4-(Dimethylamino)pyridine as a Powerful Auxiliary Reagent in the

Dec 6, 2010 - Gold nanotubes of small particle sizes down to 5 nm and high aspect ratios were synthesized in ion track etched polycarbonate following ...
18 downloads 0 Views 3MB Size
Download PDF
pubs.acs.org/Langmuir © 2010 American Chemical Society

4-(Dimethylamino)pyridine as a Powerful Auxiliary Reagent in the Electroless Synthesis of Gold Nanotubes Falk Muench,* Ulrike Kunz, Cornelia Neetzel, Stefan Lauterbach, Hans-Joachim Kleebe, and Wolfgang Ensinger Technische Universit€ at Darmstadt, Department of Materials and Geoscience, Petersenstrasse 23, 64287 Darmstadt, Germany Received October 5, 2010. Revised Manuscript Received November 12, 2010 Gold nanotubes of small particle sizes down to 5 nm and high aspect ratios were synthesized in ion track etched polycarbonate following a rational reaction design. 4-(Dimethylamino)pyridine (DMAP) was employed to adjust the electroless deposition by interfering with the autocatalytically active gold surface. Modification of the pH value and DMAP concentration led to a wide range of products which were characterized by SEM, TEM, and EDS. Filigree nanotubes of 10-15 nm wall thickness and 5.0 ( 2.1 nm grain size were obtained as well as robust and free-standing structures proving homogeneous deposition along the whole template length of 30 μm. Template-supported gold nanotubes were applied in the UV-vis monitored reduction of 4-nitrophenol by sodium borohydride under pseudofirst-order conditions. They proved to be a reliable microfluidic system of excellent catalytic activity coming up with an apparent rate constant of 1.3  10-2 s-1. Despite a high flow rate, the reaction showed 99% conversion after a distance of just 60 μm.

Introduction During the past 15 years, metal nanotubes (NTs) have emerged as efficient materials for various exciting applications such as DNA recognition,1 nanopore biosensing,2 CO3,4 or methanol based5,6 fuel cell catalysts, synthetic permselective membranes,7 or microfluidic devices.8 The diverse synthetic approaches leading to these structures9,10 have one thing in common: They comprise the use of templates, since metals with their quite isotropic crystal structure do not form tubes as freely as highly anisotropic materials like graphite or transition metal sulfides.9-11 As main methods, electroless plating,1-3 electroplating,12-15 and sacrificial templating4 can be named. For the synthesis of metal NTs,1-4,7,8,12-15 mainly hard and passive templates are used, resulting in oriented, embedded, and well-manageable nanostructures. The most prominent templates *Corresponding author. Tel þ49-6151-166387. Fax: þ49-6151-166378. E-mail: [email protected]. (1) Kohli, P.; Harrell, C. C.; Cao, Z.; Gasparac, R.; Tan, W.; Martin, C. R. Science 2004, 305, 984. (2) Siwy, Z.; Trofin, L.; Kohli, P.; Baker, L. A.; Trautmann, C.; Martin, C. R. J. Am. Chem. Soc. 2005, 127, 5000. (3) Sanchez-Castillo, M. A.; Couto, C.; Kim, W. B.; Dumesic, J. A. Angew. Chem., Int. Ed. 2004, 43, 1140. (4) Kim, W. B.; Voitl, T.; Rodriguez-Rivera, G. J.; Dumesic, J. A. Science 2004, 305, 1280. (5) Guo, Y.-G.; Hu, J.-S.; Zhang, H.-M.; Liang, H.-P.; Wan, L.-J.; Bai, C.-L. Adv. Mater. 2005, 17, 746. (6) Bi, Y.; Lu, G. Electrochem. Commun. 2009, 11, 45. (7) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Science 1997, 278, 655. (8) Kim, B. Y.; Swearingen, C. B.; Ho, J. A.; Romanova, E. V.; Bohn, P. W.; Sweedler, J. V. J. Am. Chem. Soc. 2007, 129, 7620. (9) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim, F.; Yan, H. Adv. Mater. 2003, 15, 353. (10) Rao, C. N. R.; Govindaraj, A. Adv. Mater. 2009, 21, 4208. (11) Remskar, M. Adv. Mater. 2004, 16, 1497. (12) Wang, H.-W.; Shieh, C.-F.; Chen, H.-Y.; Shiu, W.-C.; Russo, B.; Cao, G. Nanotechnology 2006, 17, 2689. (13) Cheng, C.-L.; Lin, J.-S.; Chen, Y.-F. Mater. Lett. 2008, 62, 1666. (14) Hendren, W. R.; Murphy, A.; Evans, P.; O’Connor, D.; Wurtz, G. A.; Zayats, A. V.; Atkinson, R.; Pollard, R. J. J. Phys.: Condens. Matter 2008, 20, 362203. (15) McPhillips, J.; Murphy, A.; Jonsson, M. P.; Hendren, W. R.; Atkinson, R.; H€oo€k, F.; Zayats, A. V.; Pollard, R. J. ACS Nano 2010, 4, 2210.

430 DOI: 10.1021/la104015a

are nanochannel-containing anodized metal oxides12-15 and ion track etched polymer films16;porous solids which are not directly involved in the redox process. In electroless plating, the whole template surface is simultaneously covered with metal, rendering this method well suited for the fabrication of high aspect ratio NTs and other complex-shaped structures. Electroplating requires electrical contacting of the template and tends to produce nanowires17 or nanowire-nanotube junctions, especially at small tube diameters and extended reaction times.12,13 By modifications such as introducing additional templating steps, the electrochemical fabrication of metal NTs can be improved. By this way, narrow AuNTs of less than 400 nm length could be obtained.14,15 Sacrificial templating uses wirelike structures to reduce a nobler metal on their surface, yielding tubes after the template residues are removed.4 This method does not require porous templates and can be scaled up more easily.18 However, the NT composition, diameter, length, and wall thickness cannot be controlled as freely as in the case of electroless plating.18 Despite its high potential, electroless metal deposition often lacks control on the nanoscale. For instance, the particles of which typical AuNTs are composed of have a minimum size of around 30 nm.19 This limits the synthesis of small and regular tubular structures. An increase of the morphological control in the gold system is desirable, since many applications benefit from small Au crystals20 and inner tube diameters.7 Metal nanoparticle (NP) synthesis relies on the adjustment of the particle surface reactivity in order to avoid agglomeration or to favor the growth of singular crystal faces.21 In the presented (16) Apel, P. Radiat. Meas. 2001, 34, 559. (17) Karim, S.; Toimil Molares, M. E.; Maurer, F.; Miehe, G.; Ensinger, W.; Liu, J.; Cornelius, T. W.; Neumann, R. Appl. Phys. A: Mater. Sci. Process. 2006, 84, 403. (18) Sun, Y.; Xia, Y. Adv. Mater. 2002, 14, 833. (19) De Leo, M.; Pereira, F. C.; Moretto, L. M.; Scopece, P.; Polizzi, S.; Ugo, P. Chem. Mater. 2007, 19, 5955. (20) Delvaux, M.; Walcarius, A.; Demoustier-Champagne, S. Anal. Chim. Acta 2004, 525, 221. (21) Xia, Y.; Xiong, Y.; Lim, B.; Skrabalak, S. E. Angew. Chem., Int. Ed. 2009, 48, 60.

Published on Web 12/06/2010

Langmuir 2011, 27(1), 430–435

Muench et al.

Article

work, we pursue this concept by transferring a compound used in NP design to the related field of electroless plating: The protective agent 4-(dimethylamino)pyridine (DMAP). This electron-rich pyridine inhibits particle agglomeration by the formation of a hydrophilic shell and the introduction of charge.22-26 Its reversible binding to gold surfaces can be triggered by protonation and allows the exchange of the DMAP by other species.22-26 DMAP comes up with a combination of beneficial properties. Since it is water-soluble, it can be introduced in regular electroless plating baths as a controlling agent next to the metal complex and the reducer. It blocks Au surfaces effectively and should constrict the heterogeneously autocatalyzed Au deposition without irreversibly binding to reactive sites. Furthermore, the adsorption behavior can be tuned by adjusting the pH value. These characteristics gave rise to the reinterpretation of DMAP as an ideal compound for controlling electroless gold plating. Following these thoughts, we herein elucidate the pH-dependent effect of DMAP on the growth rate and morphology of electrolessly synthesized AuNTs. Since DMAP can be easily removed from gold surfaces, the obtained structures should be convenient heterogeneous catalysts.27 This issue was examined by testing the polymer embedded AuNTs in the reduction of 4-nitrophenol by sodium borohydride.

Experimental Section General, Chemicals. Glassware was cleaned with aqua regia prior to use. The sensitization, activation, and plating solutions were freshly prepared. All procedures were performed with purified water (Milli-Q 18 MΩ water). The following chemicals were used without further purification: 4-(dimethylamino)pyridine (Fluka, puriss.); 4-nitrophenol (Fluka, puriss. p.a.); acetic acid 99%-100% (Sigma-Aldrich, puriss.); AgNO3 (Gr€ ussing, p.a.); ammonia 33% in water (Merck, puriss.); dichloromethane (Sigma-Aldrich, puriss. p.a.); ethanol (Brenntag, 99.5%); formaldehyde solution 37% in water, methanol stabilized (Gr€ ussing, p.a.); methanol (Sigma-Aldrich, laboratory reagent); NaBH4 (Merck, for synthesis); Na2SO3 (Merck, p.a.); SnCl2 (Merck, for synthesis); sodium hydroxide solution 32% in water (Fluka, :: puriss. p.a.); trifluoroacetic acid (Riedel-de Haen, >99%). A commercial electroplating solution (El-Form Galvano Goldbad, Sch€ utz Dental GmbH) was used as the Au source (15 g 99.9% Au per liter, present as (NH4)3(Au(SO3)2)). Template Synthesis. Polycarbonate foils (Makrofol, Bayer MaterialScience AG, nominal thickness 30 μm) were irradiated with Au ions (energy: 11 MeV/u; ion fluence: 1  108 ions/cm2) at the Helmholtz Center for Heavy Ion Research (GSI). Subsequently, they were irradiated with UV light in the presence of air (1 h per side, UV source provides 1.5 and 4.0 Wm2 in the ranges 280-320 and 320-400 nm, respectively) and etched in stirred sodium hydroxide solution (6 M, 50 °C, time depending on desired diameter, ranging from 30 to 500 nm). The as-prepared templates were thoroughly washed with water and dried. Template Activation. First, the polycarbonate membranes were immersed in a Sn(II) solution for at least 45 min (0.042 M SnCl2, 0.071 M trifluoroacetic acid in MeOH:water = 1:1). To remove excessive Sn(II), the foils were washed twice with ethanol before immersing them in Ag(I) solution for 3 min (0.059 M AgNO3, 0.230 M NH3). In this step, the membranes turned slightly brownish, indicating the precipitation of AgNPs. The (22) (23) 4674. (24) (25) 508. (26) (27)

Gandubert, V. J.; Lennox, R. B. Langmuir 2005, 21, 6532. Rucareanu, S.; Gandubert, V. J.; Lennox, R. B. Chem. Mater. 2006, 18, Vivek, J. P.; Burgess, I. J. J. Phys. Chem. C 2008, 112, 2872. Gittins, D. I.; Susha, A. S.; Schoeler, B.; Caruso, F. Adv. Mater. 2002, 14, Gittins, D. I.; Caruso, F. Angew. Chem., Int. Ed. 2001, 40, 3001. Yu, A.; Liang, Z.; Cho, J.; Caruso, F. Nano Lett. 2003, 3, 1203.

Langmuir 2011, 27(1), 430–435

Ag-covered templates were washed with ethanol and water and immediately used for the electroless gold plating. Growth of Au Nanostructures. The deposition solutions contained 7.00 mM Au(I), 125 mM Na2SO3, 625 mM formaldehyde, and different amounts of DMAP. The pH of all solutions was adjusted using concentrated sodium hydroxide solution. Deposition took place at room temperature for all experiments except the preparation of the catalyst membrane, which was conducted at 8 °C in order to slow down the reaction and achieve a fine structure. After the desired reaction time, the AuNP covered templates were taken out, washed with water, and dried. Catalysis. Before the catalytic experiments, the gold-covered membranes were washed with diluted acetic acid and water in order to remove adsorbed DMAP. The reaction solution was prepared by mixing a solution containing 4-nitrophenol with a freshly prepared sodium borohydride solution. The initially light yellow color of the 4-nitrophenol solution turned to an intense yellow due to deprotonation of the 4-nitrophenol molecule in the basic environment provided by sodium borohydride. The final reaction solution contained 8.4  10-4 M 4-nitrophenol and a large excess of sodium borohydride (60 mM) to ensure pseudofirst-order kinetics despite the hydrolysis of sodium borohydride as a side reaction. A glass syringe attached to a membrane filtration holder (Nucleopore sample kit for diagnostic cytology) containing gold-covered templates was used to press the reaction solution through the NTs. Analytics. TEM (FEI CM20, 230 kV acceleration voltage, LaB6-cathode): The tube-containing templates were embedded in Araldit 502 (polymerization at 60 °C for 16 h) and examined as ultrathin sections (70 nm, Reichert-Jung Ultracut E ultramicrotome, DKK diamond knife). SEM (JEOL JSM-7401F, 5 kV acceleration voltage): Prior to the measurement, the template was removed with dichloromethane. The freed metal structures were collected on silicon wafer pieces sputter-coated with a thin Au film. UV-vis spectroscopy (ATi Unicam UV/vis spectrometer UV4): The UVvis spectra were recorded using a high scan rate of 240 nm/min to evade problems caused by the formation of hydrogen bubbles. The observed range was 250-500 nm.

Results and Discussion Prior to the experimental results, a brief summary of the applied Au deposition procedure and the adsorption chemistry of DMAP is given. In electroless plating, a metal film is selectively deposited on a substrate surface immersed in a solution containing a metal salt and a reducing agent which together form a metastable redox pair. Metastability is necessary to suppress homogeneous nucleation leading to uncontrolled metal precipitation. In our case, formaldehyde is used as the reducer in combination with sulfite-stabilized Au(I) as the oxidizing agent. Accompanied by the consumption of hydroxide ions, formaldehyde is oxidized to formic acid while the sulfito complex is reduced to elemental Au: 2AuðSO3 Þ2 3 - þ HCHO þ 3OH - f 2Au þ 4SO3 2 þ HCOO - þ 2H2 O

ð1Þ

As reaction 1 is autocatalyzed by Au, the deposition is sustaining itself once started. To achieve initial activity, the polycarbonate surface is covered with AgNPs. During contact with the plating solution, the AgNPs are converted into AuNPs acting as nuclei for the Au film growth. In Scheme 1, the complete route from the polymer film to the NT-containing template is summarized. DMAP is a relatively basic molecule (pKa = 9.70)28 and can be protonated once. Depending on its protonation state, it interacts (28) Dean, J. A. Lange’s Handbook of Chemistry, 15th ed.; McGraw-Hill: New York, 1999.

DOI: 10.1021/la104015a

431

Article

Muench et al.

Scheme 1. Synthetic Steps in the Fabrication of a AuNT Membranea

a Steps: (1) The polymer is irradiated by heavy ions leaving cylindrical damage zones along their tracks. (2) The ion tracks are favorably etched out with NaOH solution yielding the nanochannel template. (3) Immersion in a SnCl2 solution leads to adsorption of Sn(II) by the polar functional groups of the polymer (sensitization). (4) The membrane is transferred to an ammoniac AgNO3 solution. AgNPs precipitate on the template surface due to Ag(I) reduction by Sn(II) (activation). (5) When in contact with the electroless plating solution, a Au film evolves on the activated template surface.

Scheme 2. Equilibria between DMAP, Elemental Au, and Protons

with Au surfaces in different ways (Scheme 2). In neutral state, it binds to elemental Au by its pyridine nitrogen, leading to a perpendicular orientation toward the surface.29 The corresponding pyridinium ion HDMAPþ loses its properties as a nitrogen ligand but is still able to protect AuNPs to a minor degree by adsorption in a flat geometry.22,29 Therefore, the pH value of the solution affecting the ratio of the DMAP species is an important experimental parameter. If surface reactions are ignored, the DMAP and its protonated derivative are in equilibrium at a pH of 9.7, and the relative amount of the unprotonated species can be calculated according to eq 2: ½DMAP ¼ 10 - pKa þpH ½HDMAPþ 

ð2Þ

As the DMAP molecule adsorbs on Au forming a protective shell of around 1 nm thickness,22 one expects the deposition reaction rate to decrease with increasing surface coverage. To clarify this issue, the DMAP concentration was varied while keeping the pH value of the reaction solutions constant at 9.9. The applied pH value is convenient for the electroless synthesis of AuNTs19 and ensures the presence of unprotonated DMAP. (29) Barlow, B. C.; Burgess, I. J. Langmuir 2007, 23, 1555.

432 DOI: 10.1021/la104015a

According to the expectation, the reaction rate drops by increasing concentration of the protective agent. In the DMAPfree reference experiment (Figure 1A), polycrystalline Au wires of a few micrometers length were found next to a thick Au surface film. The fast reaction blocks the region near the pore openings and inhibits tube formation inside the template. In the presence of DMAP, NTs and thinner surface films were obtained. Because of their fragility, the tubes suffer from deformation and fragmentation during the template removal (Figure 1B-D). Next to kinetic effects, distinct changes in the size and shape of the AuNPs are observed. The NPs created by conversion of the initial silver layer forming the outer tube surface are nearly spherical and reduce in size from the 42 mM sample (10 nm) to the 132 mM sample (slightly below 5 nm). Also, increasing DMAP contents improve the life span of the metastable deposition baths by effectively suppressing homogeneous nucleation. On top of the Au film arising from the initial stages of the Au deposition, needlelike particles are found (see arrow in the inset of Figure 1C). The growth of anisotropic particles can be understood in terms of the shape-control ability of surfactants selectively adsorbing on different crystal faces21 and is most pronounced in the case of intermediate DMAP concentrations. Interestingly, the Au needle film is subject to self-organization. Flowlike orientations on the template surface are found as well as polygonal arrangements near pore openings. For the evaluation of the pH dependence, two pH values where chosen including regimes with both DMAP and HDMAPþ as the dominant solution species (see eq 2). Figure 2 displays Au structures yielded in the presence of 90 mM DMAP at pH values of 8.8 and 11.6. With increasing pH value, the deposition speed as well as the Au particle size considerably decreases. In DMAP-free experiments, the reaction rate rises with the hydroxide concentration according to eq 1. The observed reverse trend can be explained by the gain of unprotonated DMAP overcompensating the rate acceleration by Au surface blocking. Also, the particle shape undergoes distinct changes. At pH 8.8 large edged particles are formed instead of needles (Figure 2A), while a pH value of 11.6 led to fragile and thin Au films whose corresponding tubes instantly collapse in case of template removal (Figure 2B). By choosing appropriate pH values, DMAP concentrations, initial AgNP coverages, and reaction times, it is possible to produce homogeneous AuNTs of different particle size, shape, and wall thickness in a reproducible manner. As the general trend, high pH values and DMAP concentrations cause reduced particles sizes and deposition rates. Even very small nanotubes with outer diameters of 30 nm could be obtained. The synthesis of freestanding NTs spanning the whole template length proves the deposition homogeneity and the accessibility of high aspect ratios (Figure 3) which is difficult to achieve with electroplating.12,13 Many common capping agents like thiols or polyvinylpyrrolidone30 bind strongly to metal NPs and form bulky shells around them. This is an advantage if reliable agglomeration protection is concerned but can turn into a flaw whenever neat metal surfaces are needed, such as in case of catalytic27,31 or surface-enhanced Raman32 experiments. Since DMAP can be removed by simple rinsing with diluted acids,22 NPs grown in its presence have the potential to be highly efficient catalysts.27 (30) Borodko, Y.; Habas, S. E.; Koebel, M.; Yang, P.; Frei, H.; Somorjai, G. A. J. Phys. Chem. B 2006, 110, 23052. (31) Kumar, S. S.; Kumar, C. S.; Mathiyarasu, J.; Phani, K. L. Langmuir 2007, 23, 3401. (32) Tao, A.; Kim, F.; Hess, C.; Goldberger, J.; He, R.; Sun, Y.; Xia, Y.; Yang, P. Nano Lett. 2003, 9, 1229.

Langmuir 2011, 27(1), 430–435

Muench et al.

Article

Figure 1. SEM images of Au nanostructures obtained at different DMAP concentrations at pH 9.9 (A, 0 mM; B, 42 mM; C, 89 mM; D, 130 mM). Insets show magnified Au wires (A), tubes (B, C), or collapsed tubes (D). The arrow in part C marks anisotropic crystals grown on top of the surface Au film inside broken tubes.

Figure 3. SEM images of a field of free-standing NTs (left) and a relatively small NT with homogeneous walls (right) proving the successful deposition inside the whole template.

The polymer template is flexible and allows easy handling of the aligned hollow nanostructures inside of it, rendering it an ideal system for flow-through catalysis. Tracking this concept, we tested the embedded AuNTs in a model reaction, the formation

of 4-aminophenol by the reduction of 4-nitrophenol with sodium borohydride.31,33-36 This comfortable reaction does not require elevated temperatures, it is not proceeding in the absence of a catalyst, and both the product and educt can be readily detected by UV-vis spectrometry. Mechanistic investigations on Au- and PtNPs show that the reduction kinetics can be fitted by a Langmuir-Hinshelwood model, suggesting both reagents are present on the catalytically active NPs in the form of adsorbates.37 In more detail, borohydride and 4-nitrophenol are in a concentration-dependent equilibrium with surface-bound hydrogen and adsorbed nitrophenol. In the rate-determining step, these species

(33) Huang, J.; Vongehr, S.; Tang, S.; Lu, H.; Shen, J.; Meng, X. Langmuir 2009, 25, 11890. (34) Kuroda, K.; Ishida, T.; Haruta, M. J. Mol. Catal. A 2009, 298, 7. (35) Panigrahi, S.; Basu, S.; Praharaj, S.; Pande, S.; Jana, S.; Pal, A.; Ghosh, S. K.; Pal, T. J. Phys. Chem. C 2007, 111, 4596.

(36) Gao, Y.; Ding, X.; Zheng, Z.; Cheng, X.; Peng, Y. Chem. Commun. 2007, 3720. (37) Wunder, S.; Polzer, F.; Lu, Y.; Mei, Y.; Ballauff, M. J. Phys. Chem. C 2010, 114, 8814.

Figure 2. (A) SEM image of a closed AuNT lying on the rough surface film composed of particles of around 50 nm size (pH value of 8.8). (B) SEM image of a thin and damaged surface film obtained at a pH value of 11.6; the arrows point toward pore openings.

Langmuir 2011, 27(1), 430–435

DOI: 10.1021/la104015a

433

Article

Figure 4. (A-C) TEM images of microtome cuts of the polymer embedded AuNTs. (D) Histogram of the particle size distribution. (E) Representative EDS spectrum of the NT wall demonstrating Au to be the dominant element. The Cu and O signals are due to the TEM grid and the organic matrix.

react to 4-aminophenol, which then desorbs and restores the active catalyst surface. To ensure easy reactant flow, polymer templates with large pores of 500 nm size were used. The deposition reaction was carried out with a DMAP concentration of 92 mM at a moderate pH value of 9.6 to minimize damage to the template membrane and to yield small particles at the same time. High pH values caused membrane embrittlement, whereas low DMAP contents and pH values increased the particle size leading to reduced activity.35 In Figure 4, the characterization of the optimized product is given. The approximately 10-15 nm thin NT walls consist of partially agglomerated particles of 5.0 ( 2.1 nm size. The particles consist of Au, although a low amount of residual Ag cannot be ruled out. Compared to the standard reaction,19 grain size was reduced by at least a factor of 6. Also, these filigree extended structures would be extremely difficult to obtain by electrodeposition which relies on more massive metal walls to maintain electrical conductivity. The experimental setup consisted of a glass syringe in combination with an adapter used for cell filtrations. The filter of 1.0 cm solution-accessible diameter was exchanged by cleaned AuNTcontaining polycarbonate membranes. For all experiments, a flow rate of 1 mL/min was applied. The solution composition was chosen similar to related experiments.34,36 Figure 5 summarizes the experimental results. In the beginning, the UV-vis spectrum displays the features expected from a solution containing the 4-nitrophenolate ion. Its bright yellow color corresponds to the intense peak at 400 nm. The spectrum did not change over time until a catalyst-containing membrane 434 DOI: 10.1021/la104015a

Muench et al.

Figure 5. (A) UV-vis spectra of the reaction solution containing 4-nitrophenol and sodium borohydride (reference) and the eluents yielded by one and two flow-throughs. The educt peak at 400 nm is diminishing due to the reduction reaction, while the evolving peak at 300 nm indicates the product formation.37 (B) Apparent pseudo-first-order rate constant as slope of the linear fit of ln(A/At=0) over time.

was applied, proving the restrained nature of the reaction and the AuNPs being the reason for its initiation. After one passage through the membrane, the solution color faded to pale yellow. In the associated UV-vis spectrum, the absorbance of the educt peak is reduced by 85%. At the same time, a new peak at 300 nm arises, indicating the formation of 4-aminophenolate as the product. Because the product does not interfere with the main educt peak and the absorbance is linearly correlated with the concentration, the percental decrease of the educt peak directly equals the conversion of the reaction. After a second flow-through, 99% of the 4-nitrophenol was reduced and the eluent turned colorless. The performance did not drop noticeably during several experiments, and particle leaching could not be detected, indicating the NT membrane to be a stable supported catalyst. This is an advantage over dispersed polymer-NP composites which have to be recovered from the reaction solution and can show aging after few cycles.34 Another benefit is the absence of a protecting agent in our case, while the metal NPs of the mentioned composites are surrounded by polymer shells constraining their reactivity.31,36 A direct comparison of our experiment with other results is complicated due to the differing setup. Like in the mentioned systems31,33-36 the large excess of borohydride guarantees firstorder kinetics with respect to 4-nitrophenol. This leads to a linear correlation of the educt concentration and time on a half-logarithmic scale and allows the calculation of an apparent firstorder rate constant. We performed the calculation for a solution volume of 3 mL, corresponding to related experiments in UV-vis cuvettes.33,34 The as-obtained apparent rate constant adopts a value of 1.3  10-2 s-1. It is among the best results published for Langmuir 2011, 27(1), 430–435

Muench et al.

gold so far (e.g., 4.39  10-2 s-1 for poly(styrenesulfonate)stabilized AuNPs31 or 7.9  10-3 and 7.4  10-3 s-1 for AuNPcovered poly(methyl methacrylate) spheres 34 and polymer nanocapsules,36 respectively). Also, unsupported AuNPs of different mean particle sizes (8-55 nm) have been applied in the nitrophenol reduction. Compared to similarly sized exchange resin supported AuNPs, the activity of the AuNP sol was improved35 as expected by the increase of accessible surface and the more homogeneous catalyst distribution. However, all determined apparent rate constants were relatively low even in the case of small NPs (∼2  10-3 s-1 for a AuNP sol and 3  10-4 s-1 for supported AuNPs of 8 nm mean size).35

Conclusion Because of its water solubility, reversible binding to Au surfaces and its protonation-driven equilibrium with a less strongly protective pyridiniuim ion, the NP capping agent DMAP was reinterpreted as an auxiliary reagent to enhance control over the electroless gold plating process in the disulfitoaurate(I)formaldehyde system. By altering the pH value and the DMAP concentration, deposition speed and grain size can be adjusted. Compared to the standard reaction, the reduced reaction rates allow the formation of homogeneous high aspect ratio NTs consisting of AuNPs of significantly reduced size down to less than 5 nm. Furthermore, DMAP serves as a stabilizing agent and suppresses bath decomposition. Thus, it acts as a powerful

Langmuir 2011, 27(1), 430–435

Article

additive in electroless gold plating baths and allows the reliable synthesis of nanostructured Au films and NTs. Polymer-embedded AuNTs show excellent activity in the reduction of 4-nitrophenol by sodium borohydride, proving optimized electroless plating procedures to be simple and effective methods for the fabrication of tailored nanomaterials. Despite the relatively high flow rate, the model reaction was almost quantitatively completed after a distance of 60 μm. The obtained results emphasize the qualification of electrolessly synthesized metal thin films as catalysts in micro- and nanofluidic devices. A striking advantage of the applied combination of electroless plating and ion track etched polymer templates is its high flexibility, allowing the independent variation of the pore density, length, diameter, and shape as well as the morphology of the Au film. Further studies will point toward the highly controlled electroless deposition of other noble metals for applications in catalysis and sensing. Acknowledgment. We thank Prof. R. Neumann (GSI Helmholtz Centre for Heavy Ion Research) for access to the ion accelerator and the facilities for template preparation. The supply of gold solution by Sch€ utz Dental GmbH is gratefully acknowledged. Note Added after ASAP Publication. This article posted ASAP on December 6, 2010. The third sentence in the abstract and the Figure 3 caption have been revised. The correct version posted on December 8, 2010.

DOI: 10.1021/la104015a

435