Tailored Engineering of Bimetallic Plasmonic Au@Ag Core@Shell Nanoparticles

Abstract

A distinctive synthetic method for the efficient synthesis of multifunctional bimetallic plasmonic Au@Ag core@shell nanoparticles (NPs) with tunable size, morphology, and localized surface plasmon resonance (LSPR) using Triton X-100/hexanol-1/deionized water/cyclohexane-based water-in-oil (W/O) microemulsion (ME) is described. The W/O ME acted as a “true nanoreactor” for the synthesis of Au@Ag core@shell NPs by providing a confined and controlled environment and suppressing the nucleation, growth, agglomeration, and aggregation of the NPs. High-resolution transmission electron microscopic analysis of the synthesized Au@Ag core@shell NPs revealed an “unusual core@shell” contrast, and the selected area electron diffraction and Moiré patterns showed that Au layers are paralleled to Ag layers, thus indicating the formation of Au@Ag core@shell NPs. Interestingly, the UV–visible spectrum of the Au@Ag core@shell NPs exhibited enthralling plasmonic properties by introducing a high-frequency quadrupolar LSPR mode originated from the isolated Au@Ag NPs along with a low-frequency dipolar LSPR mode originated from the coupled Au@Ag NPs. The effective plasmonic enhancement of the Au@Ag core@shell NPs is attributed to the extreme enhancement of the localized electromagnetic field by coupling of the localized surface plasmons of the Au core and Ag shell. The mechanisms for the nucleation and growth of Au@Ag core@shell NPs in W/O ME have been proposed. A unique electron transfer phenomenon between the Au core and Ag shell is elucidated for better understanding and manipulation of the electronic properties, which evinced the development of Au@Ag core@shell NPs through suppression of the galvanic replacement reaction.

Introduction

The emergence of plasmonics in nanoscale materials has renewed interests in the fundamental understanding of localized surface plasmon resonance (LSPR).¹ Novel metal plasmonic nanoparticles (NPs) have been the subject of painstaking research owing to their captivating plasmonic, electronic, and optical properties, including LSPR, surface-enhanced Mie scattering, metal-enhanced photoluminescence, and surface-enhanced Raman scattering.¹⁻³ Among all of the novel metal NPs, gold (Au) and silver (Ag) NPs received the utmost attention in terms of their synthetic refinement and highly fascinating plasmonic properties.⁴ Moreover, Au NPs possess a very high molar extinction coefficient and also a very high resistance to oxidation. On the other hand, Ag NPs possess the highest molar extinction coefficient of any metal but show severe susceptibility to oxidation.⁵ In recent years, new bimetallic heteronanostructured core@shell-type nanomaterials were prepared by combining Au and Ag. They possess higher plasmonic efficiency, superior electromagnetic enhancement, ideal optical properties, and very high molar coefficient of Ag and high biocompatibility, chemical stability and reactivity, and easy surface modification of Au. In view of the multifunctional properties, they can be used in advanced applications.^5,6 In addition to contribution of the properties from the individual components, a highly intriguing synergistic phenomenon was also observed.⁶

Numerous attempts were made to synthesize highly monodisperse Ag@Au core@shell NPs as well as Au@Ag core@shell NPs where the expectation was that the Ag will contribute enhanced plasmonic properties (high molar extinction coefficient), while the Au will impart biomolecular reactivity as well as chemical stability against aggregation and oxidation.⁶⁻¹³ Synthesis of a Ag@Au core@shell-type structure is associated with the problem that while depositing Au on a Ag core by reduction of Au³⁺ ions, Ag from the core is oxidized by the Au³⁺ ions, which is also known as the galvanic replacement reaction.¹⁴⁻¹⁷ The end result is typically quasi-core@shell NPs or hollow-structured core@shell NPs that have deformities such as gaps or holes in the Au shell, formation of homonanostructured alloy, or even complete removal of the Ag core.^7,14−19 Such particles exhibit high suppression of the optical and electronic properties of Ag by the Au shell; further, they are inherently unstable to the outside environment, which makes them nonideal for any applications.⁵ To synthesize highly monodispersed Ag@Au core@shell NPs, the galvanic replacement reaction should be restrained or eradicated, which is an immense challenge.^7,20,21

A promising alternative is the creation of an inverse core@shell structure, Au@Ag, where Au is used as the core and Ag as the shell, which is the key to harness the unique electronic interaction between Au and Ag. The Au core serves as a platform for regulating the overall particle size and also modifies the electronic characteristics of the deposited Ag shell to increase the stability against oxidation and galvanic replacement reaction.^5,20,22 Such a Au@Ag core@shell-type structure has several advantages such as particle dispersity is higher²³ and the LSPR features of the Ag are not likely to be suppressed by an outer Au shell.²⁰ It has been demonstrated that the Ag shell in the Au@Ag core@shell-type structure shows both an enhanced resistance to oxidation and the galvanic replacement reaction by the Au core.^5,20

To date, there have been numerous reports on the synthesis of Au@Ag core@shell NPs,^{1−4,24−35} but to the best of our knowledge, the appropriate mechanism of the formation of core and shell is yet to be clearly understood. In our laboratory, we developed a method by using water-in-oil (W/O) microemulsion (ME) as a template for the synthesis of NPs including core@shell NPs.^34,36 ZnO@Ag NPs prepared in this method were found to exhibit interesting properties.^36a−36e

The reverse micelles stabilized by surfactant and cosurfactant species at the water/oil interface of a microemulsion provide a confined aqueous medium and constitute a nanoreactor, the size of which can be controlled by appropriate choice of the W/O ratio. Nanoparticles of desired sizes can be synthesized by choosing the appropriate size of the nanoreactor.³⁴⁻³⁶ Such nanoreactors offer interesting opportunities not only for forming core@shell NPs with a controllable size and shape but also for stabilizing the resulting NPs by forming steric barriers using the surfactants and cosurfactant species around the NPs in the system.

Herein, an efficacious and versatile W/O ME technique for the successful synthesis of highly stable Au@Ag core@shell NPs has been demonstrated, using a W/O ME system containing reverse micelles as “nanoreactors”. Fascinating plasmonic and optical properties of the Au@Ag core@shell NPs are reported. Hydrodynamic diameter from dynamic light scattering (DLS) measurements at different time intervals in association with high-resolution transmission electron microscopic (HRTEM) images and elemental analyses were critically analyzed to establish the mechanism of the formation of Au@Ag core@shell NPs. An attempt has been made to elucidate the mechanism of the formation of the Au core and Ag shell during the formulation process via the electron transfer phenomenon, which is the key to create high stability and optical activity.

Results and Discussion

Triton X-100-Based Water-in-Oil Microemulsion as Versatile Nanoreactors

The composition of W/O ME is shown in Table S1. As depicted in Figure 1, the Z-average hydrodynamic diameter (D_h) of the synthesized reverse micelles containing W/O ME was around 14 nm, and these were highly uniform and stable.

Figure 1.

Size distribution plot of W/O ME of Triton X-100 synthesized at W_o = 7.21. The insert shows the pictorial representation of the D_h of the nanoreactors.

The surfactant- and cosurfactant-stabilized nanosized reverse micelles containing W/O ME rendered a unique and controlled microheterogeneous environment for the formation of NPs. The reverse micelles containing W/O ME also restrained the particles from agglomeration and aggregation during the exchange of their contents by the fusion-redispersion process^35,37−40 and thus the reverse micelles containing W/O ME were regarded as “nanoreactors”.

The formation of W/O ME depends on some formulation variables,⁴⁰ and these are also salient features for the synthesis of NPs. Among various ionic (cationic and anionic) surfactants, a nonionic surfactant, Triton X-100 was used for the formulation of W/O ME since the headgroup of the surfactant is uncharged.⁴⁰ For the reason that there are no electrostatic charges from the headgroups, the interactions between these nonionic headgroups are dominated by steric and osmotic forces, which make the W/O ME more uniform, stable, and highly relevant for the use as nanoreactors for the synthesis of NPs.⁴⁰ The headgroup of the nonionic surfactant (Triton X-100) does not interact with the synthesized NPs since they are uncharged and thus able to stabilize the synthesized NPs by controlling the growth process. However, Triton X-100 is a single-chain surfactant and it is not able to diminish the surface tension to the ultralow levels required for the formulation of W/O ME. Thus, further, the short-chain alcohol, hexanol-1 was used as a cosurfactant, which helped to further reduce the surface tension between the oily and aqueous phases and also fluidize the surfactant film.⁴⁰ As a result, the entropy of the W/O ME system increased, leading to its thermodynamic stability, which further assisted to enhance the stability of the reverse micelles.⁴⁰ A comparatively short-chain hydrocarbon, cyclohexane was used as oil phase since oils with shorter hydrocarbon chains easily form W/O ME as compared to oils with long hydrocarbon chains. Moreover, the solubilizing potential of cyclohexane for Triton X-100 and hexanol-1 is distinctly precise and the W/O ME forming region is also enriched by cyclohexane.⁴⁰ Thus, Triton X-100/hexanol-1/deionized (DI) water/cyclohexane-based W/O ME as shown in Figure 1 is highly convenient for exploiting as a nanoreactor for the synthesis of NPs.

Synthesis of Heteronanostructured Au@Ag Core@shell Nanoparticles Using Triton X-100-Based Water-in-Oil Microemulsion as Nanoreactors

Noble plasmonic metal Au and Ag have the identical face-centered cubic (fcc) crystal structure. Their lattice constants, Au (0.408) and Ag (0.409), and atomic radii, Au (1.74) and Ag (1.65), are very analogous.^25,33 Considering these factors, it is pertinent to point out that the major advantage of using Au and Ag is that they will unequivocally initiate the formation of an Au@Ag core@shell-type structure at the initial stage with a nonuniform shell. However, by virtue of their similar crystal structure, lattice constant, and atomic radius, they are quite in favor to initiate the galvanic replacement reaction between the metallic Au and aqueous Ag interface after some period, and consequently at the final stage, they may form Au/Ag alloy-type or Au@Ag hollow nanostructured NPs.^4,41−45 Nonetheless, in this study, this predicament was skillfully unraveled by the technique using nanosized reverse micelles containing Triton X-100-based W/O ME for the synthesis of Au@Ag core@shell NPs since the nanosized reverse micelles provided a stable, controlled, and confined environment that favored the formation of Au@Ag core@shell NPs by controlling the nucleation and growth process and precluded the formation of Au/Ag alloy or Au@Ag hollow-structured NPs by suppressing the galvanic replacement reaction.

The single ME reactant addition scheme was employed for the synthesis of Au@Ag core@shell NPs because it promotes more intramicellar reduction, nucleation, and growth and lowered the possibilities of the galvanic replacement reaction between the metallic gold atoms and aqueous silver ions and also controlled the size of the particles. The composition of W/O and the composition of the components used for the synthesis of Au@Ag core@shell NPs are shown in Tables S1 and S2.

The synthesized Au@Ag NPs were characterized using high-resolution transmission electron microscopic (HRTEM) and selected area electron diffraction (SAED) pattern analyses, which unveiled a distinct core and shell contrast and also an exceptional feature known as Moiré pattern. The HRTEM images of the synthesized Au@Ag core@shell NPs taken at different magnifications for demonstrating better contrast are shown in Figure 2.

Figure 2.

HRTEM micrographs of Au@Ag core@shell NPs synthesized in W/O ME by selecting different zones.

Fairly uniform spherical particles are identified. The HRTEM images as depicted in Figure 2 revealed an “unusual core@shell” contrast in which a dark Au core is located in the central portion and the bright Ag shell is on the outer portion occupying the periphery of the core.

The bottom middle image is of one particle, showing a darker central zone and a lighter outer zone, with a diameter around 20 nm. A clear boundary between the Au core and Ag shell has been distinguished by the outer brighter, denser layer and the central darker, sparser contrast in the HRTEM images. Additionally, the boundary between the Au core and Ag shell is sharp; this suggests that the Ag shell is pure Ag. The difference in atomic number and thus attenuation of electrons of Au and Ag atoms are the main factors that provided sufficient contrast to distinguish the Au core and Ag shell.^3,24,25 The plasmon excitation efficiency of Ag NPs is more pronounced than that of Au NPs, which is also responsible for the unusual core@shell contrast. It is evident from the images that a majority of the core@shell NPs are almost uniform and spherical in morphology.

To further support the core@shell structure of Au@Ag bimetallic NPs, field emission scanning electron microscopy–energy dispersive X-ray spectroscopic (FESEM–EDS) analysis was performed (Figure S1). FESEM–EDS measurements showed that the core is made up of elemental gold and the peripheral region is made up of silver atoms, thus indicating the Au@Ag core@shell structure. From selecting a zone or an area of FESEM image of Au@Ag core@shell NPs, it was clearly seen that the Au atoms were mostly concentrated in the center of the NPs, whereas the Ag atoms were situated on the circumference of the NPs. This clearly indicates the formation of an almost homogeneous core@shell structure of the synthesized Au@Ag heteronanostructures (Table S3).

Interestingly, the HRTEM images of the Au@Ag core@shell NPs as shown in Figure S2 exhibited variable contrasts within each NP, which is attributed to the Moiré pattern that allowed us to characterize in more detail the crystal interface between the core and shell lattices in Au@Ag NPs.⁴⁶ The core@shell structure of the Au@Ag core@shell NPs was vividly revealed by the Moiré pattern of the core region of the particle formed by overlap of the core and the shell.⁴⁷

In general, Moiré patterns originate from translational or rotational interference. The formation of the Moiré pattern in the Au@Ag core@shell NPs as shown in Figure S2 also indicates that the core and the shell are two highly crystalline materials with slightly different lattice constants that are rotated relative to each other by a small angle. The formation of these patterns is also attributable to the difference in the plane interval between the Au core and Ag shell.⁴⁸ However, most of the Au@Ag NPs were characterized by a stripe-like Moiré pattern. Stripe-like patterns are generally associated with the superposition of two lattices with the same cell parameter in one direction and a slight difference in another and there is a slight rotation of the NPs from the exact zone axis.^46,49

The SAED pattern of Au@Ag core@shell NPs as shown in Figure S3 illustrated the appearance of the Debye–Scherrer ring pattern in the SAED pattern that clearly indicated the polycrystalline nature of the Au@Ag core@shell NPs. The SAED pattern exhibited Debye–Scherrer rings corresponding to the fcc structure originated from the Au core and the Ag shell including the 110 and 111 planes. Au and Ag lattices are irresolvable because of the close lattice constants of Au and Ag (0.408 and 0.409 Å for Au and Ag, respectively).⁴⁹ The SAED pattern is crucial in resolving the structure of the Au@Ag core@shell NPs. The thick Debye–Scherrer rings from the SAED pattern come from the core and shell structure of the NPs, which indicated the formation of Au@Ag core@shell NPs.⁵⁰ Thus, the Ag shell is grown on the Au core.

The Z-average D_h of Au NPs was around 82 nm and of Ag NPs was around 105 nm as shown in Figure 3. Moreover, the Z-average D_h of synthesized Au@Ag core@shell NPs was around 115 nm. The increase in the Z-average D_h of the synthesized Au@Ag core@shell NPs with respect to the Au NPs indicated the formation of a Ag shell on the Au core. The correlogram of the synthesized Au@Ag core@shell NPs is shown in Figure S4a. The baseline of the correlogram is smooth and straight, which indicates the absence of larger particles and sedimentation in the solution of the Au@Ag NPs. The sharp decay of the correlation functions also indicates that the particles are smaller in size and nearly monodisperse. The size distribution by volume of the synthesized Au@Ag NPs is given in Figure S4b. The polydispersity index (PdI) and standard deviation of Au@Ag core@shell NPs are shown in Table S4. The PdI value was 0.497, which suggested nearly monodispersity of the synthesized Au@Ag core@shell NPs.

Figure 3.

Size distribution plot of Au, Ag, and Au@Ag core@shell NPs synthesized in W/O ME. The insert shows a pictorial representation of the Z-average D_h of the Au@Ag core@shell NPs.

It may be noted that the Z-average D_h of the synthesized Au@Ag core@shell NPs is the diameter of the reverse micellar nanoreactor in which it is formed. The diameter determined by HRTEM is the actual diameter of the metallic NPs because the small molecular size organic compounds, such as Triton X-100, hexanol-1, and cyclohexane, are electron-transparent and therefore they did not show up in the HRTEM micrographs and thus only provided the average true diameter of the NPs. However, dynamic light scattering (DLS) provided the average D_h of the NPs and so the average particle size was found to be smaller in HRTEM compared to that by DLS (Scheme S1).^51a

Another factor behind the larger particle size in the case of DLS measurements compared to HRTEM is that the DLS principle emphasizes more on the interactions during the dynamic condition. The particle size explicated by using DLS theory emphasized the aspect that the particles underwent random interparticle attraction due to the van der Waals and plasmonic interactions and also through interparticle repulsion due to the electrostatic and steric interactions as illustrated in Scheme S1.^51a The HRTEM images provide the “true diameter” of the particles though determined on statistically small samples.

The synthesized Au@Ag core@shell NPs were analyzed using powder X-ray diffraction (PXRD) to study the distribution of elements within the Au@Ag NPs and to analyze the crystalline nature of the synthesized Au@Ag NPs. The PXRD pattern and the deconvoluted PXRD pattern of the synthesized Au@Ag NPs are shown in Figure 4a,b.

Figure 4.

(a) PXRD pattern and (b) deconvoluted PXRD pattern of the synthesized powdered Au@Ag core@shell NPs.

The deconvoluted PXRD pattern of the synthesized Au@Ag core@shell NPs as depicted in Figure 4b shows sharp peaks at 2θ = 27.85, 32.25, 38.16, 44.31, 46.23, 54.85, 57.46, 64.59, 76.86, 77.55, 81.74, and 85.71°, and these peaks correspond to the lattice planes (111), (200), (111), (200), (220), (311), (222), (220), (300), (311), (222), and (422) respectively, consistent with the characteristic of Au@Ag core@shell NPs.^51b−51j The diffraction peaks at 27.85, 32.25, 38.16, 44.31, 46.23, 54.85, 57.46, 64.59, 76.86, 77.55, 81.74, and 85.71° are indexed with the (111), (200), (111), (200), (220), (311), (222), (220), (300), (311), (222), and (422) planes, respectively, of metallic fcc Au (JCPDS No. 04-0784) and Ag (JCPDS No. 04-0783).^51b−51j Interestingly, the Au@Ag core@shell NPs showed significant peaks at 32.25 and 38.16° for Au and Ag. Some weak peaks were also observed for the fcc Au and Ag. Metallic Au and Ag have similar fcc crystal structures, lattice constants, and atomic radii. Thus, the PXRD diffraction peak positions of Au and Ag are similar, and Au could not be distinguished from Ag by this PXRD characterization, which is in good agreement with the literature data.^51b−51j

The PXRD pattern confirmed the existence of fcc Au and Ag in the Au@Ag core@shell NPs. In the PXRD pattern of the Au@Ag core@shell NPs, the Au and Ag reflections were quite similar to the standard reflections of the monometallic NPs of Au and Ag. However, some Au and Ag reflections were slightly blue-shifted with respect to the standard reflections of the monometallic NPs of Au and Ag. It was considered that the Ag shell was formed on the Au core, which was the reason behind the slight shift of some of the Au and Ag reflections with respect to the standard values. Thus, it also confirmed the formation of Au@Ag core@shell NPs.

The crystalline behavior of the synthesized Au@Ag core@shell NPs can be analyzed by the PXRD pattern. An intrinsic limitation of PXRD is the fact that amorphous phases cannot be detected but the crystalline phases provide sharp reflection peaks.^51b−51h The sharp PXRD pattern of both Au and Ag thus clearly showed that the synthesized Au@Ag core@shell NPs were polycrystalline in nature. The results are totally in agreement with the SAED pattern.

Enticing Plasmonic Properties of Au, Ag, and Au@Ag Core@Shell Nanoparticles

The absorption spectrum of Triton X-100-based W/O ME at W_o = 7.21 is shown in Figure S5. The absorption spectrum clearly elucidated that the components of the W/O ME exhibited no peak in the 300–800 nm range of the absorption spectrum. The absorption spectrum of Au, Ag, and Au@Ag core@shell NPs after synthesis in W/O ME at W_o = 7.21 is shown in Figure 5. It elucidates the plasmonic properties of the NPs. The unique plasmonic properties of the Au@Ag core@shell NPs are their ability to absorb incident electromagnetic radiation due to localized surface plasmon resonance (LSPR), which is caused by the collective oscillation of conduction electrons at the surface (or more accurately, at the interface between metal nanoparticles and their dielectric surrounding) upon excitation by the incident light.^26,27,52 The LSPR effect can lead to strong confinement of the electromagnetic field and thus great enhancement of the local electric field near the metal surface within a subwavelength distance.⁵² One of the key advantages of plasmonic Au, Ag, and Au@Ag core@shell NPs is their ability to produce engineered multipolar LSPR modes such as dipolar, quadrupolar, octupolar, and hexadecapolar modes.⁵³ The sensitivity of the multipolar LSPR modes is highly dependent on the effect of size, shape, and composition.

Figure 5.

Absorption spectra of Au, Ag, and Au@Ag core@shell NPs synthesized in W/O ME.

The Au NPs with Z-average D_h of around 82 nm exhibited a dipolar LSPR mode at 541 nm as depicted in Figure 5. The Au NPs also generated a higher-order quadrupolar LSPR mode at 324 nm. This is in agreement with previous works.^1,26,28,54 The peaks below 400 nm (216 and 320 nm) were also assigned to the ligand-to-metal charge transfer (π → σ*) Cl pπ → 5d_x²–y² band.⁵⁵ Importantly, the appearance of the quadrupolar and dipolar LSPR modes indicated the formation of Au NPs. Furthermore, the Ag NPs with a Z-average D_h of around 105 nm exhibited a dipolar LSPR mode at 406 nm and a quadrupolar LSPR mode at 295 nm and thus confirmed the formation of Ag NPs. Subsequently, the Au@Ag core@shell NPs with a Z-average D_h of around 115 nm exhibited two bands at 322 and 527 nm as depicted in Figure 5.

This dipolar and quadrupolar LSPR modes are a distinct feature of the dimerlike structure with a nanogap. For a dimerlike system with a nanogap, which is not in physical contact with neighboring NPs, no charge can be completely separated, leading to high-frequency quadrupolar LSPR mode. When the particle contact is made in the overlapping interparticles, at a point that leads to a low-frequency dipolar LSPR mode which gave rise to the charge transfer in the conductive junction of interparticles.^1,56−60 The band at 527 nm for Au@Ag core@shell NPs in the absorption spectrum is attributed to dipolar LSPR mode where the particles are coupled with each other through plasmon coupling and the band at 322 nm is also attributed to quadrupolar LSPR mode originating from the dipole oscillation of the isolated Au@Ag core@shell NPs.^1,56−60

The appearance of the quadrupole mode in the absorption spectrum of Au, Ag, and Au@Ag core@shell NPs is due to the size effect of the NPs and the retardation effects. Spherical plasmonic NPs with an average diameter of around 10–65 nm exhibited only a single dipolar LSPR band, but above this size regime, the phase retardation effects gained more relevance and the higher-order modes became non-negligible and thus started to appear.^53,61,62 In accordance with the results for Au, Ag, and Au@Ag core@shell NPs, the Z-average D_h’s were around 82, 105, and 115 nm, respectively, and thus the phase retardation effects became more and more significant and the quadrupolar LSPR mode started to develop at shorter wavelengths. The general resonance conditions for multipolar modes in the “quasi-static” regime occur for ε = −ε_medium(l + 1)/l; thus, the resonance condition for a quadrupolar mode is ε = −(2/3)ε_medium.⁶¹ When the condition for quadrupolar LSPR mode is fulfilled, a smaller shoulder of the dipolar LSPR mode at a lower wavelength started to manifest. This phenomenon was observed in the case of Au, Ag, and Au@Ag core@shell NPs as shown in Figure 5. As the size of the NPs increases, the contribution from the quadrupolar mode increases. In the case of dipolar LSPR mode for plasmonic Au, Ag, and Au@Ag core@shell NPs, the metal surface is in contact with the dielectric medium of dielectric function, ε_medium. Moreover, in the case of quadrupolar LSPR mode, short charge waves prevent the interactions between distant charges of the dielectric medium; therefore, each small region on the surface of the metal behaves as in a planar bulk metal.⁶¹ This fact can be understood more precisely by perceiving the electric field and charge distribution condition at the surface of plasmonic NPs⁶² as shown in Figure S6.

The higher-order modes, however, need to be considered precisely since in size range larger than 60 nm where these modes are active for plasmonic metallic Au, Ag, and Au@Ag core@shell NPs, the optical properties are additionally influenced by phase retardation effects. One of this retardation effects, the “energy-shifting” effect, arises when the particle size is no longer negligible compared to the wavelength of the incident radiation.⁶¹ When the dimensions of the particles are much smaller than the wavelength of the incident light, such as in the range of 10–67 nm, all electrons in the entire particle experience a roughly uniform electric field, leading to the excitation of the dipolar LSPR mode. As the size increases above 60 nm, the wavelength of the incoming radiation cannot be considered as infinite, thus, the light cannot polarize homogeneously and the field is no longer uniform throughout the NPs, which results in phase retardation effects and starts to develop quadrupolar mode.⁵³ More precisely, these effects appear when the diameter 2R of the particle is around 1/10 of the mode wavelength λ_m = λ/(ε_medium)1/2 of the radiation in the medium surrounding the particle.⁶¹

When a dipolar LSPR mode is excited in spherical plasmonic NPs, the distance between opposite charges can be approximated by the diameter of the NPs. Thus, one side of the particle will experience any charge occurring at the opposite side of the particle with phase retardation equal to 4πR/λ_m.^61,62 The oscillation period of the dipolar LSPR mode increases to take such retardation into account. When quadrupolar LSPR mode is involved, the distance between the opposite charges on the surface of the NPs is smaller than the diameter of the NPs and thus the phase retardation is smaller than in the dipolar LSPR mode.^61,62 This phenomenon is clearly indicated in Figure S6. Indeed, the phase retardation effect in spherical metallic plasmonic Au, Ag, and Au@Ag core@shell NPs scales roughly as 4πR/(l × λ_m) with l = 1, 2, 3, ... for dipole, quadrupole, octupole modes, and so on.

Another phenomenon that is consequential for the emergence of dipolar and quadrupolar LSPR modes is known as the radiation scattering effect.⁶¹ For particles with an average diameter greater than 60 nm, increase in the radiation scattering is observed, as the resonant scattering of the multipole resonance becomes a dominating spectral component. In this size regime, the scattering cross section exceeds their geometric cross section.⁵³ Indeed, the electrons are accelerated as a consequence of the electromagnetic field of the incident radiation.

During irradiation, NPs also start radiating, and this causes them to lose energy.⁶¹ Part of the energy of the plasmonic oscillation is converted into photons. This phenomenon leads to a broadening of the dipolar LSPR band in the absorption spectrum and thus to an apparent decrease in its intensity with respect to the background (mostly given, at the lowest wavelengths, by nonresonant scattering).⁶¹ Thus, in large particles, as opposed to small particles, the interaction with the incoming light is dominated by radiative processes (meaning the photons are scattered).

In pursuance of precisely interpreting a dipolar and quadrupolar LSPR mode phenomenon and the formation of Au@Ag core@shell NPs, deconvolution of the absorption spectrum of Au, Ag, and Au@Ag core@shell NPs is executed. The deconvoluted absorption spectra of Au and Ag NPs are exhibited in Figures S7 and S8. The broad dipolar LSPR mode consisted of bands at 550, 542, and 701 nm, and the broad quadrupolar LSPR mode consisted of bands at 307, 320, 335, and 341 nm. The deconvoluted absorption spectrum of Ag NPs depicted that the broad dipolar LSPR mode consisted of bands at 403, 441, and 603 nm and the broad quadrupolar LSPR mode consisted of bands at 295 and 299 nm.

More interestingly, the deconvoluted absorption spectrum of Au@Ag core@shell NPs as shown in Figure 6 interpreted that the deconvoluted broad dipolar LSPR mode of Au@Ag NPs consisted of bands at 526 and 560 nm, which are blue-shifted with respect to the deconvoluted broad dipolar LSPR mode of Au NPs consisted of bands at 542 and 701 nm. Furthermore, the deconvoluted broad quadrupolar LSPR mode of Au@Ag NPs consisted of bands at 306, 315, 329, and 337 nm, which are also slightly blue-shifted with respect to the deconvoluted broad quadrupolar LSPR mode of Au NPs consisted of bands at 307, 320, 335, and 341 nm. The deconvoluted broad quadrupolar LSPR mode of Au@Ag NPs consisted of bands at 297 and 302 nm, which are slightly red-shifted with respect to the deconvoluted broad quadrupolar LSPR mode of Ag NPs consisted of bands at 295 and 299 nm. The parameters of the deconvoluted spectra of Au, Ag, and Au@Ag core@shell NPs in Figures S7, S8, and 6 are given in Table S5.

Figure 6.

Deconvoluted absorption spectrum of Au@Ag core@shell NPs synthesized in W/O ME at W_o = 7.21 as shown in Figure 5 (fitted using the Gaussian function).

Importantly, the deconvoluted absorption spectra of Au@Ag core@shell NPs as shown in Figure 6 exhibited a band at 406 nm, which is characteristic of Ag and clearly provided the evidence for the formation of a Ag shell on the Au core. Additionally, upon formation of the Ag shell on top of Au NPs, the original dipolar LSPR band of Au NPs at 541 nm gradually blue-shifted toward 527 nm, which was the original dipolar LSPR mode of Au@Ag NPs as shown in Figure 5, and also the deconvoluted broad dipolar LSPR mode of Au NPs consisted of bands at 542 and 701 nm, which were blue-shifted toward 526 and 560 nm that were the bands of the deconvoluted broad dipolar LSPR mode of Au@Ag NPs as shown in Figures S7 and 6. This further indicated the formation of Au@Ag NPs.^{1,3,25,28−30} This spectral feature is due to Ag interband transitions in this spectral region and is inherent to the Ag dielectric properties.^4,24 Ag coating on the Au surface, which resulted in a strong blue shift of the main LSPR, showed not only the deposition of a different metal but also a drastic change in the morphology of the Au@Ag core@shell NPs.²⁴ The gradual blue shift of the LSPR band of Au@Ag core@shell NPs has also been interpreted due to the damping of Au LSPR by the surface of silver atoms.^4,26

The size of the Au, Ag, and Au@Ag core@shell NPs has a dramatic effect on the LSPR and the optical properties of the NPs.⁶² The broadening of the dipolar LSPR band and an apparent decrease in its intensity for Au, Ag, and Au@Ag core@shell NPs were observed along with the appearance of quadrupolar mode, as a consequence of the presence of particles with Z-average D_h greater than around 60 nm. The deconvoluted absorption spectra of Au@Ag core@shell NPs as shown in Figure 6 also correlated well with the fact that due to the presence of particles of average diameter greater than 60 nm, the dipolar LSPR mode becomes broader and further weakens in intensity. During the plasmonic oscillation of these particles, some part of the energy is converted into photons and the interaction with light is dominated by the scattering of photons, which results in the broadening of the dipolar LSPR mode. Moreover, the quadrupolar LSPR mode becomes stronger in intensity and narrower in width as the size of the NPs increases.⁶¹ Therefore, as the size of the NPs increases, the contribution from the quadrupolar LSPR mode increases and as a result the dipolar LSPR mode becomes more influenced by the size of the particles. The dipolar LSPR modes are, therefore, influenced more by the particle size due to retardation effects compared to the quadrupolar LSPR mode.⁶¹

Probable Mechanism for the Formation of Au@Ag Core@Shell Nanoparticles via Single Reactant Microemulsion Addition Scheme

The formation of Au@Ag core@shell NPs in reverse micelles containing W/O ME is a complex process, and the formation mechanism is still not clear. The size distribution at different stages of the formation of Au@Ag nanoparticles via the single reactant ME addition scheme has been thoroughly investigated; moreover, the distribution profiles were critically examined, correlated with HRTEM results in applicable cases, and compared with the available literature to understand the mechanism of nucleation and growth for the synthesis of Au and Au@Ag core@shell NPs in W/O ME-based nanoreactors.

The mechanism of formation of Au NPs in reverse micelles containing W/O ME-based nanoreactors is illustrated in Scheme S2, which involves a three-step process. The first step is considered to involve the reduction and nucleation steps as shown in Scheme S2a. Primarily, the NaAuCl₄ precursor solution was incorporated into the reverse micelles containing W/O ME system and the reactants were directly moved and dissolved into the water pools of the reverse micelles. Thereafter, the NaBH₄ precursor solution was added as the reducing agent into the same W/O ME system containing the NaAuCl₄ precursor solution, which further directly moved into the water pools of the reverse micelles, and both the reactants collided, reacted with each other, and additionally produced Au atoms.

The reverse micelles containing Au atoms were in random Brownian motion that resulted in collision with sufficient energy to lead to the opening of the surfactant and cosurfactant bilayer through which rapid exchange of the Au atoms occurred. The exchange of the Au atoms was too fast and thus the atoms underwent simultaneous nucleation within 1–5 min, which initiated the formation of Au NPs. The size distribution plot as shown in Scheme S3a exhibited one strong size distribution at around 67.93 nm and a weak size distribution at around 8.79 nm.

Consequently, the growth process took place. The second step is the growth of the Au NPs through Ostwald ripening within 6–15 min as illustrated in Scheme S2b. During the random collision and exchange process of the reverse micelles containing Au NPs in the W/O ME system, some Au atoms on the surface of the Au NPs preferentially detach and diffuse in the water core of the nanoreactors in the W/O ME system within 6–10 min as shown in Scheme S2b. This is the first stage of the Ostwald ripening phenomenon. The size distribution plot of the first stage of the Ostwald ripening process within 6–10 min as depicted in Scheme S2b exhibited new size distributions at around 1.31 and 104.77 nm, and the strong size distribution at 67.93 nm was shifted to 33.70 nm from the reduction–nucleation step to the first stage of the Ostwald ripening process. This indicated that due to the random collision and exchange process some Au atoms on the surface of the Au NPs were detached and diffused in the water core of the nanoreactors. Thus, the new size distribution at 1.31 nm appeared because of the diffused Au atoms, the new size distribution at 104.77 nm appeared via aggregation of some of the Au NPs from which the Au atoms were diffused, and the strong size distribution at 67.93 nm was decreased to 33.70 nm via the detachment and diffusion of the Au atoms from the surface of the Au NPs from the reduction–nucleation step to the first stage of the Ostwald ripening process.

The water core of the nanoreactors of the W/O ME system has a larger saturated solubility due to the free Au atoms, and the free Au atoms tend to acquire lower surface energies. Thus, the free Au atoms are redeposited on the surface of the Au NPs to reach a more thermodynamically stable state, which eventually leads to the growth of Au NPs within 11–15 min as shown in Scheme S2b. This is the second stage of the Ostwald ripening phenomenon. The size distribution plot of the second stage of the Ostwald ripening process within 11–15 min as depicted in Scheme S2b showed that the weak size distribution at 1.31 nm disappeared, new size distribution at 5.09 nm appeared, and the strong distribution at 33.70 nm was shifted to 70.52 nm from the first stage to the second stage of the Ostwald ripening process. This indicated that the free Au atoms were redeposited on the surface of the Au NPs, which led to the growth of Au NPs. Thus, the weak size distribution at 1.31 nm due to diffused Au atoms disappeared, the new size distributions at 5.09 nm appeared via the aggregation of the diffused Au atoms, and the strong size distribution at 33.70 nm increased to 70.52 nm via the redeposition of the Au atoms on the surface Au NPs from the first stage to the second stage of the Ostwald ripening process.

The third step is the growth of the particles by the diffusion-controlled growth process as illustrated in Scheme S2c. The size distribution plot exhibited two different types of particles of average D_h’s of around 6.58 and 80.39 nm. The growth of the particles from 5.09 to 6.58 nm and 70.52 to 80.39 nm was considered to occur through self-sharpening growth by diffusion. However, diffusion-controlled growth usually only occurred over a longer period, in this case during 16–20 min, due to the constant reduction of gold that occurs within the solution where the limiting factor is the concentration of gold metal within the solution itself.⁶³

After the synthesis of Au NPs, the Au NPs were used as a core to further synthesize the Au@Ag core@shell NPs as shown in Scheme 1. The first step of the process is considered as a reduction–nucleation process as illustrated in Scheme 1a. At the initial stage, the nucleation process commenced within 1–5 min when the AgNO₃ precursor solution was added into the W/O ME solution containing Au NPs. As evident from the color change and spectroscopic measurements, the Ag⁺ ions reduced into Ag atoms via the galvanic replacement reaction and a Ag shell started to form around the Au NPs. Nevertheless, some major drawbacks are that the rate of nucleation is very slow and if the process is continued further, it will have higher possibilities to form Au/Ag alloy NPs or hollow nanostructured Au@Ag core@shell NPs. To solve the issues, NaBH₄ solution was added into the same W/O ME system. After the addition of NaBH₄ solution, the Ag⁺ ions underwent a fast reduction process and there was a rapid increase in the formation of Ag atoms in the W/O ME system. Afterward the Ag atoms readily underwent “burst nucleation”, which significantly reduced the concentration of free Ag atoms in the W/O ME system and simultaneously increased the rate of formation of a thin Ag shell around the Au core. At this point, the rate of nucleation is described as “effectively infinite” and the nucleation process follows the “LaMer mechanism”.⁶³⁻⁶⁵ The size distribution plot for the nucleation process as shown in Scheme 1a exhibited one weak size distribution at around 2.58 nm and one strong size distribution at around 81.69 nm. This indicated that two different types of particles existed in the ME system. The nonuniformity of the second peak also suggested that at that instant the particles underwent continuous and fast reduction and then rapid nucleation.

Scheme 1. Schematic Illustration of the Formation of Au@Ag Core@Shell NPs in W/O ME via the Single ME Reactant Addition Scheme along with the Time-Dependent Size Distribution Plot.

(a) Reduction and nucleation process (1–5 min), (b) growth by the autocatalysis process (6–10 min), (c) growth by the Oswald ripening process (11–20 min), (d) growth by the intraparticle coupling process (30–35 min), and (e) growth and stabilization via the steric stabilization process (3 h).

Most of the major reactions for the synthesis of Au@Ag core@shell NPs proceeded at the nucleation process. As depicted in Scheme 1a, a thin and nonuniform Ag shell was formed on the surface of Au NPs after the nucleation process. Then, the second step of the process proceeded through autocatalysis as depicted in Scheme 1b. The free Ag atoms in the same W/O ME solution which did not undergo the nucleation process diffuse to the surface of the Au@Ag core@shell NPs with a nonuniform shell. The energy of these free Ag atoms within the solution is lower than that of the synthesized Au@Ag core@shell NPs with nonuniform shell thickness, and that is why they slowly diffuse on the particle surface and change the shape and size of the particles.⁶³ As shown in the size distribution in Scheme 1b, there was a slight increase in the average D_h of the particles from around 81.69 to 94.96 nm, which indicated that the free Ag atoms diffused to the surface of the particles via diffusion and a uniform and highly stable spherical Ag shell formed on the surface of the Au NPs.

The third step of the process is denoted as “Ostwald ripening”, which is a two-step process as shown in Scheme 1c. In the reverse micelles of the W/O ME system containing Au@Ag core@shell NPs, there are some Au@Ag NPs in which smaller Au@Ag NPs particles are agglomerated on the surface of the larger Au@Ag NPs particles. These Au@Ag NPs have higher size distribution, and these are energetically more unstable than the smaller, nonagglomerated particles. The smaller, nonagglomerated particles are well ordered and packed in the interior since smaller particles have a lower surface area-to-volume ratio, resulting in a lower-energy state (and a lower surface energy).

Therefore, the W/O ME system containing Au@Ag NPs attempts to lower its overall energy via random collision and exchange of the reverse micelles containing Au@Ag NPs and the smaller Au@Ag NPs that are agglomerated on the surface of larger Au@Ag NPs often tend to detach and diffuse in the water core of the nanoreactors of the W/O ME system. This is induced by low thermodynamic stability of the smaller Au@Ag NPs that are agglomerated on the surface of larger Au@Ag NPs. Thus, the number of free smaller Au@Ag NPs is increased in the W/O ME system. This is the first stage of the Ostwald ripening phenomenon as shown in Scheme S1c. When the W/O ME system is supersaturated with the free smaller Au@Ag NPs, the free smaller Au@Ag NPs possess overarching desire to acquire lower surface energies. Thus, the free smaller Au@Ag NPs are aggregated on the surface of the larger Au@Ag NPs and also aggregated with each other in the reverse micelles of the W/O ME system to reach a more thermodynamically stable state. Thus, the number of smaller Au@Ag NPs is started to decrease until they disappeared and the larger Au@Ag NPs grow even larger in the W/O ME system.⁶³ This is the second stage of the Ostwald ripening phenomenonon shown in Scheme S1c.

The size distribution plots of the Ostwald ripening process that occurred within 11–20 min are depicted in Scheme 1c. The size distribution plot of the first stage of the Ostwald ripening process from 11 to 15 min as shown in Scheme 1c exhibited one new size distribution at 12.08 nm, and the strong size distribution at 94.96 nm was shifted to 89.18 nm from the autocatalysis step to the first stage of the Ostwald ripening process. This indicated that due to random collision and exchange of the reverse micelles containing Au@Ag NPs of the W/O ME system, some smaller Au@Ag NPs that were agglomerated on the surface of larger Au@Ag NPs detached and diffused in the water core of the nanoreactors, which was induced by low thermodynamic stability of the smaller Au@Ag NPs that were agglomerated on the surface of larger Au@Ag NPs. Thus, the new size distribution at 12.08 nm appeared because of the free diffused smaller Au@Ag NPs and the strong size distribution at 94.96 nm decreased to 89.18 nm via the diffusion of the smaller Au@Ag NPs that are agglomerated on the surface of larger Au@Ag NPs from the autocatalysis step to the first stage of the Ostwald ripening process.

The size distribution plot of the second stage of the Ostwald ripening process from 16 to 20 min as shown in Scheme 1c exhibited that the size distribution at 12.08 nm disappeared, one new size distribution at 6.53 nm appeared, and the strong size distribution at 89.18 nm shifted to 96.08 nm from the first stage to the second stage of the Ostwald ripening process. This indicated that the W/O ME system was supersaturated with free smaller Au@Ag NPs and thus tended to acquire lower surface energies. Thus, the free smaller Au@Ag NPs were aggregated on the surface of the larger Au@Ag NPs, also aggregated with each other in the W/O ME system, and further reached a more thermodynamically stable state. Thus, the size distribution at 12.08 nm due to free diffused smaller Au@Ag NPs disappeared, new size distribution at 6.53 nm appeared via the aggregation of the free smaller Au@Ag NPs with each other, and the strong size distribution at 89.18 nm increased to 96.08 nm via the aggregation of the free smaller Au@Ag NPs on the surface of the larger Au@Ag NPs from the first stage to the second stage of the Ostwald ripening process.

The fourth step of the process is denoted as intraparticle coupling growth as shown in Scheme 1d. In the intraparticle coupling growth process within 30–35 min, there was a narrow size distribution with a slight increase in the average D_h of the particles with respect to the Oswald ripening process from 6.53 to 8.79 nm and 96.08 to 102.34 nm as shown in Scheme 1d. This slight increase in the average D_h indicated that the particles underwent intraparticle coupling. It also corroborated that the single ME reactant addition scheme favored the intraparticle coupling growth and thus escalated the formation of stable and uniform Au@Ag core@shell NPs.

The fifth and last step is denoted as the steric stabilization as shown in Scheme 1e. The growth and stabilization of the synthesized Au@Ag core@shell NPs were observed within 3 h. After 3 h of the growth reaction, there was only one strong size distribution at around 130.94 nm as shown in Scheme 1e. The disappearance of the smaller particles and the increase in the average D_h of the larger particles from 102.34 to 130.94 nm indicate that the smaller particles aggregate with the larger particles. When the smaller and larger particles along with the reverse micelles approached each other, the particles underwent random collision, exchange, and interaction. Using the concept of entropic stabilization theory, it is assumed that the second surface of the smaller particles approaching the surface of the larger particles is impenetrable. Thus, the particles present in the interaction lose configurational entropy. This reduction in entropy increases the change in the Gibbs free energy ΔG by producing the net effect of repulsion between the particles. If the change in the Gibbs free energy is positive, stabilization results, and if the change is negative, aggregation and precipitation take place. The size distribution plot as depicted in Scheme 1e indicated that the W/O ME favored the reduction in entropy and increased the change in the Gibbs free energy and thus prevented the particles from further reaction through the coagulation process. This integration is termed as steric stabilization of the particles.^38,63

Probable Mechanism for the Formation of Au@Ag Core@Shell NPs via Electron Transition Phenomenon

The reaction mechanism for the reduction process of the [AuCl₄]⁻ species to metallic Au NPs is investigated in some previous reports, but the appropriate formation mechanism is little understood.^55,66−76 An attempt has been made here to illustrate the chemical stability of Au@Ag NPs using the unique electron transition phenomenon between the Au core and Ag shell.⁵⁵

Au NPs were generated from [AuCl₄]⁻ {Au³⁺} precursor reduction through a stepwise mechanism.^66,67,70,72 The formulation mechanism of metallic Au NPs is divided into three representative stages: reduction–nucleation, disproportion, and association as illustrated in Scheme S3. The initial stage was the reduction of [AuCl₄]⁻ {Au³⁺} species to [AuCl₃]⁻ {Au²⁺} species as a means of the one-electron reduction–nucleation process. After that, nucleation-dominant surface growth of the [AuCl₃]⁻ {Au²⁺} species took place.

In the second stage, a disproportionation of [AuCl₃]⁻ {Au²⁺} species was eventuated to produce [AuCl₂]⁻ {Au⁺} species. The reduction of [AuCl₃]⁻ {Au²⁺} species to [AuCl₂]⁻ {Au⁺} species is a slower step than that of [AuCl₄]⁻ {Au³⁺} species to [AuCl₃]⁻ {Au²⁺}. Therefore, the disproportionation reaction is the rate-determining process in the formation of Au NPs.

The [AuCl₂]⁻ {Au¹⁺} species are more stable than [AuCl₄]⁻ {Au³⁺} species. For this reason, [AuCl₂]⁻ {Au⁺} species were promptly reduced to Au⁰ atoms.

In the subsequent third stage, the Au⁰ atoms started to nucleate and after the autocatalytic surface growth and aggregative particle growth process, Au NPs were produced. The process including nucleation, autocatalytic surface growth, and aggregative growth of the Au⁰ atoms is delineated as the association process. In this stage, the growth of aggregative particles dominantly occurred. The reduction of [AuCl₂]⁻ {Au¹⁺} species to Au⁰ atoms is a slower process than that of [AuCl₄]⁻ to [AuCl₂]⁻, and the reduction of [AuCl₂]⁻ species to Au⁰ atoms and the association of Au⁰ atoms to form Au NPs proceed concurrently. After the association process, Au NPs are formed.

When the AgNO₃ precursor solution was introduced into the W/O ME system containing Au NPs, an intriguing electron transition phenomenon occurred at the interface of the Au atoms and Ag ions as depicted in Scheme 2.^5,6,20 The Au atoms started to form Au³⁺ ions by releasing three electrons stepwise, and then the Ag⁺ ions accepted the electrons that were released by Au atoms and form Ag atoms, which furthermore diffuse to the surface of the Au NPs. This phenomenon is known as the galvanic replacement reaction.

Scheme 2. Schematic Illustration of the Morphological and Structural Changes at Different Stages during the Formation of Au@Ag Core@Shell NPs.

It is feasible to synthesize Au@Ag core@shell NPs via the galvanic replacement reaction, but there are some drawbacks of this process: it is a very slow process, the formation of uniform Ag shell on the Au surface is quite difficult, and this process has higher possibilities to form a hollow nanostructured Au@Ag core@shell-type structure or Au/Ag alloy-type structure rather than a Au@Ag core@shell-type structure. Hence, to avoid the formation of hollow nanostructured NPs or alloy-type NPs, NaBH₄ was introduced into the Au NPs containing W/O ME solution after addition of the AgNO₃ precursor solution so that the Ag⁺ reduced rapidly to Ag atoms and suppressed the galvanic replacement reaction. Then, two reactions proceeded simultaneously. One is, the  Ag⁺ ions reduced by NaBH₄ into Ag atoms and then the Ag atoms diffused on the surface of the Au NPs. Moreover, the other one is the nonreacted Ag⁺ ions that underwent galvanic replacement reaction and formed Ag atoms also diffused on  the surface of the Au NPs, but this was a slow process. The two reactions proceeded concurrently, and the Ag shell started to form around the Au NPs.

The electronic and chemical properties of the Ag shell are highly enhanced by coupling of the Ag shell to the Au core due to a unique charge transfer that increased the electron density within the Ag shell, yielding a negative Ag oxidation state, Ag^δ−, which suppressed the galvanic replacement reaction at the Ag shell surface and further increased the stability of the Ag shell against oxidation.²⁰ The Au core served as a platform for regulating the overall particle size and also modified the electronic characteristics of the deposited Ag shell.^5,20 The electron-rich Ag shell increased the stability of the Au@Ag core@shell NPs and restricted the formation of the Au–Ag alloy.⁶ The electronic interaction between Au and Ag is counterintuitive since Au is traditionally thought of as “highly electronegative”, which implies that Au will, in general, withdraw electrons from Ag.⁵ However, realistically, the electronic interaction between Au and Ag is quite complex. When Au is the core material, Au gains the non-d orbital charge and loses the d orbital charge, and this d orbital charge, that is lost by Au, is gained by Ag and thus increases the electron density of the Ag shell. This charge redistribution is essentially what causes the Ag to gain enhanced stability properties in the Au@Ag core@shell NPs system.^5,20,22

Conclusions

In summary, the study demonstrates a facile and efficient approach for the successful synthesis and stabilization of Au@Ag core@shell NPs using the W/O ME technique via the single ME reactant addition scheme. Detailed HRTEM and DLS analyses corroborated the uniform growth of the Ag shell on the Au core inside the confined region of the reverse micelles. Consequently, successful synthesis of the Au@Ag core@shell NPs showed that the reverse micelles containing W/O ME system acted as “true nanoreactors” for the synthesis of Au@Ag core@shell NPs, as they provided a controlled and confined medium and controlled the size of the synthesized particles. More importantly, the Au@Ag core@shell NPs exhibited remarkably engrossing plasmonic properties. Two resonance absorption bands were found existing in the absorption spectrum of the Au@Ag core@shell NPs. The mode at lower wavelength that is ascribed to the quadrupolar LSPR mode is a characteristic of isolated Au@Ag core@shell NPs where the electric field energy is concentrated at the interface between the core and shell metals. Additionally, the mode at higher wavelength that is ascribed to the dipolar LSPR mode is a characteristic of coupled Au@Ag core@shell NPs where the electric field energy is concentrated at the outer surface of the NPs.⁴⁶ Likewise, the LSPR of the core and shell can affect the overall LSPR of the Au@Ag core@shell NPs with respect to electromagnetic field enhancement, which is also evident from the deconvoluted spectrum of the Au@Ag core@shell NPs. The deconvoluted absorption spectra impart a strong evidence of the formation of a uniform Ag shell on the Au core and also provide a perspective of how the size of the NPs can dominate the plasmonic properties. The formation mechanism of Au@Ag core@shell NPs in W/O ME via the single ME reactant addition scheme also contributes a better perspicacity of the stepwise reduction, nucleation, and growth process. The unique electron transfer phenomenon provides an intuitive understanding about the electronic transition properties and the enhanced stability of the particles. Bimetallic plasmonic Au@Ag core@shell NPs exhibit remarkably engrossing plasmonic properties and thus it is still a challenging as well as an enthralling field for the nanoscientists. These NPs are expected to have applications in biosensors, photonics, biomedical fields, optics, bioimaging, drug delivery systems, and optoelectronic devices.^{1−4,24−33}

Experimental Details

Materials

Triton X-100 (polyoxyethylene octyl phenyl ether), cyclohexane, hexanol-1, and sodium tetrachloroaurate(III) dehydrate, (NaAuCl₄·2H₂O, >99%) were purchased from Sigma-Aldrich, Germany. Silver nitrate (AgNO₃) was obtained from Merck, and sodium borohydride (NaBH₄, >98%) was obtained from Acros Organics, Germany. All of the chemicals were of analytical grade and were used as received without further purification. All solutions were prepared using deionized (DI) water (conductivity: >0.055 μS m^–1 at 25.0 °C) from an HPLC-grade water purification system (BOECO, Germany).

Preparation of Water-in-Oil Microemulsion

W/O ME was prepared by mixing Triton X-100 (surfactant), cyclohexane (oil), hexanol-1 (cosurfactant), and DI water in the mass ratio of 1.5012:0.9929:0.3116:12.2305, resulting in a molar W/O ratio of 7.21. The mixture was sonicated for 30 min at ambient temperature to obtain a clear, thermodynamically stable, isotropic liquid mixture as illustrated in Scheme S4. The composition of the prepared W/O ME is shown in Table S1.

Synthesis of Au NPs

The W/O ME was prepared using the composition as shown in Table S1. W/O ME solution (4.00 mL) was taken in a glass vial. Furthermore, 10.40 μL of the prepared aqueous solution of 53.8 × 10^–3 M NaAuCl₄ was injected into the W/O ME solution using a micropipette. Then, the solution was vigorously stirred for 2 min using the sonicator, and a light yellow solution was obtained, which indicated the incorporation of NaAuCl₄ solution into the W/O ME. Then, 6.40 μL of 278.9 × 10^–3 M NaBH₄ solution was added into the same W/O ME using a micropipette to reduce the Au³⁺ ions. The solution was further vigorously stirred for 15 min using the sonicator. The color of the solution changed from light yellow to flamingo pink, which implied the formation of Au NPs.

Synthesis of Au@Ag Core@Shell NPs

After preparation of Au NPs, 5.30 μL of the prepared aqueous solution of 105 × 10^–3 M AgNO₃ was injected into the same W/O ME solution containing the Au NPs and vigorously stirred for 2 min using the sonicator. Subsequently, 4.20 μL of the 278.9 × 10^–3 M NaBH₄ solution was added to this W/O ME solution as a reducing agent and the solution was further vigorously stirred for 15 min. The color of the solution changed from flamingo pink to violet, which indicated the formation of Au@Ag core@shell NPs. The composition of the components used for the synthesis of Au@Ag core@shell NPs is given in the Table S2.

Synthesis of Ag NPs

First, 5.30 μL of the prepared aqueous solution of 105 × 10^–3 M AgNO₃ was injected into 4.00 mL of the W/O ME solution using a micropipette and mixed together for 2 min using a sonicator. After that, 4.20 μL of the 278.9 × 10^–3 M NaBH₄ solution was added to the same W/O ME solution as a reducing agent and the solution was further vigorously stirred for 10 min. The color of the solution changed from transparent to bright yellow, which was indicative of the formation of Ag NPs.

Characterization

The morphology, size, and the core–shell contrast of the synthesized Au@Ag core@shell NPs were investigated using high-resolution transmission electron microscopic analysis (model: JEOL JEM 2100 HRTEM). The HRTEM analysis of the sample was performed by drop and dry method using a drop of the ME containing the NPs some days after the synthesis. A few drops of the W/O ME solution containing Au@Ag core@shell NPs were cast on a copper-coated grid to form a thin layer and allowed to dry for a few minutes. HRTEM micrographs were taken at different magnifications at 200 kV. The selected area electron diffraction (SAED) patterns of the Au@Ag core@shell NPs were also analyzed along with HRTEM analysis.

The energy dispersive X-ray spectroscopic (EDS) analysis was carried out using the spectrometer attached with a field emission scanning electron microscope (FESEM, model: JSM-7600F). A copper stab was used to cast the sample. The EDS spectrum was taken by selecting several spots or a zone of a specific particle. The accelerating voltage was 10 keV, and the counting rate was varied from 1375 to 3343 cps.

The hydrodynamic diameter (D_h) of the synthesized NPs was determined using Zetasizer Nano ZS90 (ZEN3690, Malvern instruments Ltd., U.K.) by employing the dynamic light scattering technique. A He–Ne laser beam of wavelength 632.8 nm was used, and the measurements were made at a fixed scattering angle of 90°. A measuring glass cell of 10 mm diameter was used throughout the experiment. The accuracy of the D_h determined by DLS measurements was approximately ±2%.

The synthesized Au@Ag core@shell NPs had been examined using X-ray powder diffraction (model: D8 Advance, Bruker, Germany) to study the distribution of the elements. To prepare the sample for analysis, ethanol was added into the Au@Ag core@shell NPs that were synthesized using W/O ME as nanoreactors so that the samples can be easily separated from the nanoreactors. The sample was washed using ethanol several times so that the surfactant molecules can be totally removed from the Au@Ag core@shell NPs. Then, these washed Au@Ag NPs were dried at 80 °C for 30 min in an oven, and we got the powdered Au@Ag NPs. The dried powdered Au@Ag NPs were analyzed using Cu Kα₁ radiation (λ = 1.5406). The phase composition was analyzed with a wide range of Bragg angles, 2θ ranging from 0 to 90°, to determine the crystalline structure of the Au@Ag core@shell NPs.

UV–visible spectra of the synthesized NPs were measured using a Spectro UV–visible double beam spectrophotometer (model: UVD-500, Labomed) in the range of 200–900 nm to study the plasmonic properties. Triton X-100-based microemulsion was used as the reference.

Glossary

Abbreviations

W/O

water-in-oil

ME

microemulsion

NPs

nanoparticles

HRTEM

high-resolution transmission electron microscope

SAED

selected area electron diffraction

LSPR

localized surface plasmon resonance

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.9b01897.

• Table S1, composition and water-to-surfactant molar ratio of TX-100; Table S2, composition of the components used for the synthesis of Au@Ag core@shell NPs; Table S3, summary of FESEM–EDS results; Table S4, parameters of Au@Ag core@shell NPs from size distribution plot of Figure 2; Table S5, parameters of the deconvoluted absorption spectrum of Au, Ag and Au@Ag core@shell NPs; Figure S1, EDS spectrum of Au@Ag core@shell NPs; Figure S2, HRTEM images of Au@Ag core@shell NPs indicating the stripe-like Moiré pattern; Figure S3, SAED patterns of Au@Ag core@shell NPs; Figure S4, correlogram and size distribution by volume of Au@Ag core@shell NPs; Figure S5, absorption spectra of Triton X-100-based W/O ME; Figure S6, electric field and charge distribution at the surface of NPs; Figure S7, deconvoluted absorption spectrum of Au NPs; Figure S8, deconvoluted absorption spectrum of Ag NPs and Scheme S1, representation of two Au@Ag core@shell NPs into the reverse micelle system and their major interactions; Scheme S2, schematic illustration of the formation of Au NPs in W/O ME via single ME reactant addition scheme along with size distribution plot; Scheme S3, morphological and structural changes at different stages during the formation of Au NPs; Scheme S4, schematic illustration of the formation of TX-100/cyclohexane/hexanol-1/deionized water-based water-in-oil ME (PDF)

Author Contributions

All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.