- Email: [email protected]
Optical and electrical properties of polymer metal nanocomposites prepared by magnetron co-sputtering
Thin Solid Films 515 (2006) 801 – 804 www.elsevier.com/locate/tsf
Optical and electrical properties of polymer metal nanocomposites prepared by magnetron co-sputtering Ulrich Schu¨rmann, Haile Takele, Vladimir Zaporojtchenko, Franz Faupel * Chair for Multicomponent Materials, Technische Fakulta¨t der CAU, Kaiserstrasse 2, D-24143 Kiel, Germany Available online 23 January 2006
Abstract Polymer-metal nanocomposite films were prepared by co-sputtering from two independent magnetron sources as well as from a silverpolytetrafluoroethylene (PTFE) composite target. By sputtering from two magnetrons we prepared both, samples with a metal gradient and homogenous composite films. The dielectric constant depends on the metal filling factor and increases approaching the percolation threshold. The resistivity drops from 107 to 10 3 V cm over a narrow range of metal content. The thin composite films show a strong optical absorption in the visible region and a red shift with increasing silver filling. We have produced Bragg reflectors using multilayer systems of composites and pure sputtered PTFE. D 2005 Elsevier B.V. All rights reserved. PACS: 52.80.Pi; 73.61.Ph; 78.66.Sq Keywords: PTFE; Silver nanoparticles; Electrical properties; Optical properties; Bragg reflector
1. Introduction Nanocomposite materials consisting of dispersed metal nanoclusters in a matrix of insulating material show extraordinary physical properties and have been proposed for optical [1,2], electrical [3] and medical applications [4] as well as for data storage [5]. These properties of the composites are very sensitive to small changes in the amount of metal and in the size and shape of the nanoparticles. This sensitivity leads to drastic changes in the electrical and optical properties of the material, which can thus be used as sensors or switching devices. Polymers could be a good choice as a matrix for the metal nanoparticles in order to stabilize the particle size and the growth along with easy processability and low-cost fabrication. Different techniques are available for producing such composite films [6]. Several techniques are based on physical vapor deposition (PVD) including the combination of evaporation and sputtering of metals with plasma polymerization [7] as well as magnetron sputtering from a composite target [8] and coevaporation [9,10].
* Corresponding author. Tel.: +49 431 880 6226; fax: +49 431 880 6229. E-mail address: [email protected] (F. Faupel). 0040-6090/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2005.12.249
In the present work we used a PVD technique based on cosputtering from two magnetron sources as well as from a multicomponent target to analyze optical and electrical properties with focus on the properties near the percolation threshold. High deposition rates can be realized with this sputter technique. Another advantage of magnetron sputtering is the good uniformity of the composite films. The advantage compared to chemical deposition techniques is the absence of residual solvents in the prepared thin films. A RF magnetron source was used for sputtering the polytetrafluoroethylene (PTFE) to prevent charging of the polymer target. Our work was focused on the dependence of the optical and electrical properties specially near the percolation threshold on metal content and preparation conditions. With alternating layers of pure sputtered polymer and composite we were able to prepare Bragg reflectors due to the varying dielectric constant of the two materials. 2. Experimental Composites were prepared with different configurations and arrangements of the sputter sources. Both sputtering from a composite target and from two magnetron sources were performed.
802
U. Schu¨rmann et al. / Thin Solid Films 515 (2006) 801 – 804
All experiments were carried out in a metal vacuum chamber, which was initially evacuated to a pressure below 10 7 Torr. Quartz-crystal monitors were installed in the chamber to control the deposition rates. Polymer sputtering was performed with a RF magnetron system. A Dressler CESAR 136 RF-Generator (13.560 MHz) was coupled to the planar magnetron sputter source ION’X-2WUHV_9109 (Thin Film Consulting). Normally, a RF power of 50 W was applied to the PTFE target to achieve deposition rates around 10 nm min 1. For DC sputtering of pure silver an Advanced Energy MDX 500 sputter source was coupled to the planar magnetron sputter source ION’X-2WUHV_9011 (Thin Film Consulting). The DC power ranged between 4 and 12 W. The targets had a diameter of 2 in. and they were mounted on a cooled holder. The distance between the target and the substrate was 80 mm. Silver/PTFE-multicomponent targets were tested applying different set-ups. We changed the geometry of silver foils or wires on the top of a PTFE sputter target. Thin silver foils clamped onto the surface have the advantage of an easy filling factor control in the deposited film by changing the silver area on the sputter target. The so-called filing factor is defined as the metal volume fraction in the composite material. [6] We can reach high filling factors up to values of over 50% and vary the silver amount easily. However, some problems occurred during sputtering. The homogeneity of the films was not sufficient due to the large areas on the target covered by silver foil. Because of the small thickness of the used silver foils these were partly used up after a few experiments, so the reproducibility decreased. While this problem could be eliminated by using thicker silver foils, additionally, some sputtered silver covered part of the polymer target surface after a few experiments. This resputtering effect increased the filling factor of the deposited composite films with time. Another set-up using silver pins of 1 mm diameter plugged into the PTFE target minimized this problem. The pins were arranged in rings surrounding the center of the target. The ring geometry was chosen because the sputtered area on the target had also a ring-like shape due to the magnetron of the sputter sources. Depending on the number of silver pins and their position one can prepare composite films with a wide range of filling factors. The filling factor and the deposition rate could also be adjusted by changing the deposition parameters such as the input power and the sputter gas (argon) pressure. By increasing the input power the deposition rate of the polymer increased stronger than the metal rate and the filling factor decreased whereas an increase of the pressure had the opposite effect. Such a sputtering setup with a multicomponent target is useful in the case of preparing nanocomposites with more then one metal component, e.g. alloys out of gold, silver, or copper for optical applications. One can easily control the ratio of different metals inside the polymer matrix just by changing the ratio of the metal pins on the target. Better reproducibility and control of the filling factor in the case of one metal component can be achieved by using independent magnetron sputter sources aiming at the substrate from opposite directions with an angle of 50- to the substrate plane. This setup is described in detail in our previous article [11]. Here we could tailor the filling factor by adjusting the
deposition ratio of the metal and the polymer from the two sputter sources. With a rotatable sample holder we could achieve homogenous samples. Without rotation we got samples with a thickness gradient and a filling factor gradient due to the arrangement of the sources. We used energy dispersive X-ray spectroscopy (EDX) to determine the filling factor. Uniform composite samples with different filling factors which were characterized gravimetrically were used as standards for the EDX measurements. The exact procedure was described recently [11]. Layers of composite films with a maximum thickness of 100 nm were deposited on glass slides or transparent polymer foils for UV –Vis measurements. The optical properties were studied using an UV/Vis/NIR-spectrometer Lambda900 (Perkin Elmer). We used a Keithley Model 6485 picoammeter to carry out the resistance measurements. Capacitor measurements to determine the dielectric constant were performed with the automatic RCL meter PM 6306 (Fluke). Transmission electron microscopy (TEM) measurements were carried out using a Philips CM 30 microscope. Films with a maximum thickness of about 50 nm were deposited onto carbon-covered copper grids. Slices of the multilayer films for the cross-sectional TEM images were cut using a Reichert ultracut S ultramicrotome. 3. Results and discussion The morphology of a polymer/metal composite with a filling factor below the percolation threshold prepared by cosputtering from two magnetron sources is shown in the TEM image in Fig. 1. The clusters have a size between 5 and 10 nm. They are spherical and are separated from each other. The electrical properties changed with the filling factor from a polymer-like to a metal-like material. Thereby, the transition is sharp and occurs in a relatively narrow range near the percolation threshold, where a path of percolated particles is formed through the material for the first time. The resistivity changes over several orders of magnitude for a change in the amount of metal of only a few percent. We prepared samples with a gradient in the filling factor and in the thickness. The thickness thereby varied from around 200 nm at the silver side to around 400 nm at the polymer side. On these samples we observed a resistivity drop from values around 5 106 to
Fig. 1. TEM micrograph of a silver-PTFE nanocomposite ( f å 0.2) prepared by co-sputtering from two magnetron sources.
U. Schu¨rmann et al. / Thin Solid Films 515 (2006) 801 – 804
803
Fig. 2. Change of the resistivity (solid squares) and plasmon resonance wavelength (open circles) with the filling factor.
5 10 3 V cm. The percolation threshold was located at a filling factor of f å 0.35. This is in agreement with other works. For example, a theoretical approach for spherical clusters gives the percolation threshold with f å 0.29 [12]. In experiments with gold nanoclusters in different polymer matrices the percolation threshold was found at f å 0.43 (Teflon AF) and f å 0.20 (Poly(a-methylstyrene)) [13]. The dielectric constant ( of the pure sputtered PTFE and of the composite were determined using a capacitor set-up with a completely metalized (Cu/Al) polymer substrate as bottom electrode. The top electrode consisted of silver sputtered through a mask with small holes (diameter 1 mm). ( changed from 2.1 T 0.1 for the pure polymer to more than 5.0 near the percolation threshold. The optical properties were also studied as a function of the filling factor. Even films with a thickness of a few nanometers had an intensive yellow color (Fig. 2). This color effect is based
upon the so-called surface plasmon resonance or particle plasmon resonance (PPR). When a metal-dielectric nanocomposite is excited by light, photons are coupled at the metaldielectric interface, causing an induced charge density oscillation that creates a strong absorption maximum at a particular wavelength. For silver and other noble metal particles in polymers the resonance frequencies are in the visible range. The position of the absorption maximum depends on the cluster morphology like size, shape, and particle distance as well as on the matrix material. The absorption maxima for small filling factors should be found theoretically between 350 and 420 nm depending on the surrounding dielectric matrix [14]. We observed a red shift of the absorption maximum from 405 nm for small filling factors to more than 500 nm for filling factors above the percolation threshold. There was only a slight shift of the particle plasmon resonance frequency observed for
Fig. 3. Optical properties: UV – Vis reflectance spectrum of Bragg multilayer system with five polymer/composite double layers; inset: UV – Vis example spectra for two different filling factors with absorption peaks due to the plasmon resonance.
804
U. Schu¨rmann et al. / Thin Solid Films 515 (2006) 801 – 804
swelling of the matrix. Therefore such kind of Bragg reflectors may be used as optical sensors [2]. 4. Conclusion
Fig. 4. Cross-sectional TEM micrograph of a multilayer system with four double layers. (a) composite layer with silver cluster, (b) sputtered PTFE, (c) polymer substrate.
small filling factors. Near the percolation a stronger dependence of the resonance frequency on the filling factor was observed. With increasing metal concentration inside the composite film near the percolation threshold the gaps between single clusters become smaller and the cluster size and shape can also change due to coalescence of the particles. The resulting significant red shift of the absorption maximum is in good agreement with theoretical models [14]. The changes in the resistivity and the shift of the plasmon resonance with the filling factor is shown in Fig. 2. Representative spectra for two different filling factors are given in the inset of Fig. 3. We prepared multilayer structures composed of alternating layers of pure sputtered PTFE and an Ag-PTFE-composite with low and high refractive indices, which leads to a Bragg reflector set-up. Fig. 4 shows a cross-sectional TEM image of such a multilayer set-up with four double layers. The selective reflectivity results from the constructive interference of reflections from the different layer interfaces. The refractive index n of the sputtered polymer was determined with ellipsometry as n = 1.374 [11]. The value of the refractive index of the composite is larger than that of the pure polymer and it changes with the silver fraction. Because the position of the maximum of reflectance depends on the layer thickness and on the refractive indices of the alternating layer, one can tailor the wavelength of reflectance by changing these parameters. So far we have no exact values for the refractive index of the composites with changing filling factor. Therefore, the determination of the reflection maximum is only an estimate. However, a qualitative change in the parameter’s thickness and filling factor gave the predicted shift. By increasing the number of the layers of such a Bragg reflector the reflectance can be enhanced to even higher values. Also, the peak becomes sharper by improving the Bragg conditions and increasing the number of the layers. A typical spectrum of a very simple Bragg reflector containing five double layers is given in Fig. 3 to demonstrate the feasibility. And even with only five double layers we reached a reflectance of more than 80%. The maximum was located in the near infrared region, but by changing the parameters it was also possible to shift it to the visible region. It is known that the refractive index of such composites is sensitive to the vapor of organic solvents due to
We have shown that it is possible to synthesize nanocomposites consisting of a polymer and silver by sputtering from a composite target as well as from two independent magnetron sources. Different geometries were checked in case of the composite target in order to optimize deposition rate, filling factor, metal distribution, and reproducibility. An arrangement of the metal as rings of pins showed the best results. In any case, there is the problem of silver resputtering, which changes the target slightly after every experiment. In order to eliminate this problem, we developed a more reliable process based on co-sputtering from two targets. We obtained relatively high deposition rates and a good reproducibility and uniformity of the deposited films. The dielectric constant increases with increasing filling factor up to the percolation threshold, where the electrical and optical properties change drastically from insulating to metallic. The resistivity and the optical absorption can be tailored via the filling factor. Using the different refractive indices of the pure polymer and the composite, one can thus easily produce Bragg reflectors with tailored properties. Acknowledgements The authors are grateful to S. Rehders for his technical support in developing the vacuum chambers and to the German Science Foundation (DFG) for financial support under grant number FA 234/10-1. References [1] A. Biswas, O.C. Aktas, U. Schu¨rmann, U. Saeed, V. Zaporojtchenko, F. Faupel, Appl. Phys. Lett. 84 (14) (2004) 2655. [2] A. Convertino, A. Capobianchi, A. Valentini, E.N.M. Cirillo, Adv. Mater. 15 (2003) 1103. [3] P.M. Ajayan, L.S. Schadler, P.V. Braun, Nanocomposite Science and Technology, Wiley-VCH, Weinheim, 2003. [4] F.-R.F. Fan, A.J. Bard, J. Phys. Chem., B 106 (2002) 279. [5] J. Ouyang, C.-W. Chu, R. Szmanda, L. Ma, Y. Yang, Nat. Mater. 3 (2004) 918. [6] A. Heilmann, Polymer Films with Embedded Metal Nanoparticles, Springer, Berlin, 2003. [7] H. Biederman, D. Slavinska, Surf. Coat. Technol., A 125 (2000) 371. [8] R.A. Roy, R. Messier, S.V. Krishnaswamy, Thin Solid Films 109 (1983) 27. [9] K. Behnke, T. Strunskus, V. Zaporojtchenko, F. Faupel, Proc. Microbiol. Mater. (2000) 1052. [10] A. Biswas, Z. Ma´rton, J. Kruse, J. Kanzow, V. Zaporojtchenko, F. Faupel, T. Strunskus, Nano Lett. 3 (1) (2003) 69. [11] U. Schu¨rmann, W. Hartung, H. Takele, V. Zaporojtchenko, F. Faupel, Nanotechnology 16 (2005) 1078. [12] E.J. Garboczi, K.A. Snyder, J.F. Douglas, M.F. Thorpe, Phys. Rev., E 52 (1) (1995) 819. [13] H. Takele, U. Schu¨rmann, V. Zaporojtchenko, F. Faupel, accepted for publication in Europ. Journ. Phys. [14] U. Kreibig, M. Vollmer, Optical Properties of Metal Clusters, Springer, Berlin, 1995.
Optical and electrical properties of polymer metal nanocomposites prepared by magnetron co-sputtering Ulrich Schu¨rmann, Haile Takele, Vladimir Zaporojtchenko, Franz Faupel * Chair for Multicomponent Materials, Technische Fakulta¨t der CAU, Kaiserstrasse 2, D-24143 Kiel, Germany Available online 23 January 2006
Abstract Polymer-metal nanocomposite films were prepared by co-sputtering from two independent magnetron sources as well as from a silverpolytetrafluoroethylene (PTFE) composite target. By sputtering from two magnetrons we prepared both, samples with a metal gradient and homogenous composite films. The dielectric constant depends on the metal filling factor and increases approaching the percolation threshold. The resistivity drops from 107 to 10 3 V cm over a narrow range of metal content. The thin composite films show a strong optical absorption in the visible region and a red shift with increasing silver filling. We have produced Bragg reflectors using multilayer systems of composites and pure sputtered PTFE. D 2005 Elsevier B.V. All rights reserved. PACS: 52.80.Pi; 73.61.Ph; 78.66.Sq Keywords: PTFE; Silver nanoparticles; Electrical properties; Optical properties; Bragg reflector
1. Introduction Nanocomposite materials consisting of dispersed metal nanoclusters in a matrix of insulating material show extraordinary physical properties and have been proposed for optical [1,2], electrical [3] and medical applications [4] as well as for data storage [5]. These properties of the composites are very sensitive to small changes in the amount of metal and in the size and shape of the nanoparticles. This sensitivity leads to drastic changes in the electrical and optical properties of the material, which can thus be used as sensors or switching devices. Polymers could be a good choice as a matrix for the metal nanoparticles in order to stabilize the particle size and the growth along with easy processability and low-cost fabrication. Different techniques are available for producing such composite films [6]. Several techniques are based on physical vapor deposition (PVD) including the combination of evaporation and sputtering of metals with plasma polymerization [7] as well as magnetron sputtering from a composite target [8] and coevaporation [9,10].
* Corresponding author. Tel.: +49 431 880 6226; fax: +49 431 880 6229. E-mail address: [email protected] (F. Faupel). 0040-6090/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2005.12.249
In the present work we used a PVD technique based on cosputtering from two magnetron sources as well as from a multicomponent target to analyze optical and electrical properties with focus on the properties near the percolation threshold. High deposition rates can be realized with this sputter technique. Another advantage of magnetron sputtering is the good uniformity of the composite films. The advantage compared to chemical deposition techniques is the absence of residual solvents in the prepared thin films. A RF magnetron source was used for sputtering the polytetrafluoroethylene (PTFE) to prevent charging of the polymer target. Our work was focused on the dependence of the optical and electrical properties specially near the percolation threshold on metal content and preparation conditions. With alternating layers of pure sputtered polymer and composite we were able to prepare Bragg reflectors due to the varying dielectric constant of the two materials. 2. Experimental Composites were prepared with different configurations and arrangements of the sputter sources. Both sputtering from a composite target and from two magnetron sources were performed.
802
U. Schu¨rmann et al. / Thin Solid Films 515 (2006) 801 – 804
All experiments were carried out in a metal vacuum chamber, which was initially evacuated to a pressure below 10 7 Torr. Quartz-crystal monitors were installed in the chamber to control the deposition rates. Polymer sputtering was performed with a RF magnetron system. A Dressler CESAR 136 RF-Generator (13.560 MHz) was coupled to the planar magnetron sputter source ION’X-2WUHV_9109 (Thin Film Consulting). Normally, a RF power of 50 W was applied to the PTFE target to achieve deposition rates around 10 nm min 1. For DC sputtering of pure silver an Advanced Energy MDX 500 sputter source was coupled to the planar magnetron sputter source ION’X-2WUHV_9011 (Thin Film Consulting). The DC power ranged between 4 and 12 W. The targets had a diameter of 2 in. and they were mounted on a cooled holder. The distance between the target and the substrate was 80 mm. Silver/PTFE-multicomponent targets were tested applying different set-ups. We changed the geometry of silver foils or wires on the top of a PTFE sputter target. Thin silver foils clamped onto the surface have the advantage of an easy filling factor control in the deposited film by changing the silver area on the sputter target. The so-called filing factor is defined as the metal volume fraction in the composite material. [6] We can reach high filling factors up to values of over 50% and vary the silver amount easily. However, some problems occurred during sputtering. The homogeneity of the films was not sufficient due to the large areas on the target covered by silver foil. Because of the small thickness of the used silver foils these were partly used up after a few experiments, so the reproducibility decreased. While this problem could be eliminated by using thicker silver foils, additionally, some sputtered silver covered part of the polymer target surface after a few experiments. This resputtering effect increased the filling factor of the deposited composite films with time. Another set-up using silver pins of 1 mm diameter plugged into the PTFE target minimized this problem. The pins were arranged in rings surrounding the center of the target. The ring geometry was chosen because the sputtered area on the target had also a ring-like shape due to the magnetron of the sputter sources. Depending on the number of silver pins and their position one can prepare composite films with a wide range of filling factors. The filling factor and the deposition rate could also be adjusted by changing the deposition parameters such as the input power and the sputter gas (argon) pressure. By increasing the input power the deposition rate of the polymer increased stronger than the metal rate and the filling factor decreased whereas an increase of the pressure had the opposite effect. Such a sputtering setup with a multicomponent target is useful in the case of preparing nanocomposites with more then one metal component, e.g. alloys out of gold, silver, or copper for optical applications. One can easily control the ratio of different metals inside the polymer matrix just by changing the ratio of the metal pins on the target. Better reproducibility and control of the filling factor in the case of one metal component can be achieved by using independent magnetron sputter sources aiming at the substrate from opposite directions with an angle of 50- to the substrate plane. This setup is described in detail in our previous article [11]. Here we could tailor the filling factor by adjusting the
deposition ratio of the metal and the polymer from the two sputter sources. With a rotatable sample holder we could achieve homogenous samples. Without rotation we got samples with a thickness gradient and a filling factor gradient due to the arrangement of the sources. We used energy dispersive X-ray spectroscopy (EDX) to determine the filling factor. Uniform composite samples with different filling factors which were characterized gravimetrically were used as standards for the EDX measurements. The exact procedure was described recently [11]. Layers of composite films with a maximum thickness of 100 nm were deposited on glass slides or transparent polymer foils for UV –Vis measurements. The optical properties were studied using an UV/Vis/NIR-spectrometer Lambda900 (Perkin Elmer). We used a Keithley Model 6485 picoammeter to carry out the resistance measurements. Capacitor measurements to determine the dielectric constant were performed with the automatic RCL meter PM 6306 (Fluke). Transmission electron microscopy (TEM) measurements were carried out using a Philips CM 30 microscope. Films with a maximum thickness of about 50 nm were deposited onto carbon-covered copper grids. Slices of the multilayer films for the cross-sectional TEM images were cut using a Reichert ultracut S ultramicrotome. 3. Results and discussion The morphology of a polymer/metal composite with a filling factor below the percolation threshold prepared by cosputtering from two magnetron sources is shown in the TEM image in Fig. 1. The clusters have a size between 5 and 10 nm. They are spherical and are separated from each other. The electrical properties changed with the filling factor from a polymer-like to a metal-like material. Thereby, the transition is sharp and occurs in a relatively narrow range near the percolation threshold, where a path of percolated particles is formed through the material for the first time. The resistivity changes over several orders of magnitude for a change in the amount of metal of only a few percent. We prepared samples with a gradient in the filling factor and in the thickness. The thickness thereby varied from around 200 nm at the silver side to around 400 nm at the polymer side. On these samples we observed a resistivity drop from values around 5 106 to
Fig. 1. TEM micrograph of a silver-PTFE nanocomposite ( f å 0.2) prepared by co-sputtering from two magnetron sources.
U. Schu¨rmann et al. / Thin Solid Films 515 (2006) 801 – 804
803
Fig. 2. Change of the resistivity (solid squares) and plasmon resonance wavelength (open circles) with the filling factor.
5 10 3 V cm. The percolation threshold was located at a filling factor of f å 0.35. This is in agreement with other works. For example, a theoretical approach for spherical clusters gives the percolation threshold with f å 0.29 [12]. In experiments with gold nanoclusters in different polymer matrices the percolation threshold was found at f å 0.43 (Teflon AF) and f å 0.20 (Poly(a-methylstyrene)) [13]. The dielectric constant ( of the pure sputtered PTFE and of the composite were determined using a capacitor set-up with a completely metalized (Cu/Al) polymer substrate as bottom electrode. The top electrode consisted of silver sputtered through a mask with small holes (diameter 1 mm). ( changed from 2.1 T 0.1 for the pure polymer to more than 5.0 near the percolation threshold. The optical properties were also studied as a function of the filling factor. Even films with a thickness of a few nanometers had an intensive yellow color (Fig. 2). This color effect is based
upon the so-called surface plasmon resonance or particle plasmon resonance (PPR). When a metal-dielectric nanocomposite is excited by light, photons are coupled at the metaldielectric interface, causing an induced charge density oscillation that creates a strong absorption maximum at a particular wavelength. For silver and other noble metal particles in polymers the resonance frequencies are in the visible range. The position of the absorption maximum depends on the cluster morphology like size, shape, and particle distance as well as on the matrix material. The absorption maxima for small filling factors should be found theoretically between 350 and 420 nm depending on the surrounding dielectric matrix [14]. We observed a red shift of the absorption maximum from 405 nm for small filling factors to more than 500 nm for filling factors above the percolation threshold. There was only a slight shift of the particle plasmon resonance frequency observed for
Fig. 3. Optical properties: UV – Vis reflectance spectrum of Bragg multilayer system with five polymer/composite double layers; inset: UV – Vis example spectra for two different filling factors with absorption peaks due to the plasmon resonance.
804
U. Schu¨rmann et al. / Thin Solid Films 515 (2006) 801 – 804
swelling of the matrix. Therefore such kind of Bragg reflectors may be used as optical sensors [2]. 4. Conclusion
Fig. 4. Cross-sectional TEM micrograph of a multilayer system with four double layers. (a) composite layer with silver cluster, (b) sputtered PTFE, (c) polymer substrate.
small filling factors. Near the percolation a stronger dependence of the resonance frequency on the filling factor was observed. With increasing metal concentration inside the composite film near the percolation threshold the gaps between single clusters become smaller and the cluster size and shape can also change due to coalescence of the particles. The resulting significant red shift of the absorption maximum is in good agreement with theoretical models [14]. The changes in the resistivity and the shift of the plasmon resonance with the filling factor is shown in Fig. 2. Representative spectra for two different filling factors are given in the inset of Fig. 3. We prepared multilayer structures composed of alternating layers of pure sputtered PTFE and an Ag-PTFE-composite with low and high refractive indices, which leads to a Bragg reflector set-up. Fig. 4 shows a cross-sectional TEM image of such a multilayer set-up with four double layers. The selective reflectivity results from the constructive interference of reflections from the different layer interfaces. The refractive index n of the sputtered polymer was determined with ellipsometry as n = 1.374 [11]. The value of the refractive index of the composite is larger than that of the pure polymer and it changes with the silver fraction. Because the position of the maximum of reflectance depends on the layer thickness and on the refractive indices of the alternating layer, one can tailor the wavelength of reflectance by changing these parameters. So far we have no exact values for the refractive index of the composites with changing filling factor. Therefore, the determination of the reflection maximum is only an estimate. However, a qualitative change in the parameter’s thickness and filling factor gave the predicted shift. By increasing the number of the layers of such a Bragg reflector the reflectance can be enhanced to even higher values. Also, the peak becomes sharper by improving the Bragg conditions and increasing the number of the layers. A typical spectrum of a very simple Bragg reflector containing five double layers is given in Fig. 3 to demonstrate the feasibility. And even with only five double layers we reached a reflectance of more than 80%. The maximum was located in the near infrared region, but by changing the parameters it was also possible to shift it to the visible region. It is known that the refractive index of such composites is sensitive to the vapor of organic solvents due to
We have shown that it is possible to synthesize nanocomposites consisting of a polymer and silver by sputtering from a composite target as well as from two independent magnetron sources. Different geometries were checked in case of the composite target in order to optimize deposition rate, filling factor, metal distribution, and reproducibility. An arrangement of the metal as rings of pins showed the best results. In any case, there is the problem of silver resputtering, which changes the target slightly after every experiment. In order to eliminate this problem, we developed a more reliable process based on co-sputtering from two targets. We obtained relatively high deposition rates and a good reproducibility and uniformity of the deposited films. The dielectric constant increases with increasing filling factor up to the percolation threshold, where the electrical and optical properties change drastically from insulating to metallic. The resistivity and the optical absorption can be tailored via the filling factor. Using the different refractive indices of the pure polymer and the composite, one can thus easily produce Bragg reflectors with tailored properties. Acknowledgements The authors are grateful to S. Rehders for his technical support in developing the vacuum chambers and to the German Science Foundation (DFG) for financial support under grant number FA 234/10-1. References [1] A. Biswas, O.C. Aktas, U. Schu¨rmann, U. Saeed, V. Zaporojtchenko, F. Faupel, Appl. Phys. Lett. 84 (14) (2004) 2655. [2] A. Convertino, A. Capobianchi, A. Valentini, E.N.M. Cirillo, Adv. Mater. 15 (2003) 1103. [3] P.M. Ajayan, L.S. Schadler, P.V. Braun, Nanocomposite Science and Technology, Wiley-VCH, Weinheim, 2003. [4] F.-R.F. Fan, A.J. Bard, J. Phys. Chem., B 106 (2002) 279. [5] J. Ouyang, C.-W. Chu, R. Szmanda, L. Ma, Y. Yang, Nat. Mater. 3 (2004) 918. [6] A. Heilmann, Polymer Films with Embedded Metal Nanoparticles, Springer, Berlin, 2003. [7] H. Biederman, D. Slavinska, Surf. Coat. Technol., A 125 (2000) 371. [8] R.A. Roy, R. Messier, S.V. Krishnaswamy, Thin Solid Films 109 (1983) 27. [9] K. Behnke, T. Strunskus, V. Zaporojtchenko, F. Faupel, Proc. Microbiol. Mater. (2000) 1052. [10] A. Biswas, Z. Ma´rton, J. Kruse, J. Kanzow, V. Zaporojtchenko, F. Faupel, T. Strunskus, Nano Lett. 3 (1) (2003) 69. [11] U. Schu¨rmann, W. Hartung, H. Takele, V. Zaporojtchenko, F. Faupel, Nanotechnology 16 (2005) 1078. [12] E.J. Garboczi, K.A. Snyder, J.F. Douglas, M.F. Thorpe, Phys. Rev., E 52 (1) (1995) 819. [13] H. Takele, U. Schu¨rmann, V. Zaporojtchenko, F. Faupel, accepted for publication in Europ. Journ. Phys. [14] U. Kreibig, M. Vollmer, Optical Properties of Metal Clusters, Springer, Berlin, 1995.