Resistance Changes Due to Thermal Coalescence in Colloidal Au

3715

2005, 109, 3715-3718 Published on Web 02/15/2005

Resistance Changes Due to Thermal Coalescence in Colloidal Au/Organic Linker Molecule Multilayer Films Lakshmi Supriya and Richard O. Claus* Macromolecular Science and Engineering Program, Fiber & Electro-Optics Research Center, Mail Code 0356, Virginia Polytechnic Institute and State UniVersity, Blacksburg, Virginia 24061 ReceiVed: December 13, 2004; In Final Form: February 1, 2005

A decrease in the resistance of colloidal Au multilayer films was observed upon heating. These multilayer Au films were fabricated by a layer-by-layer approach, using Au colloids and a bifunctional linker molecule, 1,6 hexanedithiol (HD) on polymer substrates. The resistance of the film prior to heating was 1 MΩ. The films were heated at three different temperatures, 120, 160, and 180 °C. After heating for 12 h, the resistance decreased by 6 orders of magnitude to about 50 Ω. This decrease in resistance was faster at higher temperatures. X-ray photoelectron spectroscopy (XPS) of the unheated films revealed two S 2p peaks corresponding to the Au-S thiolate peak and an oxidized S peak. Upon heating, the relative intensity of the oxidized S peak increased and that of the Au-S peak decreased, indicating an oxidation and desorption of the linker molecules. Scanning electron microscope (SEM) images of the heated films depict coalescence of the spherical Au particles into irregular shapes. The resistance decrease is believed to be due to the desorption of the linker molecule and subsequent coalescence of the Au particles. This method paves a way for controlling the resistance of electrodes on flexible polymer substrates.

Introduction The fabrication of metal films using colloidal particles is a well-established method. Among the various metal colloids, gold and silver colloidal particles have been the subject of many investigations.1-9 Metal colloids have properties that are intermediate between individual atoms and the bulk metals. These colloids have been used to assemble films from solution and have interesting optical and electrical properties.10-12 Applications of these films include, among others, substrates for surface plasmon resonance (SPR)13 including biosensing14 and surface enhanced Raman spectroscopy (SERS).15-17 The formation of films with bulk properties has been achieved by building multilayer films using linker molecules,18,19 increasing gold coverage by a “seeding” method,20 and formation of films using monolayer protected clusters (MPCs).21 Most previous work has centered around the formation of films on substrates such as silicon and glass. In a previous paper we have reported a method for fabricating gold films with good conductive properties on flexible polymer substrates using a solution-based approach.22 This provides a cost-effective method for fabricating electrodes for polymer electronic devices. One method for the fabrication of metal films from solution is the layer-by-layer assembly process. Alternate immersion of the substrate in a colloid solution and a linker molecule solution is used to build the film. The linker molecules are bifunctional and chosen so their end groups have an affinity for the metal. For example, to fabricate Au or Ag films, the end groups that are usually used are -SH, -NH2, or -CN. The substrate is * Corresponding author. Tel: 540-231-7203. Fax: 540-231-4561. E-mail [email protected].

10.1021/jp044321z CCC: $30.25

first modified with silanes having these functional end groups. Immersion in the gold solution causes Au to be adsorbed onto the substrate. Next, immersion in the linker molecule solution causes adsorption of the molecule, which can then adsorb more Au. This process when continued leads to the formation of a continuous film. Using this method Natan et al.19 have fabricated Au films using different linker molecules. The electrical conductivity of the film was found to be related to the length of the linker molecule. Films fabricated using short linkers, less than 8 Å, were extremely conductive with a visual appearance similar to that of bulk Au. Using longer linker molecules, the resistance was found to be 105 times higher. This paper describes the effect of heating on the Au films fabricated on flexible polymer substrates by the layer-by-layer assembly process. When the films were heated, the resistance was observed to decrease sharply. This decrease in resistance occurs due to the thermal desorption of the linker molecules from the film, leading to coalescence of the Au particles, and forming interconnected pathways without the insulating linker molecule in between. The change in the conductivity is observed to occur at as low a temperature as 120 °C; however, the time required increases with decreased temperatures. Experimental Section Fabrication of Au Film. The synthesis and deposition of Au colloids has been described in detail previously.22 In brief, the polymer substrate is plasma treated and treated with an amine terminated silane, which forms covalent siloxane linkages to the polymer substrate. The size of the Au colloidal particles is 18 nm as measured by transmission electron microscopy. The gold film is fabricated by alternate immersion of the silane © 2005 American Chemical Society

3716 J. Phys. Chem. B, Vol. 109, No. 9, 2005

Letters

TABLE 1: Resistance of 10-Bilayer Au-HD as a Function of Time and Temperature time

120 °C

160 °C

180 °C

0 15 min 30 min 90 min 4h 12 h

1.56 MΩ 172 kΩ 125 kΩ 92 kΩ 16 kΩ 64 Ω

1 MΩ 124 kΩ 98 kΩ 83 kΩ 38 kΩ 41 Ω

1.6 MΩ 147 kΩ 39 kΩ 7 kΩ 110 Ω 57 Ω

coated substrate first in Au solution and in 4 mM 1, 6-hexanedithiol (HD) (Aldrich) solution in ethanol. This procedure was repeated to form 10 bilayers of Au-HD. All the heating experiments were performed in air. Characterization. X-ray photoelectron spectroscopy (XPS) data were acquired by a Perkin-Elmer 5400 X-ray photoelectron spectrometer. All the spectra were corrected with reference to the Au 4f peak at 84.0 eV. UV-vis spectra were recorded on a Hitachi U-2001 spectrophotometer. The films were examined by a Leo 1550 field-emission scanning electron microscope (FE-SEM) operating at an accelerating voltage of 5 kV, and a Nanoscope IIIa atomic force microscope (AFM) operating in the tapping mode. The resistance of the film was measured in the two point probe manner by touching the leads of a digital multimeter (Sperry DM-350A) to two places on the film. An average of three measurements was taken for each sample. These values were corroborated by resistance values measured from the slope of a current-voltage (I-V) curve. The I-V curves were obtained using a Keithley 236 source-measure unit; contacts were made by two leads attached to the films, by touching, about 1 cm apart. Results and Discussion Au films were fabricated on a flexible polymer substrate such as Kapton, a polyimide. The surface of the polymer was modified using (aminopropyl)trimethoxysilane (APS) to facilitate the deposition of Au. Details of the surface modification process and Au deposition are given elsewhere.22 Films were fabricated using HD as the linker molecule. After 10 bilayers, the resistance of the film was measured to be about 1 MΩ for a rectangular film (2.5 cm × 1 cm). The films were heated at three different temperatures, 120, 160, and 180 °C, for different lengths of time, in air. Table 1 gives the change in resistance as a function of time for the three different temperatures. The resistance is observed to decrease faster at higher temperatures. The resistivity of the films was computed from the resistance and the thickness of the film.23 The thickness of a 10-bilayer Au-HD film was obtained by cross-sectional SEM images and

was found to be 108 nm. There was no significant change in the thickness observable before and after heating. The resistivity of the unheated film was about 10.8 Ω‚cm, and a lowest value obtained after heating at 180 °C was 4.7 × 10-5 Ω‚cm, whereas typical values of the resistivity after heating for 12 h at different temperatures were about 5.4 × 10-4 Ω‚cm. This value is about 200 times more than that of bulk Au (2.4 × 10-6 Ω‚cm). The effect of heating on the films was characterized by XPS. The samples were heated at 180 °C for 1, 4, and 12 h. Figure 1 shows XPS data (C 1s and S 2p peaks, takeoff angle 90°) for 10-bilayer Au-HD heated at 180 °C for different lengths of time. The unheated sample shows a C 1s peak at 284.6 eV (C-C) and two S 2p peaks at 162.1 and 168.3 eV. The lower energy peak is attributed to S bound to Au as thiolate,24,25 and the higher energy peak is due to an oxidized form of S. It has been observed previously that thiol monolayers self-assembled on Au are oxidized when exposed to the atmosphere.26 The origin of the higher energy peak is believed to be due to the oxidation of the thiol. The C 1s spectra of the heated samples acquired by XPS show the hydrocarbon peak at ca. 284.6 eV and also a lower energy peak centered about 282.0 eV. This anomalous peak is attributed to differential charging occurring in the sample. When a neutralizer was used during spectral acquisition, the Au 4f, C 1s, and O 1s peaks broadened or narrowed depending upon the energy of the neutralizer, and when the energy of the neutralizer was changed, the lower energy peak in the C 1s spectra was removed. This differential charging indicates the presence of well-connected conductive Au regions and regions where the insulating linker molecule has clustered. In the unheated sample, the Au and the insulator molecule are evenly distributed and this prevents the differential charging. This further demonstrates the desorption of the linker molecules from the Au surface leading to coalescence. With an increase in the time of heating, the higher energy S 2p peak increases in intensity compared to the lower energy peak. This indicates oxidation of the thiols attached to the Au particles. Delamarche et al.27 have studied the thermal stability of alkanethiol self-assembled monolayers on Au. They have observed oxidation of thiols and desorption of the alkyl chain above temperatures of 100 °C. Other studies have also shown that there is an oxidation/desorption of thiols assembled on Au substrates.28,29 The intensity of the S peaks (both oxidized and thiolate) decreases upon heating, indicating desorption of molecules. The S/Au ratio in the unheated sample is 0.18 whereas in the sample heated for 12 h it was 0.1. The oxidized S is not attached to the Au and gets desorbed from the surface. This desorption of the

Figure 1. XPS spectra of a 10-bilayer Au-HD film heated at 180 °C for different lengths of time: (a) C 1s (b) S 2p (takeoff angle 90°).

Letters

J. Phys. Chem. B, Vol. 109, No. 9, 2005 3717

Figure 2. SEM of a 10-bilayer Au-HD film (a) unheated and (b) heated at 180 °C for 12 h.

Figure 3. UV-vis spectra of a 10-bilayer Au-HD film unheated and heated at 180 °C for 15 min and 1, 4, and 12 h.

linker molecule causes a coalescence of the Au particles leading to the formation of an interconnected network without the linker molecule. The XPS data for the sample heated at 120 °C also show similar results. However, the decrease in the relative intensity of the lower energy S peak compared to the oxidized S peak is slower, and even after 12 h of heating, there is some amount of S bound to Au as thiolate present, in contrast to the film heated at 180 °C. The S/Au ratio in the unheated film and the film heated for 12 h is 0.22 and 0.14, respectively, leading to the conclusion that there is desorption of the linker molecule but at a slower rate. Figure 2 depicts the SEM images of the unheated film and the film heated at 180 °C. The unheated film shows well-defined spherical Au particles. Upon heating, the spherical particles coalesce into irregular shapes and clusters. However, they are still connected but without the linker molecule, hence leading to higher conductivity. Similar effects were observed using an AFM to image the samples. The heating of the films was also characterized by UV-vis spectroscopy. Figure 3 shows the spectra acquired for 10 bilayers of Au-HD on glass heated at 180 °C for different lengths of time. The spectrum for the unheated sample shows a broad peak with a maximum at 557 nm. The peaks for the Au colloidal film occur because of excitation of surface plasmons. Upon heating, the peak first broadens (15 min, 1 h) and then becomes narrower, with the peak for the 12 h heated sample being narrow and blue-shifted (λmax ) 530 nm). The broadening of the peak is associated with aggregation of the particles.9 When the sample is heated, the particles first come close together, forming aggregates. This causes a broadening of the absorption peak. Upon further heating, the particles coalesce to form larger particles that break up the aggregation. This causes a narrowing of the peak. The peak for the unheated sample is red-shifted

compared to the peak for the sample heated for 12 h, because although the individual particle size is smaller in the unheated sample, there is a greater degree of aggregation. Upon heating further, the particle shapes become more irregular, larger than in the unheated sample and less aggregated. The AFM images taken at different times while heating also show a merging of particles. Current-voltage measurements of the heated and unheated samples show linear behavior indicating the resistance is ohmic. The adhesion properties of the film were qualitatively measured by using the scotch tape peel test. The adhesive tape was pressed firmly onto the film and was slowly removed and the amount of material attached to the tape is an indicator of the adhesion of the film. The unheated films did not have very good adhesion properties. Manual rubbing of the films with fingers removed portions of the film easily, although rinsing the film in water or ethanol did not remove the film, neither did slight bending or twisting the film. Using the adhesive tape test, a major portion of the Au film was transferred to the tape in the unheated samples, indicating poor adhesion of the film. However, the adhesion improved dramatically upon heating. As the time heating was increased, the adhesion of the films was much better to the substrate. For the samples heated at 180 °C, a negligible amount of the film was transferred to the tape and even scratching the surface using a sharp object did not change the resistance values or remove the film. Conclusions In conclusion, it was observed that there is a dramatic change in resistance of nearly 6 orders of magnitude upon heating gold films fabricated on flexible polymer substrates by a layer-bylayer assembly process using organic linker molecules. This decrease in the resistance depends on the temperature and is faster at higher temperatures. Even at temperatures as low as 120 °C, there was an observed decrease, although the time required for the decrease was greater. The lowest resistance observed after heating at 180 °C for 12 h was ca. 50 Ω. This low resistance was achievable at temperatures as low as 120 °C; however, the time required was greater. XPS and SEM results indicate desorption of the linker molecules and a coalescence of the Au particles leading to increased conductivity. Although the fabrication of conductive films by thermal annealing of drop-cast alkanethiolate coated Au MPCs has been reported before,21 this occurs at temperatures around 300 °C, excluding the possibility of formation of these films on lowtemperature polymer materials. This procedure provides an easy method for controlling the resistance of conductive films, and the low temperature makes it attractive for use on polymer substrates. Compared to the “seeding” method of fabricating

3718 J. Phys. Chem. B, Vol. 109, No. 9, 2005 low conductivity films, a series of different resistances can be obtained by simply changing the time and temperature of heating, and by changing the linker molecule. This can be beneficial when a series of different resistors are required in the fabrication of a device. Another advantage is that this method allows the possibility of localized heating of the films to form patterned conducting regions on the film. Acknowledgment. Financial aid from the U.S. Army Research Laboratory and U.S. Army Research Office under contract/grant number DAAD19-02-1-0275 Macromolecular Architecture for Performance (MAP) MURI is acknowledged. We also thank Stephen McCartney, Materials Research Institute, Virginia Tech, for help with the SEM images and Frank Cromer, Department of Chemistry, Virginia Tech, for helpful discussions of the XPS data. References and Notes (1) Tan, B. J.; Sherwood, P. M. A.; Klabunde, K. J. Langmuir 1990, 6, 105. (2) Doron, A.; Katz, E.; Willner, I. Langmuir 1995, 11, 1313. (3) Fan, H.; Lopez, G. P. Langmuir 1997, 13, 119. (4) Nagayama, N.; Itagaki, K.; Yokoyama, M. AdV. Mater. 1997, 9, 71. (5) Mayya, K. S.; Sastry, M. Langmuir 1998, 14, 6344. (6) Park, S.-H.; Im, M. J.-H.; Im, J.-W.; Chun, B.-H.; Kim, J.-H. Microchem. J. 1999, 63, 71. (7) Suyal, G.; Mennig, M.; Schmidt, H. J. Mater. Sci. 2003, 38, 1645. (8) Solecka-Cermakova, K.; Vlckova, B.; Lednicky, F. J. Phys. Chem. 1996, 100, 4954. (9) Grabar, K. C.; Freeman, R. G.; Hommer, M.; Natan, M. J. Anal. Chem. 1995, 67, 735. (10) Halperin, W. P. ReV. Mod. Phys. 1986, 58, 533.

Letters (11) Henglein, A. J. Phys. Chem. 1993, 97, 5457. (12) Schmitt, J.; Ma¨chtle, P.; Eck, D.; Mo¨hwald, H.; Helm, C. A. Langmuir 1999, 15, 3256. (13) Lyon, A. V.; Musick, M. D.; Smith, P. C.; Reiss, B. D.; Pena, D. J.; Natan, M. J. Sens. Actuators, B: Chem. 1999, B54, 118. (14) Natan, M. J.; Lyon, L. A. In Metal Nanoparticles; Feldheim, D. L.; Foss, C. A., Jr., Eds.; Marcel Dekker: New York, 2002; p 183. (15) Angel, S. M.; Myrick, M. L.; Milanovich, F. P. Appl. Spectrosc. 1990, 44, 335. (16) Aroca, R.; Scraba, M.; Mink, J. Spectrochim. Acta, Part A: Mol. Biomol. Spectrosc. 1991, 47, 263. (17) Ahern, A. M. Garrell, R. L. Langmuir 1988, 4, 1162. (18) Musick, M. D.; Keating, C. D.; Keefe, M. H.; Natan, M. J. Chem. Mater. 1997, 9, 1499. (19) Musick, M. D.; Keating, C. D.; Lyon, L. A.; Botsko, S. T.; Pen˜a, D. J.; Holliway, W. D.; McEvoy, T.; Richardson, J. N.; Natan, M. J. Chem. Mater. 2000, 12, 2869. (20) Brown, K. R.; Lyon, L. A.; Fox, A. P.; Reiss, B. D.; Natan, M. J. Chem. Mater. 2000, 12, 314. (21) Wuelfling, W. P.; Zamborini, F. P.; Templeton, A. C.; Wen, X.; Yoon, H.; Murray, R. W. Chem. Mater. 2001, 13, 87. (22) Supriya, L.; Claus, R. O. Langmuir 2004, 20, 8870. (23) Using an average value of R ) 50 Ω for the resistance, and the area A of the film ) 1 cm (width) × 108 nm (thickness), and length l ) 1 cm, the resistivity (F) after heating was computed using the formula, F ) RA/l. (24) Castner, D. J.; Hinds, K.; Grainger, D. W. Langmuir 1996, 12, 5083. (25) Ishida, T.; Choi, N.; Mizutani, W.; Tokumoto, H.; Kojima, I.; Azehara, H.; Hokari, H.; Akiba, U.; Fujihara, M. Langmuir 1999, 15, 6799. (26) Lee, M.-T.; Hsueh, C.-C.; Freund, M. S.; Ferguson, G. S. Langmuir 1998, 14, 6419. (27) Delamarche, E.; Michel, B.; Kang, H.; Greber, Ch. Langmuir 1994, 10, 4103. (28) Kodama, C.; Hayashi, T.; Nozoye, H. Appl. Surf. Sci. 2001, 169170, 264. (29) Kondoh, H.; Kodama, C.; Sumida, H.; Nozoye, H. J. Chem. Phys. 1999, 111, 1175.