Stress In Polycrystalline Films Research Articles

Mechanisms of Stress Generation in Thin Films and Coatings

Analysis of the literature data on the reasons for the development of mechanical stresses in epitaxial, polycrystalline, and amorphous films during their formation and under various external influences is carried out. The mechanism of the appearance of internal stresses during heteroepitaxial growth of films caused by the mismatch of the crystal lattice constants of the film and substrate is described. The relationship between the development of mismatch stresses in heteroepitaxial films and changes in the nature of their growth is shown. Models of the occurrence of compressive and tensile stresses in polycrystalline films due to the formation and coalescence of islands at the initial stage of their growth are considered. The regularities of the evolution of internal stresses in continuous films are discussed depending on the conditions of their deposition, as well as their chemical composition, structure, and mechanical properties. The mechanisms of development of internal stresses in thin films associated with the formation of point defects in them, the incorporation of impurities, and phase transformations occurring in the deposition process are reviewed. External factors that lead to the appearance of stresses in thin films during their storage and operation are considered in detail.

Механизмы возникновения напряжений в тонких пленках и покрытиях

The paper considers current conceptions of generation of mechanical stresses in epitaxial, polycrystalline and amorphous films during their growth and under different external actions. The mechanism of stress generation in geteroepitaxial films due to misfit in crystal lattices of the film and substrate is described. The relation between arising of the misfit stress in heterostructures and variation of their growth mode is shown. The mechanisms of generation of compressive and tensile stresses in polycrystalline films caused by nucleation and coalescence of islands at the beginning of their growth are considered. Different aspects of evolution of intrinsic stresses in continuous films are discussed in dependence of their deposition conditions, chemical composition, microstructure and mechanical properties. Special attention is given to consideration of generation mechanisms of intrinsic stresses in thin films concerned with formation of pint defects, incorporation of impurities and phase transformations during deposition. Factors leading to arising extrinsic stresses in thin films during their storage and operation are described in details.

Disclosing the origin of the postcoalescence compressive stress in polycrystalline films by nanoscale stress mapping

A method based on atomic force microscopy, which allows mapping the residual stress at nanoscale on the surface of crystalline solids [Polop et al., Nanoscale 9, 13938 (2017)], sheds light on the controversial origin of the intrinsic compression that arises in continuous polycrystalline films under high atomic mobility conditions. The maps of residual stress reveal that the compression is concentrated in narrow strips adjacent and parallel to the grain-boundary triple junctions, but not inside the grain boundary as usually assumed. We explain these findings in the light of Mullins's theory for surface diffusion of adatoms towards grain boundaries. As the surface slope at the grain-boundary triple junctions is constrained by the balance between interfacial tensions, the kinetic surface profile is different from the mechanical equilibrium profile predicted by the Laplace-Young equation. Where the curvatures of both profiles differ, an intrinsic stress is generated in the form of Laplace pressure. The average value of the stress profile is compressive. In turn, the resulting stress induces a Srolovitz-type surface diffusion, which counters the Mullins-type diffusion. The competition between both diffusion mechanisms addresses the main fingerprints of the intrinsic stress behavior in polycrystalline films, namely, the stabilization of the stress profile during growth, its reversibility with the flux, and the kinetics of the stress relaxation and recovery.

Intrinsic Compressive Stress in Polycrystalline Films is Localized at Edges of the Grain Boundaries.

The intrinsic compression that arises in polycrystalline thin films under high atomic mobility conditions has been attributed to the insertion or trapping of adatoms inside grain boundaries. This compression is a consequence of the stress field resulting from imperfections in the solid and causes the thermomechanical fatigue that is estimated to be responsible for 90% of mechanical failures in current devices. We directly measure the local distribution of residual intrinsic stress in polycrystalline thin films on nanometer scales, using a pioneering method based on atomic force microscopy. Our results demonstrate that, at odds with expectations, compression is not generated inside grain boundaries but at the edges of gaps where the boundaries intercept the surface. We describe a model wherein this compressive stress is caused by Mullins-type surface diffusion towards the boundaries, generating a kinetic surface profile different from the mechanical equilibrium profile by the Laplace-Young equation. Where the curvatures of both profiles differ, an intrinsic stress is generated in the form of Laplace pressure. The Srolovitz-type surface diffusion that results from the stress counters the Mullins-type diffusion and stabilizes the kinetic surface profile, giving rise to a steady compression regime. The proposed mechanism of competition between surface diffusions would explain the flux and time dependency of compressive stress in polycrystalline thin films.

Clamping effect by the substrate on the intrinsic stress in polycrystalline films

Mechanical constrictions imposed by a massive substrate to the interactions between coalescing grains within a polycrystalline film deposited on top are investigated by finite element modelling. Such interactions, which underlie the origin of the post-coalescence compression, as suggested recently, induce the twisted zipping (a kind of elastic deformation combining radial and shear strain) of grain boundaries during coalescence. In particular, the substrate clamps the azimuthal interactions that give rise to the crystallography reorientation of the grains (which happens by rotation and/or shear strain) to form pseudocoherent low-angle and coincident-site-lattice grain boundaries. Depending of the mechanical constants of the film/substrate system as well as the aspect-ratio of the structural features in the polycrystalline film, different mechanical responses to the clamping effect are identified, they are: inhibited coalescence of early islands with the formation of a void network between them, coalescence with crystalline defects of larger features growing preferentially along the interface with the substrate, and defect-free coalescence of high aspect-ratio features (columns) by grain boundary twisted zipping.

Stress engineering using low oxygen background pressures during Volmer–Weber growth of polycrystalline nickel films

A robust strategy for controlling the level of residual stress in polycrystalline films remains elusive, owing to the complex coevolution of the surface, microstructure, and intrinsic stress during Volmer–Weber film growth. Recent improvements in the understanding of stress evolution mechanisms have led to the possibility of engineering the intrinsic stress through the control of thin film growth conditions. Here, the authors demonstrate stress engineering during deposition of polycrystalline Ni films through control of the oxygen partial pressure. The physical mechanisms of stress management during codeposition of nickel and oxygen are investigated using in situ stress measurements and ex situ structural and chemical characterizations. The intrinsic stress in Ni films is affected by grain growth during deposition (which causes a tensile stress) and by Ni adatom trapping at grain boundaries and oxygen incorporation in the Ni lattice (which cause a compressive stress). The authors show direct evidence that a small amount of oxygen suppresses grain growth during deposition. They suggest that the presence of chemisorbed oxygen limits surface diffusion of Ni adatoms, thereby limiting adatom trapping at grain boundaries. The presence of oxygen therefore affects the mechanisms for development of both tensile and compressive stresses, providing a direct method for engineering the residual stress in as-deposited Ni films. Finally, the authors demonstrate a process for evaporative deposition of “zero” stress Ni films by introducing a very low level of background impurities, with the resultant films containing only 1.2 at. % oxygen.

Effects of oblique-angle deposition on intrinsic stress evolution during polycrystalline film growth

Precise control of the residual stress in polycrystalline films, which remains a central but difficult task in a variety of emerging technologies, requires synergistic manipulation of multiple processing parameters. In addition to the substrate temperature and the rate of deposition, the angle of incidence of the deposition flux is known to affect structure evolution processes during film formation. However, its role in influencing the evolution of the intrinsic stress has not been determined. In this work, we investigate the effects of oblique-angle deposition on intrinsic stress evolution in polycrystalline gold and nickel films, using in situ stress measurements as well as ex situ surface and microstructure characterization. We find that as the angle between the surface normal and the direction of the deposition flux increases, the thickness at which film coalescence occurs increases and the post-coalescence stress shifts toward the tensile direction. We suggest that the first trend is associated with the effects of shadowing on the nucleation density in the pre-coalescence regime, and we attribute the second trend to an increase in surface roughness and shadowing effects associated with the dome-shaped surfaces of individual grains. According to this view, oblique-angle deposition on a mesoscopically rough surface eliminates or reduces condensation at and near grain boundaries, and therefore lowers the rate of adatom–grain boundary attachment, resulting in reduction in the compressive component of the intrinsic stress. The stress evolution map built from this work suggests routes to achieve specific stress levels in polycrystalline films.

The flow stress in polycrystalline films: Dimensional constraints and strengthening effects

An analytical model is presented in order to derive a general expression for the flow stress in polycrystalline films which encompasses and correlates dimensional constraints and strengthening effects. The model is based on the Thompson approach, which is extended to take into account both different grain aspect ratios and distinct strengthening contributions. It allows an accurate prediction of the growth textures in polycrystalline CdTe thick films when grain growth is driven by strain energy minimization. The model also matches the experimental data concerning the grain size and film thickness dependences of the yield stress in polycrystalline Cu thin films either deposited on a substrate or freestanding. Interestingly, the yield stress is found to be fitted by a modified Hall–Petch relation resulting in a d − n dependence in which the exponent n varies between ½ and 1 as a function of the grain size for a given thickness.

Chemistry-induced intrinsic stress variations during the chemical vapor deposition of polycrystalline diamond

Intrinsic tensile stresses in polycrystalline films are often attributed to the coalescence of neighboring grains during the early stages of film growth, where the energy decrease associated with converting two free surfaces into a grain boundary provides the driving force for creating tensile stress. Several recent models have analyzed this energy trade off to establish relationships between the stress and the surface∕interfacial energy driving force, the elastic properties of the film, and the grain size. To investigate these predictions, experiments were conducted with diamond films produced by chemical vapor deposition. A multistep processing procedure was used to produce films with significant variations in the tensile stress, but with essentially identical grain sizes. The experimental results demonstrate that modest changes in the deposition chemistry can lead to significant changes in the resultant tensile stresses. Two general approaches were considered to reconcile this data with existing models of stress evolution. Geometric effects associated with the shape of the growing crystal were evaluated with a finite element model of stress evolution, and variations in the surface∕interfacial energy driving force were assessed in terms of both chemical changes in the deposition atmosphere and differences in the crystal growth morphology. These attempts to explain the experimental results were only partially successful, which suggests that other factors probably affect intrinsic tensile stress evolution due to grain boundary formation.

Intrinsic compressive stress in polycrystalline films with negligible grain boundary diffusion

The model developed here describes compressive stress evolution during the growth of continuous, polycrystalline films (i.e., beyond the point where individual islands have coalesced into a continuous film). These stresses are attributed to the insertion of excess adatoms at grain boundaries. Steady state occurs when the strain energy at the top of the film is balanced by the local excess chemical potential of surface adatmos. Strain gradients associated with this compressive stress mechanism depend on the kinetics of the process. In the absence of grain boundary diffusion, these strain profiles are determined by the ratio of the atom insertion and growth rates. The steady-state strain and the strain evolution kinetics also depend on the two key length scales, the grain size, and the film thickness. The ratio of these two lengths (i.e., the grain aspect ratio) can also have a significant influence on the thermodynamic driving force for strain evolution if the grain sizes are sufficiently small. The model is fit to existing data for the growth of AlN films. However, more detailed comparisons will require experiments that are specifically designed to test this model.

Defect annealing in polycrystalline silicon films

Studies of the isochronal annealing of paramagnetic vacancy-type defects (VV centres) in polycrystalline silicon films by electron spin resonance are reported. Silicon films of 1 and 10 μm thickness were deposited by the thermal decomposition of silane at 700–800°C in N 2 and H 2 atmospheres. The number N VV (cm -2) of VV centres in the films after their deposition was found to be approximately (3–4)x10 14 cm -2. During annealing in the 450–750°C temperature range a considerable increase in N VV was observed, the increase depending on substrate temperature and structure (amorphous or monocrystalline). From this we conclude that intrinsic stresses in polycrystalline films are responsible for the increase in N VV. On comparing the data for the annealing of polycrystalline films with the results on the annealing of amorphous silicon produced by ion implantation we reach the conclusion that the disappearance of VV centres in the second case is due to an interaction with mobile point defects rather than to the thermal decomposition of the VV centres.