In late 1980s, two technological problems motivated us to study cracking in complex material systems. The first problem was the design of all-ceramic composites for high temperature applications. Ceramics can sustain high temperature, but are susceptible to fracture. One strategy to toughen ceramics, quite ironically, is to introduce some weakness into them. For example, in a two-ceramic composite, with one ceramic being the matrix, and the other the fibers, the interfaces between the matrix and the fibers control the overall behavior of the composite. Let’s say the composite carries a load along the direction of the fibers, and cracks exist in the matrix. When the interface is sufficiently weak, the cracks grow in the matrix, causing the fiber-matrix interface to debond without breaking the fiber. The weak interface protects the fibers. When the fibers do finally break, the fibers slide against the matrix, absorbing a large amount of energy. To effectively design composites, A.G. Evans and others were developing methods to measure interfacial fracture energy. We were fortunate to be part of the team Evans put together, helping to develop the mechanics of interfacial fracture.

The second problem was durability of electronic devices. An electronic device integrates diverse materials (semiconductors, dielectrics, conductors, etc.). Stresses arise in the materials for many reasons, such as temperature change and material transport. These stresses can cause cracking in individual materials, and debonding between dissimilar materials. When we wrote our article, the microelectronic industry was confronting these problems largely by trial and error. There was a clear need to apply mechanics principles in failure analysis and product design. In the last decade, researchers have developed testing procedures specifically for electronic device structures, some of which have been successfully put to industrial use. Our article has found new friends in the industry and academia, and contributed to the process of introducing fracture mechanics to this critical technology.

The following more recent articles have described later developments in the field:

• A.G. Evans and J.W. Hutchinson, "The thermomechanical integrity of thin films and multilayers," Acta Metall. Mater. 43, 2507-2530 (1995).
• J.W. Hutchinson and A.G. Evans, "Mechanics of Materials: Top-Down Approaches to Fracture," Acta Materialia 48, 125-135 (2000)
• R.F. Cook and Z. Suo, "Mechanisms active during fracture under constraint. MRS Bulletin," 27, 45-51 (2002).
• Z. Suo, "Fracture in Thin Films." In Encyclopedia of Materials: Science and Technology. Second edition, Elsevier Science Limited, Oxford, UK, 2001.

The field has been active, driven by applications in many technologies, by unsolved basic problems, and by recently available experimental tools. The following provides a partial list of promising directions for future research.

• Fracture in structures of complex geometries. Our article is mainly for layered materials and blanket films. Practical devices have more complex structural components, such as pads, lines, and studs. Although finite element programs can analyze structures of any geometry, it is desirable to have a catalogue of solutions for structures of increasing complexity. An understanding of these solutions will give insight to the engineer, particularly in the earlier design phase, when he or she has to choose among many structures.
• Structures containing compliant materials. Various polymers and porous materials are being introduced in microelectronics and optoelectronics. For example, organic materials with low dielectric constants enhance the performance of microprocessors, and polymeric substrates are being developed for large-area electronics. These compliant materials are often used together with stiff materials such as metals and ceramics. The large stiffness mismatch among these materials can give rise to unusual fracture behaviors.
• The roles of inelastic deformation. Plasticity and creep are known to be important to fracture in bulk materials. Small structures with inelastic components can exhibit far richer behaviors. Both the material arrangement and the feature size play important roles. A recent example involves fracture in a brittle layer caused by ratcheting plastic deformation in an adjacent ductile layer. The phenomenon has been observed in both microelectronic devices and thermal barrier coatings under cyclic temperature.
• Evolving small structures. Processes like oxidation, phase transition, and atomic diffusion change structures over time, and often are the root cause of fracture. Examples include failure of thermal barrier coatings, and electromigration-induced fracture. Understanding of these processes is important to the lifetime prediction.
• Crack nucleation. A small structure may take more time to nucleate a crack than to grow the crack. Theoretical and experimental studies of crack nucleation processes remain a great challenge. Modern surface science tools such as atomic force microscopy and scanning tunneling microscopy offer opportunities to study events leading to crack nucleation in unprecedented detail.

John W. Hutchinson
Division of Engineering and Applied Sciences
Harvard University