Early stages of growth of gold layers sputter deposited on glass and silicon substrates
© Malinsky et al.; licensee Springer. 2012
Received: 10 January 2012
Accepted: 6 May 2012
Published: 6 May 2012
Extremely thin gold layers were sputter deposited on glass and silicon substrates, and their thickness and morphology were studied by Rutherford backscattering (RBS) and atomic force microscopy (AFM) methods. The deposited layers change from discontinuous to continuous ones for longer deposition times. While the deposition rate on the silicon substrate is constant, nearly independent on the layer thickness, the rate on the glass substrate increases with increasing layer thickness. The observed dependence can be explained by a simple kinetic model, taking into account different sticking probabilities of gold atoms on a bare glass substrate and regions with gold coverage. Detailed analysis of the shape of the RBS gold signal shows that in the initial stages of the deposition, the gold layers on the glass substrate consist of gold islands with significantly different thicknesses. These findings were confirmed by AFM measurements, too. Gold coverage of the silicon substrate is rather homogeneous, consisting of tiny gold grains, but a pronounced worm-like structure is formed for the layer thickness at electrical continuity threshold. On the glass substrate, the gold clusters of different sizes are clearly observed. For later deposition stages, a clear tendency of the gold atoms to aggregate into larger clusters of approximately the same size is observed. At later deposition stages, gold clusters of up to 100 nm in diameter are formed.
Keywordssputtering gold layer glass silicon RBS
Coating of substrates with thin metal films is of great importance for contemporary technologies (optical coatings, corrosion protection, semiconductor devices). Different methods, such as sputtering, pulsed laser deposition, vacuum evaporation, vapor phase techniques, or molecular beam epitaxy, are used for thin film creation. Growing rate and resulting morphology of the thin metal films depends on the deposition technique used and several other factors including the properties of substrate, adsorbed atoms, and their interaction strength with the substrate surface. Initial phase of the film growth and its morphology is a persisting problem in the physics of thin films and in a wide range of technological applications, too. Three basic mechanisms of the film growth have been described and observed .
Experimental examination of the film morphology in the initial phases of growth is not easy despite the wide spectrum of diagnostic techniques available for this purpose, including AES, LEED, RHEED, XPS, atomic force microscopy (AFM), SEM, Rutherford backscattering (RBS), and TEM. The properties of the metal films deposited by various techniques on different substrates have been the subject of many studies: experimental and theoretical (see e.g. [2–6]). Gold is often used in these studies because of its inert character and the interesting physicochemical properties of gold nanoparticles.
In this work, standard RBS and AFM methods are used for the study of growth and morphology of very thin gold layers prepared by sputtering on two different substrates: ‘common’ glass and monocrystalline Si. The study is a continuation of our previous works in the field.
Substrate and Au deposition
The gold films were deposited onto two substrates: standard Si (100) n-type wafer (5-cm diameter) and glass (25 × 25 mm2, 0.1-mm thick, Marienfeld, Lauda-Königshofen, Germany). The native SiO2 was present on the Si substrate. The substrates were cleaned with methanol; the drying of samples was performed with nitrogen flow. Deposition was performed by diode sputtering from a 99.99 % Au target (supplied by Goodfellow Ltd, Friedberg, Germany) using a Bal-Tec SCD 050 device (BalTec Maschinenbau AG, Pfäffikon, Switzerland). The sputtering conditions were a discharge power of 7.5 W, a total pressure of about 5 Pa of argon (99.995 % purity), and the electrodes of 48 cm2 in area at a distance of 50 mm. The deposition was performed at room temperature, and the deposition times from 5 to 150 s were chosen to create the layers from discontinuous to continuous ones. The deposition onto both substrates was performed simultaneously.
RBS analyses were performed on a Tandetron 4130MC accelerator in the Nuclear Physics Institute in Rez using 1.75-MeV 4He ions. The measurements were performed in IBM geometry with an incident angle of 0° and a laboratory scattering angle of 170°. Scattered particles were registered with a surface barrier detector connected to a standard spectrometric chain and acquisition system. The typical energy resolution of the spectrometer was FWHM = 15 keV. The ion flux was measured using a monitor with a rotating target situated in front of the sample. Sufficiently long measuring times were chosen to reduce typical statistical errors in gold signal to about 1 %. The RBS spectra were evaluated using a SIMNRA code .
Surface morphology and roughness of the pristine and gold-deposited Si and glass samples were examined by AFM technique using a VEECO CP II device in tapping mode (Bruker Corporation, Santa Barbara, CA, USA). A Si probe RTESPA-CP with a spring constant of 20 to 80 N m−1 was used. The mean roughness value (Ra) represents the arithmetic average of the deviations from the center plane of the sample. AFM technique combined with standard scratch method was also used for the determination of the gold layer thickness .
Results and discussion
Parameters of Equation 2
(1015 cm−2 s−1)
2.00 ± 0.02
59.70 ± 3.0
0.59 ± 0.02
2.16 ± 0.02
28.9 ± 7.5
Widths of gold signal calculated as central second moment of measured distribution (see Figure 4 )
Pristine glass surface exhibits very low surface roughness compared to the pristine Si surface (see Figures 2, 5, and 6). A rather complex surface morphology is observed on the gold-covered glass. The presence of isolated gold grains of various sizes is observed for all deposition times. With increasing deposition time, the mean size and the density of these grains increase, but the initial size differences (observed at deposition times of 5 and 15 s) are gradually smeared out. The later effect is reflected in the evolution of the Ra which increases rapidly in initial deposition stages and achieves saturation for later ones (Figure 6). The gold grains with diameters of up to 100 nm are formed in the later deposition stages. With increasing deposition time, the gold layer becomes electrically conductive. These findings are in agreement with above-mentioned RBS results, namely with the evolution of the width and asymmetry of the RBS gold signal (see also Table 2). The appearance of the gold grains and their growth may be due to the above-mentioned preferential capture of the incoming gold atoms on the already existing gold islands or to surface diffusion of deposited gold atoms and their aggregation into larger grains.
Thickness of gold layers sputtered onto silicon and glass substrates as a function of the deposition time was measured by standard RBS method and AFM combined with scratch technique. The layer thickness is a monotonously increasing function of the deposition time, as could be expected. The deposition rate on the silicon substrate determined by the RBS practically does not depend on the instantaneous layer thickness. For the glass substrate, however, the deposition rate is low for short deposition times and discontinuous gold coverage, and with increasing deposition time and more homogenous coverage, it increases to a value comparable with that found on the silicon substrate. The observed rate dependence can be explained by different sticking probabilities of gold atoms on the bare glass substrate and on regions with gold coverage. The results on the layer thicknesses obtained by AFM method differ from those by RBS significantly, the differences being due to different principles of both methods, namely, in contrast to the RBS results, the AFM method gives higher layer thicknesses on the glass substrate than on the silicon substrate. The AFM images taken on bare substrates and those coated with gold for different deposition times show great differences in the morphology of the gold layers deposited on both substrates. On the silicon substrate, rather homogenous gold layers are observed, the surface morphology of which has only little changes with increasing sputtering time. At the deposition stage roughly corresponding to the outbreak of layer electrical conductivity, a worm-like structure in the gold layer is formed, and at later deposition stages, a globular layer structure is observed. On the glass substrate, however, the presence of isolated gold grains of various sizes is observed for short deposition times. With increasing deposition time, the mean size and the density of these grains increase, but the initial differences in the grain size are gradually smeared out. The AFM results are in good agreement with the results of RBS measurements, especially with detailed analysis of the form of the gold RBS signal, indicating significant heterogeneity of the gold layers deposited for short times on the glass substrate.
Present results, together with our previous ones, contribute to a better understanding of initial stages of the formation of metal layers on different substrates. The results can be of interest for specialists developing metamaterials, photonic crystal structures, and nanostructured materials with enhanced biocompatibility.
The authors are indebted to the Grant Agency of the Czech Republic for the financial support under the project 108/12/G108.
- Venables JA, Spiller GDT, Hanbucken M: Nucleation and growth of thin films. Rep Prog Phys 1984, 47: 399–459. 10.1088/0034-4885/47/4/002View Article
- Muzard S, Templier C, Delafond J, Girard JC, Thiaudiere D, Pranevicius L, Galdikas A: Development of the microstructure of sputter deposited gold on amorphous carbon. Surf Coat Technol 1998, 98: 100–101.
- Resta V, Afonso CN, Piscopiello E, Van Tendeloo G: Role of substrate on nucleation and morphology of gold nanoparticles produced by pulsed laser deposition. Phys Rev B 2009, 79(1–6):235409.View Article
- Ratsch C, Venables JA: Nucleation theory and the early stages of the thin film growth. J Vac Sci Technol, A 2003, 21: S96-S109. 10.1116/1.1600454View Article
- Chaloupka A, Simek P, Sutta P, Svorcik V: Influence of substrate on properties of gold nanolayers. Mater Lett 2010, 64: 1316–1318. 10.1016/j.matlet.2010.03.019View Article
- Švorčík V, Siegel J, Šutta P, Mistrík J, Janíček P, Worsch P, Kolská Z: Annealing of gold nano-structures sputtered on glass substrate. Appl Phys A 2011, 102: 605–610. 10.1007/s00339-010-6167-1View Article
- Mayer M: SIMNRA User’s Guide, Report IPP 9/113. Report of Max-Planck-Institut für Plasmaphysik, Garching, Germany; 1997.
- Švorčík V, Hubáček T, Slepička P, Siegel J, Kolská Z, Bláhová O, Macková A, Hnatowicz V: Characterization of carbon nanolayers flash evaporated on PET and PTFE. Carbon 2009, 47: 1770–1778. 10.1016/j.carbon.2009.03.001View Article
- Stoquert JP, Szorenyi T: Determination of the number and size of inhomogeneities in thin films by ion beam analysis. Phys Rev B 2002, 66(1–8):144108.View Article
- Barradas NP: Can quantum dots be analysed with microbeam RBS? Nucl Instr Meth B 2007, 261: 435–438. 10.1016/j.nimb.2007.04.183View Article
- Švorčík V, Zehentner J, Rybka V, Slepička P, Hnatowicz V: Characterization of thin gold layers on polyethyleneterephthalate: transition from discontinuous to continuous, homogenous layer. Appl Phys A 2002, 75: 541–544. 10.1007/s003390101024
- Slepička P, Švorčík V, Šlouf M, Rybka V, Špirková M: Characterization of metal nanolayers sputtered on poly(ethyleneterephtalate). Optoelectron Adv Mater Rapid Com 2008, 2: 153–160.
- Siegel J, Kvítek O, Slepička P, Náhlík J, Heitz J, Švorčík V: Structural, electrical and optical studies of gold nanostructures formed by Ar plasma-assisted sputtering. Nucl Instrum Meth B 2012, 272: 193–197.View Article
- Švorčík V, Slepička P, Švorčíková J, Špírková M, Zehentner J, Hnatowicz V: Characterization of evaporated and sputtered thin Au layers on poly(ethylene terephtalate). J Appl Polym Sci 2006, 99: 1698–1704. 10.1002/app.22666View Article
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.