Semiconductors 1853-1919: an historical study of selenium and some related materials

Abstract

The history of semiconductors began in 1833 when Faraday noted that the conductivity of silver sulphide increased with temperature. This study ends in 1919 when Pohl went to Göttingen. Five periods are identified: 1833-1895; ^ 1895-1919; 1919-1931; 1931-1948; 1948-present. Only towards the close of the second period was the semiconductor 'class' recognised; until then researches on selenium did not interact with those on other materials. Faraday's discovery excited little interest, but photoconductivity in selenium, 1873, engendered considerable activity during which rectification and photovoltaicity were noted, Allotropic and electrolytic theories were suggested, (by 1877), with the latter being extensively developed. Theories of electricity were insufficiently agreed to form other than a qualitative base for conduction in solids; until the electron theory and Arrhenius's picture, electrolysis was not an unsatisfactory basis on which to construct an account of the properties of selenium. The understanding of electrolysis and the electron theory made an alternative description of the action of selenium desirable and attractive. Initially, to cl912, an allotropic theory was developed to be replaced, by 1914, by an electron theory for there was no direct empirical evidence in support of the former. With the adoption of an electron theory selenium was no longer unique and its properties were compared with those of other non-metallic, non-electrolytic conductors, (cl915). With the electron at the root of all electrical properties a new class of materials was defined, the halbleiter, but at the same time the shortcomings of the 'electron gas' theory were highlighted. By 1919 work on polycrystalline aelenium and naturally occurring materials had extended electrical knowledge; but Pohl's work on single crystals and the development of the quantum theory paved the way for the modern understanding; an understanding whose theoretical beginning was Wilson's theory of 1931.