Seedless, copper-induced synthesis of stable Ag/Cu bimetallic nanoparticles and their optical properties

Abstract

In this study, we demonstrate a sensitive and selective method for the seedless synthesis of an Ag@Cu bimetallic nano-structured material based on the competitive coordination chemistry of cysteine with Cu²⁺ and Ag⁺. The subsequent addition of Ag⁺ to the cysteine–Cu²⁺, leads to the formation of a perfectly transparent orange–yellow color at 425 nm in the UV-visible region after ca. 3 h of mixing time. These changes are ascribed to the formation of Ag@Cu bimetallic nanoparticles. Visual observations and transmission electron microscope data indicate that the reaction mixture containing different orders of reactants (cysteine + Ag⁺, cysteine + Cu²⁺, Ag⁺ + Cu²⁺ + cysteine, and cysteine + Cu²⁺ + Ag⁺) have different colors as well as different morphologies. The reaction proceeds through the reduction of Ag⁺ ions into Ag⁰ by the side chain HS-moiety of the cysteine–Cu²⁺ complex. The resulting cystine–Cu²⁺ complex is subsequently adsorbed onto the surface of Ag⁰. The reduction of Cu²⁺ occurred on the surface of the Ag⁰ by under potential deposition. On the basis of this, the formation of Ag@Cu bimetallic nanoparticles can be visualized with the naked eye through the colorless-to-orange–yellow color change. Cysteine could not reduce the Cu²⁺ ions into metallic copper under normal conditions because Cu²⁺ has a strong affinity towards coordination through the thiol moiety.

---

1. Introduction

Cysteine has three N, O, and S donor atoms and/or coordination sites, i.e., HS–, NH₂– and –COOH. Its metal complexes, composites, and nanomaterials have wide applications in various fields (nanotechnology, surface science, biochemistry, medical science, purification of water, etc.).^1–6 It has been established that non-polar amino acids coordinate with transition metals via a chelate binding mode involving both carboxylate and amine groups, which is entirely dependent on the reaction media. Polar side chain amino acids provide extra opportunities and/or coordination sites for complexation with metals and act as bi-dentate and tridentate ligands.⁷ As far as biological systems are concerned, cysteine is an important ligand in terms of the functions of copper proteins and cysteine residues play a very crucial role in the stability and function of several proteins for coordination with Cu⁺ or Cu²⁺ ions.^8–10 Due to this, the synthesis and structural elucidation of cysteine–metal complexes, especially copper has been the matter of debate for over a decade.^11–15 Scarpa et al. discussed the stability and catalytic role of cupric ions in the anaerobic thiol oxidation of cysteine–cuprous complexes.¹³ Gardea-Torresdey et al. synthesized stable Cu(II) cysteine complexes (1 : 2, 1 : 4, and 1 : 6; Cu(II) : cysteine) in ethanol, suggesting that Cu(II) binding occurred via the thiol ligand of cysteine, and that copper remained in the Cu(II) oxidation state in all of the Cu(II)–cysteine complexes.¹⁵ Hammouti and his coworkers studied the inhibition effect of five amino acids (valine, glycine, arginine, lysine and cysteine) on the corrosion of copper under nitric acid in solution and reported that non polar side chain (valine and glycine) and polar side chain (arginine, lysine and cysteine) amino acids enhanced and inhibited the corrosion phenomenon, respectively.¹⁶

Ngeontae et al. used Cu²⁺-modulated cysteamine-capped CdS quantum dots for the detection of cyanide in water and suggested that due to the binding of Cu²⁺ to the cystamine surface of the quantum dots, the fluorescence intensity was quenched.¹⁷ Bimetallic nanoparticles are more attractive over monometallic nanocrystals because they exhibit improved electronic, optical and catalytic properties in chemistry, bio-chemistry and nano-biotechnology due to new bi-functional or synergistic effects.^18–21 The most important examples of bimetallic nanoparticles are Ag–Au, Ag–Ti, Pd–Pt, Au–Pd and Ag–Cu. Stroyuk et al. used photo catalytic reduction for the synthesis of various bi-, and tri-metallic (ZnO@Cu, ZnO@Pb, ZnO@Ag@Cu, ZnO@Ag@Cd and ZnO@Ag@Zn) nanocomposites and determined their optical activity.²² Sobczak and his coworkers have demonstrated that Ag–Cu bimetallic catalysts exhibit superior activity in low-temperatures compared to monometallic nanoparticles in the total catalytic oxidation of methanol to CO₂.²³ Zhang and his coworkers employed the oleylamine thermal reduction process to the synthesis of copper–silver bimetallic (Cu, Cu₅₀Ag, Cu₂₀Ag, Cu₁₀Ag, Cu₅Ag, CuAg, CuAg₂, and Ag) nanoparticles using Cu(CH₃COO)₂ and AgNO₃ as the precursors.²⁴ They determined the catalytic properties of these bimetallic nanoparticles and used them as model catalysts for silicon conversion in the Rochow reaction. Liu et al. used a seedless chemical reduction method for the synthesis of star-shaped Au@Ag bimetallic nanoparticles. The modified chitosan was chosen as a biocompatible surfactant to maintain the shape and increase the dispersion of nanoparticles.²⁵

In nanotechnology, metal ion reduction potentials and the stability of the complexes are important for the nucleation and growth redox processes, which leads to the formation of mono-, and poly-dispersed metal nanoparticles. Our goal in this study was to use a simple seedless coordination and/or chemical reduction method for the synthesis of Ag@Cu bimetallic nanoparticles at room temperature using cysteine as a reductant, which has there potential reaction sites. To the best of our knowledge, use of a reaction mixture containing three reactants, Ag⁺ + Cu²⁺ + cysteine, was reported for the first time for the synthesis of nanostructures. Here, we report the result of UV-visible, SEM, TEM, EDX, and XRD characterization of Ag@Cu nanoparticles.

2. Experimental section

2.1. Materials

All of the glassware were washed with aqua regia (3 : 1 HCl and HNO₃) and rinsed many times with double distilled (first time from alkaline KMnO₄) deionized water as well as acetone, followed by subsequent drying in an oven. Silver-nitrate (AgNO₃, Merck India, 99.99%), copper(II) nitrate pentahydrate (Cu(NO₃)₂·5H₂O, BDH, 99.9%), L-cysteine hydrochloride anhydrous (C₃H₇NO₂S·HCl, Sigma Aldrich, ≥99.0%), and cetyltrimethylammonium bromide (C₁₉H₄₂BrN, 99.8%, Fluka) were used without further purification. Stock solutions of AgNO₃ = 0.01 mol dm⁻³, Cu(NO₃)₂ = 0.01 mol dm⁻³, cysteine = 0.01 mol dm⁻³, and CTAB = 0.01 mol dm⁻³ were prepared in distilled water by direct weighing. The total volume was 50 cm³ in each experiment. Owing to the aerial oxidation of AgNO₃ in water, solutions were prepared daily, and stored in amber colored bottle.

2.2. UV-visible absorption measurements

UV-visible spectra were recorded on a UV-visible (UV-vis, Perkin Elmer, Lambda 35) double beam spectrophotometer having a deuterium and tungsten iodine lamp between the wavelength range of 200 to 700 nm at room temperature as a function of reactants concentration at different reaction time intervals. Quartz cuvettes of 1 cm path length were used for the measurements.

2.3. TEM measurements

In order to determine the morphology of the synthesized nano materials (AgNPs, and Ag@Cu bimetallic nanocomposites), transmission electron microscopy (Hitachi, H7 100) models was used operating at 200 kV. For TEM and SAED analysis, samples were prepared by placing a drop of the resulting sols onto a carbon-coated Cu grid followed by the slow evaporation of solvent at room temperature in open air.

2.4. XRD measurements

To confirm the crystalline nature, XRD spectra were recorded using Ni-filtered Cu Kα radiation (λ = 1.54056 Å) of a Rigaku X-ray diffractometer operating at 40 kV and 150 mA at a scanning rate of 0.02° per step in the 2 h range of 10° ≤ 2θ ≤ 80°. Samples of nanomaterials were collected by centrifugation around 10 000 rpm for 20 minutes at room temperature and washed several times with distilled water as well as with ethanol and acetone alternately to remove remaining reactants (Ag⁺ ions, Cu²⁺ ions, CTAB, NO₃⁻ ions) and then dried in a vacuum at room temperature for 3 h prior to use.

2.5. SEM and EDX measurements

Elemental composition and surface morphology (homogeneity and particle size) were determined using energy dispersive X-ray spectroscopy (EDX) on a TECHNAI-320 KV JAPAN, operating at 80 kV system equipped with energy dispersion X-ray spectroscopy and field emission scanning electron microscope QUANTA FEG 450, FEI Company, Eindhoven, The Netherlands.

2.6. Kinetic measurements

The kinetic measurements were carried out by mixing the required [cysteine], [Cu²⁺], and [CTAB] which are maintained at a constant temperature to a solution of [Ag⁺], and water (for dilution). The progress of the reaction (formation of colored Ag@Cu sols) was followed spectrophotometrically by pipetting aliquots of the sols at definite time intervals and their absorbance was measured at 425 nm. At this wavelength maximum, all reactants and stabilizers have no significant absorbance. Apparent pseudo first rate constants k_obs were calculated from the initial part of the slopes of the plots of ln(a/(1 − a)) versus time, where a = A_t/A_α (absorbance A_t at time t and A_α is the final absorbance) with a fixed time method.²⁶

2.7. pH measurements

Stability, morphology of nanoparticles strongly depend on the pH of the reaction media and the growth can be stopped by adding small amounts of mineral acids.²⁷ The pH also plays a significant role in the synthesis of nano-composites with amino acids.^28,29 An Accumet, Fisher Scientific digital pH meter 910 fitted with a combination electrode was used for pH measurements.

3. Results and discussion

3.1. General considerations and mixing order of Ag⁺ and Cu²⁺ ions

Preliminary observations showed that the oxidation of cysteine by Ag⁺ ions in the presence of Cu²⁺ ions has complicated reaction features and the formation of the yellow-orange color and/or yellow and turbid depending on the mixing order of the Ag⁺ ions and Cu²⁺ ions with cysteine. Silver ions are colorless, therefore the formation of silver nanoparticles was observed through color change since small silver nanoparticles are yellow.³⁰ A series of experiments were performed to find the actual role of Ag⁺ or Cu²⁺ ions under different mixing orders, such as cysteine + Ag⁺, cysteine + Cu²⁺, cysteine + Cu²⁺ + Ag⁺, and cysteine + Ag⁺ + Cu²⁺ in the synthesis Ag@Cu nanomaterials (Table 1). Interestingly, we did not observe the appearance of any color in the reaction mixture containing only Ag⁺ + Cu²⁺ ions . Yellow turbidity and pale blue colors were present for cysteine + Ag⁺ and cysteine + Cu²⁺, respectively (Table 1). Under our experimental conditions, reversing the order of reactant (Ag⁺ + Cu²⁺ + cysteine or Cu²⁺ + Ag⁺ + cysteine) addition does not alter the visual observations regarding the formation of colored sols. On the basis of these results, we may confidently state that the mixing order of Ag⁺ and Cu²⁺ in a solution of cysteine obviously changes the reaction path, nucleation and growth processes, which might be attributed to the coordinating nature and/or competitions between Ag⁺ and Cu²⁺ reacting with cysteine.

Table 1 Impact of reactant order of mixing on the appearance and pH of Ag@Cu bimetallic nanoparticles

Order of mixing	Observations	pH
Cysteine + Ag^+	White turbid	5.2
Cysteine + Cu^2+	Pale blue	5.2
Cu^2+ + Ag^+	No color	5.8
Cysteine + Cu^2+ + Ag^+	Yellow color; unstable	5.3
Cu^2+ + Ag^+ + cysteine	Yellow color; unstable	5.1
Cysteine + Cu^2+ + CTAB	Pale blue	5.5
Cysteine + Ag^+ + CTAB	Yellow-orange; stable	5.4
Cysteine + Cu^2+ + Ag^+ + CTAB	Brown; stable	5.4

---

In aqueous solution, cationic, zwitterionic, and anionic species of cysteine exist in solution (eqn (1)–(3)) due to the presence of pH sensitive –COOH, NH₂, and –SH groups. The concentrations of these species depends on the pH. Between pH ca. 3.0 and 8.0, zwitterionic species is the principal species, it has its maximum concentration at pH 6.0, the isoelectric point. The pH of the reaction mixture containing cysteine, Ag⁺, Cu²⁺, and CTAB were also measured, which was found to be nearly constant in presence of cysteine (weak acid; pK₁ = 1.71 (COOH); pK₂ = 8.33 (SH); pK₃ = 10.78 (NH₃)). The observed pH values with or without CTAB are summarized in Table 1. Our experimental conditions (pH = 5.2 to 5.4) are resists the formation of Cu oxide.

It has been established that the pH of the reaction media decides the coordination mode of amino acids with metal ions. α-Amino acids with coordinating side chains can function as tridentate ligands, leading to the formation of N,O,N–, N,O,O– and N,O,S– coordinating metal complexes under different experimental conditions.³¹ Concerning the biological role of polar chain amino acids as ligands, cysteine is of particular importance because of the high affinity of its thiol group for soft metal ions. For cysteine, the involvement of –SH, –COOH, and NH₂ groups with the coordination of metal ions have been the subject of debates from several decades and finally accepted that all three groups (–SH, –NH₂ and –COOH) are involved in the complex formation at different pH and ambient conditions.^32–34 Thus, cysteine would act as a uni-, bi-, and tri-dentate ligand, depending on the pH of the working solutions. In addition, the sulfur atom of its thiol group will facilitate the formation of polynuclear complexes with Cu²⁺ in presence of ethanol.^15,34 Under our experimental conditions (pH = 5.2 to 5.4), the zwitterionic species of cysteine is a major and reactive species.³⁵

3.2. Cu²⁺–, and Ag⁺–cysteine systems

Both the reduction of Cu²⁺ into Cu⁰ (formation of CuNPs; route (I)) and complexation between cysteine and Cu²⁺ (route (II)) are feasible in the cysteine–Cu²⁺ system (Scheme 1). Therefore, in order to gain insight into Scheme 1, Cu²⁺ ions (from 2.0 × 10⁻⁴ to 20.0 × 10⁻⁴ mol dm⁻³) were added to a series of reaction mixtures containing fixed [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³ in the absence and presence of a constant amount of [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³ at 30 °C. The most interesting features of the present observations are the appearance of pale blue (Table 1), instead of the brown-red color (characteristic of CuNPs). We have reported the synthesis and morphology of the perfectly transparent wine red colored CuNPs using ascorbic acid and hydrazine in the presence of CTAB and starch at room temperature.^36,37 In the present study, we did not observe the appearance of a red color, which might be due to the weak reducing nature of cysteine (standard oxidation–reduction potential of cysteine/cystine couple = −0.22 V³⁸) in comparison to ascorbic acid and hydrazine (reduction potential of N₂H₄/N₂ and ascorbic acid/dehydroascorbic acid couples are +0.23 and −0.09 V, respectively).

Scheme 1 Probable role of Cu²⁺ ions with cysteine.

Furthermore, the presence of radicals in situ was checked by adding acrylonitrile (10 cm³) in the same solution containing [Cu²⁺ ] = 10.0 × 10⁻⁴ mol dm⁻³, and [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³ at 30 °C. No white gel and/or precipitate formation was observed as the reaction proceeded, ruling out the possibility of Cu²⁺ ions reduction by cysteine. Gardea-Torresdey reported the synthesis and structural analysis of stable 1 : 2, 1 : 4, and 1 : 6 of Cu(II) : cysteine complexes in ethanol. They also suggested that Cu²⁺ ions coordinate with cysteine via the thiol and the oxidation state of copper remained as Cu²⁺ in all of the cysteine complexes.¹⁵ Thus on the basis of observed results (Table 1) and previous data, we may confidently state that cysteine could not reduce the Cu²⁺ ions into metallic Cu (Scheme 1; route (I)) under our experimental conditions. Reduction of Cu²⁺ by cysteine did not proceed unless Ag⁺ ions were added. On the other hand, a perfectly transparent yellow-orange colored stable silver sol and white precipitate were formed in presence of Ag⁺ ions, indicating Ag-nanoparticle formation and the reduction of Ag⁺ ions by cysteine (Table 1). These results are in good agreement with our previous observations on the formation of silver quantum dots in the cysteine–Ag⁺ redox system.³⁵

3.3. Surface plasmon resonance band of Ag@Cu bimetallic nanocomposites

In order to prepare a perfectly transparent colored Ag@Cu bimetallic sol by a competitive coordination method, the required aqueous solution of AgNO₃ and Cu(NO₃)₂ were first mixed at room temperature, then a moderate amount of reducing agent, cysteine solution, was added with and without CTAB. Interestingly, an initial precursor solution with faint blue color changed to yellowish-red after ca. 6 h, indicating the formation of nanoparticles. The resulting color was stable for several months under our experimental conditions. It is well known that optically, small size particles display surface plasmon resonance (SPR) bands with intensity and energy strongly dependent on the morphology. The formation of nanosized Ag@Cu bimetallic particles can be confirmed by optical spectroscopy. Fig. 1 and 2 show a representative set of UV-visible spectra and optical images of Ag@Cu nanoparticles in aqueous solution at different time intervals. The formation of bimetallic nanoparticles is indicated by the presence of a single SPR band (Fig. 1). The absence of two or more bands rules out the possibility for a mixture of two different metal nanoparticles, Cu and Ag.³⁹ The spectra of Ag@Cu nanoparticles display the SPR band at ca. 425 nm. The intensities and shape of the spectra increased and changed with time from a weak shoulder to a sharp band. A subtle shift of the SPR band to the shorter wavelength (blue-shift; indicated by arrow) is evident with the reaction time of the Ag@Cu nanoparticle formation. Fig. 2 (optical images; color of the solution changed from pale yellow, yellow, orange dark red to red) are in good correspondence with the results of the UV-visible spectra (Fig. 1). Interestingly, our spectrum does not show any characteristic absorption bands for CuO at around 800 nm.⁴⁰

Fig. 1 Time resolved UV-visible spectra of Ag/Cu nanocomposites. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [Cu²⁺] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, [Ag⁺] = 20.0 × 10⁻⁴ mol dm⁻³. Insert display TEM image of irregular quasi-spherical Ag@Cu nanocomposites at Cu : Ag ratio = 1 : 2.

Fig. 2 Optical images of Ag/Cu nanocomposites at different time intervals. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [Cu²⁺] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, [Ag⁺] = 20.0 × 10⁻⁴ mol dm⁻³, and time = 10 (A), 120 (B), 180 (C), 360 (D) and 840 min (E).

In order to further confirm the formation of Ag@Cu nanoparticles, the pure Ag-, and Cu-nanoparticles were also prepared under similar experimental conditions using cysteine and hydrazine (the reduction product is nitrogen, which has no impact on the morphology of the nanomaterials and also provides an inert atmosphere) as reducing agents, respectively. Fig. 3 shows the UV-visible spectra of the as prepared nanoparticles along with Ag@Cu nanoparticles in their final stage of preparation. Interestingly, it was seen that the absorption profile shows a high degree of well-defined SPR broad shoulder at 450 nm and a SPR sharp band at 590 nm, respectively, for the for the monometallic Ag- and Cu- nanoparticles. An entirely different behavior of SPR was observed when the cysteine solution was added in the reaction mixture containing Ag⁺ and Cu²⁺ ions (Fig. 3), indicating the formation of Ag@Cu bimetallic nanoparticles instead of individual corresponding nanoparticles under different experimental conditions.^41–44 Inspection of Fig. 3 clearly suggests that the position of the SPR band strongly depends on the composition of the Ag⁺ and Cu²⁺ ions (blue shift of ca. 75 nm in the absorbance of pure monometallic AgNPs from 450 to 375 nm and red shift of ca. 25 nm in the absorbance of Ag@Cu bimetallic nanoparticles from 375 to 400 nm for 1 : 1 and 1 : 2 Ag/Cu, respectively (as indicated by the arrow)).⁴⁵ Finally, we may confidently state that Ag@Cu bimetallic colloids show only one SPR at ca. 400 nm, which can be attributed to the plasmon resonance of silver particles alone. The appearance of only one absorption band corresponding to silver indicates that homogeneously mixed colloidal particles of the two metals are formed and the surface of the CuNPs are completely covered by silver. As a result, the optical properties of the composite nanoparticles are dominated by the Ag shell (Fig. 2 and 3).⁴⁶

Fig. 3 UV-visible spectra of pure monometallic (Ag and Cu) and bimetallic Ag/Cu nanocomposites. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [Cu²⁺] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, [Ag⁺] = 10.0 × 10⁻⁴ mol dm⁻³, [hydrazine] = 10.0 × 10⁻⁴ mol dm⁻³. For bimetallic Cu : Ag ratio = 1 : 1 (blue line) and 1 : 2 (red line).

3.4. SEM and TEM images of Ag@Cu bimetallic nanocomposites

Fig. 4 shows the SEM images of as synthesized Ag@Cu bimetallic nano-composites at two different Ag/Cu mole ratios. It is clear from these images that the small sized particles were deposited on the surface of the bigger particles (Fig. 4(A)). As the Ag : Cu mole ratio increases (from 2 : 1 to 4 : 1), the amount of deposition increases, leading to the formation of various particles with different morphology (Fig. 4(B)), and finally they formed layered, bimetallic Ag@Cu (Fig. 4(B)). It is also clear that particles are different in size, varying from 15 to 60 nm with a preferred bimodal distribution and small particles aggregate onto the surface of large particles.

Fig. 4 SEM images of Ag/Cu nanocomposites at different molar ratios of Cu²⁺ : Ag⁺. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, Cu²⁺ : Ag⁺ = 1 : 2 (A), and 1 : 4 (B).

The aggregated quantum dots of silver nanoparticles have been synthesized in neutral conditions in previous studies using cysteine as a reducing agent in the presence of CTAB.³⁵ In order to gain insight into the morphology, TEM measurements were also performed for the same Ag/Cu mole ratios (Fig. 5). A typical TEM image of the Ag@Cu bimetallic nanoparticles that were produced by competitive chemical reduction clearly shows the formation of bimetallic nanoparticles. When no copper was used, quantum dots of metallic silver were obtained, but the above reactions with low Ag : Cu ratios yielded irregular quasi-spherical nanoparticles. Fig. 5(A) presents the few irregular quasi-spherical nanoparticles obtained with an Ag : Cu ratio of 2 : 1. Inspection of Fig. 5(B) clearly indicates that some parts of some grains in the TEM image are darker, which means that these grains contain the two component materials and thus that silver/copper bimetallic particles were formed. The particles were observed to be poly-dispersed and aggregated, which might be due to the reduction of Cu²⁺ onto the surface of Ag⁰ and leads to the formation of Ag@Cu bimetallic particles (Fig. 5(A); diameter = ca. 10 to 40 nm). When the [Ag⁺] precursor is increased to [Ag⁺] = 10.0 × 10⁻⁴ mol dm⁻³ and 40.0 × 10⁻⁴ mol dm⁻³, the number of Ag@Cu nanoparticles increased, which might be due to the increase in nucleation sites (Fig. 5(B)). Therefore, the formation of Ag@Cu bimetallic irregular quasi-spherical NPs is strongly related to proportion of Ag⁺ ions. The capping is prominent on each particle and the same may also be responsible for inter particle binding. Each irregular quasi-spherical shaped particle is a group of several Ag/Cu nanoparticles. Fig. 5(B) shows the TEM images obtained at the different Ag : Cu ratio of 4 : 1 in this investigation, and the poly-dispersed large number of nanoparticles. The shape, size and size distribution of bimetallic Ag@Cu (irregular quasi-spherical shaped and poly-dispersed) are strongly related to the amount of [Ag⁺] ions. Interestingly, the number of particles increases with [Ag⁺] (Fig. 5(B)), indicates that the number of the nucleation sites increases, which in turn, provides more sites for Cu²⁺ ion reduction.

Fig. 5 TEM images of Ag/Cu nanocomposites. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, Cu²⁺ : Ag⁺ = 1 : 2 (A), and 1 : 4 (B). The darker color of some parts of the grains shows that the grains contain more than one material, and thus that silver/copper alloy nanoparticles were formed.

The high-resolution TEM (HRTEM) image of the Ag@Cu nanoparticles in Fig. 6 shows that Cu has a spacing distance of 0.21 nm, which corresponds to the interplanar distance of the (111) plane of the Cu, while the lattice fringe spacing of 0.24 nm corresponds to the lattice fringe distance of the (111) plane in Ag. The HRTEM image of Ag@Cu nanoparticles shows twinning, indicating that the nanoparticles were constructed from multiple crystal domains and that they have [111] plane twins.^25,47 The non-uniform black part of the center of the nanoparticles indicates the irregular growth of the nanoparticles, providing further evidence of three-dimensional growth. Large irregular quasi-spherical shapes are considered to be the outcome of smaller inter-particle fusions and are thus responsible for the prominent blue shift depicted in Fig. 1 and 3. These observations are in good agreement to the results of Bakshi et al. regarding the formation of bimetallic nanoparticles and capping action of proteins and phospholipids.^45,48,49

Fig. 6 High resolution TEM images (A and B) and electron diffraction ring patterns (C) of Ag/Cu nanocomposites. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, Cu²⁺ : Ag⁺ = 1 : 2 (A), and 1 : 4 (B).

3.5. EDX and XRD of Ag@Cu bimetallic nanocomposites

To confirm the bimetallic structure and the composition of the Ag@Cu nanoparticles, TEM-EDX analysis is also carried out. The EDX profile of cysteine capped Ag@Cu bimetallic nanoparticles (Fig. 7) indicates that the sample contains only silver and copper, with no peaks of oxygen and sulphur. In contrast, a bromine peak is observed in the EDX profile of the nanoparticles prepared at higher [Ag⁺] = 40.0 × 10⁻⁴ mol dm⁻³ (Fig. 7(B)). The EDX spectra of the nanoparticles also demonstrates the presence of only Ag and Cu elements with an approximate atomic ratio of 3 : 1. Comparison between the two EDX spectra indicates that the cysteine has been washed off the surface of nanoparticles during treatment and that AgBr is formed at higher Ag⁺ concentrations. The absence of the O peak in the EDX, confirms the formation of pure Ag@Cu nanoparticles with no oxide, which might be due the strong capping actions of cysteine and CTAB (TEM images; Fig. 5 and 6).

Fig. 7 EDX of Ag/Cu nanocomposites at different molar ratio of Cu²⁺ : Ag⁺. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, Cu²⁺ : Ag⁺ = 1 : 2 (A), and 1 : 4 (B).

Fig. 8 and 9 shows the XRD results of the synthesized Ag@Cu nanoparticles obtained by adding different [Ag⁺] (=20.0 × 10⁻⁴ and 40.0 × 10⁻⁴ mol dm⁻³) at a fixed [Cu²⁺] = 10.0 × 10⁻⁴ mol dm⁻³, [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, and [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³. For both samples obtained after adding the cysteine precursor, the eight characteristic peaks appeared at 2θ = 28.2°, 30.5°, 38.2°, 43.6°, 54.8°, 64.4°, 74.4°, and 78.8°. Out of these, the three main characteristic peaks at 43.6°, 54.8°, and 74.4°, which correspond to crystal facets of (111), (200), and (220) of the pure copper phase, respectively. Several other peaks at 2θ values of 28.2°, 30.5°, 38.2°, 64.4°, and 78.8° correspond to the lattice planes of (110), (111), (200), (220), and (311) of the silver phase (JCPDS card no. 04-0783),⁵⁰ suggesting that the samples are composed of both Cu and Ag phases. No additional impurities were detected either in the EDS or in the XRD profile. Each crystallographic facet contains energetically distinct sites based on atom density. The silver and copper nanoparticles both contain high atom density facets such as (111) that are known to be highly reactive.^24,51 Since it is clear from visual observations that Ag@Cu nanoparticles have not been oxidized for at least six months, it was demonstrated that the organic coating on the metal nanoparticles is effective in preventing them from oxidation.

Fig. 8 XRD of Ag/Cu nanocomposites. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [Cu²⁺] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, [Ag⁺] = 20.0 × 10⁻⁴ mol dm⁻³.

Fig. 9 XRD of Ag/Cu nanocomposites. Reaction conditions: [cysteine] = 10.0 × 10⁻⁴ mol dm⁻³, [Cu²⁺] = 10.0 × 10⁻⁴ mol dm⁻³, [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³, [Ag⁺] = 40.0 × 10⁻⁴ mol dm⁻³.

3.6. Kinetics and mechanism of Ag@Cu bimetallic nanocomposites formation

In our experiments, when cysteine was added to a solution containing Ag⁺ and Cu²⁺ ions, the color of the solutions turned from colorless to brown-yellow (Table 1), which was attributed to the presence of Ag@Cu bimetallic nanocomposites. In order to see the effect of Ag⁺/Cu²⁺ on the morphology and rate constant of Ag@Cu bimetallic nanoparticles, a series of experiments were carried out under different conditions. Table 2 indicated that different ratios of Ag⁺ and Cu²⁺ are responsible to the formation of the perfectly transparent, stable, and different colored silver/copper sols. Surprisingly, the absorbance of the resulting sols decreases with increasing Ag⁺/Cu²⁺ ratio. These results are also depicted graphically in Fig. S1 (ESI†). On the other hand, decreasing the order of Ag⁺/Cu²⁺ ratio has no significant effect on the rate constant. In the reaction mixture, there is a competition between Ag⁺ and Cu²⁺ ions to coordinates with cysteine. The positively charged metal ions acts as a Lewis acid, and the ligand, with one or more lone pairs of electrons, acts as Lewis base. Highly charged Cu²⁺ has the greatest tendency to act as a Lewis acid, and consequently, Cu²⁺ has a strong affinity to form complex ions compared to Ag⁺ (formation constants of [Ag(NH₃)₂]⁺, and [Cu(NH₃)₄]²⁺ are 1.1 × 10⁷, and 2.1 × 10¹³, respectively). Ag⁺ ions had a higher reducing potential than the Cu²⁺ ions (Table 3).

Table 2 Effects of molar ratios of reactants on the visual appearance and rate constant of the formation of Ag@Cu bimetallic nanoparticles in the presence of [CTAB] = 10.0 × 10⁻⁴ mol dm⁻³

[Ag^+/Cu^2+]^a	10^4 [cysteine] (mol dm^−3)	Absorbance, color, and stability	10^6 kobs (s^−1)
a [Cu^2+] is constant (10.0 × 10^−4 mol dm^−3) and [Ag^+] varying (from 2.0 × 10^−4 to 50.0 × 10^−4 mol dm^−3).b [Ag^+] is constant (10.0 × 10^−4 mol dm^−3) and [Cu^2+] varying (from 2.0 × 10^−4 to 10.0 × 10^−4 mol dm^−3).
0.2	10.0	0.71; pale yellow; unstable	4.4
0.5	10.0	0.68; yellow; unstable	4.2
1.0	10.0	0.64; dark yellow; stable	3.9
1.5	10.0	0.61; dark yellow; stable	3.6
2.0	10.0	0.60; brown; stable	3.3
2.5	10.0	0.52; brown; stable	3.0
3.0	10.0	0.41; brown; stable	2.7
4.0	10.0	0.32; brown; stable	2.4
5.0	10.0	0.24; brown; stable	2.0
5.0^b	10.0	0.64; dark yellow; stable	4.0
2.0	10.0	0.64; brown yellow; stable	3.9
1.0	10.0	0.64; brown yellow; stable	3.9
0.66	10.0	0.63; brown; stable	3.8
0.5	10.0	0.63; brown; stable	3.9
0.33	10.0	0.63; brown; stable	3.9

---

Table 3 Impacts of complex formation on the redox potential and free energy change (ΔG) of Ag⁺ ions, Cu²⁺ ions, and cysteine

Redox system	Standard chemical potentials	ΔG
Ag^+ + e^− → Ag^0	E0 = +0.799 V	−0.799 F
[Ag(NH3)2]^+ + e^− → Ag^0	E0 = +0.38 V	−0.38 F
Cu^2+ + 2e^− → Cu^0	E0 = +0.34 V	−0.68 F
[Cu(NH3)4]^2+ + 2e^− → Cu^0	E0 = +0.047 V	−0.094 F
Cysteine + e^− → cystine	E0 = −0.22 V	+0.22 F
Cysteine–Cu^2+ + e^− → cystine + Cu^+	E0 = +0.16 V	−0.16 F
Cu^2+ + e^− → Cu^+	E0 = +0.15 V	−0.15 F

---

It is well known that the coordination of metals and the coordination number vary with the types of metals bound in the protein and the metal cysteine bond is strong, with stability constants ranging from 10¹⁸ and 10²².⁵² Vallee and co-workers to have proposed that metal release is facilitated by the breakage of the cysteine–metal thiolate bond to form the corresponding cystine–cystine under oxidizing conditions.⁵³ Cysteine protected copper nanoparticles were synthesized by adding CuCl₂ + cysteine solution in a hydrazine solution at pH = 6.5, and used as a colorimetric sensor for Hg²⁺ ion determination. They suggested the adsorption of cysteine on the surface of Cu-nanoparticles by a side chain –SH group.⁵⁴ First, a cysteine–Cu(II) complex is formed (Scheme 1: vide supra), secondly, Ag⁺ ions reduced by a – coordinated –SH group of Cu²⁺–cysteine (cysteine–metal thiolate bond) to generate Ag-nanoparticles and a cystine–Cu²⁺ complex (Scheme 2).⁵⁵

Scheme 2 Mechanism of the reduction of Ag⁺ ions by cysteine–Cu²⁺.

As can be seen in Fig. 1, the reaction progress is typically very slow. We did not observe the formation of any significant color up to 100 min reaction time (Fig. S2; ESI†), which can be rationalized due to the complex formation between cysteine–Cu²⁺ and the Ag⁺ ions (Scheme 2, eqn (4)). Absorbance of Ag@Cu nanoparticles is found to be directly proportional to the [cysteine] (Fig. S3†), which might be due to the higher nucleation reaction sites. As a result, the reduction potential of the –SH group decreases, which in turn, decreases the overall reaction rates (reduction of Ag⁺ ions by cysteine–Cu²⁺; complex formation decrease in the potential (Table 3), and consequently, the reduction process may be rather involved). The reduction reaction of metallic ions is sensitive to the pH of the working solution. The majority of redox systems in aqueous solutions involve H⁺ and OH⁻ ions. As a result, the pH of the reaction medium can have a major impact upon the redox potential of the solutes, as predicted by the Nernst equation.⁵⁶ The overall cysteine–Cu²⁺–Ag⁺ redox reaction can be articulated in eqn (8).

Cysteine–Cu²⁺ + Ag⁺ → cystine–Cu²⁺ + Ag⁰ + H⁺ (8)

According to the Nernst’s law, the overall chemical potential can be written as:

Eqn (10) clearly suggests that the higher the pH, the higher the overall potential, and accordingly the faster the reaction will be. However, H⁺ ions will be successively involved during the reaction, and the reaction would gradually be getting slower. Zhou et al.⁵⁷ assigned the wave at −0.63 V to the reduction of the cysteine–mercury thiolate. These investigators, Zhao et al.⁵⁸ and Vallee et al.⁵⁹ also pointed out that the metal–thiolate bonds can be readily broken under oxidation to form cystine. Complete protonation of the –COOH and –NH₂ groups lowers the reduction potential to such extent that the oxidation site may effectively remain at the sulphur atom. Thus, the Cu²⁺–cysteine–Ag⁺ complex undergoes one-electron oxidation–reduction mechanism, leading to the formation of cystyl–Cu²⁺ radical and Ag⁰ (eqn (5); rate-determining step). In the next reaction, the cystyl–Cu²⁺ radical immediately gets converted into cystine–Cu²⁺ after fast dimerization (eqn (6)).⁵⁹

Finally, the cystine–Cu²⁺ complex is adsorbed onto the surface of the Ag-nanoparticle, and therefore, coordinated Cu²⁺ ions would be reduced on the surface of the Ag-nanoparticles by under potential deposition which act as seeds or active sites for further growth (Scheme 3). Owing to the catalytic effect of copper atoms, Ag grew from these active sites on the surface of the Ag-nanoparticles, generating irregular quasi-spherical shaped Ag/Cu bimetallic nanocomposites (Fig. 5; TEM image). The proposed mechanism is in accordance with the results of Liu, et al.²⁵ and Ungureanu et al.⁶⁰ for the formation of Au@Ag bimetallic nanoparticles by the seedless approach.

Scheme 3 Mechanism for the formation of Ag@Cu bimetallic nanoparticles.

Scheme 3 represents the adsorption of the cystine–Cu²⁺ complex onto the surface of Ag-nanoparticles, which immediately convert to Ag@Cu bimetallic nanoparticles. The schematic adsorption of Cu²⁺ on the surface of AgNPs, its reduction to Cu⁰, Ag growth, and TEM image are also summarized in Scheme 3. The reduction potential of Ag⁺/Ag metal = +0.799 V and Cu²⁺/Cu metal = +0.34 V. The aqueous reaction mixture containing AgNO₃ (10.0 cm³; 0.01 mol dm⁻³) and Cu(NO₃)₂ (5.0 cm³; 0.01 mol dm⁻³) remains colorless for a long period, suggesting that the reduction of Ag⁺ ions by Cu²⁺ ions does not occur under normal conditions. On the basis of these results, from the UV-visible absorbance spectra, and optical images, we can reasonably infer that the growth of Ag⁰ alone occurs on the surface of Ag/Cu (Scheme 3) rather than forming more nucleation sites.^61–64 Thus, it is not surprising that the optical properties of composite nanoparticles are dominated by metallic Ag.⁴⁶ These findings provide additional evidence for the formation of pure Ag@Cu bimetallic nanoparticles. Evident from Fig. 7, these nanoparticles have good monodispersity and most of the particles are spherical. The population of nanoparticles in Ag⁺/Cu²⁺ = 4/1 is evidently greater than that in Ag⁺/Cu²⁺ = 2/1.

4. Conclusions

In conclusion, we reported a simple seedless competitive chemical reduction method for the synthesis of stable Ag@Cu bimetallic nanoparticles using an aqueous solution of Ag⁺, Cu²⁺, and cysteine. EDX and XRD confirmed that nanoparticles consist only of silver and copper in a pure crystalline state without Cu₂O. The TEM and SEM revealed that the Ag⁺ : Cu²⁺ ratio has a significant impact on the morphology (from few irregular quasi-spherical to a large number of interconnected particles) of the resulting Ag@Cu nanoparticles. The results of the visible light spectroscopy presented in this paper show the reduction potential dependent nucleation path in the nanostructures synthesis.

Acknowledgements

This work was supported by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, under grant No. (130-28-D1436). The authors, therefore, gratefully acknowledge the DSR technical and financial support.

References

1. F. Arnesano, E. Balatri, L. Banci, I. Bertini and D. R. Winge, Structure, 2005, 13, 713–722.
2. S. Panigrahi, S. Kundu, S. Basu, S. Praharaj, S. Jana, S. T Pande, S. K. Ghosh, A. Pal and T. Pal, Nanotechnology, 2006, 17, 5461–5468.
3. H. Zheng, M. Chruszcz, P. Lasota, L. Lebioda and W. Minor, J. Inorg. Biochem., 2008, 102, 1765–1776.
4. Z. Wu, F. A. Fernandez-Lima and D. H. Russell, J. Am. Soc. Mass Spectrom., 2010, 21, 522–533.
5. M. Vandenbossche, M. Casetta, M. Jimenez, S. Bellayer and M. Traisnel, J. Environ. Manage., 2014, 132, 107–112.
6. M. Vandenbossche, H. Vezin, N. Touati, M. Jimenez, M. Casetta and M. Traisnel, J. Environ. Manage., 2014, 143, 99–105.
7. M. S. Masoud and O. H. A. El-Hamid, Transition Met. Chem., 1989, 14, 233–234.
8. H. Zheng, M. Chruszcz, P. Lasota, L. Lebioda and W. Minor, J. Inorg. Biochem., 2008, 102, 1765–1776.
9. M. Y. Combariza, A. M. Fahey, A. Milshteyn and R. W. Vachet, Int. J. Mass Spectrom., 2005, 244, 109–124.
10. Z. X. Tian, A. Pawlow, J. C Poutsma and S. R. Kass, J. Am. Chem. Soc., 2007, 129, 5403–5407.
11. R. J. Gale and C. A. Walker, Inorg. Chim. Acta, 1977, 21, 151–161.
12. O. Yamauchi, A. Odani and M. Takani, J. Chem. Soc., Dalton Trans., 2002, 3411–3421.
13. A. Rigo, A. Corazza, M. L. di Paolo, M. Rosetto, R. Ugolini and M. Scarpa, J. Inorg. Biochem., 2004, 98, 1495–1501.
14. K. Barouni, L. Bazzi, R. Salghi, M. Mihit, B. Hammouti, A. Albourine and S. El Issami, Mater. Lett., 2008, 62, 3325–3327.
15. K. M. Dokken, J. G. Parsons, J. McClureb and J. L. Gardea-Torresdey, Inorg. Chim. Acta, 2009, 362, 395–401.
16. K. Barouni, L. Bazzi, R. Salghi, M. Mihit, B. Hammouti, A. Albourine and S. E. Issami, Mater. Lett., 2008, 62, 3325–3327.
17. T. Noipa, T. Tuntulani and W. Ngeontae, Talanta, 2013, 105, 320–326.
18. N. Toshima and P. Lu, Chem. Lett., 1996, 25, 729–730.
19. M. P. Mallin and C. J. Murphy, Nano Lett., 2002, 2, 1235–1237.
20. S. Link and M. A. El-Sayed, J. Phys. Chem. B, 1999, 103, 8410–8426.
21. S. Link, Z. L. Wang and M. A. El-Sayed, J. Phys. Chem. B, 1999, 103, 3529–3533.
22. A. L. Stroyuk, V. V. Shvalagin and S. Y. Kuchmii, J. Photochem. Photobiol., A, 2005, 173, 185–194.
23. J. Czaplinska, I. Sobczak and M. Ziolek, J. Phys. Chem. C, 2014, 118, 12796–12810.
24. Z. Zhang, Y. Ji, J. Li, Z. Zhong and F. Su, RSC Adv., 2015, 5, 54364–54371.
25. L.-C. Cheng, J.-H. Huang, H. M. Chen, T.-C. Lai, K.-Y. Yang, R.-S. Liu, M. Hsiao, C.-H. Chen, L.-J. Her and D. P. Tsai, J. Mater. Chem., 2012, 22, 2244–2253.
26. Z. Khan, S. A. Al-Thabaiti, A. Y. Obaid, Z. A. Khan and A. O. Al-Youbi, J. Colloid Interface Sci., 2012, 367, 101–108.
27. B. G. Ershov, E. Janata and A. Henglein, J. Phys. Chem., 1993, 97, 4589–4594.
28. Y. Huang, Y. Lin and H. Chang, Nanotechnology, 2006, 17, 4885–4894.
29. Y. F. Huang, K. M. Huang and H. T. Chang, J. Colloid Interface Sci., 2006, 301, 145–154.
30. A. Henglein, J. Phys. Chem., 1993, 97, 5457–5471.
31. K. Severin, R. Bergs and W. Beck, Angew. Chem., Int. Ed., 1998, 37, 1634–1654.
32. P. Manikandan, B. Epel and D. Goldfarb, Inorg. Chem., 2001, 40, 781–787.
33. M. Belcastro, T. Marino, N. Russo and M. Toscano, J. Mass Spectrom., 2005, 40, 300–306.
34. G. Berthon, Pure Appl. Chem., 1995, 67, 1117–1240.
35. Z. Khan and A. Talib, Colloids Surf., B, 2010, 76, 164–169.
36. S. A. Al-Thabaiti, A. Y. Obaid, Z. Khan, O. Bashir and S. Hussain, Colloid Polym. Sci., 2015, 293, 2543–2554.
37. O. Bashir, S. Hussain, S. A. Al-Thabaiti and Z. Khan, Spectrochim. Acta, Part A, 2015, 140, 265–273.
38. P. C. Jocelyn, Eur. J. Biochem., 1967, 2, 327–331.
39. N. N. Kariuki, J. Luo, M. M. Maye, S. A. Hassan, T. Menard, H. R. Naslund, Y. Lin, C. Wang, M. H. Engelhard and C.-J. Zhong, Langmuir, 2004, 20, 11240–11246.
40. B. Nikoobakht and M. A. El-Sayed, Langmuir, 2001, 17, 6368–6374.
41. T. P. Ang and W. S. Chin, J. Phys. Chem. B, 2005, 109, 22228–22236.
42. Y.-S. Li, Y.-C. Lu, K.-S. Chou and F.-J. Liu, Mater. Res. Bull., 2010, 45, 1837–1843.
43. M. Valodkar, S. Modi, A. Pal and S. Thakore, Mater. Res. Bull., 2011, 46, 384–389.
44. J. Czaplinsk, I. Sobczak and M. Ziolek, J. Phys. Chem. C, 2014, 118, 12796–12810.
45. M. S. Bakshi, F. Possmayer and N. O. Petersen, J. Phys. Chem. C, 2007, 111, 14113–14124.
46. L. Lu, H. Wang, Y. Zhao, S. Xi, H. Zhang, J. Hu and B. Zhao, Chem. Commun., 2002, 144–145.
47. Z. L. Wang, J. Phys. Chem. B, 2000, 104, 1153–1175.
48. M. S. Bakshi, H. Kaur, P. Khullar, T. S. Banipal, G. Kaur and N. Singh, J. Phys. Chem. C, 2011, 115, 2982–2992.
49. M. S. Bakshi, F. Possmayer and N. O. Petersen, J. Phys. Chem. C, 2007, 111, 14113–14124.
50. W. X. Zhang, X. N. Yang, Q. Zhu, K. Wang, J. B. Lu, M. Chen and Z. H. Yang, Ind. Eng. Chem. Res., 2014, 53, 16316–16323.
51. D. W. Hatchett and H. S. White, J. Phys. Chem., 1996, 100, 9854–9859.
52. E. H. Fischer and E. W. Davie, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 3333–3334.
53. W. Maret and B. L. Vallee, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 3478–3482.
54. R. A. Soomro, A. Nafady, Sirajuddin, N. Memon, T. H. Sherazi and N. H. Kalwar, Talanta, 2014, 130, 415–422.
55. C. J. Hawkins and D. D. Perrin, Inorg. Chem., 1963, 2, 843–849.
56. D. V. Goia and E. Matijevic, New J. Chem., 1998, 1203–1215.
57. F. Song, A. L. Briseno and F. Zhou, Langmuir, 2001, 17, 4081–4089.
58. R. Zhao, J. Lind, G. Merbnyi and T. E. Eriksen, J. Am. Chem. Soc., 1994, 116, 12010–12015.
59. L.-J. Jiang, W. Maret and B. L. Vallee, Proc. Natl. Acad. Sci. U. S. A., 1997, 95, 3483–3488.
60. C. Ungureanu, R. Kroes, W. Petersen, T. A. M. Groothuis, T. Ungureanu, H. Janssen, F. W. V. B. Leeuwen, R. P. H. Kooyman, S. Manohar and T. G. V. Leeuwen, Nano Lett., 2011, 11, 1887–1895.
61. Y. Sun, B. Mayers and Y. Xia, Adv. Mater., 2003, 15, 641–646.
62. Y. Sun and Y. Xia, J. Am. Chem. Soc., 2004, 126, 3892–3901.
63. N. R. Jana, L. Grearheart and C. J. Murphy, Chem. Mater., 2001, 13, 2313–2320.
64. E. Hao and G. C. Schatz, J. Chem. Phys., 2004, 120, 357–365.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra26732c

---