Synergism of gold and silver invites enhanced fluorescence for practical applications

Abstract

Synergism of gold and silver, causing enhanced fluorescence, has been reported in cluster science with higher photochemical stability and practical applications. The electronic factor, nuclearity, size effect etc., bring autofluorescence, doping/alloying, aggregation induced fluorescence, core–shell interaction, oxidation induced interaction and silver effect. The plausible interplaying mechanisms behind the synergism in the bimetallic clusters has been focused upon. Appropriate selection of the template is mandatory to achieving AuAg bimetallic clusters. Such templates are scarce and only a few are reported in the literature while templates to obtain individual Au and Ag clusters are numerous. Semi/complete reduced gold and non/complete reduced silver are one of the important features of such bimetallic clusters. Mingled Au and Ag has a profound effect on the stability, electronic structure and band energy of the bimetallic clusters. The arrangement of Au and Ag atoms in the bimetallic clusters is also a matter of interest. The bimetallic AuAg clusters are found to be superior to not only individual Au/Ag clusters but also carbon and semiconductor quantum dots, considering their emissive nature, toxicity, ease of synthesis, robustness etc. Water miscible as well as water immiscible solvents are equally efficient for the production of AuAg bimetallic clusters. Finally, such bimetallic clusters have proved to be unique candidates in the context of practical applications, namely sensing, catalysis, surface enhanced Raman spectroscopy (SERS), metal enhanced fluorescence (MEF), bio-imaging, synthesis of anti-bacterial cotton/papers etc. The ratio of Au and Ag not only tune the fluorescence behavior but also toxicity, as described.

---

Mainak Ganguly

Dr Mainak Ganguly received his Ph.D. degree under the supervision of Prof. Tarasankar Pal and Dr Anjali Pal from Indian Institute of Technology, Kharagpur, India in 2014. He is currently working as a post-doctoral researcher in the Department of Chemistry, Furman University, USA. His research interests include synthesis of noble metal fluorescent clusters using biocompatible scaffolds, metal enhanced fluorescence and sensing.

Jayasmita Jana

Jayasmita Jana received her B.Sc. and M.Sc. degree from Vidyasagar University (West Bengal), India. She is currently pursuing her Ph.D. degree under the supervision of Prof. Tarasankar Pal at Indian Institute of Technology, Kharagpur, India. Her area of research is synthesis and various applications of fluorescent carbon dots.

Anjali Pal

Dr Anjali Pal, M.Sc., Ph.D. graduated from Calcutta University in chemistry and at present is an associate professor in the department of civil engineering, Indian Institute of Technology, Kharagpur. Dr Pal is actively engaged in teaching and research in the field of environmental engineering and science. She has published more than 150 research papers. Dr Pal has received an International Hall of Fame award (USA), R&D-100 award (USA) and convention award from the Indian Chemical Society. She has visited many countries as a visiting professor. She is working in the field of adsolubilization, catalysis and spectroscopy mainly for environmental remediation.

Tarasankar Pal

Professor Tarasankar Pal, M.Sc., Ph.D., D.Sc. has been working in the Department of Chemistry, Indian Institute of Technology, Kharagpur since 1984. He is a fellow of the Indian National Academy of Sciences, Allahabad. He has published more than 320 papers. Recently the ACS highlighted him among 20 Indian authors for high-quality research. Prof. Pal has received numerous national and international awards which includes ISCAS and Prof. J. N. Mukherjee endowment gold medals, R&D-100 Award from the USA, CRSI medal and medals from the Indian Chemical Society. His research interest includes synthesis and applications of metal and semiconductor nanoparticles for catalytic, electrocatalytic and spectroscopic applications. Original contribution in surface-enhanced Raman scattering (SERS), synergistic fluorescence enhancement by Ag and Au particles and supercapacitor fields are noteworthy. His ‘arsenic detection kit’ and ‘bench marked model reaction’ to test any metal nanoparticle catalyst deserve special mention.

---

1. Introduction

The synergistic effect is the resultant effect of two or more species when the resultant effect is greater than the sum of their individual effects. Synergism is just like epistasis, a phenomenon that consists of the effect of one gene being dependent on one or more ‘modifier genes’ (genetic background). Different consequences are observed with epistatic mutations in combination rather than individually [Fig. 1].¹

Fig. 1 The gene for total baldness is epistatic to those for blond or red hair. The baldness phenotype supersedes genes for hair colour and so the effects are non-additive.

Synergism is a well known in the field of catalysis. Metals such as Ag, Cu, Ni, Pd and Co bestow pronounced effects in promoting oxygen activation in gold-catalyzed oxidation reactions.² Gold and silver, being present in the same group (11), show synergistic effects very frequently unlike the other elements in the periodic table. Again, Liu et al.³ reported that Au–Ag alloy nanoparticles exhibit exceptionally higher catalytic activity for low-temperature CO oxidation compared with monometallic gold. Interestingly, such alloys have a lower surface-to-volume ratio in comparison to monometallic gold, explaining the essence of the synergistic effect. Not only in catalysis, but also in other fields of research synergism between gold and silver is observable. Phytofabricated Ag–Au nanoparticles was found to have improved antibacterial and anticandidal potential compared with their monometallic counterparts with specific reference to some pathogenic bacteria and Candida sp.⁴

Luminescent noble metal (gold, silver) nanoclusters (NCs), with discrete electronic states, have become highly promising over the last few years, possessing notable optical properties for extensive use in sensing,^5–8 imaging^9,10 and biolabeling.^11–13 As the sizes of Ag and Au NCs are analogous to the Fermi wavelength of an electron (i.e., the electron de Broglie wavelength at the Fermi level: ∼0.5 nm for Au and Ag), they bestow molecular-like behavior including size-dependent fluorescence.¹⁴ For defining the size-dependent electronic structure and relative electronic transitions of the small NCs, the Jellium E_Fermi/N_1/3 energy scaling law can be employed as a model.¹⁵

Template-based synthetic approaches have been employed to form fluorescent Au and Ag NCs from their corresponding salts using reducing agents such as NaBH₄ and templates namely poly(amidoamine) (PAMAM) dendrimers, polyglycerol-block-poly(acrylic acid) copolymers, proteins, DNA, thiols etc.^14,16–30 However, the templates for the gold–silver bimetallic silver clusters are limited in the literature. Au NCs are easier to synthesize than Ag NCs. This is mainly due to the poor stability associated with Ag NCs as they are vulnerable to oxidation. A suitable choice of templates is found to be somewhat useful to prevent rapid aggregation and oxidation of the Ag clusters. In comparison to the Au NC, the Ag NC show intense fluorescence, augmenting its efficacy to different key applications.¹⁶ Recently, extensive research on AuAg NCs has been conducted to overcome the shortcomings of individual Au NCs and Ag NCs.

Gold–silver synergism in fluorescence is becoming a very promising phenomenon and many researchers have ventured into this area recently. Bimetallic particles, made of silver and gold, can easily be synthesized in different size regimes and compositions for their similar lattice constants. Thus, mixed crystals over the almost whole concentration range of Au and Ag can be produced.^31–33 Bimetallic clusters of silver and gold have proven to be a promising route to bypass the usually weak emission (in comparison to other common fluorophores) of gold clusters as the bimetallic clusters exhibit a noticeably enhanced quantum yield.^34,35 Such an intense emission of AuAg NCs is the key feature of various useful applications. Of late, luminescent Au/Ag NCs have been employed as novel luminescent probes for sensing toxic metals, bioimaging and biosensing.^36–44 To achieve various improved applications, the research community has become interested in developing facile synthetic protocols of strongly luminescent and stable AuAg NCs.^45,46 Among the diverse techniques, mass spectrometry has become a crucial analytical tool to comprehend the atomic composition of such NCs. Moreover, the theoretical prediction of the structure of the these bimetallic clusters is also in progress.⁴⁷

Plausible mechanisms of synergism in AuAg clusters, suitable templates for the synthesis of such clusters, their oxidation states, electronic-band structure as well as applications have been summarized in this review.

2. Mechanism for synergistic evolution of fluorescence

There are ample debates regarding the mechanism of fluorescence enhancement, when both gold and silver are present in the NCs.

2.1. Autofluorescence

Emissive bimetallic silver–gold nanoparticles in aqueous medium were demonstrated by Ristig et al.⁴⁸ in one pot. 11-mercaptoundecanoic acid-functionalized nanoparticles were prepared with various Au, Ag molar ratios (90 : 10 to 10 : 0). The particle sizes were very small (1.8 ± 0.4 nm) qualifying the criterion of NCs. They reported such fluorescence as autofluorescence. The size of the bimetallic particles has been explained to be the reason for interband transition, accounting for the emissive property. The purified dispersions of bimetallic particles, exhibiting a wide range of color variation, were synthesized. This color variation was controlled by the variation of Ag/Au content; solutions with high Au content were colorless while solutions with higher amounts of Ag appeared light brown in color under visible light. Pure gold nanoparticles and samples with molar compositions of Ag : Au ranging from 10 : 90 and 60 : 40 showed a decent fluorescence signal when a 254/365 nm fluorescence lamp was employed. However, the absorption and emission maxima were distinctly altered with the composition as shown in Fig. 2. A red shift in the optical spectral profile with a higher silver content is also found from the work of Guével et al.⁴⁹ Particles with a greater silver content (>60 mol%) and pure silver nanoparticles could not exhibit any autofluorescence presumably due to the slightly larger particle size.⁴⁸ So, Ristig et al. believed that the decreasing size is the corollary of the AuAg bimetallic particle formation, resulting in fluorescence evolution. The emission-color ranged from bright orange (high Au content) to red (high Ag content).

Fig. 2 Excitation and emission spectra of nanoparticles with different molar Ag : Au compositions. The numbers in parentheses indicate the maxima of the absorption and emission bands, respectively, for the six samples. [Ref. 48] – ©The Royal Society of Chemistry.

2.2. Doping/alloying

Doping is the intentional introduction of impurities into a pure substance in order to tune its properties. Semiconductors are frequently converted to conductors as a result of doping which causes a dramatic change in the band gap energy. The remarkable alteration of the emissive property of Au NCs via Ag doping is presumably due to the change of band gap energy.

Ag NCs have proven to be more fluorescent than Au NCs.^50–53 However, the main limitations to biological applications are due to the high reactivity and Ag related cytotoxicity with the discharge of Ag ions from Ag NCs. Thus, Ag NCs are less popular for practical applications. Moreover, oxidation of Ag NCs causes a rapid decrease in the emissive property. In order to develop brighter metal clusters, silver has been added by Guével et al.⁴⁹ during the synthesis of Au NCs, protected by glutathione (GSH). Thus Ag doped AuGSH (AuAgGSH) has been produced ([Au]/[Ag] = 50.1). AuAgGSH are non-cytotoxic and possess a stronger emissive property. An enhancement of the emission intensity by a factor of 4 to 5 with silver compared to that of AuGSH (excitation at 400 nm and emission at 615 nm) has been obtained [Fig. 3].

Fig. 3 (a) Excitation (dashed line)/emission (solid line) spectra of AuAgGSH. (b) Normalised excitation spectra of AuGSH (blue line) and AuAgGSH (red line) with λ_em = 615 nm. The picture in the inset shows the fluorescence of AuGSH and AuAgGSH diluted to the same optical absorption under UV light (λ_ex = 366 nm). (c) The relative fluorescence intensity of diluted AuAgGSH solutions prepared with different amounts of silver salt during the synthesis. Results show an increase of the fluorescence of the clusters in the presence of silver independently of the dilution. Ref. 49 © 2012 Royal Society of Chemistry. Reproduced from ref. 49 with permission from The Royal Society of Chemistry.

Several noble metal clusters exhibit a much longer fluorescence lifetime (one or two magnitudes higher than organic dyes and quantum dots).^14,54 Multiple components of the decay profile for lifetime measurements are dependent on the excitation wavelength. AuGSH and AuAgGSH, synthesized by Guével et al.,⁴⁹ showed a fluorescence lifetime having a biexponential decay: a short component s1 (15 ± 2 ns) and a long component s2 (250 ± 20 ns) [Fig. 4]. The absolute value of either lifetime component of AuGSH remained unaffected with the variation of emission wavelength, pH and the existence of silver. Interestingly, there was a distinguishable impact on the long lifetime component s2, when Ag was doped. Ag caused a notable increase of the amplitude of s2. Again, ∼95 nm red shifted emission was favored because of silver doping. Patel et al. connected the long component to the cluster–ligand electron transfer, associated with the redox process.⁵⁵

Fig. 4 Relative amounts of the long lifetime component s2 (∼250 ns) for AuGSH and AuAgGSH at two different emission wavelengths with λ_ex = 405 nm. The presence of silver and an emission wavelength in the red region of AuAgGSH at higher wavelength led to a “quasi monoexponential” long lifetime decay. Lifetime decays were fitted with global lifetimes. Ref. 49 © 2012 Royal Society of Chemistry. Reproduced from ref. 49 with permission from The Royal Society of Chemistry.

Zhang et al.²⁴ also revealed a rapid synthesis of GSH-capped AuAg NCs (GS-AuAg NCs) via microwave irradiation. The doped GS-AuAg NCs have a much higher quantum yield (7.8% for GS-AuAg and 2.2% for GS-Au NCs) in comparison to GS-Au NCs. Their work has demonstrated that Ag/Au 0.2 brings the maximum quantum yield [Table 1]. Au–Ag charge redistribution to compensate the electron-withdrawing properties of the thiolate has been attributed to the cluster stability. Absorption spectra of the samples contain peaks at 500 nm as well as shoulder peaks at ∼400 nm. Stepwise multiple bands and localized surface plasmon resonance (LSPR) bands, characteristic of plasmonic gold or silver nanocrystals,⁵⁶ do not appear in the absorption spectra contradictory to nanocrystal formation.²⁴

Table 1 The photoemission wavelength and quantum yield of the GS–Au/Ag NCs with various Ag : Au molar ratios. The quantum yield was obtained with quinine sulfate (in 0.5 M H₂SO₄, 54%) as the reference

Ag/Au	Emission wavelength (nm)	Quantum yield (%)
0	610	2.2
0.1	614	4.6
0.2	618	7.8
0.5	614	4.3
1	614	4.3

---

Alloy formation between gold and silver produces a distinctly different metal core, causing a large fluorescence enhancement. Udayabhaskararao et al.⁵⁷ synthesized Ag₇Au₆ (a 13-Atom Alloy Quantum Cluster, QC), protected with mercaptosuccinic acid from the Ag_7,8 cluster. The excitation and emission maxima were at 670 and 770 nm for Ag_7,8, while the excitation and emission maxima are at 390 and 650 nm, respectively for the alloy cluster. The Ag_7,8 cluster shows weak fluorescence (quantum yield 8 × 10⁻³) at room temperature (300 K) and the quantum yield is 3.5 × 10⁻² for Ag₇Au₈ clusters under similar conditions. Alteration in luminescence property between Ag_7,8 and Ag₇Au₈ were attributed to the core-modifications owing to alloying. The lesser extent of electron donation from the Au core of the QCs to thiol may be a vital point in this issue. One of the possible structures of the alloy QC, as reported by Udayabhaskararao et al.,⁵⁷ is a distorted icosahedral core of C_2v symmetry and the thiolates are bonded in a bridged form (–Au/Ag–thiolate–Ag/Au) [Fig. 5].

Fig. 5 Changes observed during synthesis. (A and B) Solutions of Ag₇,₈ (A) and the alloy QC (B) after synthesis under visible light. Insets: the same samples under UV light. (C) UV/vis profile of (a) Ag₇,₈ and (b) alloy QC measured in water; arrows indicate the well-defined optical features of the cluster. (D) Luminescence spectra of (a) Ag7,8 and (b) alloy QC in water at 300 K. (E) Photographs of (a, a1) alloy QC in water and (b, b1) in the solid state under visible and UV light. (F) Comparison of the PAGE of (a, a1) Ag₈ and (b, b1) alloy QC. Photographs of gel in visible (a, b) and UV (a1, b1) light. Band positions are marked with circles on the gels. (G) Simulated absorption spectrum using time-dependent density functional theory (TDDFT). The data are compared with the experimental spectrum (full line). The region in between 1.4 to 2.3 eV is expanded. Inset: one of the optimized structures of the model cluster, Ag₇Au₆(SCH₃)₁₀, with which the spectrum was simulated. Ag large light gray, Au black, S small dark gray, CH₃ small light gray. Reprinted with permission from ref. 57. © 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Doping and alloying are similar to some extent. Both of them indicate mixing. However, more specifically for doping, the dopant has a remarkably smaller concentration unlike alloying. The perturbation of the properties of individual substances is expected to be greater in the case of doping.

2.3. Aggregation induced fluorescence

Dou et al.⁴⁴ made strongly emissive thiolated Au@Ag nanoclusters (NCs) where Ag(I) acts as a bridge between small Au(I)–thiolate motifs on the weakly luminescent thiolated Au nanoclusters. Thus large Au(I)Ag(I)–thiolate motifs are located at the surface of nanoclusters engendering highly emissive substance via aggregation induced emission (AIE).

The luminescence lifetime value assisted in understanding the luminescence property of the as-synthesized Au@Ag NCs. Luminescence decay profile indicated the predominance of long lifetime (in the microsecond scale) components in the luminescent Au@Ag NCs: 2.21 ms (56%), 0.641 ms (31.4%), 0.121 ms (10.3%), and 11.8 ns (2.3%). The microsecond-scale lifetime values were also reported for aggregation induced luminescence of Au NCs.²⁰ These data support emission of the so-produced Au@Ag NCs is due to AIE of Au(I)/Ag(I)–thiolate complexes on the surface of NC. It was actually phosphorescence involving the metal-centered triplet states. Clearly this luminescence property was unlike the nanosecond emission from the singlet excited states, previously stated for luminescent DNA-protected Ag NCs.³⁶

AIE of Au@Ag NCs originated from the complexes on the NC surface and the size/structure of the Au(I)/Ag(I)–thiolate complexes became the main reason for their strong luminescence behavior in the visible to near-infrared region. Dou et al.⁴⁴ assumed that the intense red emission of the Au@Ag NCs is caused by the linking of tiny Au(I)–thiolate motifs on the parental Au NCs via the Ag(I) linkers. As a result, large Au(I)/Ag(I)–thiolate motifs are formed exhibiting strong luminescence via AIE. Since the Ag(I) ions serve as linkers in forming the large Au(I)/Ag(I)–thiolate complexes on the Au NCs, the removal of Ag(I) linkers breaks those emission-active species [large Au(I)/Ag(I)–thiolate complexes] and scrubs off those Ag(I) ions from the NC surface. Thus, the Au@Ag NCs revert back to the parental Au NCs, annulling their strong luminescence in solution. The Ag(I) linker, present on the surface of luminescent Au@Ag NCs, is detached by the introduction of a specific thiolate ligand, cysteine (Cys), that is capable of interacting efficiently with Ag(I) ions to form Ag(I)–thiolate complexes.⁵⁸ Thus, the large red emission of Au@Ag NCs is instantaneously destroyed with the introduction of Cys [Fig. 6].

Fig. 6 (a) Schematic illustration of the light-up process for the synthesis of highly luminescent Au@Ag NCs by using Ag(I) ions as linkers connecting the small Au(I)–thiolate motifs on the parental Au NC surface. (b) UV-vis absorption (solid lines) and luminescence spectra (dashed lines, λ_ex = 520 nm): spectra of the parental Au₁₈(SG)₁₄ NCs (black lines) and luminescent Au@Ag NCs (red lines). (Insets) Digital photos of the parental Au₁₈(SG)₁₄ NCs (item 1 and 2) and luminescent Au@Ag NCs (item 3 and 4), under visible (item 1 and 3) and UV (item 2 and 4) light. (c) Luminescence decay profiles (top panel) of the luminescent Au@Ag NCs. The red line is a tetra-exponential fit of the experimental data. The bottom panel shows the residuals of fitting. Ref. 44 © 2013 Royal Society of Chemistry. Reproduced from ref. 44 with permission from The Royal Society of Chemistry.

2.4. Core–shell interaction in the giant cluster

With the gradual decrease of the particle size, the continuous density of states splits to separate energy levels resulting in dramatic optical, electrical and chemical behaviors (in comparison to nanoparticles) when the particle size approaches the Fermi wavelength of electrons (∼0.5 nm is the electron de Broglie wavelength at the Fermi level for Au and Ag).¹⁶ So, clusters are usually very small often sub-nanometer in size. However, aggregation induced emission,⁴⁴ as discussed earlier, is interesting in this context. Another intriguing report was made by Ganguly et al.,³⁴ revealing a fluorescent giant cluster, protected by glutathione. An Au(I)_core–(Ag₂/Ag₃)_shell model was proposed. The fluorescent species were Ag₂/Ag₃ cluster, decorated on the Au(I) surface resulting in highly fluorescent giant clusters (∼500 nm) [Scheme 1]. Electron withdrawal from Ag(0)_shell to Au(I)_core was ascribed to the long term stability. Along with glutathione, Au(I) also acts as a scaffold for fluorescent silver clusters. Maretti et al.⁵⁹ demonstrated that the positive surface helps to stabilize Ag NCs and thus Au⁺ acts as a stabilizer for giant clusters. Such Au(I)_core–(Ag₂/Ag₃)_shell giant clusters are so stable that they can be obtained as a luminescent solid after vacuum drying. Again, they could be re-dispersed in different water miscible solvents to engender fluorescent solutions.⁶⁰ To synthesize these giant clusters, the optimal silver(I) concentration should be 1.7 times higher than that of gold(III). So, precursor [Ag] > [Au] is quite different from other literature reports. However, further increases in the precursor [Ag(I)] destabilized the giant clusters due to the aggregation of tiny fluorescent silver clusters on the surface of Au(I).⁶⁰ Ganguly et al.⁶¹ were also able to produce a weakly fluorescent Au(I)–glutathione specimen. The product, obtained via a sluggish synthetic protocol, possessed a poor quantum yield and ∼55 nm red-shifted emission peak. This observation indicates the synergistic effect of gold and silver inviting new fluorescence behavior.

Scheme 1 Synergistic evolution of highly fluorescent Ag₂ and Ag₃ clusters on the Au(I) surface by solar light irradiation of Ag(I)–glutathione and Au(I)–glutathione. Reprinted with permission from (ref. 60). © 2013 American Chemical Society.

Metal enhanced fluorescence (MEF) demands large-aggregated particles with higher scattering cross-section so that there is no lossy surface wave.^62–66 Concentrating the electric field around the metalized surface in the presence of a weakly fluorescing fluorophore to enhance the rate of excitation (lightening rod effect) and higher radioactive decay rates of such fluorophores in the presence of the metal causing the higher rate of emission are the key issues for MEF. On the contrary, fluorescent clusters are very tiny and often sub-nanometer in size containing discrete energy levels. Again, emissive giant clusters are, as a whole, micro particles. However, it does not fulfil the criterion of MEF. Aqueous glutathione is not at all fluorescent, at least under ∼400 nm excitation (the optimal excitation wavelength for giant clusters). Moreover, using glutathione (reducing and capping agent for the giant clusters), separate Ag NCs and Au NCs have been produced by many groups.^67,68

2.5. Oxidation and interaction induced fluorescence enhancement

Glutathione protected Au₂₅(SG)₁₈ NCs was prepared by Wu et al.⁶⁹ They titrated the solution against Ag⁺ and obtained an intriguing fluorescence enhancement. By using hydrogen peroxide also, they observed a similar enhancement and explained the silver induced fluorescence enhancement to be due to the oxidation of gold. However, stronger oxidizing ions, e.g., Au³⁺ and Hg²⁺ unlike Au⁺ did not enhance the fluorescence. They explained this on the basis of the effect of heavy ions or their chemically reduced species. Adequate Ag⁺ caused further fluorescence enhancement of Au₂₅(SG)₁₈ after the oxidation of Au₂₅(SG)₁₈ NCs by H₂O₂ or Ag⁺.

It was revealed that the amounts of silver ions required to obtain the highest emission after the completion of Au₂₅(SG)₁₈ oxidation was 5 equivalents (with respect to Au₂₅) suggesting that a certain structured complex was responsible for the most intense emission. Added Ag⁺ followed the chelation-enhanced fluorescence (CHEF) mechanism for the increase of fluorescence of Au₂₅(SG)₁₈ NCs.^70–72 The red-shifted emission with the silver addition, after the completion of oxidation of Au₂₅(SG)₁₈, supported the association of Ag⁺ with Au₂₅.

Wu et al.⁶⁹ also studied the consequence of the reduced silver [neutral silver, Ag(0), from the stoichiometric redox reaction between Au₂₅(SG)₁₈ and Ag⁺] on the emission of Au₂₅(SG)₁₈. In both these circumstances, the post-oxidation processes became similar. But, an alteration was observed in the fluorescence increase during the oxidation course. In one case (oxidized directly by Ag⁺), 188% fluorescence enhancement took place. In the other case, 126% enhancement (oxidized by H₂O₂) was observed. As there was no other suspected species except for mainly Ag(0) during the redox process, the plausible cause was explained to be the interaction between oxidized Au₂₅ and Ag(0). The phenomenon was described using the MEF effect. However, XPS analyses could not detect Ag(0) species due to facile oxidation of Ag(0) and capping agent could not inhibit such oxidation under the experimental conditions.

2.6. Silver effect

Silver effect is one of the probable reasons for the huge fluorescence enhancement of AuAg bimetallic clusters, as reported by Zhang et al.⁷³ They followed the idea of Wang et al.⁷⁴ where it was explained that the silver effect makes the AuAg bimetallic particle a more efficient catalyst than the Au particle.

Two opposing opinions, however, were found regarding the “silver effect” in fluorescence. Such an effect could increase⁷⁵ and decrease⁷⁶ the luminescence intensity of bimetallic AuAg NCs with respect to Au NCs.

3. Templates for synergistic fluorescence enhancement

Over the last decade, researchers have been actively involved in discovering various templates for synthesizing gold and silver NCs. A lot of materials including polymers, DNA, various thiolates etc. have been found to be efficient for the synthesis of Au and Ag NCs, individually. However, to have the synergistic effect of Au and Ag for the enhancement of fluorescence, only a few templates have been reported to date.

Glutathione is mostly used for the synthesis of strongly fluorescent AuAg bimetallic clusters. Ganguly et al.^20,34,60,77 achieved success for Ag(I)@Ag(0) fabrication using glutathione via UV irradiation, sunlight exposure and a modified hydrothermal (MHT) protocol. They also used S-lactoylglutathione for synthetic purposes.⁶⁰ They were able to prepare the bimetallic NCs by employing green chemical approaches.⁶⁰ Dou et al.⁴⁴ also reported glutathione as a template for AuAg NCs, exhibiting AIE. Zhang et al.²⁴ presented the rapid microwave-assisted technique for the synthesis of strongly luminescent AuAg nanoclusters using glutathione. Le Guével et al. also used glutathione for the synthesis of AuAg bimetallic clusters.⁴⁹ By adding Ag⁺ to the pre-synthesized AuGSH, they made such bimetallic clusters. Again, Wu et al.⁶⁹ used glutathione to show oxidation-interaction induced fluorescence enhancement for bimetallic clusters.

An interesting Ag₇Au₆ alloy NC⁵⁷ was made by employing interfacial etching of mercaptosuccinic acid (MSA)-stabilized Ag nanoparticles (NPs) and a galvanic exchange reaction. However, the synthetic procedure demands toxic toluene and multistep washings via centrifugation. Zhou et al.³⁵ also prepared highly fluorescent AgAu alloy NCs, protected with MSA by core etching of Ag NPs and a galvanic exchange reaction without any organic solvent as well as multistep centrifuge washing. Again, to withstand hostile environments the MSA–AgAu NCs were derivatized by poly(ethylene glycol) (PEG) using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) chemistry.^78,79 Such PEGylated MSA–AgAu NCs exhibited efficient fluorescence enhancement in the presence of Al³⁺ [Fig. 7].

Fig. 7 Schematic representation for the formation of AuAg NCs that show fluorescence enhancement in the presence of Al³⁺. Reprinted with permission from (ref. 35). © 2013 American Chemical Society.

Ristig et al.⁴⁸ used 11-mercaptoundecanoic acid (11-MUA) to synthesize fluorescent AuAg alloy by co-reduction of gold and silver with NaBH₄ in aqueous medium and 11-MUA caps the particle simultaneously. Paramanik and Patra⁸⁰ prepared highly efficient blue bimetallic AuAg alloy emitter clusters using an easy one-pot bottom-up approach with 11-MUA. They also presented an alternative top-down protocol for the formation of the same bimetallic clusters.

A bimetallic AuAg alloy nanocluster was synthesized on the macroscale via sequential reduction employing simple mortar grinding. In this report, chitosan biopolymer was used for both stabilization and reducing purposes.⁸¹

Kumara and Dass⁸² reported Au_38−nAg_n(SCH₂CH₂PH)₂₄, synthesized involving two steps. The first step was the formation of a crude product containing polydisperse AuAg clusters, demonstrated by Negishi et al.⁸³ The next step involved the thermo-chemical treatment of the crude product in the presence of excess thiol to form Au_38−nAg_n(SCH₂CH₂PH)₂₄ alloy nanomolecules.

Mohanty et al.⁸⁴ reported the preparation protocol of luminescent AuAg alloy quantum clusters (QCs) using protein template, bovine serum albumin (BSA). They mixed the as-synthesized protein-protected Au and Ag clusters to form AuAg alloy clusters within BSA. Mass-spectrometric analyses of the AuAg alloy quantum clusters from a 1 : 1 molar ratio reaction mixture of Au_QC@BSA and Ag_QC@BSA indicated the alloy clusters to be Au_38−xAg_x@BSA. Zhang et al.⁷³ also prepared bimetallic AuAg NCs, using BSA as the protein stabilization and reduction agent. They used HAuCl₄, AgNO₃ and NaOH for synthetic purposes (no preformed gold or silver clusters) and a one-pot biomineralization synthetic route.

Chen et al.²¹ reported a one-pot synthetic route of fluorescent DNA–AuAg NCs employing HAuCl₄, AgNO₃, and NaBH₄ (reducing agent) in the presence of 5′-CCCTTAATCCCC-3′, used as a template. They described the relative molar ratios of Au³⁺ to Ag⁺ and [NaBH₄] to make stable and emissive DNA–AuAg NCs.

4. Oxidation states

The oxidation states of the gold and silver in bimetallic clusters are 0 or +1, depending upon the nature of the template and the experimental conditions. In such emissive bimetallic clusters, Au(III) precursors become partially or completely reduced to 0 or +1 [Table 2]. However, Ag(I) precursors may or may not be reduced. A tabular representation regarding oxidation states of Au and Ag in bimetallic clusters has been introduced, as reported by different groups.

Table 2 Oxidation states of Au and Ag in different bimetallic AuAg clusters

Material	Oxidation state of Au	Oxidation state of Ag	Binding energy of Au 4f7/2 (eV)	Binding energy of Ag 3d5/2 (eV)	Reference
MSA–AgAu NCs	0, +1	0	84.7	368.1	Zhou et al.^35
DNA–AgAu NCs	0, +1	0, +1	83.6, 85.0	368.7	Chen et al.^21
MUA–AgAu NCs	0, +1	0	85.03	367.54	Paramanik and Patra^80
GSH–AgAu NCs	+1	+1	84.4	367.9	Dou et al.^44
BSA–AgAu NCs	0	0	84.1	368.2	Mohanty et al.^84
GSH–AgAu NCs	0, +1	+1	83.8	367.9	Zhang et al.^24
MSA–AgAu NCs	0, +1	0	84.7	368.1	Udayabhaskararao et al.^57
BSA–AgAu NCs	0	0, +1	84.1	367.8	Zhang et al.^73
GSH–AgAu NCs	+1	0	85	368.5	Ganguly et al.^23

---

5. Geometry and electronic structure

Although the gold–silver alloyed cluster is much superior to gold or silver clusters, the electronic structure of the alloyed cluster is still a question of intense investigation. Malola and Hakkinen⁸⁵ used DFT (density functional theory) to understand the electronic structure and bonding in Au_144−xAg_x(SR)₆₀ on the basis of the structural model of the icosahedral Au₁₄₄(SR)₆₀ that contains a metal core of 114-atom and 60 symmetry-equivalent surface sites and 30 RSAuSR units protect the core. The optimized structures of the considered clusters (1–5) of composition Au_144−xAg_x(SR)₆₀ are as follows [Fig. 8(A)].

Fig. 8 (A) Optimized structures of the considered clusters 1–5 of composition Au_144−xAg_x(SH)₆₀ (see the main text for the detailed descriptions) brown, gold; gray, silver; yellow, sulfur. The hydrogen atoms are not shown for clarity. (B) Formation energy per metal atom, E_form, of clusters 1–5. The smallest value indicates optimal formation of cluster 4. Adapted with permission from (ref. 85). © 2011 American Chemical Society.

(1) Au₁₄₄(SR)₆₀ [written as Au₁₁₄(RSAuSR)₃₀] in the “Divide and Protect” scheme.

(2) Ag₅₄Au₆₀(RSAuSR)₃₀ where the inner 54-atom Mackay icosahedron of the metal core are formed by the Ag atom;

(3) Au₅₄Ag₆₀(RSAuSR)₃₀ contains 60 Ag atoms, randomly populated in both the Mackay icosahedron core and the anti-Mackay surface layer (30 each);

(4) Au₅₄Ag₆₀(RSAuSR)₃₀ contains anti-Mackay surface sites of the metal core of 60 Ag; and

(5) Au₁₁₄(RSAgSR)₃₀, where in the thiolate layer, Ag substitutes Au.

However, it is the energetically optimal cluster that possesses the lowest (positive) formation energy. The minimal value of E_form comes from cluster 4 (∼0.07 eV/metal atom) and the energy value is ∼ half of the E_form of Au₁₄₄(SR)₆₀ [Fig. 8(B)]. Thus, Au and Ag like to be in a separate atom shell in the core. The theoretical investigations also indicate that the intensity of Au(5d)–Au(6sp) and Ag(4d)–Ag(5sp) interband transitions in the plasmonic region depends on the mixing ratio of Au to Ag. The lowest energy feature in the absorption spectra is due to gold-based transitions while the contribution of silver becomes dominant at higher energies of excitation. For the case of smaller nanoclusters, Ag₁₃Au₁₂(SH)₁₈⁻¹, where all silver atoms are present in the core, bestowed more low-energy “superatomic” P–D transitions than all-gold Au₂₅(SR)₁₈⁻¹ as reported by Aikens.⁸⁶

Again, Negishi et al.⁸⁷ discovered the technique to form stable anionic Au_25−xAg_x(SR)₁₈⁻¹ clusters and the maximal x was found to be 12. The affinity of silver to be populated at the surface-site of the metal core in the well-investigated structure of Au₂₅(SR)₁₈⁻¹ is also known.

A theoretical prediction of the structure of the Au₆Ag₇(SR)₁₀ (R = CH₃) cluster, protected by mercaptosuccinic acid, was performed by Tlahuice-Flores.⁴⁷ Using a DFT approach, he followed an isoelectronic replacement of seven Au atoms by Ag atoms on the [Au₁₃(SR)₁₀]⁺ cluster. With a target in mind to point out the lowest energy isomers, he revealed the most plausible structure to be an octahedral Ag₆ core that is covered by two monoatomic dimer motifs and one Au₂Ag₁(SR)₄ staple-like motif. Ag atoms prefer to stay at the inner (core) portion and the surface has Au atoms attached by staple-like motifs.

Zhang et al.⁸⁸ performed a first-principles investigation of an Ag-doped gold cluster. A stable Au₂₀ cluster with a distinct electronic structure can be manipulated by including Ag. Consequently, the gap between the HOMO and LUMO of Au_20−nAg_nclusters is remarkably altered with the electronic states of Ag in the HOMO and LUMO. The imaginary part of the dielectric function implied that the Ag atom shifts the HOMO–LUMO transition of Au clusters towards the lower energy range and it is related to the concentration of Ag.

Ground state geometries of Au_20−nAg_nwere calculated by Zhang et al.⁸⁸ Electronic and optical properties of Au NCs are greatly related to the stability of the NCs. The binding energy/atom [E_b(n)] of Au NCs was calculated to predict the cluster-stability using the following equation.

[E_b(n)] = [E(Ag) + nE(Au) − nE(Au_nAg)]/(n + 1)

E(Ag), E(Au), and E(Au_nAg) stand for the total energies of the most stable Ag, Au, and AuAg clusters, respectively.

All of the clusters prefer the lowest spin state. The 2.40 eV binding energy of Au₂₀ from the study of Zhang et al.⁸⁸ is very close to the report of Idrobo et al.⁸⁹ The increased size of gold clusters is related to the increase in binding energy.⁹⁰ Enhanced binding energy denotes improved stability because of increased core electron configurations. Owing to the incorporation of Ag in Au NCs, the obtained binding energies of Au₁₉Ag₁, Au₁₈Ag₂, Au₁₇Ag₃, and Au₁₆Ag₄ became 2.68, 2.68, 2.68, and 2.69 eV, respectively. The increased Ag atom caused little perturbation in the binding energy. However, the binding energy of AuAg alloy was higher in comparison to Au₂₀. Thus, the introduction of Ag atom improves the structural stability. Again, the Ag–Au bond is stronger in comparison to the Au–Au bond, contributing an additional stability for the σ-bonding interaction that occurs due to the overlap between the valence Au 6s (5d) orbital and vacant Ag 4p orbital.

The electronic structure of Au_20−nAg_n clusters can be predicted by the density of states (DOS) calculated by Zhang et al.⁸⁸ The Au₂₀ cluster possesses a large HOMO–LUMO gap.^89,91 However, the theoretical band gap of Au₂₀ was found to be 1.47 eV (0.31 eV fewer than the experimental result of 1.78 eV) for the DFT induced underestimation of electronic states.⁹² The Au d states (in the range of −6 and 0 eV) are dominated in HOMO, consisting of s and d states. The DOS alters with the value of n. The band gap of pure Au₂₀ clusters decreases with the incorporation of Ag indicating the effect of Ag in the band gap. It was found that 1.42, 1.34, 1.40, and 1.68 eV were the band gaps of Au₁₉Ag₁, Au₁₈Ag₂, Au₁₇Ag₃ and Au₁₆Ag₄, respectively.⁸⁸ The shift of LUMO (of Au₁₉Ag₁, Au₁₈Ag₂, and Au₁₇Ag₃) towards the lower energy region reduces the band gap owing to the change of electronic properties. Successive incorporation of Ag in the Au₂₀ clusters causes significant p states in LUMO, resulting in the shift of LUMO and narrowing of the band gap. On the other hand, improvement of structural stability and enclosing the electronic configurations describes the increase of the band gap for the case of Au₁₆Ag₄.⁹³ Meanwhile, the p and s states are obviously enhanced in the LUMO, and these electronic states can have effects on the optical properties.

So, Ag insertion greatly perturbs the HOMO–LUMO transition of pure gold clusters. Such a deviation in the energy level can be explained in two stages. Firstly, for Au₁₉Ag₁ and Au₁₈Ag₂, both HOMO and LUMO are contributed by Ag electronic states. Again, LUMO is slightly shifted towards the low energy region, responsible for the decrease of the transition level. Secondly, for Au₁₇Ag₃ and Au₁₆Ag₄, increasing the number of Ag atoms creates more dispersive d states lifting LUMO to the high energy region and the band gap alters. The broadening of the band gap for Au₁₆Ag₄ implies improvement in the structural stability and binding energy [Fig. 9].⁹³

Fig. 9 (A) Calculated ground state geometries of Au_20−nAg_n, (B) the partial DOS of (a) Au_20−nAg_n clusters, (C) the outline of optical transition of Au_20−nAg_n clusters calculated by energy level. Ref. 88 is an open access article distributed under the Creative Commons Attribution License (CC BY) which permits unrestricted use, distribution, and reproduction in any medium.

6. Application

AgAu bimetallic clusters have many practical applications.

6.1. Sensing

6.1.1. Metal ion. Pollution owing to heavy metals has become an increasing problem worldwide with harmful effects on human health as well as the environment.⁹⁴

In contrast to the wide applicability of silver compounds in numerous industries (photography, electronics, optical instruments, mirrors, etc.), Ag⁺ is termed as a toxic heavy metal. Ag⁺ disables sulphydryl enzymes, explained by the HSAB theory, being amassed in the body. It has been found that ∼2500 tons of Ag⁺ is released annually into the environment through industrial wastes and emissions. Amongst these, 150 tons are released into the sludge of waste-water while 80 tons are released into surface waters together with other contaminants. So, selective detection of Ag⁺ is an important issue in modern research.^95,96

Mercury, one of the most hazardous heavy metals, is expansively found in air, water, soil, and food.⁹⁷ The accumulation of Hg in the human body causes several problems including damage to the digestive, excretory, and central nervous systems. As a result, a variety of serious diseases such as tremors, deafness, arthritis, loss of muscle coordination/sensation/memory, motor disorders etc. occur. Again, copper ions carry out pivotal roles in many essential biological activities including metabolism, growth, and immune system development.⁹⁸ Copper deficiency is thus a cause of various illnesses. Besides, a high concentration of copper in tissues also creates lethal effects. Long-standing contact of high [Cu²⁺] causes cellular toxicity, damaged kidney and liver.^99,100 Therefore, it is of great interest to develop a simple, rapid, and highly sensitive detection method for mercury and copper ions.

Aluminum is the most abundant metal in the earth’s crust and is used expansively in our daily life as well as industrially. However, excess Al³⁺ affects plant performance, the growth of fish, and causes a number of diseases (Parkinson’s disease, Alzheimer’s disease etc.) in humans. So, easy recognition of Al³⁺ is an important issue now-a-days.^101–104

Pb²⁺ is one of the highly poisonous heavy metal cations. Ingestion of even very small quantities causes numerous health hazards namely anaemia, memory loss, and slow nerve conduction velocity especially in children. A report by the CDC (Centers of Disease Control) reveals that ≥100 mg L⁻¹ (0.48 mM) [Pb] in the blood is deleterious to children.^105,106 Thus, sensitive detection of Pb²⁺ is essential.

By employing the phenomenon that synergism of Au and Ag enhances fluorescence, Wu et al.⁶⁹ showed intriguing fluorescence enhancement of Au₂₅(SG)₁₈ in the presence of Ag⁺ and designed a Ag⁺ sensor. They explained it using oxidation and interaction induced fluorescence enhancement (discussed earlier). A good linear relationship existed between [Ag⁺] and the emission intensity (20.2 nm to 11.1 μm). A detection limit of ∼200 nM was obtained by their protocol. It is satisfactorily lower than the maximum allowable level (MAL) of Ag⁺ (460 nm) in drinking water as required by the U.S. Environmental Protection Agency (EPA).

Zhang et al.⁷³ designed a fluorimetric technique for the separate recognition of Hg²⁺ and Cu²⁺ ions using BSA protected AuAg NCs. The fluorescence quenching of AuAg NCs with Hg²⁺ ions at different concentrations is demonstrated in Fig. 10. A plot of logarithmic concentration of Hg²⁺ vs. fluorescence quenching efficiency indicates that Hg²⁺ ions can be sensed over a linear concentration range of 0.20 to 2500 nM (R² = 0.9903) and LOD (limit of detection) of 0.10 nM (3σ rule). Likewise, a linear correlation for Cu²⁺ was reported from 0.50 nM to 2500 nM (R² = 0.9960) and a LOD of 0.30 nM. Structural analysis from TEM imaging for such bimetallic AuAg NCs with/without Cu²⁺ or Hg²⁺ ions and in the presence of EDTA revealed insignificant changes in the size and morphology for the case of Hg²⁺. Fluorescence microscopy indicated that quenched fluorescence with Hg²⁺ ions does not alter with the introduction of EDTA. On the contrary, Cu²⁺ ions induce the aggregation of the protein template of Au–Ag NCs. Then, EDTA causes a deaggregation effect on the resulting mixture with the restoration of the fluorescence. The Hg²⁺ induced quenching fluorescence of AuAg NCs was explained by the interaction between Hg²⁺ ions and Au of AuAg NCs, resulting in metallophilic bonding of their 5d¹⁰ centers.^75,107 On the contrary, histidyl and carboxyl groups of the protein scaffold of AuAg NCs react with Cu²⁺ ions, forming the protein–Cu²⁺ adduct.¹⁰⁸ Cu²⁺stimulated protein–protein cross-linking is responsible for aggregation of the protein scaffold. Amino acid residues of the protein scaffolds can also reduce Cu²⁺ to Cu⁺ ions, originating metallophilic actions of 3d¹⁰(Cu⁺)–4d¹⁰(Ag⁺) with the fluorescence quenching. EDTA, being a strong chelating ligand, possesses a higher affinity for copper than the protein–Cu²⁺complex.¹⁰⁹ Thus, the deaggregation of the protein–Cu²⁺complex takes place with the restoration of the vanished fluorescence of AuAg NCs. However, strong metallophilic Au–Hg²⁺ interaction wins over Hg²⁺–EDTA chelation with the permanent loss of fluorescence in the presence of Hg²⁺. This method was also employed to detect Hg²⁺ and Cu²⁺ selectively in blood samples down to 0.30 nM and 0.60 nM levels.

Fig. 10 Fluorescence spectra of Au–AgNCs (0.417 mM) upon the addition of (A) Hg²⁺ ions (0, 0.20, 1.0, 5.0, 25, 125, 625, 1250, 2500 nM) and (C) Cu²⁺ ions (0, 0.5, 1.0, 5.0, 25, 125, 625, 1250, 2500 nM) at λ_ex 370 nm, corresponding to fluorescence quenching efficiencies versus the logarithmic concentrations of (B) Hg²⁺ ions and (D) Cu²⁺ ions in water (insert: photographs under UV light). Reprinted with permission from (ref. 73). © 2014 American Chemical Society.

A strongly fluorescent Au(I)_core–(Ag₂/Ag₃)_shell giant cluster, reported by Ganguly et al.,⁶¹ loses its emissive property in the presence of Hg²⁺. Along with fluorescence quenching, red shifting of the emission maxima is also noticed due to Hg²⁺ [Scheme 2]. d¹⁰(Hg²⁺)–4d¹⁰(Ag^δ+) metallophilic interaction and aggregation of the silver clusters were ascribed to the observed fluorescence quenching phenomenon and red shifting, respectively. Cu(II) and Fe(III) also quench the fluorescence of giant clusters. Na₂–EDTA and ammonium bifluoride were employed to overcome Cu(II) and Fe(III) induced quenching. Thus, a highly selective Hg(II) sensor was designed. A linear correlation was found between (I₀ − I)/I₀ and [Hg(II)] over the range 0–10 μM (R² = 0.98) while the LOD [at an S/N ratio of 3 for Hg(II)] was reported to be 6 nM. It is much lower than the permissible level of the United States Environmental Protection Agency (EPA).¹¹⁰ This method was also found to be useful for the estimation of Hg²⁺ in real environmental samples.

Scheme 2 Giant fluorescent Au(I)_core–(Ag₂/Ag₃)_shell clusters as a highly selective sensing platform for Hg(II). Ref. 34 © 2014 Royal Society of Chemistry. Reproduced from ref. 34 with permission from The Royal Society of Chemistry.

The photoluminescence intensity of microwave-assisted glutathione capped AuAg bimetallic clusters was reported to be reduced successively with the gradual increase of the concentration of Cu²⁺ ion.²⁴ No other metal ion bestowed any significant quenching, helping to design the Cu²⁺sensor. The linear correlation between I/I₀ [I and I₀ are the luminescence intensity with and without Cu²⁺ ion] and [Cu²⁺] exists in the range 0 to 100 nM (y = 1–0.00467x, R² = 0.996) with a limit of detection (LOD) of 2 nM (S/N = 3).

According to the report of Zhou et al.,³⁵ the PEGylated MSA–AgAu NCs can be employed extremely selectively as well as a sensitive sensing platform in an aqueous medium of Al³⁺ (detection limit of 0.8 μM that is much lower than the permitted level 7.4 μM of World Health Organization, WHO).¹¹¹ Introduction of Al³⁺ to the MSA–AgAu NCs causes aggregation of the particle as revealed by DLS analysis. The hydrodynamic diameter changed from 15 to 72 nm with 50 μM Al³⁺. To offer more chelating sites with Al³⁺ and completion with the ligand shell of metal NCs, free succinic acid (SA) or MSA was added in the reaction mixture. However, no significant reduction in the fluorescence intensity and hydrodynamic diameters of PEGylated MSA–AgAu NCs took place calling into question the Al³⁺ mediated aggregation induced fluorescence enhancement. Zhou et al.³⁵ explained the fluorescence enhancement due to the AlAu alloy formation. XPS, DLS, and ζ-potential were employed to support the formation of such an alloy. In other words, deposition of Al³⁺ on the surface of the bimetallic core is the main factor for fluorescence enhancement and selective Al³⁺ sensing.

Glutathione capped Au(I)_core–(Ag₂/Ag₃)_shell giant cluster¹¹² loses its fluorescence selectively in the presence of DMSO that replaces glutathione. Pb(II) binds DMSO because of the strong affinity between Pb(II) and S. Pb(II) brings back the lost fluorescence by removing DMSO from the surface of the giant clusters and glutathione takes its original position. Thus, a highly selective and sensitive Pb²⁺ sensor [Scheme 3] was made by Ganguly et al.¹¹² Like Pb(II), Ag(I) and Hg(II) possess a strong affinity towards the sulfur of DMSO.¹¹³ But, extra Ag(I) ion destabilizes the fluorescent silver clusters via aggregation and Hg(II) forms an amalgam with silver. Thus, Ag(I) and Hg(II) could not regenerate the lost fluorescence (due to DMSO). The fluorescence property is re-established only in the presence of Pb(II). A linear correlation between I₀/I [I₀ and I are the fluorescence intensity before and after the addition of Pb²⁺] and [Pb²⁺] exists in the range 0–20 μM and the LOD was found to be 2 × 10⁻⁷ M. However, EDTA again quenches the Pb²⁺ induced fluorescence by forming a strong chelating complex with Pb²⁺.

Scheme 3 Schematic representation of DMSO induced fluorescence quenching, Pb²⁺ induced regaining of fluorescence of Ag(I)@Au(0)–GSH. Na₂–EDTA eliminates the Pb²⁺ induced enhanced fluorescence. Reprinted with permission from (ref. 112). © 2014 American Chemical Society.

6.1.2. Anion. Sulfide anion (S²⁻), a toxic component, is available considerably in various industrial areas like tanneries, petroleum refineries, food processing plants, and paper/pulp manufacturing plants. Here sulfide ion acts frequently as a reactant. S²⁻ ion is also generated as a by-product for different manufacturing/industrial activities. For living organisms, S²⁻ is very dangerous. It also works to remove the dissolved oxygen and produces hydrogen sulfide in aqueous medium. The S²⁻ induced toxicity is noteworthy in humans due to exposure in various occupational settings. A continuous escalation of [S²⁻] in the human body causes various physiological discomfort and biochemical difficulties. Again, protonation further increases its toxicity. Along with health problems, degradation of concrete and corrosion of metal surfaces occur with excess [S²⁻].^114–118

To detect the toxic and corrosive S²⁻, Ganguly et al.⁷⁷ made highly fluorescent faint yellow colored Au(I)@Ag particles via a modified hydrothermal technique in glutathione (GSH) matrix. Here, tiny Ag₂, Ag₃ clusters embedded on Au(I) are the origin of fluorescence. Due to the smaller size associated with a high diffusing capability, S²⁻ selectively covers the surface of Au(I)@Ag surface generating an orange coloration and removes GSH from the fluorescent particle. Due to a higher reducing property of S²⁻ than GSH, Au(I) is reduced to Au(0) causing the aggregation of silver clusters and eventually destroys the fluorescence. Thus, naked eye and fluorimetric detection of S²⁻ become possible. In fluorimetric sensing via selective quenching, I⁻ and S₂O₃²⁻ interfere as they are also quenchers under the experimental conditions. Pb²⁺ removes interference by causing fluorescence enhancement in the case of I⁻ and S₂O₃²⁻. However, S²⁻ induced quenching remains unaffected in the presence of Pb²⁺. A linear correlation was found between I₀/I (I₀ and I are the fluorescence intensities of the solutions containing Au(I)@Ag particles without and with S²⁻, respectively) and [S²⁻] through the range of 0–10 μM (R² = 0.985) and the LOD was 10 nM (S/N ratio of 3). Not only in water but also in water miscible solvents, sulfide detection is possible by this regime. Ganguly et al. also investigated the effect of straight chain organothiols with different chain lengths in fluorescence quenching and found that with the increase of chain length of thiols, the quenching capability decreases. This phenomenon supports the strong quenching by sulfide to be related to its smaller size and higher diffusing power [Fig. 11].

Fig. 11 (A) Fluorescence spectral profile and (B) relative fluorescence intensity (I = fluorescence intensity of alkanethiol-passivated HFL and I₀ = fluorescence intensity of HFL) showing the effect of chain length of alkane thiols on the fluorescence quenching of HFL. Conditions: 0.5 mL HFL diluted to 3 mL with distilled water, [alkanethiol] = 2.5 × 10⁻⁴ M, λ_ex = 390 nm. Ref. 77 © 2015 Royal Society of Chemistry. Reproduced from ref. 77 with permission from The Royal Society of Chemistry.

Chen et al.²¹ synthesized DNA–AuAg NCs employing HAuCl₄, AgNO₃, NaBH₄ and 5′-CCCTTAATCCCC-3′. S²⁻ ions quenches the emission of the DNA–AuAg NCs due to the interaction between the Au/Ag atoms/ions and S²⁻ ions. The solubility products (K_sp) of Au₂S and Ag₂S are reported to be 1.58 × 10⁻⁷³ and 8.0 × 10⁻⁵¹ M², respectively.^119,120 Owing to the formation of Au₂S and Ag₂S, the DNA–AuAg NCs are broken down and the DNA scaffolds are converted to random coiled structures. The emission at 630 nm of such DNA–AuAg NCs is quenched with the successive introduction of S²⁻ ions. Two linear regions 0–0.01 μM and 0.01–9 μM were found (I⁰_F − I_F)/I⁰_F [I⁰_F and I_F are the fluorescence intensity before and after the addition of S²⁻] versus [S²⁻]. The LOD for sulfide in this process has been reported to be 0.83 nM.

6.1.3. Water miscible solvent. Dimethyl sulfoxide (DMSO) is a familiar organic solvent as well as a natural product. DMSO is found in foods and beverages (wine, coffee, tea etc.) even at the micromolar level. Dimethyl sulfide (DMS) is a reductive product of DMSO and inauspicious for foods/beverages because of its offensive smell.^121,122 So, selective and sensitive DMSO sensing is important.

Green chemically synthesized fluorescent Au(I)@(Ag₂/Ag₃)–thiolate core–shell particles¹¹² (synthesized from HAuCl₄, AgNO₃ and GSH under the Sun) were used for selective detection of dimethyl sulfoxide (DMSO) via fluorescence quenching. No other water-miscible solvent causes quenching of fluorescence. DMSO removes capped GSH and GSSG from the particle surface resulting in fluorescence quenching. DMSO can be recognized even when DMSO molecules are 4.09 × 10⁻⁵ times the number of water molecules in a solution. With the increasing [DMSO], the fluorescence of the fluorescent particles is monotonously decreased until DMSO : H₂O [molar ratio] = 0.01 : 1. DMSO sensing can be performed in water when the molar concentration of water is 2.4 × 10⁴ to 1 × 10² (a wide range) times higher than [DMSO] and the LOD is 178 ppm. When [DMSO]/[H₂O] (molar ratio) > 0.01, further increase of [DMSO] causes fluorescence enhancement along with a continuing red shift. [DMSO]/[H₂O] (molar ratio) = 0.08 causes maximum fluorescence intensity with a red shift of ∼29 nm. Then, emission intensity is reduced with the further increase of [DMSO] in water.

6.1.4. Amino acid. Cysteine (Cys) carries out vital roles in physiological functions such as protein synthesis, metabolism and detoxification by its involvement in the pathways of reversible redox reactions.^123–127 It is also employed widely in several pharmaceutical synthetic routes including antibiotics and medicines for skin lesions. Scarcity of Cys creates many problems, e.g., hematopoiesis decrease, leukocyte loss, psoriasis, etc. Alzheimer’s disease and Parkinson’s disease are also favored by changes in the Cys level in the human body. Thus, selective and sensitive recognition of Cys is vital.

Green chemically synthesized, GSH capped Au(I)@Ag₂/Ag₃ core–shell giant cluster was used as a highly selective and sensitive Cys sensor in the solution phase.¹²⁸ Giant clusters can have robust hydrophilic moieties where the adsorbate monolayer is accumulated with the sterically demanding –GS group.¹²⁹ When Cys is introduced, GSH (molecular weight 307 Da) is replaced via place exchange reaction by Cys (an amino acid of low molecular weight, ∼121 Da). Such a place exchange reaction is facilitated by the release of steric crowding, destroying the fluorescence property of the giant cluster. The place exchange reaction is also favored by the strong penetrating power of Cys. A report by Tehrani et al.¹³⁰ demonstrated that GSH is prone to forming intramolecular hydrogen bonds with noble metals, contributing to the long-term stability of the fluorescent silver clusters. However, Cys is unable to produce such hydrogen bonding and a non-fluorescent solution is generated due to the removal of GSH by Cys from the cluster surface. A linear correlation was found between (I₀ − I)/I₀ (I₀ and I is the fluorescence intensity of giant cluster without and with Cys, respectively) and [Cys] in the range of 50 nM to 10 mM and the LOD was 50 nM.

6.2. Cytotoxicity and bioimaging

Barcikowski et al.¹³¹ investigated biological studies with ligand-free AuAg nanoalloys and found an enhanced cytotoxicity with a higher silver content. They correlated this phenomenon with released [Ag⁺] that is possibly increased due to noble metal gold, making a local electrochemical element in the presence of Ag.

It is found that the enhanced dose of silver in alloyed AgAu nanoparticles does not cause a monotonous decrease in cell viability. Importantly, the nanoparticles with Ag : Au = 90 : 10 were reported to be more toxic in comparison to pure silver nanoparticles (Ag : Au = 100 : 0). 11-MUA capped AuAg clusters, synthesized by Ristig et al.,⁴⁸ reported a similar observation. Again, Barcikowski et al.¹³¹ demonstrated a similar fact for alloyed silver–gold nanoparticles, obtained from laser ablation in mammalian gametes. Ristig et al.⁴⁸ also incubated the HeLa cells in their 11-MUA capped AuAg clusters (Ag : Au = 30 : 70 and Ag : Au = 50 : 50) for 24 h in serum supplemented medium and found red emitters are located at the cytoplasm. It is an efficacious internalization of bimetallic particles inside the cells. The red emissive dots in the cells are not separate nanoparticles. They are due to the gathering of nanoparticles in endosomal vesicles [Fig. 12 and 13].

Fig. 12 MTT assays of HeLa cells after incubation with different concentrations of alloyed silver–gold nanoparticles for 24 h (A) and 72 h (B). [Ref. 48] – published by The Royal Society of Chemistry.

Fig. 13 CLSM images of HeLa cells after incubation with Ag : Au 30 : 70 (A) and Ag : Au 50 : 50 (B) nanoparticles for 24 h in serum supplemented medium. The nanoparticles have a red colour and the cell cytoskeleton (actin) is stained green. [Ref. 48] – © The Royal Society of Chemistry.

MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay of hydrothermally synthesized fluorescent Au(I)@Ag particles (Au : Ag = 10 : 17) showed that the particles are bio-compatible.⁷⁷ The optical study on both primary [human fibroblast (foreskin origin) cells] and cancer cell (MG-63 cells) lines indicate that for primary cells, fluorescent particles have an inclination to accrue in the cytoplasm. On the other hand, the fluorescent particles are likely to be accumulated inside the nucleus for cancer cells. This observation was enlightened using the electric properties of a cancer cell. The cytoplasm of a cancer cell has a paucity of electrons in comparison to a primary/normal cell. Due to the positive zeta potential of the emissive particles (+11.9 mV), the cytoplasm repels the particles strongly in a cancer cell unlike the primary cell. This study increases the opportunity of bimetallic nanoclusters to be used as nanoprobes for the in vitro diagnostics of cancer cells employing a traditional biopsy strategy, related to fluorescence microscopy.

6.3. SERS

Surface-enhanced Raman scattering (SERS) is a surface-sensitive method by which molecules exhibit a higher Raman scattering after being absorbed on to metal surfaces. According to electromagnetic theory (EM), localized surface plasmon in the presence of a metal surface is responsible to SERS. On the other hand, chemical theory describes a charge transfer complex for SERS. The extent of SERS is estimated using the enhancement factor (EF) with the following expression.¹³²

EF = (I_SERS/N_ads)/(I_bulk/N_bulk) [I_SERS = the intensity of a vibrational mode in the surface enhanced spectrum, I_bulk = intensity of the Raman spectrum of same vibrational mode in the bulk material, N_ads = number of adsorbed molecules on the SERS-active substrate, and N_bulk = the number of sampled molecules in the bulk material].

The bimetallic AuAg alloy nanoclusters have also been employed for SERS experiments. As the surface area of NCs is significantly higher in comparison to the corresponding NPs/bulk materials, a greater degree of absorption on the cluster surface takes place, causing SERS. Paramanik and Patra⁸⁰ showed the MUA protected AuAg nanoclusters in the context of SERS when the probe is methylene blue (MB).

In the Raman spectra of MB, strong peaks for C–C stretching of the ring carbon (1618 cm⁻¹) and peaks for the symmetric, asymmetric stretching and bending of the C–N–C skeleton (1395 cm⁻¹, 1450 cm⁻¹ and 445 cm⁻¹) were found¹³³ When AuNCs are added to MB (non-fluorescent), an insignificant Raman signal is obtained at 1618 cm⁻¹ or 445 cm⁻¹. Interestingly, AuAg alloy clusters cause remarkable SERS enhancement with an EF (enhancement factor) of 1.44 × 10⁶ for MB. MB dye adsorbs on the surface of AuAg clusters, concentrating the local electromagnetic field. A study by Paramanik and Patra⁸⁰ revealed that AuAg alloy NCs have a higher potential for SERS activity in comparison to pure Au NCs.

6.4. Antibacterial activity

Antibacterial activity from noble metals is well-known and related to the morphology of the particles.^60,134 Due to bacteriostatic and bactericidal effects, noble metal particles are extensively employed for antimicrobial surfaces on textiles, fibers, and polymers. They are also used for the dressing of wounds and burns, microbial resistant catheters, surgical masks and tools, tissue engineering scaffolds, etc. Moreover, for the production of antimicrobial paints, water purification, and effluent treatment membranes, the antibacterial property of noble metals is gaining in significance day-by-day.

Fluorescent giant clusters [Au(I)_core–(Ag₂/Ag₃)_shell], prepared by Ganguly et al.,⁶⁰ were impregnated in cotton and paper to produce fluorescent cotton and paper with antibacterial activity. Paper substrate and cotton wool containing fluorescing giant clusters exhibited decent antibacterial activity against both Gram-positive (S. aureus) and Gram-negative (E. coli) bacterial strains. Clear zones were found around the fluorescent cotton wool and paper substrate when the disk diffusion assay method was performed. The fluorescent particle treated cotton wool/paper showed a decrease in the bacterial number for both Gram-positive and Gram-negative bacteria. It was reported that the fluorescent cotton wool caused 99.91% reduction in viability of the Gram-negative bacteria after 3 h of incubation and 98.6% was obtained for the Gram-positive bacteria after 3 h. It was concluded that fluorescent cotton wool, impregnated with giant clusters, was marginally less effective for Gram-positive bacteria than Gram-negative bacteria within the initial 3 h of incubation [Fig. 14].

Fig. 14 Antibacterial activity of the fluorescent cotton wool in agar plates for E. coli (a) and S. aureus (b). Antibacterial activity of the fluorescent paper substrate using agar plates for E. coli (a1) and S. aureus (b1). Reprinted with permission from (ref. 60). © 2013 American Chemical Society.

6.5. Reduction of nitrophenol

A trustworthy benchmark reaction is 4-nitrophenol reduction in the presence of metal catalyst. Murugadoss et al.⁸¹ synthesized gold silver alloy NCs employing HAuCl₄, AgNO₃, chitosan and NaOH by simple mortar grinding. This NC was found to be an efficient catalyst for the reduction of 4-nitrophenol (4NP). Reduction of 4NP by NaBH₄ has been considered to be a model reaction to judge the catalytic property of noble metal particles.¹³⁵ Without metal catalyst, the reaction is remarkably slow. Murugadoss et al.⁸¹ monitored reaction kinetics spectrophotometrically. The absorbance of 4-nitrophenolate ions at 400 nm decreased with time, when their NCs were introduced in the reaction mixture. Excess NaBH₄ was used to make reduction kinetics pseudo first-order with respect to the 4NP and linear plots were obtained for ln(C_t/C₀) vs. time [Fig. 15]. During the catalysis an induction period was found. An induction period is the necessary time for the adsorption of 4NPs on the catalyst surface. Interestingly, the induction period is gradually enhanced with increasing Au content in the bimetallic cluster. The highest rate constant for the reduction of 4NP is 8.41 × 10³ s⁻¹ L mol⁻¹, when NCs contained equal amounts of Au and Ag. From Arrhenius plots for this NCs (Au/Ag = ∼1) catalyst, it was shown to have a significantly low activation energy. The extraordinary catalytic property of this AuAg alloy NCs is because of the existence of the Au/Ag interface at the surface of the NC, helping faster adsorption of the 4NP on the catalyst surface.^136,137

Fig. 15 (a) Decay of the logarithmic absorbance of 4NP as a function of time for different NC catalysts. (b) Arrhenius plots for reactions catalyzed by Ag and Au–Ag NCs. Ref. 81 © 2012 Royal Society of Chemistry. [Ref. 81] – reproduced by permission of The Royal Society of Chemistry.

6.6. Surface enhanced fluorescence

An efficient substrate was synthesized by Dong et al.¹³⁸ to produce surface-enhanced fluorescence. It was made of Ag/Au bimetallic nanostructures on the surface of copper. Multi-stage galvanic replacement reaction was employed at room temperature with Ag cluster in the chloroauric acid (HAuCl₄) solution. The bimetallic cluster on the Cu surface bestowed a higher degree of surface enhanced fluorescence when rhodamine 6G acted as a fluorophore and deposited as a probe on the substrate surface. The growth of nanostructure as well as fluorescence enhancement was found to be intensely connected. By tuning the reaction conditions (reaction time, and concentration of reaction solutions), they established the optimum conditions for the maximum emission.¹³⁸

7. Superiority of AuAg bimetallic clusters

Fluorescent silver clusters are unstable due to oxidation and aggregation. Thus, they lose their fluorescing nature rapidly with time. They are also incapable of withstanding heat, UV exposure etc. In other words, they are not robust. Moreover, Ag⁺ present in the Ag NCs is toxic limiting their biological applications. Fluorescent gold clusters are less bright than silver clusters. Again, preparation of pure Au NCs is an expensive process. So, to have real practical applications, pure Au NCs are not very promising.

Researchers are also involved in the synthesis of carbon dots.^139,140 The synthesis of carbon dots is cheap and they are mostly cytocompatible as the precursor compounds are cytocompatible. However, the main problem is that in most of the synthetic strategies the particle size of the carbon dots cannot be controlled properly and the quantum yield is poor.¹⁴¹ Semiconductor quantum dots^142,143 are known for their strong luminescent property. However, high toxicity associated with severe health hazards limits their in vivo applications.

Bimetallic AuAg NCs are superior to individual Au NCs or Ag NCs due to their high emissive behavior and photostabilty.⁴⁹ The toxicity of AuAg NCs can be controlled by altering the ratio of Au and Ag in the bimetallic clusters.⁴⁸ They are very robust. They can be vacuum dried to obtain fluorescent solids and can be reversibly transferred in solid and liquid phase.⁶⁰ Usually, fluorescent Ag NCs/Au NCs change the emission maximum with variation of the solvent.¹⁴⁴ However, Au(I)_core–(Ag₂,Ag₃)_shell giant clusters are virtually unchanged in terms of their emission maxima with solvent variation. Different extents of fluorescent enhancement were observed in different water-miscible solvents in comparison to water [Fig. 16]. The enhancement was probably due to release of self-quenching and hydrogen bonding. Not only in water miscible solvents, but also in water immiscible solvents, Au(I)_core–(Ag₂,Ag₃)_shell giant clusters can be prepared using the water pool of the AOT reverse micelle.⁶⁰ Such giant clusters can also be impregnated in cotton wool and paper to obtain fluorescent cotton and paper with antibacterial activity. Bimetallic clusters are also superior to monometallic clusters in the context of different useful applications such as sensing, surface enhanced Raman scattering, catalytic activity etc.

Fig. 16 Fluorescence spectra of the fluorescing solution containing giant clusters in different water miscible organic solvents. Reprinted with permission from (ref. 60). © 2013 American Chemical Society.

Zhou et al.¹⁴⁵ examined the excited state dynamics of a highly luminescent cluster [Au₆Ag₂(C) (dppy)₆](BF₄)₄ in three different solvents (CH₂Cl₂, CH₃CN, CH₃OH). They employed femtosecond transient absorption spectroscopy associated with quantum chemical calculations in this context. It is a Au(I)–Ag(I) cluster with hypercoordinated carbon. The cluster possessed an octahedral Au₆ core and two Ag atoms. The two Ag atoms cap the two opposite Au₃ triangles and six peripheral diphenylphoshpino-2-pyridine (dppy) ligands. The cluster was strongly emissive in the solid state as well as the solution phase. Solvents of altered polarity and hydrogen bond formation capability (CH₂Cl₂, CH₃CN, CH₃OH) were used to monitor the excited state dynamics and the electronic structure. The shift of the steady state absorption spectra in the visible range was observed with alteration of the solvent systems. Interestingly, the emission wavelength was unaltered [Fig. 17 and 18]. Actually, the carbon hypercoordinated Au(I)–Ag(I) cluster exhibits an ultrafast intersystem crossing rate (1–3 ps) and solvent dependent energy transfer. This is the reason for different emissive behaviors. Again, the population of various states and the phosphorescence behaviors of the carbon hypercoordinated clusters can be manipulated with the variation of solvent.

Fig. 17 Crystal Structure of [Au₆Ag₂(C) (dppy)₆]⁴⁺ (A) and simplified optimized structure of ground state [Au₆Ag₂(C) (PH₂py)₆]⁴⁺ (B). Au atoms are colored yellow, Ag atoms green, N atoms blue, C atoms gray, P atoms orange, H atoms light gray. Reprinted with permission from (ref. 145). © 2015 American Chemical Society.

Fig. 18 (A) Normalized steady state absorption spectra of [Au₆Ag₂(C)-(dppy)₆](BF₄)₄ dissolved in three solvents, where the concentrations were adjusted to be the same (∼8 × 10⁻⁵ mol L⁻¹) for all measurements. (B) Luminescence spectra of [Au₆Ag₂(C)-(dppy)₆](BF₄)₄ with excitation of 350 nm in three solvents. Reprinted with permission from (ref. 145). © 2015 American Chemical Society.

8. Summary and outlook

To produce cost-effective commercially available AuAg bimetallic clusters, employment of the synergism of Au and Ag has become a recent endeavor. Such clusters can easily bypass the weak fluorescence of Au clusters. Ag clusters are more fluorescent than Au clusters, while Ag clusters are unstable due to oxidation and aggregation. Photostability is also an issue of cluster science for achieving biological utility. Enhanced emission/improved quantum yield, photo-stability, withstanding power in various environments are bestowed on the bimetallic clusters. Tuning of the activation color and emission intensity are intensely related to the Au and Ag content in the bimetallic clusters. The ratio of Au to Ag in the bimetallic clusters imparts not only the emissive property of the clusters, but also the toxicity. Though Ag is known to be more toxic than Au, an optimum ratio of Au and Ag for the highest toxicity has been determined. Few-atom bimetallic clusters and giant clusters are equally important in terms of their applicability. However, synthetic protocols of the former demands strong reducing agents (NaBH₄) while preparation of the latter requires mild reaction conditions with no NaBH₄ at all. Employment of NaBH₄ to the so-produced giant clusters destroys fluorescence with the evolution of a plasmon band of the nanocrystals due to the facile aggregation of nuclei. Though solvents influence greatly the fluorescing property of the bimetallic clusters, non-aqueous water miscible solvents only increase the quantum yield of the giant clusters keeping the emission wavelength virtually intact. Synthesis in water immiscible solvents using the water-pool of the reverse micelle widens the horizon for applications. Electronic and band structures are significantly perturbed with the gradual incorporation of Ag in Au clusters, bestowing many extraordinary features of the bimetallic clusters. Bimetallic clusters are not only useful in fluorescence based applications such as sensing, bio-imaging, metal enhanced fluorescence etc., but are also efficient for SERS activity, catalysis, bactericidal effect, etc. Extensive research to introduce the bimetallic clusters for in vitro, in vivo and ex vivo applications is still required. The exact mechanism of synergism has been unclear to date. A number of mechanisms have been proposed. However, unambiguous mechanisms, based on direct experimental evidence, have not yet been established. These limitations prohibit the applications of AuAg clusters, and need to be tackled to solve many practical problems. Tuning of the size and shape as well as the morphology may be an important topic of research for the days to come. Along with experimental chemistry, thought provoking theoretical knowledge has to be augmented; mandatory to a better understanding of the system. Theories with some specific systems are of course reported. However, these are insufficient to estimate the general physico-chemical behavior of the bimetallic clusters. Incorporation of the scaffolds in the theory is vital in this context, although these are too complicated with many variables. Soft interaction between the templates and clusters enhances emission. Thus, DNA wins over thiols in a lot of cases. However, use of oligonucleotides for the synthesis of bimetallic AuAg clusters is scarcely reported in the literature. Hopefully, DNA based bimetallic clusters will begin an era of cluster science in the near future. The magic of the bimetallic clusters, with or without magic numbers of nuclearity, may be a remarkable asset for upcoming areas of investigation.

References

1. http://www.en.wikipedia.org/wiki/Epistasis.
2. A. Wang, X. Y. Liu, C.-Y. Mou and T. Zhang, J. Catal., 2013, 308, 258–271.
3. J.-H. Liu, A.-Q. Wang, Y.-S. Chi, H.-P. Lin and C.-Y. Mou, J. Phys. Chem. B, 2005, 109, 40–43.
4. D. Nagaonkar and M. Rai, Adv. Mater. Lett., 2015, 6, 334–341.
5. Y. C. Shiang, C. C. Huang and H. T. Chang, Chem. Commun., 2009, 3437–3439.
6. H. Wei, Z. D. Wang, L. M. Yang, S. L. Tian, C. J. Hou and Y. Lu, Analyst, 2010, 135, 1406–1410.
7. L. Shang and S. J. Dong, Biosens. Bioelectron., 2009, 24, 1569–1573.
8. C. C. Huang, C. T. Chen, Y. C. Shiang, Z. H. Lin and H. T. Chang, Anal. Chem., 2009, 81, 875–882.
9. J. Yu, S. A. Patel and R. M. Dickson, Angew. Chem., Int. Ed., 2007, 46, 2028–2030.
10. A. Retnakumari, S. Setua, D. Menon, P. Ravindran, H. Muhammed, T. Pradeep, S. Nair and M. Koyakutty, Nanotechnology, 2010, 21, 055103.
11. C. I. Richards, S. Choi, J. C. Hsiang, Y. Antoku, T. Vosch, A. Bongiorno, Y. L. Tzeng and R. M. Dickson, J. Am. Chem. Soc., 2008, 130, 5038–5039.
12. J. H. Yu, S. Choi and R. M. Dickson, Angew. Chem., Int. Ed., 2009, 48, 318–320.
13. C. A. J. Lin, T. Y. Yang, C. H. Lee, S. H. Huang, R. A. Sperling, M. Zanella, J. K. Li, J. L. Shen, H. H. Wang, H. I. Yeh, W. J. Parak and W. H. Chang, ACS Nano, 2009, 3, 395–401.
14. X. L. Guevel, B. Hotzer, G. Jung, K. Hollemeyer, V. Trouillet and M. Schneider, J. Phys. Chem. C, 2011, 115, 10955–10963.
15. J. Zheng, P. R. Nicovich and R. M. Dickson, Annu. Rev. Phys. Chem., 2007, 58, 409–431.
16. H. Xu and K. S. Suslick, Adv. Mater., 2010, 22, 1078–1082.
17. M. Ganguly, C. Bradsher, P. Goodwin and J. T. Petty, J. Phys. Chem. C, 2015, 119, 27829.
18. J. Xie, Y. Zheng and J. Y. Ying, J. Am. Chem. Soc., 2009, 131, 888–889.
19. J. Zheng, J. T. Petty and R. M. Dickson, J. Am. Chem. Soc., 2003, 125, 7780–7781.
20. Z. Luo, X. Yuan, Y. Yu, Q. Zhang, D. T. Leong, J. Y. Lee and J. Xie, J. Am. Chem. Soc., 2012, 134, 16662–16670.
21. W.-Y. Chen, G.-Y. Lan and H.-T. Chang, Anal. Chem., 2011, 83, 9450–9455.
22. C.-C. Huang, Z. Yang, K.-H. Lee and H.-T. Chang, Angew. Chem., Int. Ed., 2007, 46, 6824–6828.
23. M. Ganguly, A. Pal, Y. Negishi and T. Pal, Langmuir, 2013, 29, 2033–2043.
24. J. Zhang, Y. Yuan, Y. Wang, F. Sun, G. Liang, Z. Jiang and S.-H. Yu, Nano Res., 2015, 8, 2329.
25. W. A. de Heer, Rev. Mod. Phys., 1993, 65, 611–676.
26. M. Harb, F. Rabilloud, D. Simon, A. Rydlo, S. Lecoultre, F. Conus, V. Rodrigues and C. Felix, J. Chem. Phys., 2008, 129, 194108.
27. K. Clemenger, Phys. Rev. B: Condens. Matter Mater. Phys., 1985, 32, 1359–1362.
28. J. T. Petty, S. P. Story, J.-C. Hsiang and R. M. Dickson, J. Phys. Chem. Lett., 2013, 4, 1148–1155.
29. T. P. Bigioni, R. L. Whetten and O. Dag, J. Phys. Chem. B, 2000, 104, 6983–6986.
30. J. Zheng, C. Zhang and R. M. Dickson, Phys. Rev. Lett., 2004, 93, 077402.
31. S. Link, Z. L. Wang and M. A. El-Sayed, J. Phys. Chem. B, 1999, 103, 3529–3533.
32. D. Mahl, J. Diendorf, S. Ristig, C. Greulich, Z. A. Li, M. Farle, M. Koeller and M. Epple, J. Nanopart. Res., 2012, 14, 1153.
33. D. Tiedemann, U. Taylor, C. Rehbock, J. Jakobi, S. Klein, W. A. Kues, S. Barcikowski and D. Rath, Analyst, 2014, 139, 931–942.
34. M. Ganguly, C. Mondal, J. Pal, A. Pal, Y. Negishi and T. Pal, Dalton Trans., 2014, 43, 11557–11565.
35. T. Y. Zhou, L. P. Lin, M. C. Rong, Y. Q. Jiang and X. Chen, Anal. Chem., 2013, 85, 9839–9844.
36. S. Choi, R. M. Dickson and J. Yu, Chem. Soc. Rev., 2012, 41, 1867–1891.
37. W. Guo, J. Yuan and E. Wang, Chem. Commun., 2009, 3395–3397.
38. M. Yu, C. Zhou, J. Liu, J. D. Hankins and J. Zheng, J. Am. Chem. Soc., 2011, 133, 11014–11017.
39. L. Shang, S. Dong and G. U. Nienhaus, Nano Today, 2011, 6, 401–418.
40. X. Yuan, Z. Luo, Y. Yu, Q. Yao and J. Xie, Chem.–Asian J., 2013, 8, 858–871.
41. D. M. Chevrier, A. Chatt and P. Zhang, J. Nanophotonics, 2012, 6, 064504.
42. J. Yu, S. A. Patel and R. M. Dickson, Angew. Chem., Int. Ed., 2007, 46, 2028–2030.
43. B. Sengupta, K. Springer, J. G. Buckman, S. P. Story, O. H. Abe, Z. W. Hasan, Z. D. Prudowsky, S. E. Rudisill, N. N. Degtyareva and J. T. Petty, J. Phys. Chem. C, 2009, 113, 19518–19524.
44. X. Dou, X. Yuan, Y. Yu, Z. Luo, Q. Yao, D. T. Leong and J. Xie, Nanoscale, 2014, 6, 157–161.
45. Y. Yu, Q. Yao, Z. Luo, X. Yuan, J. Y. Lee and J. Xie, Nanoscale, 2013, 5, 4606–4620.
46. P. Maity, S. Xie, M. Yamauchi and T. Tsukuda, Nanoscale, 2012, 4, 4027–4037.
47. A. Tlahuice-Flores, Phys. Chem. Chem. Phys., 2014, 16, 18083–18087.
48. S. Ristig, D. Kozlova, W. Meyer-Zaika and M. Epple, J. Mater. Chem. B, 2014, 2, 7887–7895.
49. X. Le Guével, V. Trouillet, C. Spies, K. Li, T. Laaksonen, D. Auerbach, G. Jung and M. Schneider, Nanoscale, 2012, 4, 7624–7631.
50. S. Kumar, M. D. Bolan and T. P. Bigioni, J. Am. Chem. Soc., 2010, 132, 13141–13143.
51. T. U. B. Rao and T. Pradeep, Angew. Chem., Int. Ed., 2010, 49, 4218.
52. C. I. Richards, S. Choi, J. C. Hsiang, Y. Antoku, T. Vosch, A. Bongiorno, Y. L. Tzeng and R. M. Dickson, J. Am. Chem. Soc., 2008, 130, 5038–5039.
53. J. Sharma, H. C. Yeh, H. Yoo, J. H. Werner and J. S. Martinez, Chem. Commun., 2010, 46, 3280–3282.
54. C. Zhou, C. Sun, M. X. Yu, Y. P. Qin, J. G. Wang, M. Kim and J. Zheng, J. Phys. Chem. C, 2010, 114, 7727–7732.
55. S. A. Patel, M. Cozzuol, J. M. Hales, C. I. Richards, M. Sartin, J. C. Hsiang, T. Vosch, J. W. Perry and R. M. Dickson, J. Phys. Chem. C, 2009, 113, 20264–20270.
56. E. Hutter, J. H. Fendler and D. Roy, J. Phys. Chem. B, 2001, 105, 11159–11168.
57. T. Udayabhaskararao, Y. Sun, N. Goswami, S. K. Pal, K. Balasubramanian and T. Pradeep, Angew. Chem., Int. Ed., 2012, 51, 2155–2159.
58. B. O. Leung, F. Jalilehvand, V. Mah, M. Parvez and Q. Wu, Inorg. Chem., 2013, 52, 4593–4602.
59. L. Maretti, P. S. Billone, Y. Liu and J. C. Scaiano, J. Am. Chem. Soc., 2009, 131, 13972–13980.
60. M. Ganguly, J. Pal, S. Das, C. Mondal, A. Pal, Y. Negishi and T. Pal, Langmuir, 2013, 29, 10945–10958.
61. M. Ganguly, C. Mondal, J. Pal, A. Pal, Y. Negishi and T. Pal, Dalton Trans., 2014, 43, 11557–11565.
62. M. Ganguly, C. Mondal, J. Chowdhury, J. Pal, A. Pal and T. Pal, Dalton Trans., 2014, 43, 1032–1047.
63. M. Ganguly, J. Pal, C. Mondal, A. Pal and T. Pal, Dalton Trans., 2015, 44, 4370–4379.
64. J. Lukomska, J. Malicka, Z. Gryczynski, Z. Leonenko and J. R. Lakowicz, Biopolymers, 2005, 77, 31–37.
65. J. R. Lakowicz, Y. Shen, S. Auria, J. Malicka, J. Fang, Z. Gryczynski and I. Gryczynski, Anal. Biochem., 2002, 301, 261–277.
66. M. Ganguly, A. Pal and T. Pal, J. Phys. Chem. C, 2011, 115, 22138–22147.
67. K. G. Stamplecoskie and P. V. Kamat, J. Am. Chem. Soc., 2014, 136, 11093–11099.
68. A. Baksi, M. S. Bootharaju, X. Chen, H. Hakkinen and T. Pradeep, J. Phys. Chem. C, 2014, 118, 21722.
69. Z. Wu, M. Wang, J. Yang, X. Zheng, W. Cai, G. Meng, H. Qian, H. Wang and R. Jin, Small, 2012, 8, 2028–2035.
70. J. Kang, M. Choi, J. Y. Kwon, E. Y. Lee and J. Yoon, J. Org. Chem., 2002, 67, 4384–4386.
71. M. Schmittel and H. Lin, Inorg. Chem., 2007, 46, 9139–9145.
72. A. S. Al-Kady, M. Gaber, M. M. Hussein and E.-Z. M. Ebeid, J. Phys. Chem. A, 2009, 113, 9474–9484.
73. N. Zhang, Y. Si, Z. Sun, L. Chen, R. Li, Y. Qiao and H. Wang, Anal. Chem., 2014, 86, 11714–11721.
74. D. Wang, R. Cai, S. Sharma, J. Jirak, S. K. Thummanapelli, N. G. Akhmedov, H. Zhang, X. Liu, J. L. Petersen and X. Shi, J. Am. Chem. Soc., 2012, 134, 9012–9019.
75. J. Xie, Y. Zheng and J. Y. Ying, Chem. Commun., 2010, 46, 961–963.
76. J. Sun, H. Wu and Y. Jin, Nanoscale, 2014, 6, 5449–5457.
77. M. Ganguly, J. Jana, B. Das, S. Dhara, A. Pal and T. Pal, Dalton Trans., 2015, 44, 11457–11469.
78. C.-A. J. Lin, R. A. Sperling, J. K. Li, T.-Y. Yang, P.-Y. Li, M. Zanella, W. H. Chang and W. J. Parak, Small, 2008, 4, 334–341.
79. C.-A. J. Lin, T.-Y. Yang, C.-H. Lee, S. H. Huang, R. A. Sperling, M. Zanella, J. K. Li, J.-L. Shen, H.-H. Wang, H.-I. Yeh, W. J. Parak and W. H. Chang, ACS Nano, 2009, 3, 395–401.
80. B. Paramanik and A. Patra, J. Mater. Chem. C, 2014, 2, 3005–3012.
81. A. Murugadoss, N. Kai and H. Sakurai, Nanoscale, 2012, 4, 1280–1282.
82. C. Kumara and A. Dass, Nanoscale, 2012, 4, 4084–4086.
83. Y. Negishi, T. Iwai and M. Ide, Chem. Commun., 2010, 46, 4713–4715.
84. J. S. Mohanty, P. L. Xavier, K. Chaudhari, M. S. Bootharaju, N. Goswami, S. K. Pal and T. Pradeep, Nanoscale, 2012, 4, 4255–4262.
85. S. Malola and H. Hakkinen, J. Phys. Chem.Lett., 2011, 2, 2316–2321.
86. C. M. Aikens, J. Phys. Chem. C, 2008, 112, 19797.
87. Y. Negishi, T. Iwai and M. Ide, Chem. Commun., 2010, 46, 4713–4715.
88. X.-D. Zhang, M.-L. Guo, D. Wu, P.-X. Liu, Y.-M. Sun, L.-A. Zhang, Y. She, Q.-F. Liu and F.-Y. Fan, Int. J. Mol. Sci., 2011, 12, 2972–2981.
89. J. Idrobo, W. Walkosz, S. Yip, S. Öğüt, J. Wang and J. Jellinek, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 205422.
90. X. Li, H. Wang, X. Yang, Z. Zhu and Y. Tang, J. Chem. Phys., 2007, 126, 084505.
91. P. Gruene, D. Rayner, B. Redlich, A. van der Meer, J. Lyon, G. Meijer and A. Fielicke, Science, 2008, 321, 674–676.
92. X. Zhang, M. Guo, W. Li and C. Liu, J. Appl. Phys., 2008, 103, 063721.
93. S. Zorriasatein, K. Joshi and D. Kanhere, J. Chem. Phys., 2008, 128, 184314.
94. X. Fu, T. Lou, Z. Chen, M. Lin, W. Feng and L. Chen, ACS Appl. Mater. Interfaces, 2012, 4, 1080–1086.
95. H. T. Ratte, Environ. Toxicol. Chem., 1999, 18, 89–108.
96. K. Ma, H. Wang, X. Li, B. Xu and W. Tian, Anal. Bioanal. Chem., 2015, 407, 2625–2630.
97. C. Guo and J. Irudayaraj, Anal. Chem., 2011, 83, 2883–2889.
98. R. Uauy, M. Olivares and M. Gonzalez, Am. J. Clin. Nutr., 1998, 67, 952S–959S.
99. M. E. Letelier, A. M. Lepe, M. Faúndez, J. Salazar, R. Marín, P. Aracena and H. Speisky, Chem.–Biol. Interact., 2005, 151, 71–82.
100. L. M. Gaetke and C. K. Chow, Toxicology, 2003, 189, 147–163.
101. E. Delhaize and P. R. Ryan, Plant Physiol., 1995, 107, 315–321.
102. N. E. Alstad, B. M. Kjelsberg, L. A. Vøllestad, E. Lydersen and A. Poléo, Environ. Pollut., 2005, 133, 333–342.
103. D. P. Perl, D. C. Gajdusek, R. M. Garruto, R. T. Yanagihara and C. J. Gibbs, Science, 1982, 217, 1053–1055.
104. D. Crapper, S. Krishnan and A. Dalton, Science, 1973, 180, 511–513.
105. J. S. Lin-Fu, Lead Poisoning: A Century of Discovery and Rediscovery, in Human Lead Exposure, ed. H. L. Needleman, Lewis Publishing, Boca Raton, FL, 1992.
106. Centers for Disease, Control, and Prevention, Preventing Lead Poisoning in Young Children, Atlanta, GA, 2005.
107. B. Yoon, H. Häkkinen, U. Landman, A. S. Wörz, J.-M. Antonietti, S. Abbet, K. Judai and U. Heiz, Science, 2005, 307, 403–407.
108. R. A. Løvstad, BioMetals, 2004, 17, 111–113.
109. M. H. Muhammed, P. K. Verma, S. K. Pal, A. Retnakumari, M. Koyakutty, S. Nair and T. Pradeep, Chem.–Eur. J., 2010, 16, 10103–10112.
110. B. P. Joshi, C. R. Lohani and K.-H. Lee, Org. Biomol. Chem., 2010, 8, 3220–3226.
111. WHO, Guidelines for Drinking-Water Quality, World Health Organization, Geneva, Switzerland, 4th edn, 2011.
112. M. Ganguly, C. Mondal, J. Jana, A. Pal and T. Pal, Langmuir, 2014, 30, 348–357.
113. S. J. Mora, P. Lee, D. Shooter and R. Eschenbruch, Am. J. Enol. Vitic., 1993, 44, 327–332.
114. A. H. Gore, S. B. Vatre, P. V. Anbhule, S.-H. Han, S. R. Patila and G. B. Kolekar, Analyst, 2013, 138, 1329–1333.
115. X. F. Yang, L. P. Wang, H. M. Xu and M. L. Zhao, Anal. Chim. Acta, 2009, 631, 91–95.
116. Z.-X. Wang, C.-L. Zheng and S.-N. Ding, RSC Adv., 2014, 4, 9825–9829.
117. M. I. Prodromidis, P. G. Veitsistas and M. I. Karayannis, Anal. Chem., 2000, 72, 3995–4002.
118. M. M. F. Choi, Analyst, 1998, 123, 1631–1634.
119. T. Morris, H. Copeland and G. Szulczewski, Langmuir, 2002, 18, 535–539.
120. D. V. Goia, J. Mater. Chem., 2004, 14, 451–458.
121. S. J. Mora, P. Lee, D. Shooter and R. Eschenbruch, Am. J. Enol. Vitic., 1993, 44, 327–332.
122. T. W. Pearson, H. J. Dawson and H. B. Lackey, J. Agric. Food Chem., 1981, 29, 1089–1091.
123. Z. A. Wood, E. Schroder, J. R. Harris and L. B. Poole, Trends Biochem. Sci., 2003, 28, 32–40.
124. R. Carmel and D. W. Jacobsen, Homocysteine in Health and Disease, Cambridge University Press, Cambridge, UK, 2001.
125. T. Perez-Ruiz, C. Martinez-Lozano, V. Tomas and G. Lambertos, Talanta, 1991, 38, 1235–1239.
126. S. Shahrokhian, Anal. Chem., 2001, 73, 5972–5978.
127. W.-S. Liao, F.-W. Wu, Y.-M. Wu and X.-J. Wang, Microchim. Acta, 2008, 162, 147–152.
128. M. Ganguly, J. Jana, C. Mondal, A. Pal and T. Pal, Phys. Chem. Chem. Phys., 2014, 16, 18185–18197.
129. T. G. Schaaff, G. Knight, M. N. Shafigullin, R. F. Borkman and R. L. Whetten, J. Phys. Chem. B, 1998, 102, 10643–10646.
130. Z. A. Tehrani, Z. Jamshidi, M. J. Javan and A. Fattahi, J. Phys. Chem. A, 2012, 116, 4338–4347.
131. D. Tiedemann, U. Taylor, C. Rehbock, J. Jakobi, S. Klein, W. A. Kues, S. Barcikowski and D. Rath, Analyst, 2014, 139, 931–942.
132. P. L. Stiles, J. A. Dieringer, N. C. Shah and R. P. van Duyne, Annu. Rev. Anal. Chem., 2008, 1, 601–626.
133. R. R. Naujok, R. V. Duevel and R. M. Corn, Langmuir, 1993, 9, 1771–1774.
134. Z. Lu, K. Rong, J. Li, H. Yang and R. Chen, J. Mater. Sci.: Mater. Med., 2013, 24, 1465–1471.
135. T. Aditya, A. Pal and T. Pal, Chem. Commun., 2015, 51, 9410–9431.
136. Y. Shin, A. Dohnalkova and Y. Lin, J. Phys. Chem. C, 2010, 114, 5985–5989.
137. J. Huang, S. Vongehr, S. Tang, H. Lu, J. Shen and X. Meng, Langmuir, 2009, 25, 11890–11896.
138. J. Dong, Y. Ye, W. Zhang, Z. Ren, Y. Huo and H. Zhenga, Luminescence, 2015, 30, 1090.
139. K. Wang, Z. Gao, G. Gao, Y. Wo, Y. Wang, G. Shenand and D. Cui, Nanoscale Res. Lett., 2013, 8, 122.
140. J. Jana, M. Ganguly and T. Pal, Phys. Chem. Chem. Phys., 2015, 17, 2394–2403.
141. Y. Wang and A. Hu, J. Mater. Chem. C, 2014, 2, 6921–6939.
142. R. Hardman, Environ. Health Perspect., 2006, 114, 165–172.
143. A. M. Derfus, W. C. W. Chan and S. N. Bhatia, Nano Lett., 2004, 4, 11–18.
144. I. Díez, M. I. Kanyuk, A. P. Demchenko, A. Walther, H. Jiang, O. Ikkala and R. H. Ras, Nanoscale, 2012, 4, 4434–4437.
145. M. Zhou, Z. Lei, Q. Guo, Q.-M. Wang and A. Xia, J. Phys. Chem. C, 2015, 119, 14980–14988.

---