Formation Mechanism, Patterning, and Physical Properties of Gold

Mar 16, 2012 - Gold nanoparticle (AuNP) films stacked with individual AuNPs have been shown to exhibit novel electric, plasmonic, and photoelectric ...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/JPCC

Formation Mechanism, Patterning, and Physical Properties of GoldNanoparticle Films Assembled by an Interaction-Controlled Centrifugal Method I-Chi Ni,† Su-Ching Yang,† Cheng-Wei Jiang,‡ Chih-Shin Luo,§ Watson Kuo,‡ Kuan-Jiuh Lin,§ and Shien-Der Tzeng*,† †

Department of Physics, National Dong Hwa University, Hualien 974, Taiwan Department of Physics, National Chung Hsing University, Taichung 402, Taiwan § Department of Chemistry, National Chung Hsing University, Taichung 402, Taiwan ‡

S Supporting Information *

ABSTRACT: Gold nanoparticle (AuNP) films stacked with individual AuNPs have been shown to exhibit novel electric, plasmonic, and photoelectric properties for wide applications. Here, we developed an efficient centrifugal method to assemble desirable large area monolayer, multilayer, and three-dimensional (3D) patterned AuNP films. The formation mechanism of AuNP films under different colloidal interactions was studied. The optimal energy barrier is about 10 kBT for assembling high quality monolayer AuNP films. The shift of localized surface plasmon resonance bands of the films follows a near-exponential distance decay with interparticle spacing s. A red-shift of about 190 nm reveals the strong near-field coupling at s ∼ 0.9 nm. The electrical resistance exponentially increases with s, and exhibits Coulomb charging behavior at low temperature. Furthermore, patterning of AuNP films based on the lift-off technique was achieved and yielded 2D/3D complex structures with submicrometer critical dimension. This assembly method provides a feasible approach in developing future nanodevices and functional nanostructures.

■

INTRODUCTION Metal nanoparticle films consisting of nanoparticles modified with self-assembled monolayers (SAMs) have many interesting electrical and optical properties.1−11 Many of these properties are strongly affected by the interparticle spacing (s) that can be controlled by the lengths of ligand or linker molecules. For instance, the electrical behavior of gold nanoparticle (AuNP) films exhibited a metal-to-insulator transition behavior at s ∼ 1.1 nm.1,2 Meanwhile, the localized surface plasmon resonance (LSPR) bands of the AuNP films can be tunable through nearfield coupling of adjacent AuNPs.3,4 The plasmons may enrich the conductivity of the NP films, which depends on the lengths, functional groups, and electrical characteristics of the SAM molecules.5,6 Furthermore, AuNP films hold great potential for sensing applications, such as chemiresistor-type gas sensors.11−15 Accordingly, it is important to pursue a simple, efficient method to fabricate mono- and multilayer conductive, © 2012 American Chemical Society

plasmonic AuNP films for novel electric, optoelectronic, and nanophotonic devices and chemical sensing applications. To ensure their high performance in such applications, largearea or patterned AuNP films have to be prepared with wellcontrolled particle sizes and shapes as well as with the precision of stacking interparticle spacing in a subnanometer scale. For a long time, self-assembly has been proven a powerful approach to this goal.4,8,16−18 Indeed, some self-assembly techniques demonstrated the formation of monolayer uniform AuNP films. For instance, Langmuir−Blodgett (LB),19,20 solvent evaporation,21−25 and electrostatic26 methods and entropy-driven assembly27 can prepare long-range-ordered nanoparticle monolayers at the liquid−air interface, while convective assembly28 and electrophoresis methods29,30 can also create Received: November 18, 2011 Revised: March 9, 2012 Published: March 16, 2012 8095

dx.doi.org/10.1021/jp211126v | J. Phys. Chem. C 2012, 116, 8095−8101

The Journal of Physical Chemistry C

Article

large-area AuNP films on various substrates. On the other hand, multilayer AuNP films with well-controlled thickness could be achieved by sequential transfer of LB films onto substrates,31 or by the approaches of layer-by-layer (LbL) assembly.1,7,9,32−34 However, these methods either remain unsatisfactory in largescaled uniformity or require many deposition cycles to form a desirable film and are thus time-consuming. To date, a simple, one-step, and high throughput process for fabricating large area monolayers and well-thickness-controlled multilayers is still lacking. There are two key factors limiting the throughput of NP film formation by self-assembly approaches. One is the colloidal interactions between NPs. According to the DLVO (Derjaguin−Landau−Verwey−Overbeek) theory,35 the force between two colloidal particles combines the effects of the van der Waals attraction and the electrostatic repulsion. For small NPs (such as the 12-nm-sized AuNPs used in this work), they are generally thought to be too small to sediment because of strong electrostatic repulsion, which is a benefit in preventing the aggregation of monodispersed NPs in colloidal solution. However, strong electrostatic repulsion becomes unfavorable in the deposition process because the arrival of subsequent NPs is impeded by the surface-adsorbed predecessors, resulting in a low adsorption density in such an interaction-controlled regime. To dilute the electrostatic repulsion, one may apply a neutralization process using surfactants,36 counterions,37 or linker molecules.7 The other key factor is the adsorption kinetics in the diffusion-controlled regime.33,35 Zhai et al. reported the acceleration of AuNP adsorption with the aid of centrifugal force,37 which mainly plays the role of a driving force for NP transport to the substrate.38 However, their method was slowed down by alternately dipping the substrates into the counterion solution to reduce the electrostatic repulsion. In this work, both key factors mentioned above are carefully managed. We not only take advantage of centrifugal force to increase the adsorption rate but also add the idea of interaction control to reduce the electrostatic repulsion during NP deposition. We successfully demonstrated that large area monolayer and multilayer AuNP films with well-controlled thickness can be efficiently assembled in one deposition process. It typically takes less than 30 min, which is a substantial decrease in processing time as compared with other LbL fabrication methods (typically a few hours or days). Moreover, our method economically saves material, since excess AuNPs can be redispersed in water and completely recycled. This assembly process can be combined with conventional photo (or electron-beam) lithography and liftoff techniques to form patterned AuNP structures. The optical and electric properties of the assembled AuNP films with various interparticle spacings were also studied in this work for performance assessment in optoelectronics.

unattached n-mercaptoalkanecarboxylic acids in the colloidal solution. The average radius of the metal core RAu = 6.04 ± 0.40 nm and the interparticle spacing s of MUA modified gold nanoparticles (MUA-AuNPs) = 1.88 ± 0.40 nm were measured with a transmission electron microscope (TEM, JEOL JEM1400). The adsorption density D (particles per area) was acquired by scanning electron microscope (SEM, JSM-6500F, JEOL) images of AuNP monolayer films assembled on the silicon substrates. The zeta potential (ζ) of AuNPs is related to the pH values and ionic strengths (I) of the colloidal solution, as measured by a zeta potential analyzer (Zetasizer Nano ZS, Malvern Instruments). Silicon and glass (EAGLE XG, Corning) substrates were modified with (3-aminopropyl)-trimethoxysilane (APTMS, Fluka) to increase the adhesion to AuNPs. Before the deposition process, the colloidal interaction between AuNPs was controlled by the pH value and the ionic strength I of the colloidal solution. Both decreasing the pH value and increasing I could reduce the energy barrier UB, which is defined as the maximum of the potential energy U and can be computed by the DLVO theory35 (see Supporting Table S1, Supporting Information). Then the gold colloidal solution (typically 5 mL with concentration CNP ∼ 3.3 × 1012 cm−3) was added in a 30 mL centrifuge tube, together with the APTMSmodified substrate laid on a PDMS support in vertical orientation, as shown in Figure 1a. After being centrifuged at

EXPERIMENTAL SECTION Gold nanoparticles used in this work were first synthesized by sodium citrate reduction of HAuCl4,39 and then modified with 3-mercaptopropionic acid [HS−(CH2)2−COOH, MPA], 6mercaptohexanoic acid [HS−(CH2)5−COOH, MHA], 8mercaptooctanoic acid [HS−(CH2)7−COOH, MOA], or 11mercaptoundecanoic acid [HS−(CH2)10−COOH, MUA] by two-step functionalization approaches.40 The resulting molecular encapsulated AuNPs were centrifuged and redissolved in deionized (DI) water several times to remove residual ions and

RESULTS AND DISCUSSION To study the formation mechanism of AuNP films, we controlled UB in the range 3−100 kBT and measured the fractional coverage (θ) of the assembled AuNP monolayer films composed of MUA-AuNPs, where θ is defined as πa2D and a (∼7.0 nm for our MUA-AuNPs) is the effective radius, which is defined as a half of the average center-to-center distance between two adjacent AuNPs in the film. The relationship between θ and UB can be concluded as in Figure 2a. Figure 2b− e shows typical SEM images at UB = 3.6 kBT, 9.3 kBT, 28 kBT,

Figure 1. Schematic diagram of the centrifugal assembly method to assemble (c) multilayer or (d) monolayer AuNP films. The tube is partially filled with PDMS, serving as a substrate support to the vertical direction. The AuNPs are modified with self-assembled monolayers (SAMs) to control the interparticle spacing for the achievement of tunable conductive and plasmonic properties.

8500g for 20 min in the centrifuge (Z233MK-2, Hermle Labortechnik), AuNPs were fully or partially deposited on the substrate, as shown in Figure 1b. Subsequently, two routes lead to different morphologies of the AuNP films: For UB < 8 kBT, AuNPs would be fully deposited on the substrate, and we could directly get a multilayer AuNP film by gently pulling out the sample and drying it in the air, as shown in Figure 1c. On the other hand, for all UB, if the sample was taken out, kept wet, and then immersed in DI water for 30 s to remove the AuNPs above the first layer, we got a monolayer AuNP film, as shown in Figure 1d.

■

■

8096

dx.doi.org/10.1021/jp211126v | J. Phys. Chem. C 2012, 116, 8095−8101

The Journal of Physical Chemistry C

Article

bridge between two spheres42 yielded an estimated capillary attraction force between two 12 nm AuNPs separated within 10 nm also of the order of 10 nN, and the force increases when the interparticle distance decreases. Thus, two individual AuNPs may undergo surface aggregation if the interparticle distance is less than ∼10 nm, as illustrated in Figure 2h. Indeed, the SEM images of the AuNP films assembled at UB = 18−103 kBT (see Supporting Figure S3, Supporting Information) revealed that unconnected AuNPs on the substrate have an interparticle distance larger than 10 nm. To be more precise, a critical surface aggregation distance of about 8−10 nm was inferred from AuNP radial distribution functions in this regime. At the crossover where UB = 8−13 kBT, AuNPs may greatly reduce the interparticle distance and are highly concentrated near the substrate under the centrifugal driving force. The fractional coverage θ can be even larger than the jamming limit (54.7%), which is the maximum fractional coverage achievable in an irreversible random sequential adsorption (RSA) process for hard-sphere-like nanoparticles.35 Two possible pathways to the high coverage growth can be inferred: first, individually adsorbed AuNPs could laterally move on the surface for clustering, or second, AuNPs can become clustered before adsorption. The monolayer film, even having a highest θ of 74%, was of approximate red color before drying, an indication that AuNPs remained individually adsorbed without clustering because of finite electrostatic repulsion. Thus, the pathway of lateral move on the surface is more likely in explaining the high coverage, and drying afterward introduces large scale AuNP aggregations. Indeed, such aggregation is guaranteed by θ = 74%, which yields the average interparticle distance to be about 1.5 nm (core-to-core interparticle spacing s ∼ 3.4 nm), much shorter than the critical surface aggregation distance (8−10 nm) given by previous discussions. As a result, adsorbed AuNPs aggregated in the drying procedure under the driving force of robust capillary attraction, as schematically illustrated in Figure 2g. While the surface aggregation produced many vacancies in the monolayer film, nevertheless the film remained continuous and electrically conductive, as shown in Figure 2c. In the clustering regime where UB was adjusted to less than 8 kBT, the electrostatic repulsion was too weak to prevent AuNPs forming clusters in the colloidal solution, and the color of solution soon changed from ruby red to purple. In the subsequent film deposition, all the AuNPs were deposited on the substrate, resulting in an entirely clear supernatant. The AuNP film was grown with AuNP clusters full of vacancies, and AuNP clusters were only partially in contact with the substrate, as schematically illustrated in Figure 2f. When such a sample was rinsed in DI water, the electrostatic repulsion between AuNPs was rebuilt to make the AuNP clusters dissociated and constituent AuNPs were redispersed in the water except for those that had attached to the substrate. This resulted in a monolayer film with a relatively small θ. We note that, although the film deposited in the clustering regime may have a similar value of θ to those obtained in the dispersing regime (for example, Figure 2b and d), the former presents many discontinuous large AuNP clusters, which is very different from the small clusters in the latter. Besides forming a high coverage monolayer by tuning UB at the crossover, this interaction-controlled centrifugal assembly method is capable of rapidly forming AuNP-multilayer films with high uniformity and good thickness control. This noticeable merit is crucial in various optical and electrical applications, since the thickness of the AuNP-multilayer film

Figure 2. (a) Relationship between θ and UB. Inset: electrostatic repulsion energy Uel(r), van der Waals attraction energy UW(r), and the potential energy U(r) between two colloidal particles (effective radius a = 7.0 nm) calculated using I = 1.5 mM and ζ = −72 mV (i.e., the assembly condition of part d). (b−e) Typical SEM images of monolayer AuNP films deposited with UB ∼ (b) 3.6 kBT, (c) 9.3 kBT, (d) 28 kBT, and (e) 103 kBT. The scale bars indicate 100 nm. (f−i) Schematic diagram of AuNPs assembled with the experimental conditions of parts b−e under the centrifugal force, and the acquired monolayer films after drying.

and 103 kBT, respectively. They show that by, decreasing UB, we were able to tune continuously the AuNP deposition from the dispersing regime to the clustering regime, upon a crossover at UB ∼ 10 kBT. The possible formation mechanism was schematically illustrated in Figure 2f−i. In the dispersing regime where UB > 13 kBT, we found that the centrifugation process successfully concentrated most AuNPs near the substrate but only a small portion of these AuNPs were adsorbed on the substrate surface. Figure 2e exhibits that, even under a large centrifugal driving force, all AuNPs were separately adsorbed on the surface at UB = 103 kBT because of strong electrostatic repulsion, as schematically illustrated in Figure 2i. When UB decreases, reduced electrostatic repulsion yields a more compact AuNP distribution in equilibrium, resulting in an increase in θ. As can be seen in Figure 2d, AuNPs assembled at UB = 28 kBT appeared as slight aggregations on the surface. Such surface aggregation mainly occurred in the drying procedure, in which the color of the AuNP monolayer (assembled on glass substrate) changed from pink to light purple. A plausible explanation is that a robust attractive capillary force is generated via the liquid bridge formed between two neighboring AuNPs so as to overcome the adhesion between AuNP and substrate, resulting in the surface movement and following aggregation of AuNPs. From the studies on interaction between carboxylic acid- and amineterminated SAMs,41 the adhesion force between the COOH modified 12 nm (diameter) AuNP and NH2 modified substrate was estimated to be on the order of 10 nN. By comparison, a theoretical consideration on the capillary force due to a liquid 8097

dx.doi.org/10.1021/jp211126v | J. Phys. Chem. C 2012, 116, 8095−8101

The Journal of Physical Chemistry C

Article

can influence not only their conductance1 but also their photon absorbance and plasmon properties.9,34 We took advantage of the clustering regime (UB < 8 kBT), in which all AuNPs in solution can totally be deposited on the substrate surface to form a multilayer film. Therefore, the film thickness is closely related to the total AuNP numbers in solution, which can be controlled by the AuNP concentration (CNP) and the depth of colloidal solution (t) during assembly. For instance, a film thickness d of about 100 nm, as shown in Figure 3a, was

Figure 4. (a) A photograph of a gold colloidal solution (left) and an MPA-AuNP film on APTMS/glass substrate (right). (b) The relationship between Δλmax and s. Δλmax is the shift of resonance peak from corresponding gold colloid. The data is fit for singleexponential decay Δλmax/λmax = a exp(−x/τ) with amplitude a = 0.84 and decay constant τ = 0.09, where x is the interparticle spacing scaled by the core diameter 2RAu ∼ 12.08 nm. (c) Extinction spectra of colloids and films of AuNPs modified with MUA, MOA, MHA, and MPA. The spectrum is normalized by the extinction at λmax. Here, all AuNP films were about two layers obtained with CNP ∼ 3.3 × 1012 cm−3 and t ∼ 4 mm, typically.

Figure 3. Cross-sectional view SEM images of the multilayer AuNP films with thickness d ∼ (a) 100 nm, (b) 170 nm, and (c) 1.5 μm. The scale bars indicate 100 nm in parts a and b and 500 nm in part c.

obtained with CNP ∼ 3.3 × 1012 cm−3 and t ∼ 16 mm. Figure 3b shows that d of about 170 nm was obtained with CNP ∼ 1.0 × 1013 cm−3 and t ∼ 9 mm. Both of them gave a packing factor (PF = (4/3)πa3CNP(t/d)) approximating 0.74. The high packing factor implies that the loosely structured and vacancycontaining AuNP multilayers formed in the centrifugal procedure, as illustrated in Figure 2f, would collapse into close-packed films after drying. Figure 3c shows that even a μmthick AuNP film can be achieved in a one-step assembly process (with CNP ∼ 5 × 1013 cm−3 and t ∼ 16 mm). Figure 4a shows the area of AuNP films could be larger than 1 cm × 2 cm, which was just limited by the size of our centrifugal tubes. Besides, the color of AuNPs changes from red (as dispersed in solution) to blue (as assembled on glass substrate) due to the interparticle coupling effect on the LSPR of AuNPs.3,4,34,43 Figure 4b and c shows the LSPR properties of the AuNP colloids and films. In Figure 4c, the peak positions of the LSPR bands (λmax) for both pristine and modified AuNP colloids were all around 525 ± 2 nm, indicating the modification of AuNPs had only a weak influence on their LSPR properties. However, λmax red-shifted to about 719, 668, 644, and 608 nm, respectively, for assembled AuNP films (∼2 layers) consisting of MPA-, MHA-, MOA-, and MUA-AuNPs, a hint for strong near-field coupling on the LSPR governed by the interparticle spacing s, about 0.90, 1.27, 1.51, and 1.88 nm, respectively (see the Supporting Information for details). The plasmon-resonance red-shift with decreasing interparticle spacing s in nanoparticle arrays has been theoretically predicted for small s,44 and demonstrated by experiments.3,34 Figure 4b shows that the shift of λmax (Δλmax) behaves in an exponentially dependent way with s, and the fractional plasmon shift could be written as Δλmax/λmax = a exp(−x/τ), where x is s scaled by 2RAu. For our two-layerd amorphous AuNP films, a and τ were about 0.84 and 0.09, respectively. However, smaller amplitude a and larger decay constant τ (∼0.2) in near-exponential distance decay have been reported in experiments and simulations on pairs of gold nanodiscs and nanoparticles.43,45,46 The

discrepancy may be due to the fact that Au nanodiscs and nanoparticles in prior literature had a lower dielectric constant of their surrounding medium (mostly air, organics, and oxide). However, AuNPs in AuNP films are mostly embedded in a host of AuNPs and have a relatively higher effective medium dielectric constant,34 resulting in a larger and more sensitive shift in λmax.46 Our nearly nonselective deposition for AuNP multilayer films should lead to potential patterning of such films by the conventional lift-off process, as schematically shown in Figure 5a, if the film has good adhesion and stability to undergo the lift-off. For a demonstration of this concept, micro- or nanoscale photoresist patterns were first created by the conventional photo lithography or electron-beam (e-beam) lithography method on SiO2/Si substrates. The exposed oxide surfaces were then modified with APTMS molecules to assist the film adhesion. After centrifugal assembly and well drying, the film was put in acetone with ultrasonication for some minutes (i.e., the lift-off process, see the Supporting Information for details). Figure 5b−d shows two-dimensional (2D) patterned AuNP films fabricated by the above process, confirming the robustness of the film on the NH2-modified surfaces and revealing a high spatial resolution much better than 100 nm in such patterning. The same fabrication process can also form 3D AuNP structures by depositing thicker AuNP film, as the assembled pillared structure of AuNPs shown in Figure 5e. As the deposited AuNP film is much thicker than the patterned photoresist, AuNP film can form an air bridge after the photoresist underneath is gently remove by acetone, as demonstrated in Figure 5f. The prior patterning process also can easily be integrated with other top-down fabrication inprior, for example, a set of premade electrodes, as shown in Figure 6a and b, proving it superior in fabricating functional nanodevices for a wide array of applications. 8098

dx.doi.org/10.1021/jp211126v | J. Phys. Chem. C 2012, 116, 8095−8101

The Journal of Physical Chemistry C

Article

Figure 6. (a) A SEM image of a rectangular AuNP film assembled on SiO2/Si substrate with Cr/Au electrodes. (b) A SEM image of a typical AuNP device for the measurements of charge conduction properties. The film was MUA-AuNPs (∼2 layers) assembled on SiO2/Si substrate with Cr/Au multiple electrodes. (c) Resistances of AuNP devices with different surface modifications at room temperature. (d) IV curves for a MUA-AuNP device at temperatures from 10 to 300 K. The inset shows low-bias resistance R vs 1/T plot. The fitting uses data with temperature T ranging from 50 to 300 K. The scale bars indicate 50 μm in part a and 200 nm in part b.

spacing s: R ∝ exp(βs). The tunneling decay constant β obtained from the data fitting is about 1.18 Å−1, allowing a comparison with the value in the alkanedithiol case. On alkanedithiol single molecular junctions and on assembled AuNPs cross-linked with different alkanedithiol linkers HS− (CH2)n−SH, previous works reported R ∝ exp(βnn) with βn ∼ 1.0 ± 0.1.1,47 Here n is the number of carbon atoms of the linker molecule, and βn the tunneling decay constant per carbon atom. Given the specific interparticle spacing per additional carbon of about 1.2 Å,3,48 β for alkanedithiols is about βn/1.2 Å = 0.75−0.92 Å−1, significantly smaller than β for our mercaptoalkanecarboxylic acids. However, our β-value (∼1.18 Å−1) falls in the range between the through-bond decay constant βTB ∼ 0.91 Å−1 and the through-space decay constant βTS ∼ 1.31 Å−1 for alkanethiols.49 Thus, unlike that in the alkanedithiol case, the charge transport mechanism in our case may partly be contributed from through-space tunneling.50 Figure 6d shows the current−voltage (IV) curves of a MUAAuNP device at temperatures from 10 to 300 K. The nonlinear IV characteristics at low temperatures can arise from singleelectron charging of AuNPs. Zabet-Khosousi et al.1 have reported the low-bias resistance R would exhibit Coulomb charging behavior R ∝ exp(EC/kBT) at temperatures above ∼100 K, where EC is the charging energy. For a single AuNP embedded in AuNP assemblies, the charging energy can be written as EC = (e2/8πεε0)[(1/RAu) − (1/(s + RAu))],1,51 in which e is the charge of an electron, ε the dielectric constant, s the interparticle spacing, and RAu the AuNP core radius. As ε = 2.6 is assumed for MUA, we get EC = 10.9 meV for our MUAAuNP device. The inset in Figure 6d shows that log(R) well follows T−1 (Arrhenius) behavior at temperatures above 50 K. At lower temperatures, it diverges from Arrhenius behavior and would tend to T−2 behavior.1 From the fitting of data with T above 50 K, we get EC ∼ 8.7 mV, which is of the same order as the prior calculated value.

Figure 5. (a) Schematic drawing of fabricating a patterned nanostructure by the combination of the bottom-up centrifugal assembly method and the top-down lithography method. In this work, a patterned photoresist was first created by the conventional photo or e-beam lithography method. Then the exposed surfaces were modified with APTMS. After deposition of AuNP film by the centrifugal assembly method, the film on the photoresist was lifted off, resulting in a patterned AuNP film. (b−d) SEM images of 2D patterned AuNP films. The patterns were created by (b) the photo lithography method and (c, d) the e-beam lithography method. The scale bars indicate 20 μm in part b; 200 nm (upper) and 5 μm (lower) in part c; and 200 nm (upper) and 2 μm (lower) in part d. (e, f) Side-view (10° tilt angle) SEM images of 3D AuNP assembled structures. The scale bars indicate 200 nm in parts e and f.

For studying their charge conduction properties, AuNP films with different surface modifications were assembled on SiO2/Si substrates with premade electrodes, which were 20 nm/50 nm Cr/Au electrodes gapped with hundreds of nanometers fabricated by using e-beam lithography and a lift-off technique. Two types of devices were prepared: devices with multiple electrodes, as shown in Figure 6b, and devices with parallel electrodes. However, both types showed almost the same transport behavior. For the same AuNP surface modification, the resistances (R) at room temperature measured between different pairs of electrodes were on the same order of magnitude. On the other hand, a great variation in R for AuNP devices with different surface modifications was found, as shown in Figure 6c. This strong impact of the molecule linkage on the charge conduction is mainly attributed to the exponential increase of tunneling resistance with interparticle 8099

dx.doi.org/10.1021/jp211126v | J. Phys. Chem. C 2012, 116, 8095−8101

The Journal of Physical Chemistry C

■

Table 1. Important Parameters of the AuNP Films Described in Text type of AuNP film carbon number, n interparticle spacing, s (nm) LSPR shift, Δλmax (nm) resistivity at 25 °Ca (Ω-cm)

MPA

MHA

MOA

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +886-3-863-3737. Fax: +886-3-863-3690.

MUA

3

6

8

11

Notes

0.90

1.27

1.51

1.88

The authors declare no competing financial interest.

194 ± 17

143 ± 9

119 ± 8

83 ± 13

10−4−10−3

10−1−100

100−101

103−104

■

ACKNOWLEDGMENTS This work was financially supported by the National Science Council of Taiwan under Grant Nos. NSC99-2112-M-259-005MY3 and NSC99-2112-M-005-007-MY3.

■

a

The resistivity was measured using 20-μm-width two-layerd AuNP films on parallel electrodes with separations from 5 to 50 μm.

■

REFERENCES

(1) Zabet-Khosousi, A.; Trudeau, P.-E.; Suganuma, Y.; Dhirani, A.-A.; Statt, B. Phys. Rev. Lett. 2006, 96, No. 156403. (2) Brust, M.; Bethell, D.; Schffrin, D. J.; Kiely, C. J. Adv. Mater. 1995, 7, 795−797. (3) Chen, C.-F.; Tzeng, S.-D.; Chen, H.-Y.; Lin, K.-J.; Gwo, S. J. Am. Chem. Soc. 2008, 130, 824−826. (4) Ghosh, S. K.; Pal, T. Chem. Rev. 2007, 107, 4797−4862. (5) Nakanishi, H.; Bishop, K. J. M.; Kowalczyk, B.; Nitzan, A.; Weiss, E. A.; Tretiakov, K. V.; Apodaca, M. M.; Klajn, R.; Stoddart, J. F.; Grzybowski, B. A. Nature 2009, 460, 371−375. (6) Banerjee, P.; Conklin, D.; Nanayakkara, S.; Park, T.-H.; Therien, M. J.; Bonnell, D. A. ACS Nano 2010, 4, 1019−1025. (7) Brust, M.; Bethell, D.; Kiely, C. J.; Schiffrin, D. J. Langmuir 1998, 14, 5425−5429. (8) Shipway, A.; Katz, E.; Willner, I. ChemPhysChem 2000, 1, 18−52. (9) Lin, M.-H.; Chen, H.-Y.; Gwo, S. J. Am. Chem. Soc. 2010, 132, 11259−11263. (10) Daniel, M.-C.; Astruc, D. Chem. Rev. 2004, 104, 293−346. (11) Joseph, Y.; Besnard, I.; Rosenberger, M.; Guse, B.; Nothofer, H.G.; Wessels, J. M.; Wild, U.; Knop-Gericke, A.; Su, D.; Schlögl, R.; Yasuda, A.; Vossmeyer, T. J. Phys. Chem. B 2003, 107, 7406−7413. (12) Wohltjen, H.; Snow, A. W. Anal. Chem. 1998, 70, 2856−2859. (13) Zamborini, F. P.; Leopold, M. C.; Hicks, J. F.; Kulesza, P. J.; Malik, M. A.; Murray, R. W. J. Am. Chem. Soc. 2002, 124, 8958−8964. (14) Krasteva, N.; Besnard, I.; Guse, B.; Bauer, R. E.; Müllen, K.; Yasuda, A.; Vossmeyer, T. Nano Lett. 2002, 2, 551−555. (15) Chow, E.; Müller, K.-H.; Davies, E.; Raguse, B.; Wieczorek, L.; Cooper, J. S.; Hubble, L. J. J. Phys. Chem. C 2010, 114, 17529−17534. (16) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103−1169. (17) Caruso, F. Colloids and Colloid Assemblies; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2004. (18) Claridge, S. A.; Castleman, A. W. Jr.; Khanna, S. N.; Murray, C. B.; Sen, A.; Weiss, P. S. ACS Nano 2009, 3, 244−255. (19) Tao, A. R.; Huang, J. X.; Yang, P. D. Acc. Chem. Res. 2008, 41, 1662−1673. (20) Tsai, H.-J.; Lee, Y.-L. Soft Matter 2009, 5, 2962−2970. (21) Bigioni, T. P.; Lin, X.-M.; Nguyen, T. T.; Corwin, E. I.; Witten, T. A.; Jaeger, H. M. Nat. Mater. 2006, 5, 265−270. (22) Lin, X. M.; Jaeger, H. M.; Sorenson, C. M.; Klabunde, K. J. J. Phys. Chem. B 2001, 105, 3353−3357. (23) Santhanam, V; Liu, J.; Agarwal, R.; Andres, R. P. Langmuir 2003, 19, 7881−7887. (24) Martin, M. N.; Basham, J. I.; Chando, P.; Eah, S.-K. Langmuir 2010, 26, 7410−7417. (25) Desireddy, A.; Joshi, C. P.; Sestak, M.; Little, S.; Kumar, S.; Podraza, N. J.; Marsillac, S.; Collins, R. W.; Bigioni, T. P. Thin Solid Films 2011, 519, 6077−6084. (26) Sastry, M.; Rao, M.; Ganesh, K. N. Acc. Chem. Res. 2002, 35, 847−855. (27) Dong, A.; Ye, X.; Chen, J.; Murray, C. B. Nano Lett. 2011, 11, 1804−1809. (28) Prevo, B. G.; Velev, O. D. Langmuir 2004, 20, 2099−2107.

CONCLUSIONS We have demonstrated that desirable monolayer and multilayer AuNP films can be achieved by using a simple centrifugal method combined with energy barrier control. When the energy barrier UB is about 10 kBT, the fractional coverage of monolayer film can reach up to 74%, which is larger than the jamming limit (54.7%). By controlling UB < 8 kBT, multilayer AuNP films could be achievable in a one-step deposition process. The thickness of multilayer AuNP films can be tunable via the control of concentration of nanoparticles or the thickness of colloid solution. It was found that AuNPs are highly close-packed in the dried multilayer film. The sample size of AuNP films can reach up to centimeter scales. When the centrifugal method combines with conventional photo or electron-beam lithography methods and lift-off techniques, 2D and 3D designed complex AuNP assembled structures can be fabricated. The optical and electrical properties of the film can also be well controlled. Both the localized surface plasmon resonance bands and the electrical conduction properties of AuNP films can be tunable by interparticle spacing s, which can be controlled by the surface modification of AuNPs. A significant red-shift that follows a near-exponential distance decay with s reveals a strong near-field coupling on the LSPR. The electrical conduction exhibits Coulomb charging behavior, displaying that the assembled films have great potential for applications to nanoelectronic devices. We envision the versatile optical and electrical properties of these AuNP films to make a strong impact on future nanoscaled optoelectronics. In summary, this centrifugal assembly method is quite promising to assemble 2D/3D functional nanostructures and nanodevices. Although only cm-sized samples were demonstrated because of the size limitation of our centrifugal tube, it is certainly achievable to form large-scaled films by using commercially available fixed-angle rotors or swinging-bucket rotors equipped with larger tubes. Moreover, although only gold nanoparticles in aqueous solutions are demonstrated here as building blocks to fabricate patterned nanostructures, it is practical to create a greater variety of nanomaterial-based artificial structures by applying the present centrifugal assembly method on other nanomaterials dispersed in aqueous or organic solvents.

■

Article

ASSOCIATED CONTENT

S Supporting Information *

Experimental details, energy barrier calculations, characterization of the films, and data analysis details. This material is available free of charge via the Internet at http://pubs.acs.org. 8100

dx.doi.org/10.1021/jp211126v | J. Phys. Chem. C 2012, 116, 8095−8101

The Journal of Physical Chemistry C

Article

(29) Isozaki, K.; Ochiai, T.; Taguchi, T.; Nittoh, K.; Miki, K. Appl. Phys. Lett. 2010, 97, No. 221101. (30) Giersig, M.; Mulvaney, P. Langmuir 1993, 9, 3408−3413. (31) Mayya, K. S.; Patil, V.; Sastry, M. Langmuir 1997, 13, 2575− 2577. (32) Musick, M. D.; Keating, C. D.; Lyon, L. A.; Botsko, S. L.; Peña, D. J.; Holliway, W. D.; McEvoy, T. M.; Richardson, J. N.; Natan, M. J. Chem. Mater. 2000, 12, 2869−2881. (33) Qi, Z.; Honma, I.; Ichihara, M.; Zhou, H. Adv. Funct. Mater. 2006, 16, 377−386. (34) Ung, T.; Liz-Marzán, L. M.; Mulvaney, P. J. Phys. Chem. B 2001, 105, 3441−3452. (35) Yuan, Y.; Oberholzer, M. R.; Lenhoff, A. M. Colloids Surf., A 2000, 165, 125−141. (36) Khatri, O. P.; Murase, K.; Sugimura, H. Langmuir 2008, 24, 3787−3793. (37) Zhai, J.; Wang, Y.; Zhai, Y.; Dong, S. Nanotechnology 2009, 20, No. 055609. (38) Dokou, E.; Barteau, M. A.; Wagner, N. J.; Lenhoff, A. M. J. Colloid Interface Sci. 2001, 240, 9−16. (39) Slot, J. W.; Geuze, H. J. Eur. J. Cell Biol. 1985, 38, 87−93. (40) Lin, S. Y.; Tsai, Y. T.; Chen, C. C.; Lin, C. M.; Chen, C. H. J. Phys. Chem. B 2004, 108, 2134−2139. (41) Thomas, R. C.; Houston, J. E.; Crooks, R. M.; Kim, T.; Michalske, T. A. J. Am. Chem. Soc. 1995, 117, 3830−3834. (42) Rabinovich, Y. I.; Esayanur, M. S.; Moudgil, B. M. Langmuir 2005, 21, 10992−10997. (43) Su, K.-H.; Wei, Q.-H.; Zhang, X.; Mock, J. J.; Smith, D. R.; Schultz, S. Nano Lett. 2003, 3, 1087−1090. (44) Zhao, L. L.; Kelly, K. L.; Schatz, G. C. J. Phys. Chem. B 2003, 107, 7343−7350. (45) Reinhard, B. M.; Siu, M.; Agarwal, H.; Alivisatos, A. P.; Liphardt, J. Nano Lett. 2005, 5, 2246−2252. (46) Jain, P. K.; Huang, W.; El-Sayed, M. A. Nano Lett. 2007, 7, 2080−2088. (47) Xu, B.; Tao, N. J. Science 2003, 301, 1221−1223. (48) Martin, J. E.; Wilcoxon, J. P.; Odinek, J.; Provencio, P. J. Phys. Chem. B 2000, 104, 9475−9486. (49) Slowinski, K; Chamberlain, R. V.; Miller, C. J.; Majda, M. J. Am. Chem. Soc. 1997, 119, 11910−11919. (50) Salomon, A.; Cahen, D.; Lindsay, S.; Tomfohr, J.; Engelkes, V. B.; Frisbie, C. D. Adv. Mater. 2003, 15, 1881−1890. (51) Abeles, B.; Sheng, P.; Coutts, M.; Arie, Y. Adv. Phys. 1975, 24, 407−461.

8101

dx.doi.org/10.1021/jp211126v | J. Phys. Chem. C 2012, 116, 8095−8101