Roughness effects on thin film physical properties

The emergence of thin film nanoscience as an identifiable field of investigation parallels the development and adaptation of surface structural and chemical probes to provide in situ real-time atomic-scale data during film growth. The next step in this evolutionary process, which evolving right now, is the addition of theory and modeling as standard approaches, used in combination with the above physical tools, to complement, design, and analyze experiments. The beautifully hierarchical complexity associated with thin film micro-nanostructural and surface morphological evolution, which arises from a diverse set of competitive kinetic instabilities operating on very different time and length scales, assures that thin film nanoscience and technology will remain a vital and exciting field long into the future. Thus, an important part of my research focus is the following:  

• Thin film growth properties: Under different film preparation conditions (substrate temperature, pressure, growth rate, presence of impurities, shadowing effects etc.) and different growth methods (physical and plasma vapour deposition, sputtering, chemical vapour deposition, nanocluster deposition), one may obtain a wide variety of different surface/interface morphologies and film microstructure which are inherently related to dynamic growth mechanisms [1]. These, in turn, have a significant and generally different influence on physical properties. Although epitaxial growth of a film in the layer-by-layer mode can commence in some cases within a certain temperature regime, such a growth mode may not always occur and instead growth front can be rough in the form of mounds (multilayer step structure; unstable growth) or due to noise induced roughening during the growth can lead to the formation of self-affine fractal morphologies [2]. Besides kinetic effects during growth, stress relaxation in between film interfaces strongly alters growth characteristics and has to be taken certainly into consideration [3].
• Magneto-transport (spin-polarized) properties: Systematic studies in spin polarised electron transport in magnetic multilayers which show giant magneto-resistance (GMR) effect, indicated that interface roughness as well as bulk impurities strongly influence the magnitude of the effect in a way that depends on the spin asymmetry for bulk and interfacial electron scattering [4]. Since the magnitude of the GMR effect is linked to the interface topology via the exchange coupling in magnetic multilayers, the interface quality is of crucial importance for magneto-electronic devices. Besides interface roughness, the presence of magnetic pin holes, that cause ferromagnetic alignment of parts of the sample, as well as precursors of these in the form of large spacer film fluctuations (which lead also to partial ferromagnetic alignment) reduce GMR. In addition, the presence of interface roughness in GMR multilayers can mediate 90° coupling, which can lead to perpendicular alignment of the magnetic moments of adjacent magnetic films at zero field, instead of an anti-parallel or parallel one, and thus reducing significantly GMR [5].
• Magnetic properties: Besides magneto-transport, the amount of disorder at the surface/interface can also influences magnetic properties of thin magnetic films, like for instance coercivity, magnetic domain structure, and magnetization reversal [6-7]. These magnetic properties are important in applications related to magnetic recording industry, magnetic memories, etc. The coercivity of chemically etched NiFeCo films was found to increase with increasing surface roughness. By use of x-ray resonant magnetic scattering, the hysteretic behaviour of CoFe thin films with varying roughness was investigated, and it was found that the coercivity increases with increasing surface roughness. For thin Co films deposited on plasma etched Si, the uniaxial magnetic anisotropy has been to decrease with increasing surface roughness. The coercivity increases with increasing surface roughness also for ultra-thin Co films on Ar⁺ sputtered Cu substrates. Finally, surface/interface roughness has been shown to have a significant influence on the demagnetizing field [7].
• Electrical transport properties: Rough surfaces/interfaces of thin films contribute to electron scattering and thus to the film electrical conductivity. Oscillations in electrical resistivity vs. thickness (quantum size effects: QSE) have been observed i.e., in Pt films, superposed metal films of Ag, In and Ga on a thick base of Ag or Au [8]. The conductivity of epitaxial Ag/Si films was shown to be proportional to the square of the interface roughness amplitude w and to have a complex dependence on the lateral roughness correlation length x . For quantum wells, it has been shown that the mobility of the two-dimensional electrons in AlAs/GaAs quantum wells is strongly influenced by interface roughness. Roughness was also shown to have an influence on the Hall mobility of metal-oxide-semiconductor (MOS) field effect transistor devices, and on the electron mobility of high-mobility Si inversion layers - which improves if the exponential correlation function at the Si/SiO₂ interface is taken into account. The interface roughness has also been shown to affect the electron sub-band structure of InAs/GaSb semiconducting quantum wells. Moreover, the degree of roughness irregularity at short lateral length scales has significant influence on the conductivity of metallic/semiconducting thin films [8]. Nonetheless, additional complexity of the transport properties arises due to electron scattering by grain boundaries, in addition to scattering from surface/interface roughness [8].

• Mechanical  properties: The problem of how surface roughness influence the adhesion between elastic solids and a hard solid substrate is important from both the fundamental and technological point of view as for example in polymer/metal junctions. This topic was studied initially by Fuller and Tabor [9], and it was shown that a relatively small surface roughness can remove the adhesion. In their model it was considered a Gaussian distribution of asperity heights with all asperities having the same radius of curvature. The contact force was obtained by applying the contact theory by Johnson et al. [10] to each individual asperity. However, this approach considers surface roughness over a single lateral length scale. The maximum pull-off force is expressed as a function of a single parameter which determines (the statistically averaged) competition between compressive forces from higher asperities that try to pull the surfaces apart, and the adhesive forces from lower asperities that try to hold the surfaces together. On the other hand, random rough surfaces, which are commonly encountered for solid substrate surfaces, posses roughness over many different length scales rather than a single one. In this case for random self-affine rough surfaces [11, 12] it has been shown that when the local fractal dimension D is larger than 2.5 the adhesive force may vanish or at least be reduced significantly. At any rate, for real polymers viscous-elastic [11] effects are present which alter the interaction between polymer-metal substrate. In this case modifications are required since surface roughness introduces fluctuating forces with a wide distribution of frequencies [11]. Notably, the determination of the real contact area is a fundamental problem with important technological implications. The latter include, for example, heat transfer phenomena between solid bodies, electrical transport, sliding friction, adhesive forces between solid bodies in direct contact [11, 12] etc.

• Wetting phenomena:  The phenomenon of wetting of solid substrates exposed to a gas (under thermodynamic equilibrium conditions) is a topic of intense research from both the fundamental [13] and application [3-5] point of view. The wetting of a substrate by a liquid is driven by the strong substrate/particle (van der Waals) attraction forces. Currently there is a rather clear microscopic understanding of wetting on flat solid substrates [13]. In this case, the liquid film thickness is described as a function of substrate/particle and inter-particle interactions for specified thermodynamic parameters (pressure and temperature). Experiments using noble gases [13] on different substrates confirmed that the thickness of the wetting layer grows with increasing substrate/particle attraction (for fixed thermodynamic parameters), as well as that complete wetting (diverging liquid film thickness) occurs for stronger substrate/particle attraction than inter-particle interactions (and thermodynamic conditions approaching liquid-gas coexistence). The presence of surface roughness and/or chemical heterogeneities further complicate wetting phenomena [14].

• Superconductivity (proximity effects in SN junctions):  Proximity effects at the junction of a normal metal film and a thin superconducting film constitute currently a topic of intense research in the field of superconductivity (e.g., anisotropic high-temperature Cu-oxide superconductors) [15, 16]. The penetration of Cooper pairs from the superconductor into the metal and electrons from the normal metal into superconductor determines the proximity effects [15, 16], which manifest themselves by reducing the critical temperature of a thin superconducting film covered by a thick normal metal film [15, 16]. We have that the proximity effect is  influenced predominantly  by the degree of interface irregularity at short wave lengths as expressed by the roughness exponent H and the roughness ratio of roughness amplitude (w) /correlation length (x) [17].  Therefore, in future studies of rough SN-junctions the precise roughness nature at all roughness wavelengths should be properly quantified (measurement of H, w , and x , e.g., by x-ray reflectivity, scanning probe microscopy etc.) in order to gauge precisely morphology contributions on proximity effects.

• Electrical double layers: A diverse variety of important applications in electrochemistry, colloid science, biophysics, semicoductor technology etc. are based on the Gouy-Chapman theory of electrolyte plasma near a flat charged wall which represents the electrode [17, 18]. For rough interfaces, we can not simply consider metal/electrolyte interface roughness by replacing in the equation for the capacitance the flat surface area by the true surface area due to roughness because the characteristic lateral roughness length scales L can compete with system characteristic length scales such as the Debye length  leading to different functional dependence on potential and electrolyte concentration [18, 19]. 

Related references

[1] P. Meakin, Fractals, Scaling, and Growth Far from Equilibrium (Cambridge University Press, Cambridge, 1998)

[2] M. Kardar, G. Parisi, and Y. C. Zhang, Phys. Rev. Lett. 56 (889 (1986); T. Halpin-Heally, Phys. Rep. 254, 215 (1995); J. Krim and G. Palasantzas, Int. J. of Mod. Phys. B 9, 599 (1995).

[3] D.J. Srolovitz, Act. Mater. 37, 621 (1989); G. Palasantzas and J. Th. M. De Hosson, Appl. Phys. Lett. 78, 3044 (2001).

[4] See references there in: J. Barnas and Y. Bruynseraede, Europhys. Lett. 32, 167 (1995); Phys. Rev. B 53, 408 (1996), S. Zhang and P. M. Levy, Phys. Rev. Lett. 77, 916 (1996), J. Barnas and G. Palasantzas, J. Appl. Phys. 82, 3950 (1997); R. Schad et al., Phys. Rev. B 59 1242 (1999).

[5] J. C. Slonczewski, Phys. Rev. Lett. 67, 3172 (1991).

[6] Ultrathin Magnetic Structures I and II, ed. by J. A. C. Bland and B. Heinrich (Springer, New York, 1994).

[7] See references there in: Y.-P. Zhao, R.M. Gamache, G.-C. Wang, T.-M. Lu, G. Palasantzas, J.Th.M. De Hosson, J. Appl. Phys. 89, 13 25 (2001); Y. P. Zhao, G. Palasantzas, G. -C. Wang, T. -M. Lu, and J. Th. M. De Hosson, Phys. Rev. B 60, 1216 (1999)

[8] See references there in: Y. P. Zhao, G. -C. Wang, and T. -M. Lu, Characterization of amorphous and crystalline rough surfaces-principles and applications, (Experimental Methods in the Physical Science Vol. 37, Academic Press, 2000); G. Fishman and D. Calecki, Phys. Rev. Lett. 62, 1302 (1989); A. E. Meyerovich and A. Stepaniats, Phys. Rev. B 60, 9129 (1999); H. Sakaki et al., Appl. Phys. Lett. 51, 1934 (1987); G. Palasantzas and J. Barnas, Phys. Rev. B 56, 7726 (1997); G. Palasantzas, Y.P. Zhou, G.C. Wang, T.M. Lu, J. Barnas, J.Th.M. De Hosson, Phys. Rev. B 61, 11109 (2000).

[9] K. N. G. Fuller and D. Tabor, Proc. R. Soc. London, Ser. A 345, 327 (1975).

[10] K. L. Johnson, K. Kendall, and A. D. Roberts, Proc. R. Soc. London, Ser. A 324, 301 (1971).

[11] B. N. J. Persson and E. Tosatti, J. Chem Phys. 115, 5597  (2001); J. Chem. Phys. 115, 3040 (2001); B. N. J. Persson, J. Chem. Phys. 115, 3840 (2001).

[13] S. Dietrich, in Phase Transitions and Critical Phenomena, edited by C. Domb and J. Lebowitz (Academic Press, London, 1988), Vol. 12, pp. 1–128; R. Evans, in Liquids at Interfaces, Proceedings of the Les Houches Summer School, Session XLVIII, edited by J. Charvolin, J. F. Joanny, and J. Zinn-Justin (Elsevier, Amsterdam, 1990); H. Gau et al., Science 283, 46 (1999); K. Kargupta et al., Phys. Rev. Lett. 86, 4536 (2001);  J. Bico et al., Europhys. Lett. 47, 220 (1999); S. Dietrich and M. Schick, Phys. Rev. B 33, 4952 (1986); J. L. Seguin et al., Phys. Rev. Lett. 51, 122 (1983); M. Bienfait et al., Phys. Rev. B 29, 983 (1984); J. Krim, J. G. Dash, and J. Suzanne, Phys. Rev. Lett. 52, 640 (1984).

[14] M. Kardar and J. O. Indekeu, Europhys. Lett. 12, 161 (1990); R. R. Netz and D. Andelman, Phys. Rev. E 55, 687 (1997); G. Palasantzas, Phys. Rev. B 51, 14612 (1995); G. Palasantzas and G. Backx, Phys. Rev. B 55, 9371 (1997); G. Palasantzas and G. Backx , Phys. Rev. E 66, 021604-1 (2002).

[15] P. G. de Gennes, in Superconductivity of Metals and Alloys (W. A. Benjamin, New York, 1966); G. Deutscher and P. G. de Gennes, in Superconductivity, ed. by R. D. Parks (Marcel Dekker, Inc. New York, 1969) Vol. 2.; K. A. Delin and A. W. Kleinsasser, Supercond. Sci. Technol. 9, 1996, 227.

[16] E. Polturak, G. Koren, D. Coden, O. Nesher, R. G. Mints, and I. Snapiro, Phys. Rev. B 57, 1998, 14068; A. A. Abrikosov, in Fundamentals of the Theory of Metals (North-Holland, Amsterdam, 1988);  R. G. Mints and I. B. Snapiro, Phys. Rev. B 57, 1998, 10318.

[17] G. Palsantzas, Solid State Commun. 112, 97 (1999); G. Palsantzas and J. Th. M. Hosson, Physica C 330, 99 (2000); G. Palasantzas and J. Th. M. De Hosson,  Physica C 355, 211 (2001).

[18] P. Delahay, Double Layer and Electrode Kinetics (Wiley, New York, 1965); E. Gileadi, E. Kirowa-Eisner, and J. Penciner, Interfacial Electrochemistry (Addison-Wesley, Reading, Mass., 1975); B. B. Damaskin and O. A. Petrii, Introduction to Electrochemical Kinetics (Vyshaya Shkola, Moscow, 1975); J. N. Israelachvili, Intermolecular and Surface Forces (Academic, London, 1990); D. Andelman, in Handbook of Physics of Biological Systems, edited by R. Lipowsky (Elsevier, New York, 1994), p. 577; S. Liu, in Condensed Matter Physics Aspects of Electrochemistry, edited by M. P. Tosi and A. A. Kornyshev (World Sci-entific, Singapore, 1991), p. 329; G. Gouy, J. Phys. (Paris) 9, 457 (1910); D. L. Chapman, Philos. Mag. 25, 475 (1913); P. Debye and W. Huckel, Phys. Z. 24, 185 (1924); 24, 305 (1924); P. Debye, Phys. Z. 25,97 (1924).

[18] L. I. Daikhin, A. A. Kornyshev, and M. Urbakh, Phys. Rev. E 53, 6192 (1996).

[19] G. Palasantzas and G. Backx Self-affine roughness effects on the double-layer charge density and capacitance in the non-linear regime. J. Chem. Phys. 118, 4631 (2003); G. Palasantzas and G. Backx Self-affine and mound roughness effects on the double layer charge capacitance. J. Appl. Phys. 92, 7175 (2002); G. Palasantzas and G.M.E.A. Backx, 'Roughness effects on the double-layer charge capacitance: The case of Helmholtz layer induced electrode roughness attenuation '  Surf. Sci. 540, 401 (2003).