Top-Down Paradigm to Miniaturization

Abstract

The current development of electronics is ultimately a consequence of the invention of the MOS-FET. The idea upon which the FET is based, of controlling the conductivity of a region by varying its surface potential, precedes by many years its practical realization and industrial implementation. In fact, the invention of the metal–insulator–semiconductor (MIS) FET goes as far back as to 1926 for the depletion-mode FET [14] and to 1935 for the enhancement-mode FET [15]. Quite ironically, in an attempt to produce MIS-FETs on semiconducting germanium, Bardeen, Brattain, and Shockley discovered the bipolar transistor in 1947 (an insider view of the history of the transistor discovery¹ is given in [16]). It was only after the change of substrate (from germanium to silicon) that the first FET could actually be prepared (by Khang and Atalla in 1960 [17]) by changing the insulator to SiO₂ and the semiconductor to silicon – and the MOS-FET opened the main avenue to the IC era.² Most ICs are indeed constituted by MOS-FETs, connected in a way that allows assigned electrical functions to be performed – the more sophisticated the functions, the more complex the circuit. The early IC market was however slow to develop and its exponential growth was triggered by Fairchild’s decision to sell IC chips at a price lower than the sum of the prices of the individual components necessary to make an equivalent circuit [20].