim, K. Colloids Surf., A 2005, 262, 125. (30) (a) Wang, T.; Hu, J.; Yang, W.; Zhang, H. Electrochem. Commun. 2008, 10, 814. (b) Zhang, D. W.; Chen, C. H.; Zhang, J.; Ren, F. Chem. Mater. 2005, 17, 5242– 5245.

7470 DOI: 10.1021/la904199f

Figure 1. (A) UV-vis absorbance spectra of PAAm-capped copper nanoparticles that were synthesized in aqueous solution with 0.22 mL of PAAm and different amounts (or moles) of NaOH. (B) Absorption maxima of the copper nanoparticles versus amount (or moles) of NaOH used. nanoparticle suspension on a carbon-coated copper grid and then drying them in air. The dried grid was then placed under a JEOL 1200 EX transmission electron microscope. The images were taken at an acceleration voltage of 120 kV. Surface-enhanced Raman scattering (SERS) spectra were obtained on a microRaman instrument (Renishaw 1000, Gloucestershire, U.K.) equipped with a He/Ne laser (632.8 nm) and a CCD detector operated at room temperature. The laser power at the sample position was typically 2.0 mW for SERS spectra and normal Raman spectra.

3. Results and Discussion 3.1. Synthesis of Copper Nanoparticles and Formation of Copper versus Copper Oxide Nanoparticles in Aqueous Solution. As in the synthesis of many types of colloidal nanomaterials, copper nanoparticles also require organic ligands to prevent them from irreversible aggregation in solution. Here, we used PAAm to prepare and stabilize small Cu nanoparticles. Besides providing long-term stability to the nanoparticles by preventing particle agglomeration, polymer capping agents such as PAAm make the particles dispersible in aqueous solution.31 In our study, we found that several factors, including the amount of NaOH solution, concentration of PAAm, and reaction time, affect the composition, size, morphology, and degree of agglomeration of the resulting copper nanoparticles. Figure 1 shows the UV-vis absorption spectra for PAAmcapped copper nanoparticles and their absorption maxima versus molar concentration of NaOH used. Figure 1a displays the UV-vis spectra of colloidal Cu nanoparticles recorded 30 min after addition of hydrazine into Cu2þ solution in various amounts of 0.5 M NaOH in the presence of 0.22 mL of PAAm. The results (31) Sardar, R.; Park, J. W.; Shumaker-Parry, J. S. Langmuir 2007, 23, 11883.

Langmuir 2010, 26(10), 7469–7474

Wang and Asefa

Article

Scheme 1. Schematic Illustration of the Procedure Used for the Synthesis of PAAm-Capped Colloidal Cu Nanoparticles

showed that the intensity and position of the absorption band changed with the amount of NaOH used. When the NaOH amount was between 0.6 and 0.75 mL, the absorbance maximum was between 560 and 568 nm, a characteristic plasmon absorption band for copper nanoparticles.32 The intensity of the plasmon peak reached its maximum when the NaOH amount was increased to 0.7 mL. This corresponds to the formation of the highest possible yield of copper nanoparticles. When the amount of NaOH was adjusted above 0.8 mL, a peak at around 575 nm, whose intensity became stronger with increasing amount of NaOH solution, started to appear. Since a peak at g575 nm was reported to correspond to metallic copper nanoparticles that are surrounded by a copper(I) oxide shell,33 our result here indicated that the synthesis yields a mixture of Cu and Cu2O particles or Cu2O-coated Cu nanoparticles when >0.8 mL of 0.5 M NaOH solution (or 0.4 mmol OH-) was used. The study demonstrated that water-soluble and polymer-coated pure copper nanoparticles in the presence of PAAm capping ligand could be synthesized by using an appropriate amount, 0.60-0.75 mL of 0.5 M NaOH (or 0.30-0.38 mmol OH-). These results also clearly implied that adjusting the molar concentration of NaOH in the reaction was crucial for the synthesis to result in either pure copper or copper-oxide-containing copper nanoparticles and to produce the highest possible yield of copper nanoparticles. 3.2. Growth of PAAm-Stabilized Copper Nanoparticles in Aqueous Solution. The growth of colloidal Cu nanoparticles during the reaction went through several steps as observed from a set of color changes, from light yellow to wine red (Scheme 1). A brown colored solution was first formed instantly as soon as an aqueous NaOH was added into a mixture of CuSO4 and PAAm. Further addition of hydrazine into this reaction mixture produced a yellow solution containing Cu2O seed particles within 30 min. The presence of Cu2O nanoparticles was proved by UV-vis absorption spectroscopy (please see below). These Cu2O nanoparticles were further reduced by excess N2H4 3 H2O and yielded a red colored solution containing colloidal copper nanoparticles within 35 min of reaction. The color of the solution turned to deep red after stirring the reaction mixture at 60 °C for ∼90 min, indicating the growth of the colloidal Cu nanoparticles. The intermediate products during the growth process of the nanoparticles were characterized at intervals of 10 min by UV-vis absorption spectroscopy. Figure 2 displays the UV-vis absorption spectra that were taken at different stages of the continuous transformation of copper ions into PAAm-coated copper nanoparticles. After 30 min of reaction, a yellow solution, which displayed a very weak absorption peak at 578 nm corresponding to the plasmon resonance of Cu2O nanoparticles, was observed (Figure 2). However, after 40 min of overall (32) Hambrock, J.; Becker, R.; Birkner, A.; Weiss, J.; Fischer, R. A. Chem. Commun. 2002, 68. (33) Yanase, A.; Komiyama, H. Surf. Sci. 1991, 248, 11.

Langmuir 2010, 26(10), 7469–7474

Figure 2. UV-vis spectral evolution during the formation of PAAm-capped colloidal Cu nanoparticles from the reduction of Cu2þ ions with hydrazine in water at 60 °C in the presence of 0.22 mL of PAAm.

reaction time, the solution turned red and showed a well-defined absorbance band at ∼560 nm corresponding to the plasmon resonance of Cu nanoparticles. This clearly indicated that the Cu nanoparticles began to form between 30 and 40 min of reaction time. Between 40 and 90 min reaction time, the absorption spectra exhibited sharp plasmon resonance bands, which were slightly red-shifted from 560 to 568 nm, suggesting the growth of nanoparticle size. Furthermore, the intensity of the absorption peak continued to increase upon increasing the reaction time up to 90 min. However, after 90 min, no further evolution of the absorption band or growth of the nanoparticles was observed, indicating the reduction of copper ions came to completion. These results suggest that size controlled synthesis of copper nanoparticles with narrow size distribution would be possible with this approach. TEM images further confirmed that the nucleation and growth of copper nanoparticles progressed with the reaction time. Figure 3 shows the TEM images of the copper nanoparticles at different reaction times after injecting hydrazine into the Cu2þ/ PAAm solution. As described above, nucleation of the nanoparticles began at ∼30 min after injection of hydrazine, as indicated by the sudden change of the solution color from brown to yellow. As the reaction proceeded for an additional 5 min, the solution color rapidly changed from yellow to red. Samples collected right after 35 min of reaction time showed an average size of 40 nm copper nanoparticles (Figure 3a). The rate of further growth of the size of the nanoparticles became slower as evidenced by the stable red color of the solution. This was further confirmed by UV-vis absorption and TEM. For instance, as shown in Figure 3b, the average size of the nanoparticles after 60 min reaction time became only 50 nm. Nevertheless, this method was still proven to produce copper nanoparticles with variable average sizes, at least between 40 to 50 nm, simply by changing the reaction time. For the copper nanoparticles synthesized with 0.22 mL of PAAm after 60 min of reaction time, the pH of the solution was ∼10.8 and the absorption maximum was at 561 nm. When the pH was increased above 12.0, the UV-vis absorption remained to have only one absorption peak but the absorption maxima shifted from 561 to 578 nm. This result suggests that metallic copper nanoparticles were oxidized to Cu2O or a mixture of Cu and Cu2O nanoparticles at higher pHs. This is corroborated by previous reports which indicate that an absorption maximum at g575 nm corresponds to metallic copper nanoparticles that are surrounded by a copper(I) oxide shell.33 Although the copper nanoparticles underwent oxidation, it is interesting to note that their shape DOI: 10.1021/la904199f

7471

Article

Wang and Asefa

Figure 3. TEM images of PAAm-capped copper nanoparticles synthesized with 0.22 mL of PAAm after (a) 35 min and (b) 60 min of reaction time.

Figure 4. TEM image of PAAm-capped copper nanorods synthesized with 0.27 mL of PAAm.

remained almost unchanged and the size appeared slightly more polydisperse after increasing the pH to 12.5 (Supporting Information Figure S2). 3.3. Concentration of PAAm and Synthesis of Copper Nanospheres or Nanorods. It has been widely reported that the size and shape of noble metal nanoparticles can be controlled by changing the types and concentration of capping agent.10,15,34-36 Here also, as the TEM image in Figure 4 shows, nanorods mixed with nanoparticles were formed with our synthetic method, by increasing the relative concentration of PAAm in the solution compared to the one used above. The UV-vis absorption, however, showed barely any shift on the absorption maxima (Supporting Information Figure S3). The significant change in the absorption maxima was observed only when the metal copper nanoparticles were oxidized into Cu2O nanoparticles. The slight increase of PAAm concentration was favorable for the formation of Cu nanorods. This proves that the use of a different amount of PAAm leads to nanoparticles with significantly different morphologies. However, our attempt, by further increasing the PAAm concentration, to enlarge the aspect ratio of the nanorods (34) Johnson, C. J.; Dujardin, E.; Davis, S. A.; Murphy, C. J.; Mann, S. J. Mater. Chem. 2002, 12, 1765. (35) Gai, P. L.; Harmer, M. A. Nano Lett. 2002, 2, 771. (36) (a) Sun, Y.; Mayers, B.; Herricks, T.; Xia, Y. Nano Lett. 2003, 3, 955. (b) Sun, Y.; Gates, B.; Mayers, B.; Xia, Y. Nano Lett. 2002, 2, 165. (c) Sun, Y.; Xia, Y. Adv. Mater. 2002, 14, 833. (d) Sun, Y.; Yin, Y.; Mayers, B. T.; Herricks, T.; Xia, Y. Chem. Mater. 2002, 14, 4736.

7472 DOI: 10.1021/la904199f

or increase their yield with respect to the nanoparticles did not prove to be successful. It is worth noting that Cu nanorods have been demonstrated to form in AOT-water-oil systems by changing the water-to-oil or salt concentration of the reaction.15 In another case, capping agent EDA concentration played an important role in determining Cu nanowire and disklike morphologies. An appropriate amount of EDA resulted in the formation of Cu nanowires. When EDA was overused, however, the axial 1D growth of nanowires could be switched totally to disklike morphologies.10 However, this growth mechanism of such Cu nanowires dependent on the concentration of capping agents has not yet been well explored. A number of capping reagents have been examined to control the growth rates of metal surfaces kinetically and thus achieve 1D growth.34-36 For instance, uniform gold nanorods with controllable aspect ratios are prepared by using CTAB as the capping reagent. In general, the 1D nanostructures synthesized with CTAB as the capping reagent are twinned in crystal structure. Johnson et al.34 and Gai and Harmer35 have proposed a mechanism in which gold nanorods were assumed to evolve from multiply twinned particles (MTPs) with a decahedral shape. In another example, silver nanorods and nanowires are grown with PVP as capping reagent.36 It has been believed that each silver nanowire or nanorod evolved from a MTP of silver with the assistance of PVP at the initial stage of the Ostwald ripening process. The MTPs of silver consist of 10 {111} facets. As with the singly twinned seeds, anisotropic growth of decahedral Ag seeds can be induced to facilitate the formation of 1D nanocrystals such as nanorods and nanowires. The anisotropic growth was maintained by selectively covering the {100} facets with PVP while leaving the {111} facets largely uncovered by PVP and thus highly reactive. In addition to MTPs, another important factor for the formation of 1D nanorods is selective binding of capping agent to the different crystallographic planes of the metal. In general, the presence of a capping agent can change the order of free energies for different crystallographic planes and thus their relative growth rates. The plane with a lower addition rate will be exposed more on the nanocrystal surface. For example, PVP is a polymeric capping agent whose oxygen atoms bind most strongly to the {100} facets of Ag.36a The role of PAAm on the formation of Cu nanorods was similar to that of CTAB on the formation of gold nanorods34,35 and PVP on the formation of silver nanowires.36a Thus, the absence of Cu nanorods synthesized with 0.27 mL of PAAm Langmuir 2010, 26(10), 7469–7474

Wang and Asefa

Article Table 1. Raman Shifts (cm-1) and Assignments for 4-Mpya NR (cm-1)

SERS (cm-1)

assignment

EF

718 710 β(CC)/ν(C-S) 2.0  103 1003 1006 ring breathing 1.2  104 1120 1098 ring breathing/C-S 3.0  104 1214 1222 β(CH)/δ(NH) 5.1  104 1475 1477 ν(CdC/CdN) 1.0  104 a NR peaks are from 0.1 M 4-Mpy and SERS peaks are from 1  10-5 M of 4-Mpy on colloidal Cu nanoparticles synthesized from 0.22 mL of PAAm after 40 min reaction time. Assignments were obtained from refs 42-44.

Figure 5. (a) Normal Raman (NR) spectra of neat 4-Mpy with 0.1 M concentration. SERS spectra for 1  10-5 M 4-Mpy that are adsorbed on colloidal copper nanoparticles synthesized from 0.22 mL of PAAm after (b) 40 min, (c) 60 min, and (d) 90 min reaction time.

in the final product could be attributed to two possibilities: (i) PAAm with this concentration was favorable for the formation of MTPs and (ii) PAAm might bind more strongly to the {100} facets of Cu nanorods. This preferential capping can drive the addition of Cu atoms to the other crystal facets to form Cu nanorods when seeds are enough. 3.4. Effect of Temperature on the Synthesis of Copper Nanospheres/Nanorods. In addition to the synthesis at 60 °C above, experiments at lower and higher temperatures to determine the effect of temperature on the synthesis and structure of copper nanomaterials were conducted. The results suggested that the reaction at room temperature (23 °C) could also lead to copper nanomaterials. However, the yield of the copper nanoparticles was very low at 23 °C when 0.22 mL of PAAm was used. On the other hand, when 0.27 mL of PAAm was used, the synthesis did not form nanorods at 23 °C as it did at 60 °C. This is probably because the relatively lower reaction temperature could not provide enough energy required for the activation of specific faces of the copper nanoparticles for anisotropic growth into nanorods. For the synthesis at a higher temperature of 80 °C, the reaction took place quickly and the nucleation appeared to happen in 5 min. However, the size distribution of the resulting copper nanoparticles was rather more polydisperse compared to those synthesized at lower temperature. So, we used an optimized reaction temperature of between 50 and 60 °C in order to obtain more uniformly sized Cu nanoparticles in a higher yield. 3.5. SERS Activity. Noble metal nanostructures have been demonstrated to be effective SERS-active substrates. According to the electromagnetic theory of SERS, SERS enhancements depend on the excitation of the localized surface plasmon resonance, which is influenced by several significant parameters, such as the size, shape, and nature of aggregates of the nanomaterials.37-39 Therefore, it is important to develop monodisperse size metal nanoparticles in order to control or optimize the factors influencing the localized surface plasmon resonance and subsequently maximize their SERS signals. The colloidal Cu nanoparticles we have synthesized could provide an ideal substrate for SERS and to study the compositional dependence of chemical adsorption and reactions on the surface of Cu nanoparticles. Here we studied and (37) Bao, F.; Li, J.; Ren, B.; Yao, J.; Gu, R.; Tian, Z. J. Phys. Chem. C 2008, 112, 345. (38) Hubenthal, F.; Sanchez, D. B.; Borg, N.; Schmidt, H.; Kronfeldt, H. D.; Tr€ager, F. Appl. Phys. B: Lasers Opt. 2009, 95, 351. (39) Haynes, C. L.; McFarland, A. D.; Van Duyne, R. P. Anal. Chem. 2005, 77 (17), 338A.

Langmuir 2010, 26(10), 7469–7474

evaluated the SERS activity of our as-prepared Cu nanoparticles using 4-Mpy as the SERS reporter molecule (Figure 5). Figure 5a demonstrates a normal Raman (NR) spectrum of neat 4-Mpy. Figure 5b shows a typical SERS spectrum of 4-Mpy adsorbed on colloidal Cu nanoparticles. The SERS spectra consist of several observable bands at 1594, 1477, 1222, 1098, 1006, and 710 cm-1, which are intrinsic to 4-Mpy. Detailed peak frequency assignments are given in Table 1. The peak at 1594 cm-1 can be attributed to the ring stretch mode of the 4-Mpy molecule with deprotonated nitrogen. A previous SERS study of 4-Mpy on a silver substrate at different pH values40 showed that this peak at 1594 cm-1 was particularly sensitive to the environment of the molecule. Generally, 4-Mpy on Ag substrate gives a peak at 1623 cm-1 at pH < 1.00. However, when the pH is raised above 1.00, a new band near 1580 cm-1 also appears and increases in intensity with pH. At pH = 12.00, the peak at 1623 cm-1 disappears completely and the peak at 1580 cm-1 becomes very strong. In our study of the SERS spectra of 4-Mpy with colloidal Cu nanoparticles, the pH value was set above 12.00 as the nanoparticles are not stable or oxidize below that pH. Thus, we observed only a single peak at 1594 cm-1. However, a few remarkable spectral changes were observed upon adsorption of 4-Mpy on the colloidal Cu nanoparticles. For example, as shown in Figure 5b-d, a marked downshift of the ν(C-S) mode at 710 cm-1 and a dramatic increase in intensity of the ν(C-S) mode at 1006 cm-1 in comparison to the Raman spectrum of bulk 4-Mpy were exhibited. A similar downward shift and enhancement have been observed for 4-Mpy adsorbed on other metal substrates such as Au,41 Ag,42-45 and Pt,46 which has been interpreted to be due to the coordination of 4-Mpy with the metal surface through its sulfur atom. This further suggests that 4-Mpy is chemisorbed on the colloidal Cu nanoparticles also via its S atom. To evaluate the SERS activity of as-prepared Cu nanoparticles, it is useful to obtain the enhancement factor (EF). The EF can be calculated according to the following expression (eq 1):47 EF ¼ ðI SERS =I Raman ÞðN bulk =N ads Þ

ð1Þ

where ISERS and IRaman are the intensity of a vibrational mode in the SERS spectrum and normal Raman spectrum, respectively. Nbulk is the number of 4-Mpy molecules in the bulk, and Nads is (40) Hu, J. W.; Zhao, B.; Xu, W. Q.; Li, B. F.; Fan, Y. G. Spectrochim. Acta, Part A 2002, 58, 2827. (41) Yu, H. Z.; Xia, N.; Liu, Z. F. Anal. Chem. 1999, 71, 1354. (42) Baldwin, J.; Sch€uhler, N.; Butler, I. S.; Andrews, M. P. Langmuir 1996, 12, 6389. (43) Wang, Z. J.; Rothberg, L. J. J. Phys. Chem. B 2005, 109, 3387. (44) Baldwin, J. A.; Vlckova, B.; Andrews, M. P.; Butler, I. S. Langmuir 1997, 13, 3744. (45) Hu, J. W.; Zhao, B.; Xu, W. Q.; Fan, Y. G.; Li, B. F.; Ozaki, Y. J. Phys. Chem. B 2002, 106, 6500. (46) Bryant, M. A.; Joa, S. L.; Pemberton, J. E. Langmuir 1992, 8, 753. (47) C-ulha, M.; Kahraman, M.; Tokman, N.; T€urkoglu, G. J. Phys. Chem. C 2008, 112, 10338.

DOI: 10.1021/la904199f

7473

Article

the number of 4-Mpy molecules adsorbed on the Cu nanoparticles. Thus, the determination of the EF requires that the spectra from the adsorbed and free molecules be measured under identical conditions. Raman spectra of a 10-5 M concentration of 4-Mpy adsorbed on colloidal Cu nanoparticles and a 0.1 M concentration of 4-Mpy solution were measured to obtain information on the band intensities of adsorbed and bulk molecules directly under the same experimental conditions. Both experiments were within the laser focal volume (1 μm in diameter). Assuming the theoretical surface areas of 4-Mpy 40 nm size copper nanoparticles, we estimated that the theoretical maximum number of 4-Mpy molecules that could be adsorbed on our colloidal Cu nanoparticles in 1 mL would be 2.19  1017 (see detailed calculation in the Supporting Information). However, since we have, in fact, used only 1 mL of 10-5 M 4-Mpy in the colloidal Cu solution for SERS detection, the actual maximum number of adsorbed 4-Mpy is 6.02  1015 assuming complete adsorption of 4-Mpy on copper. On the other hand, the theoretical maximum adsorbed number was bigger than the actual number we used in the experiment. It is, therefore, obvious that the copper nanoparticle surface was large enough to adsorb all of the 1 mL of 4-Mpy (10-5 M). Here, assuming that all 4-Mpy molecules from the concentration of 10-5 M were adsorbed on the Cu nanoparticles, we used the concentration 10-5 M as the adsorbed species to calculate the EF. Since the concentration of bulk 4-Mpy solution was 0.10 M, we obtained an Nbulk/Nads ratio of 104. Combined with the intensity ratio, this makes the SERS EF for the peak at 1006 cm-1 of 4-Mpy to be 1.2  104. Thus, the average SERS EF of our polymer-capped colloidal Cu nanoparticles is estimated to be in the order of 104 (Table 1). These results indicate that Cu nanostructures, especially Cu nanoparticle colloidal systems, are attractive substrates for SERS detection of molecular species. However, the SERS spectrum of the copper nanorods exhibited strong fluorescence, which covered the SERS signals of 4-Mpy molecules (Supporting Information Figure S4). The observation of strong fluorescence in case of colloidal Cu nanorods and strong SERS in case of colloidal Cu nanoparticles can be explained on the basis of possible differences in the degree of interactions of the organic molecules with the different facets of the shaped copper nanostructures. For instance, for silver nanowires capped with PVP,36a it has been demonstrated that PVP interacts more strongly with the {100} facets (i.e., the side surfaces of a silver nanowire) than with the {111} facets (i.e., the ends of a silver nanowire). Attaching 1,12-dodecanedithiol molecules to the surfaces of a silver nanowire verified such a difference in interaction strength. The results demonstrated that the sides were completely passivated by PVP while the nanowire ends were only partially passivated (or essentially uncovered) by PVP.36a In light of these previous results, it is inconceivable that the sides of the copper nanorods could also be tightly passivated by PAAm while the nanorod ends were largely uncovered and remained more attractive (or reactive) toward molecules such as 4-Mpy. Thus, 4-Mpy molecules are expected to be preferentially adsorbed on the {111} facets (i.e., the ends) of Cu nanorods. This results in a strong laser excited fluorescence from the {100} facets of Cu nanorods, which overwhelms the SERS signals. However, for Cu (48) Hayazawa, N.; Tarun, A.; Inouye, Y.; Kawata, S. J. Appl. Phys. 2002, 92, 6983.

7474 DOI: 10.1021/la904199f

Wang and Asefa

nanoparticles, 4-Mpy molecules could cover up nearly all the crystal faces of Cu by replacing PAAm. This results in the fluorescence energy being able to transfer between the 4-Mpy molecules and the Cu nanoparticles, leading to the quenching of the fluorescence intensity and the amplification of the Raman enhancement factors.48 In addition, previous reports showed that the morphologies of as-obtained Au49 and Ag50 substrates could have a great effect on the SERS activities of the nanomaterials for organic molecules. For instance, Ag nanoparticles showed stronger SERS signals than Ag nanowires did.50 SERS enhancement from spherical Ag nanoparticles relies on interparticle coupling or aggregations, where the junctions of the adjacent two particles render the SERS-active sites.51 From this viewpoint, the nanorods with the least junctions have the smaller enhancement ability. Our observation that Cu nanoparticles gave rise to stronger SERS enhancement than Cu nanorods did is also consistent with this observation for Ag nanoparticles and nanorods.50 The SERS enhancement for the Cu nanoparticles compared to Cu nanorods could also be due to the aggregation of the Cu nanoparticles as shown in Figure 3. As mentioned above, stronger SERS enhancement due to nanoparticle aggregation has indeed been well-recognized for nanomaterials such as Ag nanoparticles.52

4. Conclusions In conclusion, we have successfully developed a facile, aqueousphase procedure for the synthesis of stable, polymer-coated copper nanoparticles with well-controlled size using PAAm as a capping agent. The size and shape of the Cu nanoparticles were controlled by changing the relative concentration of PAAm in the solution. We employed UV-vis spectroscopy to monitor the growth process of the nanoparticles. The results showed that yellow Cu2O seed nanoparticles were first formed at the beginning of the chemical reaction, which were then converted to Cu nanoparticles as the reaction progressed. The TEM studies indicated that the average size of the nanoparticles increased from 40 to 50 nm when the reaction time was increased from 30 to 90 min. The as-synthesized colloidal copper nanoparticles were proved to serve as effective SERS-active substrates with SERS enhancement factors in the order of 104. Acknowledgment. We gratefully acknowledge the financial assistance by the U.S. National Science Foundation (NSF), CAREER Grant No. CHE-0645348 and NSF DMR-0804846 for this work. Supporting Information is available online or from the authors. Supporting Information Available: UV-vis absorption spectra of Cu NPs synthesized using different PAAm concentrations; SERS spectra for 1  10-5 M 4-Mpy absorbed on copper nanorods. This material is available free of charge via the Internet at http://pubs.acs.org. (49) Wang, T.; Hu, X.; Dong, S. J. Phys. Chem. B 2006, 110, 16930. (50) Zhang, J.; Li, X.; Sun, X.; Li, Y. J. Phys. Chem. B 2005, 109, 12544. (51) Nikoobakht, B.; El-Sayed, M. A. J. Phys. Chem. A 2003, 107, 3372. (52) (a) Schwartzberg, A. M.; Grant, C. D.; Wolcott, A.; Talley, C. E.; Huser, T. R.; Bogomolni, R.; Zhang, J. Z. J. Phys. Chem. B 2004, 108, 19191. (b) Li, X.; Xu, W.; Zhang, J.; Jia, H.; Yang, B.; Zhao, B.; Li, B.; Ozaki, Y. Langmuir 2004, 20, 1298. (c) Lu, L.; Zhang, H.; Sun, G.; Xi, S.; Wang, H.; Li, X.; Wang, X.; Zhao, B. Langmuir 2003, 19, 9490.

Langmuir 2010, 26(10), 7469–7474