Thermoelectric Arrays

doi:10.1117/3.415351.ch3

Book: Uncooled Thermal Imaging Arrays, Systems, and Applications

Author(s): Paul W. Kruse

Published: 2001

https://doi.org/10.1117/3.415351.ch3

Author Affiliations +

Abstract

Although there are three thermoelectric effects observed in metals and semiconductors, only one, the “Seebeck effect,” is employed in uncooled infrared arrays. Thus, the term “thermoelectric array” or “thermoelectric effect” herein refers only to the Seebeck effect. The three thermoelectric effects are manifested in an electrical circuit that includes two dissimilar metals or semiconductors or a metal and semiconductor. These dissimilar metals must be electrically and mechanically joined, such as by soldering, at one or more points. An example is the joining of two wires of materials A and B at two points (see Fig. 3-1). Let wire A be cut and a high impedance voltmeter inserted at that point. If the two junctions are at the same temperature, no voltage will be measured. However, if one junction is warmer than the other, a voltage (referred to as the "Seebeck" voltage after its discoverer) will be detected. For small temperature differences such as are found in infrared arrays, the voltage depends linearly on the temperature difference between the two junctions. The Seebeck effect in semiconductors is more complex than that in metals and in general far exceeds the value in metals. Consider a semiconductor wire to which a metal wire is connected at either end. Assume that the semiconductor is, say, p-type, and is operating in the extrinsic regime. Then, the free hole concentration at any point along the semiconductor wire is determined by the temperature at that point. That concentration is expressed in terms of the Fermi level, which therefore varies along the length of the wire.