Microcantilevers Modified with Ferrocene-Terminated Self-Assembled

Publication Date (Web): December 30, 2010 ... The hydrophobic anions PF6− and ClO4−, which form 1:1 contact ion pairs with ... Abstract | Full Tex...
25 downloads 0 Views 2MB Size
ARTICLE pubs.acs.org/JPCC

Microcantilevers Modified with Ferrocene-Terminated Self-Assembled Monolayers: Effect of Molecular Structure and Electrolyte Anion on the Redox-Induced Surface Stress Lana L. Norman and Antonella Badia* Department of Chemistry, Universite de Montreal, FQRNT Center for Self-Assembled Chemical Structures, and Regroupement quebecois sur les materiaux de pointe, C.P. 6128 succursale Centre-ville, Montreal, QC H3C 3J7 Canada

bS Supporting Information ABSTRACT: The redox-activated deflection of microcantilevers has attracted interest for chemical sensing and nanoactuation. However, the development and optimization of this type of microcantilever transduction requires a better understanding of the effect of the particular system parameters on the surface stress changes that cause the measured bending response. We investigate here the effects of the adsorbate structure and electrolyte anion on the surface stress generated by the electrochemical oxidation of ferrocene-terminated self-assembled monolayers (SAMs) chemisorbed to the surface of gold-coated microcantilevers. Ferrocenylalkanethiolate-modified cantilevers are an interesting system for study as the electroactive monolayer can induce charge-normalized surface stress changes that are at least 10-fold greater than those generated by multilayers of the conducting polymers commonly used for electroactuation. The resonance angle shifts measured by surface plasmon resonance spectroscopy in the presence of ClO4- suggest that the extent of the oxidationinduced SAM reorganization is the same for short (n = 6) and long (n = 12) chain ferrocenylalkanethiolate (FcCnSAu) SAMs and for a long chain carbonyl derivative (Fc(CO)C11SAu). The magnitude of the measured cantilever deflection is however not the same for the different SAMs, reflecting differences in the tensile contributions to the overall surface stress of the SAM elasticity and wettability versus the compressive lateral pressure generated by the collective molecular reorientations induced by the pairing of anions to the surface-confined, oxidized ferrocenium cations. The hydrophobic anions PF6- and ClO4-, which form 1:1 contact ion pairs with the SAM-bound ferrocenium cations, give reversible cantilever deflections and the largest compressive surface stress changes. By contrast, oxidation of the FcC11SAu SAM in NO3- and F-, which exhibit the poorest ion-pairing abilities of the anions investigated, results in an irreversible deformation of the cantilever bending and smaller stress changes. The surface phenomena that give rise to the observed differences in the surface stress as a function of the ion pairing ability are discussed.

’ INTRODUCTION The microcantilevers used in scanning force microscopy are the most simplified microelectromechanical-based devices available for label-free analyte sensing in gas or liquid environment.1 The most common operating mode for (bio)chemical sensing experiments is the so-called “surface stress” or “static” mode, where a vertical bending or deflection of the cantilever arises from surface stress changes caused by molecular interactions or transformations occurring on only one of the cantilever faces. Despite the promise of potential applications in medical diagnostics, environmental monitoring, and chemical detection, microcantilever transducers have yet to compete with established detection methods such as surface plasmon resonance (SPR) spectroscopy, quartz crystal microbalance (QCM), and electrochemistry even though commercial platforms exist (e.g., Concentrís GmbH and Cantion A/S), the limits of detection are comparable or better than those of these other techniques, and the analysis can be multiplexed.1,2 The main reason is that the molecular origins of the surface stress changes that cause the r 2010 American Chemical Society

observed nanomechanical deflections must be better understood and used to optimize analytical performance.1-3 It is only recently that researchers have stepped back from investigating complex biomolecular interactions2 to explicitly address some of the physicochemical parameters affecting the stress response in a number of different systems (e.g., surface morphology,3,4 surface charge state,5 surface modification chemistries,3 and sensing layer elastic properties6). Electrochemical reactions represent one type of transformation that has been used to drive the deflection of metal-coated cantilevers for sensing or actuator applications. Examples include underpotential metal deposition,7 oxidation-reduction of metal ions,8 doping/dedoping of conducting polymers,9,10 electrocapillary effects,11 and redox-controlled movement of bistable [3] rotaxanes.12 The voltage potential applied to the cantilever Received: August 31, 2010 Revised: December 7, 2010 Published: December 30, 2010 1985

dx.doi.org/10.1021/jp108284w | J. Phys. Chem. C 2011, 115, 1985–1995

The Journal of Physical Chemistry C

ARTICLE

Scheme 1. (a) Schematic Illustration of the Redox-Induced Deflection of a Ferrocenylalkanethiolate-Modified Microcantilevera and (b) Schematic Illustration of the Monolayer Reorganization Induced by the Oxidation of the Ferrocene to Ferrocenium and Anion Pairing, As Suggested by Spectroelectrochemical Investigations30-33

a A vertical bending of the cantilever, Δz, due to a surface stress change is detected by the lateral displacement, ΔS, of a focused laser beam on a position-sensitive detector or PSD.

(working electrode) is the external stimulus that triggers the electrochemical processes. We recently used a self-assembled ferrocenylundecanethiolate monolayer on a gold-coated cantilever (FcC11SAu) as a model system to investigate the surface stress changes generated by a surface-confined faradaic reaction (Scheme 1a).13,14 In this article, we characterize how the adsorbate structure and electrochemical environment affect the redox-induced surface stress of gold-coated microcantilevers modified with an electroactive self-assembled monolayer (SAM). ω-Ferrocenylalkanethiolate SAMs on gold (FcRSAu) are probably the most extensively studied redox-active assemblies to date.15 Numerous reports have focused on the electrochemical properties and interfacial electron transfer dynamics of clustered and isolated ferrocenes in single-component and mixed systems.16-27 Oxidation of the SAM-bound ferrocene (Fc) to ferrocenium (Fcþ) proceeds via coupled electron-transfer and anion-pairing reactions: Fc f Fcþ þ e-

ð1Þ

Fcþ þ X - f Fcþ X -

ð2Þ

Ferrocene oxidation (1) and anion-pairing (2) are accompanied by changes in surface wettability,28,29 SAM volume,21 and SAM structure30-33 that can cause the cantilever to bend and must be taken into account in the interpretation of the surface stress results. The formation of surface ion pairs between the electrogenerated Fcþ and X- stabilizes the oxidized cation.27 Hydrophobic anions, such as PF6-, ClO4-, and BF4-, pair more effectively with the poorly solvated Fcþ than hydrophilic ones, such as NO3- and F-.22,27 The nature of the anion, X-, strongly affects the redox response in CV (i.e., half-wave potential, formal width at halfmaximum of the anodic peak, number of surface ferrocenes oxidized) and stability of the FcRSAu SAM upon repeated potential cycling in aqueous solution.22,25,27 The effect of the electrolyte anion on the electron transfer across ferrocene-terminated SAMs

may be related to differences in the physicochemical changes induced by the electron transfer/anion pairing reactions. In an investigation of the mediated oxidation of ascorbic acid in solution through FcC4COOC9SAu SAMs, Valincius et al. found that the formation of interfacial FcþX- pairs inhibits electrocatalytic electron transfer, while the absence of ion pairing favors the process.27 The electron transfer inhibition by hydrophobic anions involving tight (high energy) ion pairs was explained in terms of the formation of an electrically neutral, rigid two-dimensional FcþXlayer or network at the SAM/solution interface which restricts the thermal movement of the electron accepting ferrocenium terminus and results in a low transition state entropy of the electron transfer reaction activated complex.27 In the absence of surface-bound ion pairs for hydrophilic anions, the net positive charge of the oxidized SAM may be shielded by water dipoles and diffusely distributed counterions located outside of the SAM, resulting in a less constrained ferrocenium terminus and a significantly mediated electron transfer rate.27 Related to this, Orlwoski et al. compared the reorganization energy for the electrochemical oxidation of ferrocene-peptide SAMs in the presence of PF6-, ClO4-, and BF4-.34 The highest reorganization energy was observed for the BF4- counterion, which is attributed to a perturbation of the monolayer structure by penetration of the more weakly associated BF4- anions into the film.34 By contrast, the strongly associating PF6- and ClO4- maintain contact interactions with Fcþ and do not penetrate the monolayer, resulting in lower reorganization energies.34 Moreover, electrochemical quartz microbalance (EQCM) studies have found that the surface coverage of FcRSAu affects the redox-induced anion association, uptake of solvent, and film structural change.35,36 Overall, these findings imply that the SAM molecular ordering/packing density and the ability of the complexing anion to induce stabilizing organizational changes within the oxidized film may play important roles in the magnitude of the bending response of FcRSAu-modified cantilevers and are therefore obvious parameters to investigate. We previously reported that the oxidation of a FcC11SAu SAM in perchlorate electrolyte generates a compressive surface stress change of -0.20 ( 0.04 N m-1.13 These cantilever experiments were carried out under conditions (i.e., strong anion pairing and maximum ferrocene surface coverage) where solvent and ion penetration into the SAM are significantly inhibited so that the measured stress changes could be straightforwardly related to the oxidation of the SAM-bound ferrocene to ferrocenium. The microcantilever deflection versus quantity of electrogenerated ferrocenium obtained in cyclic voltammetry (CV) and potential step/hold experiments, as well as the surface stress changes obtained for mixed FcC11SAu/C11SAu SAMs containing different populations of clustered and isolated ferrocenes, permitted us to establish the molecular basis of the stress generation.13 As discussed in an earlier paper,13 our results are consistent with the cantilever bending arising from the monolayer volume expansion created by the collective reorientational motions induced by the complexation of perchlorate ions to the surface-immobilized ferroceniums (Scheme 1b). The cantilever responds to the lateral pressure exerted by an ensemble of reorienting Fcþ-bearing alkylthiolates upon each other rather than individual anion pairing events.13 The FcC11SAu-modified cantilevers exhibit a striking 10- to 15-fold greater charge-normalized surface stress (4500 N m-1/ C cm-2)13 compared to the redox-activated deflection of cantilevers coated with polyaniline (∼300 N m-1/C cm-2)9 and polypyrrole (∼350 N m-1/C cm-2)10 multilayers (see Supporting 1986

dx.doi.org/10.1021/jp108284w |J. Phys. Chem. C 2011, 115, 1985–1995

The Journal of Physical Chemistry C Information). We attribute the larger stress change observed for the FcC11SAu microcantilever to greater steric constraints in the closer-packed FcC11SAu SAM compared to the conducting polymer films and to an efficient coupling between the chemisorbed FcC11S- molecules and the Au-coated microcantilever transducer versus the physisorbed polyaniline and polypyrrole.13 Although larger deflections are theoretically obtainable by building thicker films of conducting polymers (i.e., increase the number of redox sites per square area), this is not possible for ferrocene-terminated SAMs. Other system properties must therefore be tailored to produce a larger actuation for cantilevers modified with an electroactive monolayer film. To this end, we systematically investigate here the effect of the alkyl chain spacer and electrolyte anion on the magnitude and reversibility of the potential-induced bending of ferrocenylalkanethiolate-functionalized gold-coated cantilevers. The chain composition of the ferrocene-to-thiol spacer has been shown to affect the intermolecular interactions and electron transfer properties of the redox-active SAM.37,38 By varying the length of the spacer and type of ferrocene-spacer linker, one can examine the impact of molecular structure on the redox-elicited microcantilever response. The present work considers SAMs formed from the following ferrocenylalkanethiols: Fc(CO)(CH2)11SH, Fc(CH2)12SH, Fc(CH2)11SH, and Fc(CH2)6SH, which are respectively denoted Fc(CO)C11SH, FcC12SH, FcC11SH, and FcC6SH. We also compare the potential-controlled response of FcC11SAu modified microcantilevers in the presence of anions that associate with the ferrocenium cation to varying degrees.

’ EXPERIMENTAL SECTION Materials. The ferrocenylalkylthiols Fc(CO)C11SH, FcC12SH, FcC11SH, and FcC6SH were prepared according to the procedure of Creager and Rowe,37 starting from ferrocene (g98%, Fluka) and either 12-bromododecanoic acid (g98%, Fluka), 11-bromoundecanoic acid (g98%, Fluka), or 6-bromohexanoic acid (g97%, Fluka). The purity and identity of the products were verified by thin layer chromatography (silica gel, hexanes/ethyl acetate 99:1 v/v) and 1H NMR spectroscopy (400 MHz, CDCl3). The following reagents were used without any further purification: 1-dodecanethiol (C12SH, g97%, Fluka), sodium perchlorate (NaClO4, 98%, Sigma Aldrich), sodium fluoride (NaF, 99þ%, Sigma Aldrich), sodium nitrate (NaNO3, 99.0%, Sigma Aldrich), sodium tetrafluoroborate (NaBF4, Alfa Inorganics), and sodium hexafluorophosphate (NaPF6, 99%, Strem Chemicals). All aqueous electrolyte solutions were prepared with ultrapure water (18.2 MΩ 3 cm) obtained by further purification of distilled water with a Milli-Q Gradient system (Millipore, Bedford, MA). The perchlorate electrolyte solution contained 0.010 M HClO4 and 0.10 M NaClO4, whereas the 0.10 M NaPF6, 0.10 M NaBF4, 0.10 M NaNO3, and 0.10 M NaF electrolyte solutions contained no added acid. A refractive index of 1.33 was measured at 20 C and λ = 589 nm (AR200 digital hand-held refractometer, Reichert Analytical Instruments, USA) for all of the electrolyte solutions. The solutions were purged with N2 gas for at least 20 min prior to the electrochemical measurements to minimize oxygen levels. Preparation of Gold-Covered Substrates and SAMs. The original reflective gold coating was stripped off of the silicon nitride microcantilever chips (model # MLCT-AUHW, Veeco Probes, Camarillo, CA) by immersion in a dilute aqua regia

ARTICLE

(3:1:6 HCl/HNO3/H2O) solution for ∼5 min, followed by a thorough rinse with ultrapure water. B270 glass slides (Esco Products, Inc., Oak Ridge, NJ) were substituted for the microcantilevers in ellipsometry, electrochemical surface plasmon resonance (ESPR) spectroscopy, polarization modulationinfrared reflection-absorption spectroscopy (PM-IRRAS), and cyclic voltammetry (CV). Prior to the deposition of metal, the microcantilever chips and glass slides were immersed in piranha solution (3:1 H2SO4 and 30% H2O2) for ∼3 min, rinsed extensively with ultrapure water followed by absolute ethanol, and dried with nitrogen. One face of the clean microcantilevers was rendered electrically conductive by the deposition of 5 nm of Ti (99.99%, Alfa Aesar, USA) followed by 85 nm of Au (99.99%, Kitco Metals Inc., Montreal, QC) in a VE-90 thermal evaporator equipped with a calibrated quartz crystal deposition monitor (Thermionics Vacuum Products, Port Townsend, WA). The B270 glass slides were coated with 1.2 nm of Ti and 48 nm of Au for ESPR spectroscopy, ellipsometry, and CV experiments and with 5 nm of Ti and 100 nm of Au for PM-IRRAS. The gold-coated substrates (glass slides and cantilever chips) were immersed for ∼12 h in one of the following 1 mM thiol solutions: FcC12SH, FcC11SH, and FcC6SH (80:20 v/v absolute EtOH/THF), C12SH (100% EtOH), or Fc(CO)C11SH (95% EtOH). Each SAM-modified substrate was removed from the incubation solution, rinsed copiously with absolute ethanol, dried with nitrogen, and used immediately in an experiment. Spectroscopic Ellipsometry. The thicknesses of the ferroceneterminated SAMs and the optical constants (n and k) of the supporting gold layer were determined using a multiwavelength ellipsometer equipped with a QTH lamp and rotating compensator (model M-2000 V, J.A. Woollam Co., Inc., Lincoln, NE). All measurements were performed in air at an incident angle of 70 and a wavelength range of 370-1000 nm. Five to six different spots on each substrate surface were analyzed and the results averaged. The average film thickness was calculated from plots of Ψ and Δ versus wavelength (λ) using a four-layer model: glass (0.92 mm)/Ti (1.2 nm)/Au (48 nm)/SAM and the Levenberg-Marquardt nonlinear optimization algorithm of the vendor's WVASE32 software. The complex refractive index (^n = n - ki) of the freshly evaporated gold film was first calculated as a function of λ. The titanium and gold film thicknesses were fixed to those measured by the calibrated quartz crystal monitor during thermal evaporation. The n(λ) and k(λ) values provided in the vendor's materials database for polycrystalline titanium and BK7 glass were used in the fitting process. The SAM thicknesses were calculated using the optical parameters determined for the bare gold reference and the Cauchy dispersion equation: n(λ/μm) = A þ B/λ2 þ C/λ4, where A = 1.45, B = 0.01, and C = 0. Polarization Modulation-Infrared Reflection-Absorption Spectroscopy (PM-IRRAS). The FcRSAu SAMs were characterized by PM-IRRAS using a Bio Rad FTS-6000 Fourier transform infrared spectrometer equipped with a liquid-nitrogen-cooled mercury-cadmium-telluride detector. The incident angle was ca. 70 for both p- and s-polarization. Spectra were collected at a resolution of 2 cm-1 using 10 coadded scans. Cyclic Voltammetry (CV). All CV experiments were carried out using an Epsilon potentiostat (Bioanalytical Systems, Inc., West Lafayette, IN). A custom-built, one-compartment threeelectrode cell was employed (Figure S1 of Supporting Information), where a FcRSAu-coated glass slide served as the working electrode for quantitative surface analysis, the counter electrode was a platinum wire (99.99%, Alfa Aesar), and all potentials are 1987

dx.doi.org/10.1021/jp108284w |J. Phys. Chem. C 2011, 115, 1985–1995

The Journal of Physical Chemistry C reported with respect to an aqueous Ag/AgCl reference electrode (3 M NaCl, Bioanalytical Systems). CV Deconvolution Method. The analysis of the CVs required a mathematical deconvolution (OriginPro 7.5) of the background-current-corrected anodic segment of the CV, as described by Lee et al.24 for binary SAMs. The lower-potential peak (shoulder peak assigned to ferrocenes at defect sites, vide infra) was fit to a Gaussian distribution, and a Lorentzian function was applied to the higher-potential peak (main peak assigned to clustered ferrocenes, vide infra). Three fitting parameters, peak position, peak width, and peak area, were used to fit the peaks. Electrochemical SPR (ESPR) Spectroscopy. Redox-induced changes in the FcRSAu SAM thicknesses were quantified with a SR7000 SPR instrument (Reichert, Inc., Depew, NY). A custombuilt electrochemical cell (Figure S1), fitted with reference and counter electrodes, was mounted onto the SAM-functionalized surface of the gold-coated slide (working electrode). The CVs were acquired at a scan rate of 10 mV s-1 in an aqueous solution of 0.10 M NaClO4/0.01 M HClO4. The SR7000 SPR instrument employs the Kretschmann-type attenuated total reflection configuration, where surface plasmons are excited with p-polarized incident light from a 15 mW GaAlAs emitter (peak emission λ = 780 nm) which is focused through a sapphire prism onto the underside of the gold-coated glass. The glass slide was optically coupled to the base of the sapphire prism using immersion oil (Cargille type A liquid, nD = 1.515). The total internally reflected light from the gold/solution interface is detected with a 3696pixel CCD linear array, and the optical pixel signals are digitized with a 14 bit analogue-to-digital converter. A National Instruments Labview interface (SR7000 Alpha Instrument version 2.24) is used for data acquisition and transfer. The pixel of minimum reflectivity was tracked with a time resolution of 2-3 s as a function of the applied potential. All experiments were carried out at 25 C, and the temperature at the gold/solution interface was controlled to within (0.015 C by a Peltier device. The change in the pixel of minimum reflectivity resulting from the oxidation of ferrocene to ferrocenium was converted to a resonance angle change (ΔΘm) using a pixel-to-incident angle relation of 1 pixel = 0.00502 established through calibration of the SR7000 instrument with solutions of different refractive indices.39 The change in SAM thickness (ΔdFcfFcþ) caused by oxidation of the ferrocene to ferrocenium was derived from the resulting ΔΘm using Fresnel multilayer modeling (Winspall software version 2.20, MPI-P, Mainz, Germany). Listed in Table 1 are the parameters used to calculate ΔdFcfFcþ. Electrochemical Microcantilever Measurements. The static mode of operation was employed to monitor the deflection of the FcRSAu-coated microcantilevers as a function of applied potential (scan rate of 5 mV s-1). The custom-built, reflecting beam deflection setup integrating the microcantilever chip with a potentiostat and the conversion of the position-sensitive detector (PSD) voltage signal to a vertical cantilever deflection are detailed in an earlier publication.13 The nanometer-scale microcantilever deflection, Δz, is directly proportional to the induced surface stress change, Δσ, and calculated as outlined in the Supporting Information of ref 13.40 In accordance with common sign convention, the compressive surface stress is expressed as a negative value.41 The measured electrochemical current arises mainly from the chip substrate, given the relative FcRSAu-covered surface areas of the cantilevers, ClO4- > BF4- > NO3- > F-.22,23,25,27 While PF6-, ClO4-, and BF4- give reversible and stable microcantilever deflections, oxidation of the FcC11SAu SAM in NO3- and Fcauses an irreversible deformation of the cantilever deflection. The likely source of this deformation is addressed below. It is important to note that, although ΔσFcfFcþ and ΓFcþ follow the same trend, there is a nonlinear relation between ΔσFcfFcþ and ΓFcþ for the series of anions investigated (Table 4), indicating that the surface stress changes that cause the cantilever to deflect do not directly originate from the electrogeneration of the ferrocenium cation but from processes that accompany the electron transfer step. The ionic size and solvation free energy of PF6- are similar to those of ClO4-.25 Both anions form strong surface ion pairs with the ferroceniums that stabilize the electrogenerated cations.27 A 1:1 stoichiometric association of ClO4- and PF6- with the ferroceniums has been found by EQCM.27,35,56 The ΓFcþ of 4.8 ((0.2)  10-10 mol cm-2 obtained with PF6- concurs with the theoretical coverage, indicating that all of the available surface ferrocenes are electrochemically oxidized. The SAM and magnitude of the cantilever deflection are stable over at least three successive redox cycles (Figure 3a). Between the first and the third CV scan, there is an average variation of 4% in the anodic peak current and a change of 0.5% in the peak deflection amplitude. The ΔσFcfFcþ of -0.21 ( 0.05 N m-1 measured for the FcþPF6- pair is statistically equivalent (t-test) to the value of -0.20 ( 0.04 N m-1 reported for the FcþClO4system.13 BF4- anions pair with the electrochemically generated ferroceniums to a weaker extent than PF6-,22,27 as reflected by the lower ΓFcþ value and positive shift of ∼125 mV in E1/2. A decreased ΔσFcfFcþ is measured (i.e., -0.13 ( 0.03 N m-1), suggesting that the weaker association of BF4- with FcþC11SAu does not induce the same degree of monolayer reorganization as ClO4- and PF6-. A marked difference in the FcC11SAu-modified microcantilever response is evident with the hydrophilic NO3- and Fanions. Significant decreases in the anodic peak currents (14% for NO3- and 45% for F-) and ΔσFcfFcþ (24% for NO3- and 43% for F-) are observed over the three potential cycles shown in Figure 3d,e. Moreover, the cantilevers do not return to their initial zero-stress (open circuit) positions at the end of the first oxidation-reduction cycle. It has been shown that the anion solvation directly influences the stability of ferrocenium-terminated SAMs.22,25,27 More hydrophilic anions are transported with large amounts of water, inhibiting the extent of anion interaction with the ferrocenium cation.22,25,27 Consequently,

b

the ferrocene cation is subject to nucleophilic attack, resulting in the demetallization of the ferrocene and loss of electroactivity.47 Demetallization of the terminal ferrocene is a probable source of the current decrease and deformation of the cantilever deflection observed upon repeated oxidation and reduction. The ΔσFcfFcþ values measured during the second anodic scan (where the baselines before and after the oxidation-reduction cycle are the same) are -0.11 ( 0.03 N m-1 and -0.06 ( 0.03 N m-1 in the presence of NO3- and F-, respectively. For the oxidation of SAM-bound ferrocenes in NO3- and F-, deducing the origin(s) of the surface stress is more complicated. Coulomb repulsion between neighboring ferrocenium moieties (i.e., less effective ion pairing), potential-induced ion and solvent penetration into the SAM, anion adsorption to the positively charged, underlying gold surface, and a destabilization of the SAM are possible contributors.22,27,36,62,63 Altogether, the trend in the measured ΔσFcfFcþ values indicates that anion pairing yields reversible perturbations of the film structure that stabilize the oxidized state of the FcC11SAu SAM, resulting in larger microcantilever deflections, while a poor ion pairing produces destabilizing perturbations and smaller, irreversible microcantilever deflections.

’ CONCLUSIONS This work identifies some of the system properties that influence the response characteristics of micromechanical cantilevers modified with electroactive SAMs. Our findings show that the alkyl chain spacer and electrolyte anion influence the amplitude and reversibility of the electrochemically induced bending of cantilevers functionalized with model ferrocenylalkylthiolate SAMs. The resonance angle changes measured by ESPR in the presence of ClO4- indicate that the extent of the oxidationinduced SAM reorganization is the same for short (n = 6) and long (n = 12) chain ferrocenylalkanethiolates (FcCnSAu) and for a long chain carbonyl derivative (Fc(CO)C11SAu). However, the magnitude of the measured cantilever deflection is not the same, reflecting differences in the tensile contributions of the SAM elasticity and wettability versus compressive SAM reorganization to the overall surface stress of the different modified cantilevers. The magnitude of the oxidation-induced compressive surface stress change follows the anion-pairing ability. PF6- and ClO4-, which form 1:1 contact ion pairs with the SAM-bound ferrocenium cations, give reversible FcC11SAu microcantilever deflections and the largest surface stress changes. By contrast, the oxidation of FcC11SAu SAMs in NO3- and F-, which exhibit the 1993

dx.doi.org/10.1021/jp108284w |J. Phys. Chem. C 2011, 115, 1985–1995

The Journal of Physical Chemistry C poorest ion-pairing abilities of the anions investigated, results in an irreversible deformation of the cantilever bending and smaller compressive surface stress changes. In the case of the most hydrophobic PF6- and ClO4- counterions, the compressive surface stress change is likely dominated by collective molecular reorientations caused by the tight complexation of these anions to the SAM-bound ferroceniums to form a rigid two-dimensional ionic network at the SAM/solution interface. For the more hydrophilic NO3- and F- anions, the surface stress change is most probably the result of several phenomena: electrostatic interactions between neighboring ferrocenium moieties, potential-induced ion and solvent penetration into the SAM, anion adsorption to the underlying gold surface, and a perturbation of the SAM integrity. The results of this investigation highlight the importance of knowing how the experimental parameters and properties of the chemical system affect the measured surface stress for the optimization of the microcantilever peformance.

’ ASSOCIATED CONTENT

bS

Supporting Information. Calculation of the charge-normalized surface stress, schematic of the custom-built electrochemical cell, and deconvolution of the voltammetric anodic scans of FcRSAu-modified cantilever substrates. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Tel.: 514-343-7057. Fax: 514-343-7586. E-mail: antonella.badia@ umontreal.ca.

’ ACKNOWLEDGMENT This work was supported by NSERC (Canada), CFI (Canada), FQRNT (Quebec), Canada Research Chairs program, and Universite de Montreal. L.L.N. acknowledges financial support from the Groupe de recherche en technologie des couches minces of the regroupement quebecois sur les materiaux de pointe. ’ REFERENCES

 . Analyst 2009, 134, 1274–1290. (1) Zougagh, M.; Ríos, A (2) Alvarez, M.; Lechuga, L. M. Analyst 2010, 135, 827–836. (3) Ji, H.-F.; Armon, B. D. Anal. Chem. 2010, 82, 1634–1642. (4) Tabard-Cossa, V.; Godin, M.; Burgress, I. J.; Monga, T.; Lennox, R. B.; Gr€utter, P. Anal. Chem. 2007, 79, 8136–8143. (5) Arroyo-Hernandez, M.; Tamayo, J.; Costa-Kr€amer, J. L. Langmuir 2009, 25, 10633–10638. (6) Sushko, M. L.; Harding, J. H.; Shluger, A. L.; McKendry, R. A.; Watari, M. Adv. Mater. 2008, 20, 3848–3853. (7) Brunt, T. A.; Rayment, T.; O'Shea, S. J.; Welland, M. E. Langmuir 1996, 12, 5942–5946. (8) Tian, F.; Boiadjiev, V. I.; Pinnaduwage, L. A.; Brown, G. M.; Thundat, T. J. Vac. Sci. Technol., A 2005, 23, 1022–1028. (9) Lahav, M.; Durkan, C.; Gabai, R.; Katz, E.; Willner, I.; Welland, M. E. Angew. Chem., Int. Ed. 2001, 40, 4095–4097. (10) Tabard-Cossa, V.; Godin, M.; Gr€utter, P.; Burgess, I.; Lennox, R. B. J. Phys. Chem. B 2005, 109, 17531–17537. (11) Raiteri, R.; Butt, H.-J. J. Phys. Chem. 1995, 99, 15728–15732. (12) Juluri, B. K.; Kumar, A. S.; Liu, Y.; Ye, T.; Yang, Y.-W.; Flood, A. H.; Fang, L.; Stoddart, J. F.; Weiss, P. S.; Huang, T. J. ACS Nano 2009, 3, 291–300.

ARTICLE

(13) Norman, L. L.; Badia, A. J. Am. Chem. Soc. 2009, 131, 2328– 2337. (14) Quist, F.; Tabard-Cossa, V.; Badia, A. J. Phys. Chem. B 2003, 107, 10691–10695. (15) Finklea, H. O. In Electroanalytical Chemistry: A Series of Advances; Bard, A. J., Rubinstein, I., Eds.; Marcel Dekker: New York, 1996; Vol. 19, pp 109-335. (16) Auletta, T.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Langmuir 2002, 18, 1288–1293. (17) Chidsey, C. E. D. Science 1991, 251, 919–922. (18) Chidsey, C. E. D.; Bertozzi, C. R.; Putvinski, T. M.; Mujsce, A. M. J. Am. Chem. Soc. 1990, 112, 4301–4306. (19) Collard, D. M.; Fox, M. A. Langmuir 1991, 7, 1192–1197. (20) Smalley, J. F.; Feldberg, S. W.; Chidsey, C. E. D.; Linford, M. R.; Newton, M. D.; Liu, Y.-P. J. Phys. Chem. 1995, 99, 13141–13149. (21) Crua~ nes, M. T.; Drickamer, H. G.; Faulkner, L. R. Langmuir 1995, 11, 4089–4097. (22) Ju, H.; Leech, D. Phys. Chem. Chem. Phys. 1999, 1, 1549–1554. (23) Kondo, T.; Okamura, M.; Uosaki, K. J. Organomet. Chem. 2001, 637-639, 841–844. (24) Lee, L. Y. S.; Sutherland, T. C.; Rucareanu, S.; Lennox, R. B. Langmuir 2006, 22, 4438–4444. (25) Rowe, G. K.; Creager, S. E. Langmuir 1991, 7, 2307–2312. (26) Uosaki, K.; Sato, Y.; Kita, H. Langmuir 1991, 7, 1510–1514. (27) Valincius, G.; Niaura, G.; Kazakevicien_e, B.; Talaikyt_e, Z.; Kazem_ekait_e, M.; Butkus, E.; Razumas, V. Langmuir 2004, 20, 6631– 6638. (28) Abbott, N. L.; Whitesides, G. M. Langmuir 1994, 10, 1493– 1497. (29) Sondag-Huethorst, J. A. M.; Fokkink, L. G. J. Langmuir 1994, 10, 4380–4387. (30) Kazakevicien_e, B.; Valincius, G.; Niaura, G.; Talaikyt_e, Z.; Kazem_ekait_e, M.; Razumas, V.; Plausinaitis, D.; Teiserskien_e, A.; Lisauskas, V. Langmuir 2007, 23, 4965–4971. (31) Viana, A. S.; Jones, A. H.; Abrantes, L. M.; Kalaji, M. J. Electroanal. Chem. 2001, 500, 290–298. (32) Kondo, T.; Horiuchi, S.; Yagi, I.; Ye, S.; Uosaki, K. J. Am. Chem. Soc. 1999, 121, 391–398. (33) Ye, S.; Sato, Y.; Uosaki, K. Langmuir 1997, 13, 3157–3161. (34) Orlowski, G. A.; Chowdhury, S.; Kraatz, H.-B. Langmuir 2007, 23, 12765–12770. (35) Shimazu, K.; Yagi, I.; Sato, Y.; Uosaki, K. J. Electroanal. Chem. 1994, 372, 117–124. (36) Ye, S.; Haba, T.; Sato, Y.; Shimuza, K.; Uosaki, K. Phys. Chem. Chem. Phys. 1999, 1, 3653–3659. (37) Creager, S. E.; Rowe, G. K. J. Electroanal. Chem. 1994, 370, 203–211. (38) Napper, A. M.; Liu, H.; Waldeck, D. H. J. Phys. Chem. B 2001, 105, 7699–7707. (39) Norman, L. L.; Badia, A. Langmuir 2007, 23, 10198–10208. (40) Godin, M.; Tabard-Cossa, V.; Gr€utter, P. Appl. Phys. Lett. 2001, 79, 551–553. (41) Ibach, H. Surf. Sci. Rep. 1997, 29, 195–263. (42) Watcharinyanon, S.; Moons, E.; Johansson, L. S. O. J. Phys. Chem. C 2009, 113, 1972–1979. (43) Walczak, M. M.; Popenoe, D. D.; Deinhammer, R. S.; Lamp, B. D.; Chung, C.; Porter, M. D. Langmuir 1991, 7, 2687–2693. (44) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559–3568. (45) Abrantes, L. M.; Kalaji, M.; Viana, A. S. Russ. J. Electrochem. 2002, 38, 39–43. (46) Lee, M.; Chung, C. Bull. Korean Chem. Soc. 1999, 20, 1–3. (47) Popenoe, D. D.; Deinhammer, R. S.; Porter, M. D. Langmuir 1992, 8, 2521–2530. (48) Sato, Y.; Frey, B. L.; Uosaki, K.; Corn, R. M. Bull. Chem. Soc. Jpn. 1994, 67, 21–25. (49) Shimazu, K.; Ye, S.; Sato, Y.; Uosaki, K. J. Electroanal. Chem. 1994, 375, 409–413. 1994

dx.doi.org/10.1021/jp108284w |J. Phys. Chem. C 2011, 115, 1985–1995

The Journal of Physical Chemistry C

ARTICLE

(50) Viana, A. S.; Abrantes, L. M.; Jin, G.; Floate, S.; Nichols, R. J.; Kalaji, M. Phys. Chem. Chem. Phys. 2001, 3, 3411–3419. (51) Bard, A. J.; Faulkner, L. R. Electrochemical Methods Fundamentals and Applications, 2nd ed.; J. Wiley & Sons: New York, 2001. (52) Ohtsuka, T.; Sato, K.; Uosaki, K. Langmuir 1994, 10, 3658– 3662. (53) Yao, X.; Wang, J.; Zhou, F.; Wang, J.; Tao, N. J. Phys. Chem. B 2004, 108, 7206–7212. (54) Calvente, J. J.; Andreu, R.; Molero, M.; Lopez-Perez, G.; Domínguez, M. J. Phys. Chem. B 2001, 105, 9557–9568. (55) Lee, L. Y. S.; Lennox, R. B. Langmuir 2007, 23, 292–296. (56) Viana, A. S.; Kalaji, M.; Abrantes, L. M. Langmuir 2003, 19, 9542–9544. (57) Yao, X.; Yang, M.; Wang, Y.; Hu, Z. Sens. Actuators, B 2007, 122, 351–356. (58) Umeda, K.-i.; Fukui, K.-i. Langmuir 2010, 26, 9104–9110. (59) Kondo, T.; Takechi, M.; Sato, Y.; Uosaki, K. J. Electroanal. Chem. 1995, 381, 203–209. (60) Green, J.-B. D.; McDermott, M. T.; Porter, M. D. J. Phys. Chem. 1996, 100, 13342–13345. (61) Willman, K. W.; Murray, R. W. Anal. Chem. 1983, 55, 1139– 1142. (62) De Long, H. C.; Donohue, J. J.; Buttry, D. A. Langmuir 1991, 7, 2196–2202. (63) Rowe, G. K.; Creager, S. E. J. Phys. Chem. 1994, 98, 5500–5507.

1995

dx.doi.org/10.1021/jp108284w |J. Phys. Chem. C 2011, 115, 1985–1995