Enhancement of Au Nanoparticles Formed by in Situ Electrodeposition

Nov 10, 2008 - ... spectroscopy (EIS), scanning electron microscopy (SEM), and energy dispersive X-ray (EDX) analysis were used to characterize the fi...
1 downloads 0 Views 505KB Size
Download PDF
J. Phys. Chem. B 2008, 112, 15513–15520

15513

Enhancement of Au Nanoparticles Formed by in Situ Electrodeposition on Direct Electrochemistry of Myoglobin Loaded into Layer-by-Layer Films of Chitosan and Silica Nanoparticles Xihong Guo,† Dong Zheng,‡ and Naifei Hu*,† Department of Chemistry, Beijing Normal UniVersity, Beijing, 100875, P. R. China, and Analytical and Testing Center, Beijing Normal UniVersity, Beijing, 100875, P. R. China ReceiVed: August 20, 2008; ReVised Manuscript ReceiVed: September 28, 2008

In the present work, a new kind of myoglobin (Mb)/Au nanoparticles composite film was fabricated on pyrolytic graphite (PG) electrodes. Oppositely charged chitosan (CS) and silica (SiO2) nanoparticles were alternately adsorbed on the PG surface by the electrostatic interaction between them, forming {CS/SiO2}5 layer-by-layer films. Mb and HAuCl4 in solution were then simultaneously loaded into {CS/SiO2}5 films. The loaded Au(III) in the films were electrochemically reduced into Au nanoparticles, forming nanocomposite films, designated as {CS/SiO2}5-Mb-Au. Various techniques such as cyclic voltammetry (CV), square wave voltammetry (SWV), quartz crystal microbalance (QCM), electrochemical impedance spectroscopy (EIS), scanning electron microscopy (SEM), and energy dispersive X-ray (EDX) analysis were used to characterize the films. Compared with {CS/SiO2}5-Mb films without Au nanoparticles inside, the {CS/SiO2}5-Mb-Au films exhibited much better behavior in electrochemistry and electrocatalysis of Mb, mainly because the Au nanoparticles formed inside the films were located in proximity to Mb and acted as electron bridges between Mb molecules, making more Mb molecules in the films become electroactive. In addition, the permeability or porosity of the films also played an important role in realizing the direct electrochemistry of Mb. This system provides a novel platform to develop electrochemical biosensors based on the direct electron transfer of redox enzymes without using mediators. Introduction The study of direct chemistry of redox proteins can present a model for the mechanism study of electron transfer between enzymes in biological systems and establish a foundation for fabricating new kinds of biosensors, bioreactors, and biomedical devices without using mediators1,2 and has attracted increasing interest among researchers. The films modified on an electrode surface may provide a favorable microenvironment for the proteins to transfer electrons directly with electrodes, and various types of films incorporating proteins have thus been developed for realizing direct electrochemistry of redox proteins.3,4 Among different film-forming materials, Au nanoparticles show great advantages in enhancing electron exchange of proteins with electrodes mainly due to their good biocompatibility and conductivity.5-8 For example, in 1996, Natan and co-workers reported a reversible electrochemistry of cytochrome c at SnO2 electrodes modified with 12 nm diameter Au nanoparticles.5 Since then, a great deal of literatures have been published to study the direct electron transfer of various redox proteins or enzymes using Au nanoparticles as a promoter. Layer-by-layer assembly, originally developed by Decher and co-workers,9,10 has remarkable advantages over other filmforming approaches in the precise control of film composition and thickness at molecular or nanometer level with relative simplicity and versatility. The layer-by-layer films are also used toimmobilizeredoxproteinsandrealizetheirdirectelectrochemistry.11-14 Particularly, different kinds of Au nanoparticles were assembled * Corresponding author: e-mail [email protected]; Tel (+86) 105880-5498; Fax (+86) 10-5880-2075. † Department of Chemistry. ‡ Analytical and Testing Center.

layer-by-layer with myoglobin (Mb) on the surface of pyrolytic graphite (PG) electrodes by our group, and the electron transfer of Mb in the films was greatly enhanced.15-17 Recently, a new type of protein film, called protein-loaded layer-by-layer film, has been developed18 and used to study the protein electrochemistry. This kind of protein film is fabricated essentially by two steps. First, layer-by-layer films of polyelectrolytes or nanoparticles are assembled on the surface of solid substrates mainly by electrostatic interaction. In the next step, the assembled films are placed into a protein solution to load or incorporate the protein, forming protein-loaded layer-by-layer films. Compared to the regular protein layer-by-layer films directly assembled by proteins and polyelectrolytes or nanoparticles, protein-loaded films demonstrate unique advantages in keeping the native structure and bioactivity of the proteins, since the loading of proteins is established on the basis of spontaneous adsorption or/and absorption of the proteins, which is separated from the process of layer-by-layer assembly.19-21 Some redox proteins loaded in the layer-by-layer films demonstrated enhanced electron transfer with underlying electrodes.22-25 Layer-by-layer films have also been used as nanoreactors to prepare metal nanoparticles by postdeposition in situ.26 The in situ deposition method displays advantages over the conventional layer-by-layer films with presynthesized metal nanoparticles such as better controllability or tunability in film structure, composition and particle size, and the flexibility of using a variety of chemical or electrochemical approaches in the postdeposition. For instance, in the work of Rubner’s group, silver ions in solution were first loaded into polyelectrolyte multilayer films and then reduced in a hydrogen atmosphere at an elevated temperature, forming Ag nanoparticles inside the

10.1021/jp807452z CCC: $40.75  2008 American Chemical Society Published on Web 11/11/2008

15514 J. Phys. Chem. B, Vol. 112, No. 48, 2008 films.27,28 Because the incorporated silver ions were well distributed along the polymer chains and the networks around silver ions limited particle aggregation, the Ag nanoparticles in the films were dispersed uniformly throughout the films with small size. Similar results were also observed by Bruening and co-workers.29 This inspires us to combine the layer-by-layer selfassembly technique with in situ electrochemical deposition of gold nanoparticles to realize the direct electrochemistry of loaded proteins in multilayer films. While the formation of Au nanoparticles electrodeposited in various kinds of layer-by-layer films on electrode surface has been described,30-32 to the best of our knowledge in situ electrodeposition of Au nanoparticles within multilayer films after infiltration or loading of Au(III) ions into the films and its enhancement on protein electrochemistry has not been reported up to now. In the present work, oppositely charged chitosan (CS) and SiO2 nanoparticles were first assembled into {CS/SiO2}n layerby-layer films. Mb and Au(III) ions in their mixture solution were then simultaneously and spontaneously loaded into the films, followed by electrochemical reduction of Au (III) into Au nanoparticles. The formed {CS/SiO2}5-Mb-Au films showed much better behavior in Mb electrochemistry and electrocatalysis than the Mb-loaded {CS/SiO2}5 films with no Au nanoparticles inside. Various techniques, such as quartz crystal microbalance (QCM), scanning electron microscopy (SEM), energy dispersive X-ray (EDX) analysis, and different electrochemical approaches, were used to characterize the films. The mechanism of electron transfer of Mb in {CS/SiO2}5Mb-Au films, especially the mechanism of enhancement of Au nanoparticles on direct electrochemistry of Mb in the films, was discussed in detail. Experimental Section 1. Chemicals. Horse heart myoglobin (Mb, MW 17 800) and chitosan (CS, the degree of acetylation is less than 15%, MW ∼ 200 000) were purchased from Sigma. The sixth-generation poly(amidoamine) dendrimer (PAMAM) with surface amino groups (10 wt % in methanol), poly(diallyldimethylammonium) (PDDA, 20%, MW 200 000-350 000), poly(styrenesulfonate) (PSS, MW ∼ 70 000), HAuCl4 · 3H2O, and 3-mercapto-1propanesulfonate (MPS, 90%) were obtained from Aldrich. The SiO2 nanoparticles (average diameter 15 ( 5 nm) were purchased from Zhoushan Nanoparticle Technology. K3Fe(CN)6 was obtained from Beijing Chemical Plant. All other chemicals were of reagent grade. Solutions were prepared with twicedistilled water. Buffers were 0.05 M sodium acetate (pH 5.0), potassium dihydrogen phosphate (pH 7.0), or boric acid (pH 9.0) solutions, containing 0.1 M NaCl. The pH of the buffers was adjusted to the desired value by dilute HCl or NaOH. The preparation procedure of presynthesized Au nanoparticles was the same as our previous work.15 In brief, an aqueous solution of chloroauric acid (50 mL, 0.01%) was heated to boiling, followed by addition of trisodium citrate solution (0.2 mL, 1%) and fresh NaBH4 solution (0.75 mL, 0.075%). The dispersion was kept boiling for 5 min. After cooling, the dispersion was stored at 4 °C and used directly for the assembly of multilayer films. 2. Film Assembly. Basal plane pyrolytic graphite (PG, Advanced Ceramics, geometric area 0.16 cm2) electrodes were used for electrochemical studies. Prior to assembly, the PG electrodes were abraded with 320-grit metallographic sandpaper, ultrasonicated in water for 30 s, and then dried in air. The electrodes were then alternately immersed in CS solution (1 mg mL-1 at pH 5.0) and aqueous suspension of SiO2 nanopar-

Guo et al. ticles (3 mg mL-1 at pH 9.0) for 20 min with intermediate water washing, forming {CS/SiO2}n layer-by-layer films on the PG surface. The films were then immersed in the mixture solution of 0.5 mM HAuCl4 and 1 mg mL-1 Mb at pH 7.0 for about 10 h to allow the Au(III) ions and Mb molecules to diffuse into the films. After being washed with water and dried in air, the loaded films were placed in pH 7.0 buffers for cyclic potential scans between +0.3 and -0.1 V at 0.05 V s-1 for 2-3 min until the steady-state voltammograms were obtained. This electrochemical deposition procedure transformed the Au(III) loaded in the films into Au nanoparticles, and the resultant films were designated as {CS/SiO2}n-Mb-Au. The similarly formed films with immersion in HAuCl4 solution without Mb or with immersion in Mb solution without HAuCl4 were designated as {CS/SiO2}n-Au or {CS/SiO2}n-Mb. For assembly of {PAMAM/SiO2}n or {PDDA/PSS}n films and their corresponding Mb-Au or Mb loaded films, a similar assembly procedure was adopted with 1 mg mL-1 PAMAM or 3 mg mL-1 PDDA and PSS solutions. For quartz crystal microbalance (QCM) study, the clean QCM gold electrodes were first immersed in MPS solution for 24 h to chemisorb MPS as a precursor layer, making the electrode surface become negatively charged. The subsequent assembly of {CS/SiO2}n films on the surface of Au/MPS was the same as on the PG electrodes. The films fabricated on PG disks were also used as samples for scanning electron microscopy (SEM) measurements. 3. Apparatus and Procedures. A CHI 621B or CHI 660A electrochemical workstation (CH Instruments) was used for electrochemical measurements. A three-electrode cell was used with a saturated calomel electrode (SCE) as the reference, a platinum wire as the counter, and the PG disk with films as the working electrode. The buffers were purged with highly purified nitrogen for at least 10 min prior to a series of electrochemical experiments, and the nitrogen atmosphere was then maintained above the cell during the whole experiments. Cyclic voltammetry (CV) and square wave voltammetry (SWV) of different film electrodes was performed in pH 7.0 buffers containing no Mb. QCM measurements were performed with a CHI 420 electrochemical analyzer (CH Instruments). After each adsorption step, the QCM gold electrodes were washed with water and dried under a nitrogen stream, and the frequency change was then measured in air by QCM. The SEM images and the spectra of energy dispersive X-ray (EDX) were obtained using a scanning electron microscope (Hitachi, S-4800) equipped with an energy dispersive X-ray analyzer (Horiba, EMAX-350). Results 1. Fabrication of {CS/SiO2}5-Mb-Au Films. CS, as a type of polysaccharide, shows good biocompatibility and has been widely used as a film-forming material in immobilization of proteins.33,34 Biocompatible SiO2 nanoparticles with their good rigidity are also a suitable candidate in immobilizing proteins.34-36 CS and SiO2 nanoparticles were thus used in this work to assemble layer-by-layer films. With its pKa at about 6.0,37,38 CS carries positive charges at pH 5.0, while SiO2 nanoparticles have negative surface charges at pH 9.0 with the isoelectric point at pH 2.039 or 3.0.40 The {CS/SiO2}n layer-by-layer films were thus assembled mainly by electrostatic interaction between oppositely charged CS and SiO2. The assembly procedure was monitored and confirmed by QCM (Figure 1). The frequency decrease (-∆F) showed a roughly linear relationship with adsorption

Enhancement of Au Nanoparticles

Figure 1. QCM frequency shift with adsorption step for {CS/SiO2}n films on a Au/MPS surface: (9) CS and (b) SiO2 adsorption steps.

step, indicating that the {CS/SiO2}n layer-by-layer films are successfully assembled and the assembly is in a regular fashion with nearly equal adsorption amount of CS/SiO2 for each bilayer. When the assembled {CS/SiO2}5 films with n ) 5 were immersed in the mixture of Mb and HAuCl4 solutions at pH 7.0, Mb molecules and Au(III) ions were gradually loaded into the films. After about 10 h, the loaded films were transferred in pH 7.0 blank buffers for continuous CV scans between +0.3 and -0.1 V at 0.05 V s-1 for 2-3 min until the stable voltammograms were obtained. This electrochemical deposition step would transfer the Au(III) ions in the films into Au(0) nanoparticles, since the scanning potential (+0.3 to -0.1 V) is much more negative than the reduction potential of AuCl4-/Au redox pair at about 0.88 V (vs Ag/AgCl),31,41 and the films after the electrodeposition are designated as {CS/SiO2}5-Mb-Au. In this potential window, Mb did not undergo any redox reaction, and the properties of Mb in the films would not be influenced by the potential scan. The formation of Au nanoparticles in the {CS/SiO2}5Mb–Au films was confirmed by SEM. Small bright spots with the size of less than 20 nm were observed on the surface of the films (Figure 2A), which should be attributed to Au nanoparticles since no such bright spots were observed for {CS/ SiO2}5-Mb films with only Mb loaded and with the same electrodeposition procedure (Supporting Information Figure S1). To further support the speculation that the small bright spots are Au nanoparticles, EDX was used to characterize the sports. When the test region of EDX was focused on a bright spot, a characteristic peak of Au at 2.2 keV42,43 was observed, accompanied by the large peaks of Si and O (Figure 2B). When the test region of EDX was focused on the other place, no Au peak could be detected, while the peaks of Si and O were still obviously observed (Figure 2C). By EDX analysis combined with SEM, it is very clear that the bright spots observed on the surface of {CS/SiO2}5-Mb-Au films are Au nanoparticles, which are formed in situ by electrochemical reduction of Au(III). 2. Electrochemical and Electrocatalytic Properties. When {CS/SiO2}5-Mb-Au films were transferred in pH 7.0 buffers containing no Mb, and CV scans were performed between +0.1 and -0.8 V, a well-defined and quasi-reversible reductionoxidation peak pair was observed at about -0.34 V vs SCE, characteristic of Mb Fe(III)/Fe(II) redox couple3,44 (Figure 3a). The CV peaks of {CS/SiO2}5-Mb-Au films showed roughly equal heights of reduction and oxidation peaks and nearly symmetric peak shapes, and both reduction and oxidation peak currents increased linearly with scan rates from 0.05 to 2.0 V s-1, all suggesting that the voltammetric behavior of Mb in the films is diffusionless and surface-confined.45 In this case, the surface concentration of electroactive Mb (Γ*, mol cm-2) could be estimated by integration of CV reduction peak (Q, coulomb) and applying the equation of Q ) NFAΓ* according to Faraday’s

J. Phys. Chem. B, Vol. 112, No. 48, 2008 15515 law,45 where A is the geometric area of electrode (0.16 cm2), N is the number of electrons transferred (1), and F is Faraday’s constant. The value of Γ* was estimated to be about 1.24 × 10-10 mol cm-2. In this work, we assume that the electroactive Mb would be a representative of total Mb loaded into the films, and the loading behavior of electroactive Mb would reflect the behavior of total Mb in the films.25 In contrast, {CS/SiO2}5 and {CS/SiO2}5-Au films showed no CV peaks in the same position (Figure 3c,d). These results confirm that the peak pair at about -0.34 V comes from Mb in {CS/SiO2}5-Mb-Au films, and Mb is indeed loaded into the films. In addition, the {CS/SiO2}5-Mb films also showed a CV peak pair at the similar position for Mb Fe(III)/Fe(II) couple but with much smaller peak heights (Figure 3b) than those of {CS/SiO2}5-Mb-Au films under the same conditions, and the Γ* value was only about 1.35 × 10-11 mol cm-2. This result implies that the Au nanoparticles formed in the {CS/ SiO2}5-Mb-Au films must play a central role in enhancing the direct electrochemistry of Mb in the films. Square wave voltammetry (SWV) combined with nonlinear regression analysis was used to estimate the apparent heterogeneous electron transfer rate constant (ks) for {CS/ SiO2}5-Mb–Au and {CS/SiO2}5-Mb films. The working model in regression was the combination of the single-species surfaceconfined SWV model46 and the formal potential dispersion model, as described in detail previously.47,48 The analysis of SWV data for the two Mb films showed accuracy of fits on the 5-E°′ dispersion model over a range of amplitudes and frequencies. For {CS/SiO2}5-Mb-Au films, the ks was about 28.8 ( 4.5 s-1, a little bit smaller than that for {CS/SiO2}5-Mb films (38.6 ( 5.6 s-1). Considering the estimation error and the limitation of the method, these two ks values should have no substantial difference. The electrocatalytic activity of different films toward O2 was also examined (Figure 4A). With the same amount of air injected, the {CS/SiO2}5-Mb-Au films showed a larger increase in reduction peak at about -0.4 V than the {CS/ SiO2}5-Mb films (Figure 4A, curves a and b), indicating that Au nanoparticles in the former films increase the amount of electroactive Mb and thus enhance the catalytic response toward oxygen. In addition, the {CS/SiO2}5-Au films without Mb inside also showed some electrocatalytic reactivity toward oxygen since the reduction peak potential of O2 at {CS/ SiO2}5-Au films (Figure 4A, curve c) was more positive than that at {CS/SiO2}5 films (Figure 4A, curve d). It was reported that Au nanoparticles could electrocatalyze reduction of O2.49,50 The better electrocatalytic behavior of {CS/SiO2}5-Mb-Au films than that of {CS/SiO2}5-Mb films may thus also partially attributed to the Au nanoparticles in the former films. Amperometry was also employed to compare the electrocatalytic performance of three different films toward hydrogen peroxide (Figure 4B). For all three films, the stepped increase of amperometric current with the addition of H2O2 at a constant potential of -0.1 V was observed, and the steady-state reduction current showed a linear relationship with the concentration of H2O2, indicating that these films have electrocatalytic reactivity toward hydrogen peroxide. However, the slope of calibration curve for the films showed a sequence of {CS/SiO2}5-Mb-Au > {CS/SiO2}5-Mb > {CS/SiO2}5-Au. The much larger slope for {CS/SiO2}5-Mb-Au films suggests that the combination of Mb and Au nanoparticles in the films provides the best electrocatalytic environment for H2O2.

15516 J. Phys. Chem. B, Vol. 112, No. 48, 2008

Guo et al.

Figure 2. (A) SEM top views of {CS/SiO2}5-Mb-Au films assembled on PG disks. (B) EDX spectra focused on the location with a bright spot (a) in (A). (C) EDX spectra focused on the location with no bright spot (b) in (A).

Figure 3. CVs at 0.2 V s-1 for (a) {CS/SiO2}5-Mb-Au, (b) {CS/ SiO2}5-Mb, (c) {CS/SiO2}5-Au, and (d) {CS/SiO2}5 films in pH 7.0 buffers.

3. Influence of Various Factors on Mb Electrochemistry in the Films. Various factors influencing the CV response of electroactive Mb in {CS/SiO2}5-Mb-Au films were investigated. The effect of immersion time (t) of {CS/SiO2}5 films in the mixture solution containing 0.5 mM HAuCl4 and 1 mg mL-1 Mb at pH 7.0 was first examined. The reduction peak current (Ipc) of formed {CS/SiO2}5-Mb-Au films initially increased with the immersion time and then tended to level off at about 5 h (Figure 5). Overnight immersion (about 10 h) was then selected to ensure the saturated loading of Mb and Au(III) ions into the films. The influence of the number of bilayers (n) for {CS/ SiO2}n-Mb-Au films on surface concentration of electroactive Mb in the films (Γ*) was also tested. The Γ* with n ) 5 showed the largest value (Supporting Information Figure S2A), and n ) 5 was thus selected as the optimal number of bilayers in our experiments. The Γ* value was also affected by the concentration of HAuCl4 (CHAuCl4) in the Mb/HAuCl4 loading solution. Γ* increased with CHAuCl4 initially and then reached the steady state (Supporting Information Figure S2B); 0.5 mM HAuCl4 was thus used in the loading experiments. The pH of the Mb-HAuCl4 loading solution had a significant influence on the amount of electroactive Mb in {CS/ SiO2}n-Mb–Au films. The Γ* value with loading solution at pH 5.0 was much larger than that at pH 7.0 or 9.0 (Supporting Information Figure S3), probably because the positively charged Mb molecules at pH 5.0 are more easily loaded into the films.22,24 The {CS/SiO2}n-Mb films without Au nanoparticles inside showed the similar pH-dependent loading behavior (Supporting Information Figure S3). Considering that the difference in Γ* value between {CS/SiO2}n–Mb-Au and {CS/SiO2}n-Mb films was more remarkable at pH 7.0 than at other pHs, in most of our experiments, pH 7.0 was used in loading solution.

4. Comparative Studies. To further understand how Au nanoparticles in the films influence the electron transfer of Mb, two other types of Mb-Au hybrid films with different loading sequence were designed and fabricated. For the first type, the Mb was initially loaded into {CS/SiO2}5 films, the formed {CS/ SiO2}5-Mb films were then transferred into HAuCl4 solution for 10 h to incorporate Au(III) ions, followed by electrodeposition under the same condition, and the formed films were designed as {CS/SiO2}5-Mb+Au. For the second type, the Au(III) ions were initially loaded into {CS/SiO2}5 films followed by electrodeposition; the formed {CS/SiO2}5-Au films were then immersed in Mb solution for 10 h to incorporate Mb, designed as {CS/SiO2}5-Au+Mb. While the CVs of these two types of films in blank buffers at pH 7.0 also demonstrated the reversible peak pair of Mb Fe(III)/Fe(II) couple in the similar position, the reduction and oxidation peak currents or Γ* values were much smaller than those of {CS/SiO2}5-Mb-Au films (Supporting Information Figure S4). Since the only difference of these Mb-Au composite films was the loading sequence of Mb and Au(III) into the {CS/SiO2}5 films, these CV results suggest that the loading of Mb and Au(III) at the same time is more favorable to getting better CV response of Mb than the separate loading. For further comparison, Au nanoparticles were first prepared by reduction of HAuCl4 with citrate and NaBH4, and the presynthesized Au nanoparticles were assembled with CS into {CS/AuNPs}n layer-by-layer films, followed by Mb loading from Mb solution at pH 7.0, designated as {CS/ AuNPs}n-Mb. The {CS/AuNPs}n-Mb films also showed a CV peak pair at about -0.35 V, but with much smaller peak heights than those of {CS/SiO2}5-Mb-Au films under the same condition (Supporting Information Figure S4, curve e). 5. Influence of Permeability of Films. The porosity or permeability of layer-by-layer films is of great importance in protein loading.22-24 To investigate the permeability of {CS/ SiO2}5 films, two different types of layer-by-layer films, {PAMAM/SiO2}5 and {PDDA/PSS}5 films, were assembled on the electrode surface for comparison. The dendrimer PAMAM is an organic nanoparticle with a three-dimensional and spherical structure and good biocompatibility.51,52 The sixth-generation PAMAM with its diameter of 7 nm has a pKa value at about 9.4 for its protonated surface amino groups53 and thus possesses positive charges at pH 7.0. They were expected to be assembled with negatively charged SiO2 nanoparticles through electrostatic interaction, forming porous {PAMAM/SiO2}5 multilayer films. In contrast, the soft and oppositely charged polyelectrolytes PDDA and PSS were expected to be assembled into more dense and less porous {PDDA/PSS}5 layer-by-layer films. The perme-

Enhancement of Au Nanoparticles

J. Phys. Chem. B, Vol. 112, No. 48, 2008 15517

Figure 4. (A) CVs at 0.2 V s-1 for (a) {CS/SiO2}5-Mb-Au, (b) {CS/SiO2}5-Mb, (c) {CS/SiO2}5-Au, and (d) {CS/SiO2}5 films in 10 mL of pH 7.0 buffers after 40 mL of air was injected. (B) Amperometric responses of (a) {CS/SiO2}5-Mb-Au, (b) {CS/SiO2}5-Mb, and (c) {CS/SiO2}5-Au films at -0.1 V in pH 7.0 buffers with an increment of 5 µM H2O2 every 40 s.

Figure 5. Dependence of reduction peak current (Ipc) of {CS/ SiO2}5-Mb-Au films at 0.2 V s-1 in pH 7.0 buffers on immersion time (t) for {CS/SiO2}5 films in the mixture solution containing HAuCl4 and Mb at pH 7.0.

ability of these three films was investigated by CV with small molecular Fe(CN)63- as an electroactive probe in solution (Supporting Information Figure S5). The CV response showed an obvious difference at the different film electrodes with the sequence of {PAMAM/SiO2}5 > {CS/SiO2}5 > {PDDA/PSS}5 in reduction peak current. As expected, the {PAMAM/SiO2}5 films demonstrated the highest porosity since both PAMAM and SiO2 nanoparticles were rigid and hard spheres, and the {PDDA/PSS}5 films assembled with two soft and flexible polyelectrolytes displayed the poorest permeability, while the porosity of {CS/SiO2}5 films assembled with one polyelectrolyte and one nanoparticle was in between. Mb and HAuCl4 could also be loaded into {PAMAM/SiO2}5 and {PDDA/PSS}5 films. After the electrodeposition with the same procedure, the {PAMAM/SiO2}5-Mb-Au and {PDDA/ PSS}5-Mb-Au films were formed, respectively. Under the same conditions, the CV peak current or Γ* value showed the sequence of {PAMAM/SiO2}5-Mb-Au > {CS/SiO2}5-MbAu > {PDDA/PSS}5-Mb-Au (Figure 6A), in accordance with the film permeability. In addition, Mb-loaded films without Au nanoparticles inside displayed the same sequence in Γ* value (Figure 6B). All these results suggest that the porosity or permeability of multilayer films plays a key role in protein incorporation. The films with more porous structure tend to load more Mb and show larger Γ* value, as also observed previously.22-24 When the pore size of films is large enough for proteins, the film thickness may also influence the loading of proteins, and generally the thicker films would load more proteins. However, in this study, with only a few number of bilayers (n ) 5) for the films, the effect of film thickness would be very limited, and the influence of film porosity or permeability on Mb loading would be predominant. The {PAMAM/ SiO2}5-Mb-Au films also demonstrated much higher Γ* value than the {PAMAM/SiO2}5-Mb films (Figure 6B), further confirming that the Au nanoparticles formed inside the films

by incorporation of HAuCl4 and following electrodeposition can greatly enhance the electroactivity of Mb in the films. The permeability of three types of Mb-Au loaded films was also examined by electrochemical impedance spectroscopy (EIS) with Fe(CN)63-/4- as the redox probe at its formal potential of 0.17 V vs SCE with frequencies from 100 kHz to 100 mHz using a 10 mV peak-to-peak sinusoidal perturbation (Figure 7A). The EIS response in the form of Nyquist diagram for all three films showed a semicircle in high-frequency range. The diameter of the semicircle usually equals the charge transfer resistance (Rct) of the probe in electron transfer.54 For the film system, Rct mainly reflects the restricted diffusion of the probe through the film phase and relates directly to the accessibility of the underlying electrode or the film permeability.55,56 The Rct values were estimated by using the Randles equivalent circuit57 as the model and showed a sequence of {PAMAM/SiO2}5-Mb-Au (745 Ω) < {CS/SiO2}5-Mb-Au (1160 Ω) < {PDDA/ PSS}5-Mb-Au (6169 Ω), suggesting that the permeability of the loaded films keeps the same sequence as the unloaded films. In addition, the Rct value for {CS/SiO2}5-Mb-Au films was 1160 Ω, larger than 682 Ω for {CS/SiO2}5-Mb films (Figure 7B). 4. Discussion Mb in the {CS/SiO2}5-Mb-Au films shows much better electrochemical response (Figure 3) and electrocatalytic performance (Figure 4) than that in {CS/SiO2}5-Mb films, indicating that Au nanoparticles presented inside {CS/SiO2}5Mb–Au films play a key role in enhancing the Mb electrochemistry. However, the existence of Au nanoparticles in the films seems unable to increase the apparent heterogeneous electron transfer rate constant (ks) of Mb. The ks values for {CS/ SiO2}5-Mb-Au and {CS/SiO2}5-Mb films estimated by SWV are 28.8 and 38.6 s-1, respectively. Considering the estimation error and the limitation of the method, these two ks values are believed to have no substantial difference. It is thus difficult to conclude that the electron transfer of Mb for {CS/SiO2}5-Mb film system is obviously faster than that for {CS/ SiO2}5-Mb-Au films. This is further confirmed by the CV peak separation (∆Ep ) Epa - Epc) data. For {CS/SiO2}5-Mb-Au films, ∆Ep is about 65 mV at 0.2 V s-1, a little bit smaller than 75 mV for {CS/SiO2}5-Mb films. In general, ∆Ep reflects the reversibility of a redox system and is also an indicator of electron transfer rate.58 All these results suggest that the formation of Au nanoparticles in the {CS/SiO2}5-Mb-Au films does not increase the electron transfer rate of Mb in the films significantly. Therefore, the better electrochemical activity of Mb in {CS/ SiO2}5-Mb-Au films is most probably attributed to the larger amount of electroactive Mb in the films. In viscous film phase,

15518 J. Phys. Chem. B, Vol. 112, No. 48, 2008

Guo et al.

Figure 6. (A) CVs at 0.2 V s-1 in pH 7.0 buffers for (a) {PAMAM/SiO2}5-Mb-Au, (b) {CS/SiO2}5-Mb-Au, and (c) {PDDA/PSS}5-Mb-Au films after the {PAMAM/SiO2}5, {CS/SiO2}5, and {PDDA/PSS}5 films were immersed in the mixture solution of HAuCl4 and Mb at pH 7.0 for 10 h, respectively, followed by electrodeposition. (B) The surface concentration of electroactive Mb (Γ*) measured by CV in pH 7.0 buffers at 0.2 V s-1 for Mb loaded films (filled column) and Mb-Au loaded films (blank column): (a) {PAMAM/SiO2}5, (b) {CS/SiO2}5, and (c) {PDDA/PSS}5 films.

Figure 7. (A) EIS results for (a) {PAMAM/SiO2}5-Mb-Au, (b) {CS/SiO2}5-Mb-Au, and (c) {PDDA/PSS}5-Mb-Au films in 5 mM Fe(CN)63-/4solution. (B) EIS results for (a) {CS/SiO2}5-Mb-Au and (b) {CS/SiO2}5-Mb films in 5 mM Fe(CN)63-/4- solution.

it is impossible for macromolecular Mb to move to the electrode surface through physical diffusion and then exchange electrons with electrodes during the period of time of CV scan, since the diffusion process in the films is very slow and needs at least a few hours (Figure 5) while the CV scan usually takes only a few minutes. It is thus most probable for Mb to take the electron hopping or electron self-exchange mechanism in electron transfer.14,59 That is, while Mb molecules in the films do not move or displace significantly from their original positions, two neighboring Mb molecules in the films can exchange electrons with each other, and the electron transfer of Mb with electrodes can then be extended sequentially from electrode surface to the Mb molecules in the outer layers by successive “electron hopping” among these neighboring Mb molecules. This mechanism needs Mb molecules to have a certain degree of flexibility so that the neighboring Mb molecules can take the appropriate “face-to-face” orientation and exchange electron through their exposed heme edge. However, the electron hopping requires that the distance between neighboring Mb molecules is small enough.58 This is why only a fraction of Mb molecules in the films show electroactive because quite large amounts of Mb molecules in the films have too large distance with their neighboring Mb. For {CS/SiO2}5-Mb-Au films, however, conductive Au nanoparticles formed inside the films may act as electron bridges between those neighboring Mb molecules with relatively large distance, making the electron hopping between these Mb molecules become possible. Since some originally “inactive” Mb molecules become electrochemically “active” with the help of conductive Au nanoparticles, the {CS/ SiO2}5-Mb-Au films show much higher Γ* value than {CS/ SiO2}5-Mb films. However, the presynthesized Au nanoparticles seem not as effective as those prepared in situ by electrodeposition in enhancement of Mb electrochemistry, which is reflected by the

much smaller CV response of {CS/AuNPs}n-Mb films than that of {CS/SiO2}5-Mb-Au films (Supporting Information Figure S4). The exact reason for this is not yet clear, but it is probably related to the better proximity between Au nanoparticles and Mb molecules in the latter films since efficient electron hopping between Au nanoparticles and Mb molecules also requires very close distances between them. The possible difference in loading amount and distribution of Mb for these two films may also have some influence on the CV responses. For preparation of {CS/SiO2}5-Mb-Au films, both HAuCl4 and Mb in their mixture solution are loaded into {CS/SiO2}5 films simultaneously and uniformly distributed in the films. A lot of very small Au(III) ions must be located around each Mb molecule in close proximity. After electrodeposition, some Au nanoparticles synthesized in situ should thus be very close to Mb molecules, making more Mb become electroactive in comparison with {CS/AuNPs}n-Mb films. The loading mode or sequence of Au(III) ions and Mb into {CS/SiO2}5 films also influences the subsequent CV response of Mb in the corresponding films. Both {CS/SiO2}5-Mb+Au and {CS/SiO2}5-Au+Mb films with separate loading of Au(III) ions and Mb demonstrate smaller CV response of Mb than {CS/ SiO2}5-Mb-Au films with simultaneous loading of HAuCl4 and Mb followed by electrodeposition (Supporting Information Figure S4). This is probably related to the favorable orientation of Mb molecules toward Au nanoparticles in {CS/SiO2}5Mb-Au films. The simultaneous loading may make uniform distribution of HAuCl4 and Mb at molecular level in the films, and some small Au(III) ions must be located in the proximity of the Mb cleft, where the electroactive heme prosthetic group resides. After electrodeposition, these Au(III) ions are transferred into Au nanoparticles, which may keep their proximity to the heme redox center of Mb. The cleft region of Mb is surrounded by positively charged amino acid groups,60 which is also

Enhancement of Au Nanoparticles beneficial to attract negatively charged Au nanoparticles in this region. These Au nanoparticles in the vicinity of the heme group of Mb may act as conducting nanowires or electron bridges in electron hopping and play a central role in enhancing electron transfer of Mb. For {CS/SiO2}5-Au+Mb films, however, Mb molecules are loaded into the films after Au nanoparticles are formed. In this case, it may be difficult for the large Au nanoparticles in the films to approach the active center of Mb, thus resulting in the smaller CV response of Mb. A similar explanation may also be applied to the smaller CV response of {CS/SiO2}5-Mb+Au films. While further experiments are needed to support the above speculation, the present results show that the loading mode or sequence of Au(III) ions and Mb into {CS/SiO2}5 films, and the subsequent spatial arrangement or orientation between Au nanoparticles and Mb is important in realizing the direct electron transfer of Mb in the films. The permeability or porosity of layer-by-layer films also plays an important role in Mb loading.22-24 The comparative experiments show that different types of multilayer films have different porosity, and the more porous the films are, the more amount of Mb can be loaded (Figure 6, Supporting Information Figure S5). Theoretically, the electron hopping between Mb molecules in the films can be extended to an infinite distance from electrode surface to the outer layers of films. However, the electroactivity of Mb in the {CS/SiO2}5-Mb-Au films can only be extended to a very limited number of bilayers even with the help of Au nanoparticles (Supporting Information Figure S2A). This can be explained by charge compensation in electron hopping. In order to compensate the charge alteration caused by electron transfer and keep the electroneutralisation of the films, the transportation of small counterions between external solution and film phase is not only necessary but may also become a limiting factor.59 It is the limited mobility of small counterions in the Mb films that restricts the electroactivity of Mb to a very limited number of bilayers. The mobility of counterions within the film phase depends on the porosity or permeability of the films. The permeability of Mb-loaded films measured by EIS with electroactive probe shows the sequence of {PAMAM/ SiO2}5-Mb-Au > {CS/SiO2}5-Mb-Au > {PDDA/PSS}5Mb-Au (Figure 7A), in good agreement with the sequence of CV response of Mb (Figure 6). This suggests that even with Au nanoparticles inside the films the permeability of Mb-loaded films can greatly influence the electrochemical activity of Mb in the films. For the same {CS/SiO2}5 film system, {CS/SiO2}5-Mb films show smaller Rct value than {CS/SiO2}5-Mb-Au films (Figure 7B), suggesting that the permeability of the former is better than the latter. Considering that the difference in Rct between these two films is not significant but the difference in Γ* is quite remarkable (Figures 3 and 6B), it would be reasonable to conclude that while both factors of Au nanoparticles synthesized in situ in {CS/SiO2}5-Mb-Au films and the film permeability influence the electrochemical activity of Mb in the films, the Au nanoparticles seem play a more important role. Conclusion In this work, the gold nanoparticles with a size of less than 20 nm are electrochemically deposited in situ in {CS/ SiO2}5-Mb-Au films and are most probably distributed uniformly throughout the films and located in close vicinity to Mb molecules. This makes some of the Au nanoparticles in the films become electron bridges between neighboring Mb molecules, and more Mb becomes electroactive through the electron hopping mechanism. The {CS/SiO2}5-Mb-Au films thus

J. Phys. Chem. B, Vol. 112, No. 48, 2008 15519 demonstrate much better behavior in electrochemistry and electrocatalysis for Mb than the corresponding {CS/SiO2}5-Mb films with no Au nanoparticles inside. The comparative studies show that the Au nanoparticles electrodeposited in situ in the films perform much better than that presynthesized in enhancing Mb electrochemistry, and the loading sequence of HAuCl4 and Mb into {CS/SiO2}5 films also influences the CV response of Mb significantly. All these suggest that the relative distance and orientation between Au nanoparticles and Mb molecules in the films play a key role in improving electron hopping among Mb molecules. Moreover, film permeability or porosity is also an important factor in realizing the direct electrochemistry of Mb in film phase. The understanding of the electron transfer mechanism of redox proteins in films, especially the role of Au nanoparticles in the enhancement of protein electrochemistry, will open a new way to develop the third generation of electrochemical biosensors based on the direct electrochemistry of redox enzymes without using mediators. Acknowledgment. The financial support from the National Natural Science Foundation of China (NSFC 20775009 and 20475008) is acknowledged. Supporting Information Available: Five figures showing the SEM top views of {CS/SiO2}n-Mb films, the influence of the number of bilayers, the concentration of HAuCl4, and the pH of loading solution on Γ* for {CS/SiO2}n-Mb-Au and {CS/ SiO2}n-Mb films, CVs of five different Mb films, and CVs of Fe(CN)63- at {PAMAM/SiO2}5, {CS/SiO2}5, and {PDDA/PSS}5 film electrodes. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Chaplin, M. F.; Bucke, C. Enzyme Technology; Cambridge University Press: Cambridge, UK,1990. (2) Armstrong, F. A.; Wilson, G. S. Electrochim. Acta 2000, 45, 2623. (3) Rusling, J. F.; Zhang, Z. In Handbook of Surfaces and Interfaces of Materials; Nalwa, R. W., Ed.; Academic Press: San Diego, CA, 2001; Vol. 5, p 33. (4) Hu, N. Pure Appl. Chem. 2001, 73, 1979. (5) Brown, K. R.; Fox, A. P.; Natan, M. J. J. Am. Chem. Soc. 1996, 118, 1154. (6) Willner, I.; Willner, B.; Katz, E. Bioelectrochemistry 2007, 70, 2. (7) Guo, S.; Wang, E. Anal. Chim. Acta 2007, 598, 181. (8) Jensen, P. S.; Chi, Q.; Grumsen, F. B.; Abad, J. M.; Horsewell, A.; Schiffrin, D. J.; Ulstrup, J. J. Phys. Chem. C 2007, 111, 6124. (9) Decher, G. Science 1997, 277, 1232. (10) Decher, G.; Hong, J. D.; Schmitt, J. Thin Solid Films 1992, 210/ 211, 831. (11) Lvov, Y. M.; Lu, Z.; Schenkman, J. B.; Zu, X.; Rusling, J. F. J. Am. Chem. Soc. 1998, 120, 4073. (12) Zhou, Y.; Hu, N.; Zeng, Y.; Rusling, J. F. Langmuir 2002, 18, 8573. (13) Ma, H.; Hu, N.; Rusling, J. F. Langmuir 2000, 16, 4969. (14) Beissenhirtz, M. K.; Scheller, F. W.; Stocklein, W. F. M.; Kurth, D. G.; Mohwald, H.; Lisdat, F. Angew. Chem., Int. Ed. 2004, 43, 4357. (15) Zhang, H.; Lu, H.; Hu, N. J. Phys. Chem. B 2006, 110, 2171. (16) Zhang, H.; Hu, N. Biosens. Bioelectron. 2007, 23, 393. (17) Zhang, H.; Hu, N. J. Phys. Chem. B 2007, 111, 10583. (18) Salloum, D. S.; Schlenoff, J. B. Biomacromolecules 2004, 5, 1089. (19) Weidinger, I. M.; Murgida, D. H.; Dong, W.; Mo¨hwald, H.; Hildebrandt, P. J. Phys. Chem. B 2006, 110, 522. (20) Mckenzie, K.; Marken, F. Langmuir 2003, 19, 4327. (21) Zhang, S.; Yang, W.; Niu, Y.; Li, Y.; Zhang, M.; Sun, C. Anal. Bioanal. Chem. 2006, 384, 736. (22) Lu, H.; Hu, N. J. Phys. Chem. B 2006, 110, 23710. (23) Wang, G.; Liu, Y.; Hu, N. Electrochim. Acta 2007, 53, 2071. (24) Guo, X.; Zhang, H.; Hu, F. Nanotechnology 2008, 19, 055709. (25) Hu, Y.; Hu, N. J. Phys. Chem. B 2008, 112, 9523. (26) Shi, X.; Shen, M.; Mo¨hwald, H. Prog. Polym. Sci. 2004, 29, 987. (27) Joly, S.; Kane, R.; Radzilowski, L.; Wang, T.; Wu, A.; Cohen, R. E.; Rubner, M. F. Langmuir 2000, 16, 1354.

15520 J. Phys. Chem. B, Vol. 112, No. 48, 2008 (28) Wang, T. C.; Rubner, M. F.; Cohen, R. E. Langmuir 2002, 18, 3370. (29) Dai, J.; Bruening, M. L. Nano Lett. 2002, 2, 497. (30) Zhang, X.; Shi, F.; Liu, H.; Fu, Y.; Wang, Z.; Jiang, L.; Li, X. J. Am. Chem. Soc. 2004, 126, 3064. (31) Shi, F.; Wang, Z.; Zhang, X. AdV. Mater. 2005, 17, 1005. (32) Huang, M.; Shen, Y.; Cheng, W.; Shao, Y.; Sun, X.; Liu, B.; Dong, S. Anal. Chim. Acta 2005, 535, 15. (33) Huang, H.; Hu, N.; Zeng, Y.; Zhou, G. Anal. Biochem. 2002, 308, 141. (34) dos Santos, D. S., Jr.; Riul, A., Jr.; Malmegrin, R. R.; Fonseca, F. J.; Oliveira, O. N., Jr.; Mattoso, L. H. C. Macromol. Biosci. 2003, 3, 591. (35) He, P.; Hu, N.; Rusling, J. F. Langmuir 2004, 20, 722. (36) He, P.; Hu, N. Electroanalysis 2004, 16, 1122. (37) Denuziere, A.; Ferriera, D.; Domard, A. Carbohydr. Polym. 1996, 29, 317. (38) Rinaudo, M.; Milas, M.; Le Dung, P. Int. J. Biol. Macromol. 1993, 15, 281. (39) Franks, G. V. J. Colloid Interface Sci. 2002, 249, 44. (40) Fisher, M. L.; Colic, M.; Rao, M. P.; Lange, F. F. J. Am. Ceram. Soc. 2001, 84, 713. (41) Finot, M. O.; Braybrook, G. D.; McDermott, M. T. J. Electroanal. Chem. 1999, 466, 234. (42) Qian, L.; Yang, X. J. Phys. Chem. B 2006, 110, 16672. (43) Wang, B.; Chen, K.; Jiang, S.; Reincke, F.; Tong, W.; Wang, D.; Gao, C. Biomacromolecules 2006, 7, 1203. (44) Rusling, J. F. Acc. Chem. Res. 1998, 31, 363.

Guo et al. (45) Murray, R. W. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1984; Vol. 13, p 191. (46) O’Dea, J. J.; Osteryoung, J. G. Anal. Chem. 1993, 65, 3090. (47) Nassar, A.-E. F.; Zhang, Z.; Hu, N.; Rusling, J. F.; Kumosinski, T. F. J. Phys. Chem. B 1997, 101, 2224. (48) Zhang, Z.; Rusling, J. F. Biophys. Chem. 1997, 63, 133. (49) El-Deab, M. S.; Ohsaka, T. Electrochem. Commun. 2002, 4, 288. (50) El-Deab, M. S.; Sotomura, T.; Ohsaka, T. Electrochem. Commun. 2005, 7, 29. (51) Ottaviani, M. F.; Cossu, E.; Turro, N.; Tomalia, D. A. J. Am. Chem. Soc. 1995, 117, 4387. (52) Patri, A. K.; Majoros, I. J.; Baker, J. R., Jr. Curr. Opin. Chem. Biol. 2002, 6, 466. (53) van Duijvenbode, R. C.; Borkovec, M.; Koper, G. J. M. Polymer 1998, 39, 2657. (54) Katz, E.; Willner, I. Electroanalysis 2003, 15, 913. (55) Sabatani, E.; Rubinstein, I. J. Phys. Chem. 1987, 91, 6663. (56) Sabatani, E.; Cohen-Boulakia, J.; Bruening, M.; Rubinstein, I. Langmuir 1993, 9, 2974. (57) Randles, J. E. B. Discuss. Faraday Soc. 1947, 1, 11. (58) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamental and Applications, 2nd ed.; Wiley: New York, 2001. (59) Majda, M. In Molecular Design of Electrode Surfaces; Murray R. W., Ed.; Wiley: New York, 1992; p 159. (60) Protein Data Bank at http://www.rcsb.org/pdb.

JP807452Z