Contact mechanics

Contact problems are central to Solid Mechanics, because contact is the principal method of applying loads to a deformable body and the resulting stress concentration is often the most critical point in the body. Contact is characterized by unilateral inequalities, describing the physical impossibility of tensile contact tractions (except under special circumstances) and of material interpenetration. Additional inequalities and/or non-linearities are introduced when friction laws are taken into account. These complex boundary conditions can lead to problems with existence and uniqueness of quasi-static solution and to lack of convergence of numerical algorithms. In frictional problems, there can also be lack of stability, leading to stick±slip motion and frictional vibrations. If the material is non-linear, the solution of contact problems is greatly complicated, but recent work has shown that indentation of a power-law material by a power law punch is self-similar, even in the presence of friction, so that the complete history of loading in such cases can be described by the (usually numerical) solution of a single problem. Real contacting surfaces are rough, leading to the concentration of contact in a cluster of microscopic actual contact areas. This a€ects the conduction of heat and electricity across the interface as well as the mechanical contact process. Adequate description of such systems requires a random process or statistical treatment and recent results suggest that surfaces possess fractal properties that can be used to obtain a more ecient mathematical characterization. Contact problems are very sensitive to minor pro®le changes in the contacting bodies and hence are also a€ected by thermoelastic distortion. Important applications include cases where non-uniform temperatures are associated with frictional heating or the conduction of heat across a non-uniform interface. The resulting coupled thermomechanical problem can be unstable, leading to a rich range of physical phenomena. Other recent developments are also brie ̄y surveyed, including examples of anisotropic materials, elastodynamic problems and fretting fatigue. # 1999 Published by Elsevier Science Ltd. All rights reserved. International Journal of Solids and Structures 37 (2000) 29±43 0020-7683/00/$ see front matter # 1999 Published by Elsevier Science Ltd. All rights reserved. PII: S0020-7683(99 )00075-X www.elsevier.com/locate/ijsolstr * Corresponding author. Tel.: +1-313-936-0406; fax: +1-313-647-3170. E-mail addresses: [email protected] (J.R. Barber), [email protected] (M. Ciavarella) 1 Presently at CNR-IRIS Computational Mechanics of Solids, str. Croce®sso, 2/B 70126, Bari, Italy. A casual survey of the kinds of engineering applications to which the techniques of Solid Mechanics are applied will show that the vast majority of solid bodies are loaded by being pressed against another body. The only alternatives comprise loading of the boundary by ̄uid pressure or various kinds of body force such as gravitational or magnetic forces, but even in such cases, the reaction force required to maintain equilibrium will almost invariably be provided at a contact interface. When we also recall that contacts between bodies generally constitute stress concentrations and are therefore likely sites for material failure, it is not surprising that Contact Mechanics has occupied a central place in the development of Solid Mechanics over the years and continues to do so today. Additional interest in the subject is generated by the fact that the inevitable roughness of contacting surfaces generates a very complex local structure at which extreme conditions are likely to occur, particularly if sliding takes place, leading to frictional heating and very high local temperatures. Historically, the development of the subject stems from the famous paper of Heinrich Hertz (1882) giving the solution for the frictionless contact of two elastic bodies of ellipsoidal pro®le. Hertz' analysis still forms the basis of the design procedures used in many industrial situations involving elastic contact. Since 1882, the subject of contact mechanics has seen considerable development. Two major threads can be distinguished Ð from a mathematical standpoint, emphasis has been placed on the extension of Hertz' analysis to other geometries and constitutive laws and on the proof of theorems of existence and uniqueness of solution, whereas engineers have focussed on the solution of particular problems in an attempt to understand and in ̄uence phenomena that occur in practical contacting systems, both on the macro and the micro scale. Gladwell (1980) provides a compendious treatment of the various contact geometries that had been treated up to that time, including an invaluable survey of the rich Russian literature on the subject. Johnson (1985) gives an excellent overview of the range of contact problems that have come under consideration and achieves a nice balance between mathematical rigour and engineering practicality. In an attempt to obtain a similar balance, we ®rst revisit the de®ning characteristics of contact mechanics in the mathematical framework of problems involving unilateral inequalities. Particular attention is given to the additional features associated with the presence of friction at the interface, where nonexistence, non-uniqueness and instability of the quasi-static solution can be obtained with suciently high friction coecients. We then introduce the concept of self-similarity, which provides a powerful method for indentation problems even for non-linear materials and other aspects of contact beyond the elastic limit are also discussed. The specialized areas of anisotropic and elastodynamic contact are brie ̄y summarized and the paper concludes with a discussion of recent developments in the characterization and contact of rough surfaces and in thermoelastic contact. 1. Unilateral inequalities The essence of a contact problem lies in the fact that any point on the boundary of each body must either be in contact or not in contact. If it is not in contact, the gap g between it and the other body must be positive ( g>0), whereas if it is in contact, g = 0, by de®nition. A dual relation involves the contact pressure p between the bodies which must be positive ( p>0) where there is contact and zero where there is no contact. The inequalities serve to determine which points will be in contact and which not. If the contact area is prescribed, it can be shown from classical existence and uniqueness proofs that the equations alone are sucient to de®ne the stresses and displacements throughout the bodies, but there is then of course no guarantee that the solution will satisfy the inequalities. Fichera (1964, 1972) proved that the complete problem, including the inequalities, has a unique solution when the material is linear elastic and many related proofs have since been advanced for other classes of contact J.R. Barber, M. Ciavarella / International Journal of Solids and Structures 37 (2000) 29±43 30