pH Dependence of the Size and Crystallographic Orientation of the

Mar 16, 2011 - The present method provides a way out for the preparation of GNPs with homogeneous shape resolving the problem of simultaneous formatio...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/Langmuir

pH Dependence of the Size and Crystallographic Orientation of the Gold Nanoparticles Prepared by Seed-Mediated Growth Mohammad R. Rahman, Farhana S. Saleh, Takeyoshi Okajima, and Takeo Ohsaka* Department of Electronic Chemistry, Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, 4259-G1-5 Nagatsuta, Midori-ku, Yokohama 226-8502, Japan ABSTRACT: The effect of the pH of the growth solution on the size and crystallographic orientation of gold nanoparticles (GNPs) was studied during the course of the preparation of surface-confined spherical GNPs following a two-step protocol (electrochemical and chemical). GNPs were first electrodeposited onto a clean glassy carbon (GC) electrode and these GNPs were used as seeds. Seed-mediated growth of the electrodeposited GNPs was performed in a solution of H[AuCl4] at various pHs (5.0 to 0.5) using NH2OH as a reducing agent. The thus-prepared GNPs were characterized by electrochemical, UVvisible absorption spectral, SEM, and TEM studies. The nucleation (i.e., formation of the new seeds) was found to dominate over growth (i.e., enlargement of the seed particles) process at higher pH during NH2OH seeding, whereas only growth was recognized at low pH (0.5). Nonspherical byproducts were noticed when the seed-mediated growth was performed at higher pHs, but at pH 0.5 only spherical GNPs were observed. The present method provides a way out for the preparation of GNPs with homogeneous shape resolving the problem of simultaneous formation of nonspherical byproducts during the seed-mediated growth as well as for the preparation of GNPs with a Au(111) facet ratio as high as 97%. On the basis of the obtained results, the mechanism of the growth process at low pH is also discussed. Interestingly, an enhanced electrochemical response was obtained for the oxidation of H2O2 using the GNPs prepared at pH 0.5.

’ INTRODUCTION Gold-nanoparticles (GNPs) were found to have extraordinary electrocatalytic properties for potential applications in a number of fields.14 Two major parameters control the chemical, physical, and electrocatalytic properties of the GNPs, i.e., the particle size and shape (in other words, the crystallographic orientation) of the prepared nanometer scale gold (Au). Currently, shapecontrolled synthesis of nanoparticles has been achieved either by using geometric templates5 or by using some additives, such as polymers6 or inorganic anions,7 to regulate the particle growth. However, all of these methods require harsh conditions or laborious work in order to remove the unwanted residues from the target products, and thus exploring new strategies for shapecontrolled synthesis without any additives has attracted great interest.8 However, electrochemical methods are suitable for the control of size and geometry of the prepared nanoparticles in the absence of additives; much attention has been focused on the dependence of the catalytic and sensing properties on the size, shape, and exposed crystallographic facets of Au nanostructures.915 A common approach of controlling the morphology of nanoparticles is the seed-mediated growth method.1618 It has been found that seed-mediated growth of colloidal GNPs is noteworthy in several respects: (i) it produces particles of an improved monodispersion relative to the Frens method,19 (ii) it allows smaller particles to grow into larger particles of a predetermined size, and (iii) it can be applied successfully to surface-confined GNPs. r 2011 American Chemical Society

Hydroxylamine and ascorbic acid are common reducing agents used in the seed-mediated growth of gold nanoparticles.1618 The particle size can be manipulated by varying the concentration ratio of H[AuCl4]:seeds. In either one-step or step-by-step preparation of large-size spherical gold nanoparticles through the seed-mediated growth process by using hydroxylamine or ascorbic acid, a certain percentage of nonspherical byproducts such as nanorods, triangles, and hexagonal nanoplates were frequently observed.17 Recently, for the first time, we have reported the preparation of surface-confined monodispersed (spherical) GNPs enriched in a Au(111) facet by hydroxylamine seeding at a certain pH (0.5).20 In continuation of the previous work, a detailed investigation of the effect of various pHs on the size and crystallographic orientation of the surface-confined GNPs prepared by seed-mediated growth has been reported in the present work. The GNPs prepared by seed-mediated growth at various pHs were characterized by electrochemical, UVvisible absorption spectral, SEM, and TEM measurements. In addition, a probable mechanism is also proposed for the seed-mediated growth of the GNPs at low pH. Finally, the GC electrode modified with GNPs (enriched in Au(111) facet) was employed for the electrocatalytic oxidation of the H2O2. Received: January 13, 2011 Revised: February 22, 2011 Published: March 16, 2011 5126

dx.doi.org/10.1021/la200150h | Langmuir 2011, 27, 5126–5135

Langmuir

ARTICLE

Scheme 1. Fabrication of Various c-GNP/GC Electrodes

’ EXPERIMENTAL SECTION Materials. All of the reagents (of analytical grade) used in this work were purchased from either Kanto Chemicals Co. Ltd. (Tokyo, Japan) or Wako Pure Chemicals Industries Ltd. (Osaka, Japan) and used without further purification. All of the solutions were prepared with Milli-Q (18 MΩ cm) deionized water. A fresh solution of H[AuCl4] was prepared daily prior to use. Instrumentation. All electrochemical measurements were performed using a two-compartment three-electrode cell with a glassy carbon (GC) or a polycrystalline gold (poly-Au) or GNP-deposited GC working electrode, a spiral Pt wire counter electrode and Ag|AgCl| KCl(sat.) reference electrode. The counter electrode compartment was separated from the working electrode compartment by a sintered frit. Voltammetric responses were recorded using a computer-controlled ALS CHI-760D electrochemical analyzer (BAS, Tokyo, Japan), driven with a general-purpose electrochemical system software (BAS). Scanning electron microscopy (SEM) analysis of the deposited GNPs was carried out using a Hitachi S4700 field emission scanning electron microscope (Hitachi, Japan). JEM-2010 field emission transmission electron microscope (JEOL, Japan) was used to capture the TEM images. The UV absorption spectra were measured using a JASCO V-550 UV/vis spectrometer (JASCO, Japan). The pHs of the solutions were measured using a standard pH meter (IM-55G, TOA Electronics Ltd., Japan). Prior to each electrochemical experiment, N2 gas was bubbled directly into the cell for 30 min to obtain N2 saturated solution and all of the electrochemical measurements were carried out under this gas. All of the measurements were accomplished at room temperature (25 ( 1 C).

Preparation of the GNP Modified GC Electrodes. Scheme 1 shows the two-step protocol used in the fabrication of the desired electrodes. In the first step, GNPs were electrodeposited onto a clean GC electrode (3.0 mm in diameter) from an acidic bath of 0.5 M H2SO4 solution containing 0.1 mM Na[AuCl4] by applying a potential step from 1.1 to 0.0 V vs Ag|AgCl|KCl(sat.) for 10 s. The electrodeposited GNP modified GC electrode will be referred to as e-GNP/GC electrode. The electrode was electrochemically pretreated in 0.1 M H2SO4 solution by repeating the potential scan in the range of 0.20 to 1.50 V vs Ag|AgCl|KCl(sat.) at 0.1 V s1 until cyclic voltammogram (CV) characteristics for a clean Au electrode were obtained. In the second step, seed-mediated growth on the electrodeposited GNPs of the e-GNP/GC electrode was performed in a solution of H[AuCl4] containing NH2OH (0.3 mM of each) under stirred conditions (20 rpm) for 2030 min at various pHs (5.0, 3.6, 2.6, 1.4, and 0.5). Henceforth, GNPmodified electrodes prepared through seed-mediated growth at pH 5.0, 3.6, 2.6, 1.4, and 0.5 will be referred to as c-GNP/GC A, c-GNP/GC B, c-GNP/GC C, c-GNP/GC D, and c-GNP/GC E, respectively.

’ RESULTS AND DISCUSSION Electrochemical Characterization of the e-GNP/GC and Various c-GNP/GC Electrodes. Figure 1 shows the characteristic

currentpotential (IE) curves of the e-GNP/GC and various c-GNP/GC electrodes in a deoxygenated (i.e., N2-saturated) 0.1 M H2SO4 solution at a potential scan rate of 0.1 V s1. In all cases, redox peaks corresponding to the oxidation of the Au surface in the potential range of 1.1 to 1.4 V (except for the GC in Figure 1g) and the reduction of the Au surface oxide at ∼0.90 V 5127

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir

ARTICLE

Figure 1. CVs obtained at (a) c-GNP/GC A, (b) c-GNP/GC B, (c) c-GNP/GC C, (d) c-GNP/GC D, (e) c-GNP/GC E, (f) e-GNP/GC, and (g) bare GC electrodes in N2-saturated 0.1 M H2SO4 solution. Potential scan rate: 0.1 V s1.

were observed. This figure clearly shows the existence of different patterns for the formation of the gold oxide monolayer for different GNP-modified electrodes, i.e., a significant difference in the compositional ratios for the different single-crystallographic orientation facets constituting the GNPs in each case, albeit no quantitative estimation for each facet could be given. Moreover, different values of the real surface areas (Areal) of the deposited GNPs were estimated from the amounts of charge consumed during the reduction of the surface oxide monolayer of the Au (the peaks at ∼0.90 V in Figure 1af) using a reported value of 390 μC cm2 (Table 1).21 The differences in the Areal of the deposited GNPs obtained for various c-GNP/GC electrodes may originate from the differences in the size and population of

the prepared GNPs. The characteristic CV of the Au(111) singlecrystal electrode showed a sharp anodic peak at ∼1.4 V corresponding to the oxidation of the surface Au;20,22 two oxidation peaks for the formation of a monolayer of oxides on the Au(110) and Au(100) single-crystal electrodes have been reported.21 In Figure 1, it can be seen that the CVs of all of the c-GNP/GC electrodes provide a small but sharp anodic peak which is very similar to that observed at the Au(111) single-crystal electrode in the H2SO4 solution. The observed small cathodic shift of the anodic peak for the c-GNP/GC electrodes compared to that observed for Au(111) single crystal electrode may arise from the presence of two other low index facets. Thus, the results indicate that the deposition protocol of the GNPs adopted here for the 5128

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir

ARTICLE

Table 1. Characterization of the Different c-GNP/GC Electrodes Obtained under Various Conditions and Comparison of the Relative Ratios of the Different Low Index Facets Estimated from the Reductive Desorption of Cysteine from the Au(111) and Au (100) þ Au (110) Surface Domains of the Chemically Deposited GNPs on the GC Electrodes deposition time c-GNP/GC electrodes

Echem/(s)

A

10

20

B C

10 10

20 20

Chem/(min)

Areal of

pH of the solution during

deposited Au surface area

Chem deposition

(Areal)/(103 cm2)

QAu(111)a / (%)

Au(111)/(103 cm2)

5.0

9.1

88

8.0

3.6 2.6

10.6 18.0

91 78

9.6 14.0

D

10

20

1.4

8.0

90

7.2

E

10

30

0.5

4.3

97

4.2

a

QAu(111) represents the relative ratio of the amount of charge consumed during the reductive desorption of cysteine SAMs from the QAu(111) surface domain.

Figure 2. SEM images of 10 s electrodeposited GNPs on the GC electrode (a) and after growth of the electrodeposited GNPs ((be) 20 min and (f) 30 min) at different pHs: (b) 5.0, (c) 3.6, (d) 2.6, (e) 1.4, and (f) 0.5. Scale bar: (af) 1 μm and (insets of c and d) 100 nm.

fabrication of c-GNP/GC electrodes could populate GNPs enriched in Au(111) facet. The intensity of the sharp anodic peaks obtained at different c-GNP/GC electrodes was found to vary with the real surface area of the deposited GNPs. SEM Characterization of the Prepared Au Nanoparticles at the e-GNP/GC and Various c-GNP/GC Electrodes. Morphology of the electrochemically and chemically deposited GNPs was first investigated using SEM. Figure 2 shows the SEM micrographs of the e-GNP/GC and different c-GNP/GC electrodes. The size, shape, and population of the deposited GNPs were found to vary with the preparation conditions of the electrodes. Particle size distributions of ca. 1030, 1050, 2080, 50200, 20100, and 5080 nm were observed for e-GNP/GC, c-GNP/ GC A, c-GNP/GC B, c-GNP/GC C, c-GNP/GC D, and c-GNP/ GC E electrodes, respectively. A highest population of the GNPs

was observed when the growth was performed at pH 5.0, but the population decreased with decreasing the pH of the growth solution (from 5.0 to 0.5). The growth rate increased with decreasing pH from 5.0 to 2.6 but started to decrease from pH 1.4. Growth at pH 0.5 provided larger particles but similar particle density compared to that observed at the e-GNP/GC electrode, suggesting that only growth actually occurs at this pH. The explanation for the variation in the size and population of the GNPs prepared from the growth solutions of various pHs will be given in the following sections. A certain amount of nonspherical byproducts such as nanorods, triangles, and hexagonal nanoparticles were observed along with the spherical ones when the growth was performed at different pHs except at pH 0.5. Nanorods are evident at all pH while triangles and hexagonal nanoparticles can be observed at pH 2.6 and 1.4. The shape of the particle is almost spherical on the c-GNP/GC E electrode which may be responsible for the higher ratio of the Au(111) facet at this electrode. UVvis Absorption Spectral and TEM Studies of the H[AuCl4] Solution at Different pH’s in the Presence of NH2OH. SEM images (Figure 2) showed an increase in particle density during the seed-mediated growth on the electrodeposited GNPs at higher pHs, which may arise from the new nucleation during the reduction of H[AuCl4] by NH2OH. But Natan et al. reported that, at room temperature, mixtures of NH2OH and H[AuCl4] did not lead to the formation of colloidal Au particles in the solution without any seed,23 although NH2OH is thermodynamically capable of reducing Au3þ to bulk metal.24 However, the nucleation was recognized by Murphy et al. for a faster reduction process even in the absence of any seed using ascorbic acid as reducing agent.8b In the present work, NH2OH was added to the H[AuCl4] solution all at once under the stirred condition. The higher population of the GNPs at the c-GNP/GC electrodes (except for the c-GNP/GC E electrode) compared to that at the e-GNP/GC electrode may originate from either of the following two reasons: (i) a deposition of GNPs on the electrode surface which were produced from the nucleation of Au in the growth solution in the absence of any seed, and (ii) a direct nucleation at the electrode surface due to the weak reducing capability of the GC surface. The second option can be ruled out since the particle density at c-GNP/GC E electrode was found to be similar to that observed at the e-GNP/GC electrode. Besides, no GNP was noticed to be formed on a clean bare GC electrode in the growth solution of pH 0.5. To investigate the formation of the GNPs in the growth solution, UVvis absorption measurements were performed for the H[AuCl4] solutions of different pHs in the presence of NH2OH. 5129

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir

ARTICLE

Figure 4. TEM images for the GNPs obtained in the H[AuCl4] solutions containing equimolar (0.3 mM) NH2OH at various pHs: (a) 5.0, (b) 3.6, and (c) 2.6.

Figure 3. UVvis absorption spectra of the Au nanoparticles formed during the reduction of the H[AuCl4] by NH2OH in the growth solutions of different pHs: (a) 5.0, (b) 3.6, and (c) 2.6.

Figure 3 shows the typical UVvis absorption spectra. No absorption peak was observed for the solutions of pH 1.4 and 0.5. But the solutions of pH 5.0, 3.6, and 2.6 gave the absorption peaks at 545, 570, and 600 nm, respectively. The

observed absorption peaks confirm the formation of GNPs in the growth solutions in the absence of any seed.8b,2528 The relative intensity of the absorption peaks was found to decrease with decreasing pH from 5.0 to 2.6, suggesting a highest population of the GNPs in the growth solution of pH 5.0. Therefore, the highest population of the GNPs observed at the electrode c-GNP/GC A originates from the deposition of the GNPs onto the electrode surface in the growth solution of pH 5.0 (Figure 2b). Since the population of the GNPs in the growth solutions was decreased with decreasing pH, the population of the particles on the electrode surfaces was also noticed to decrease (Figure 2ce). Eventually, the population of the GNPs at the electrode c-GNP/GC was found to be similar to that at the electrode e-GNP/GC E, as no GNP was formed in the growth solution of pH 0.5 (Figure 2a,f). Formation of the GNPs in the growth solutions of various pHs in the absence of any seed was further verified by performing TEM measurements (Figure 4). No particle was recognized from the TEM images of the solutions of pH 1.4 and 0.5. Interestingly, 5130

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir

ARTICLE

Figure 5. CVs obtained at the e-GNP/GC (a) A, (b) B, (c) C, (d) D, and (e) E electrodes in N2-saturated 0.5 M KOH solution for the reductive desorption of the chemisorbed cysteine. Potential scan rate: 0.05 V s1.

the particle size distribution of ca. 24, 510, and 1030 nm could be noticed in the growth solutions of pH 5.0, 3.6, and 2.6, respectively (Figure 4ac). The results support the formation of GNPs in the growth solutions of these pHs in the absence of any seed. The formation of new GNPs at these pHs in the absence of any seed may originate from the faster reduction of H[AuCl4] by NH2OH in these stirred solutions. The sizes of the particles can be noticed to increase with decreasing pH from 5.0 to 2.6, revealing a higher growth rate at lower pH. Therefore, the size of the particles was found to increase in the following order: c-GNP/GC A < c-GNP/GC B < c-GNP/GC C (Figure 2bd). Again, the growth rate was found to decrease in the solutions of pHs 1.4 and 0.5, which is apparent from the size of the particles on the electrodes c-GNP/GC D and c-GNP/GC E (Figure 2e,f),

as most of the reducing agent (NH2OH) exists as its protonated form (NH3OHþ). Measurements of the Reductive Desorption Patterns of Cysteine-Self-Assembled Monolayer (SAM) Formed on the Various c-GNP/GC Electrodes. It has been reported that the polycrystalline Au surfaces are composed mainly of three lowindex crystallographic orientations, i.e., the Au(111), Au(100), and Au(110) domains.29,30 Each single-crystalline domain exhibits different binding strengths toward an attached selfassembled thiol. For instance, for a short-chain thiol species like cysteine, a multiple reductive desorption pattern was recorded for the reductive desorption from a polycrystalline Au electrode.29,30 The different peaks were assigned to the different crystallographic surface domains of the poly-Au 5131

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir

ARTICLE

Figure 6. Geometric shapes of (a) triangular, (b) rod-like, (c) truncated octahedral, and (d) decahedral nanoparticles.

electrode.2933 The relative ratio of the peak current intensities is reasonably proportional to the relative ratio of the lowindex single-crystallographic orientations constituting the composition of the Au substrate. In other words, it reflects the population (i.e., the percentage) of the different facets at the Au substrate surface. In order to give an assessment of the ratio of the different Au facets, the reductive desorption of cysteine was performed for different c-GNP/GC electrodes. Figure 5 shows the CVs measured in deoxygenated 0.5 M KOH solution, for the reductive desorption of the SAM of cysteine formed on the c-GNP/GC electrodes (AE). The cathodic peak observed at ca. 0.70 V corresponding to the desorption of cysteine from Au(111) facet was evident in all cases. A broad cathodic peak at ca. 0.95 V for the c-GNP/GC electrodes (except for the electrode c-GNP/GC E) and an additional peak at ca. 1.0 V for the c-GNP/GC C electrode were also recognized. The former corresponds to the desorption of cysteine from a Au(100) facet and the latter to that from Au(110) facet.29,30 The only major cathodic peak observed at ca. 0.70 V for the electrode c-GNP/GC E revealed a higher ratio of Au(111) facet at this electrode. Ultimately, from the amounts of charge consumed during the reductive desorption of cysteine at ca. 0.70 V and ca. 0.95 (and 1.0) V, the ratios of Au(111) facet to Au(100) þ Au(110) facets were calculated for all of the electrodes (Table 1). The electrode c-GNP/GC E was found to possess the highest ratio of Au(111) facet (ca. 97%) which is very high compared to previous reports.11,13 Correlation of the Crystallographic Orientation with the Different Shapes of the Prepared GNPs. The reductive desorption of cysteine from different c-GNP/GC electrodes showed the dependence of the relative ratio of their Au(111)

surface domain on the pH of the growth solution. Different values of the relative ratio of Au(111) facet for the prepared GNPs at the c-GNP/GC electrodes can be explained in terms of the geometrical shapes originating from different growth rates of different crystalline planes of the seed particles. According to the featured-structure of face-centered cubic (fcc) crystals, the triangular particle was proposed to be modeled as a truncated-tetrahedral shape with a {111} base bound to the substrate and three {100} side faces truncated at corners by less extended {111} faces, just as the model shown in Figure 6a (since the {111} base has larger area than each {100} face, it tends to bind the substrate for energetic stability).34 Since the area of {100} face is larger than that of {111} face enclosed to the surface of this particle, the growth rate (G) of the {111} face (G{111}) is much larger than that of the {100} (G{100}), which eventually leads to the disappearance of {111} faces. It is reported that the surface of the rodlike gold particle includes {110} as well as {100} faces, where ten {111} end faces and five {100} or {110} side faces are arranged around a common Æ110æ central axis, as modeled in Figure 6b.35 It originates from decahedral penta-twinned seed crystals with five {111} twin boundaries arranged radially to the Æ110æ direction of elongation, and the growth in the {100} or {110} side faces would be inhibited compared to that in the {111} end faces. Therefore, the preferential growth of the {111} faces over {100} or {110} also accounts for rod-like particles. However, a truncated octahedral (TO) shape cluster is bound by eight {111} and six {100} planes.36 If the growth rate of the {111} plane (G{111}) is significantly higher that of the {100} (G{100}), then the {111} facets will vanish, forming a cubic-like nanoparticle. On the contrary, if G{100} . G{111}, then the {100} facets will vanish, forming an octahedral 5132

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir

Figure 7. TEM images of (a) c-GNP/GC C and (c) c-GNP/GC E electrodes. Part (b) represents the HRTEM image of trancated end of one triangular particle and the inset of (b) shows the TEM image of a triangular particle. Parts (d and e) show HRTEM images of two different particles from (c). Part (f) represents the HRTEM image of an edge of a gold nanodecahedron.

nanoparticle. If G{111} ≈ G{100}, then the area ratio between the {111} and {100} facets remains constant, resulting in the formation of the larger TO nanoparticles (Figure 6c) which look hexagonal in SEM and TEM.37 Over and above these, twinning is frequently observed for fcc structured metallic nanocrystals. Twinning is the result of sharing a common crystallographic plane by two subgrains in such a way that one subgrain is the mirror reflection of the other by the twin plane. This kind of fcc structured nanocrsytal usually has {111} twins. In this case, usually decahedron is formed (Figure 6d) which is assembled from five tetrahedral subunits with the {111} crystallographic orientation and this type of particle looks spherical in SEM and TEM.20,3739 According to the above discussion, it can be concluded that the high ratio of Au(111) facet originates from the spherical particles at the electrode c-GNP/GC E, whereas the ratio is lower at the other electrodes due to the presence of triangular, rod-like or hexagonal particles. Formation of the homogeneous spherical nanoparticles may be facilitated at lower pH due to the slower growth rate. For further confirmation

ARTICLE

of the enrichment of the Au(111) facet at the c-GNP/GC E electrode, TEM measurements of the prepared GNPs were performed. TEM Studies of the Prepared Au Nanoparticles at the c-GNP/GC Electrodes. Further confirmation for the shape of the chemically deposited nanoparticles and their crystallographic orientation was made from HRTEM measurements. Figure 7 shows the TEM images of the c-GNP/GC C (Figure 7a) and c-GNP/GC E (Figure 7c) electrodes where the particle size distribution of ca. 30200 and 5090 nm can be observed for c-GNP/GC C and c-GNP/GC E electrodes, respectively. Rod- and triangular-shaped particles can easily be recognized after the seed-mediated growth at pH 2.6 (c-GNP/GC C electrode). The truncated corners of the triangular-shaped particles are confirmed from the HRTEM image of one particle (Figure 7b). However, the particles are found to be spherical when the growth was performed at pH 0.5 (c-GNP/GC E). The images (Figure 7d,e) of two nanoparticles reveal that the particle is roughly pentagonal and composed of five domains. Similar results have also been reported for the decahedral particles which are composed of five tetrahedral subunits with the fcc crystal structure and {111} crystallographic orientation.40 The particle is rounded compared with the ideal pentagonal projection in the proximity of the twin planes, and in some cases, small re-entrant facets occur at the boundaries. Figure 7f shows the HRTEM image taken at one edge of a gold nanodecahedron, in which no dislocations are observed, apart from just a slight distortion at the twin plane. It is difficult to quantify the deviation from ideal decahedral geometry from HRTEM images alone; however, previous microscopic analysis of these samples and electron-holographic analyses indicate that the {111} faceting and the nominal decahedral geometry are preserved.40,41 Thus, we suggest that the enrichment in the Au(111) facet may be related to the surface morphology of the deposited nanoparticles. Mechanism of the Seed-Mediated Growth. Attempts have been made by a number of researchers to understand the process of nucleation and growth for nanocrystals.42,43 In the present work, the size, shape, and population of the GNPs prepared through seed-mediated growth were found to depend largely on the pH of the growth solution. At higher pH, the nucleation predominated over the growth and vice versa at low pH. The effect of the concentration of Hþ on the rate of the nucleation and growth process can be involved with the “equilibrium reaction 1”44 related to the deprotonation of NH3OHþ Ka

NH3 OHþ h NH2 OH þ Hþ fKa ¼ 5:9  106 mol dm3 , pKa ¼ 5:2g

ð1Þ

From the fact that the NH2OH species exist predominantly at higher pH (5.0) (ca. 40%) compared to that at lower pHs and they act as a strong reducing agent and thermodynamically are capable of reducing [AuCl4] to Au0 as well as the SEM images in the Figure 2, the nucleation is considered to be more favorable at higher pH. Consequently, a higher population of the GNPs was recognized on the electrodes at higher pHs. At lower pHs, the reducing ability of the NH2OH decreases due to its higher degree of protonation, resulting in the retardation of the nucleation process. Hence, the population of the GNPs 5133

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir

ARTICLE

seed particles provide electrons for the reduction of the [AuCl2] (reaction 6) by its self-oxidation (reaction 7) resulting in the growth of the seeds.46 Au0  3e þ 4Cl f ½AuCl4 

Figure 8. CVs obtained at (a,b) poly-Au and (c, d) c-GNP/GC E electrodes in N2-saturated 0.5 M KOH solutions containing (a, c) 0.0 and (b, d) 2.0 mM H2O2 initially. Potential scan rate: 0.1 V s1.

on the surface of the GC electrode was found to decrease with decreasing the pH of the growth solution. The mechanism of the reduction of H[AuCl4] by NH2OH in acidic media was proposed as follows:45 k1

½AuCl4  þ NH2 OH sf ½AuCl2  þ HNO þ 2Cl þ 2Hþ ð2Þ k2

½AuCl4  þ NH3 OHþ þ Cl sf ½AuCl2  þ HNO þ 3Cl þ 3Hþ

ð3Þ fast

2HNO sf N2 O þ H2 O

ð4Þ

Since no Cl ion is present initially in the present study, the reduction of [AuCl4] can be considered to be initiated by a small amount of NH2OH present at pH 0.5 according to reaction 2, where [AuCl2] and Cl ions are formed. In the second step, the formation of these two species continues through the reduction of [AuCl4] by NH3OHþ in the presence of the Cl ions formed during the first step (reaction 3). In the present work, the observed growth of the seed particles may proceed through either of the following two processes: (i) spontaneous disproportionation of the [AuCl2]; 3½AuCl2  h 2Au0 þ ½AuCl4  þ 2Cl

ð5Þ

or, (ii) a direct reduction of the [AuCl2]; ½AuCl2  þ e f Au0 þ 2Cl

ð6Þ

Since no colloidal gold was formed in the growth solution of pH 0.5 in the absence of any seed, the growth of the electrodeposited GNPs via reaction 5 can be ruled out. It can be assumed that a so-called local cell reaction takes place on the electrodeposited GNPs on the GC electrode due to the different local potentials at different sites which arise from an initial electric field asymmetry. Therefore, we consider that the electrodeposited

ð7Þ

Electrochemical Oxidation of H2O2. The electrochemical behavior of H2O2 is important in the clinical diagnostic, food, and pharmaceutical industries, as well as in environmental monitoring, bioassays, etc., as it is the product of the metabolism in the biological systems and many enzyme-based biosensors.47,48 It is, however, difficult to detect H2O2 accurately by electrochemical techniques because the electrode reaction of H2O2 is very irreproducible at most kinds of electrodes having poor activity (their activity is further reduced after they are used) toward its electrochemical oxidation/reduction. Catalytic decomposition of H2O2 in contact with the metallic electrodes is also a notable obstacle toward its quantification.49 The bare poly-Au electrode has very poor activity toward the oxidation of H2O2 in both acidic and alkaline media. Even Pt electrode does not show a considerable activity for this reaction. The oxidation of 2.0 mM H2O2 at the bare poly-Au electrode in N2-saturated 0.5 M KOH solution was carried out and the results are shown in Figure 8. The anodic and cathodic peaks at 0.3 and 0.1 V obtained in both the absence (Figure 8a) and presence (Figure 8b) of H2O2 are assigned to the oxidation of the poly-Au electrode surface to form the oxide layer and its reduction, respectively. A potential-independent plateau oxidation current (jpla) was obtained above 0 V during the anodic scan. The difference in the oxidation currents in the presence and absence of H2O2 can be attributed to the net oxidation current of H2O2. The cathodic peak at ca. 0.14 V obtained at the bare poly-Au electrode is assigned to the reduction of the O2 produced in the anodic cycle (Figure 8b). Figure 8c,d shows the CVs obtained at the c-GNP/GC E electrode in N2-saturated 0.5 M KOH solution containing (c) 0.0 and (d) 2.0 mM H2O2. Remarkable large peak current for the oxidation of H2O2 to O2 and also a negative shift of the onset potential of the oxidation was observed at the modified electrode, i.e., 70 mV, compared to that at the poly-Au electrode. The oxidation peak current (jpa) obtained at the electrode c-GNP/ GC E at ca. 0.8 V (Epa) is more than 5 times higher than the jpla obtained at the poly-Au electrode at 0.1 V for the same concentration of H2O2. The enhancement of the oxidation current at the c-GNP/GC E electrode can be correlated to the catalytic activity of Au(111) facet in the oxidation of H2O2.29 With a continuous potential cycle, the anodic and cathodic peak currents and potentials changed negligibly. The CV obtained at the c-GNP/GC E electrode (the peak separation was 182 mV) indicates a quasi-reversible response of the H2O2/O2 redox couple with the formal potential (E0/) of 165 mV vs Ag/ AgCl/KCl (sat.) in 0.5 M KOH solution. Therefore, the most notable thing is that the introduction of Au(111)-like nanoparticles on the GC electrode results in an enhanced electrochemical oxidation of H2O2.

’ CONCLUSIONS A two-step (electrochemical and chemical) protocol was employed for the preparation of spherical GNPs with a homogeneous size distribution. Seed-mediated growth of the surfaceconfined GNPs was investigated at various pHs. The prepared GNPs were characterized electrochemically and the surface 5134

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135

Langmuir morphology was studied using SEM. The nucleation was found to predominate over the growth process at higher pH and vice versa at lower pH. Triangular, rod-like, hexagonal nanoparticles were observed as byproducts when the growth was performed at different pHs. Interestingly, only spherical-shaped GNPs were evident from the SEM studies when the growth was performed at pH 0.5 which provides a way out for the preparation of surfaceconfined monodispersed GNPs. The reductive desorption of a chemisorbed short-chain thiol, i.e., cysteine, from the GNPs prepared at different pHs revealed a high ratio of Au(111) facet (97%) for the GNPs prepared at pH 0.5. Enrichment of the Au(111) surface domain at such low pH was further confirmed from the TEM measurements; TEM images revealed the GNPs of decahedral shape which are composed of five tetrahedral subunits with the fcc crystal structure and {111} crystallographic orientation. Finally, a probable mechanism for the seed-mediated growth at low pH was proposed. Application of the thus prepared GNPs (enriched in Au(111) facet) toward the electrochemical oxidation of H2O2 revealed an enhanced catalytic activity compared to that observed at the poly-Au electrode.

’ AUTHOR INFORMATION Corresponding Author

*Phone: þ81-45-9245404. Fax: þ81-45-9245489. E-mail: [email protected].

’ ACKNOWLEDGMENT The present work was financially supported by Grants-in-Aid for Scientific Research (A) (No. 19206079) to T.O., from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan and by the JapanChina Research Program on Enzyme-based Biofuel Cells sponsored by Japan Science and Technology (JST) and Natural Science Foundation of China (NSFC). M.R.R. and F.S.S. thank the Government of Japan for a MEXT scholarship. ’ REFERENCES (1) Guo, X.; Hu, N. J. Phys. Chem. C 2009, 113, 9831. (2) Buso, D.; Palmer, L.; Bello, V.; Mattei, G.; Post, M.; Mulvaney, P.; Martucci, A. J. Mater. Chem. 2009, 19, 2051. (3) Xiao, L. X.; Wildgoose, G. G.; Compton, R. G. Anal. Chim. Acta 2008, 620, 44. (4) Ishida, T.; Kinoshita, N.; Okatsu, H.; Akita, T.; Takei, T.; Haruta, M. Angew. Chem., Int. Ed. 2008, 47, 9265. (5) (a) van der Zande, B. M. I.; Bohmer, M. R.; Fokkink, L. G. J.; Schonenberger, C. Langmuir 2000, 16, 451. (b) Spatz, J. P.; Mossmer, S.; Hartmann, C.; Moller, M.; Herzog, T.; Krieger, M.; Boyen, H. G.; Ziemann, P.; Kabius, B. Langmuir 2000, 16, 407. (6) (a) Ahmadi, T. S.; Wang, Z. L.; Green, T. C.; Henglein, A.; ElSayed, M. A. Science 1996, 272, 1924. (b) Kumar, S.; Yang, H.; Zou, S. J. Phys. Chem. C 2007, 111, 12933. (7) Filankembo, A.; Giorgio, S.; Lisiecki, I.; Pileni, M. P. J. Phys. Chem. B 2003, 107, 492. (8) Caswell, K. K.; Bender, C. M.; Murphy, C. J. Nano Lett. 2003, 3, 667. (9) Liu, H.; Tian, Y. Electroanalysis 2008, 20, 1227. (10) El-Deab, M. S.; Sotomura, T.; Ohsaka, T. Electrochem. Commun. 2005, 7, 29. (11) Chowdhury, A. N.; Alam, M. T.; Okajima, T.; Ohsaka, T. J. Electroanal. Chem. 2009, 634, 35. (12) Jena, B. K.; Raj, C. R. Anal. Chem. 2008, 80, 4836. (13) Gao, F.; El-Deab, M. S.; Okajima, T.; Ohsaka, T. J. Electrochem. Soc. 2005, 152, A1226.

ARTICLE

(14) Othman, S.; El-Deab, M. S.; Okajima, T.; Ohsaka, T. Electrochem. Commun. 2009, 11, 1273. (15) Oyama, T.; Okajima, T.; Ohsaka, T. J. Electrochem. Soc. 2007, 154, D322. (16) Brown, K. R.; Natan, M. J. Langmuir 1998, 14, 726. (17) (a) Jana, N. R.; Gearheart, L.; Murphy, C. J. Langmuir 2001, 17, 6782. (b) Jana, N. R.; Gearheart, L.; Murphy, C. J. Chem. Mater. 2001, 13, 2313. (18) Watzky, M. A.; Finke, R. G. Chem. Mater. 1997, 9, 3083. (19) Frens, G. Nature Phys. Sci. 1973, 241, 20. (20) Rahman, M. R.; Okajima, T.; Ohsaka, T. Chem. Commun. 2010, 46, 5172. (21) Trasatti, S.; Petrii, O. A. Pure Appl. Chem. 1991, 63, 711. (22) Hamelin, A. J. Electroanal. Chem. 1996, 407, 1. (23) Brown, K. R.; Walter, D. G.; Natan, M. J. Chem. Mater. 2000, 12, 306. (24) E0 = 1.0 V vs NHE for [AuCl4] þ 3e f Au0 þ 4Cl and E1/2 = 0.9 V vs NHE for the oxidation of NH2OH. Encyclopedia of Electrochemistry of the Elements; Bard, A. J., Ed.; Marcel Dekker, Inc.: New York, 1975; Vols. 4 and 8. (25) Link, S.; El-Sayed, M. A. Int. Rev. Phys. Chem. 2000, 19, 409. (26) Link, S.; El-Sayed, M. A. Annu. Rev. Phys. Chem. 2003, 54, 331. (27) Kreibig, U.; Vollmer, M. Optical Properties of Metal Clusters; Springer: Berlin, 1995. (28) Bohren, C. F.; Huffman, D. R. Absorption and Scattering of Light by Small Particles; Wiley: New York, 1983. (29) El-Deab, M. S.; Arihara, K.; Ohsaka, T. J. Electrochem. Soc. 2004, 151, E213. (30) Arihara, K.; Ariga, T.; Takashima, N.; Arihara, K.; Okajima, T.; Kitamura, F.; Tokuda, K.; Ohsaka, T. Phys. Chem. Chem. Phys. 2003, 5, 3758. (31) Walczak, M. M.; Alves, C. A.; Lamp, B. D.; Porter, M. D. J. Electroanal. Chem. 1995, 396, 103. (32) Zhong, C. -J.; Zak, J.; Porter, M. D. J. Electroanal. Chem. 1997, 421, 9. (33) Yang, D. -F.; Wilde, C. P.; Morin, M. Langmuir 1996, 12, 6570. (34) Cao, L.; Zhu, T.; Liu, Z. J. Colloid Interface Sci. 2006, 293, 69. (35) Johnson, C. J.; Dujardin, E.; Davis, S. A.; Murphy, C. J.; Mann, S. J. Mater. Chem. 2002, 12, 1765. (36) Petroski, J. M.; Wang, Z. L.; Green, T. C.; El-Sayed, M. A. J. Phys. Chem. B 1998, 102, 3316. (37) Wang, Z. L. J. Phys. Chem. B 2000, 104, 1153. (38) Ino, S. J. Phys. Soc. Jpn. 1966, 21, 346. (39) Allpress, J. G.; Sanders, J. V. Surf. Sci. 1967, 7, 1. (40) Kinoshita, T.; Seino, S.; Okitsu, K.; Nakayama, T.; Nakagawa, T.; Yamamoto, T. A. J. Alloys Compd. 2003, 359, 46. (41) Pastoriza-Santos, I.; Sanchez-Iglesias, A.; Garcıa de Abajo, F. J.; Liz- Marzan, L. M. Adv. Funct. Mater. 2007, 17, 1443. (42) Nikoobakht, B.; El-Sayed, M. A. Chem. Mater. 2003, 15, 1957. (43) (a) Gai, P. L.; Harmer, M. A. Nano Lett. 2002, 2, 770. (b) Sun, Y.; Mayers, B.; Herricks, T.; Xia, Y. Nano Lett. 2003, 3, 955. (44) Goyal, B.; Prakash, A.; Mehrotra, R. N. Indian J. Chem. Sect. A 1999, 38, 541. (45) Soni, V.; Mehrotra, R. N. Trends Metal Chem. 2003, 28, 893. (46) E0 = 1.0 and 1.15 V E0 = 1.0 V vs NHE for [AuCl4] þ 3e f Au0 þ 4Cl and [AuCl2] þ e f Au0 þ 2Cl, respectively. Bard, A. J., Ed., Encyclopedia of Electrochemistry of the Elements; Marcel Dekker, Inc.: New York, 1975; Vol. 4. (47) Hedenmo, M.; Narvaez, A.; Dominguez, E.; Katakis, I. J. Electroanal. Chem. 1997, 425, 1. (48) Forzani, E. S.; Rivas, G. A.; Solis, V. M. J. Electroanal. Chem. 1995, 382, 33. (49) Merkulova, N. D.; Zhutaeva, G. V.; Shumilova, N. A.; Bagotzky, V. S. Electrochim. Acta 1973, 18, 169.

5135

dx.doi.org/10.1021/la200150h |Langmuir 2011, 27, 5126–5135