Usually, polymers are not deposited by physical vapor deposition. Certain polymers, however, lend themselves to evaporation or sputtering. In our evaporation processes, the nanoparticle containing polymer matrix is produced either by evaporation of the monomers and polycondensation on the substrate or by thermal cracking of suitable polymers like Teflon AF, and repolymerization of the fragments on the substrate. Vapor phase co-deposition of noble metals and various polymers (Nylon 6, TAF, and PMMA) was successfully used to produce polymer/metal nanocomposites with a wide range of the metal volume filling factor (see below). For example, Fig. 1 shows the morphology variations of Nylon/Ag composite films at different metal filling factors. As the Ag concentration increases from 4 to 50 %, the average size of Ag nanoclusters in the composite also increases from 2 to 20 nm.

Fig. 1  TEM micrograph of Nylon/Ag nanocomposites of about 60 nm thickness at metal filling factors of (a) 4.4%, (b) 14 %, (c) 21%, and (d) 41 5%.

The co-evaporation technique can also be used to prepare nanocomposite films with alloy nanoparticles of well-defined composition. This was demonstrated, for instance, by producing nanocomposites containing bimetallic Au_xAg_1-x and Cu_xAg_1-x nanoparticles in a Teflon AF matrix to tune the position of the surface plasmon resonance. For alloy nanoparticles, the bulk phase diagram is not always valid since surface and interfacial energies contribute on the nanoscale.

Disadvantages of the co-evaporation procedure are not only the low condensation probability of metal atoms but also a reduction of the molecular weight of the polymers after deposition. The latter effect gives rise to poor mechanical properties. For functional applications, however, this is often not considered to be a significant problem, otherwise we use alternative methods (see preparation web page).

Co-evaporation of the metal and polymer components can also be used to produce non-spherical metallic nanoparticles. An example of magnetic nano-rods embedded in a protecting Teflon AF matric is show below.

Fig. 2  (left) TEM image of ternary magnetic Fe-Ni-Co nano-rods in a Teflon AF matrix deposited by co-evaporation. (right) Kinetic Monte Carlo showing the metallic filling factor vs the deposition rate ration of the components.

We have also applied kinetic Monte Carlo simulations to clarify and control the formation of the nanostructure. Based on early studies within the group, we now cooperate with Prof. Bonitz form theoretical physics in Kiel.

Selected publications

Takele, H.; Jebril, S.; Strunskus, T.; Zaporojchenko, V.; Adelung, R.; Faupel, F.: Tuning of electrical and structural properties of metal-polymer nanocomposite films prepared by co-evaporation technique, Appl. Phys. A 92(2) (2008) 345-350.

Biswas, A.; Márton, Z.; Kruse, J.; Kanzow, J.; Zaporojtchenko, V.; Faupel, F.; Strunskus, T: Controlled generation of Ni nanoparticles in the capping layers of Teflon AF by vapor phase tandem evaporation, Nano Letters 3(1) (2003) 69-73.

Takele, H.; Schürmann, U.; Greve, H.; Paretkar, D.; Zaporojtchenko, V.; Faupel, F.: Controlled growth of Au nanoparticles in co-evaporated metal/polymer composite films and their optical and electrical properties, European Physical Journal:Applied Physics 33(2) (2006) 83-89.

Zaporojtchenko, V.; Strunskus, T.; Behnke, K.; Bechtolsheim, C. v.; Thran, A.; Faupel, F.: Formation of metal-polymer interfaces by metal evaporation: influence of deposition parameters and defects, Microelectronic Engineering 50(1-4)(2000) 465-471.

Monte Carlo simulations

Rosenthal, L.; Strunskus, T.; Faupel, F.; Abraham, J.W.; Bonitz, M.: Kinetic Monte Carlo Simulations of Cluster Growth and Diffusion in Metal-Polymer Nanocomposites, Complex Plasmas, Springer Series on Atomic, Optical, and Plasma Physics82 (2014) 321-370.

Rosenthal, L.; Greve, H.; Zaporojtchenko, V.; Strunskus, T.; Faupel, F.; Bonitz, M.: Formation of magnetic nanocolums during vapor phase deposition of a metal-polymer nanocomposite: experiments and kinetic Monte Carlo simulations,Journal of Applied Physics 114 (2013) 044305.

Faupel, F.; Willecke, R.; Thran, A.: Diffusion of metals in polymers, Materials Science & Engineering R-Reports 22(1)(1998) 20090,1-55.

Thran, A.; Faupel, F.: Computer simulations of metal diffusion in polymers, Defect and Diffusion Forum 903 (1997) 143-147.

b) Sputtering

Alternatively, we produce our nanocomposites by co- and tandem sputtering, involving rf magnetron sputtering of selected polymers such as Teflon from a polymer target and dc sputtering from a metal or alloy target. For magnetic metals, rf sputtering is also used. The resulting polymer matrix is crosslinked, which is often advantageous and may, in particular, improve the mechanical properties compared to the original polymer. In order to find suitable polymers for sputter deposition, we investigated in detail the interaction of low energy ions with polymers. The crucial point is the competition between chain scission and crosslinking. Polymers with a high sputter rate should exhibit a low crosslinking tendency. Teflon (PTFE) turned out to exhibit a sputter yield which is about two orders of magnitude above that of most other polymers. This polymer, as also reported by Biedermann [Vacuum 59, 594 (2000)], is particularly suitable for sputtering and preparation of metal-containing nanocomposites. Besides polymers also inorganic materials can serve as a dielectric matrix. Therefore also rf sputter processes for inorganic materials like SiO2, TiO2 and AlN are well established. Composites with AlN as piezoelectric matrix are investigated in the SFB855.

c) Plasma Polymerization

Plasma polymerization is another method to obtain polymeric matrices. Here, we start from hexamethyldisiloxane (HMDSO) as a precursor to produce the polymeric matrix. Addition of oxygen to the plasma process allows to vary the properties of the matrix from polymer like (without oxygen, leading to a high carbon content) to SiOx like (using a 40% oxygen/argon mixture). The combination of the plasma polymerization process with the nanoparticle formation by a cluster source, also allows the preparation of nanocomposites.

d) Gas aggregation cluster source

Sputtering at a high gas pressure leads to formation of nanoparticles in the gas phase. In our group we have developed and built a gas aggregation source based on magnetron sputtering from 2-inch targets. Different to other or commercially available clusters sources our source works without external cooling and allows the deposition of nanoparticles with a high rate and narrow size distribution. In a cluster source, the nanoparticles form in the gas phase and not on the surface of the growing composites. This, among other advantages, allows chemical reactions between the metallic component and the matrix to be widely eliminated. Moreover, the source also allows to deposit highly porous films made up of aggregated nanoparticles, which are presently investigated. One of the sources is shown in Fig. 2.

Fig. 2 One of our cluster sources

• B. Gojdka, V. Hrkac, T. Strunskus, L. Kienle, F.Faupel, Nanotechnology 22, 465704 (2011).

• Plasmonic and photoswitchable nanocomposites
• Magnetic high frequency composites with cut-off frequencies > 1 GHz
• Antibacterial coatings with controlled release
• Sensors for volatile organic compounds
• Magnetic field sensors based on magnetoelectric layered composites

Optical nanocomposites

If the nanoparticles in a dielectric matrix are small compared to the wavelength of light, scattering is avoided, and the composites are transparent even at high metal filling factors. Thus the index of refraction can be tuned over a wide range. Furthermore, nanoparticles of metals exhibiting free electron behavior such as noble and alkali metals show a strong absorption maximum due to collective resonant oscillations of the conduction electrons, i.e. a particle plasmon resonance, which occurs generally in the visible region or the near infrared region of the optical spectrum. This gives rise to characteristic colors.

The optical properties of such nanocomposites are currently of considerable interest, because other applications besides making colored glass are emerging - such as sensing, e.g., by single-molecule detection, enhancement of nonlinear processes, synthesis of materials with otherwise unavailable optical properties, e.g., materials with µr >>1 at optical wavelengths, and many more. Interest in plasmonic properties of nanocomposites for possible uses in integrated optical components and circuits is also steadily growing.

The plasmon resonance frequency depends on many factors including size, shape, and filling factor of the metallic nanoparticles, as well as dielectric properties of the matrix. The plasmon resonance frequency can also be tuned by choosing alloy nanoparticles and adjusting the alloy composition. This is shown in Fig. 2 for Au-Ag alloy particles in a Teflon AF matrix prepared by vapor phase co-deposition of the constituents from three independent sources simultaneously.

Fig. 3 Tuning of the plasmon resonance in Ag-Au particles

As mentioned above, the index of refraction of the present nanocomposites can be tuned via the filling factor and type of filler. This can be made instrumental to build Bragg reflectors consisting of alternating layers of pure polymer and metal/polymer nanocomposite. An example based on Ag/polymer composites prepared by Teflon sputtering and co-sputtering of Ag, respectively is given in. More details on optical nanocomposites prepared by our present techniques are described in our following papers (see also publication list for additional papers):

• A. Biswas, O. C. Aktas, U. Schürmann, U. Saeed, V. Zaporojtchenko, T. Strunskus and F. Faupel, Appl. Phys. Lett., 84, 2655 (2004).
• U. Schürmann, W.A. Hartung, H. Takele, V. Zaporojtchenko, and F. Faupel, Nanotechnology, 16, 1078 (2005).
• U. Schürmann, H. Takele, V. Zaporojtchenko, and F. Faupel, Thin Solid Films, 515, 2, 801, 2 (2006).
• H. Takele, U. Schürmann, H. Greve, D. Paretkar, V. Zaporojtchenko, and F. Faupel, Eur. Phys. J. Appl. Phys. (EPJAP), 33, 83 (2006).
• Vladimir Kochergin, Vladimir Zaporojtchenko, Haile Takele, Franz Faupel and Helmut Föll, J. Appl. Phys., 101, 024302 (2007).

In cooperation with the group of Prof. Mady Elbahri we have developed two new applications of optical nanocomposites. Both are based on a coupling of plasmons of the nanoparticles with the plasmons of a metallic film. Using suitable filling factors, metal film thickness and very important proper matching conditions this can be used to create highly transparent conductors (showing a superior conductivity to ITO at comparable transparancy) or as another extreme a perfect absorber. Both systems have been patented and further details can be found on the homepage of Prof Elbahri and in the following publications:

• An Omnidirectional Transparent Conducting-Metal-Based Plasmonic Nanocomposite, Mady Elbahri, Mehdi Keshavarz Hedayati, Venkata Sai Kiran Chakravadhanula, Mohammad Jamali, Thomas Strunkus, Vladimir Zaporojtchenko and F. Faupel, Avanced Materials, Volume: 23, Issue: 17, Pages: 1993-1997.
• Design of a perfect black absorber at visible frequencies using plasmonic metamaterials, Mehdi Keshavarz Hedayati, Mojtaba Javaherirahim, Babak Mozooni, Ramzy Abdelaziz, Ali Tavassolizadeh, Venkata Sai Kiran Chakravadhanula, Vladimir Zaporojtchenko, Thomas Strunkus, Franz Faupel, Mady Elbahri, Advanced Materials 2011 (In press).

Nanocomposites also play a key role in one of our projects within the Collaborative Research Center SFB 677 "Function by Switching". Here, the nanocomposites are combined with photoswitchable molecules. These so-called chromophores change their properties reversibly upon irradiation with light of two different wavelengths. Very interesting new electro-optical properties arise through interactions between chromophores and the surface plasmon resonance of the metallic nanoparticles. Our recent major progress in this field is reported in the listed publications.

• Reversible light-controlled conductance switching of azobenzene based metal/polymer nanocomposites, Christina Pakula, Vladimir Zaporojtchenko, Thomas Strunskus, Dordaneh Zargarani, Rainer Herges and Franz Faupel, Nanotechnology, Volume 21, Issue 46, pp. Art.Nr.:465201 (2010).
• Optical switching behavior of azobenzene/PMMA blends with high chromophore concentration, Christina Pakula, Christian Hanisch, Vladimir Zaporojtchenko, Thomas Strunskus, Claudia Bornholdt, Dordaneh Zargarani, Rainer Herges and Franz Faupel, Journal of Materials Science: Volume 46, Issue 8 (2011), Page 2488.
• Reversible light-induced capacitance switching of azobenzene ether/PMMA blends, Vladimir Zaporojtchenko, Christina Pakula, Sri Wahyuni Basuki, Thomas Strunskus, Dordaneh Zargarani, Rainer Herges and Franz Faupel Applied Physics A: Volume 102, Issue 2 (2011), Page 421.
• Free volume changes on optical switching in azobenzene-PMMA blends studied by a pulsed low-energy positron beam, S. Harms, K. Rätzke, C. Pakula, V. Zaporojtchenko, T.Strunskus, W. Egger, P. Sperr, F. Faupel, Journal Polymer Science, Part B: Polymer Physics Volume 49, Issue 6, (2011) 404.

Magnetic nanocomposites

(a) Magnetic core materials for high frequencies up to the GHz range

The ongoing demands for miniaturization in mobile electronic devices as well as steadily increasing frequencies of electronic circuits into the GHz range have stimulated the research on high frequency magnetic components. Our approach is very attractive for the development of high-frequency (up to > 1 GHz) soft-magnetic materials which would be readily integrable in monolithic microwave integrated circuit (MMIC) devices. Eddy currents, one of the main loss mechanisms determining the cut-off frequency of inductor materials, can be effectively suppressed by embedding magnetic nanoparticles in a dielectric matrix. The ferromagnetic resonance frequency, which is the other main parameter determining the high frequency behavior of an inductor, can be maximized by the possibility to tune the magnetic properties of the material by using alloy particles with well controlled composition, size and shape. Fig.3 shows very promising permeability data of a composite consisting of ternary magnetic clusters in a hydrophobic protecting fluoropolymer matrix. The resonance frequency is already around 5 GHz and the quality factor is ~12 at 1 GHz without any tuning of the deposition conditions and can further be improved. Meanwhile, quality factors > 50 were achieved at 1 GHz (publication in preparation)

Fig. 4 Real (µ’) and imaginary (µ”) part of the permeability as well as the quality factor Q of a sputtered Fe-Ni-Co/Teflon composite.

One field of application for this type of magnetic materials is as core material of integrated micro­inductor devices. Current devices for commercial applications are realized as planar spiral inductors without a magnetic core. Their drawbacks are their large lateral size and thus higher costs as well as the spatial propagation of stray fields leading to eddy currents in the substrate and possibly to coupling effects between neighboring devices. A toroidal inductor (Fig. 4) is advantageous because of the minimi­zation of stray fields due to a closed magnetic core ring. In a first series of experiments the aforementioned magnetic core material has been successfully integrated into such toroidal microinductors by the group of Prof. E. Quandt.

Fig. 5 (Left) Schematic of a toroidal microinductor; (Right) Cross-section through a real device

(c) Formation of nanorods

We recently have found that by co-deposition of Fe-Ni-Co alloys together with a fluoropolymer like Teflon AF at elevated substrate temperatures, extremely thin rods with a very large aspect ratio can be obtained above a critical ratio of the metal/organic deposition rates. The growth is attributed to a self-organization process which is based on the very low interaction energy between metal and polymer as well as a critical threshold in the flux ratio of the two constituents. Fig. 5 shows a cross-sectional transmission electron microscopy image of such a nanorod film. This route to nanorods might have applications in high density data storage and other fields.

Fig. 6 (a) TEM image of Fe-Ni-Co nanorods on top of a layer of Ag clusters in an evaporated Teflon AF matrix. (b) Metal filling factor vs. metal/organic deposition rates

More details on magnetic nanocomposites prepared by our present techniques are described in our following papers (see also publication list for additional papers):

• H. Greve, C. Pochstein, H. Takele, V. Zaporojtchenko A. Gerber, M. Frommberger, E. Quandt, and F. Faupel, Appl. Phys. Lett. 89, 242501 (2006).
• Henry Greve, Abhijit Biswas, Ulrich Schürmann, Vladimir Zaporojtchenko, and Franz Faupel, Appl. Phys. Lett., 88, 123103 (2006).
• A. Biswas, Z. Marton, J. Kanzow, J. Kruse, V. Zaporojtchenko, F. Faupel, and T. Strunskus, Nano Letters, 3, 1 (2003).

Antibacterial coatings

Currently, we are exploring co-sputtering of Ag and other noble metals and alloys together with an organic component as a method to produce antibacterial coatings, where the precious metals are only incorporated into a thin surface layer. Moreover, the active metals are finely dispersed as nanoparticles, thus saving additional material and providing a very large effective surface for metal ion release. Fig. 6 gives an example of a first nanocomposite showing strong antibacterial activity. The antimicrobial effect of the nanocomposite coatings was evaluated by means of two different assays in cooperation with Prof. R. Podschun from the University Hospital Schleswig-Holstien,Campus Kiel. Bactericidal activity due to silver release from the surface was determined by a modification of conventional disc diffusion methods. Inhibition of bacterial growth on the coated surface was investigated by a modified proliferation assay. Staphylococcus aureus and S. epidermidis were used as test bacteria as these species commonly cause infections associated with medical polymer devices. The antibacterial efficiency of the coatings against different bacteria was demonstrated at extremely small noble metal consumption: Au ~ 1 mg/m2 and Ag ~ 0.1 g/m2. The maximum antibacterial effect was seen for Au-Ag/PTFE nanocomposite, followed by Ag/PTFE nanocomposite. In the case of Ag-Au/PTFE nanocomoposite, a huge acceleration of the silver ion release of one order of magnitude was observed when a trace amount of Au (less than 1mg/m2 of coating) was deposited additionally on the Ag/PTFE composite surface. This can be explained by the formation of galvanically coupled silver and gold nanoparticles. The gold nanoparticles lead to enhanced Ag ion formation because in the galvanic pair silver is more active than gold

Also, it was found that the release of Ag ions correlates with the antibacterial efficiency of Ag composite coatings and can be controlled by the coating thickness and volume fraction of Ag and composition of nanoparticles.

Fig. 7 (left) Zone of growth inhibition of Staphylococcus aureus due to ion release from an Ag-Au/ PTFE coating, placed top down onto an agar plate inoculated with S. aureus, after overnight incubation. (right) Growth inhibition on Ag/ PTFE coat determined by a modified proliferation assay of Bechert et al. Growth data are given for bacterial cells (colony forming units, CFU ml-1) shedded into the surrounding growth medium.

Further details are given in the following publications, where the second paper deals with TiO2 based nanocomposite.

• V. Zaporojtchenko, R. Podschun, U. Schürmann, A. Kulkarni, and F. Faupel, Nanotechnology 17, 4904 (2006)
• V.S.K. Chakravadhanula, T. Hrkac, V. Zaporojtchenko, R. Podschun, V.G. Kotnur, A. Kulkarni, T. Strunskus, L. Kienle and F. Faupel, Journal of Nanoscience and Nanotechnology, Vol. 11, 4893–4899, (2011)

Sensors for organic vapors

Nanocomposites containing metallic filler in a dielectric matrix also exhibit very interesting electronic properties. For instance, the conductivity varies from insulating to metallic as function of the metal concentration and drops over many orders of magnitude near the percolation threshold. Below, but near the percolation threshold, charge transport is governed by thermally activated hopping from nanoparticle to nanoparticle. In this regime near the metal-insulator transition, the conductivity depends exponentially on the cluster separation. This can be used for sensing applications as illustrated for the case of quasi two-dimensional sensors for organic vapors in Fig. 7.

The principle is based on the swelling properties of polymer films by the incorporation of organic vapors which increases the separation of the metal nanoparticles. The sensors were prepared by taking advantage of the high process control in the deposition of metal clusters via thermal evaporation on top of a thin polymer film. These sensors show a reversible signal, which is sensitive to the kind of organic vapor present in the surrounding as well as to the amount (vapor pressure) of the vapors. Figure 7 shows an example for 5 different organic vapors. One notes that the response to the vapors is different for every vapor. This is based on the fact that different organic vapors have different solubilities in a given polymer. Hence, by choosing the suitable polymer matrix, one can tune the sensor to be the most sensitive for a specific vapor. In addition, by combining several different polymer matrices in a sensor array, one can build a highly selective sensing device taking the response ratios of the different sensors as a fingerprint for a particular vapor. A corresponding publication is in preparation.

Fig. 8 Response of a Au-Nylon nanocomposite sensor to different organic vapors. The change in resistance is related to the different solubility parameters of the vapors. The response is different for other polymer substrates for the same vapors at the same pressure.

• C. Hanisch, A. Kulkarni, V. Zaporojtchenko and F. Faupel, Journal of Physics: Conference Series 100 Volume 100 (2008) 052043.
• C. Hanisch, N. Ni, A. Kulkarni, V. Zaporojtchenko, T. Strunskus and F. Faupel, Journal of Materials Science (2010),Volume 46(2), pages 438-445.

For further information please contact Dr. Thomas Strunskus.