One reason why semiconductors are so useful for device applications is that their electrical properties can be modified significantly by the incorporation of small amounts of impurities or other kinds of defects. However, while one type of defect can make a semiconductor useful for fabricating a device, another type can have undesirable effects which render the device useless. The quantity of defects necessary to change the properties of a semiconductor is often considerably less than one defect atom per million host atoms.

As a result, our ability to control the defects in a semiconducting material often determines whether it can be used in device applications. To control the amount and nature of defects in a material typically involves developing a process for growing a relatively defect-free sample. Then the desired amounts of defects are introduced either during the growth process or after growth. There is extensive literature devoted to the study of defects in semiconductors. It would take more than a whole book to review all the properties of defects, so in this chapter we will limit ourselves to the study of the electronic properties of defects.

We shall begin by classifying the different kinds of defects found in semiconductors. Next we will separate defects into two broad categories. Impurities whose electronic energies can be calculated by means of the "effective mass approximation" are referred to as shallow impurities, while defects whose energies cannot be calculated with that approximation are known as deep centers. One method capable of...