Universal One-Pot and Scalable Synthesis of SERS EncodedNanoparticlesBernat Mir-Simon,† Irene Reche-Perez,†,‡ Luca Guerrini,†,‡ Nicolas Pazos-Perez,*,†

and Ramon A. Alvarez-Puebla*,†,‡,§

†Medcom Advance, Viladecans Business Park, Edificio Brasil, Bertran i Musitu 83-85, 08840 Viladecans, Barcelona, Spain‡Departamento de Quimica Fisica e Inorganica, Universitat Rovira i Virgili and Centro de Tecnologia Quimica de Cataluna, Carrer deMarcel·lí Domingo s/n, 43007 Tarragona, Spain§ICREA, Passeig Lluís Companys 23, 08010 Barcelona, Spain

*S Supporting Information

ABSTRACT: Encoded particles are one of the most powerfulapproaches for multiplex high-throughput screening. Surface-enhanced Raman scattering (SERS) based codification can, inprinciple, avoid many of the intrinsic limitations due toconventional alternatives, as it decreases the reading time andparticle size while allowing for almost unlimited codification.Unfortunately, methods for the synthetic preparation of theseparticles are tedious; often subjected to limited reproducibility(associated with large fluctuations in the size distributions of thepolymers employed in the standard protocols); and to date,limited to a small amount of molecules. Herein, we report a universal, one-pot, inexpensive, and scalable synthetic protocol forthe fabrication of SERS-encoded nanoparticles. This synthetic strategy is highly reproducible, independent of the chemical natureand size of the Raman code used (31 different codes were tested) and scalable in the liter range without affecting the finalproperties of the encoded structures. Furthermore, the SERS efficiency of the fabricated encoded nanoparticles is superior to thatof the materials produced by conventional methods, while showing a remarkable reproducibility from batch to batch. Thisencoding strategy can easily be applied to nanoparticles of different materials and shapes.

■ INTRODUCTION

Encoded nanoparticles are among the most powerfulalternatives for high-throughput multiplex screening1,2 inmicroarray technology,3 diagnosis,4,5 and bioimaging.6 Thesematerials are simple and cost-effective platforms that allow forfast, sensitive, and reliable analyses.2,7−14 During the pastdecade, several encoded particles have been prepared15−17

using codification strategies based on changes in particleshape,18 composition,19 physical marks,17 or spectroscopicproperties (e.g., luminescence or vibrational finger-prints).6,20−22 Among all of them, those based on surface-enhanced Raman scattering (SERS) are gaining importance23

because of (i) virtually unlimited multiplexing capabilityassociated with the unique vibrational fingerprints of thedifferent codes; (ii) short detection times (milliseconds) thanksto the intrinsic sensitivity of the SERS phenomenon;24 (iii)small size, allowing for bioimaging;25−27 and (iv) photostabilityand low toxicity (as compared to those of dyes or quantumdots).28

In essence, a SERS-encoded nanoparticle (also called a SERStag) comprises a plasmonic nucleus, responsible for thegeneration of the electric field necessary for the Ramanamplification; a Raman probe (i.e., code), responsible for theunique vibrational fingerprint of the encoded particle; and a

coating layer. This external coating is of key importance as it (i)prevents the code from leaching out into the medium, thusavoiding toxic effects or vibrational cross-contamination withthe codes of other particles; (ii) protects the plasmonic particlefrom contaminations of the medium that could give rise tovibrational noise that would hinder the particle readout; (iii)increases the colloidal stability of the particle; (iv) provides aconvenient surface for further chemical functionalization; and(v) protects the plasmonic core from interacting with otherplasmonic particles, avoiding plasmon coupling and thus theuncontrolled generation of hotspots. Although polymers havebeen reported as particle coatings,29−31 the unique properties ofsilica (i.e., known surface chemistry, biocompatibility, opticaltransparency, and colloidal stability) make this material themost efficient protective layer for nanoparticles by far.32,33

Silica coating of nanoparticles requires the colloidalstabilization of the particles in ethanolic solution prior to thehydrolysis/condensation of tetraethyl orthosilicate (TEOS).Although a range of polymers has been proposed for thistask,32,34 the most common remains polyvinylpyrrolidone

Received: November 19, 2014Revised: January 7, 2015Published: January 12, 2015

Article

pubs.acs.org/cm

© 2015 American Chemical Society 950 DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

(PVP).32,34 On the other hand, surfactants such ascetyltrimethylammonium bromide (CTAB) are also usedcommonly for this reaction.32 Nevertheless, with the exceptionof nanogapped core−shell nanoparticles,35,36 the mostimportant factor for the generation of active SERS-encodedparticles is intimate contact between the Raman code and theplasmonic structure.This requirement introduces further complexity into the

coating process associated with the surface chemistry proper-ties. Both PVP and CTAB form solid layers of coating on thesurface of the particles, limiting or even preventing theinteraction of the encoding agent with the metallic core whenadded to the solution.37,38 Therefore, to increase the codeadsorption efficiency on the plasmonic structure and, thus, theSERS signal, PVP and CTAB species need to be removed fromthe metallic surfaces.On the other hand, this removal usually results in a dramatic

reduction of the colloidal stability, which is further aggravatedby the nonpolar nature of most codes, leading to uncontrolledparticle agglomeration24,39,40 or even to irreversible precip-itation. Aggregation of labeled nanoparticles into clusters ofdifferent sizes and geometries does generate very active SERSstructures but with a highly inhomogeneous SERS response.Moreover, these fabrication methods normally work for a verylimited number of encoding molecules as, in many cases,precipitation of the whole colloid occurs upon addition of thecode (except for very low amounts of label). In fact, thisexplains why, in most of the literature, examples of SERS-encoded particles include small numbers of codes, usually justthree or four.As an alternative to conventional polymers or surfactants,

thiolated poly(ethylene glycol) (PEG) has been successfullyemployed for the controlled silica coating of single metallicnanoparticles. The high polarity and porosity of this polymerefficiently stabilize particles in alcohol and water while allowingfor the at least partial diffusion of the code to the metallicsurface.41

Importantly, polymers commonly suffer from large fluctua-tions in size distribution from batch to batch even for the samecommercial brand. As a result, the synthetic protocol forencoding particles using these materials often needs to beretuned when a new polymer is purchased. Additionally, thehigh price of thiolated PEG hinders its use in the large-scalepreparation of encoded particles as required for real-lifeapplications. In contrast to most of the reported procedures,which typically involve complex steps, herein, we demonstratean easy and fast one-pot approach for the production of SERS-encoded nanoparticles. As the initial plasmonic material, citrate-capped gold nanoparticles were selected, as most of thepublished research on SERS-encoded particles has been carried

out on spherical gold nanoparticles.27,42,43 However, theencoding protocol can be readily applied to other metals,such as citrate-reduced silver colloids, and nanoparticle shapes,such as gold nanostars.This versatile strategy allows for the SERS codification of

particles with every molecule with affinity toward the metalsurface, independently of its chemical nature, as exemplifiedhere in the fabrication of 31 different encoded particles usingexactly the same standard procedure. The method relies on thecontrolled coabsorption of mercaptoundecanoic acid (MUA)and the SERS code on the metallic surfaces. In addition to theease of preparation, scalability to the liter range, and stability inaqueous solutions, our SERS-encoded particles show consid-erably higher optical efficiencies than those fabricated usingPEG or PVP polymers.

■ RESULTS AND DISCUSSION

A schematic outline of the universal protocol for the fabricationof encoded particles is presented in Figure 1. The process canbe divided into four different steps: (1) synthesis of citrate-capped gold nanoparticles of ca. 50-nm diameter (see Figure S1in the Supporting Information for transmission electronmicroscopy (TEM) images and size distribution histograms),(2) MUA functionalization, (3) SERS codification, and (4)silica coating. Because of the low stability of colloidal solutionsupon functionalization with the SERS code, a stabilization stepis required prior to the codification. MUA was chosen as thestabilizing agent because it binds covalently to the gold surfacethrough the thiol group while providing particle stability withboth a long aliphatic chain (steric repulsion) and a terminalcarboxylic group (electrostatic repulsion). On the other hand,because of its aliphatic nature, its Raman cross-section is almostnegligible as compared to those of aromatic compounds(Figure S2, Supporting Information).44,45 Nevertheless, thepresence of a thiol group implies that MUA should be added inan adequate proportion to avoid the formation of a compactmonolayer that could passivate the metallic surface, preventingthe retention of the SERS codes, and with extreme care to avoidheterogeneous adsorption of the molecule by some of thecolloids in the solution. Thus, in a second step, MUA wasrapidly added under vigorous stirring at basic pH to yieldMUA-functionalized gold nanoparticles (Au@MUA).To maximize the final SERS efficiency of the encoded

nanoparticles, the MUA surface coverage was decreased asmuch as possible to provide maximum accessibility to the metalsurface while preserving the overall colloidal stability whenexposed to an excess of the SERS code. The addition of thecode is depicted as the third step in Figure 1. Among all of theinvestigated encoding molecules, 2-mercaptopyridine (MPy)was observed to induce the fastest colloidal aggregation upon

Figure 1. Schematic representation of the synthetic procedure for the production of SERS-encoded nanoparticles showing the different stepsinvolved in the synthesis. First, citrate-capped Au nanoparticles are produced. Second, MUA is used to stabilize the particles in basic media. Third, anexcess of SERS code is added to encode the particles. Fourth, a silica shell is growth on the particles to ensure stability for long periods of time andavoid undesired plasmon coupling.

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

951

addition to the bare citrate-capped gold nanoparticles. For thisreason, the optimization of the protocol was performed usingthis molecule (i.e., the worst colloidal stability scenario). MUAhas been reported to occupy an area of 0.22 nm2,corresponding to ca. 4.5 molecules nm−2.46 Therefore,experiments decreasing the amount of MUA were designedfor concentrations of between 5.0 and 0.1 molecules nm−2.After sufficient time had been allowed for MUA adsorption toreach its thermodynamic equilibrium (30 min), MPy was addedin a large excess (15 molecules nm−2) to yield thecorresponding Au@MUA/MPy nanoparticles. Aggregationevents in the colloidal system were monitored by UV−vis−NIR spectroscopy (Figure 2) via comparison of the localized

surface plasmon resonance (LSPR) at 535 nm, associated withisolated gold nanoparticles, and the absorption feature at 750nm, attributed to plasmonic contributions of interactingparticles and thus indicative of aggregation (Figure 2C).Concurrently, SERS was also monitored in the same samples toestimate the amount of adsorbed code (Figure 2B). The SERSmeasurements were performed by focusing a 785-nm laser ontothe colloidal suspension ([Au] = 1 mM) with a long-working-distance objective. At concentrations between 5 and 0.8 MUA

molecules nm−2 (stable colloidal regime), the extinction spectraof Au@MUA/MPy exhibit the characteristic LSPR ofmonodisperse spherical gold nanoparticles in suspension. At0.7 molecules nm−2, the appearance of a shoulder at ca. 700 nmwas observed, indicating the significant formation of nano-particle aggregates. A further decrease of the MUA surfacecoverage led to a dramatic perturbation of the colloidal stabilityupon addition of MPy, as clearly revealed by the dominantplasmonic contribution at longer wavelengths. Therefore, therange of colloidal stability was identified as being between 5 and0.8 MUA molecules nm−2. Conversely, the monitoring of theSERS intensity of the MPy ring breathing mode at 1001 cm−1

(Figure 2D) reveals the existence of three SERS regimes. Thefirst corresponds to particle aggregation (below 0.8 MUAmolecules nm−2) and, as expected, shows a marked increase ofthe intensity due to uncontrolled plasmon coupling (see alsothe very large standard deviation). The second, between 1.4and 0.8 MUA molecules nm−2, reveals a progressive decrease ofthe SERS intensity of MPy as the MUA content increases. Inthis regime, an increase in MUA surface coverage is directlyreflected in the decrement of code adsorption onto the metalsurface. In the third regime, 1.6−5 MUA molecules nm−2, theSERS intensity remains constant as MUA forms a progressivelyfuller monolayer and only a fixed amount of MPy moleculescan diffuse to the nanoparticle surfaces. Notably, for smallmolecules such as MPy, the SERS intensity never decays tozero. This is because, unlike for the crystalline arrangement ofdensely packed films of alkanethiols on gold surfaces,47 theCoulomb repulsions between the negatively charged carboxylicgroups of MUA limit the lateral interactions of the hydrophobicalkyl chains, preventing the formation of a thick molecularpacking on the surface.48,49 Clearly, accessibility to the metalsurface in this particular regime is highly dependent on thechemical and geometrical properties of the SERS code. Forinstance, for a MUA concentration of 4 molecules nm−2, almostno SERS signals were observed for large molecular codes(Figure S3, Supporting Information). Therefore, 0.8 moleculesnm−2 was identified as the optimum MUA concentration forthe production of our SERS-encoded gold nanoparticles,preventing colloidal aggregation and maximizing the finalSERS signal.Finally, to ensure stability for long periods of time, protect

both the SERS code and the plasmonic core, and generate areadably functionalizable external surface, the Au@MUA/MPynanoparticles were encapsulated in a silica matrix. Silica coatingwas performed using a modification of the well-known Stobermethod50 by exploiting the ability of ligands with terminalcarboxylic acid groups, such as MUA, to induce silicagrowth.51−55 To this end, appropriate amounts of ethanoland NH4OH were added to the Au@MUA/MPy aqueoussuspension to maintain an adequate pH of the solution andprovide the correct EtOH/H2O molar ratio (1.84) for theStober process. Then, TEOS was added to initiate silica growth.The solution was allowed to react for 14 h at room temperaturebefore being subjected to several washing cycles (Figure 3B−Dpresents characteristic TEM images of Au@MUA/MPy@SiO2nanoparticles). A thick silica shell of ca. 30 nm was grown ontop of the labeled nanoparticles to prevent the localization ofintense local fields at the interparticle junctions of possibleSERS-encoded nanoparticle agglomerates. Figure 3A displaysthe extinction spectra of the colloidal suspension after eachfabrication step. As can be seen, the shift of the LSPRs ofindividual nanoparticles clearly reflects the changes in the

Figure 2. (A) Extinction spectra of Au@MUA/MPy nanoparticleswith different amounts of MUA (molecules nm−2). MPy was alwaysadded in a large excess to guarantee the full coating of thenanoparticles. Below 0.8 MUA molecules nm−2, it is possible torecognize an unstable colloidal regime (nanoparticles start aggregat-ing). (B) SERS spectra of Au@MUA/MPy colloids ([Au] = 1 mM)with different amounts of MUA (molecules nm−2). (C) Intensity ratiobetween the absorption at 535 nm and the absorption at 750 nm forAu@MUA/MPy nanoparticles as a function of different amounts ofMUA. (D) SERS intensity of Au@MUA/MPy colloids ([Au] = 1mM) as a function of different amounts of MUA.

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

952

refractive index associated with each functionalization step(Figure 3A, inset). First, the submonolayer deposition of MUAinduces minimal changes in the plasmon maximum; then fullcoating with MPy results in a ca. 3-nm shift; and finally, thegrowth of a thick silica layer is responsible for the largedisplacement up to 545 nm. Importantly, the extinction spectrado not reveal any significant broadening of the LSPR, indicatingthat the nanoparticles preserve their colloidal stability duringthe whole process with no appreciable formation of aggregates.It is worth noting that, independently of the SERS encodingprocess and the related colloidal stability, homogeneous silicacoatings could be achieved only for MUA concentration above0.7 molecules nm−2 (Figure S4, Supporting Information).The described protocol was successfully extended to a large

set of different codes, including thiolated and nonthiolatedaromatic small molecules and dyes (phenothiazines, rhod-amines, oxazines, triarylmethanes, tri- and tetrazoles, etc.),confirming the universal applicability of this synthetic strategy.Figure 4 shows the SERS signatures and TEM images of 11representative SERS-encoded nanoparticles, whereas another20 codes are reported in the Supporting Information (FigureS5). Notably, the one-pot synthetic method was successfullyemployed in the fabrication of larger volumes of SERS-encodednanoparticles (in the liter range; Figure S6, SupportingInformation) without impacting the final characteristics of thesubstrate, which clearly demonstrates the scalability of theprocess.To evaluate the optical efficiency of our protocol, we

compared the SERS intensities provided by seven of our SERS-encoded nanoparticles with those yielded by their analogouscounterparts fabricated (when possible) with the most commonpolymer-based procedures used in the literature: PVP andthiolated PEG.41,56 Both polymer-based strategies rely onproviding the required stability to the Au particles during thecodification step. To this end, the reported recipes41,56 werefollowed and optimized to achieve the highest possible SERSsignal. When PVP was used, citrate-stabilized Au nanoparticleswere first produced and then subsequently transferred to EtOHusing PVP. Prior to code addition, the particles were extensivelywashed to remove as much as possible of the PVP from thesurface of the particles while maintaining the colloidal stability.These washing cycles are critical to maximize the adsorption ofthe code onto the metallic surface.

In the case of thiolated PEG, it was previously shown that thepolymeric layer can significantly hinder the diffusion ofmolecules onto the metal surface.56 Consequently, because ofthe very limited number of Raman molecules capable ofbinding PEG-coated nanoparticles, the use of the correspond-ing Au@PEG@code particles is largely limited to surface-enhanced resonant Raman scattering (SERRS) conditions; i.e.when the electronic excitation of the dye is in resonance withthe laser beam.41,57 An identical limitation applies if thecodification step is performed prior to the PEG-coating step(Au@code@PEG)31 because, in this case, except for fewfortunate cases of “nanoparticle-stabilizing” SERS code, thenumber of molecules per unit area that can be adsorbed ontothe metal without perturbing the colloidal stability is very small.We therefore first tried to improve the accessibility of theRaman label to the metal surface (i.e., increasing the number ofcode molecules per nanoparticle) by progressively lowering theamount of added PEG-SH from 4 molecules nm−2, as reportedin the literature,41 to 0.2 molecules nm−2. However, nano-particle aggregation was already observed at a polymerconcentration less than 3.5 molecules nm−2, even prior to theaddition of the excess of MPy (Figure S7, SupportingInformation). Moreover, no distinguishable SERS signalswere recorded upon functionalization of the PEG-coated gold

Figure 3. (A) Extinction spectra of Au particles after each step of theMUA-based protocol. Inset: Detail of the plasmon absorption maxima.(B−D) Representative TEM images at different magnifications of theMPy SERS-encoded gold nanoparticles.

Figure 4. SERS spectra of 11 representative SERS-encoded nano-particle colloids ([Au] = 1 mM). From top to bottom: 2-mercaptopyridine (MPy), benzenethiol (BT), mercaptobenzoic acid(MBA), 4-nitrobenzenethiol (4-NBT), 3,4-dicholorobenzenethiol(DBT), 3-fluorothiophenol (3-FTP), 4-fluorothiophenol (4-FTP),3,5-bis(trifluoromethyl)benzenethiol (3-FMBT), methylene blue(MB), Nile blue A (NBA), and rhodamine 6G (R6G). Right column:Representative TEM images of the corresponding SERS-encoded goldnanoparticles.

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

953

nanoparticles with the excess of MPy unless in the presence ofcolloidal aggregation (Figure S7, Supporting Information).Discarding the possibility to increase the MPy surface coverageby decreasing the PEG-SH concentration, we pursued adifferent strategy.56,57 In this case, CTAB-stabilized Aunanoparticles of a similar size were prepared instead ofcitrate-capped colloids. The surfactant double layer offers aneffective stabilizing shell, preventing nanoparticle aggregationwhen the colloids are exposed to lower PEG amounts(corresponding to 1.8 molecules nm−2 under our optimizedexperimental conditions). The resultant PEG/CTAB-stabilizednanoparticles were then subjected to, first, several washingcycles to remove excess surfactant and, second, exposure to theexcess of MPy in the codification step without impacting thecolloidal stability. This optimized protocol allowed for thesynthesis of SERS-encoded nanoparticles with much higherSERS efficiencies, generally larger than that observed for thePVP approach, but still significantly lower than that provided bythe MUA-based method (Figure 5A and Figure S8, SupportingInformation). Furthermore, it is important to stress that, evenwhen optimized, the combined CTAB/PEG approach retainsintrinsic limitations and problems, such as the necessity formultiple cycles of centrifugation, separation, and resuspension(which are often critical for colloidal stability and also representpractical obstacles for large-scale production) and the use oflarge amounts of the highly cytotoxic CTAB.58

Table 1 summarizes the SERS efficiency ratios calculated bycomparing the SERS intensities of seven different SERS-encoded nanoparticles obtained by the MUA-based protocolwith those yielded by the PEG-SH and PVP approaches (IMUA/IPVP and IMUA/IPEG‑SH, respectively). For all the investigatedcases, the SERS intensity achieved with the MUA-basedprotocol lies between 140 to 10 times higher than in the caseof PVP, and from 10 to 2 times higher than in the case ofthiolated PEG. Such drastic differences from one Raman labelto another can be ascribed to the different chemical natures andmolecular sizes of the codes which clearly affect their ability todiffuse through the external polymeric layer and finally adsorbonto the metal surface.As previously indicated, modification of the nanoparticle

surface chemistry by adsorption of Raman labels normallyresults in a decrease of the colloidal stability in suspension.However, in a very few cases, the specific chemical nature of thecode molecule can still preserve such stability, with the codeacting as a stabilizing agent. This is the case, for instance, for 4-mercaptobenzoic acid (MBA). As for MUA, the carboxylicgroups of MBA bound to the metal surface are oriented towardthe bulk solution, providing, at basic pH, the necessaryelectrostatic repulsion to avoid nanoparticle aggregation and,in the subsequent silica-coating step, promote silica growth. Asa result, this SERS-encoded nanoparticle can be synthesized atfull MBA coverage with no need for external stabilizing agents.We therefore assess the SERS performance of MBA-encodednanoparticles produced by our MUA synthetic method (0.8MUA molecules nm−2) with respect to the same SERS tagsobtained at full MBA coverage (i.e., 0 MUA molecules nm−2,maximum SERS efficiency). Figure 5C shows the correspond-ing SERS spectra as well as those of the analogous SERS tagssynthesized by the PVP and thiolated-PEG methods. Notably,MUA functionalization results in a ca. 42% loss with respect tothe maximum SERS efficiency, which is much less than thoseobserved for the PVP (ca. 97% reduction) and thiolated-PEG(92% reduction) methods. The SERS enhancement factor for

gold spherical nanoparticles fully coated with MBA was alsocalculated by direct comparison with the normal Ramanspectrum of MBA in 0.1 M ethanol solution (Figure S9,Supporting Information). The estimated enhancement factor at

Figure 5. Comparison of the SERS efficiencies of encodednanoparticle suspensions synthesized by the three different PVP,PEG-SH, and MUA approaches ([Au] = 1 mM). (A) SERS spectra inthe 900−1210 cm−1 range for MPy-encoded nanoparticles. (B) SERSintensity (%) for encoded nanoparticles using 1NT, NBA, MBA, BT,MPy, TFBT, and 4NBT as Raman labels (corresponding SERS spectraare shown in Figure S6, Supporting Information). (C) SERS spectra inthe 1000−1630 cm−1 range for MBA-encoded nanoparticles preparedusing a full monolayer of MBA (purple) and by the three differentapproaches.

Table 1. SERS Efficiency Ratios Calculated by Dividing theSERS Intensities of Different SERS-Encoded NanoparticlesObtained by the MUA-Based Method by Those Synthesizedby the PVP and PEG-SH Approaches

code IMUA/IPVP IMUA/IPEG‑SH

NBA 22.2 3.6BT 14.9 3.8MPy 26.5 2.71NT 10 3.1MBA 19.4 7TFBT 79 2.94NBT 140 2

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

954

785 nm is 4.53 × 104 M (see Supporting Information for alldetails about this calculation).Finally, in addition to spherical gold nanoparticles, we

demonstrate the straightforward application of the MUA-basedencoding protocol to other SERS-enhancing platforms such asgold nanostars prepared by the standard PVP protocol59 (i.e.,same material but different shape) or citrate-reduced silvercolloids (i.e., similar shape but different material). Goldnanostars were selected as a representative example of highlySERS-active spiked nanostructures. Figure 6 presents repre-

sentative SERS spectra and TEM images of different encodedsilver nanoparticles and gold nanostars. The enhancementfactors associated with these structures were found to be 9.12 ×105 M for the gold nanostars and 6.73 × 107 M for the silverspherical particles.

■ CONCLUSIONS

In summary, herein, we describe a universal, one-pot,inexpensive, and scalable synthetic protocol for the fabricationof SERS-encoded nanoparticles. This method relies on thefunctionalization of plasmonic nanoparticles with a submono-layer of mercaptoundecanoic acid, providing high colloidalstability during the codification process while allowing Ramanlabels to easily diffuse onto the metal. Furthermore, in asubsequent step, the carboxylic groups of MUA also act asfunctional sites promoting silica growth on the outer shell ofthe nanoparticles. This synthetic strategy was found to besuccessfully applicable to every Raman code we tested (31codes) and scalable up to 2 L without affecting the finalproperties of the encoded structures. It is worth noticing thatthe MUA-based method avoids the use of high-molecular-weight polymers and their associated issues of reproducibilityfrom batch-to-batch even for different manufacturers. TheSERS efficiencies of the so-fabricated encoded nanoparticleswere found to be 2−140 times higher than those of thecorresponding SERS tags prepared by the common polymer-based methods (i.e., using PVP and thiolated PEG). The MUA-based protocol can be readily to metallic nanoparticles, such ascitrate-reduced silver colloids or gold nanostars.

■ EXPERIMENTAL SECTIONMaterials and Methods. Gold(III) chloride trihydrate (99.9%,

HAuCl4·3H2O), dehydrated trisodium citrate (≥99.5%, C6H5Na3O7·2H2O), ammonia solution (29%, NH4OH), L-ascorbic acid (≥99.0%,AA), tetraethoxysilane (99.999%, TEOS), ethanol (99.5%, EtOH),polyvinylpyrrolidone (average MW = 58000, PVPk58), polyvinylpyrro-lidone (average MW = 8000, PVPk8), cetyltrimethylammoniumbromide (99.72%, CTAB), 11-mercaptoundecanoic acid (95%,MUA), 2-mercaptopyridine (97%, MPy), 4-nitrobenzenethiol (80%,4NBT), 4-mercaptophenol (97%, 4MP), 4-mercaptobenzoic acid(99%, MBA), 3,5-bis(trifluoromethyl)benzenethiol (97%,35BTFMB), 4-fluorothiophenol (98%, 4FTP), 2,3,5,6-tetrafluoroben-zenethiol (97%, 2356TFBT), 2-(trifluoromethyl)benzenethiol (96%,2TFMBT), 3-fluorothiophenol (95%, 3FTP), Nile blue A (95%,NBA), 2-fluorothiophenol (97%, 2FTP), toluidine blue O (≥84%,TB), benzenethiol (97%, BT), 4-(((3-mercapto-5-(2-methoxyphenyl)-4H-1,2,4-triazol-4-yl)imino)methyl)phenol (97%, MMPHTYIMP), 4-(((3-mercapto-5-(2-pyridinyl)-4H-1,2,4-triazol-4-yl)imino)methyl)-benzoic acid (MPHTYIMBA), 4-(((3-mercapto-5-(2-pyridinyl)-4H-1,2,4-triazol-4-yl)imino)methyl)-1,2-benzenediol acid (MPHTYIMB-DO), 1-(4-hydroxyphenyl)-1H-tetrazole-5-thiol (97%, HPHTT),1,1′,4′,1″-terphenyl-4-thiol (97%, TPT), 1-naphtalenethiol (99%,1NT), 2-naphtalenethiol (99%, 2NT), 5-(4-methoxyphenil)-1,3,4-oxidazole-2-thiol (97%, MPOT), methylene blue (≥82%, MB), 3,4-dichlorobenzenethiol (97%, DCBT), biphenyl-4-thiol (97%, BPT), 7-mercapto-4-methylcoumarin (≥97%, MMC), biphenyl-4−4′-dithiol(95%, BPDT), thiosalicylic acid (97%, TSA), 5-amino-1,3,4-thiadiazole-2-thiol (87%, ATT), 4-aminothiophenol (97%, 4ATP), 2-phenylethanethiol (98%, 2PET), and crystal violet (≥90%, CV) werepurchased from Sigma-Aldrich Gmbh (Munich, Germany). Allreactants were used without further purification. Mili-Q water (18MΩ cm−1) was used in all aqueous solutions, and all glassware wascleaned with aqua regia before the experiments.

Synthesis of Citrate-Stabilized Spherical Gold Nanopar-ticles. Spherical gold nanoparticles of approximately 50-nm diameterwere produced by a modification of the well-known Turkevichmethod.60 Briefly, 308 μL of an aqueous solution of HAuCl4 (0.081M) was added to a boiling aqueous solution of sodium citrate (100mL, 0.27 mM) under vigorous stirring. Heating and stirring werecontinued for 30 min. A condenser was utilized to prevent theevaporation of the solvent. During this time, the color of the solutiongradually changed from colorless to purple to finally become deep red.After this time, heating was continued, and the condenser wasremoved to allow evaporation of the solvent to half its initial volume,to achieve a final gold concentration of 5 × 10−4 M.

Mercaptoundecanoic Acid Functionalization of Gold Nano-particles. To provide colloidal stability to the Au nanoparticles duringthe encoding process and later promote the growth of silica, 50 mL ofthe as-produced spherical gold nanoparticles was functionalized with asmall amount of MUA (0.8 molecules nm−2). To this end, a solutioncontaining NH4OH (879 μL, 29% aqueous solution) and MUA (1mL, 3.99 × 10−5 M in EtOH) was prepared. This solution was thenrapidly added under vigorous stirring to the gold nanoparticle sol (50mL). Agitation was continued for 30 min to ensure MUAfunctionalization of the Au surface.

Gold Nanoparticle Codification. With the aim of demonstratingthe versatility of the presented method, 31 different SERS-activemolecules were used, namely, MPy, 4NBT, 4MP, MBA, 35BTFMB,4FTP, 2356TFBT, 2TFMBT, 3FTP, NBA, 2FTP, TB, BT,MMPHTYIMP, HPHTT, TPT, 1NT, 2NT, MPOT, MB, DCBT,BPT, MMC, BPDT, CV, 2PET, 4ATP, ATT, TSA, MPHTYIMBDO,and MPHTYIMBA. The exact same procedure was used for eachmolecule. Briefly, a solution containing EtOH (324.89 mL) andNH4OH (5.73 mL, 29% aqueous solution) was prepared. Thissolution was then rapidly added under vigorous stirring to the MUA-stabilized gold nanoparticles (51.88 mL). Next, a stock solution of theSERS-active molecule was prepared (10−2 M, in EtOH), and 74.8 μLof this solution was added to the MUA-functionalized Au particles(382 mL) under strong magnetic stirring for 30 min (to ensure proper

Figure 6. SERS spectra of encoded Ag nanoparticles (yellow curve;[Ag] = 1 mM; Raman probe = 3,4-dicholorobenzenethiol, DBT) andgold nanostars (red curve; [Au] = 1 mM; Raman probe = 2-mercaptopyridine, MPy). Representative TEM images of thecorresponding encoded nanoparticles are also included.

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

955

Au functionalization). The amount of SERS-active molecules addedwas calculated to be 15 molecules nm−2.Silica Encapsulation. The silica encapsulation of the encoded

nanoparticles was achieved through a modified Stober method usingthe MUA carboxylic group to promote the growth of silica as follows:The proper concentrations of H2O, NH4OH, and EtOH for silicagrowth on the MUA/SERS-encoded particle solution were previouslyadjusted, during the codification step, to yield final concentrations of7.94, 0.128, and 14.60 M, respectively (EtOH/H2O molar ratio of1.84). Then, TEOS (13.2 μL) was added, and the solution wasenergetically shaken and left undisturbed at room temperature for 14h. Finally, the resulting core−shell nanoparticles were cleaned toremove excess reactants by centrifugation (3 × 6000 rpm, 20 min) andredispersed in ethanol. To concentrate the solution (10−3 M) toperform the SERS characterization, 15.3 mL of this solution wascentrifuged again (2 × 6000 rpm, 20 min) and redispersed in water.After the final centrifugation step, everything was resuspended in afinal volume of 1 mL.Synthesis of PVP-Based Spherical Encoded Gold Nano-

particles. Different SERS-active molecules were used, namely, 1NT,MPy, 4NBT, MBA, NBA, TB, and 2356TFBT. Spherical goldnanoparticles of approximately 50-nm diameter were produced aspreviously described. To 75 mL of a PVPk58 solution (0.69 mM) wasadded 50 mL of the citrate Au particles dropwise, and the mixture wasleft to react overnight under stirring. Next, the solution wascentrifuged (5400 rpm, 25 min) and redispersed in 50 mL of EtOH([Au] = 0.5 mM). To remove as much as possible of the excessPVPk58, this process was repeated four times. Then, the SERS codemolecule (74.8 μL, 10−2 M) was added under stirring for 2 h. Finally,silica coating was carried out through adjustment of the finalconcentrations as follows (in a 50 mL solution): [Au] = 0.5 mM,[H2O] = 10.55 M, [NH3] = 0.2 M, [EtOH] = 13.4 M, and [TEOS] =1.12 mM. The reaction mixture was allowed to react for 24 h. Whenthe reaction time was completed, the particles were centrifuged andwashed with ethanol. To concentrate the solution (10−3 M) toperform the SERS measurements, 2 mL of this solution wascentrifuged again (2 × 6000 rpm, 20 min) and redispersed in water.After the final centrifugation step, everything was resuspended in afinal volume of 1 mL.Synthesis of PEG-SH-Based Spherical Encoded Gold Nano-

particles. Different SERS-active molecules were used, namely, 1NT,MPy, 4NBT, MBA, NBA, TB, and 2356TFBT. First, CTAB-stabilizedgold nanospheres of 50-nm diameter were prepared by the seededgrowth approach previously described in the literature61 as follows: Aseed solution was made by preparing an aqueous solution (20 mL)containing HAuCl4 (2.5 × 10−4 M) and sodium citrate (2.5 × 10−4

M). As the mixture was vigorously stirred, NaBH4 (600 μL, 0.1 M)solution was added, and the mixture underwent a rapid color change tored, which indicates the formation of gold particles. The seeds wereleft stirring under the open atmosphere for 1 h to allow the NaBH4 todecompose. Next, a growth solution was prepared by dissolving CTAB(from Merck, 100 mL, 0.1 M) and potassium iodide (0.3 mg/g ofCTAB) in Milli-Q water and then adding HAuCl4 (510 μL, 0.103 M)and ascorbic acid (735 μL, 0.1 M). After each addition, the bottleswere vigorously shaken. Next, 187 μL of seeds was added, and thesolution was again vigorously shaken. The flask was left undisturbed at28 °C for 48 h. After this time, a small amount of gold particles wasobserved as sediment in the bottom of the flask. Because particles withdiameters of less than 100 nm are stable in solution, this precipitatemust be composed of larger Au structures coming from seeds with adifferent crystallographic structure. Carefully, the supernatant wascollected, and the precipitate discarded to ensure the monodispersityo f t h e p a r t i c l e s . The s e cond s t ep i n vo l v ed o - [ 2 -(3mercaptopropionylamino)ethyl]-o-methylpoly(ethylene glycol)(PEG-SH, MW = 5000) capping, ethanol transfer, and silica coating.To do this, 100 mL of the as-synthesized Au spheres ([Au] = 0.25mM, [CTAB] = 0.1 M) was centrifuged for 20 min (6000 rpm), andthe precipitate was redispersed in a CTAB solution ([CTAB] = 0.5mM) to clean the CTAB as much as possible without compromisingthe colloidal stability of the particles. This process was repeated three

times to finally redisperse in a final volume of 50 mL to obtain CTABand Au concentrations of ∼0.5 mM each. Next, a stock solution ofPEG-SH was prepared and sonicated for 15 min (10−3 M, in H2O),and 89.8 μL of this solution was added to the CTAB-stabilized Auparticles (50 mL) under strong magnetic stirring for 30 min (to ensureproper Au functionalization). The amount of PEG-SH added wascalculated to be 1.8 molecules nm−2. The PEG-modified particles werecentrifuged twice to remove excess PEG-SH and redispersed inethanol (50 mL); in the second centrifugation, the particles wereredispersed in a solution (50 mL) adjusted to the following finalconcentrations: [Au] = 0.5 mM, [H2O] = 10.55 M, [NH3] = 0.2 M,and [EtOH] = 13.39 M. The third step was the encoding of thenanoparticles. For this step, a stock solution of the SERS-activemolecule was prepared (10−2 M, in EtOH), and 74.8 μL of thissolution was added to the PEG-SH-functionalized Au particles (50mL) under strong magnetic stirring for 2 h. The amount of SERS-active molecules added was calculated to be 15 molecules nm−2.Finally, TEOS (13.20 μL) was added, and the solution wasenergetically shaken and left undisturbed at room temperature 14 h.To concentrate the solution ([Au] = 1 mM) to perform the SERSmeasurements, 2 mL of this solution was centrifuged again (2 × 6000rpm, 20 min) and redispersed in water. After the final centrifugationstep, everything was resuspended in a final volume of 1 mL. Ananoparticle concentration of ca. 0.24 nM was calculated by theLambert−Beer law using an extinction coefficient for gold nano-particles of 4.2 × 1010 M−1 cm−1, derived from the literature.62

Synthesis, Codification, and Silica Coating of Silver Nano-particles. Spherical silver nanoparticles (Ag NPs) of approximately50-nm diameter were produced by adding AA (50 μL, 0.10 mM) to47.5 mL of boiling water. One minute after the addition, a solutionpreviously incubated for 5 min containing AgNO3 (0.25 mL, 1 wt %),sodium citrate (1 mL, 1 wt %), and 1.25 mL of Milli Q water wasinjected into the boiling aqueous solution. The color of the reactionsolution quickly changed from colorless to yellow. The solution wasfurther boiled for 1 h under stirring to guarantee the formation ofuniform quasispherical silver nanoparticles.63

Then, the synthesized Ag NPs were cleaned by centrifugation (5400rpm, 30 min) and redispersed in a volume of Milli Q water to achieve asilver concentration of 2.5 × 10−4 M.

The Ag NPs were then functionalized with MUA (1.8 moleculesnm−2). Specifically, 5 mL of Ag NPs ([Ag] = 2.5 × 10−4 M) wasrapidly added to a solution containing NH4OH (662 μL, 29% aqueoussolution), MUA (41.6 μL, 1.0 × 10−4 M in EtOH), and EtOH (32.50mL) under vigorous stirring. Agitation was continued for 1 h to ensurethe complete MUA functionalization of the Ag surface. Then, 3.8 μLof a 10−2 M stock solution of different SERS probes was added understrong magnetic stirring. Once more, stirring was continued foranother 1 h. The amount of SERS-active molecules added wascalculated to be 15 molecules nm−2. Finally, TEOS (12 μL, 10% v/v)was added, and the solution was energetically shaken and leftundisturbed at room temperature for 14 h. The solution wascentrifuged and redispersed in water following the same procedureas for spherical gold nanoparticles.

The nanoparticle concentration was calculated to be 0.26 nM by theLambert−Beer law using the extinction coefficient for silver nano-particles of 5.37 × 1010 M−1 cm−1, derived from the literature.64

Synthesis, Codification, and Silica Coating of Gold Nano-stars. The synthesis of gold nanostars (GNSs) was performed using amodified PVP-based method.59 In the first step, spherical gold seeds ofapproximately 10-nm diameter were produced by a modification of thewell-known Turkevich method.60 Briefly, a solution of sodium citratein Milli-Q water (150 mL, 1.93 mM) was heated with a heating mantlein a 250 mL three-necked round-bottom flask for 15 min undervigorous stirring. A condenser was utilized to prevent the evaporationof the solvent. After boiling had commenced, 308 μL of HAuCl4(0.081 M) was injected. The color of the solution changed from yellowto bluish gray and then to soft pink in 10 min and to wine red after 30min. The resulting particles were coated with negatively charged citrateions and, hence, were well suspended in H2O. Then, the seeds wereconcentrated in a 250 mL beaker to a Au concentration of 9.45 × 10−4

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

956

M with a constant flow of N2 for 120 min. In the second step, asolution of PVPk8 (15 mM) in dimethylformamide (35 mL) wassonicated for 15 min. Then, 216 μL of HAuCl4 (0.081 M) was addedto the mixture under rapid stirring at room temperature, followed bythe concentrated seeds (1.108 mL, [Au] = 9.45 × 10−4 M). Within 5min, the color of the solution changed from pink to blue, indicatingthe formation of gold nanostars. The solution was left stirringovernight. Then, the excess PVPk8 was removed by a 6-foldcentrifugation (7000 rpm, 15 min) and redispersed in ethanol(particle solution was adjusted to yield a final concentration of [Au]= 4.5 × 10−4 M).The as-synthesized GNS were then functionalized with MUA (0.8

molecules nm−2) following the previously described protocol.Specifically, a solution containing NH4OH (87.9 μL, 29% aqueoussolution) and MUA (31.6 μL, 1.0 × 10−4 M in EtOH) was rapidlyadded under vigorous stirring to 5 mL of GNS suspension ([Au] = 4.5× 10−4 M). Agitation was continued for 3 h to ensure the completeMUA adhesion on the Au surface. Next, 5.9 μL of a 10−2 M stocksolution of different SERS probes was added under strong magneticstirring. Once more, stirring was continued for another 3 h. Theamount of SERS-active molecules added was calculated to be 15molecules nm−2.Finally, the same modified Stober method as applied to spherical

gold nanoparticles was followed to obtain a homogeneous silica shellaround the Raman-labeled GNSs. The colloidal suspension wasconcentrated to a Au concentration of 1 mM prior to the SERSmeasurements.Characterization. UV−vis spectroscopy (Perkin-Elmer, Lambda

19) and transmission electron microscopy (TEM, LEO 922 EFTEMoperating at 80 kV) were applied to characterize the optical response,structure, and size of the nanoparticles during the encoding process.SERS spectra were collected in backscattering geometry with aRenishaw Invia Reflex system equipped with a 2D-CCD detector and aLeica confocal microscope. The spectrograph used a high-resolutiongrating (1200 g cm−1) with additional band-pass filter optics. A 785-nm diode laser was focused onto the colloidal solution ([Au] = 1 mM)by a long-working-distance objective (0.17 NA, working distance = 30mm). The spectra were acquired with an exposure time of 1 s(depending on Raman intensity saturation) and a laser power at thesample of ca. 300 mW, using a Renishaw StreamLine accessory.

■ ASSOCIATED CONTENT*S Supporting InformationAdditional UV−vis−NIR, TEM, and SERS characterizationdata. SERS spectra of another 20 encoded particles.Comparison of SERS intensities of different encoded particlesprepared with MUA, PVP, and PEG for different codes. Thismaterial is available free of charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Authors*E-mail: npazos@medcomadvance.com.*E-mail: ramon.alvarez@urv.cat.Author ContributionsThe manuscript was written through contributions of allauthors.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was funded by Medcom Advance SA.

■ REFERENCES(1) Doerr, A. Nat. Methods 2007, 4, 381.(2) Fenniri, H.; Alvarez-Puebla, R. Nat. Chem. Biol. 2007, 3, 247.

(3) Wilson, R.; Cossins, A. R.; Spiller, D. G. Angew. Chem., Int. Ed.2006, 45, 6104.(4) Zhao, Y.; Zhao, X.; Hu, J.; Xu, M.; Zhao, W.; Sun, L.; Zhu, C.;Xu, H.; Gu, Z. Adv. Mater. 2009, 21, 569.(5) Rodriguez-Lorenzo, L.; Fabris, L.; Alvarez-Puebla, R. A. Anal.Chim. Acta 2012, 745, 10.(6) Shtengel, G.; Galbraith, J. A.; Galbraith, C. G.; Lippincott-Schwartz, J.; Gillette, J. M.; Manley, S.; Sougrat, R.; Waterman, C. M.;Kanchanawong, P.; Davidson, M. W.; Fetter, R. D.; Hess, H. F. Proc.Natl. Acad. Sci. U.S.A. 2009, 106, 3125.(7) Doering, W. E.; Piotti, M. E.; Natan, M. J.; Freeman, R. G. Adv.Mater. 2007, 19, 3100.(8) Sha, M. Y.; Walton, I. D.; Norton, S. M.; Taylor, M.; Yamanaka,M.; Natan, M. J.; Xu, C.; Drmanac, S.; Huang, S.; Borcherding, A.;Drmanac, R.; Penn, S. G. Anal. Bioanal. Chem. 2006, 384, 658.(9) Bake, K. D.; Walt, D. R. Annu. Rev. Anal. Chem. 2008, 1, 515.(10) Freeman, R. G.; Raju, P. A.; Norton, S. M.; Walton, I. D.; Smith,P. C.; He, L.; Natan, M. J.; Sha, M. Y.; Penn, S. G. Methods Mol. Biol.2005, 303, 73.(11) Braeckmans, K.; De Smedt, S. C. Nat. Mater. 2010, 9, 697.(12) Braeckmans, K.; De Smedt, S. C.; Roelant, C.; Leblans, M.;Pauwels, R.; Demeester, J. Nat. Mater. 2003, 2, 169.(13) Braeckmans, K.; De Smedt, S. C.; Leblans, M.; Pauwels, R.;Demeester, J. Nat. Rev. Drug Discovery 2002, 1, 447.(14) Nicewarner-Pena, S. R.; Freeman, R. G.; Reiss, B. D.; He, L.;Pena, D. J.; Walton, I. D.; Cromer, R.; Keating, C. D.; Natan, M. J.Science 2001, 294, 137.(15) Finkel, N. H.; Lou, X.; Wang, C.; He, L. Anal. Chem. 2004, 76,352 A.(16) Cunin, F.; Schmedake, T. A.; Link, J. R.; Li, Y. Y.; Koh, J.;Bhatia, S. N.; Sailor, M. J. Nat. Mater. 2002, 1, 39.(17) Pregibon, D. C.; Toner, M.; Doyle, P. S. Science 2007, 315,1393.(18) Vaino, A. R.; Janda, K. D. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,7692.(19) Fenniri, H.; Ding, L.; Ribbe, A. E.; Zyrianov, Y. J. Am. Chem. Soc.2001, 123, 8151.(20) Raez, J.; Blais, D. R.; Zhang, Y.; Alvarez-Puebla, R. A.; Bravo-Vasquez, J. P.; Pezacki, J. P.; Fenniri, H. Langmuir 2007, 23, 6482.(21) Lee, J.; Bisso, P. W.; Srinivas, R. L.; Kim, J. J.; Swiston, A. J.;Doyle, P. S. Nat. Mater. 2014, 13, 524.(22) Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P.Science 1998, 281, 2013.(23) Sanles-Sobrido, M.; Exner, W.; Rodriguez-Lorenzo, L.;Rodriguez-Gonzalez, B.; Correa-Duarte, M. A.; Alvarez-Puebla, R. A.;Liz-Marzan, L. M. J. Am. Chem. Soc. 2009, 131, 2699.(24) Goddard, G.; Brown, L. O.; Habbersett, R.; Brady, C. I.; Martin,J. C.; Graves, S. W.; Freyer, J. P.; Doorn, S. K. J. Am. Chem. Soc. 2010,132, 6081.(25) Howarth, M.; Liu, W.; Puthenveetil, S.; Zheng, Y.; Marshall, L.F.; Schmidt, M. M.; Wittrup, K. D.; Bawendi, M. G.; Ting, A. Y. Nat.Methods 2008, 5, 397.(26) Serge, A.; Bertaux, N.; Rigneault, H.; Marguet, D. Nat. Methods2008, 5, 687.(27) Qian, X. M.; Peng, X. H.; Ansari, D. O.; Yin-Goen, Q.; Chen, G.Z.; Shin, D. M.; Yang, L.; Young, A. N.; Wang, M. D.; Nie, S. M. Nat.Biotechnol. 2008, 26, 83.(28) Rivera_Gil, P.; Vazquez-Vazquez, C.; Giannini, V.; Callao, M.P.; Parak, W. J.; Correa-Duarte, M. A.; Alvarez-Puebla, R. A. Angew.Chem., Int. Ed. 2013, 52, 13694.(29) Pazos-Perez, N.; Wagner, C. S.; Romo-Herrera, J. M.; Liz-Marzan, L. M.; García de Abajo, F. J.; Wittemann, A.; Fery, A.; Alvarez-Puebla, R. A. Angew. Chem., Int. Ed. 2012, 51, 12688.(30) Song, J.; Zhou, J.; Duan, H. J. Am. Chem. Soc. 2012, 134, 13458.(31) Qian, X.; Peng, X.-H.; Ansari, D. O.; Yin-Goen, Q.; Chen, G. Z.;Shin, D. M.; Yang, L.; Young, A. N.; Wang, M. D.; Nie, S. Nat.Biotechnol. 2008, 26, 83.(32) Guerrero-Martínez, A.; Perez-Juste, J.; Liz-Marzan, L. M. Adv.Mater. 2010, 22, 1182.

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

957

(33) Jiang, R.; Li, B.; Fang, C.; Wang, J. Adv. Mater. 2014, 26, 5274.(34) Liu, S.; Han, M.-Y. Chem.Asian J. 2010, 5, 36.(35) Song, J.; Duan, B.; Wang, C.; Zhou, J.; Pu, L.; Fang, Z.; Wang,P.; Lim, T. T.; Duan, H. J. Am. Chem. Soc. 2014, 136, 6838.(36) Lim, D. K.; Jeon, K. S.; Hwang, J. H.; Kim, H.; Kwon, S.; Suh, Y.D.; Nam, J. M. Nat. Nanotechnol. 2011, 6, 452.(37) Alba, M.; Pazos-Perez, N.; Vaz, B.; Formentin, P.; Tebbe, M.;Correa-Duarte, M. A.; Granero, P.; Ferre-Borrull, J.; Alvarez, R.;Pallares, J.; Fery, A.; de Lera, A. R.; Marsal, L. F.; Alvarez-Puebla, R. A.Angew. Chem., Int. Ed. 2013, 52, 6459.(38) Alvarez-Puebla, R. A.; Agarwal, A.; Manna, P.; Khanal, B. P.;Aldeanueva-Potel, P.; Carbo-Argibay, E.; Pazos-Perez, N.; Vigderman,L.; Zubarev, E. R.; Kotov, N. A.; Liz-Marzan, L. M. Proc. Natl. Acad.Sci. U.S.A. 2011, 108, 8157.(39) Su, X.; Zhang, J.; Sun, L.; Koo, T. W.; Chan, S.; Sundararajan,N.; Yamakawa, M.; Berlin, A. A. Nano Lett. 2005, 5, 49.(40) Sun, L.; Sung, K.-B.; Dentinger, C.; Lutz, B.; Nguyen, L.; Zhang,J.; Qin, H.; Yamakawa, M.; Cao, M.; Lu, Y.; Chmura, A. J.; Zhu, J.; Su,X.; Berlin, A. A.; Chan, S.; Knudsen, B. Nano Lett. 2007, 7, 351.(41) Fernandez-Lopez, C.; Mateo-Mateo, C.; Alvarez-Puebla, R. A.;Perez-Juste, J.; Pastoriza-Santos, I.; Liz-Marzan, L. M. Langmuir 2009,25, 13894.(42) Wang, Y. Q.; Yan, B.; Chen, L. X. Chem. Rev. 2013, 113, 1391.(43) Ghosh, P.; Han, G.; De, M.; Kim, C. K.; Rotello, V. M. Adv.Drug Delivery Rev. 2008, 60, 1307.(44) Goulet, P. J. G.; Aroca, R. F. Anal. Chem. 2007, 79, 2728.(45) Goulet, P. J. G.; Pieczonka, N. P. W.; Aroca, R. F. Anal. Chem.2003, 75, 1918.(46) Hiramatsu, H.; Osterloh, F. E. Langmuir 2003, 19, 7003.(47) Bryant, M. A.; Pemberton, J. E. J. Am. Chem. Soc. 1991, 113,8284.(48) Guerrini, L.; Izquierdo-Lorenzo, I.; Garcia-Ramos, J. V.;Domingo, C.; Sanchez-Cortes, S. Phys. Chem. Chem. Phys. 2009, 11,7363.(49) Guerrini, L.; Izquierdo-Lorenzo, I.; Rodriguez-Oliveros, R.;Sanchez-Gil, J. A.; Sanchez-Cortes, S.; Garcia-Ramos, J. V.; Domingo,C. Plasmonics 2010, 5, 273.(50) Stober, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26,62.(51) Xue, C.; Chen, X.; Hurst, S. J.; Mirkin, C. A. Adv. Mater. 2007,19, 4071.(52) Chen, T.; Chen, G.; Xing, S.; Wu, T.; Chen, H. Chem. Mater.2010, 22, 3826.(53) Wallace, A. F.; DeYoreo, J. J.; Dove, P. M. J. Am. Chem. Soc.2009, 131, 5244.(54) Wong, Y. J.; Zhu, L.; Teo, W. S.; Tan, Y. W.; Yang, Y.; Wang,C.; Chen, H. J. Am. Chem. Soc. 2011, 133, 11422.(55) Rycenga, M.; Langille, M. R.; Personick, M. L.; Ozel, T.; Mirkin,C. A. Nano Lett. 2012, 12, 6218.(56) Rodriguez-Lorenzo, L.; Krpetic, Z.; Barbosa, S.; Alvarez-Puebla,R. A.; Liz-Marzan, L. M.; Prior, I. A.; Brust, M. Integr. Biol. 2011, 3,922.(57) Simakova, P.; Gautier, J.; Prochazka, M.; Herve-Aubert, K.;Chourpa, I. J. Phys. Chem. C 2014, 118, 7690.(58) Murphy, C. J.; Gole, A. M.; Stone, J. W.; Sisco, P. N.; Alkilany,A. M.; Goldsmith, E. C.; Baxter, S. C. Acc. Chem. Res. 2008, 41, 1721.(59) Barbosa, S.; Agrawal, A.; Rodriguez-Lorenzo, L.; Pastoriza-Santos, I.; Alvarez-Puebla, R. A.; Kornowski, A.; Weller, H.; Liz-Marzan, L. M. Langmuir 2010, 26, 14943.(60) Enustun, B. V.; Turkevich, J. J. Am. Chem. Soc. 1963, 85, 3317.(61) Pazos-Perez, N.; Garcia de Abajo, F. J.; Fery, A.; Alvarez-Puebla,R. A. Langmuir 2012, 28, 8909.(62) Yguerabide, J.; Yguerabide, E. E. Anal. Biochem. 1998, 262, 137.(63) Li, H.; Xia, H.; Wang, D.; Tao, X. Langmuir 2013, 29, 5074.(64) Paramelle, D.; Sadovoy, A.; Gorelik, S.; Free, P.; Hobley, J.;Fernig, D. G. Analyst 2014, 139, 4855.

Chemistry of Materials Article

DOI: 10.1021/cm504251hChem. Mater. 2015, 27, 950−958

958

S1

Supporting Information

Universal One-pot and Scalable Synthesis of SERS

Encoded Nanoparticles

Bernat Mir-Simon,a Irene Reche-Perez,

b Luca Guerrini,

a,b Nicolas Pazos-Perez*

,a and Ramon

A. Alvarez-Puebla*,a,b,c

a Medcom Advance SA, Viladecans business park - Edificio Brasil, Bertran i Musitu 83 -

85, 08840 Viladecans – Barcelona, Spain. b Departamento de Quimica Fisica e Inorganica, Universitat Rovira i Virgili and Centro

de Tecnologia Quimica de Cataluña, Carrer de Marcel•lí Domingo s/n, 43007 Tarragona, Spain. c ICREA, Passeig Lluís Companys 23, 08010 Barcelona, Spain.

S2

Figure S1. Representative TEM image (A) and size distribution histograms (B) of the

synthesized citrate-capped gold nanoparticles.

S3

Figure S2. SERS spectra of Au particles coated with a full monolayer of MUA showing its

small Raman cross section.

S4

Figure S3. Comparison of the SERS spectra of CV at two different MUA molecules nm

-2 (0.8

black and 4 red). Showing that almost no signal is obtained when 4 molecules nm-2

were used.

S5

Figure S4. (A, B, and C) TEM images showing the results of encoding nanoparticles at MUA

concentrations below 0.8 molecules nm-2

(0.7, 0.6 and 0.5 respectively). (D) TEM image

showing the uncontrolled aggregation of the particles below 0.7 molecules nm-2

.

S6

Figure S5. SERS spectra of 20 representative SERS-encoded nanoparticles. Left column:

Toluidine Blue O, 2-Phenylethanethiol, 4-Mercaptophenol, Biphenyl-4-thiol, 7-Mercapto-4-

methylcoumarin, 4-Hydroxyphenyl)-1H-tetrazole-5-thiol, 2-Fluorothiophenol, Crystal Violet,

2-Naphthalenethiol, 4-(((3-Mercapto-5-(2-methoxyphenyl)-4H-1,2,4-triazol-4-

yl)imino)methyl)phenol. Right column: (2-Trifluoromethyl)benzenethiol, 4-Aminothiophenol,

1-Naphthalenethiol, 1,1',4,1''-Terphenyl-4-Thiol, Biphenyl-4,4'-dithiol, Thiosalicylic acid, 4-

(((3-Mercapto-5-(2-pyridinyl)-4H-1,2,4-triazol-4-yl)imino)methyl)-1,2-benzenediol, 4-(((3-

Mercapto-5-(2-pyridinyl)-4H-1,2,4-triazol-4-yl)imino)methyl)benzoic, 2,3,5,6-

Tetrafluorobenzenethiol, (5-(4-Methoxyphenyl)-1,3,4-oxidazole-2-thiol).

S7

Figure S6. (A) Photography of the one-pot synthesis of a liter batch of MPy encoded

nanoparticles. (B) UV-Vis, (C) SERS spectra, and (D) TEM image of the encoded

nanoparticles.

S8

Figure S7. (A) Extinction spectra of the Au@PEG-SH nanoparticles with different amount of

PEG-SH (molecules nm-2

). Below 3.5 PEG-SH molecules nm-2

, it is clearly seen that

nanoparticles start to aggregate. (B) SERS spectra of Au@PEG-SH/MPy nanoparticles with

different amount of PEG-SH, 4, 3.5, and 3 molecules nm-2

(red, blue, and magenta,

respectively).The MPy SERS spectra is only obtained below 3.5 molecules nm-2

. (C)

Comparison of the extinction spectra of Au@PEG-SH nanoparticles with different amounts of

PEG-SH (molecules nm-2

) before and after the addition of MPy. Showing that after addition

of MPy, the particles aggregate more especially below 3.5 molecules nm-2

of PEG-SH.

S9

Figure S8. SERS spectra comparison of encoded gold nanoparticles prepared using the PVP,

PEG-SH and MUA approaches (blue, red, and green respectively) of MBA, 4NBT, NBA, TB,

TFBT, and 1NT.

S10

Figure S9. Normal Raman spectrum of EtOH and MBA (0.1 M in EtOH). SERS spectrum of

Au@MBA@SiO2 suspension in EtOH ([Au] = 1 mM), [NP] = 0.244 nM). All the spectra

were acquired under the same experimental conditions (785 nm laser line, 100% laser power,

1 s exposition time, 5 accumulations).

The SERS enhancement factor was calculated using the following definition:

where NVol is the average number of molecules in the scattering volume, V, for the normal

Raman measurement, and NSurf is the average number of adsorbed molecules in the same

scattering volume for the SERS measurement.1 As normal Raman and SERS intensities (IRaman

and ISERS) we selected the height of the ν(C=C) band at ca. 1600 cm-1

. Both Raman and SERS

measurements were performed under the same experimental set-up (i.e. identical scattering

volume). Nanoparticle concentration of 0.244 nM was calculated by Lambert-Beer’s law

using the extinction coefficient for gold nanoparticles of 4.2 × 1010

M-1

cm-1

, derived from

literature.2 The molecular footprint of MBA on gold, for a full monolayer, was reported to be

0.44 nm2.3 Thus, for gold spheres of ca. 50 nm diameter functionalized with a monolayer of

MBA molecules, we estimate to have ca. 18570 molecules per particle, which corresponds to

4.53 ×10-6

M in the colloidal suspension. The estimated EF at 785 nm is therefore ca. 5.3 ×

10-4

M.

800 1000 1200 1400 1600

50k

100k

150k

200k

250k

300k

350k

400k

450k

500k

SE

RS

in

ten

sit

y (

cp

s)

Raman shift (cm-1)

EtOH

MBA 0.1 M in EtOH

Au@MBA@SiO2 ...nM in EtOH

800 1000 1200 1400 1600

0

50k

100k

150k

200k

250k

300k

SE

RS

in

ten

sit

y (

cp

s)

Raman shift (cm-1)

EtOH

MBA

Au@MBA@SiO2

S11

Figure S10. SERS spectra and TEM images of different SERS encoded gold nanostars

prepared using the MUA approach. Normalized extinction spectra of GNS@MPy@SiO2

encoded nanoparticles.

S12

Figure S11. SERS spectra and TEM images of different SERS encoded silver nanoparticles

prepared using the MUA approach. Normalized extinction spectra of Ag@DBT@SiO2

encoded nanoparticles.

REFERENCES

(1) Le Ru, E. C.; Etchegoin, P. G., Principles of Surface-Enhanced Raman Spectroscopy.

2009.

(2) Yguerabide, J.; Yguerabide, E. E., Anal. Biochem. 1998, 262, 137.

(3) Hiramatsu, H.; Osterloh, F. E., Langmuir 2003, 19, 7003.