Integrated nanoelectronics for the future.

g ordon Moore’s prediction made over 40 years ago, that the number of transistors in an integrated circuit would double roughly every 24 months, continues to be the guiding principle of the semiconductor and computing industries. As transistor count increases, each transistor becomes smaller, faster and cheaper, leading to an unprecedented increase in microprocessor performance1. Today, the so-called 65-nm technology node (related to half the metal pitch of a dynamic random access memory device) of the most advanced microprocessors in production2 has transistor gate lengths of 35 nm and a SiO2 gate oxide thickness of 1.2 nm. However, further miniaturization by traditional geometric scaling of conventional silicon devices faces many technical challenges. These include, amongst others, excessive leakage currents through the SiO2 gate oxide, exponentially increasing leakage currents between transistor source and drain, increasing source–drain access resistance, carrier-mobility degradation in the transistor channel from increasing electric field, as well as increasing device-to-device variation3. The silicon community has been responding promptly and is overcoming many of these challenges with various research breakthroughs and innovations to improve present and future generations of the complementary metal–oxide– semiconductor (CMOS) transistor technology for gigascale digital systems. Examples of these innovations include mechanically strained silicon channels, where strain enhances electron and hole mobility, as well as gate stacks composed of dielectrics with high dielectric constant (high-K dielectrics) and metal electrodes for higher drive current and lower overall gate leakage. Furthermore, non-planar designs such as tri-gate transistors are being developed to mitigate problems arising, for example, from short-channel effects. Combined, these innovations and strategies will enable continued logic CMOS transistor downscaling and improvements in performance trends until at least the middle of the next decade. Furthermore, intensive research is currently being carried out by both industry and academia on electronic materials other than silicon (for example, other semiconductor materials, semiconductor nanowires and carbon nanotubes) and their integration onto silicon wafers for future high-performance and energy-efficient very-large-scale integrated (VLSI) nanoelectronics applications beyond 2015.