Theoretical Basis of Current Instability in Transistor Structures
From the physical point of view, a semiconductor device during operation presents itself as an open, distributed, and dissipative system. This means permanent exchange of energy and matter between the transistor and the ambient space. Therefore, the transistor’s state at any moment of time can be described by a number of the distributed physical parameters: the temperature T(x,y,z), electric field E(x,y,z), electron-hole current density j(x,y,z), and other in general distributed parameters. Distribution of these parameters depends on time, due to applied external conditions, and on the internal processes in the transistor itself. Under electrical load the transistor is a nonequilibrium system in principle. The dynamic equilibrium of the semiconductor device under operation is far from thermodynamic equilibrium. This fact results in the possibility of formation of rather complex multiple thermoelectrical instabilities in the device. In particular, the transition of the semiconductor device from one state to another may become accompanied by a strong current redistribution or filamentation [15-17].
In spite of a wide variety of observed current filamentation scenarios and mechanisms to some extent all of them are based on the same physical principles. Theoretical analysis of transistors as a distributed nonlinear system was first completed in [17-19]. It has been particularly shown that the phenomenon of S-shape I-V characteristic formation in semiconductor structures presents itself as a particular case of the fundamental behavior of a nonlinear dynamic system in non-equilibrium conditions . This chapter focuses only on general principles of current filamentation phenomena in phenomenological semiconductor structures followed by a discussion of a number of general problems for different types of semiconductor devices, conductivity modulation, and spatial current instability in the following chapters.