The theory of crystal rectifiers
It has long been known that the contact between a metal and a semiconductor has a resistance which varies, in many cases considerably, with the direction of the current. A well-known example is the copper-cuprous oxide rectifier, which consists of a copper plate on which a layer of cuprous oxide has been formed, the oxide being in contact with a lead plate. It is the purpose of this paper to criticize existing theories of this effect and to suggest a new one. In any crystal rectifier we have two contacts to consider; for instance, in the copper-oxide rectifier there is the contact between the copper and the oxide, and that between the oxide and the lead. We may say at once that a condition for rectification is that at least one of these contacts shall have a resistance which is not small compared with the resistance of the oxide layer; for any theory will give, in the limit of small voltages across the contact, equal resistances in both directions; it is necessary that a significant part of the fall in potential in the rectifier should occur at the contact.