ARTICLE pubs.acs.org/Langmuir
Mechanistic Study of Silver Nanoparticle Formation on Conducting Polymer Surfaces Nathan H. Mack, James A. Bailey, Stephen K. Doorn, Chien-An Chen, Han-Mou Gau, Ping Xu, Darrick J. Williams, Elshan A. Akhadov, and Hsing-Lin Wang* Los Alamos National Laboratory, Los Alamos, New Mexico 87545, United States
Downloaded via UNIV OF WINNIPEG on December 27, 2018 at 00:20:43 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
bS Supporting Information ABSTRACT: Conducting polymer (polyaniline) sheets are shown to be active substrates to promote the growth of nanostructured silver thin films with highly tunable morphologies. Using the spontaneous electroless deposition of silver, we show that a range of nanostructured metallic features can be controllably and reproducibly formed over large surface areas. The structural morphology of the resulting metalpolymer nanocomposite is demonstrated to be sensitive to experimental parameters such as ion concentration, temperature, and polymer processing and can range from densely packed oblate nanosheets to bulk crystalline metals. The deposition mechanisms are explained using a diffusion-limited aggregation (DLA) model to describe the semi-fractal-like growth of the metal nanostructures. We find these composite films to exhibit strong surfaceenhanced Raman (SERS) activity, and the nanostructured features are optimized with respect to SERS activity using a selfassembled monolayer of mercapto-benzoic acid as a model Raman reporter. SERS enhancements are estimated to be on the order of 107. Through micro-Raman SERS mapping, these materials are shown to exhibit uniform SERS responses over macroscopic areas. These metalpolymer nanocomposites benefit from the underlying polymer’s processability to yield SERS-active materials of almost limitless shape and size and show significant promise for future SERS-based sensing and detection schemes.
’ INTRODUCTION Conducting polymers (CPs) are an exciting class of materials that enable entirely new approaches to material design and synthesis.1,2 These conjugated organic polymers exhibit unique redox properties that may be tuned according to multiple independent structural and processing characteristics.3 Their freely conducting electrons along with any intrinsic charges associated with the polymer backbone allow properly prepared CPs to complex with metal ions in solution and subsequently reduce them to zero-valent metals in the absence of any externally applied electric field.4 This electroless deposition approach is generally applicable to many metal ions that have commensurate redox potentials. For example, Ag, Au, and Pt have all been demonstrated to form metallic structures electrolessly on conducting polymer substrates and have promising applications in the facile preparation of intricate conductive patterns and surfaces.5 Recently, conducting polymers have been used to produce metal nanoparticle (MNP) composites that exhibit enhanced catalytic, electrical, and optical properties.68 These composite materials’ performance characteristics are very dependent upon the crystal structure, surface area, grain size, and morphology of their MNP components. To maximize the nanocomposite’s device performance, it is increasingly important to have a detailed understanding of the underlying mechanisms r 2011 American Chemical Society
responsible for the structure and morphology of the resulting metal nanoparticles. This may then be utilized to control the MNP structure through subtle changes in the conducting polymer’s processing method, surface chemistry, and electrochemical state. This highly tunable approach to MNP synthesis has significant implications for generating surface-enhanced Raman spectroscopy (SERS)-active materials. Recent studies on lithographically defined and aggregated metal nanoparticles reveal unique optical properties that are extremely sensitive to local interparticle and intraparticle geometries.9,10 Metal nanoparticles with sharp edges and a high radius of curvature have strong electric fields associated with them upon optical plasmon absorption, and the overlap of these fields between nearby interacting particles leads to intense SERS enhancements of up to 1010 under the proper conditions.11 In general, particle architectures that contain a large number of nanostructured intersections and junctions are found to be extremely efficient SERS scatterers.12,13 However, current fabrication processes for SERS-active materials are limited by substrate materials and structural inhomogeneities that result in highly irreproducible large SERS enhancements. Chemical- and Received: September 10, 2010 Revised: February 23, 2011 Published: March 24, 2011 4979
dx.doi.org/10.1021/la103644j | Langmuir 2011, 27, 4979–4985
Langmuir vacuum metal deposition-based approaches often trade reproducibility for sensitivity at the expense of complexity and cost, making them less suitable for large-scale ultrasensitive detection applications.14 These current deficiencies in SERS technology drive the need for novel SERS substrates that are capable of generating a wide range of particle sizes, structures, and morphologies, along with the ability to vary their structural characteristics and, by extension, their associated optical SERS properties easily. From the vast array of substrate materials available for nanoscale synthesis,15 conducting polymer thin films enable large-scale polymer processing techniques to be applied toward novel SERSactive material design. One potentially useful polymer in SERS-active metal nanocomposites is the conducting polymer polyaniline (PANI), which is readily synthesized, has well-developed processing methods, and is environmentally stable over relatively long periods.16 PANI’s unique redox properties make it ideally suited for generating complex MNP architectures through the electroless deposition of aqueous metal salt solutions (e.g., AgNO3).5 In this work, we present a detailed study of the growth mechanisms associated with Ag metal nanoparticles on a citric acid-doped PANI membrane using scanning electron microscopy (SEM) to observe the formation of nanoscale Ag particles on the PANI surface. Morphological control of nanostructured Ag particles is demonstrated by varying several experimental parameters to reveal a mechanistic understanding of the Ag particle growth process, which permits finely tuned control over the nanoscale metal morphology. The aggregate MNP surfaces are demonstrated to be highly SERSactive with large, reproducible, homogeneous signal enhancements. These novel complex nanoarchitectures represent a new approach to generating functional nanostructured surfaces that enables the fabrication of SERS-active materials not realizable using current synthesis methods.
’ EXPERIMENTAL SECTION Reagents. N-Methyl-2-pyrrolidone (99%, Aldrich), heptamethylenimine (HPMI, 98%, Acros), citric acid (99.9%, Fisher), silver nitrate (99.9999%, Aldrich), hydrochloric acid (37%, Fisher), 4-mercaptobenzoic acid (90%, Aldrich), and polyaniline emeraldine base (EB) powder (Aldrich) were used as received. PANI Membrane Preparation. Concentrated PANI solutions (20 % w/w) were prepared as follows. First, 1.15 g of EB powder was mixed with 0.747 g of heptamethyleneimine (HPMI) in a 12 mL Teflon vial for 10 min with a spatula. Then, 4.14 g of N-methyl-2-pyrrolidone (NMP) was added and stirred with a spatula for 30 min until the visible solid PANI particles were dispersed. The PANI solution was filtered through a 10 mL syringe using a cotton plug and then poured onto a glass substrate and screeded into a homogeneous thin film using a Gardner blade. The glass substrate and film were kept in a leveled oven at 55 °C for 2 days to remove the majority of organic solvents. The dried film was then detached from the glass substrate by immersing it in a ∼4 L water bath overnight with complete water exchange done after 5 h to assist in solvent exchange. After being separated from the glass substrate, the film was kept in water for an additional 4 h to ensure the complete removal of any residual NMP and HPMI. The pristine PANI films were then dried in air for 1 h, sandwiched between two KimWipes, and placed under a flat metal weight (15 kg) for 1 day to minimize the bending of the dry film. The resulting thermally cured, dense PANI film has a thickness of 1350 μm (as measured by a micrometer or vernier calipers) depending on the thickness setting of the Gardner blade. The films were then doped by immersing them into a citric acid solution
ARTICLE
(0.25 M) for 1 week, followed by three rinses in 100 mL of purified water for 130 min. SERS-Active Silver Substrate Preparation. The citric aciddoped films were cut into 5 5 mm2 pieces and immersed into freshly prepared aqueous AgNO3 solutions. Metallic nanostructures immediately and spontaneously begin to form on the PANI surface. Rinsing the PANI film in water for 5 s halts all metal growth, and the metal structures are finally dried in air. Reaction conditions such as the metal ion concentration and temperature were used to modify the resulting metal nanostructures. Raman Spectroscopy. All substrates were prepared for spectroscopic analysis by forming a self-assembled monolayer of 4-mercaptobenzoic acid (MBA) or thiobenzene (TB) on their surfaces. Each substrate was immersed in a 5 mM ethanolic solution of the thiol for 1 h and then rinsed with fresh ethanol three times and allowed to air dry. Surface-enhanced Raman spectra were recorded using approximately 1 mW of 785 nm excitation focused onto the sample through a 20 (0.5 NA) microscope objective. The scattered Raman signal was collected in a backscattering configuration through the objective, filtered, and then dispersed onto a liquid-nitrogen-cooled CCD camera through a single grating spectrometer. All spectra were recorded using a total integration time of 10 s. Raman Mapping. SERS spectra were recorded on a Kaiser microRaman apparatus using an argon ion excitation laser at 514 nm (∼100 μW). The laser was focused onto the sample through a 100 (0.9 NA) microscope objective using a 250 ms total integration time for each individual spectrum. The substrates were then rastered through the laser focal point, and an image was constructed using the integrated Raman signal of the prominent 1075 cm1 CC ring-breathing mode associated with the MBA monolayer. SEM Microscopy. SEM micrographs were recorded on an FEI Quanta 400F electron microscope operating at 20 kV. All nanostructured Ag samples were imaged as prepared. XRD Spectroscopy. X-ray diffraction measurements were made on a Rigaku Ultima III diffractometer using fine-line-sealed Cu KR tube ) X-rays. The generator is a D/MAX Ultima series with a (λ = 1.5406 Å maximum power of 3 kW. The samples were mounted onto an amorphous quartz slide. Data sets were collected in continuous scan mode in BraggBrentano slit geometry over the 2θ range from 10 to 90° with a sampling width of 0.05° and a scanning speed of 3°/min. The divergence slits were set to 2.0 and 10 mm, and the scattering and receiving slits were set to 2 and 0.6 mm.
’ RESULTS AND DISCUSSION The overall surface density of the nanostructured metals on the PANI surface is largely dictated by the underlying PANI substrate polymer morphology. Figure 1a,b shows SEM micrographs of Ag nanoparticles grown on PANI membranes that were prepared via our standard phase inversion method as well as phase inversion followed by mechanical pressing, respectively. The highly contrasting surface coverages result from slight differences in the polymer film morphology, which is very dependent on processing conditions prior to Ag deposition. The as-prepared PANI membranes (Figure 1a) exhibit a low surface coverage density of Ag metal particles, which tend to grow in spherical particle aggregates consisting of densely packed oblate substructures that have long dimensions on the order of a few micrometers and short dimensions on the order of 50 nm. This surface heterogeneity is presumably due to local differences in the doping states of the PANI polymer as well as density variations of the polymer chains at the film surface, both of which determine the nucleation site density at the PANI surface. As a demonstration of this density dependence, the as-prepared PANI 4980
dx.doi.org/10.1021/la103644j |Langmuir 2011, 27, 4979–4985
Langmuir
ARTICLE
Scheme 1. Alternative Growth Modes for Ag Particles on a PANI Surfacea
a
Figure 1. SEM micrographs of Ag particles formed on PANI substrates at room temperature under identical concentrations (20 mM AgNO3): (a) Low-density Ag nucleation on a solution-precipitated PANI film (10 μm scale bar). (Inset) High-magnification SEM micrograph of an individual nucleation site (1 μm scale bar). (b) High-density Ag nucleation on a mechanically pressed PANI film (20 μm scale bar). (Inset) High-magnification SEM micrograph of interconnected Ag nanosheets (2 μm scale bar). (c) Select SEM micrographs of Ag particles during different stages of growth (1 μm scale bar).
membrane was mechanically compressed into a densely packed film using a hydraulic press, which has a net effect of increasing the overall density of the membrane and hence the relative number of nucleation sites on the PANI surface (Figure 1b). Upon exposure to an AgNO3 solution, complete surface coverage with Ag nanosheet structures is observed. The remnants of the original nucleation sites can be resolved in the Ag structures as slight imperfections in the random nanosheet network. From these data, it is apparent that under identical Agþ ion concentrations very similar Ag structures are formed; however, the relative density of nucleation sites in this latter sample is high enough that neighboring sites grow into each other and fill in most of the PANI surface with a continuous, homogeneous nanostructured Ag surface. These large changes in the volume density of the PANI membrane significantly alter the relative concentration of nucleation sites on the polymer surface, which produces Ag particle surface coverages that range from highly scattered Ag particles to completely covered PANI surfaces. A closer inspection of individual nucleation sites during different stages of growth reveals detailed aspects of the Ag deposition mechanism (Figure 1c). The initial nanoscale Ag particles consist of a few nanosheets clustered around a nucleation site. As the clusters grow, new nanosheets are observed to add onto the nucleation site in a repeating random pattern. This unique growth process is the result of the repetitive dissection of individual oblate Ag particles, with preferential deposition occurring at the edges of the oblate structures, resulting in thin sheetlike morphologies. As new Agþ ions are added, the sheets proceed to ripen and eventually split to form multiple new Ag sheets. This process is found to be uniform around all PANI nucleation sites, resulting in homogeneous nanoscale Ag morphologies, and provided that there is a high enough nucleation site density, these structures tend to grow into each other to form a translationally invariant nanostructured Ag surface across the
(a) DLA fractal growth. (b) Surface growth.
entire PANI substrate. This result has large implications for the overall homogeneity of various optical properties, as will be demonstrated via SERS. The deposition and growth processes observed in these complex nanostructured Ag surfaces are influenced not only by the nucleation site density on the PANI surface but also by a wide range of experimental parameters such as temperature and the concentration of the AgNO3 solution. The synthetic environment (i.e., under water) makes the direct observation of the nanoparticle growth process difficult; however, the underlying physics of this process is largely governed by both the diffusion kinetics of the Agþ ions in solution and the dynamics of Agþ reduction onto the growing Ag metal particle. Therefore, we speculate on a growth mechanism based on concepts adopted from diffusion-limited aggregation (DLA) models as well as surface adsorption and reaction kinetics.17,18 In these models, randomly diffusing Agþ ions are predicted to add to nucleation sites on the basis of their diffusion rates and the sticking probability of the ion onto the growing Ag metal surface. In a diffusion-limited regime, the sticking coefficient is large and isotropic across the Ag surface, leading to highly branched fractal forest structures assuming Brownian diffusion (Scheme 1a);19 however, SEM micrographs of these Ag surfaces show fractal branching structures with a clear preference for nanosheet growth. This implies that Agþ ions deposit on the PANI surface according to a diffusion-limited model but in a highly anisotropic fashion at the nucleated Ag particles, where the sticking coefficient at the edges of the oblate sheet structures is higher than on the faces. In this semifractal growth regime, this anisotropic deposition of Agþ ions is most likely due to preferential deposition on these reactive Ag crystal faces, resulting in highly branched nanosheet crystallites. Conversely, under conditions where the sticking coefficient is low, the addition of new ions to the metal surface is dictated by surface reaction kinetics and energetics; in this case, a simple reduction of the Agþ ions at the Ag surface is presumed to be extremely fast and isotropic, leading to structures that assembled via isotropic surface growth (Scheme 1b).18 These two growth regimes are very sensitive to temperature and ion concentration and provide unique handles with which to manipulate the overall Ag particle morphology. Typical Ag particle morphologies for structures grown on citric acid-doped PANI membranes are shown in Figure 2a, where metal structures have been grown from aqueous solutions with varying Agþ concentrations. The spontaneously grown structures are observed to range from tightly packed assemblies of extremely thin Ag sheets at high ion concentration (100 mM) to loosely packed Ag sheets that are measurably thicker at intermediate concentration (20 mM) to highly faceted micrometer-sized Ag 4981
dx.doi.org/10.1021/la103644j |Langmuir 2011, 27, 4979–4985
Langmuir
Figure 2. SEM micrographs of Ag particles grown at room temperature on PANI membranes at varying AgNO3 concentrations: (a) 1, (b) 2, (c) 3, (d) 5, (e) 10, (f) 20, (g) 50, and (h) 100 mM. The scale bar is 2 μm. (Bottom plot) Representative XRD spectra of deposited Ag particles.
single crystals at low ion concentration (1 mM). These dramatically different metal morphologies hint at the underlying mechanisms associated with the reduction of metal ions at the PANI surface. In the case of relatively high Agþ concentrations, the Ag aggregates are composed of extremely densely packed Ag sheets, each with a thickness on the order of tens of nanometers (Figure 2viviii). At these relatively high ion concentrations, reactive nucleation sites on the polymer surface are immediately occupied by an Agþ ion to form an initial metal ionpolymer complex that is subsequently reduced to a zero-valent metal. This particle then acts as a preferential nucleation site for all subsequent
ARTICLE
Agþ reduction. The reduction of ions by PANI is assumed to be nearly instantaneous; therefore, diffusion-limited mechanisms dominate the observed particle morphologies, mainly in the form of highly bifurcated Ag structures similar to those in Scheme 1a. This is reflected by the densely packed thin Ag sheets observed in the SEM micrographs in which the particles tend to form new Ag sheets rather than isotropically ripen existing sheets. At lower ion concentrations, relative to the concentration of active nucleation sites on the polymer surface, the local solution surface layers become Agþ depleted because they are reduced faster than they can diffuse in from the bulk solution. The initially high diffusion rates yielding fractal growth according to the DLA model no longer hold true once this Agþ depletion layer is formed adjacent to the PANI surface. Instead, in this regime, the highly bifurcated Ag sheets begin to develop into much thicker sheets as the growth kinetics become rate-limited by ion diffusion. This ripening effect is magnified at very low relative concentration, resulting in the slow isotropic growth of Ag single crystals similar to that shown in Scheme 1b. These structural changes in the Ag particles are reflected in the XRD patterns of the respective samples, as shown in Figure 2. The patterns include numerous diffraction peaks for all of the AgNO3 concentrations studied; however, there are a few notable differences between the high-concentration densely packed Ag nanosheets and low-concentration single-crystal Ag structures. With decreasing Ag ion concentration, there is a progressive sharpening and increase in intensity of the (111) and (200) peaks that begin to resemble the spectra of bulk Ag, a result consistent with the micrometer-sized Ag single crystals observed in the SEM images at lower Ag ion concentrations. The Ag morphologies grown at ion concentrations above ∼5 mM are dominated by high-aspect-ratio oblate nanosheet structures, which are estimated to be on the order of several micrometers in diameter on their long axis and yet only tens of nanometers in thickness on the short axis. On the basis of the average fwhm of the peaks in the XRD spectra, the calculated particle size is found to vary from ∼60 nm at low Agþ concentration to ∼30 nm at high Agþ concentration,20 suggesting that the observed morphologies are actually composed of small Ag domains (as opposed to Ag single crystals) assembled into high-aspect-ratio structures. This is consistent with a broadened (111) and (200) XRD pattern, and the constant (211) and (300) peaks indicate axial growth along these respective directions. Closely related to the concentration effects, the temperature of the deposition solution plays an important role in determining the final Ag metal morphology both by dictating the diffusion rates of the Agþ to the PANI surface and by determining the sticking coefficient for newly added ions. At constant Agþ concentration (20 mM), SEM micrographs of the resulting Ag metals grown on identical PANI substrates at temperatures ranging from 2 to 80 °C yield widely varying morphologies that in many ways are similar to the results of the concentration dependence (Figure 3) and are readily apparent in the XRD spectra of the Ag substrates.20 Again, densely packed oblate sheet structures are observed at lower temperatures. These begin to increase in overall size and thickness as the temperature is increased. Above 50 °C, the Ag metal clusters trend toward bulk particles without any nanostructured characteristics. It should be noted here that SEM images were taken on regions of the substrate that were not completely covered with Ag particles in order to isolate individual Ag morphologies; however, typically the entire PANI surface is covered with these types of substructures. Additionally, the growth rate of the Ag particles on the PANI 4982
dx.doi.org/10.1021/la103644j |Langmuir 2011, 27, 4979–4985
Langmuir
ARTICLE
Figure 3. SEM micrographs of Ag particles grown at a constant concentration (20 mM AgNO3) on a PANI membrane at various temperatures: (a) 2, (b) 20, (c) 25, (d) 30, (e) 50, and (f) 80 °C. The scale bar is 5 μm.
substrate was found to be extremely sensitive to temperature. For example, at a growth temperature of 5 °C, it took 96 h to form a completely covered PANI surface, and at 50 °C, the same coverage was reached in only a few minutes. These deposition characteristics, at low temperatures, are consistent with a diffusion-limited model where the diffusion of the ions into the depletion layer near the PANI surface is the rate-limiting step during deposition because the sticking probability of the ions for a nucleated particle is high. In general, the two competing processes—diffusion and reaction kinetics—are dependent on temperature, although with opposite effects. According to the StokesEinstein relation, temperature is directly proportional to the diffusion coefficient of a particle (Agþ), thus increased temperatures will generally decrease the size of the depletion layer at the PANI surface because Agþ is able to diffuse to the surface relatively faster. This makes a highertemperature Agþ solution effectively act like a highly concentrated solution, yet from these data, the two regimes result in highly divergent Ag morphologies—oblate sheets for high concentrations and large single crystals for high temperatures. From these results, it is apparent that additional surface reaction effects with the nucleated Ag particles must be taken into account and are very relevant to the final Ag morphology. DLA-type models typically assume unit sticking probabilities for the growing particle assemblies; however, according to eq 1, the sticking coefficient (S) of a particle (ion) is inversely related to the exponential of temperature (T), meaning that higher temperatures result in significantly lowered interaction of the ions with any
Figure 4. (a) SERS spectra of MBA on Ag/PANI substrates fabricated from AgNO3 at varying concentrations. (b) SERS spectra of MBA on Ag/PANI substrates fabricated from a 20 mM AgNO3 aqueous solution at different temperatures. Insets indicate the maximum SERS intensity at 1075 cm1.
nucleated Ag metal particles.18 Sµ
1 eT
ð1Þ
This altered sticking probability results in surface growth (ripening) instead of fractal growth and is manifested as dramatic variations in the overall Ag morphology, with the end result being bulk crystalline materials observed at high temperature. This interplay between the fractal and surface growth mechanisms (Scheme 1) is dictated by the experimental conditions of the deposition process, such as temperature, concentration, and PANI nucleation site density. Using these experimental levers, we have demonstrated a unique capability for controlled engineered nanostructured Ag materials. The controlled deposition of nanostructured Ag on the PANI membranes provides an ideal platform from which to investigate the optical properties of a wide variety of particle morphologies. 4983
dx.doi.org/10.1021/la103644j |Langmuir 2011, 27, 4979–4985
Langmuir
Figure 5. (a) Spatial Raman map of an Ag/PANI substrate. (b) Histogram of SERS intensity from the spatial map.
As a sensing platform, particle morphologies with efficient SERS activity are of particular interest in extremely sensitive SERSbased sensing materials. The unique control over the Ag nanostructured morphology that is enabled via this PANI-assisted deposition process allows the optimization of the metal’s optical properties for maximal SERS enhancement. The SERS response of the Ag nanostructures was measured by utilizing a model analyte (MBA) that spontaneously forms a well-behaved selfassembled monolayer (SAM) on the metal surface.21 The measured SERS spectra for the temperature- and concentration-dependent series of PANIAg films are shown in Figure 4. The overall trend in both series indicates a maximal SERS intensity for substrates formed under conditions that generate uniform sheetlike features; at a constant nucleation density, this typically corresponds to intermediate concentrations and temperatures for the ranges investigated. These ideal conditions (1020 mM AgNO3 at 1020 °C) typically consist of Ag nanosheet morphologies with an edge dimension of approximately 50 nm, a size regime that is commensurate with an efficient SERS response from spherical Ag nanoparticle-based SERS substrates.22 Ag morphologies formed outside of the ideal deposition conditions exhibit relatively weak SERS activity—of approximately an order of magnitude—because they yield either very thin nanosheets or single-crystal Ag features; neither structure supports efficient plasmon coupling. A limited excitation profile indicates increased SERS activity as the Raman excitation energy
ARTICLE
shifts from 827 to 700 nm,20 signifying increased overlap of the excitation energy with the plasmon resonance energy of these structures. It is important to note that the observed SERS signal is extremely clean and all peaks correspond to an MBA monolayer with no extraneous resonances that could be associated with contaminants on the Ag surface or from the underlying PANI polymer substrate. With data from the optimized temperature, concentrations, and 785 nm excitation, the PANI substrates yield an enhancement factor of ∼1 107.20 Perhaps more important than the sheer SERS enhancement is the homogeneity of the SERS signal across a large-area substrate. To demonstrate these substrates as viable SERS sensing materials, a surface consisting of Ag nanosheets that completely cover the PANI surface was coated with an MBA SAM and then mapped using a scanning Raman microscope. The measurements were made over a surface area of ∼50 70 μm2 using 2.5 μm steps, and a Raman image of the substrate was constructed using the relative SERS intensity of the integrated 1075 cm1 stretching mode of the MBA monolayer (Figure 5a). The 2.5 μm step size was chosen such that minimal overlap was expected between successive point acquisitions, yet it was small enough to be commensurate with the dimensions of the large-diameter nanosheet structures. The resulting image shows a homogeneous SERS surface with the peak-to-peak intensity varying by at most 1 104 counts. A histogram of these intensities normalized to the mean indicates only 9% deviation over the entire area (Figure 5b). The SEM micrographs of these materials indicate that the Ag structures deposit over the entire area of the PANI film, thus this Raman image represents the macroscopic optical quality of the overall Ag film. Assuming that the PANI processing steps are uniform over the area of the film and the temperature and Agþ concentrations are well-defined, this facile fabrication method can in principle be extended to films of almost limitless dimensions. The exceptional uniformity and relatively large SERS enhancements make these Ag-coated PANI substrates ideal detection platforms for eventual single-molecule vibrational spectroscopy using low-cost, disposable SERS sensors. In summary, we have demonstrated the depositional control of metallic Ag particles onto PANI thin films with structures ranging from highly complex nanosheet assemblies to micrometer-sized crystalline particles. The morphologies of these structured films were found to be very sensitive to local metal ion concentration, deposition temperature, and PANI processing steps. The resulting structures can be rationalized according to diffusion mechanisms associated with the metal ions approaching the polymer surface as well as basic surface adsorption and growth mechanisms at the metal particle’s solution interface. The optimized metal structures result in uniform optical properties that are strongly correlated with SERS activity, making them uniquely suited for novel ultrasensitive detection applications.
’ ASSOCIATED CONTENT
bS
Supporting Information. Detailed descriptions of enhancement factor calculations, XRD spectra, and SERS excitation profiles. This material is available free of charge via the Internet at http://pubs.acs.org.
’ ACKNOWLEDGMENT We gratefully acknowledge financial support from Laboratory Directed Research and Development (LDRD), funded under the 4984
dx.doi.org/10.1021/la103644j |Langmuir 2011, 27, 4979–4985
Langmuir
ARTICLE
auspices of the DOE. This work is supported in part by the Department of Energy, Office of Basic Energy Sciences, Division of Materials Science and Engineering as well as the National Nanotechnology Enterprise Development Center (NNEDC). This work was performed in part at the U.S. Department of Energy, Center for Integrated Nanotechnologies, at Los Alamos National Laboratory (contract DE-AC52-06NA25396) and Sandia National Laboratories (contract DE-AC04-94AL85000).
’ REFERENCES (1) Gangopadhyay, R.; De, A. Chem. Mater. 2000, 12, 608–622. (2) Gerard, M.; Chaubey, A.; Malhotra, B. D. Biosens. Bioelectron. 2002, 17, 345–359. (3) MacDiarmid, A. G. Rev. Mod. Phys. 2001, 73, 701. (4) Dimeska, R.; Murray, P. S.; Ralph, S. F.; Wallace, G. G. Polymer 2006, 47, 4520–4530. (5) Li, W.; Jia, Q. X.; Wang, H.-L. Polymer 2006, 47, 23–26. (6) Gao, Y.; Chen, C.-A.; Gau, H.-M.; Bailey, J. A.; Akhadov, E.; Williams, D.; Wang, H.-L. Chem. Mater. 2008, 20, 2839–2844. (7) Wang, H.-L.; Li, W.; Jia, Q. X.; Akhadov, E. Chem. Mater. 2007, 19, 520–525. (8) Xu, P.; Han, X.; Zhang, B.; Mack, N. H.; Jeon, S.-H.; Wang, H.-L. Polymer 2009, 50, 2624–2629. (9) Musick, M. D.; Keating, C. D.; Lyon, L. A.; Botsko, S. L.; Pena, D. J.; Holliway, W. D.; McEvoy, T. M.; Richardson, J. N.; Natan, M. J. Chem. Mater. 2000, 12, 2869–2881. (10) Willets, K. A.; Van Duyne, R. P. Annu. Rev. Phys. Chem. 2007, 58, 267–297. (11) Doering, W. E.; Nie, S. J. Phys. Chem. B 2001, 106, 311–317. (12) Sun, Y.; Lei, C.; Gosztola, D.; Haasch, R. Langmuir 2008, 24, 11928–11934. (13) Sun, Y.; Pelton, M. J. Phys. Chem. C 2009, 113, 6061–6067. (14) Le Ru, E. C.; Blackie, E.; Meyer, M.; Etchegoin, P. G. J. Phys. Chem. C 2007, 111, 13794–13803. (15) Sun, Y. J. Phys. Chem. C 2009, 114, 857–863. (16) Genies, E. M.; Boyle, A.; Lapkowski, M.; Tsintavis, C. Synth. Met. 1990, 36, 139–182. (17) Vicsek, T. Fractal Growth Phenomena; World Scientific: Teaneck, NJ, 1989; p 355. (18) Masel, R. I. Principles of Adsorption and Reaction on Solid Surfaces; Wiley: New York, 1996. (19) Witten, T. A.; Sander, L. M. Phys. Rev. B 1983, 27, 5686. (20) See Supporting Information. (21) Diebold, E. D.; Mack, N. H.; Doorn, S. K.; Mazur, E. Langmuir 2009, 25, 1790–1794. (22) Moskovits, M. Surface-Enhanced Raman Spectroscopy: A Brief Perspective. In Surface-Enhanced Raman Scattering: Physics and Applications; Kneipp, K., Moskovits, M., Kneipp, H., Eds.; Topics in Applied Physics; Springer: New York, 2006; Vol. 103, pp 117.
4985
dx.doi.org/10.1021/la103644j |Langmuir 2011, 27, 4979–4985
