Recent progress on colloidal metal nanoparticles as

Advances in Colloid and Interface Science 233 (2016) 255–270

Contents lists available at ScienceDirect

Advances in Colloid and Interface Science journal homepage: www.elsevier.com/locate/cis

Historical perspective

Recent progress on colloidal metal nanoparticles as signal enhancers in nanosensing Sara Abalde-Cela a, Susana Carregal-Romero b,c, João Paulo Coelho d, Andrés Guerrero-Martínez d,⁎ a

Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, UK Bionanoplasmonics Laboratory, CIC biomaGUNE, Paseo de Miramón 182, 20009 Donostia-San Sebastián, Spain Oncology Area, Biodonostia Research Institute, Donostia-San Sebastián 20014, Spain d Departamento de Química Física I, Universidad Complutense de Madrid, Avda. Complutense s/n, 28040 Madrid, Spain b c

a r t i c l e

i n f o

Available online 11 June 2015 Keywords: Colloidal metal nanoparticle Nanoplasmonics Surface-enhanced spectroscopy SEF SERS Sensing

a b s t r a c t Colloidal metal nanoparticles present very special optical and electromagnetic properties at the nanoscale range. Such plasmonic properties have derived in a huge research field that encompasses the understanding of nanoparticle formation mechanisms for the ultimate goal of developing novel materials for real-life applications. Plasmonic sensing is experiencing a rapid transition by taking advantage of the characteristic properties of colloidal metal nanoparticles. However, a rational design of novel nanoplasmonic substrates, which gathers as much as the required properties for a substrate to be a ‘good’ sensor is critical through the development of applications that can be effectively transferred as applied technologies. Also, the chosen sensing technique is a key factor when planning the design of a new plasmonic-based sensor. Several factors such as composition, shape, size, particle interactions or stability among others will define the final quality of the nanomaterial as sensing platform. Herein, we review the latest and most promising state-of-the art of nanoplasmonic-based sensors in four differentiated areas regarding the surface-enhanced spectroscopy detection technique being LSPR-, SERSand SEIRA-, and SEF based platforms. © 2015 Elsevier B.V. All rights reserved.

Contents 1. 2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . LSPR-based colloidal plasmonic nanosensors . . . . . . . . . . . . . . . . . . 2.1. LSPR principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. LSPR-based nanosensing using colloidal metal NPs in solution . . . . . . . 2.3. LSPR-based nanosensing using colloidal metal NPs on substrates . . . . . . 3. SERS- and SEIRA-based colloidal plasmonic nanosensors . . . . . . . . . . . . . 3.1. SERS and SEIRA principle . . . . . . . . . . . . . . . . . . . . . . . 3.2. SERS and SEIRA-based nanosensing using colloidal metal NPs in solution . . 3.3. SERS- and SEIRA-based nanosensing using colloidal metal NPs on substrates 4. SEF-based colloidal plasmonic nanosensors . . . . . . . . . . . . . . . . . . . 4.1. SEF principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. SEF-based nanosensing using colloidal metal NPs in solution . . . . . . . 4.3. SEF-based nanosensing using colloidal metal NPs on substrates . . . . . . 5. Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

255 257 257 257 258 259 259 261 263 263 263 265 265 266 268 268

1. Introduction ⁎ Corresponding author. E-mail address: [email protected] (A. Guerrero-Martínez).

http://dx.doi.org/10.1016/j.cis.2015.05.002 0001-8686/© 2015 Elsevier B.V. All rights reserved.

Noble metals, gold and silver in particular, exhibit interesting optical properties at the nanoscale that markedly differ from those of the bulk

256

S. Abalde-Cela et al. / Advances in Colloid and Interface Science 233 (2016) 255–270

material and the isolated atomic constituents, due to the excitation of localized surface plasmon resonances (LSPRs) [1]. These optical responses can be tuned by changing different parameters such as the size, shape, and composition of the nanoparticles (NPs). The intense and highly confined electromagnetic fields induced by the LSPR at the NPs provide a very sensitive tool to detect small changes in their dielectric environments, property that is particularly attractive for sensing applications (Fig. 1) [2]. Therefore, building (bio)chemical sensors with NPs have evolved into a relevant research direction with specialized journals and recent news focus in major journals [3]. The excitation of LSPRs is accompanied by strong subwavelength confinement and resonant enhancement of light near the NP boundary [4], distance-dependent effects that find many applications in surfaceenhanced spectroscopies [5]. In this context, surface-enhanced Raman scattering (SERS), surface-enhanced infrared absorption (SEIRA) and surface-enhanced fluorescence (SEF) spectroscopies are three of the most powerful analytical techniques for the identification of molecular species at low concentrations, which can even reach single-molecule detection [6]. Additionally, SERS, SEIRA and SEF provide complete vibrational and electronic information of the molecular system under study, since the outputs are essentially infrared absorption (IR), Raman scattering and fluorescence spectra, respectively [7,8]. The enhancement of the spectroscopic signal is mainly achieved by coupling of the vibrational and electronic modes of the analyte with the LSPR generated at the surface of the NP, upon excitation with light of appropriate energy. Moreover, particle aggregates have been found to provide much higher enhancement due to coupling between the LSPRs of adjacent particles, resulting in significantly higher electromagnetic fields so-called hotspots [9]. All together, these features make NP aggregates extremely interesting nanoplasmonic structures for the sensing of chemical analytes, such as pollutants with environmental and biomedical impacts [10]. Currently, the advances in wet chemical synthesis allow the preparation of colloidal plasmonic NPs with different metal compositions (mainly gold and silver), morphologies (spheres, rods, stars and polyhedra), and nanocrystal sizes (10–100 nm) [11]. Therefore, almost any

desired LSPR response can be achieved through wet chemical synthesis of NPs [12]. The most successful methodology for controlling size and shape of gold nanoparticles (AuNPs) is the seed-mediated growth method, which comprises the reduction of a gold salt precursor with a weak reducing agent in the presence of preformed seeds, including stabilizing molecules in aqueous solution [13]. For example, highly monodisperse gold nanospheres ranging from 15 to 100 nm have been recently synthesized following this strategy [14]. The same procedure is used for the synthesis of gold nanorods, where aspect ratios ranging from 2 to 5 can be efficiently achieved [15]. In the case of silver NPs (AgNPs), the high reactivity of silver precursors and the relatively low chemical stability of silver seeds have severely limited the use of the seed-mediated growth method in the direct synthesis of AgNPs with controlled size and shape [11]. Therefore, novel strategies comprising the kinetic control of the seeded growth method by combination of several reducing agents or the controlled reduction of a silver shell on pre-formed AuNPs as cores have been recently used to synthesize AgNPs with several morphologies [16]. Most of the times, to make use of NPs for applied sensing research, access to well-stabilized colloidal NP samples in solution is necessary. Through careful choice of surface functionalization processes, such as silica [17], polymer [18], and protein [19] coating procedures, one can handle stable aqueous- or organic-soluble NPs tailored to possess desired chemical and optical properties in solution. The rational search for functionalized NP surfaces with appropriate supramolecular hosts [20] and antibodies [21] has opened the possibility of sensing guest analytes through specific physicochemical and biochemical interactions. Following molecular concepts at the nanoscale, nanocrystals coated with selected molecules have been arranged to build up solid aggregates of NPs with different levels of crystallinity [12]. Thus, the idea that molecules self-assemble in different fashions as they change from the disordered isotropic liquid state to the ordered crystalline phase has been explored in the field of plasmonic supercrystals, showing interesting collective sensing responses [22]. Amphiphilic molecules

Fig. 1. Scheme showing the different types of optical interactions between NPs and molecules. (A) LSPR shift induced by adsorbed molecules without resonance absorption coupling. (B) Resonance coupling between NP LSPRs and strong absorptions of adsorbed molecules. (C) Luminescence enhancement of fluorophores close to NPs. From Ref. [33].


have gained popularity as a surface template in the formation of selfassembled NP supercrystals [23], providing fine control over interparticle spacing that can be used to tune the plasmonic properties of the supercrystal over micrometer length scales [24]. Indeed, the types of superlattices obtained in these assemblies have been controlled by varying the length and flexibility of the capping agents [25]. Such specific assembly of NPs has served as a platform for building-up devices with applications in plasmonic sensing [23]. The fabrication of NP supercrystals with high optical activity and long-range sensing areas has been recently exploited to obtain enhanced electric field concentration to maximize the spectroscopic signal of biorelevant chemical analytes at ultralow concentrations [26]. However, although colloid chemistry principles have emerged as powerful concepts for devising novel plasmonic nanosensors, becoming a natural starting point for the bottom-up fabrication of sensing analytical devices in solution and on substrates [12], there are inherent difficulties related to the colloidal stability and crystallization of NPs that still need to be overcome: (i) In colloidal solution, the relatively low enhancements of spectroscopic signals in NP colloids, which are related to difficulties in the tuning of distance between analytes and the NP surface [27], added to problems of self-assembly control. (ii) As colloidal substrates, the low control of NP distances and packing within the aggregates and supercrystals as well as the absence of macroscopic homogeneities of the nanoplasmonic substrates, which are directly related to non-controlled drying effects of NP colloidal solutions on substrates [28]. As a consequence of these experimental uncertainties, it should be pointed that there is a lack of general and versatile commercially available LSPR-, SERS-, SEIRA- and SEF-based sensors fabricated from colloidal NPs, being only few of them obtained through lithography techniques [29]. Additionally, the lack of reusability of almost all plasmonic nanosensors considerably raises their prices and disposal needs, limiting their uses in practical detection of real samples. Thus, in order to achieve homogeneous and reliable detection of chemical analytes through surface-enhanced spectroscopies, novel nanosensors with higher stabilities, macroscopic and nanoscopic homogeneities and reusable capability of sensing are needed, and may involve the search of novel methodological tools for NP preparation. Although many different LSPR-, SERS-, SEIRA- and SEF-based nanosensors have been explored so far, we intend to present herein an overview of the most recent activities in this area, restricting ourselves to those that have been dedicated to overcome such colloidal limitations in solution and substrates. 2. LSPR-based colloidal plasmonic nanosensors 2.1. LSPR principle Nanoplasmonic sensing is based on the oscillation of the conduction electrons of NPs at the LSPR, which leads to intense light absorption and scattering effects at visible wavelengths [4]. As a simplest example, the optical properties of spherical NP colloids with a radius R can be determined by the Mie theory [30], through expressions for the extinction cross-section Cext (sum of absorption and scattering). For nanosized particles with a frequency dependent, complex dielectric function ε = ε′ + iε″, embedded in a medium of dielectric constant εm, Cext can be calculated from:

C ext ¼

3 = 24πR3 εm 2 ε″ : 2 ″ λ ðε þ 2εm Þ þ ε0

The surface polarization charge induced by the dipole created by the applied electric field of light acts as a restoring force for the free

257

electrons. Therefore, when ε′ = −2εm condition is fulfilled, the wavelength absorption by the bulk metal is condensed into the LSPR at the wavelength of the absorbing radiation λ. Therefore, not only in nanospheres but also in NPs with other morphologies such as ellipsoids, tiny changes of εm caused by (bio)chemical analytes at the surroundings of NPs may induce significant wavelength shifts of the LSPRs (Fig. 1A). Thus, LSPR-based plasmonic sensing relies on the fact that the plasmon resonance frequency is highly sensitive to the molecular environment [31], the proximity of another particle [4], and the NP surface charge [32]. Recently, on the basis of molecular orbital theory, a plasmon hybridization model has been proposed that predicts how plasmon resonances vary upon molecular interaction [33] and clustering of plasmonic particles [34] (Fig. 1). The plasmon hybridization model has been a fundamental concept in designing plasmon rulers [35], in which interparticle distances and plasmon shifts are correlated through a simple mathematical expression. In practical terms, a plasmon ruler allows monitoring single-molecule biophysics (e.g., transformation of biomolecules) even in three dimensions (3D), which may have significant impact on the future understanding of mechanisms behind biological processes. 2.2. LSPR-based nanosensing using colloidal metal NPs in solution Although the position of the LPSR in colloidal NPs is sensitive to the environment dielectric constant, normally the LSPR shifts in the presence of adsorbed analytes at low concentrations are typically in the range of a few nanometers, thus limiting the sensitivity of the nanosensors. In order to overcome such limitation, several groups have demonstrated that molecules whose absorption bands match the LSPRs of AuNPs, within the red region of the visible spectrum, show larger plasmon shifts (Fig. 1B) [33]. This resonance coupling effect has been shown by Wang and co-workers [36], who constructed hybrid nanostructures of red absorbing dyes and rod-shape AuNPs through electrostatic interactions, in which wavelength shifts of ~100 nm were observed. Interestingly, a maximum of ~ 3 nm distance between the dyes and the AuNP surface was necessary to observe significant shifts of the LSPRs. The use of monodisperse colloidal AgNPs as LSPR-based sensors through resonance coupling effects may broaden a variety of opportunities of detection in solution of molecular species that absorb at the blue and green edge of the visible spectrum. Of particular interest is the development of controlled selfassembled NPs in solution, in which the interaction between closepacked NPs offers an indirect analytical way of sensing of analytes at the interparticle gaps. One of the most representative examples is the aggregation-based immunoassay that makes use of NPs functionalized with antibodies complementary to proteins, which induce the aggregation to create colorimetric responses [37], as in home pregnancy tests. However, the control of self-assembly to produce arrays with defined geometrical arrangements of the nanoparticles in solution remains a difficult task [38], being mainly focused on the preparation of dimers of NPs. Among them, the plasmon ruler concept can be considered as one of the most successful examples in terms of control and reproducibility. Reinhard et al. have demonstrated experimentally this concept by separation of two AuNPs in solution using a conjugated DNA as linker [35]. Accurate control over the AuNP separation at the nanometer range was obtained by conjugating one colloidal solution with a single-strand DNA and conjugating a second colloidal solution with the complementary single-strand DNA. The obtained LSPR shifts of AuNPs can be directly correlated to the flexibility of the molecular spacer and thus to its structural conformation. As an alternative approach, the changes of the LSPR shape during a chemical reaction have been used to determine the concentration of analytes in solution. For example, Liz-Marzán and co-workers have developed a novel colorimetric assay for the detection of subnanomolar concentrations of biochemical inhibitors [39]. In this method, the

258


enzyme acetylcholinesterase decomposes acetylthiocholine (a central neurotransmitter widely distributed in our nervous system) to generate a thiol-derivative compound that has been used to tune the growth of rod-shaped AuNPs. Thus, at different experimental conditions of the enzymatic reaction, the final growth solution contained a mixture of AuNPs with different morphologies (rods, cubes and spheres) with controllable relative populations, strongly affecting the overall optical response of the system. Therefore, by analyzing the LSPR of rodshaped AuNPs, the concentration of paraoxon (a molecular inhibitor of acetylcholinesterase that is used in nerve gas) can be determined at the nanomolar regime (Fig. 2). As the stability and aggregation of (bio)polymer-capped NPs are highly sensitive toward external stimuli, due to the significant structural modifications that such macromolecules register under tiny changes of the (bio)chemical environment [40], a high number of reports have been devoted to LSPR-based nanosensing of parameters such as pH [41], heavy metals [42], and temperature [43], using NPs in colloidal solutions. In an elegant approach, Chanana and co-workers have designed dual-responsive NPs coated by proteins and grafted with thermosensitive polymer brushes that present pH- and thermoresponsive behavior [41]. Such preparation led to AuNPs with lower critical solution temperatures of 42 °C in pure water and around 37 °C under physiological conditions, which were followed by LSPR coupling. Jointly, the pH-induced aggregation at the isoelectric point of the protein was completely reversible and could be repeated multiple times. Additionally, the same research group synthesized insulin-capped AuNPs that showed highly sensitive and selective LSPR responses in the presence of heavy metals, such as Fe2 +, Cu2 +, Pb2 +, and Hg2 +, due to strong metal–protein interactions that favor the aggregation phenomena [42]. In connection to UV/Vis spectroscopy, circular dichroism (CD), defined as the difference in extinction of left- and right-circularly polarized light, is one of the most commonly used spectroscopic techniques for the sensing of chiral systems such as organic compounds and biomolecules [44]. Within a nanoplasmonic context, CD has been explored to date using chiral NPs resulting in intense LSPR-mediated circular dichroism (LSPR-CD) signals over a wide spectral range across the visible (Fig. 3) [45]. Although in many examples the mechanisms behind optical activity in NPs cannot be easily identified, it may in principle

be restricted to two distinct origins: (i) NPs with individual chirality and (ii) collective interactions between 3D ordered NPs (Fig. 3). Recently, interesting studies of Govorov and co-workers have shown the possibility of sensing peptides and proteins inducing LSPR-CDs in individual chiral AuNPs [46]. Moreover, a recent report of Kotov and co-workers has illustrated the promising potential of 3D chiral assemblies of rodshape AuNPs to produce LSPR-CD signals of outstanding intensity [47], which have been used in the detection of DNA at attomolar concentrations (Fig. 3).

2.3. LSPR-based nanosensing using colloidal metal NPs on substrates To apply nanoplasmonic sensing on substrates, colloidal NPs have been mainly investigated at the single particle level by dark-field microscopy [48]. In this technique, the large light scattering efficiency at the LSPR makes the visualization of single NPs down to sizes of about 10 nm in diameter. In this sense, Sönnichsen and co-workers have detected single unlabeled proteins with extremely high temporal resolution, allowing for monitoring the dynamic evolution of proteins [48]. In an advanced design, the same group has developed an efficient and cost-effective multiplexed sensor for proteins, in which the LSPR position of rod-shape AuNPs responds specifically to different proteins [49]. The efficiency of this original method lies in the nanomolar sensitivity, the sensor recycling capability, and the potential to upscale to hundreds of targets. Similar single NP approaches have been developed at the macroscopic level, in which label-free, chip-based biosensors have been fabricated by random chemisorption of AuNPs onto mercaptosilane-modified glass substrates, followed by conjugation of sensitive proteins. Chilkoti and co-workers have monitored the streptavidin binding to biotin by the wavelength shift of the LSPR peak of immobilized rod-shape AuNPs due to changes in local refractive index, obtaining a nanomolar detection limit in serum [50]. An analogous substrate has been fabricated by Lin et al., in which a fiber-based biosensor utilizing the LSPR effect is used to evaluate the amount of paraoxon in solution [51]. In this study, a bioactive layer consisting of acetylcholinesterase is immobilized by covalent coupling onto an AuNP layer, allowing the determination of paraoxon at ppb limits by inhibition of acetylcholinesterase activity.

Fig. 2. (A) Scheme showing the inhibition of acetylcholinesterase activity by paraoxon. (B) Absorbance spectra of the growth products formed in the presence of 0.05 mM acetylthiocholine, 0.5 mU mL−1 acetylcholinesterase and different concentrations of paraoxon. (C) Variation of the ratio between the maxima of the longitudinal and transverse LSPR bands, as a function of paraoxon concentration. (D) TEM images of AuNPs that show the morphologies tuned by the addition of thiocholine at different concentrations. From Ref. [39].


259

Fig. 3. (A) Molecular precursor of chiral fibers. (B–C) SEM images of nanofibers with P and M morphologies, respectively. (D) SEM micrograph of P nanofibers coated by gold nanorods. (E) TEM micrograph of M nanofibers coated by gold nanorods. (F–G) CD and UV/Vis spectra of nanofibers with P and M morphologies coated by gold nanorods in solution, respectively. (H) CD spectra for chiral gold nanorod assemblies obtained for different DNA concentrations. (I) Calibration plots obtained using UV/Vis and CD spectra of chiral gold nanorod side-by-side assemblies. (J) Cryo TEM tomography images for a chiral assembly of gold nanorods (the scale bar is 25 nm). From Refs. [45] and [47].

NPs coated with rationally selected (bio)molecules have been arranged to build up organized plasmonic aggregates, showing interesting collective optical and sensing LSPR-based responses [12]. Natan and co-workers have prepared 3D AuNP and AgNP multilayered films by using successive treatments of NP monolayers with a bifunctional cross-linker and colloidal solutions [52]. Such plasmonic substrates have shown highly sensitive LSPRs under immersion in solvents with different types of analytes, which depend on the Au or Ag composition of NPs, the number of NP layers and the length of the molecular bifunctional cross-linkers. Similarly, Ye et al. have fabricated 3D self-assembled AuNP multilayer structures by the alternate deposition of a bifunctional cross-linker and AuNPs, improving the refractive index sensitivity of the LSPR sensor [53]. The investigations show that the number of AuNP deposition cycles has a strong effect on the sensitivity of the sensor and the multilayer structure fabricated from four NP deposition cycles exhibits maximum sensitivity to the change of the environmental refractive index. These studies are showing promise, however all substrates present relatively low nanoscopic homogeneities in terms of the low control of NPs packing within the aggregates, which may render to broad and non-uniform LSPR bands. With the aim of solving that issue, GuerreroMartínez and co-workers have developed methods of syntheses of large and homogeneous substrates of supercrystals with close packing of AuNPs following supramolecular [54] and colloidal [23,24] strategies. Such plasmonic substrates show highly reproducible LSPR bands within long-range areas (~ cm2) of the substrates (Fig. 4). Within a supramolecular context, the use of thiol-functionalized nonionic surfactants to stabilize AuNPs in water induces the spontaneous

formation of polyrotaxanes at the nanoparticle surface in the presence of the macrocycle α-cyclodextrin [54]. The self-assembly between the surface supramolecules provides large and homogeneous supercrystals with hexagonal close packing of nanoparticles (Fig. 4). Once formed, the self-assembled supercrystals can be fully redispersed in water. Therefore, the reversibility of the crystallization process may offer an excellent reusable material to prepare gold nanoparticle inks and optical sensors with the potential to be recovered after use. In other example, cationic gemini surfactants were used for the reproducible and controlled synthesis of monodisperse Au and Ag nanorods, focusing on the role of the chemical structure of these surfactants on the selfassembly of highly ordered, robust 2D and 3D NP superlattices with directional optical properties (Fig. 4) [23,24]. The optimization on the preparation of colloidal supercrystals was carried out by drop casting the colloidal solution at controlled temperature and humidity conditions. 3. SERS- and SEIRA-based colloidal plasmonic nanosensors 3.1. SERS and SEIRA principle Surface-enhanced Raman scattering (SERS) is currently recognized as one of the most sensitive spectroscopic techniques, and has been extensively exploited for ultrasensitive (bio)chemical detection [55]. SERS spectroscopy basically occurs when the LSPR of a NP enhances the Raman signal of a molecule being close to the nanostructure (Fig. 5) [5]. From the early studies of the SERS effect, the experimental observations showed that more than one mechanism should contribute to the

260


Fig. 4. (A) NP thiolated stabilizer (up) and reversible supercrystals of stabilized AuNPs in water by supramolecular approaches (down). (B) TEM micrograph of a hexagonally packed bilayer of stabilized AuNPs. (C) SEM micrograph of the stabilized AuNP supercrystal. (D) Normalized UV/Vis spectra of depositions of stabilized AuNPs (1), and conventional types of AuNPs (2, 3), at the ring (solid lines) and inner (dashed lines) of the drop patterns. (E) Normalized UV/Vis spectra of the stabilized AuNP solutions before (black) and after (red) redispersion of supramolecular supercrystals in water. (F) SEM micrographs of Au@Ag nanorods deposited at optimized drop casting conditions. (G) Drop profile and SERS mapping of an analyte evaporated on Au@Ag nanorod and Au nanorod deposits. (H) Comparisons between SERS intensities on Au@Ag nanorod and Au nanorod supercrystals at the different pattern regions. From Refs. [54] and [24].

enhancement of the Raman signal. Two mechanisms are commonly considered for SERS, one of which involves enhancements in the electric near field intensity as a result of surface plasmon resonance excitation, which is called electromagnetic mechanism (EM, Fig. 5). On the other hand, the enhancement in polarizability due to chemical effects such as charge-transfer excited states gives rise to the so-called chemical effect or charge transfer effect (CT, Fig. 5) [56]. Nowadays, it is recognized that although EM is essential for SERS, CT plays a key role in the truly ultrasensitive application of the technique, such as single molecule detection [6]. The obtained SERS spectrum is the vibrational fingerprint of the system under study, and thus provides crucial information about the structure, conformation and dynamics of the analyte. Therefore, much of the impetus behind the research in SERS relies on its potential applications. The signal enhancement provided by metallic nanostructures resolves the problem of the low cross-section of the Raman process [5,6], combining the ultrasensitive potential of this technique with the rich chemical and structural information characteristic of the vibrational spectroscopy. In fact, nowadays enhancement factors have been reported up to 1014–1015 fold [57]. The enhancement factor strongly depends on both the structure (size, shape and composition) of the metallic nanostructure as well as on the nature of the molecular probe. Usually, the enhancement factor expression is simplified by the so-called E4 approximation (fourth power of field enhancement at the NP surface). This model is commonly used to explore the EM enhancement and considers the Raman scattering in the vicinity of the defined metallic nanosurface. A more detailed explanation of this model can be given by separately taking into account the enhancement of both the incident

and the emitted radiation. According to that, the SERS electromagnetic enhancement factor (EF) can be simply expressed as: E F ≈ MLoc ðωL ÞMdRad ðωR Þ where MLoc(ωL) is the local field intensity enhancement factor and MdRad(ωR) corresponds to the directional radiation enhancement factor (related to a given position). MLoc defines how stronger the field intensity is with respect to the intensity in the absence of metal substrate, and can be found by solving the electromagnetic problem under specific external excitation conditions with an incident field EInc, which yields the local field ELoc everywhere. Estimating MdRad is a priori a more difficult task, but in order to avoid further complications, it is often assumed that MdRad(ω) ≈ MLoc(ω), which means that the SERS enhancement can then be expressed as: SMEF ðωL ; ωR Þ ≈ MLoc ðωL ÞMLoc ðωR Þ ≈

jELoc ðωL Þj2 jELoc ðωR Þj2 jEInc j2

jEInc j2

:

Latter expression has been widely used in the literature, and provides a fairly simple way of estimating the electromagnetic enhancement factor from a calculation of the local field at the excitation and Raman frequencies [6]. Since SERS can be carried out under environmental or biological conditions [58], it appears as an ideal detection platform for (bio)chemical sensing, diagnostics, (bio)analytical chemistry or environmental monitoring, among other applications [59]. Although much work is still being carried out in terms of the fundamental understanding of the


261

Fig. 5. Schematic representation of SERS effect in: (A) an individual NP. (B) An interparticle gap, so-called hot-spot, area with further enhanced Raman signal. (C) Schematic representation of the interaction between the electromagnetic radiation and a spherical NP, explanation of the electromagnetic (EM) mechanism in SERS phenomenon. (D) Diagram depicting the chemical transfer (CT) mechanism showing the energy levels for a ‘metal–molecule’ system and the possible Raman resonance processes involving molecular states (path a) and molecular and metallic states (paths b and c).

SERS effect, many more publications are now entirely dedicated to its applications and the development of new substrates such as colloids, thin films, self-assembled or hybrid materials [60]. The design of efficient and flexible nanostructured substrates for SERS detection is one of the main challenges to be achieved before the technique can be widely applied. Therefore, the preparation of optical substrates with optimized properties is a very dynamic field of research, and, as there is no universal ‘best’ SERS platform, careful consideration of the analytical problem is required before choosing/designing a SERS sensor platform [10]. Soon after the discovery of SERS in the early 1970s, the parallel association to the phenomenon at longer wavelengths regarding vibrational energy changes, SEIRA (surface-enhanced infrared absorption spectroscopy) was realized. Generally speaking, it is possible to attribute to SEIRA similar mechanisms than SERS, being both complementary techniques. Electromagnetic and chemical effects play a crucial role in the SEIRA as well. Nevertheless, latter affirmation is quite general, and specific differences in SEIRA and SERS have been extensively studied by Osawa [61]. The main difference between both phenomena is that SERS happens when there is a change in the polarizability of the molecule close to the nanosurface, whereas in SEIRAS there is a need of a change in the dipole moment perpendicular to the substrate surface [62]. We would also like to refer the reader to reviews about SEIRA and a book about SEVS (surface-enhanced vibrational spectroscopy) that extensively explain the mechanism behind the SEIRA phenomenon in comparison to SERS [63,64]. Although the enhancement factors that

can theoretically be achieved by SEIRA (101–104) are considerably lower than the ones obtained by SERS (108–1014), the higher crosssection of infrared versus Raman, pushed researchers to explore further on the SEIRA technique. Therefore, metals, semimetals, conductors, polar dielectric nanostructures and composite materials can show SEIRA effect. Following metals are the most used by far for the fabrication of SEIRA substrates: Au, Ag, Cu and Pt, presumably because of the higher understanding on the spectroscopies behind such metals associated to the higher development of SERS compared to SEIRA [65]. Regarding the supporting materials, SEIRA substrates are mainly designed as thin films, and in turn, thin films fabricated by lithographic methods [66]. However, as this review is focused on colloidal substrates, we will just cover the stateof-the-art on colloidal-based SEIRA materials. 3.2. SERS and SEIRA-based nanosensing using colloidal metal NPs in solution One of the highest obstacles of SERS sensing in colloidal solution is the limited achievable sensitivity, directly related to the inability of forming stable and reproducible hot-spots (interparticle gaps with enhanced SERS effect due to the coupling of LSPRs). Strategies like magnetic optical accumulation by using hybrid materials [67], nanoparticle assemblies in liquid–liquid or liquid–air interfaces [68,69] or microfluidics and microdroplets techniques [70,71], have effectively demonstrated to overcome such sensitivity issue. SERS ultradetection of a Raman molecular probe (1-naphthalenethiol) was possible by

262


concentrating core–shell γ-Fe2O3@SiO2 coated with rod-shape AuNPs in a small spot of the sample containing the multifunctional material by applying a magnetic field [72]. A portable Raman system was used in this work, validating their application for on-field detection of analytes at trace concentrations. The same principle, adapted by different research groups varying the hybrid material components and the molecule of study, allowed DNA-sequencing and pollutants ultradetection in liquid phase. Specifically, pNIPAM, a thermosensitive coating hydrogel, was used to trap hydrophobic molecules with low affinity for Au or Ag surfaces [67]. The combination of hydrophobic trapping in the hydrogel matrix, with the collapsing ability to form hot-spots allowed for the trace detection of pentachlorophenol, an elusive chlorinated ubiquitous environmental pollutant. Also, supramolecular chemistry has shown to be effective for the detection of small molecules or analytes with low affinity for plasmonic surfaces by means of SERS [20]. For example, Baumberg's research group uses the macrocyclic host cucurbit[n]uril to form very controlled subnanometer gaps between AuNPs [73]. By this method, not only the cavity of the macrocycle traps small molecules, but also the created hot-spots are highly controllable in terms of interparticle distance and reproducibility among experiments. They have reported on the quantification of several molecules by this method, and remarkably of some small molecules being difficult to detect by SERS (Fig. 6) [74]. The creation of hot-spots by this method opens up new possibilities for liquid SERS substrates, as overcomes the sensitivity issue when using dispersed nanoparticles and the low reproducibility of hot-spots in solution regarding conventional aggregation techniques. Recently, microfluidic and microdroplet techniques are starting to rise up as promising platforms for NP synthesis as well as a detection

platform for sensing [75]. Automation, miniaturization and high throughput are unique properties desired by companies investing in novel sensing technologies [29]. The lack of reproducibility among SERS substrate batches obtained within conventional bench-top techniques can be avoided by automating the synthetic process; miniaturization in microfluidics offers higher enhancement factors due to the decrease of volume and thus an improved sensitivity; and finally, the high throughput ability of microfluidics technology leads to a rapid and accurate analysis of liquid samples. For example, Cecchini et al. developed a microdroplet-based detection tool by coupling the enhancing power of aggregated AgNPs with the high-throughput screening ability of microdroplet technique (Fig. 6) [76]. The SERRS (surface-enhanced resonance Raman scattering) signals from the silver aggregates within the microdroplets were used for sensitive and high-throughput analysis of a Raman reporter. Multiplexing is also one of the most important capabilities when developing analytical sensors. In this line, Abalde-Cela et al. used a simple approach by co-flowing AgNPs and complex liquid samples through a microfluidic chip (Fig. 6) [70]. Compared to the broad bibliography on SERS colloidal platforms, it is quite challenging to design SEIRA probe particles for analysis in solution. In order to be able to acquire the IR spectra of an analyte, all other contributions from matrix or media need to be avoided, as they would overlap the analyte signal. As an example, water exhibits a strong IR pattern owed to the OH− bending vibrational mode. According to this, the design of liquid SEIRA colloidal platforms would be quite challenging. However, it is worth mentioning that novel studies have shown that the matrix or media contribution to the spectrum can be distinguished from the analyte contribution in SEIRA platforms by adjusting their internal reflections, absorptions and scatterings [77]. Latter approach

Fig. 6. (A) Schematic of host–guest SERS analysis using ternary complexation with cucurbit[8]uril-based hot-spot. (B) SERS spectra of (i) CB[8] (5 μM) and (ii)–(v) CB[8] complexes with different molecules. (C) Real-time multiplex SERS detection in microfluidics; (i) +-shaped microfluidic device for dual multiplex online monitoring of two flows of different sources, where each flow contains a mixture of two different analytes, either CV and NB or AB and TB (CV + NB, pink; AB + TB, light green; CV, purple; NB, red; AB, blue; and TB, yellow). (ii, iii) Optical and SERS images of the microchannel showing the distribution of the analytes (experimental: dotted black line; spectral references: colored solid lines). Dye concentration, 1 mM; scan time, 70 ms. From Refs. [70] and [74].


has been reported for lithographic materials, but we just wanted to highlight that it would be theoretically possible to adapt this method to colloidal–SEIRA substrates. 3.3. SERS- and SEIRA-based nanosensing using colloidal metal NPs on substrates The use of colloid nanoparticles for the preparation of SERS sensors demands some methodological requirements. Specifically, the analyte of interest must have affinity for the plasmonic sensing platform, thus the Raman signal of the molecule can be effectively enhanced. However, a wide range of molecular functional groups shows low affinity toward Au or Ag surfaces, the most commonly used metals in SERS substrates [78]. Furthermore, the sensitivity achieved with average SERS may not be enough when detection of trace concentrations of analytes is pursued. Although polymer trapping, optical accumulation, and microfluidicsbased sensing in liquid substrates has remarkably evolved in the last years, sensor engineering plays a key role in the design of more flexible, functional and ultra-sensitive SERS substrates [59]. In this section, we will focus on SERS substrates based on NPs deposited or embedded within a supporting substrate, being thin films or solid polymer/hydrogel particles, which can effectively address the limitations of SERS liquid sensors. Initially, amorphous assemblies of NPs over substrates (glass, silicon, ITO) were obtained by simple casting a drop of the colloidal NPs over the substrate and forcing the aggregation of the NPs by adding a salt solution to destabilize the colloidal suspension and cause random hot-spots, or either by tuning the capping agent coating the specific NPs [79]. However, the assembly of the NPs in such cases ended up in random crystal formation very difficult to reproduce from one experiment to another. The next generation of substrates benefited from the accurate control and variation of external factors, such as temperature, pressure or humidity implemented to slightly control the reproducibility of hot-spot formation [24,80,81]. Later, not only to improve the control over aggregation but also to increase the size and homogeneity of substrate areas, hierarchical assembly of NPs over different substrates started to be a pursued strategy. The assembly of NPs in specific geometries allows the effective coupling of electromagnetic fields, thus controlling the nanoantenna effects of the engineered substrate [82]. However, it is not technically straightforward to build homogeneous, large size-range and reproducible hierarchical assemblies. Recently, strong interest on plasmonic supercrystals as SERS substrates is an expanding area of research [83]. Gómez-Graña et al. reported on a refined supercrystal arrangement based on rodshape AgNPs vertically aligned over a substrate acting as SERS antennas (Fig. 4) [84]. By controlling the capping molecules surrounding the nanoplasmonic surface, a novel and highly optically active SERS sensor for ultrasensitive screening of analytical targets relevant to medical and environmental science was developed. Within the same type of substrates, NPs deposited with different morphologies such as Au nanorods, Au nanostars, Au-spiked nanoparticles or Ag nanoplates have been explored aiming a homogeneous enhancing surface [26,84,85]. The control of the assembly properties, interparticle distances, composition, size, shape or surface area coverage determines the final properties of the solid substrate. Specifically, a recent work presented by Hamon et al. reports on a novel one-step patterning method achieving critical substrate aspects mentioned above [80]. Remarkably, by this method, homogeneous assemblies of rod-shape AuNPs in the millimeter scale range area were possible due to the confinement of the colloidal suspensions in micron-sized cavities and further evaporation (Fig. 7). Hybrid materials also deserve special mention in this section. The combination of polymers, hydrogels or carbon nanotubes as supporting materials for dense NP hot-spot matrixes has attracted the attention of worldwide researchers. Both in films or beads fashion, hybrid materials offer several advantages as SERS sensors [86]. By choosing the appropriate non-plasmonic component, several functionalities can be added to the final material, such as sensitivity to external triggers, magnetism,

263

electromagnetism, conductivity, biocompatibility or reusability. For example, a hybrid material comprising silver aggregates supported on the porous structure of a free-standing carbon nanotube film [87], has allowed the filtration of large volumes of fluids retaining active analytes for its direct identification of multiple analytes at the attomolar regime. It is also worth mentioning the new current of flexible, portable and user-friendly SERS sensors being developed very recently, especially those supported in paper [88]. A very interesting work was released by Polavarapu et al. demonstrating the use of plasmonic inks made of different NPs to write SERS arrays. Impressive attomolar detection limits of dye molecules were achieved by the use of such substrate [89]. Finally, Tian and co-workers reported on a novel 3D hot-spot matrix achieved by evaporating a droplet of citrate-AgNPs on a fluorosilylated silicon wafer (Fig. 7) [81]. Interparticle gaps were formed in the 3D matrix as water evaporates until the AgNPs were uniformly deposited on the wafer. Interestingly, they have made a theoretical and experimental study of the evolution of the hot-spots during the evaporation process. Peak SERS intensities were found for several analytes with different affinities for the plasmonic surface. Moreover, they demonstrate the applicability of the technique for both Ag and AuNPs. Authors claim on the universality of the platform as its use can be either extended to other plasmonic structures as well as along with different sensing techniques or analytes of different chemical nature. It is noteworthy, that this 3D platform theoretically gathers all the premises for a sensor to be extraordinary: controllable, reproducible, ultra-sensitive, homogeneous, portable and universal. SEIRA solid substrates are mainly produced by lithographic techniques as mentioned above [65]. However, the issue of this review covers analytical substrates based on colloidal metallic particles. One of the first reports on SEIRA with colloidal particles included a comparison on the efficiency of gold films versus gold colloids for the detection of Salmonella antibody/antigen complexes. Authors developed a combined method involving the filtration of gold colloids into polyethylene membranes after attachment to specific antibodies [90]. Even there is not an extensive bibliography on SEIRA, majority of the applied works report on protein–protein interactions, as a subtle change in the conformation and orientation to the surface is translated into spectral changes. In this line, Polissiou and co-workers have used dry films of colloidal gold to study the spectra of protein A and the changes after its conjugation to immunoglobulin G. Also, Halas and co-workers have made a remarkable progress on the fundamental study, development and sensing applications of SEIRA solid substrates [91]. More specifically, they have developed a substrate based on the aggregation of Au nanoshells over silicon films. The generation of infrared ‘hot-spots’ allowed for an enhancement factor in SEIRA of 104 for the molecular probe 4mercaptoaniline, one of the highest EFs in SEIRA reported up-to-date [92]. Another study of the same group describes the use of these substrates for combined SERS and SEIRA, analyzing the spectra of CTAB and p-mercaptoaniline with both enhancing spectroscopies [93]. More recently, Cai et al. have applied a novel composite material to SEIRA. They have designed Fe3O4/Au nanocomposites allowing both magnetic manipulation of the material as well as spectroscopic enhancing of the infrared absorption [94]. 4. SEF-based colloidal plasmonic nanosensors 4.1. SEF principle Fluorescence-based techniques which are key tools for labeling and sensing in analytical chemistry, biotechnology and biochemistry can also be benefited from the interaction between the free electrons of plasmonic substrates or particles and fluorophores [95]. In general, this interaction can trigger the quenching or enhancement (i.e. SEF) of fluorescence but concretely, parameters such as distance between the fluorophore and the metal [27], nature [96] and geometry of the plasmonic material (e.g., size and shape) [97], spectral overlap [98] and

264


Fig. 7. (A) Tilted SEM characterization of patterned substrates to form supercrystals at rod-shape AuNP concentration of 375 mM. (B) Optical microscopy images of (left) AuNP supercrystals (the scale bar is 10 μm) and corresponding SERS images (right — in red) obtained by mapping the intensity of the crystal violet vibrational peak integrated over 1618–1632 cm−1. (C) Average SERS spectra of crystal violet at a concentration of 10−6 M in ethanol, drop-casted on the substrate and dried under ambient conditions. Spectra were acquired on supercrystals with different shapes: drops (blue), circles (black), and squares (red). (D) (I–III) 3D hotspot matrix of noble-metal sols generated in the water evaporation process. (E) (I) Time-course SERS mapping of a 1 μL sample with 50 amol of R6G (3 × 107 molecules) and ~1010 AgNPs placed on a hydrophobic-treated silicon wafer. (II) Line SERS mapping at 610 cm−1 for the dried sample in a range of −100 to 100 μm centered on the position used for time-course mapping. From Refs. [80] and [81].

orientation of the fluorophore [7] rule the final characteristics of the radiative decay of dyes. Usually, SEF can occur at distances between 5 and 30 nm. At shorter distances, fluorescence would be quenched. Another relevant issue is the spectral overlap (Fig. 1C). The enhancement of fluorescence is more efficient when the emission of the fluorophore is red-shifted with respect to the metal plasmon band spectra and it can be quenched if it is blue-shifted [99]. It is also noteworthy that SEF takes place with all types of fluorophores, regardless if they are organic molecules, quantum dots or lanthanides [27,100]. The presence of NPs near fluorophores at nanoscale distances influences the excitation/emission process of fluorophores in two ways [101]. On one hand, the excitation process can be affected by the plasmonically-enhanced electromagnetic field near the NPs, especially at the hot-spots [102]. This may increase the absorbed light by the fluorophores. On the other hand, the emission process and non-radiative decay rates are strongly influenced by metals due to the coupling between the emission of the fluorophore and the NP LSPR modes [103], and as consequence, the quantum yield and life time of fluorophores change [104]. Excited-state fluorophores can create plasmons near metals, which can radiate with the same emission spectra than the original fluorophore but with shorter lifetimes (τ). The increase of the radiative decay rate (Γ = τ−1) is produced by the shorter lifetimes of plasmons (~50 fs) compared with fluorophores (~1 ns) [7]. Because the emission properties are a combination of both the fluorophore and the metal, the fluorophore–metal complex can be regarded as the emitting specie characterized by a new quantum yield (Qm) and lifetime (τm) with

respect to the fluorophore. Considering the following expressions for the fluorophore quantum yield (Q0) and lifetime (τ0): Q0 ¼

Γ Γ þ knr −1

τ0 ¼ ðΓ þ knr Þ

where knr is the non-radiative decay rate. The Qm and τm of the fluorophore–metal complex are given by: Qm ¼

Γ þ Γm Γ þ Γm þ knr −1

τ0 ¼ ðΓ þ Γm þ knr Þ

:

Often, the efficiency of fluorescence enhancement is characterized by the enhancement factor according to: EF ¼

I f ‐m If

where If–m represents the emission intensity of the fluorophore–metal complex and If represents the emission intensity of the free fluorophore, whose concentration matches with the concentration of fluorophores in the complex. The effect of SEF can be dramatic and quantum yields can reach almost the unity, lifetimes can decrease up to 105-fold and


enhancement factors of three orders of magnitude can be obtained [105]. Furthermore, shorter lifetimes increase the photostability of the fluorophore–metal complex, since it can undergo more excitation/ emission cycles, and enhances in parallel its brightness. These features, explain why platforms based on SEF are attractive for analytical sensing. However, the design of sensors based on SEF is challenging. The sensitivity of the coupling between the fluorophore and the LSPR modes of the metallic nanostructure is exploited for sensing, however many parameters in the environment that are not under control may affect simultaneously the fluorescence emission. For example, changes of solvent or adsorption of proteins in physiological fluids modify the refractive index and consequently, the coupling [106]. This would produce misleading signals. Moreover, the fact that SEF platforms require a controlled distance between the metal and the fluorophore involves the use of spacers. These spacers have as well their own limitations, for example in terms of synthesis and stability [107]. However, recently, promising examples of sensors based on SEF substrates had shown high sensitivity, specificity and reproducibility, even for multiplexed sensing in complex systems such as blood, which can boost further applications in the field and the development of commercial devices in the best case [108]. In the following, applications of several sensing platforms based on SEF will be discussed and compared. We refer the reader to an excellent previous report for more details about the physical aspects of SEF and the production of SEF platforms, both in solution and substrates [7]. 4.2. SEF-based nanosensing using colloidal metal NPs in solution There are only few examples of SEF platforms for analysis in solution. Nevertheless, they could have broad applications in life science and analytical chemistry. The impact of improvements in this field could be noteworthy since fluorescent probes are the main tools for real-time imaging of live cells and NPs can increase their quantum yield and photostability. For SEF nanosensing in solution, the size of colloidal NPs is particularly important. In very small metallic nanoparticles, absorption is higher than scattering of light, and quenching of the emission of the fluorophore will predominate over its enhancement. Therefore, larger nanoparticles and geometries with higher scattering crosssections are preferred [109]. One of the first reported SEF sensors in solution was based on AgNPs coated with a silica shell [110]. Such proof of concept demonstrated that the binding of fluorescently labeled streptavidin to avidin-functionalized AgNPs reduced the lifetime of the fluorophore 1.3-fold (enhancement factor ~4). The main limitations of such sensor were the polydispersity and inhomogeneous shape of AgNPs, the inhomogeneous silica shell thickness and the non-controlled and non-reversible agglomeration of the NPs upon streptavidin binding. These factors produce low sensitivities and reproducibilities. The quality of NPs and silica shells in terms of monodispersity and thickness homogeneity can be improved using new synthetic strategies in order to tailor the fluorescence enhancement [27]. However, sensors based on the streptavidin-mediated binding suffer from an intrinsic drawback. Due to non-specific adsorption, streptavidin-functionalized NPs tend to agglomerate even in the absence of biotin or avidin [40]. This non-specific adsorption is probably one of the main drawbacks for recognition using some biological linkers. Otherwise proteins can be directly attached to the metal surface and act as spacers within the fluorophore–metal complex [111]. The production of robust sensors for analytical purposes should avoid non-specific interactions. Specificity in the detection process can be substantially increased for SEF sensors in solution using oligonucleotides. For example, molecular beacons have been extensively studied (Fig. 8). These hairpin shaped oligonucleotides contain a quenched dye, a quencher and an oligonucleotide sequence that act as a probe. Such sensors can detect DNAs or RNAs that are complementary to their probe in a very specific manner, due to the opening of the hairpin structure upon the hybridization between the probe and the target

265

molecule. The detection of the target molecule triggers the reversible restoring of the fluorescence of the quenched dye [112]. Small metal nanoparticles (~ 2 nm) were firstly applied in such hairpin probes as mere quenchers due to their high absorption cross-sections (Fig. 8) [113]. However, sensors based on SEF can be designed if the colloidal NP is large enough. Cheng et al. developed a SEF sensor with thiolated hairpin oligonucleotides that undergo quenching/SEF in the absence/ presence of a target oligonucleotide (Fig. 8) [114]. The influence of the distance between the fluorophore and the metallic surface was studied using hairpin oligonucleotides of different lengths. Moreover, the influence of the NP size (from 20 to 100 nm) was demonstrated experimentally and theoretically. The largest AuNPs (100 nm) produced the highest enhancement factor (~1.5), which was in any case lower than the typical values obtained with SEF platforms in substrates. Recently, molecular beacons involving SEF have been applied for the detection of mercuric ions using thymine (T) containing probes. In this case, the hairpin structure opened in the presence of ions due to the formation of a T–Hg2+–T complex. Enhancement factors around 7 and a high specificity were reached using this approach [115]. The design of molecular beacons based on SEF can be extended to various targets such as proteins or microRNAs [116]. However, though it could be in principle feasible to use such sensors for in vivo detection [117], the need of a relatively high-sized NP well stabilized to avoid agglomeration, and the probable non-specific adsorption of proteins in physiological medium, would complicate the practical use of these sensors. There are more examples in the literature where SEF sensors are designed to undergo an enhancement in fluorescence due to a controlled increase/ decrease of the spacer length or distance. Polymers can be also used as responsive spacers [118]. The main limitation of such design is the lack of a fine control over the fluorophore–metal distance in some cases and the difficulties to achieve multiplexing. Non-invasive detection with fluorophore–metal complexes in solution has more chances to reach commercial applications as compared to invasive sensing. Recently, Hakonen and Strömberg developed an ammonium optode sensor to study tissue decomposition with AuNPs and a ratiometric fluorophore entrapped in hydrogel/ether droplets (Fig. 9) [119]. In this case, the droplets are used to concentrate both fluorophore and metal nanoparticles. The obtained limits of detection and signal-to-noise ratios were reduced as compared with other conventional analytical techniques. Moreover, this sensor configuration could be modified to sense theoretically all ions. 4.3. SEF-based nanosensing using colloidal metal NPs on substrates The most advanced SEF platforms for single molecule detection and designing of highly efficient multiplexed sensors have been built on substrates [105]. Here, it is worth differentiating between three different kinds of substrates: i) metal nanoparticles bound or deposited onto inert surfaces; ii) metallic thin films with nanoscale thickness; and iii) lithographically fabricated nanoantennas with controlled formation of hot-spots [7]. The three kinds of substrates produce different interactions between fluorophores and metals and all of them can be used for sensing [120]. Typically thin films do not modify dramatically the lifetime and emission intensity of fluorophores. In contrast, lithographically prepared substrates can trigger the largest enhancement factors but they are not cost-effective. This section will focus on the first kind of substrates that typically result in enhancement factor of fluorescence in the range of 1–100 [12]. AgNP and AuNP island films are the SEF platforms more broadly applied for sensing [108]. The formation of NP islands on a dielectric substrate concentrates optical fields due to the LSPRs of NPs and the presence of gaps between the particles that lead to enormous field amplifications and fluorescence enhancement factors (~100-fold), higher than those obtained with NPs in solution. Protein microarrays on AuNP island films have been utilized for high-throughput protein analysis. Fig. 10 schematically shows the configuration of the sensors based

266


Fig. 8. (A) Schematic representation of a molecular beacon based on metallic NPs. SEF can occur at appropriated NP–fluorophore distances and NP size. (B) Schematic representation of a molecular beacon for SEF using AuNPs and thiolated hairpin oligonucleotides of different sizes. From Ref. [114].

on antibodies (Ab) that bind specifically small proteins (e.g., cytokines) and bigger protein biomarkers (e.g., 1° human Ab) for the detection of defects in the immune system and type 1 diabetes, respectively [121]. This approach allows for multiplexing and has higher sensitivity and specificity than the ELISA assay. The size and shape of NPs within the aforementioned substrates are not homogeneous, which could compromise the reproducibility of the sensor. Recently, the formation of SEF planar platforms via the assembly of colloidal NPs with controlled geometry onto substrates has gained increasing interest. There are many possible designs such substrates based on AuNPs and quantum dots [122], Langmuir–Blodgett arrays of AuNPs [123], or copper arrays [124]. However, for detection the recognition event between the SEF platform and the analyte determines the specificity and sensitivity of the sensor. As in the case of SEF sensors in solution the most utilized molecules for this purpose are oligonucleotides and proteins [107]. Therefore, the appropriated choice of oligonucleotides or proteins and their effective linkage (avoiding non-specific adsorption) to the substrate are key factors for the further technique implementation. In a similar context to SERS, the design and development of microfluidic devices for sensing involving SEF will influence the future application of such platforms. Recently, Wang et al. successfully built a highly efficient micro/nanofluidic SEF sensor [125]. Reactive oxygen species within a flow were detected via the enhanced fluorescence of

analytes triggered by the presence of AgNPs. Multiplexing in such devices is in principle possible which may increase the number of feasible applications of SEF nanotechnology. 5. Concluding remarks In this review, we have analyzed the potential of using colloidal metal nanoparticles as advanced sensing platforms, both in solution and on solid substrates, applied to surface-enhanced spectroscopies. As shown, research in the field accomplished the design of high quality nanoplasmonic sensors showing impressively low detection limits while also being reproducible, reliable, portable and reusable. According to the latter, some of the colloidal sensing platforms described in this review could be translated into products or technologies for industrial applications [126]. In fact, even though most of the commercial devices based on surface-enhanced spectroscopies integrate planar substrates produced by photolitography, several companies have already developed detection systems based on colloidal particles. For example, agglomerated silver nanoparticles are being used on SERRS substrates for nucleic acid detection. Considering the high intensity of LSPR-mediated circular dichroism signals that have been registered by NPs both in the visible-near infrared region, using versatile and generic self-assembly strategies, we anticipate the use of such plasmonic nanoantennas as powerful chirality


267

Fig. 9. (A) Schematic representation of an optode sensor involving SEF with AuNPs and ratiometric dyes entrapped in hydrogel/ether droplets. On the right, false color images show the ammonium concentrations after 10 days. Concentrations are indicated on the color bar in mM. (B) Sensor without Au NPs. (C) Sensor with AuNPs. From Ref. [119].

probes upon specific attachment to proteins and DNA for in situ structure determination by selecting the appropriate nature, dimensions, morphology and functionalization of the nanocrystals. In such scenario, the fluorescence detected–circular dichroism (FD–CD), based on the measurement of the differential emission of light from a sample excited with circular polarized radiation and still in its infancy at the nanoscale [127], coupled to the surface enhanced fluorescence spectroscopy may emerge as a prominent technique for generation of individual and collective chiroptical activity in fluorophores with enhanced FD–CD, which might be a powerful analytical technique for detection of chiral

species. Similarly, an analogous combination has already shown very high enantioselectivity toward (bio)analytes detection through surface enhanced Raman optical activity spectroscopy (SEROA) [128]. The chemical and physical properties of the analytes to be detected are still some factors that limit the implementation of surface-enhanced spectroscopies based on colloidal particles, specially in the case of SEIRA-based nanosensing. In this regard, the (bio)functionalization of the particles rules the quality of the target binding between their surface and the analyte, which is key for the development of sensors with practical applications. The challenges in this approach are related to

Fig. 10. (A) Electron microscopy image of a gold island film. Schematic representation of plasmonic chips for detection based on protein binding: (B) the substrate is functionalized with an Ab that captures the analyte and (C) the substrate is functionalized with an antigen of the 1° Ab that acts as biomarker of type 1 diabetes. From Ref. [121].

268


the control over the favorable orientation and optimal density of the functional groups that promote the target binding, the specificity of such binding and the reusability of the sensor. Finally, it remains difficult to join all the mentioned properties in the same platform, and usually a balance between sensitivity and selectivity is very difficult to achieve. In our opinion, despite the promising works discussed herein, the development of a ‘universal’ platform based on colloidal metal nanoparticles will remain a significant challenge within the near feature. However, we believe that sensing devices based on surface-enhanced spectroscopies with colloidal particles with better performance than well-established techniques such as ELISA, PCR or standard fluorescence will arise in the next years.

Acknowledgment This work has been funded by the “I Convocatoria de Ayudas Fundación BBVA a Investigadores, Innovadores y Creadores Culturales” and the Madrid Regional Government (S2013/MIT-2807). J.P.C. acknowledges receipt of a Ciência sem Fronteiras fellowship from the CNPq of Brazil. A.G.-M. acknowledges receipt of a Ramón y Cajal Fellowship from the Spanish MINECO. S.C.R. acknowledges being involved in “Proyecto integrado de excelencia (PIE13/00048)” from Instituto de salud Carlos III.

References [1] Kreibig U, Vollmer M. Optical properties of metal clusters. Springer-Verlag; 1996. [2] Sepúlveda B, Angelomé PC, Lechuga LM, Liz-Marzán LM. LSPR-based nanobiosensors. Nano Today 2009;4:244–51. [3] Heber J. Surfing the wave. Nature 2009;461:720–2. [4] Álvarez-Puebla R, Liz-Marzán LM, García de Abajo FG. Light concentration at the nanometer scale. J Phys Chem Lett 2010;1:2428–34. [5] Moskovits M. Surface enhanced spectroscopy. Rev Mod Phys 1985;57:738–826. [6] Aroca R. Surface-enhanced vibrational spectroscopy. John Wiley & Sons; 2006. [7] Lakowicz JR, Ray K, Chowdhury M, Szmacinski H, Fu Y, Zhang J, et al. Plasmoncontrolled fluorescence: a new paradigm in fluorescence spectroscopy. Analyst 2008;133:1308–46. [8] Tuma R. Raman spectroscopy of proteins: from peptides to large assemblies. J Raman Spectrosc 2005;36:307–19. [9] Moskovits M. Imaging: spot the hotspot. Nature 2011;469:307–8. [10] Abalde-Cela S, Aldeanueva-Potel P, Mateo-Mateo C, Rodriguez-Lorenzo L, ÁlvarezPuebla RA, Liz-Marzán LM. Surface-enhanced Raman scattering biomedical applications of plasmonic colloidal particles. J R Soc Interface 2010;7:S435–50. [11] Xia Y, Xiong Y, Lim B, Skrabalak SE. Shape-controlled synthesis of metal nanocrystals: simple chemistry meets complex physics? Angew Chem Int Ed 2009;48:60–103. [12] Guerrero-Martínez A, Grzelczak M, Liz-Marzán LM. Molecular thinking for nanoplasmonic design. ACS Nano 2012;6:3655–62. [13] Nikoobakht B, El-Sayed MA. Preparation and growth mechanism of gold nanorods (NRs) using seed-mediated growth method. Chem Mater 2003;15:1957–62. [14] Zheng Y, Zhong X, Li Z, Xia Y. Successive seed-mediated growth for the synthesis of single-crystal gold nanospheres with uniform diameters controlled in the range of 5–150 nm. Part Part Syst Charact 2014;31:266–73. [15] Johnson CJ, Dujardin E, Davis SA, Murphy CJ, Mann S. Growth and form of gold nanorods prepared by seed-mediated surfactant-directed synthesis. J Mater Chem 2002;12:1765–70. [16] Gómez-Graña S, Goris B, Altantzis T, Fernández-López C, Carbó-Argibay E, Guerrero-Martínez A, et al. Au@Ag nanoparticles: halides stabilize [100] facets. J Phys Chem Lett 2013;4:2209–16. [17] Guerrero-Martínez A, Pérez-Juste J, Liz-Marzán LM. Recent progress on silica coating of nanoparticles and related nanomaterials. Adv Mater 2010;22:1182–95. [18] Carregal-Romero S, Buurma NJ, Pérez-Juste J, Liz-Marzán LM, Hervés P. Catalysis by Au@ pNIPAM nanocomposites: effect of the cross-linking density. Chem Mater 2010;22:3051–9. [19] Chanana M, Rivera Gil P, Correa-Duarte MA, Liz-Marzán LM, Parak WJ. Physicochemical properties of protein-coated gold nanoparticles in biological fluids and cells before and after proteolytic digestion. Angew Chem Int Ed 2013;52:4179–83. [20] Abalde-Cela S, Hermida-Ramón JM, Contreras-Carballada P, De Cola L, GuerreroMartínez A, Álvarez-Puebla RA, et al. SERS chiral recognition and quantification of enanotiomers through cyclodextrin supramolecular complexation. ChemPhysChem 2011;12:1529–35. [21] Endo T, Kerman K, Nagatani N, Hiepa HM, Kim DK, Yonezawa Y, et al. Multiple label-free detection of antigen–antibody reaction using localized surface plasmon resonance-based core–shell structured nanoparticle layer nanochip. Anal Chem 2006;78:6465–75. [22] Liu S, Tang ZJ. Nanoparticle assemblies for biological and chemical sensing. J Mater Chem 2010;20:24–35.

[23] Guerrero-Martínez A, Pérez-Juste J, Carbó-Argibay E, Tardajos G, Liz-Marzán LM. Gemini surfactant-directed self-assembly of monodisperse gold nanorods into standing superlattices. Angew Chem Int Ed 2009;48:9484–8. [24] Gómez-Graña S, Pérez-Juste J, Álvarez-Puebla R, Guerrero-Martínez A, Liz-Marzán LM. Self-assembly of Au@Ag nanorods mediated by gemini surfactants for highly efficient SERS-active supercrystals. Adv Opt Mater 2013;1:477–81. [25] Jones MR, Macfarlane RJ, Lee B, Zhang J, Young KL, Senesi AJ, et al. DNAnanoparticle superlattices formed from anisotropic building blocks. Nat Mater 2010;9:913–7. [26] Álvarez-Puebla RA, Agarwal A, Manna P, Khanal BP, Aldeanueva-Potel P, CarbóArgibay E, et al. Gold nanorods 3D-supercrystals as surface enhanced Raman scattering spectroscopy substrates for the rapid detection of scrambled prions. Proc Natl Acad Sci U S A 2011;108:8157–61. [27] Ma X, Fletcher K, Kipp T, Grzelczak MP, Wang Z, Guerrero-Martínez A, et al. Photoluminescence of individual Au/CdSe nanocrystal complexes with variable interparticle distances. J Phys Chem Lett 2011;2:2466–71. [28] Nguyen TAH, Hampton MA, Nguyen AV. Evaporation of nanoparticle droplets on smooth hydrophobic surfaces: the inner coffee ring deposits. J Phys Chem C 2013;117:4707–16. [29] Guicheteau JA, Farrell ME, Christesen SD, Fountain III AW, Pellegrino PM, Emmons ED, et al. Surface-enhanced Raman scattering (SERS) evaluation protocol for nanometallic surfaces. Appl Spectrosc 2013;67:396–403. [30] Bohren CF, Huffman DF. Absorption and scattering of light by small particles. Wiley; 1983. [31] Zhao J, Jensen L, Sung J, Zou S, Schatz GC, Van Duyne RP. Interaction of plasmon and molecular resonances for rhodamine 6G adsorbed on silver nanoparticles. J Am Chem Soc 2007;129:7647–56. [32] Zhao W, Brook MA, Li Y. Design of gold nanoparticle-based colorimetric biosensing assays. ChemBioChem 2008;9:2363–71. [33] Chen H, Ming T, Zhao L, Wang F, Sun L-D, Wanga J, et al. Plasmon—molecule interactions. Nano Today 2010;5:494–505. [34] Prodan E, Radloff C, Halas NJ, Nordlander PA. Hybridization model for the plasmon response of complex nanostructures. Science 2003;302:419–22. [35] Reinhard BM, Siu M, Agarwal H, Alivisatos AP, Liphardt J. Calibration of dynamic molecular rulers based on plasmon coupling between gold nanoparticles. Nano Lett 2005;5:2246–52. [36] Ni W, Yang Z, Chen H, Li L, Wang J. Coupling between molecular and plasmonic resonances in freestanding dye-gold nanorod hybrid nanostructures. J Am Chem Soc 2008;130:6692–3. [37] Thanh NTK, Rosenzweig Z. Development of an aggregation-based immunoassay for anti-protein using gold nanoparticles. Anal Chem 2002;74:1624–8. [38] Grzelczak M, Vermant J, Furst EM, Liz-Marzán LM. Directed self-assembly of nanoparticles. ACS Nano 2010;4:3591–605. [39] Coronado-Puchau M, Saa L, Grzelczak M, Pavlov V, Liz-Marzán LM. Enzymatic modulation of gold nanorods growth and application to nerve gas detection. Nano Today 2013;8:461–8. [40] Yu X, Lei DY, Amin F, Hartmann R, Acuna GP, Guerrero-Martínez A, et al. Distance control in-between plasmonic nanoparticles via biological and polymeric spacers. Nano Today 2013;8:480–93. [41] Strozyk MS, Chanana M, Pastoriza-Santos I, Pérez-Juste J, Liz-Marzán LM. Protein/ polymer-based dual-responsive gold nanoparticles with pH-dependent thermal sensitivity. Adv Funct Mater 2012;22:1436–44. [42] Chanana M, Correa-Duarte MA, Liz-Marzan LM. Insulin-coated gold nanoparticles: a plasmonic device for studying metal–protein interactions. 2011;7:2650–60. [43] Shen Y, Kuang M, Shen Z, Nieberle J, Duan H, Frey H. Gold nanoparticles coated with a thermosensitive hyperbranched polyelectrolyte: towards smart temperature and pH nanosensors. Angew Chem Int Ed 2008;47:2227–30. [44] Guerrero-Martínez A, Alonso-Gómez JL, Auguié B, Cid MM, Liz-Marzán LM. From individual to collective chirality in metal nanoparticles. Nano Today 2011;6: 381–400. [45] Guerrero-Martínez A, Auguié B, Alonso-Gómez JL, Džolić Z, Gómez-Graña S, Žinić M, et al. Intense optical activity from three-dimensional chiral ordering of plasmonic nanoantennas. Angew Chem Int Ed 2011;50:5499–503. [46] Slocik JM, Govorov AO, Naik RR. Plasmonic circular dichroism of peptidefunctionalized gold nanoparticles. Nano Lett 2011;11:701–5. [47] Ma W, Kuang H, Xu L, Ding L, Xu C, Wang L, et al. Attomolar DNA detection with chiral nanorod assemblies. Nat Commun 2013;4:2689. [48] Ament I, Prasad J, Henkel A, Schmachtel S, Sönnichsen C. Single unlabeled protein detection on individual plasmonic nanoparticles. Nano Lett 2012; 12:1092–5. [49] Rosman C, Prasad J, Neiser A, Henkel A, Edgar J, Sönnichsen C. Multiplexed plasmon sensor for rapid label-free analyte detection. Nano Lett 2013;13:3243–7. [50] Marinakos SM, Chen S, Chilkoti A. Plasmonic detection of a model analyte in serum by a gold nanorod sensor. Anal Chem 2007;79:5278–83. [51] Lin T-J, Huang K-T, Liu CY. Determination of organophosphorous pesticides by a novel biosensor based on localized surface plasmon resonance. Biosens Bioelectron 2006;22:513–8. [52] Musick MD, Keating CD, Lyon LA, Botsko SL, Pena DJ, Holliway WD, et al. Metal films prepared by stepwise assembly. 2. Construction and characterization of colloidal Au and Ag multilayers. Chem Mater 2000;12:2869–81. [53] Ye J, Bonroy K, Nelis D, Frederix F, D'Haen J, Maes G, et al. Enhanced localized surface plasmon resonance sensing on three-dimensional gold nanoparticles assemblies. Colloids Surf A 2008;321:313–7. [54] Coelho JP, González-Rubio G, Delices A, Barcina JO, Salgado C, Ávila D, et al. Polyrotaxane-mediated self-assembly of gold nanospheres into fully reversible supercrystal. Angew Chem Int Ed 2014;53:12751–5.

S. Abalde-Cela et al. / Advances in Colloid and Interface Science 233 (2016) 255–270 [55] Qian XM, Nie SM. Single-molecule and single-nanoparticle SERS: from fundamental mechanisms to biomedical applications. Chem Soc Rev 2008;37:912–20. [56] Le Ru ECE, Etchegoin P. Principles of surface enhanced Raman spectroscopy and related plasmonic effects. Amsterdam: Elsevier; 2009. [57] Kneipp K, Kneipp H, Itzkan I, Dasari RR, Feld MS. Ultrasensitive chemical analysis by Raman spectroscopy. Chem Rev 1999;99:2957–76. [58] Álvarez-Puebla RA, Liz-Marzán LM. SERS-based diagnosis and biodetection. Small 2010;6:604–10. [59] Sharma B, Frontiera RR, Henry A-I, Ringe E, Van Duyne RP. SERS: materials, applications, and the future. Mater Today 2012;15:16–25. [60] Cialla D, März A, Böhme R, Theil F, Weber K, Schmitt M, et al. Surface-enhanced Raman spectroscopy (SERS): progress and trends. Anal Bioanal Chem 2012;403: 27–54. [61] Osawa M. Surface-enhanced infrared absorption. Top Appl Phys 2001;81:163–87. [62] Osawa M, Ataka K-I, Yoshii K, Nishikawa Y. Surface-enhanced infrared spectroscopy: the origin of the absorption enhancement and band selection rule in the infrared spectra of molecules adsorbed on fine metal particles. Appl Spectrosc 1993;47:1497–502. [63] Aroca RF. Surface-enhanced vibrational spectroscopy. New York: John Wiley and Sons Ltd; 2006. [64] Aroca RF, Ross DJ, Domingo C. Surface-enhanced infrared spectroscopy. Appl Spectrosc 2004;58:324A–38A. [65] Ataka K, Heberle J. Biochemical applications of surface-enhanced infrared absorption spectroscopy. Anal Bioanal Chem 2007;388:47–54. [66] Huck C, Neubrech F, Vogt J, Toma A, Gerbert D, Katzmann J, et al. Surface-enhanced infrared spectroscopy using nanometer-sized gaps. ACS Nano 2014;8:4908–14. [67] Contreras-Cáceres R, Abalde-Cela S, Guardia-Girós P, Fernández-Barbero A, PérezJuste J, Alvarez-Puebla RA, et al. Multifunctional microgel magnetic/optical traps for SERS ultradetection. Langmuir 2011;27:4520–5. [68] Cecchini MP, Turek VA, Paget J, Kornyshev AA, Edel JB. Self-assembled nanoparticle arrays for multiphase trace analyte detection. Nat Mater 2013;12:165–71. [69] Kim K, Han HS, Choi I, Lee C, Hong S, Suh SH, et al. Interfacial liquid-state surfaceenhanced Raman spectroscopy. Nat Commun 2013;4:2182. [70] Abalde-Cela S, Abell C, Álvarez-Puebla RA, Liz-Marzán LM. Real time dual-channel multiplex SERS ultradetection. J Phys Chem Lett 2013;5:73–9. [71] Lim C, Hong J, Chung BG, deMello AJ, Choo J. Optofluidic platforms based on surface-enhanced Raman scattering. Analyst 2010;135:837–44. [72] Spuch-Calvar M, Rodríguez-Lorenzo L, Morales MP, Álvarez-Puebla RA, Liz-Marzán LM. Bifunctional nanocomposites with long-term stability as SERS optical accumulators for ultrasensitive analysis. J Phys Chem C 2008;113:3373–7. [73] Taylor RW, Lee TC, Scherman OA, Esteban R, Aizpurua J, Huang FM, et al. Precise subnanometer plasmonic junctions for SERS within gold nanoparticle assemblies using cucurbit[n]uril “glue”. ACS Nano 2011;5:3878–87. [74] Kasera S, Biedermann F, Baumberg JJ, Scherman OA, Mahajan S. Quantitative SERS using the sequestration of small molecules inside precise plasmonic nanoconstructs. Nano Lett 2012;12:5924–8. [75] Theberge AB, Courtois F, Schaerli Y, Fischlechner M, Abell C, Hollfelder F, et al. Microdroplets in microfluidics: an evolving platform for discoveries in chemistry and biology. Angew Chem Int Ed 2010;49:5846–68. [76] Cecchini MP, Hong J, Lim C, Choo J, Albrecht T, De Mello AJ, et al. Ultrafast surface enhanced resonance Raman scattering detection in droplet-based microfluidic systems. Anal Chem 2011;83:3076–81. [77] Adato R, Altug H. In-situ ultra-sensitive infrared absorption spectroscopy of biomolecule interactions in real time with plasmonic nanoantennas. Nat Commun 2013;4:2154. [78] Banholzer MJ, Millstone JE, Qin L, Mirkin CA. Rationally designed nanostructures for surface-enhanced Raman spectroscopy. Chem Soc Rev 2008;37:885–97. [79] Yaffe NR, Ingram A, Graham D, Blanch EW. A multi-component optimisation of experimental parameters for maximising SERS enhancements. J Raman Spectrosc 2010;41:618–23. [80] Hamon C, Novikov S, Scarabelli L, Basabe-Desmonts L, Liz-Marzán LM. Hierarchical self-assembly of gold nanoparticles into patterned plasmonic nanostructures. ACS Nano 2014;8:10694–703. [81] Liu H, Yang Z, Meng L, Sun Y, Wang J, Yang L, et al. Three-dimensional and timeordered surface-enhanced Raman scattering hotspot matrix. J Am Chem Soc 2014;136:5332–41. [82] Freeman RG, Grabar KC, Allison KJ, Bright RM, Davis JA, Guthrie AP, et al. Selfassembled metal colloid monolayers: an approach to SERS substrates. Science 1995;267:1629–32. [83] Bryant GW, García de Abajo FJ, Aizpurua J. Mapping the plasmon resonances of metallic nanoantennas. Nano Lett 2008;8:631–6. [84] Aldeanueva-Potel P, Carbó-Argibay E, Pazos-Pérez N, Barbosa S, Pastoriza-Santos I, Álvarez-Puebla RA, et al. Spiked gold beads as substrates for single-particle SERS. ChemPhysChem 2012;13:2561–5. [85] Rodríguez-Lorenzo L, Álvarez-Puebla RA, Pastoriza-Santos I, Mazzucco S, Stéphan O, Kociak M, et al. Zeptomol detection through controlled ultrasensitive surfaceenhanced Raman scattering. J Am Chem Soc 2009;131:4616–8. [86] Anema JR, Li JF, Yang ZL, Ren B, Tian ZQ. Shell-isolated nanoparticle-enhanced Raman spectroscopy. Nature 2010;464:392–5. [87] Aldeanueva-Potel P, Correa-Duarte MA, Álvarez-Puebla R, Liz-Marzán LM. Freestanding carbon nanotube films as optical accumulators for multiplex SERRS attomolar detection. ACS Appl Mater Interfaces 2010;2:19–22. [88] Polavarapu L, Liz-Marzán LM. Towards low-cost flexible substrates for nanoplasmonic sensing. Phys Chem Chem Phys 2013;15:5288–300. [89] Polavarapu L, Porta AL, Novikov SM, Coronado-Puchau M, Liz-Marzán LM. Pen-onpaper approach toward the design of universal surface enhanced Raman scattering substrates. Small 2014;10:3065–71.

269

[90] Seelenbinder JA, Brown CW, Pivarnik P, Rand AG. Colloidal gold filtrates as metal substrates for surface-enhanced infrared absorption spectroscopy. Anal Chem 1999;71:1963–6. [91] Lal S, Grady NK, Kundu J, Levin CS, Lassiter JB, Halas NJ. Tailoring plasmonic substrates for surface enhanced spectroscopies. Chem Soc Rev 2008;37: 898–911. [92] Kundu J, Le F, Nordlander P, Halas NJ. Surface enhanced infrared absorption (SEIRA) spectroscopy on nanoshell aggregate substrates. Chem Phys Lett 2008;452:115–9. [93] Wang H, Kundu J, Halas NJ. Plasmonic nanoshell arrays combine surface-enhanced vibrational spectroscopies on a single substrate. Angew Chem Int Ed 2007;46: 9040–4. [94] Cai Q, Hu F, Lee S-T, Liao F, Li Y, Shao M. Controllable Fe3O4/Au substrate for surface-enhanced infrared absorption spectroscopy. Appl Phys Lett 2015;106: 023107. [95] Lakowicz JR. Radiative decay engineering: biophysical and biomedical applications. Anal Biochem 2001;298:1–24. [96] Mainak M, Mondal C, Chowdhury J, Pal J, Pal A, Pal T. The tuning of metal enhanced fluorescence for sensing applications. Dalton Trans 2014;43:1032–47. [97] Xue C, Xue Y, Dai L, Urbas A, Li Q. Size- and shape-dependent fluorescence quenching of gold nanoparticles on perylene dye. Adv Opt Mater 2013;1:581–7. [98] Chen Y, Munechika K, Ginger DS. Dependence of fluorescence intensity on the spectral overlap between fluorophores and plasmon resonant single silver nanoparticles. Nano Lett 2007;7:690–6. [99] Tam F, Goodrich GP, Johnson BR, Halas NJ. Plasmonic enhancement of molecular fluorescence. Nano Lett 2007;7:496–501. [100] Li ZQ, Chen S, Li JJ, Liu QQ, Sun Z, Wang ZB, et al. Plasmon-enhanced upconversion fluorescence in NaYF4:Yb/Er/Gd nanorods coated with Au nanoparticles or nanoshells. J Appl Phys 2012;111:014310. [101] Willets KA. Super-resolution imaging of interactions between molecules and plasmonic nanostructures. Phys Chem Chem Phys 2013;15:5345–54. [102] Cang H, Labno A, Lu C, Yin X, Liu M, Gladden C, et al. Probing the electromagnetic field of a 15-nanometre hotspot by single molecule imaging. Nature 2011;469: 385–8. [103] Thomas M, Greffet JJ, Carminati R, Arias-González JR. Single-molecule spontaneous emission close to absorbing nanostructures. Appl Phys Lett 2004;101: 3863–5. [104] Lakowicz JR, Shen Y, D'Auria S, Malicka J, Fang J, Gryczynski Z, et al. Radiative decay engineering. 2. Effects of silver island films on fluorescence intensity, lifetimes, and resonance energy transfer. Anal Biochem 2002;301:261–77. [105] Kinkhabwala A, Yu Z, Fan S, Avlasevich Y, Müllen K, Moerner WE. Large singlemolecule fluorescence enhancements produced by a bowtie nanoantenna. Nat Photonics 2009;3:654–7. [106] Sánchez-González A, Corni S, Mennuci B. Surface-enhanced fluorescence within a metal nanoparticle array: the role of solvent and plasmon couplings. J Phys Chem C 2011;115:5450–60. [107] Cui Q, He F, Li L, Möhwald H. Controllable metal-enhanced fluorescence in organized films and colloidal system. Adv Colloid Interface Sci 2014;207:164–77. [108] Tabakman SM, Lau L, Robinson JT, Price J, Sherlock SP, Wang H, et al. Plasmonic substrates for multiplexed protein microarrays with femtomolar sensitivity and broad dynamic range. Nat Commun 2011;2:466. [109] Bardhan R, Grady NK, Cole JR, Joshi A, Halas NJ. Fluorescence enhancement by au nanostructures: nanoshells and nanorods. ACS Nano 2009;3:744–52. [110] Aslan K, Lakowicz JR, Szmacinski H, Geddes CD. Metal-enhanced fluorescence solution-based sensing platform. J Fluoresc 2004;14:677–9. [111] Wilson R, Nicolau DV. Separation-free detection of biological molecules based on plasmon-enhanced fluorescence. Angew Chem Int Ed 2011;50:2151–4. [112] Tyagi S, Kramer FR. Molecular beacons: probes that fluoresce upon hybridization. Nat Biotechnol 1996;14:303–8. [113] Dubertret B, Calame M, Libchaber AJ. Single-mismatch detection using goldquenched fluorescent oligonucleotides. Nat Biotechnol 2001;19:365–70. [114] Cheng Y, Stakenborg T, Van Dorpe P, Lagae L, Wang M, Chen H, et al. Fluorescence near gold nanoparticles for DNA sensing. Anal Chem 2011;83:1307–14. [115] Zhou ZP, Huang HD, Chen Y, Liu F, Huang CZ, Li N. A distance-dependent metalenhanced fluorescence sensing platform based on molecular beacon design. Biosens Bioelectron 2014;52:367–73. [116] Goel G, Kumar A, Puniya AK, Chen W, Singh K. Molecular beacon: a multitask probe. J Appl Microbiol 2005;99:435–42. [117] Kang WJ, Cho YL, Chae JR, Lee JD, Choi K-J, Kim S. Molecular beacon-based bioimaging of multiple microRNAs during myogenesis. Biomaterials 2011;32:1915–22. [118] Schneider G, Decher G. Distance-dependent fluorescence quenching on gold nanoparticles ensheathed with layer-by-layer assembled polyelectrolytes. Nano Lett 2006;6:530–6. [119] Hakonen A, Strömberg N. Plasmonic nanoparticle interactions for high-performance imaging fluorosensors. Chem Commun 2011;47:3433–5. [120] Zhang B, Price J, Hong G, Tabakman SM, Wang H, Jarrell JA, et al. Multiplexed cytokine detection on plasmonic gold substrates with enhanced near-infrared fluorescence. Nano Res 2013;6:113–20. [121] Zhang B, Kumar RB, Dai H, Feldman BJ. A plasmonic chip for biomarker discovery and diagnosis of type 1 diabetes. Nat Med 2014;20:948–53. [122] Ozel T, Nizamoglu S, Sefunc MA, Samarskaya O, Ozel IO, Mutlugun E, et al. Anisotropic emission from multilayered plasmon resonator nanocomposites of isotropic semiconductor quantum dots. ACS Nano 2011;5:1328–34. [123] Aroca R, Kovacs GJ, Jennings CA, Loutfy RO, Vincett PS. Fluorescence enhancement from Langmuir–Blodgett monolayers on silver island films. Langmuir 1988;4: 518–21.

270


[124] Sugawa K, Tamura T, Tahara H, Yamaguchi D, Akiyama T, Otsuki J, et al. Metalenhanced fluorescence platforms based on plasmonic ordered copper arrays: wavelength dependence of quenching and enhancement effects. ACS Nano 2013; 7:9997–10010. [125] Wang HS, Wang C, He YK, Xiao FN, Bao WJ, Xia XH, et al. Core–shell Ag@ SiO(2) nanoparticles concentrated on a micro/nanofluidic device for surface plasmon resonance-enhanced fluorescent detection of highly reactive oxygen species. Anal Chem 2014;86:3013–9.

[126] Krishnamoorthy S. Nanostructured sensors for biomedical applications — a current perspective. Curr Opin Biotechnol 2015;34:118–24. [127] Moshe AB, Szwarcman D, Markovich G. The size dependence of chiroptical activity in colloidal quantum dots. ACS Nano 2011;5:9034–43. [128] Lombardini R, Acevedo R, Halas NJ, Johnson BR. Plasmonic enhancement of Raman optical activity in molecules near metal nanoshells: theoretical comparison of circular polarization methods. J Phys Chem C 2010;114:7390–400.