Layer-by-Layer Polypeptide Macromolecular Assemblies-Mediated

Feb 14, 2011 - The Journal of Physical Chemistry C 2013 117 (50), 26562-26572 ... Yung-Lun Lee , Ting-Xuan Lin , Feng-Ming Hsu , Jeng-Shiung Jan...
1 downloads 0 Views 2MB Size
ARTICLE pubs.acs.org/Langmuir

Layer-by-Layer Polypeptide Macromolecular Assemblies-Mediated Synthesis of Mesoporous Silica and Gold Nanoparticle/Mesoporous Silica Tubular Nanostructures Jeng-Shiung Jan,* Tzu-Han Chuang, Po-Jui Chen, and Hsisheng Teng Department of Chemical Engineering, National Cheng Kung University, No. 1, University Rd., Tainan, Taiwan 70101, Taiwan

bS Supporting Information ABSTRACT: A simple and versatile approach is proposed to use the LbL-assembled polypeptide macromolecular assemblies as mediating agents and templates for directed growth of gold nanoparticles and biomimetic silica mineralization, allowing the synthesis of polypeptide/silica and polypeptide/gold nanoparticle/silica composite materials, as well as mesoporous silica (meso-SiO2) and gold nanoparticle/mesoporous silica (Au NP/meso-SiO2). The formation of tubular nanostructures was demonstrated by silicification and growth of gold nanoparticles within macromolecular assemblies formed by poly(L-lysine) (PLL) and poly(L-glutamic acid) (PLGA) using polycarbonate membranes as templates. The experimental data revealed that the silicified macromolecular assemblies adopted mainly sheet/ turn conformation. The as-prepared mesoporous silica materials possessed well-defined tubular structures with pore size and porosity depending on the size of sheet/turn aggregates, which is a function of the molecular weight of polypeptides. The directed growth of Au NP and subsequent silica mineralization in the macromolecular assembly resulted in Au NP/meso-SiO2 tubes with uniform nanoparticle size and the as-prepared materials exhibited promising catalytic activity toward the reduction of p-nitrophenol. This approach provides a facile and general method to synthesize organic-inorganic composite materials, oxide and metal-oxide nanomaterials with different compositions and structures.

’ INTRODUCTION Nature has already developed the ability to fabricate materials such as diatoms and sponges with precisely controlled nanostructures and morphologies that exceed human engineering capabilities.1-6 Inspired by these silicification processes, the biomimetic synthesis of silica materials has significant advances in the recent years. Compared with conventional syntheses of silica, the advantage of biomimetic synthesis includes benign conditions such as neutral pH, room temperature, and aqueous environment. The silicification processes of the biological systems leads to exquisite hierarchical structures and complex morphologies, which are mediated by proteins or peptides via self-assembly and template processes.5,6 It is for this reason that nanoscale control of silica morphology and structure via template approaches has received growing attention. A variety of selfassembled structures including vesicles,7-9 micelles,10,11 platelets,12 nanofibers,13-17 and nanotapes18,19 was utilized for silica deposition. Micro- and nanostructured silicas were also synthesized using polypeptide chain conformation, polyamine-salt aggregates, and polyamine-nanoparticle aggregates as templates.20-24 Micro- and nanopatterned silica at surfaces have been demonstrated using amine-containing templates by lithography, microcontact printing, and grafting technologies.25-30 Synthetic amine-containing macromolecules were commonly used to mediate the formation of nanoscale silica. The biomimetic synthesis r 2011 American Chemical Society

of silica mediated by amine-containing macromolecules is still an important subject of study, which will continuously be pursued. The synthesis of nanostructured materials using the synergy between biomimetic mineralization and other techniques may open up new possibilities for these materials in a range of fields including catalysis, separations, gene and drug delivery, encapsulation, and so forth. LbL assembly technique is a simple and versatile approach to deposit a variety of materials on different substrates. Using this technique, free-standing structures such as films, tubes, and capsules can be obtained by depositing desired materials and subsequent removing templates.31 One approach combining LbL assembly technique and biomimetic mineralization has been utilized to prepare organic-inorganic composite microcapsules under benign conditions for cell and enzyme encapsulation.32-34 Unlike previously described composite LbL multilayer films composed of preprepared inorganic particles,35-37 the deposited inorganic layer formed by the this approach is continuous and intact, which is important for many applications such as encapsulation. Recently, several groups reported that some biomolecules (e.g., L-tyrosine and tyrosine-containing peptides) can serve both Received: March 29, 2010 Revised: December 31, 2010 Published: February 14, 2011 2834

dx.doi.org/10.1021/la103923c | Langmuir 2011, 27, 2834–2843

Langmuir as stabilizing and reducing agents to mediate the growth of metal nanoparticles in alkaline solutions.38-40 Peptide-mediated synthesis was also applied to obtain gold and silver nanoparticles on or within polyelectrolyte multilayers, organogel networks, and silk fibers.41-44 This facilitates the incorporation of functional nanoparticles into the as-prepared materials through one-step directed growth. Taking the advantage of LbL self-assembly technique and peptide-mediated synthesis of hard materials, it is a possible strategy to prepare inorganic materials with different morphologies and to incorporate additional functionality such as porosity and functional nanoparticle into the as-prepared materials, which will be essential for many applications. We have demonstrated using self-assembled polypeptides to direct the formation of inorganic materials,8,9,45 which have heightened our interest in the synthesis of complex inorganic materials using biological macromolecules as templates. Particularly relevant to this work, however, only a few studies reported the templated synthesis of porous silica under benign conditions.9,24 In this paper, we report on a polypeptide-mediated strategy that utilizes surface macromolecular assemblies to synthesize mesoporous silica through silicification under benign conditions, as well as the gold nanoparticle/mesoporous silica through directed growth of gold nanoparticles and subsequent silicification. Layer-by-layer (LbL) self-assembly technique facilitates the buildup of preorganized molecular templates with tailored structure and composition as well as the wall thickness to nanometer scale. To demonstrate the feasibility of this strategy, the synthetic cationic and anionic polypeptides were alternatively deposited on the walls of polycarbonate (PC) membrane pores via LbL assembly for mediating the formation of gold nanoparticles and silicas. The synthetic cationic and anionic polypeptides, poly(L-lysine) (PLL) and poly(L-glutamic acid) (PLGA), serve as silica mineralizing and gold reducing agent, respectively. The LbL assembled polypeptide macromolecular assemblies that possess molecular organizations (i.e., βsheet) can serve as templates to control silica nanostructures and confine the growth of gold nanoparticles. The structure and functionality of as-synthesized materials are determined by the molecular organization, size, shape, and morphology of the dual templates (i.e., macromolecular assembly and PC membrane) used.

’ EXPERIMENTAL SECTION Materials. PC membranes (product catalogue No.: HTTP04700, 47 mm diameter) with a pore diameter of 0.4 μm and a membrane thickness about 10-20 μm were obtained from Millipore. THF (ACS Reagent, Merck) and Diethyl ether (Anhydrous, ACS Reagent, J. T. Backer) were dried using Na metal. Hexane (ACS Reagent, EM Science) was dried using calcium hydride. The amino acids used in this work NεZ-L-lysine (∼99%, Z: carboxybenzyl) and γ-benzyl-L-glutamic acid (>99%) were used as received from Aldrich. Bis(1,5-cyclooctadiene)nickel(0) (98þ%), 2,20 -bipyridyl (99þ%,), hydrogen tetrachloroaurate(III) trihydrate (ACS, 99.99%), and p-nitrophenol (spectrophotometric grade) were used as received from Sigma-Aldrich. Iodotrimethylsilane (Me3SiI, 97%, saturated with Copper) and trifluoroacetic acid (99%) were supplied by Alfa Aesar. Triphosgene (98%, Merck), tetramethyl orthosilicate (99%), hydrogen bromide (33 wt % in acetic acid), and NaBH4 (96%) were used as received from Fluka. Polypeptide Synthesis. The polypeptide synthesis was performed using the zerovalent nickel initiator 2,20 -bipyridyl-Ni(1,5-cyclooctadiene) (BpyNiCOD) to polymerize Nε-Z-L-lysine and γ-benzyl-L-glutamic acid NCAs by following the literature reported

ARTICLE

procedures.46-48 Poly(Z-L-lysine) (PZLL) and poly(γ-benzyl-Lglutamic) (PBLG) acid were deprotected by using HBr and Me3SiI, respectively. The notations for poly(L-lysine) (PLL) and poly(L-glutamic acid) (PLGA) used throughout are Lysm and Glun, respectively, where m and n are the number of amino acids in one chain. PLL with different chain lengths used in this study were Lys145 (Mn = 37800, Mw/ Mn = 1.14), Lys210 (Mn = 542000, Mw/Mn = 1.02), and Lys340 (Mn = 89200, Mw/Mn = 1.10), respectively. And PLGA with different chain lengths were Glu125 (Mn = 27100, Mw/Mn = 1.19), Glu190 (Mn = 42000, Mw/Mn = 1.02), and Glu370 (Mn = 81000, Mw/Mn = 1.11), respectively. The polypeptides (PZLL and PBLG) with different molecular weights were synthesized and characterized by GPC (Figure S1 and S2, Supporting Information). NMR measurements confirmed that the polypeptides (PZLL, PBLG, PLL, and PLGA) were prepared (Figure S3, Supporting Information). PLL/PLGA Multilayer Assembly. Method I. The PLL-coated membranes were first prepared by immersing the PC membranes into a PLL solution (1 mg/mL in 0.5 M NaCl), followed by immediately sonicating for 2 min and allowing 10 min for adsorption. After thoroughly rinsing with 0.5 M NaCl aqueous solution, negatively charged PLGA was then adsorbed by immersing the PLL-coating membranes into a PLGA solution (1 mg/mL in 0.5 M NaCl) using the same procedure. The desired number of layers was deposited by cyclic adsorption of PLL and PLGA. Method II. The PLL and PLGA solutions were dissolved in sodium phosphate buffer (pH 7.4, Pierce) for depositing desired number of layers. The samples were denoted as followed. For example, the sample, (Lys210/Glu190)9I, was obtaining from silicification of 9-layer Lys210/ Glu190 coated PC membrane (that is, (Lys210/Glu190)9) using LbL assembly method I. SiO2 Tube Formation. The PLL/PLGA multilayer coated membranes were then inserted in a freshly prepared 0.5 M orthosilicic acid solution for 10-12 h to allow precipitation of silica in the PLL/PLGA multilayer. After thoroughly rinsing with DI water and drying at room temperature, the film coating on the membrane surface was removed partially using fine sandpaper. The as-prepared silica/PLL/PLGA multilayer membranes were slowly heat to 95 °C for 24 h, following by removing the membrane using dichloromethane. Pure silica tubes were obtained by calcining the silica/PLL/PLGA composite tubes in air at 500 °C for 10 h. (heating rate 2 K min-1). Gold/Silica Tube Formation. The PLL/PLGA multilayer coated membranes were immersed in a HAuCl4 solution (1.5  10-4 M, 20 mL) at pH 7 (or 11.5) for 12 h, and the solution was adjusted to pH 7 (or 11.5) and 200 μL of Au ion solution (1.5  10-2 M) was added to the solution every 12 h. After the formation of Au NPs, the membranes were taken out from the solution and inserted in a freshly prepared 0.35 M orthosilicic acid solution for 6-12 h to allow precipitation of silica in the Au NP/PLL/ PLGA multilayer. After thoroughly rinsing with DI water and drying at room temperature, the as-prepared Au NP/meso-SiO2/PLL/PLGA multilayer membranes were slowly heat to 95 °C for 24 h and the PC membranes were removed by dissolving in dichloromethane. Pure Au NP/meso-SiO2 tubes were obtained by calcining at 500 °C for 10 h in air (heating rate 2 K min-1). Gold/Silica Catalytic Tests. First 1 mg of Au NP/meso-SiO2 catalyst was dispersed in 30 mL of DI water and the suspension was stirred for overnight. Then, 30 mL of 2  10-2 M NaBH4 and 30 mL of 4  10-4 M p-nitrophenol aqueous solution were then added in the suspension and the catalytic activity was evaluated as the percentage disappearance of p-nitrophenol at a wavelength of 400 nm using UV-vis spectrophotometer. Instrumentation and Characterization. Gel permeation chromatography (GPC) measurements were performed at 55 °C before deprotection of the polypeptide using a Viscotek system equipped with three detectors, which are RI (VE3580, Viscotek), right angle light 2835

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir

ARTICLE

Scheme 1. Procedure Used for Preparing Mesoporous Silica and Au NP/Meso-Silica Tubes

scattering, and viscometer (Dual 270, Viscotek). Two ViscoGEL I-Series columns (catalog numbers I-MBHMW-J012906 and I-MBLMWH110211, Viscotek) were used for efficient separation using 0.1 M LiBr in DMF as eluent. The eluent flow rate was 1 mL/min. 1H NMR spectra were recorded at 300 MHz on a Mercury 300 Varian spectrometer using d-TFA as solvent. Field-Emission Scanning Electron Microscopy (FESEM) and energy dispersive X-ray (EDX) measurements were performed using a JEOL JSM-6700F microscope operating at 1-10 kV. Samples were collected via centrifugation, air-dried, and mounted on carbon tape for imaging. Transmission electron microscopy (TEM) measurements were performed on a Hitachi H7500 microscope with a Tungsten lamp and an excitation voltage of 120 kV. The samples were dispersed in methanol (100%, Aldrich) and placed on a 400-mesh copper grid. Infrared spectroscopy was performed on a Thermo Nicolet Nexus 670 FTIR. Background spectra were collected after 10 min of evacuation. A powder mixture of mass ratio 0.01 sample: 0.99 potassium bromide (Aldrich) was pelletized and analyzed after 10 min of evacuation. Nitrogen adsorption measurements were performed using a Micromeritics 2010 ASAP instrument at 77 K. Surface areas were calculated by the Brunauer-Emmett-Teller (BET) method. Pore volumes and pore size distributions were determined from nitrogen adsorption isotherm data using the t-plot and Barrett-Joyner-Halenda (BJH) method. Thermal gravimetric analyses (TGA) were performed using a TGA7 Instrument from Perkin-Elmer over a temperature range of 25 to 800 °C using oxygen as a carrier gas and temperature ramping rate of 5 °C min-1. The UV-vis measurements were carried out on SCINCO S-3100 UV-vis spectrophotometer and the UV-ATR measurements were performed on JASCO V-670 UV-vis spectrophotometer.

’ RESULTS AND DISCUSSION Mesoporous tubular silicas with tunable pore size and porosity were synthesized by controlling the preorganized molecular templates, which are the polypeptide macromolecular assemblies. Different molecular weights of poly(L-lysine) (PLL) and poly(L-glutamic acid) (PLGA) were selected to deposit on the walls of membrane pores. The PC membranes and LbLassembled PLL/PLGA multilayer films on walls of membrane

pores were used as sacrificial templates for the synthesis of silica hollow tubes. As shown in Scheme 1, the PLL/PLGA multilayer films were coated on porous PC membrane via LbL method. Then the PLL/PLGA-coated PC membrane was immersed in a freshly prepared orthosilicic acid solution (0.5 M, pH 3.8) for 10-12 h. The silicified PLL/PLGA-coated membranes were heated to 95 °C for 24 h and the silica/PLL/PLGA composite tubes were obtained by dissolving PC membranes. Then the composite tubes were calcined at elevated temperature to remove organic materials. As a control experiment, porous PC membranes without deposited PLL/PLGA multilayer films were immersed in a freshly prepared orthosilicic acid solution for 12 h and no silica deposition on the bare PC membranes. It is known that PLL can induce polymerization of orthosilicic acid due to electrostatic interactions or hydrogen bonding between the amine group on the polymer chain and orthosilicic acid. Therefore, it is expected that the deposited PLL/PLGA multilayer films can induce polymerization of orthosilicic acid as well and the precipitated silicas can translate the molecular organization. PLL and PLGA were deposited on the walls of membrane pores using two methods. For method I, PLL and PLGA were separately dissolved in 0.5 M NaCl aqueous solution for the polypeptide film assembly. As a comparison to method I, phosphate buffer solution (pH 7.4) was used as solvent in method II. After coating polypeptide on the membrane pores, the LbL assembled PLL/PLGA tubes were obtained and TEM characterization reveals that the PLL/PLGA tubes with well-defined structures and highly uniform coatings were observed (for example, Figure S4, Supporting Information). FTIR spectra of the PLL/PLGA tubes made by using Method I and II show that amide I and amide II characteristic peaks are at 1626-28 and 1545 cm-1, respectively (Figure S5, Supporting Information). It indicates that the PLL/PLGA macromolecular assemblies are predominantly in β-sheet conformation. In addition, the shoulder at 1650 cm-1 suggests that random coil and/or R-helix conformations with lower relative percentage were also present in the macromolecular assemblies. Our results are consistent 2836

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir

Figure 1. Transmission FTIR spectra of silica/PLL/PLGA composite tubes. Samples are the silicified (A and B) (Lys145/Glu125)9, (C and D) (Lys210/Glu125)9, (E and F) (Lys340/Glu125)9, and (G and H) (Lys340/ Glu370)9 coated PC membrane using LbL assembly method (A, C, E, and G) I and (B, D, F, and H) II, respectively.

with previous studies reported by Boulmedais and co-workers.49,50 We have attempted to measure the secondary conformation adopted by the polypeptide macromolecular assemblies using circular dichroism (CD). However, the measurements were hampered by the poor solubility and dispersion of the polypeptide tubes in aqueous solution. The silicified PLL/PLGA tubes were characterized by FTIR and the spectra reveal that the vibrational bands of silica and the amide I and II bands of the polypeptide chains were observed, confirming the incorporation of silica in the polypeptide macromolecular assemblies (Figure 1). Field-emission scanning electron microscopy (FE-SEM) images show that the well-defined silica/PLL/PLGA composite tubular structures were obtained and the outer diameter of all hollow nanotubes is about 400 nm, corresponding to the pore diameter of the PC membrane (data not shown). From FTIR and SEM analysis, the results indicate that silicas precipitate in the polypeptide macromolecular assemblies. FTIR spectra of the composite tubes made by using methods I and II show that amide I and amide II characteristic peaks are at 1626-30 and 1545 cm-1, respectively (Figure 1). It suggests that the PLL/PLGA macromolecular assemblies in the composite tubes are predominantly in β-sheet conformation. In addition, the shoulder at 1658-61 cm-1 suggests that β-turn conformation is also present in the macromolecular assemblies. Circular-dichroism (CD) spectra of the composite tubes exhibit a minimum between 225 and 230 nm, suggesting the macromolecular

ARTICLE

assemblies adopted mainly β-sheet conformation (Figure S6, Supporting Information). It is known that the macromolecular assemblies constructed from polypeptides via LbL technique adopted secondary structures similar to those found in proteins.49,50 Several groups have reported the formation of PLL/ PLGA macromolecular assemblies on surfaces can undergo conformational transition through the NH3þ--OOC interaction and adopt secondary structures.49-52 In this study, the PLL/ PLGA macromolecular assemblies in the composite tubes adopting predominantly β-sheet conformation can be explained in the following. The PLL and PLGA chains are initially in coil conformation in 0.5 M NaCl solution or phosphate buffer solution. Once they deposit on PC porous membrane, PLL and PLGA chains form macromolecular assemblies through the NH3þ--OOC interaction which induces predominantly coilsheet conformational transition, consistent with previous work.49-52 After immersing the multilayer coated membrane in an orthosilicic acid solution, the conformational transition occurred and, as a result, the silicified multilayer films adopted predominantly β-sheet/turn conformations. Previously, we reported that the PLL block in the Lys-block-Gly vesicular assemblies underwent coil-helix conformational transition in the presence of silicic acid and phosphate ions, resulting in the synthesis of microporous silicas.9 In both of these two studies, the conformational transition can be attributed to the confined environment and solution condition. The amount of silica coated on one PC membrane (averaging over 20 membranes) were measured to be 1.1 ( 0.2, 1.8 ( 0.2, and 2.7 ( 0.2 mg for samples with 5, 9, and 13 PLL/PLGA coated layers, respectively. The amount of polypeptide and silica coated on one PC membrane increases as the number of coated layer increases. Thermal gravimetric analyses (TGA) profiles reveals a 29-35% total weight loss of dry mass for all of the samples, which was associated predominantly with the removal of polypeptides from the as-synthesized materials (Figure S7, Supporting Information). On the basis of TGA data, the amount of silica condensed onto the polypeptide coated membrane was found to be only proportional to that of the polypeptide coated onto the membrane pores, instead of that of the silica precursor for silicification (Figure S7, Supporting Information). The pH effect of the silica precursor solution on the silicification process was investigated. First, the pH of the orthosilicic acid solution was varied by changing the concentration. The pH values were measured to be 5.5, 4.6, and 3.8 for 0.1, 0.3, and 0.5 M orthosilicic acid solution, respectively. The PLL/PLGA multilayer coated membranes were immersed in different concentrations of orthosilicic acid (0.1-0.5 M, pH 5.5-3.8) with varying immersing times (8-20 h). TEM images show that there is no obvious difference in the wall thickness and silica nanostructure (Figure S8, Supporting Information). TGA profiles also show almost the same organic and inorganic weight ratio for the samples (data not shown). The results indicate that the well-defined silica/PLL/ PLGA composite tubes can be synthesized by varying the concentration (or the pH) of the precursor solution. Second, the pH of the precursor solution (0.5 M) was adjusted to 7.4 or higher by using 0.5 M NaOH solution and the PLL/PLGA multilayer coated membranes were then immersed in the resultant precursor solution at different pH. However, the macromolecular assemblies cannot be silicified using this approach and only a very few silica was deposited in the PLL/PLGA multilayer film. In order to analyze the silica nanostructures templated by the polypeptide macromolecular assemblies in the wall, the silicified PLL/PLGA hollow tubes were heated to 500 °C for 10 h and the 2837

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir

ARTICLE

Figure 2. FE-SEM (A, B) and TEM (C, D) images of silica tubes obtained from silicification of (A) (Lys210/Glu190)5, (B) (Lys210/Glu190)9, (C) (Lys340/Glu125)9, and (D) (Lys145/Glu125)9 coated PC membrane using LbL assembly method I after calcination.

organic compounds (PLL and PLGA) were burned off. FE-SEM images of all calcined silica/polypeptide tubes show that the welldefined tubular structures were obtained and the outer diameters of all silica hollow nanotubes are about 400 nm after heating at higher temperature, similar to the silica/PLL/PLGA composite tubes (for example, Figure 2, parts A and B). No obvious shrinkage or collapse is observed after calcination and the tube length corresponds to the thickness of the original PC membranes. TEM images further reveal that the silica hollow tubes are porous, but the porous structure do not possess long-rang order (Figure 2, parts C and D, and Figure S9, Supporting Information). From SEM and TEM analysis, the results indicate that a highly uniform and conformal coating was successfully achieved and the average wall thickness of as-synthesized silica tubes is dependent on the number of deposited layer linearly. The mean wall thickness of silica tubes is calculated to be about 9 ( 0.5 nm per PLL/PLGA layer. Previously, the thickness of the PLL/PLGA multilayer assembled at pH 8.4 in Mes-Tris buffer solution was measured with mean thickness about 5.2 nm per bilayer.49 In our study, it is reasonable to assume the thickness of each bilayer should be 5-6 nm, indicating a thickness increment of ca. 3-4 nm for deposition of silica. It indicates the orthosilicic acid/silicate oligomers infiltrate into the coated polypeptide film, resulting in the swelling of polypeptide film and, in turn, the increment of thickness. Silicas precipitate in the film due to the presence of PLL, which is evident that the LbL-assembled PLL/ PLGA macromolecular assembly acts as a template for the formation of silica. It is known that the presence of protonated amines (NH3þ) of the lysine side chain interacts strongly with the orthosilicic acid/silicate oligomers, further enhancing silica polymerization.8,9,53 Rather silica polymerization is relatively slow in the presence of neutral PLL due to only hydrogen bonding interactions between negatively charged silicate and

electrically neutral Nε-amine groups (NH2).21 In this study, LbL assembled PLL/PLGA multilayer films were fabricated via electrostatic association (NH3þ--OOC salt bridging). Hence part of protonated amines was neutralized and the diffusion of orthosilicic acid/silicate oligomers was hindered in the polypeptide multilayers. This explains why silica polymerization in the film is slow (on the order of hours), which is consistent with previous studies.29,54 Previous studies have demonstrated that the uniform and welldefined silica hollow tubes can be synthesized by depositing silica materials on the walls of membrane pores using sol-gel method.55,56 However, the as-synthesized silica hollow tubes did not possess porosity. On the basis of the results from TEM analysis, the as-synthesized silica hollow tubes are porous and the pore size of the calcined silica/(Lys340/Glu125)9 tubes is smaller than 5 nm (Figure 2C and Figure S9A, Supporting Information), whereas the calcined silica/(Lys145/Glu125)9 tubes possess larger pore size mostly (Figure 2D and Figure S9C, Supporting Information). It indicates that the size of aggregates is a function of the PLL and PLGA molecular weight, which correlates with the size of β-sheet/turn aggregates. Nitrogen adsorption measurements were further performed to determine the influence of the LbL assemble solution condition and polypeptide molecular weight on the pore size and porosity. From the adsorption data, the total pore volumes for all samples range between 0.26 and 0.45 cm3g-1, which is the amount of nitrogen adsorbed at p/p0 value of 0.98 (Table 1). the mesopore (2-10 nm) volumes for most samples, representing nitrogen adsorbed in the pore size between 2 and 10 nm, range between 0.2 and 0.35 cm3 g-1, except (Lys340/Glu370)9I and (Lys340/Glu370)9II (Table 1). The additional nitrogen adsorbed beyond the mesopore volume is due to nitrogen adsorbed on the surface of the tubes. On the basis of BJH analysis, all of the as-synthesized silica tubes possess 2838

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir

ARTICLE

Table 1. Nitrogen Adsorption Data of Silica Hollow Tubes micropore volume

mesopore volume

BJH pore volume

total pore volume

BET surface area

sample ID

(0-2 nm) (cm3/g)a

(2-10 nm) (cm3/g)

(cm3/g)b

(cm3/g)

(m2/g)

(Lys145/Glu125)9I

0.026

0.33

0.34

0.44

373

(Lys145/Glu125)9II

0.048

0.31

0.31

0.42

442

(Lys210/Glu125)9II

0.009

0.29

0.30

0.38

492

(Lys340/Glu125)9I

0.050

0.27

0.27

0.35

373

(Lys340/Glu125)9II

0.124

0.25

0.27

0.41

555

(Lys210/Glu190)9I

0.016

0.22

0.23

0.37

421

(Lys340/Glu370)9I (Lys340/Glu370)9II

0.040 0.050

0.14 0.09

0.14 0.10

0.30 0.26

413 327

The micropore volume was calculated using the t-plot method. b The “BJH pore volume” is the total volume adsorbed over the relative pressure range of 0.1 e p/p0 e 0.9 estimated use the BJH formalism.

a

Figure 3. Nitrogen adsorption isotherms and pore size distributions of silica tubes obtained from calcination of silicified (a) (Lys145/Glu125)9, (b) (Lys340/Glu125)9, (c) (Lys210/Glu190)9, and (d) (Lys340/Glu370)9 coated PC membrane using LbL assembly method I, as well as (e) (Lys145/Glu125)9, (f) (Lys210/Glu125)9, (g) (Lys340/Glu125)9, and (h) (Lys340/Glu370)9 coated PC membrane using LbL assembly method II. Open and closed circles are the adsorption and desorption isotherms, respectively. The offsets of the BET isotherms a-d are 320, 120, 100, and 0 cm3/g-STP, respectively. And the offsets of the BET isotherms e-h are 330, 180, 120, and 0 cm3/g-STP, respectively.

mesopores with pore size distributions between 2 and 8 nm (Figure 3), consistent with previous work.18,21 It is interesting to note that the average pore size and mesopore volume decrease with the increase of polypeptide molecular weight (MW). The

four samples synthesized with higher MW of polypeptides, (Lys340/Glu125)9I, (Lys340/Glu125)9II, (Lys340/Glu370)9I, and (Lys340/Glu370)9II, possess micropore volume higher than 0.04 cm3 g-1 (Table 1). In particular, (Lys340/Glu125)9II 2839

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir

ARTICLE

Scheme 2. Illustration of the Formation of Polypeptide Macromolecular Assemblies by PLL and PLGA with Different Molecular Weights

possesses the highest micropore volume (0.12 cm3 g-1) among all of these samples. The combined micropore and mesopore volumes for those samples range between 0.23 and 0.37 cm3 g-1, except (Lys340/Glu370)9I and (Lys340/Glu370)9II. The silica tubes synthesized with lower MW of polypeptides have larger pore size and higher pore volume than those synthesized with higher MW of polypeptides. The results show that PLL and PLGA with lower MW form larger sheet/turn aggregates on the surface, which result in the templated silica tubes with larger pore size. PLL and PLGA with high MW tend to form smaller sheet/ turn aggregates due to the chain entanglement and steric hindrance. The low MW polypeptide chains have less entanglement and steric hindrance as compared with the high MW ones and this can allow the peptide chains to associate with each other via electrostatic interaction along the chains and, in turn, form larger sheet/turn aggregates (Scheme 2). The results indicate that the silicification process and polypeptide molecular weight influence the molecular organization of PLL/PLGA macromolecular assemblies and subsequently the resultant pore size and porosity of the as-synthesized silica tubes. In addition, the presence of ions and the adsorption of polypeptides on the surface can regulate and confine PLL/PLGA macromolecular assemblies, subsequently leading to the formation of much uniform sheet/turn aggregates. We are not the first one to study the synthesis of porous silica using polypeptide secondary structures as templates.9,18,21 Previously, Jan and Shantz reported the synthesis of microporous silica nanoparticles and platelets using both the self-assembled structures formed by Lys-b-Gly block copolypeptides and Rhelical PLL as templates under benign condition. The coil-helix transition of PLL block and simultaneous silica precipitation in the presence of both silicic acid and phosphate ions led to R-helix templated silicas.9 B€orner group utilized the self-assembled PEOpeptide nanotapes with a β-sheet core as templates for silicification in ethanol/water solution and the resulting silica possessed pore size between 2 and 8 nm.18 Shantz group reported that the microporous silica synthesized with solvated R-helical PLL possesses cylindrical pores of approximately 1.5 nm diameter and the mesoporous silica synthesized using PLL chains adopted β-sheet conformation as templates possesses larger pores with sizes between 2 and 8 nm.21 The pore size of β-sheet templated silica depends on the PLL concentration, or namely the size of aggregates. However, these reported β-sheet templated silicas were not synthesized under benign condition and the morphology

and pore size of the as-prepared silicas cannot be controlled. In this study, the porous silica tubes can be synthesized using the simple synthesis process (i.e., LbL assemble and silicification processes) under benign condition. In addition, the average pore size can be controlled simply by using the polypeptides with different MWs. Some of the as-synthesized silica tubes possess higher mesopore volume and narrower pore size distribution than those reported by B€orner and Shantz group. It is also worth to note that porous silicas with different morphologies such as films and capsules can be easily prepared due to the versatility of the LbL technique and feasibility of polypeptide-mediated synthesis. Metal nanoparticles such as gold are attractive catalysts which have been successfully applied to a variety of reactions such as hydrogenation, oxidation-reduction, and reforming.57,58 Using a mesoporous silica support to immobilize gold nanoparticles have been wildly used for catalytic applications in many researches.59-62 Here we first demonstrate that PLGA can serve as both reducing and stabilizing agents in aqueous solution to form monodispersed gold nanoparticles at neutral and basic conditions as evidenced by the turning of colorless solution to pink and the appearance of plasmon resonance peak of gold nanoparticles at the wavelength of 520 nm, indicating the formation of gold nanoparticles. Instead, PLL does not possess reducing capability in aqueous solution according to our study. The time-resolved UV-vis measurements were conducted to study the kinetic growth of gold nanoparticles in the presence of PLGA at neutral and basic conditions (Figure S10, Supporting Information). At basic condition, the absorption at 520 nm was recorded as a function of time and the time course shows a rapid growth of nanoparticles between 0-100 min, followed by a plateau indicating the end of reaction. Rather, at neutral condition the continuous increase of the absorption at 520 nm is observed even after 3 days. The pH-dependent reduction potential of Au(III) complexes has been reported in the literature.63-65 The Au(III) complexes, AuCl2(OH)2- and Au(OH)4-, are the dominating form at neutral and basic conditions, respectively. The reactivity of AuCl2(OH)2-, which is an indication for the reduction of Au(III) complexes, is higher than that of Au(OH)4-.65 The relatively low absorption intensity at basic condition suggests that only a portion of Au(III) complexes was reduced by PLGA, which can be probably attributed to the low reactivity of Au(OH)4-. In contrast, much more Au(0) atoms were formed at neutral condition as indicated by the high absorption intensity. In addition, the deprotonation of COOH group at different pH 2840

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir

ARTICLE

Figure 4. UV spectra of gold/silica/PLL/PLGA multilayer membrane at different reducing time. Synthesis condition: HAuCl4 solution at basic condition.

would affect the rate of redox reaction. Hence, at neutral condition, the availability of COO- group and the reactivity of Au(III) complexes are the determining factors for the reduction of gold precursor by PLGA, which led to the higher absorption intensity and longer reduction time at neutral condition. TEM analysis reveals that the gold nanoparticles with average sizes 20 and 10 nm in diameter were synthesized at neutral and basic conditions, respectively (Figure S11, Supporting Information). The synthesis scheme of Au NP/meso-SiO2 tubes is shown in Scheme 1. The PLL/PLGA coated membranes were immersed into a HAuCl4 solution (pH 10.5-11.5) for one to 3 days, followed by inserting into freshly prepared orthosilicic acid solution. FE-SEM analysis confirms that the materials possess well-defined tubular structures (Figures S12-S14, Supporting Information). The formation of gold nanoparticles was confirmed by TEM and XRD characterization (Figure 5, parts A and B, and Figure S12, Supporting Information). UV-vis analysis shows that the amount of gold nanoparticles forming in the polypeptide multilayer films can be controlled by varying the immersion time, indicated by the increase of the plasmon resonance peak absorption of gold nanoparticles at ca. 530 nm (Figure 4). The red-shift of the peak from lower wavelength to 530 nm is due to the nanoparticles embedded in the silica network and polypeptide. According to TEM analysis, the particle size of gold in the gold/ silica/(Lys340/Glu125)5 composite tubes is smaller than 4 nm and the silica is porous with pore corresponding to the same pore size range as previously described (Figure 5A). The confined growth of gold nanoparticles in the PLGA nanodomains results in the well dispersion and uniform size distribution of nanoparticles. After calcination (500 °C, 10 h), the particle size increased and was on average 5 nm (Figure 5B). The excellent thermal stability of the materials is evidenced by the minor increase of particle size and the intact of silica nanostructure after sintering process. The well-defined tubular structures were also obtained by reducing chloroauric acid at neutral condition for 1 day (Figure S13, Supporting Information). In addition, the gold/silica/polypeptide composite and gold/silica materials can be prepared by replacing poly(L-glutamic acid) with poly(L-tyrosine) (PLT) using a modified synthesis procedure (Figure 5C). The presence of gold nanoparticles in the silica materials were confirmed by UV-vis and XRD measurements and the particle size is 5 nm on average (Figure S14, Supporting Information).

Figure 5. TEM images of (A) gold/silica/polypeptide and (B) gold/ silica hollow tubes synthesized from (Lys340/Glu125)5 coated membrane, and (C) gold/silica hollow tubes synthesized from (Lys340/ PLT)5 coated membrane.

For the application of these gold/silica tubes, p-nitrophenol was used to evaluate the catalytic activity of the gold/silica tubes. The reduction of p-nitrophenol did not occur in the absence of gold/silica nanocatalyst for at least 2 days. Instead, in the presence of gold/silica nanocatalyst, the reduction of p-nitrophenol was observed as evidenced by the decrease of the absorbance band of p-nitrophenol. According to literature data, this catalyst-mediated reduction can be assumed as pseudofirstorder reaction.60 The gold/silica nanocatalyst prepared from 5-layer composite tubes with two-day reduction was used to reduce p-nitrophenol. From UV-vis measurements, the apparent rate constant can be calculated from the slope of linear relation of ln A (the intensity of absorbance) versus time (second) and the values were calculated to be 3  10-4 and 8  10-4 s-1 for gold/silica tubes obtained from (Lys340/Glu125)5 and (Lys340/PLT)5 coated membranes, respectively (Figure 6). ICP-MS measurements were performed to determine the elemental weight ratio between Si and Au in the glod/silica tubes (Experimental Section in the Supporting Information). And the weight percentages of gold in the two samples were calculated to be 0.95 and 2.41 wt %, respectively. The results demonstrate that 2841

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir

ARTICLE

Figure 6. Plot of ln A versus time for the reduction of p-nitrophenol with NaBH4 by the Au NP/meso-SiO2 tubes obtained from (A) (Lys340/Glu125)5 or (B) (Lys340/PLT)5 coated PC membrane. Initial condition: [p-nitrophenol] = 1.33  10-4 M, [NaBH4] = 6.67  10-3 M.

the Au NP/meso-SiO2 tubes synthesized using the current procedure exhibit promising catalytic property. It is expected that the structure and property of gold/silica tubes can be improved by optimizing the synthesis procedure and the sintering process. Previously, the gold nanoparticle/polyelectrolyte films were coated in the membrane pores and used as catalytic membranes for p-nitrophenol reduction.66 Using this approach, the catalytic membranes coated with gold nanoparticle/mesoporous silica in the pores can be synthesized with the advantage of robustness and durability.

’ CONCLUSION We have demonstrated that silica/polypeptide, gold nanoparticle/polypeptide, and gold nanoparticle/mesoporous silica/ polypeptide composite tubes can be prepared by polypeptidemediated formation of silica and metal nanoparticle. Mesoporous silica and gold nanoparticle/mesoporous silica tubes can be obtained by subsequent calcination. PLL/PLGA macromolecular assemblies act not only as mediating agents for formation of metal nanoparticles and silica, but also templates for directing mesoporous silica formation and confining the growth of metal nanoparticles. The as-prepared hollow tubes have well-defined diameters and lengths that are determined by that of the template pores, uniform, and conformal walls with thicknesses that are finely controlled by the number of deposited layers, and, most importantly, porosities that are dependent on the aggregate size of the sheet/turn-like PLL/PLGA aggregates, which is a function of polypeptide molecular weight. The successful preparation of metal nanoparticles with uniform size distribution in the silica porous support is also worth noting and the as-prepared gold/ silica tubes exhibit good thermal stability and catalytic activity toward the reduction of p-nitrophenol. Because of the versatility of the LbL technique and feasibility of polypeptide-mediated synthesis, it is expected the present method can be easily extended to the synthesis of multicomponent organic-inorganic composite films and capsules. Also, silica and gold nanoparticle/ silica micro/nanopatterns on surfaces can be obtained via the combination of this method and soft lithography. In addition, this study will provide insight on the directed growth of metals and mineralization of oxides in some biological systems. ’ ASSOCIATED CONTENT

bS

Supporting Information. Details of all materials and measurements. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT J.-S. Jan acknowledges funding support from National Science Council grant NSC98-2218-E-006-010, NSC99-2628-E-006003, and Center for Frontier Materials and Micro/Nano Science and Technology (CFMMNST) at National Cheng Kung University (D98-2700). J.-S. Jan also acknowledges T.-C. Wen for access to the glovebox, H. Teng for access to the nitrogen porosimetry, and J.-J. Wu for access to the UV-vis spectrophotometers. ’ REFERENCES (1) Estroff, L. A.; Hamilton, A. D. Chem. Mater. 2001, 13, 3227– 3235. (2) Poulsen, N.; Sumper, M.; Kroger, N. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 12075–12080. (3) Aizenberg, J.; Weaver, J. C.; Thanawala, M. S.; Sundar, V. C.; Morse, D. E.; Fratzl, P. Science 2005, 309, 275–278. (4) Sanchez, C.; Arribart, H.; Guille, M. M. G. Nat. Mater. 2005, 4, 277–288. (5) Jensen, M.; Keding, R.; Hoche, T.; Yue, Y. Z. J. Am. Chem. Soc. 2009, 131, 2717–2721. (6) Cha, J. N.; Shimizu, K.; Zhou, Y.; Christiansen, S. C.; Chmelka, B. F.; Stucky, G. D.; Morse, D. E. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 361–365. (7) Cha, J. N.; Stucky, G. D.; Morse, D. E.; Deming, T. J. Nature 2000, 403, 289–292. (8) Jan, J.-S.; Lee, S.; Carr, C. S.; Shantz, D. F. Chem. Mater. 2005, 17, 4310–4317. (9) Jan, J.-S.; Shantz, D. F. Adv. Mater. 2007, 19, 2951–2956. (10) Yuan, J. J.; Mykhaylyk, O. O.; Ryan, A. J.; Armes, S. P. J. Am. Chem. Soc. 2007, 129, 1717–1723. (11) Li, Y. T.; Du, J. Z.; Armes, S. P. Macromol. Rapid Commun. 2009, 30, 464–468. (12) Tomczak, M. M.; Glawe, D. D.; Drummy, L. F.; Lawrence, C. G.; Stone, M. O.; Perry, C. C.; Pochan, D. J.; Deming, T. J.; Naik, R. R. J. Am. Chem. Soc. 2005, 127, 12577–12582. (13) Meegan, J. E.; Aggeli, A.; Boden, N.; Brydson, R.; Brown, A. P.; Carrick, L.; Brough, A. R.; Hussain, A.; Ansell, R. J. Adv. Funct. Mater. 2004, 14, 31–37. (14) Yuan, J. J.; Zhu, P. X.; Fukazawa, N.; Jin, R. H. Adv. Funct. Mater. 2006, 16, 2205–2212. (15) Yuwono, V. M.; Hartgerink, J. D. Langmuir 2007, 23, 5033– 5038. 2842

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843

Langmuir (16) Holmstrom, S. C.; King, P. J. S.; Ryadnov, M. G.; Butler, M. F.; Mann, S.; Woolfson, D. N. Langmuir 2008, 24, 11778–11783. (17) Altunbas, A.; Sharma, N.; Lamm, M. S.; Yan, C. Q.; Nagarkar, R. P.; Schneider, J. P.; Pochan, D. J. ACS Nano 2010, 4, 181–188. (18) Kessel, S.; Thomas, A.; B€orner, H. G. Angew. Chem., Int. Ed. 2007, 46, 9023–9026. (19) Kessel, S.; B€orner, H. G. Macromol. Rapid Commun. 2008, 29, 419–424. (20) Wong, M. S.; Cha, J. N.; Choi, K. S.; Deming, T. J.; Stucky, G. D. Nano Lett. 2002, 2, 583–587. (21) Hawkins, K. M.; Wang, S. S.-S.; Ford, D. M.; Shantz, D. F. J. Am. Chem. Soc. 2004, 126, 9112–9119. (22) Rana, R. K.; Murthy, V. S.; Yu, J.; Wong, M. S. Adv. Mater. 2005, 17, 1145–1150. (23) Patwardhan, S. V.; Maheshwari, R.; Mukherjee, N.; Kiick, K. L.; Clarson, S. J. Biomacromolecules 2006, 7, 491–497. (24) Begum, G.; Rana, R. K.; Singh, S.; Satyanarayana, L. Chem. Mater. 2010, 22, 551–556. (25) Brott, L. L.; Naik, R. R.; Pikas, D. J.; Kirkpatrick, S. M.; Tomlin, D. W.; Whitlock, P. W.; Clarson, S. J.; Stone, M. O. Nature 2001, 413, 291–293. (26) Coffman, E. A.; Melechko, A. V.; Allison, D. P.; Simpson, M. L.; Doktycz, M. J. Langmuir 2004, 20, 8431–8436. (27) Tahir, M. N.; Theato, P.; Muller, W. E. G.; Schroder, H. C.; Borejko, A.; Faiss, S.; Janshoff, A.; Huth, J.; Tremel, W. Chem. Commun. 2005, 5533–5535. (28) Kim, D. J.; Lee, K. B.; Lee, T. G.; Shon, H. K.; Kim, W. J.; Paik, H. J.; Choi, I. S. Small 2005, 1, 992–996. (29) Wu, J. C.; Wang, Y. L.; Chen, C. C.; Chang, Y. C. Chem. Mater. 2008, 20, 6148–6156. (30) Gautier, C.; Lopez, P. J.; Hemadi, M.; Livage, J.; Coradin, T. Langmuir 2006, 22, 9092–9095. (31) Wang, Y.; Angelatos, A. S.; Caruso, F. Chem. Mater. 2008, 20, 848–858. (32) Jiang, Y. J.; Yang, D.; Zhang, L.; Sun, Q. Y.; Sun, X. H.; Li, J.; Jiang, Z. Y. Adv. Funct. Mater. 2009, 19, 150–156. (33) Yang, S. H.; Lee, K. B.; Kong, B.; Kim, J. H.; Kim, H. S.; Choi, I. S. Angew. Chem., Int. Ed. 2009, 48, 9160–9163. (34) Li, J.; Jiang, Z. Y.; Wu, H.; Zhang, L.; Long, L. H.; Jiang, Y. J. Soft Matter 2010, 6, 542–550. (35) Wang, Y. J.; Caruso, F. Chem. Mater. 2005, 17, 953–961. (36) Caruso, F.; Caruso, R. A.; Mohwald, H. Chem. Mater. 1999, 11, 3309–3314. (37) Caruso, R. A.; Susha, A.; Caruso, F. Chem. Mater. 2001, 13, 400–409. (38) Xie, J. P.; Lee, J. Y.; Wang, D. I. C.; Ting, Y. P. ACS Nano 2007, 1, 429–439. (39) Bhargava, S. K.; Booth, J. M.; Agrawal, S.; Coloe, P.; Kar, G. Langmuir 2005, 21, 5949–5956. (40) Selvakannan, P. R.; Swami, A.; Srisathiyanarayanan, D.; Shirude, P. S.; Pasricha, R.; Mandale, A. B.; Sastry, M. Langmuir 2004, 20, 7825– 7836. (41) Ray, S.; Das, A. K.; Banerjee, A. Chem. Commun. 2006, 2816– 2818. (42) Lee, H.; Lee, Y.; Statz, A. R.; Rho, J.; Park, T. G.; Messersmith, P. B. Adv. Mater. 2008, 20, 1619-þ. (43) Dong, Q.; Su, H. L.; Zhang, D. J. Phys. Chem. B 2005, 109, 17429–17434. (44) Kharlampieva, E.; Slocik, J. M.; Tsukruk, T.; Naik, R. R.; Tsukruk, V. V. Chem. Mater. 2008, 20, 5822–5831. (45) Jan, J.-S.; Shantz, D. F. Chem. Commun. 2005, 2137–2139. (46) Deming, T. J. Nature 1997, 390, 386–389. (47) Deming, T. J.; Curtin, S. A. J. Am. Chem. Soc. 2000, 122, 5710– 5717. (48) Gaspard, J.; Silas, J. A.; Shantz, D. F.; Jan, J.-S. Supramol. Chem. 2010, 22, 178–185. (49) Boulmedais, F.; Schwinte, P.; Gergely, C.; Voegel, J. C.; Schaaf, P. Langmuir 2002, 18, 4523–4525.

ARTICLE

(50) Boulmedais, F.; Bozonnet, M.; Schwinte, P.; Voegel, J. C.; Schaaf, P. Langmuir 2003, 19, 9873–9882. (51) Haynie, D. T.; Balkundi, S.; Palath, N.; Chakravarthula, K.; Dave, K. Langmuir 2004, 20, 4540–4547. (52) Yang, C. T.; Wang, Y. L.; Yu, S.; Chang, Y. C. I. Biomacromolecules 2009, 10, 58–65. (53) Menzel, H.; Horstmann, S.; Behrens, P.; Barnreuther, B.; Krueger, I.; Jahns, M. Chem. Commun. 2003, 2994–2995. (54) Xu, M. J.; Gratson, G. M.; Duoss, E. B.; Shepherd, R. F.; Lewis, J. A. Soft Matter 2006, 2, 205–209. (55) Kovtyukhova, N. I.; Mallouk, T. E.; Mayer, T. S. Adv. Mater. 2003, 15, 780–785. (56) Chen, C. C.; Liu, Y. C.; Wu, C. H.; Yeh, C. C.; Su, M. T.; Wu, Y. C. Adv. Mater. 2005, 17, 404–407. (57) Arcadi, A. Chem. Rev. 2008, 108, 3266–3325. (58) Bond, G. C.; Thompson, D. T. Catal. Rev.—Sci. Eng. 1999, 41, 319–388. (59) Gerolamo, B.; Avelino, C. A. Angew. Chem., Int. Ed. 2006, 45, 3328–3331. (60) Lee, J.; Park, J. C.; Song, H. Adv. Mater. 2008, 20, 1523–1528. (61) Wu, Z. L.; Zhou, S. H.; Zhu, H. G.; Dai, S.; Overbury, S. H. Chem. Commun. 2008, 3308–3310. (62) Gajan, D.; Guillois, K.; Delichere, P.; Basset, J. M.; Candy, J. P.; Caps, V.; Coperet, C.; Lesage, A.; Emsley, L. J. Am. Chem. Soc. 2009, 131, 14667–14669. (63) Yu, Y.-Y.; Chang, S.-S.; Lee, C.-L.; Wang, C. R. C. J. Phys. Chem. B 1997, 101, 14667–14669. (64) Pei, L.; Mori, K.; Adachi, M. Langmuir 2004, 20, 7837–7843. (65) Ji, X.; Song, X.; Li, J.; Bai, Y.; Yang, W.; Peng, X. J. Am. Chem. Soc. 2007, 129, 13939–13948. (66) Dotzauer, D. M.; Dai, J.; Sun, L.; Bruening, M. L. Nano Lett. 2006, 6, 2268–2272.

2843

dx.doi.org/10.1021/la103923c |Langmuir 2011, 27, 2834–2843