J. Phys. Chem. B 2006, 110, 25721-25728
25721
Seeded-Growth Approach to Fabrication of Silver Nanoparticle Films on Silicon for Electrochemical ATR Surface-Enhanced IR Absorption Spectroscopy Sheng-Juan Huo, Xiao-Kang Xue, Qiao-Xia Li, Su-Fan Xu, and Wen-Bin Cai* Shanghai Key Laboratory for Molecular Catalysis and InnoVatiVe Materials and Department of Chemistry, Fudan UniVersity, Shanghai 200433, China ReceiVed: June 28, 2006; In Final Form: October 4, 2006
Ag nanoparticle films (simplified as nanofilms hereafter) on Si for electrochemical ATR surface enhanced IR absorption spectroscopy (ATR-SEIRAS) have been successfully fabricated by using chemical deposition, which incorporates initial embedding of Ag seeds on the reflecting plane of an ATR Si prism and subsequent chemical plating of conductive and SEIRA-active Ag nanofilms. Two alternative methods for embedding initial Ag seeds have been developed: one is based on self-assembly of Ag colloids on an aminosilanized Si surface, whereas the other the reduction of Ag+ in a HF-containing solution. A modified silver-mirror reaction was employed for further growth of Ag seeds into Ag nanofilm electrodes with a theoretically average thickness of 40-50 nm. Both Ag seeds and as-deposited Ag nanofilms display island structure morphologies facilitating SEIRA, as revealed by AFM imaging. The cyclic voltammetric feature of the as-prepared Ag nanofilm electrodes is close to that of a polycrystalline bulk Ag electrode. With thiocyanate as a surface probe, enhancement factors of ca. 50-80 were estimated for the as-deposited Ag nanofilms as compared to a mechanically polished Ag electrode in the conventional IRAS after reasonable calibration of surface roughness factor, incident angles, surface coverage, and polarization states. As a preliminary example for extended application, the pyridine adsorption configuration at an as-deposited Ag electrode was re-examined by ATRSEIRAS. The results revealed that pyridine molecules are bound via N end to the Ag electrode with its ring plane perpendicular or slightly tilted to the local surface without rotating its C2 axis about the surface normal, consistent with the conclusion drawn by SERS in the literature.
1. Introduction Nanoparticle metal films possess unique optical properties as compared to their bulk counterparts, based on which surface enhanced spectroscopies have been developed rapidly and applied extensively.1-7 The intensification of infrared-active vibrational modes of molecules in close proximity to metallic nanoparticle films, commonly known as surface-enhanced infrared absorption (SEIRA),4,8,9 is receiving increased attention from both phenomenological10-14 and practical viewpoints.15-20 It has been employed in molecular-level analysis and characterization of both metal-ambient4,9,12,14 and metal-liquid interfaces.15-19,21,22 Specifically, surface-enhanced infrared absorption spectroscopy (SEIRAS), with attenuated total reflection (ATR) configuration, is among the most powerful analytical tools for in situ probing surface adsorption and reaction at electrodes because of its high surface sensitivity, simple selection rule, and feasibility for real-time measurement.4 To implement successfully this technique in surface electrochemistry-related research, it is essential to fabricate SEIRAactive and conductive metallic nanofilms with well-tuned size and shape on an ATR window to serve as a working electrode. Additional requirements include high stability and good adherence, and comparable electrochemical behavior as their bulk electrode counterparts. There are generally two strategies, i.e., dry process and wet process, for fabricating the nanofilms on an ATR window for in situ electrochemical SEIRA spectroscopy.4 Historically, the predominant fabrication strategy is the * Address correspondence to this author. Phone: +86-21-55664050. Fax: +86-21-65641740. E-mail:
[email protected].
dry process involving the vacuum evaporation and sputtering. Recently, the wet process including chemical and electrochemical deposition has received intense attention in considering its simplicity, economy, and more importantly better reproducibility in SEIRA-activity and absence of severely distorted bipolar band shape. So far, the wet process has been successfully applied to obtain the ATR-SEIRA-active electrodes of Au,10,23 Pt,17,24,25 Pd,25 Ru,25 Rh,25and Ni11 on Si as well as Ag26,27 and Cu28,29 on Ge. Ge merits far more efficient IR transmission than Si in the spectral region below 950 cm-1. The Ag nanofilm electrode chemically deposited on Ge for ATR-SEIRAS was first reported by Rodes and his co-workers.26,27 The basic mechanism for chemical deposition involves the dissolution of Ge, which may be given as follows:30
Ag+ + e- f Ag Ge f Ge4+ + 4e-
E0 ) 0.799 V E0 ) -0.124 V
(I) (II)
On the other hand, Ge per se possesses the inherent disadvantage of a higher conductivity and activity as compared to Si in most electrolytes over a broad window of working potentials. Consequently, a Ge substrate per se may involve in (thus interfere or distort) the electrochemical response of its supported working electrode in electrolytes and potentials of interest. Si prisms are used more conveniently and frequently than Ge ones in electrochemical ATR-SEIRAS applications. To our best knowledge, the wet process fabrication of Ag nanofilms
10.1021/jp064036a CCC: $33.50 © 2006 American Chemical Society Published on Web 11/23/2006
25722 J. Phys. Chem. B, Vol. 110, No. 51, 2006
Huo et al.
Figure 1. (Left panel) UV-vis absorption spectrum of Ag particles in solution (3 times diluted). (Right panel) Transmission electron microscope image of silver nanoparticles.
on IR transparent Si for electrochemical ATR-SEIRAS application has not been reported. Direct metathesis reaction as suggested in eqs I and II is no longer applicable for Ag metallization on Si. In this report, we will demonstrate a seededgrowth approach to the fabrication of a conductive and SEIRAactive Ag nanofilm on Si. This approach incorporates initially immobilizing discrete Ag nanoparticles on Si as catalytic seeds followed by further growth of Ag nanofilms in a modified silvermirror reaction. Without a seed layer, the silver-mirror reaction cannot produce an adhesive, shiny, and SEIRA-active Ag nanofilm on Si. It is very important that Ag seeds rather than other metals such as Pd and Au ones are originally used for seeded-growth to prevent possible contamination in the resultant Ag nanofilm electrodes.31 A Ag seed layer on Si has been attained either by self-assembly of Ag colloids on aminosilanized Si surfaces or deposition of Ag discrete nanoparticles from reduction of Ag+ in a HF-bearing solution on H-terminated Si surfaces. The first seeding tactic is derived from a Ag or Au colloidal monolayer on ITO or Si reported by Natan’s1,31,32 and Liu’s groups33 for surface enhanced Raman scattering (SERS) substrates, as well as from a Au colloidal monolayer on Si in our group for transmission SEIRA substrate.10 The second seeding tactic was based on the highly reductive nature of Si in the presence of F- (vide infra).30,34 In addition, unlike chemical deposition of a Ag nanofilm electrode on Ge where the growth of a conductive and SEIRA-active Ag nanofilm is at the continuous expense of the Ge substrate, our current approach allows chemical deposition of a Ag nanofilm electrode on Si without or with least dissolution of Si. The as-prepared Ag nanofilm electrodes presented nearly equivalent cyclic voltammetric features as compared to a bulk Ag electrode. With SCN- as the surface probe species, the two types of the abovementioned Ag nanofilm electrodes exhibit at least 1 order of magnitude enhancement in infrared absorption for the SCNadsorbate as compared to bulk Ag electrodes used in conventional infrared reflection-absorption spectroscopy (IRAS). In addition, as a first step of extending the as-deposited Ag electrodes to spectroelectrochemical application, pyridine adsorption configuration at a Ag electrode was re-examined by ATR-SEIRAS for the first time, which demonstrated a good agreement with well-documented SERS investigations. 2. Experimental Section 2.1. Preparation of Colloidal Particles. Citrate-stabilized Ag nanoparticles were prepared according to refs 32 and 35. Briefly, Ag colloid was performed by adding 1 mL of 1% w/v
SCHEME 1: Protocol for Fabricating a Ag Nanofilm Electrode on a Hydroxylized Si Substrate Based on SelfAssembly and Subsequent Seeded-Growth (Type I) and Protocol for Fabricating a Ag Nanofilm Electrode on a H-Terminated Si Substrate Based on Reduction of Ag+ in a HF-Bearing Solution and Subsequent Seeded-Growth (Type II)
aqueous silver nitrate into 100 mL of ultrapure H2O under vigorous stirring accompanied by adding 1 mL of 1% (w/v) aqueous sodium citrate in 1 min. After an additional 1 min, 1 mL of 0.075% (w/v) NaBH4 in 1% sodium citrate was added. UV-vis absorption spectroscopy (vide infra) revealed a strong plasmon absorption band at 389 nm for the Ag colloidal solution (Figure 1, left panel), and TEM indicated that the average diameter of the Ag colloids was ca. 18 ( 7 nm (Figure 1, right panel), in good agreement with literature reports.32,35 2.2. Fabrication of Ag Nanofilm Electrodes on Si. Scheme 1 depicts seeded-growth procedures used for chemical deposition of Ag nanofilm electrodes on Si. Initial immobilization of a Ag seed layer via either self-assembly of Ag colloids (type I) or reduction of Ag+ from a HF-bearing solution (type II) was employed before the reflecting plane of the Si prism was subject to further chemical plating by contacting the bath upside down. A Si hemicylindrical prism (non-doped, PASTEC, Osaka) with a 2 cm × 2.5 cm reflecting plane was cleaned by the RCA method,10,11,25 followed by immersion in a boiling piranha
Fabrication of Silver Nanoparticle Films solution (H2SO4:H2O2 ) 3:1) for 20 min to derive a hydroxyl surface. After a thorough rinse with Milli-Q ultrapure water, the Si reflecting surface was immersed in a solution of (3-aminopropyl)trimethoxysilane (APTMS, Aldrich) (5% aqueous solution) for 2 h. Rinsed thoroughly again with Milli-Q water, the aminosilanized Si surface was subsequently immersed in a Ag colloidal solution for 3 h to complete the initial seeding. Alternatively, the reflecting plane of the RCA-cleaned Si prism was subject to H-termination by contacting it with a NH4F buffer solution (1 vol of 40% HF and 10 vol of 40% NH4F) for 2 min. After being thoroughly rinsed with Milli-Q water, the reflecting plane was treated with a solution of 0.005 mol·L-1 AgNO3 + 0.06 mol·L-1 HF for 10 s to accomplish the initial seeding.34 A bright thin silver film is formed by immersing the above Ag seeded Si surface in a Ag-plating solution via a modified silver-mirror reaction, that is
CH2OH(CHOH)4CHO + 2Ag(NH3)2OH f CH2OH(CHOH)4COONH4 + 2AgV + 3NH3 + H2O (III) The Ag plating solution was made by mixing solutions in 1:1 (v/v) ratio from baths A and B (vide infra). The optimal plating time was around 60 s. For simplicity, the Ag nanofilm electrode obtained based on the self-assembly tactics was denoted as the Ag (type I) electrode, whereas that based on the reduction of Ag+ from a HF-bearing solution was denoted as the Ag (type II) electrode. For bath A, an appropriate volume of ammonia solution was added dropwise with a micropipet to a 20 mL solution containing 0.058 g of AgNO3 until the precipitate formed was just dissolved. Then 0.15 g of solid NaOH was added into the solution. Again, ammonia solution was added dropwise under sonication until a clear solution was obtained. In preparing bath B, 0.045 g of glucose and 0.04 g of tartaric acid were first dissolved in 12 mL of water, the solution was then heated to boiling for 10 min. After it was cooled, 8 mL of water and 1 mL of ethanol were added.36 2.3. TEM, ICP, and AFM. Transmission electron microscopy (TEM) images were taken with a JEOL JEM 2011 electron microscope operating at 200 kV and the Ag colloidal sample was supported on a copper grid coated with a holey carbon film for the measurement. Ultraviolet-visible absorption spectra were recorded with an Agilent 8453 spectrophotometer. Inductively coupled plasma (ICP) atomic emission spectroscopy (Varian) was used to determine the concentration of dissolved Ag+ ions from the films on Si in nitric acid, and the amount of metals was used to estimate the average thicknesses of the deposits by assuming the same densities as their bulk materials. The thickness of Ag (type I) and Ag (type II) electrodes is about 40 and 50 nm, respectively. Atomic force microscopy (AFM) images were acquired with tapping mode under ambient conditions with a Pico-SPM (Molecular Imaging, Tempe, AZ). Si cantilevers having a spring constant between 1.2 and 5.5 N m-1 were used at resonance frequencies between 60 and 90 kHz. 2.4. Electrochemistry and ATR-SEIRAS. Details of electrochemical ATR-SEIRAS have been described elsewhere.15-22 Briefly, a Magna-IR 760 spectrometer (Nicolet) was used to perform all ATR-SEIRAS measurements at a spectral resolution of 4 cm-1. Spectra of the nanofilm electrode/electrolyte interface were acquired with so-called Kretschmann attenuated-totalreflection (ATR) configuration (prism/thin metal film/solution geometry). Unpolarized infrared radiation hit on the interface at an incident angle of 70°, and the reflected beam was collected
J. Phys. Chem. B, Vol. 110, No. 51, 2006 25723 with a liquid nitrogen cooled MCT detector. A total of 256 scans were coadded for a single-beam spectrum at each potential. The spectra are shown in the absorbance unit defined as A ) -log(R/R0), were R and R0 are the single-beam spectra measured at the sample and reference potentials, respectively. The reference spectrum was taken in a potential where the interested species are desorbed. The Ag nanofilm electrodes were freshly prepared, thoroughly rinsed by ultrapure water, and reduced at a hydrogen-evolution potential to ensure a sufficiently clean surface before being subjected to (spectro)electrochemical measurements. XPS spectra also confirm that the as-deposited films consist of metallic Ag islands, containing no other detectable metals (see Figure S1 in the Supporting Information). A CHI 660B electrochemistry workstation (CH Instruments, Shanghai) was employed for a potential/current control and to record the cyclic voltammograms. All electrode potentials are cited with a saturated calomel electrode (SCE). The differential capacitance curves of the Ag nanofilm electrode in neat 0.1 M NaF and 0.1 M NaF containing 10 mM pyridine were obtained with ac voltammetry in which a 100 Hz sinusoidal signal with an amplitude of 5 mV was superposed to a linearly scanned dc potential at 10 mV s-1 starting from -1.4 to 0 V; the ac current component was extracted and analyzed with a lock-in amplifier built in a CHI 660 B electrochemistry workstation. 3. Results and Discussion 3.1. AFM Characterization. Ag colloidal solutions at pH 6 are dominantly negatively charged due to the partial capping of citrate anions, while the amino groups of self-assembled aminosilane on Si are mainly positively charged. The positively charged functional group -NH2 interacts strongly with negatively charged Ag colloidal particles, acting as an excellent organic coupler between Ag nanoparticle monolayers and hydroxylized Si surfaces. Shown in Figure 2a is the ex situ AFM image of the Ag (type I) seed layer. Sparsely dispersed 2D Ag colloids are identified as a result of significantly reduced time for the entire self-assembly process,10 yet they are sufficient to act as catalyzed seeds for further chemical plating. The heights of the feature are in reasonable agreement with the particle sizes measured by TEM results, ca. 18 ( 7 nm. Shown in Figure 2b is the ex situ AFM image of the Ag (type II) seed layer. The presumably ellipsoidal Ag nanoparticles are homogeneously distributed, with an average major axis of ca. 60 nm (lateral x-y) and a minor axis of ca. 25 nm (longitudinal z). Nevertheless, these nanoparticles are not interconnected in nature as reflected by its electrical nonconductivity and barely visible appearance. After subsequent chemical plating, both types of Ag nanofilms are silver-shiny in appearance. Surface morphology of the Ag nanofilms may affect their electrical, mechanical, and optical properties. Parts c and d of Figure 2 are the ex situ AFM images for Ag nanofilms formed as the result of two initial seeding processes followed by chemical plating for 60 s in the modified silver-mirror reaction bath, respectively. Their surface morphologies are somewhat different in terms of particle size, proximity, and aggregation. The Ag (type I) nanofilm presents a larger and yet slightly rougher morphology as compared to the other. The morphology difference between the two types of Ag nanofilms arises from different initial seed sizes and growth rates. The Ag seeds of type I demonstrate faster growth than those of type II. This may be reasonably explained by the Gibbs-Thomson equation, which suggests a higher surface free energy involved for smaller nuclei.37 Nevertheless, the island
25724 J. Phys. Chem. B, Vol. 110, No. 51, 2006
Huo et al.
Figure 2. (a) Ex situ AFM image of 2D 18 nm diameter Ag colloidal particles assembled on an aminosilanized Si substrate. (b) Ex situ AFM image of the 2D Ag seed layer by immersing the H-terminated Si surface in a solution of 0.005 mol·L-1 AgNO3 + 0.06 mol·L-1 HF for 10 s. (c) Ex situ AFM image of the chemically plated Ag nanofilm based on the seed layer as described in part a. (d) Ex situ AFM image of the chemically plated Ag nanofilm based on the seed layer as described in part b. (e) Histograms showing the particle size distribution of the Ag seeds (type I, see lower panel) as well as after-plated Ag nanofilm (see upper panel). (f) Histograms showing the particle size distribution of the Ag seeds (type II, see lower panel) as well as after-plated Ag nanofilm (see upper panel).
structure of both Ag nanofilms facilitates the interaction of IR radiation with the metal and adsorbed molecules, resulting in the strongly enhanced absorption for the allowed vibrations of the adsorbate.4,9 Shown in Figure 2e is the histogram on size distribution of the Ag seeds (type I, see lower panel) as well as the after-plated Ag nanofilm (see upper panel). The mean diameter of Ag seeds (type I) was ca.15 nm with a size distribution ranging chiefly from 10 to 30 nm, while the after-plated Ag nanofilm had a mean diameter of 45 nm with a size distribution ranging from
20 to 100 nm. Similarly, the mean Ag particle sizes increased from 29 (type II, the lower panel of Figure 2f) to 37 nm (the upper panel of Figure 2f) before and after plating. Electrochemical annealing of a Ag nanofilm electrode, that is, repetitively cycling it in between positive and negative limits sufficient for oxidation and reduction, may yield some morphological changes, the extent of which depends on the electrolytes and potential ranges used. One example of in situ AFM imaging of a Ag nanofilm subject to a mild electrochemical annealing can be found in Figure S4 in the Supporting
Fabrication of Silver Nanoparticle Films
J. Phys. Chem. B, Vol. 110, No. 51, 2006 25725
Figure 3. Cyclic voltammograms for as-deposited Ag nanofilm electrodes in 0.1 M NaOH at 50 mV s-1 (solid and dotted curves correspond to Ag (type I) and Ag (type II) nanofilm electrodes, respectively).
Information. Detailed probing in this direction will constitute the topic in further studies. 3.2. Film Adhesion. A metal film vacuum deposited on Si does not usually possess good adhesion. From time to time it may be peeled off the Si substrate in the process of cell assembly and regular cyclic voltammetry.23 Ag nanofilms fabricated with our current approach sustain a stronger stability under most mild electrochemical conditions. The tape test was used to evaluate the adherence of the Ag nanofilms to the Si substrate.37 A piece of scotch tape was thumb-pressed against the two kinds of Ag nanofilms and pulled away at a constant ratesboth the asdeposited Ag nanofilms were virtually intact. The improvement in adhesion and stability for Ag nanofilms (type I) can be ascribed to the presence of the organic glue, i.e., an aminosilanized interlayer that binds strongly to negatively charged colloidal nanoparticles with its positively charged amino ends.35 For Ag nanofilms (type II), the initial Ag seeding involves a reduction of Ag+ in a F--containing solution,30,34 i.e.,
Si + 6F- f SiF62- + 4e Ag+ + e f Ag
E0 ) -1.20 V E0 ) 0.799 V
(IV) (V)
The cathodic and anodic half-cell reactions took place at the Si surface spontaneously.30 As a result, the Ag seeds were embedded partly into the Si substrate (with a possibility of forming an interfacial Ag-Si alloy layer) to serve catalyzed nuclei for further chemical plating in a modified silver-mirror reaction. As a result, the adhesion of the final Ag nanofilm to Si was enhanced. 3.3. Voltammetry of Ag Nanofilm Electrodes. To proceed with in situ ATR-SEIRAS measurement, it is essential to ensure that Ag nanofilm electrodes prepared with our current approach have comparable electrochemical features as their bulk counterparts. Typical cyclic voltammograms for the Ag nanofilm electrodes in 0.1 M NaOH at 50 mV s-1 are shown in Figure 3, which is nearly identical with that for polycrystalline bulk Ag electrodes.38 Detailed assignment of the above voltammetric peaks can be found in the Supporting Information. Further comparison of cyclic volmmograms of as-deposited Ag nanofilm electrodes with that of a corresponding Ag bulk electrode in commonly used alkaline, acidic, and neutral electrolytes is also given in the Supporting Information (Figures S2 and S3). All results points to a close nature between them. The surface roughness factor of a Ag nanofilm can be obtained from the measurement of the charge density involved in the underpotential deposition (UPD) of Pb. Cyclic voltam-
Figure 4. Cyclic voltammograms for as-deposited Ag nanofilm electrodes in 0.5 M CH3COONa + 0.01 M CH3COOH + 0.1 M Pb(CH3COO)2 solutions at 5 mV s-1 (solid and dotted curves correspond to Ag (type I) and Ag (type II) nanofilm electrodes, respectively).
Figure 5. (Left panel) Potential dependent ATR-SEIRA spectra for adsorbed SCN- on a Ag nanofilm (type I) electrode in 5 mM KSCN + 0.1 M KClO4. (Right panel) Selected ATR-SEIRA spectra for SCNadsorbed on two Ag nanofilm electrodes (Curve a: type I, and Curve b: type II) at -0.1 V in the 5 mM KSCN + 0.1 M KClO4. All spectra were obtained with respect to reference spectra taken at -1.0 V.
mograms for two types of Ag nanofilm electrodes in 0.5 M CH3COONa + 0.01 M CH3COOH + 0.1 M Pb(CH3COO)2 were recorded at a sweep rate of 5 mV s-1 as shown in Figure 4. The total integrated charge density of the UPD deposition of Pb is 604 and 587 µC cm-2 respectively for Ag (type I) and Ag (type II) nanofilms electrodes. Jovic´evic´ et al. reported a charge density of 326 µC cm-2 for a smooth ideal Ag (111) electrode under similar conditions.39 Assuming that this value can be applied to a smooth polycrystalline Ag electrode, the ratio of the charge density values gives a roughness factor of ca. 1.9 and 1.8 for Ag (type I) and Ag (type II) nanofilm electrodes, respectively. 3.4. Evaluation of Electrochemical ATR-SEIRA. Surfaceenhanced IR absorption of the as-deposited Ag nanofilm electrodes can be examined by using SCN- as the surface probe. Figure 5 (left panel) shows the potential dependent ATR-SEIRA spectra for SCN- adsorbed on a Ag nanofilm electrode (only Ag (type I) is shown for clarity) in 5 mM KSCN +0.1 M KClO4 solution. A single beam spectrum collected at -1.0 V was taken as the reference spectrum. Similar potential dependent spectra can be obtained for a Ag nanofilm (type II) electrode. The CN stretch νCN band of adsorbed SCN- is located from 2083 to 2116 cm-1, suggesting that SCN- adsorbed via S atom on both Ag electrode surfaces.40,41 The somewhat extended peak tail (or a small shoulder at 0 V) at the lower frequency side may be due to the irregular adsorption sites.12 The positive shift of the
25726 J. Phys. Chem. B, Vol. 110, No. 51, 2006
Huo et al.
electrode potential results in an increase of the band intensity and a shift of the band position to higher frequencies. The former can be ascribed to an increased surface coverage of SCN- on Ag with potential moving positively from initially -1.0 V, i.e., around the potential of zero-charge for a polycrystalline Ag electrode.40 The increase on the donation of the σ electron from the S atom to the surface with increasing potential together with lateral adsorbate-adsorbate interaction may result in an increase in the CN stretching force constant,41 yielding a Stark tuning rate of ca. 39 cm-1 V-1. Compared in Figure 5 (right panel) are two selected ATR-SEIRA spectra for SCN- adsorbed on Ag (type I) and Ag (type II) nanofilm electrodes collected independently at -0.1 V in 0.1 M KClO4 + 5 mM KSCN. The band intensity observed for the Ag nanofilm (type I) electrode is 0.0069 in absorbance unit (Abs.) or 0.016 in relative reflectance unit (∆R/R), and 0.005 in Abs. or 0.0115 in ∆R/R for the Ag (type II) nanofilm electrode. By contrast, the band intensity of 2 × 10-4 in ∆R/R was obtained for an electropolished bulk Ag electrode in IRAS at the same potential in the same electrolyte.42 Precise calculation of the surface enhancement factor for chemically deposited Ag nanofilms on Si could be quite difficult.24 Nevertheless, if the enhancement factor is herein defined as the relative ratio of the band intensity obtained by ATR-SEIRAS and that by conventional IRAS, with appropriate calibration of the effects of the surface roughness factor, surface coverage, incident angles, and polarization states of IR radiation,11 we have
G)
IA CCCC IE 1 2 3 4
(VI)
where G is the enhancement factor, IA and IE are the band intensity of CN stretch of adsorbed SCN- species obtained in ATR-SEIRAS and IRAS, respectively, and C1, C2, C3, and C4 are calibration factors for surface roughness, incidence angle of IR irradiation, surface coverage, and polarization states, respectively. C1 is the ratio of surface roughness factor of a bulk Ag electrode and that of a Ag nanofilm electrode, assuming a reasonable roughness factor of 1.3 for an electropolished Ag electrode,43 thus C1 is about 0.69 and 0.73 for the Ag nanofilm (type I) and Ag nanofilm (type II) electrodes, respectively. C2 is calibrated approximately to be 2/3 for an incident angle of 65° on a flat CaF2 window in IRAS and for that of 70° on a Ag nanofilm in ATR-SEIRAS.44 If a 60°-beveled CaF2 window was used in IRAS, then C2 should be somewhat larger. Given that the same potential and the same electrolyte were used, C3 is approximately equal to 1. C4 is taken at least to be 2 for a p-polarized IR radiation used in IRAS and a unpolarized one used in ATR-SEIRAS.11After calibration, G was evaluated to be at least 74 and 56, respectively. For a detailed description of the above evaluation, please refer to the Supporting Information of our previous publication.11 A slightly higher band intensity, i.e., 0.008 in Abs., was reported in a previous ATR-SEIRAS measurement using p-polarized IR radiation on a vacuum-evaporated Ag nanofilm electrode on Si in an electrolyte containing 10 mM SCN-.41 Nevertheless, after calibration of the polarization effect, the current band intensity should be increased at least by a factor of 2.11 In addition, in consideration of smaller surface roughness factors for current Ag nanofilms (for an evaporated Ag nanofilm, its surface roughness factor was estimated to be ca. 34) and a lower concentration of solution SCN- (5 mM) used for our ATR-SEIRAS measurement, it can be concluded that the Ag
Figure 6. Differential capacitance curves for the Ag nanofilm electrode (type I) in 0.1 M NaF (solid line) and 10 mM pyridine + 0.1 M NaF (dashed line) solutions obtained with ac volammetry, along with the potential-dependent integrated intensities of the ν8a (1594 cm-1).
nanofilms fabricated by our wet process exhibit an even stronger SEIRA effect than those prepared by vacuum evaporation. More importantly, Ag nanofilm electrodes on Si prepared by chemical deposition based on seeded-growth merit a higher reproducibility in SEIRA effects with a cost-effective fabrication, suggestive of an attractive prospect for its application in surface electrochemistry and surface analytical chemistry. 3.5. Preliminary Application of the As-Deposited Ag Electrodes. The significant SEIRAS activity obtained facilitates probing the surface electrosorption on the as-prepared Ag electrodes. As an initial move of this application, we reexamined the adsorption configuration of a prototype molecule, i.e., pyridine on a Ag electrode. Although this system has been extensively investigated previously by electrochemical SERS,3,45-49 no relevant in situ electrochemical FTIR reports can be found on a Ag electrode in the literature. Electrochemical SERS is a powerful tool in surface and interfacial analysis, capable of directly detecting useful vibrational bands in a much broader wavenumber region (especially the lower frequencies) without subtracting a reference spectrum, thus providing a wealth of interfacial information.3,46,50,51 Yet, it requires the so-called ORC pretreatment of a Ag electrode in a chloride electrolyte, which may initiate complex and unstable surface conditions. As a result, ORC-roughened Ag electrodes may deviate significantly from their bulk counterparts in terms of electrochemical behaviors. The SERS signal obtained for a given Ag electrode is often irreversible after experiencing a very negative potential excursion.52 Furthermore, the complicated SERS mechanisms (EM and CT) make it more difficult to explain the spectra observed.3,46,50,51 By contrast, the Ag nanofilm electrode used in current ATR-SEIRAS was not subjected to (because it was not required) such an ORC pretreatment in a chloride solution (for the effect of such ORC pretreatment on SEIRA, please refer to Figure S5 in the Supporting Information), thus its electrochemical response is close to that of a Ag bulk electrode. The chemical enhancement in SERS due to the photon-driven charge transfer between the adsorbate and the metal can be neglected in SEIRAS because of the much lower IR radiation used. In addition, the SEIRAS signal is rather reversible after a negative potential excursion. The surface selection rule for SEIRAS is very straightforward, i.e., only vibrations causing dipole moment change normal to the local surface are SEIRA active.4,7 Included in Figure 6 are differential capacitance curves for the Ag nanofilm electrode (type I) recorded in the solutions of 0.1 M NaF (solid line) and 10 mM pyridine + 0.1 M NaF (dashed line) obtained with ac voltammetry. The curves obtained
Fabrication of Silver Nanoparticle Films
J. Phys. Chem. B, Vol. 110, No. 51, 2006 25727 ATR-SEIRAS on Ag. On the basis of the surface selection rule of SEIRAS, the adsorbed pyridine molecules are bound to the Ag electrode with its ring plane perpendicular or slightly tilted to the local surface without rotating its C2 axis about the surface normal. Such a configuration was also found in the case of pyridine adsorption on a Cd electrode62 and a Au electrode at potentials positive of pzc.63-65 The flat-on, edge-tilted, and R-pyridyl adsorption configurations which were previously reported for pyridine on other metals can be excluded.11,63,66 This conclusion is in accordance with that drawn by SERS3,45-49,67 and the DFT calculation.68,69 4. Conclusion
Figure 7. SEIRA spectra of the Ag nanofilm electrode (type I) from -1.0 to -0.1 V in a 0.1 M NaF + 10 mM pyridine solution (each spectrum corresponds to 256 interferograms and is referred to the singlebeam spectrum collected at -1.2 V.
were in accordance with the polycrystalline bulk electrode in the same solutions as reported in the literature.53-55 In the presence of pyridine, the capacitance in the vicinity of the point of zero charge (pzc, ca. at -0.8 V) decreased significantly, and the two capacitance humps appeared at around -1.05 and -0.40 V. The former one is typical of the adsorption/desorption of pyridine.53 The pyridine molecules are virtually desorbed from the Ag electrode surface at a potential less positive than -1.2 V, as judged by the capacitance overlapping in two curves. The latter one could be attributed to the reconstruction of the surface layer.53 The continuous increase of (pseudo)capacitance at higher potentials may be due to partial oxidation of Ag electrodes or the formation of surface hydroxides. Shown in Figure 7 are a series of SEIRA spectra for the pyridine on the as-deposited Ag nanofilm electrode as a function of applied potential from -1.0 up to -0.1 V. All the spectra are referred to the single-beam spectrum collected at -1.2 V. The spectra shown in Figure 7 show the bands at 1594 (vs),56 1481 (m), 1442 (vw), 1069 (s), 1033 (w), and 1004 (s) cm-1 corresponding to ν8a (A1), ν19a (A1), ν19b (B1), ν18a (A1), ν12 (A1), and ν1 (A1), respectively, of adsorbed pyridine.57 Here, the A1 and B1 symmetries are defined to represent in-plane vibrations yielding dipole changes along and perpendicular to the C2 axis of pyridine, respectively. The potential dependent integrated intensity of the ν8a band was plotted also in Figure 6. The intensity of this band undergoes “the course of a volcano”, that is, starts to increase from the potential of ca. -1.0 to -0.65 V and then decreases to the bottom, which suggests that maximum coverage of pyridine on Ag occurs around pzc. All other vibration modes with A1 symmetry tend to follow the same pattern mentioned above. The broadband at 1442 cm-1 at relatively positive potentials can be divided into two bands by the fitting method (see Figure S6 in the Supporting Information): the one located at 1442 cm-1 can be ascribed to ν19b (B1), the other at 1429 cm-1 probably can be assigned to the adsorbed CO32-, an impurity species commonly present in neutral and alkaline electrolytes.58,59 It is noteworthy that spectral features for the adsorbed and the bulk pyridine are different in terms of relative band intensity and position.57 The totally symmetric ring breathing mode (ν1) blue-shifts from 992 to 1004 cm-1, suggestive of the adsorption via N end.46,60,61 The ATR-SEIRA spectra are dominated by vibrational modes having A1 symmetry, while those having B1 symmetry are hardly observed. For example, the most intense IR band for liquid pyridine,57 i.e., 1442 cm-1 (ν19b), which is assigned to the B1 mode, is nearly invisible in the corresponding
In summary, we have presented an inexpensive and reproducible wet process, that is, seeded-growth approach to fabricate Ag nanofilms on Si for electrochemical ATR-SEIRAS application. Two alternative procedures were applied to the formation of an initial Ag seed layer, either via self-assembling a submonolayer of Ag colloids on an aminosilanized Si surface or embedding discrete Ag nanoparticles on a hydrogenterminated Si surface from a dilute AgNO3 and HF solution. Conductive, adhesive, and SEIRA-active Ag nanofilms were acquired with subsequent chemical plating in a modified Agmirror reaction. With SCN- as a surface probe, the Ag nanofilm electrodes were evaluated to yield enhancement factors of ca. 50-80 as compared to an electropolished Ag electrode in conventional IRAS with reasonable calibrations. ATR-SEIRA spectra were also extended to characterize the adsorption configuration of a pyridine adlayer at the Ag electrodes. Pyridine was found to bind via its N end to the Ag electrode with its ring plane perpendicular or slightly tilted to the local surface without rotating its C2 axis about the surface normal. This seeded-growth approach can be used to replace the traditional vacuum evaporation or sputtering for preparing Ag and other SEIRA-active metal nanofilm electrodes on an ATR Si window. Such an investigation is in progress and will be reported in due course. Acknowledgment. The ATR Si prism was a gift from Prof. M. Osawa in the Catalysis Research Center, Hokkaido University. The NSFC (No. 20473025), the SRFDP (No.20040246008), the NCET (No. 04-0349), and the SNPC (No.0452nm064) are gratefully acknowledged for financial support. We also thank State Key Laboratory for Physical Chemistry of Solid Surfaces in Xiamen University for partial financial support through the Coordinator, Prof. Z.Q. Tian. Supporting Information Available: Illustraions showing XPS survey spectra for As-deposited Ag nanofilms, electrochemical characterization of As-deposited Ag nanofilm electrodes, electrochemical annealing effect on Ag nanostructure, potential cycling on SEIRA stability, and peak deconvolution in SEIRA spectra selected from Figure 7. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Freeman, R. G.; Grabar, K. C.; Allison, K. J.; Bright, R. M.; Davis, J. A.; Guthrie, A. P.; Hommer, M. B.; Jackson, M. A.; Smith, P. C.; Walter, D. G.; Natan, M. J. Science 1995, 267, 1629. (2) Lu, L. H.; Randjelovic, I.; Capek, R.; Gaponik, N.; Yang, J. H.; Zhang, H. J.; Eychmuller, A. Chem. Mater. 2005, 17, 5731. (3) Tian, Z. Q.; Ren, B.; Wu, D. Y. J. Phys. Chem. B 2002, 106, 9463. (4) Osawa, M. In Handbook of Vibrational Spectroscopy; Chalmers J. M., Griffiths, P. R., Eds.; John Wiley & Sons: Chichester, U.K., 2002; Vol. 1, pp 785-799.
25728 J. Phys. Chem. B, Vol. 110, No. 51, 2006 (5) Maxwell, D. J.; Emory, S. R.; Nie, S. M. Chem. Mater. 2001, 13, 1082. (6) Roy, D.; Fendler, J. AdV. Mater. 2004, 16, 479. (7) Aroca, R. F.; Alvarez-Puebla, R. A.; Pieczonka, N.; Sanchez-Cortez, S.; Garcia-Ramos, J. V. AdV. Colloid Interface Sci. 2005, 116, 45. (8) Hartstein, A.; Kirtley, J. R.; Tsang, J. C. Phys. ReV. Lett. 1980, 45, 201. (9) Aroca, R. F.; Ross, D. J.; Domingo, C. Appl. Spectrosc. 2004, 58, 324A. (10) Huo, S. J.; Li, Q. X.; Yan, Y. G.; Chen, Y.; Cai, W. B.; Xu, Q. J.; Osawa, M. J. Phys. Chem. B 2005, 109, 15985. (11) Huo, S. J.; Xue, X. K.; Yan, Y. G.; Li, Q. X.; Ma, M.; Cai, W. B.; Xu, Q. J.; Osawa, M. J. Phys. Chem. B 2006, 110, 4162. (12) Bjerke, A. E.; Griffiths, P. R.; Theiss, W. Anal. Chem. 1999, 71, 1967. (13) Bjerke, A. E.; Griffiths, P. R. Appl. Spectrosc. 2002, 56, 1275. (14) Jensen, T. R.; Van, Duyne, R. P.; Johnson, S. A.; Maroni, V. A. Appl. Spectrosc. 2000, 54, 371. (15) Yajima, T.; Uchida, H.; Watanabe, M. J. Phys. Chem. B 2004, 108, 2654. (16) Luo, L. Q.; Futamata, M. Elctrochem. Commun. 2006, 8, 231. (17) Shao, M. H.; Adzic, R. R. Electrochim. Acta 2005, 50, 2415. (18) Shao, M. H.; Adzic, R. R. J. Phys. Chem. B 2005, 109, 16563. (19) Futamata, M.; Luo, L. Q.; Nishihara, C. Surf. Sci. 2005, 590, 196. (20) Shiroishi, H.; Ayato, Y.; Kunimatsu, K.; Okada, T. J. Electroanal. Chem. 2005, 581, 132. (21) Pronkin, S.; Wandlowski, T. Surf. Sci. 2004, 573, 109. (22) Ataka, K.; Yotsuyanagi, T.; Osawa, M. J. Phys. Chem. 1996, 100, 10664. (23) Miyake, H.; Ye, S.; Osawa, M. Elctrochem. Commun. 2002, 4, 973. (24) Miki, A.; Ye, S.; Osawa, M. Chem. Commun. 2002, 1500. (25) Yan, Y. G.; Li, Q. X.; Huo, S. J.; Ma, M.; Cai, W. B.; Osawa, M. J. Phys. Chem. B 2005, 109, 7900. (26) Rodes, A.; Orts, J. M.; Perez, J. M.; Feliu, J. M.; Aldaz, A. Elctrochem. Commun. 2003, 5, 56. (27) Delgado, J. M.; Orts, J. M.; Rodes, A. Langmuir 2005, 21, 8809. (28) Miyake, H.; Osawa, M. Chem. Lett. 2004, 33, 278. (29) Diao, Y. X.; Han, M. J.; Wan, L. J.; Itaya, K.; Uchida, T.; Miyake, H.; Yamakata, A.; Osawa, M. Langmuir 2006, 22, 3640. (30) Magagnin, L.; Maboudian, R.; Carraro, C. J. Phys. Chem. B 2002, 106, 401. (31) Bright, R. M.; Walter, D. G.; Musick, M. D.; Jackson, M. A.; Allison, K. J.; Natan, M. J. Langmuir 1996, 12, 810. (32) Brown, K. R.; Walter, D. G.; Natan, M. J. Chem. Mater. 2000, 12, 306. (33) Zhu, Z. H.; Zhu, T.; Liu, Z. F. Nanotechnology 2004, 15, 357. (34) Tong, H.; Zhu, L.; Li, M. N.; Wang, C. M. Electrochim. Acta 2003, 48, 2473. (35) Frederix, F.; Friedt, J. M.; Choi, K. H.; Laureyn, W.; Campitelli, A.; Mondelaers, D.; Maes, G.; Borghs, G. Anal. Chem. 2003, 75, 6894. (36) Liu, Z. C.; He, Q. G.; Xiao, P. F.; Liang, B.; He, N. Y.; Lu, Z. H. Acta Chim. Sin. 2002, 60, 627. (37) Jin, Y. D.; Kang, X. F.; Song, Y. H.; Zhang, B. L.; Cheng, G. J.; Dong, S. J. Anal. Chem. 2001, 73, 2843. (38) Abd El Rehim, S. S.; Hassan, H. H.; Ibrahim, M. A. M.; Amin, M. A. Monatsh. Chem. 1998, 129, 1103.
Huo et al. (39) Jovic´evic´, J. N.; Jovic´, V. D.; Despic´, A. R. Electrochim. Acta 1984, 29, 1625. (40) Weaver, M. J.; Barz, F.; Gordon, J. G., II; Philpott, M. R. Surf. Sci. 1983, 125, 409. (41) Hatta, A.; Sasaki, Y.; Sue¨taka, W. J. Electroanal. Chem. 1986, 215, 93. (42) Corrigan, D. S.; Gao, P.; Leung, L. H.; Weaver, M. J. Langmuir 1986, 2, 744. (43) Larkin, D.; Guyer, K. L.; Hupp, J. T.; Weaver, M. J. J. Electroanal. Chem. 1982, 138, 401. (44) Suzuki, Y.; Osawa, M.; Hatta, A.; Sue¨taka, W. Appl. Surf. Sci. 1988, 33/34, 875. (45) Fleischmann, M.; Hendra, P. J.; Mcquilan, A. J. Chem. Phys. Lett. 1974, 26, 163. (46) Otto, A.; Mrozek, I.; Brabhorn, H.; Akemann, W. J. Phys.: Condens. Matter. 1992, 4, 1143. (47) Jeanmaire, D. L.; Van, Duyne, R. P. J. Electroanal. Chem. 1977, 84, 1. (48) Pettinger, B. In Adsorption of Molecules at Metal Electrodes; Lipkowski, J., Ross, P. N., Eds.; VCH: New York, 1992; p 285. (49) Pemberton, J. E. In Electrochemical Interfaces: Modern Techniques for In-situ Interface Characterization of Molecules at Metal Electrodes; Abrunˇa, H. D., Ed.; VCH: New York, 1991; pp 193. (50) Moskovits, M. ReV. Mod. Phys. 1985, 57, 783. (51) Campion, A.; Kambhampati, P. Chem. Soc. ReV. 1998, 27, 241. (52) Dick, L. A.; McFarland, A. D.; Haynes, C. L.; Van Duyne, R. P. J. Phys. Chem. B 2002, 106, 853. (53) Fleischmann, M.; Robinson, J.; Waser, R. J. Electroanal. Chem. 1981, 117, 257. (54) Valette, G.; Hamelin, A. J. Electroanal. Chem. 1973, 45, 301. (55) Valette, G. J. Electroanal. Chem. 1981, 122, 285. (56) Relative intensity: vs, very strong; s, strong; m, medium; w, weak; vw, very weak. (57) Corrsin, L.; Fax, B. J.; Lord, R. C. J. Chem. Phys. 1953, 21, 1170. (58) Arihara, K.; Kitamura, F.; Ohsaka, T.; Tokuda, K. J. Electroanal. Chem. 2001, 510, 128. (59) Iwasita, T.; Rodes, A.; Pastor, E. J. J. Electroanal. Chem. 1995, 383, 181. (60) Surface Enhanced Raman Scattering; Chang, R. K., Furtak, T. E., Eds.; Plenum: New York, 1982. (61) Metiu, H. Prog. Surf. Sci. 1984, 17, 153. (62) Li, Q. X.; Yan, Y. G.; Xu, Q. J; Cai, W. B. Chem. J. Chin. UniV. In press (in Chinese). (63) Cai, W. B.; Wan, L. J.; Noda, H.; Hibino, Y.; Ataka, K.; Osawa, M. Langmiur 1998, 14, 6992. (64) Lipkowski, J.; Stolberg, L. In Adsorption of Molecules at Metal Electrodes; Lipkowski, J., Ross, P. N., Eds.; VCH: New York, 1992; p 171. (65) Li, N.; Zamlynny, V.; Lipkowski, J.; Henglein, F.; Pettinger, B. J. Electroanal. Chem. 2002, 43, 524. (66) Yan, Y. G.; Xu, Q. J.; Cai, W. B. Acta Chim. Sin. 2006, 64, 458. (67) Zuo, C.; Jagodzinski, P. W. J Phys. Chem. B 2005, 109, 1788. (68) Wu, D. Y.; Duan, S.; Ren, B.; Tian, Z. Q. J. Raman Spectrosc. 2005, 36, 533. (69) Alberto, V.; Ronald, L. B.; Richard, F.; John, R. L. J Phys. Chem. B 2003, 107, 5547.