For decades, advanced microelectronics has delivered, year by year, products with more functions and higher performance at the same costs. This has been achieved by continuous lateral shrinkage of monolithic integrated device dimensions and by relying on simple material concepts with silicon as semiconductor, silicon oxides as dielectrics and aluminium as interconnect metal. With the 100 nm length approaching, traditional trade-offs fail and we see a paradigm shift requiring sophisticated materials science from semiconductors to dielectrics and metals.

A few years after the invention of the bipolar transistor, the basic electronic semiconductor material changed from germanium to silicon. During that switch around 1960, considerable interest was focused on bulk, unstrained SiGe alloys. Advanced epitaxy methods like molecular beam epitaxy or chemical vapour deposition have enabled the growth of high quality, thin, strained SiGe layers on Si substrates since around 1985. The availability of strained SiGe/Si structures strongly stimulated the research on silicon-based heterostructure devices resulting within a few years in the fastest silicon-based transistors and other very attractive options. However, it was only in 1998 when the volume production of the SiGe heterobipolar transistor (HBT) circuits for mobile communication started, that a broad public audience became aware of this new strained layer heterostructure material which is in the main not available in bulk form. The technology involved in applying this material system will spread to other traditional and novel device areas: carbon bandgap engineering, strain adjustment techniques, quantum confinement and self-assembling.

This book is based partly on a revised version...