Defects in amorphous semiconductors

The first question to address is the definition of a defect in an amorphous material. In a crystal any departure from the perfect crystalline lattice is a defect, which could be a point defect, such as a vacancy or interstitial, an extended defect, such as a dislocation or stacking fault, or an impurity. A different definition is required in an amorphous material because there is no perfect lattice. The inevitable disorder of the random network is an integral part of the amorphous material and it is not helpful to think of this as a collection of many defects. By analogy with the crystal one can define a defect as a departure from the ideal amorphous network which is a continuous 

In one respect the defects of an amorphous solid are easier to deal with than those of a crystal. Any small deviation in the local structure of the defect in a crystal results in an identifiably different state, resulting in many possible defect structures. More than 50 point defects are known in crystalline silicon and there is probably an even larger diversity of extended defects. In the amorphous material, small differences in local structure which fall within the disorder of the ideal network cannot be resolved meaningfully. Thus one expects fewer separate classes of defects, but with their energy levels broadened out by the disorder, as illustrated schematically in Fig. 4.1. 

Similar arguments apply to impurity states. Any impurity which is bonded with its optimum valency is expected to form a part of the ideal network and contribute only to the conduction and valence bands. Oxygen, nitrogen, carbon, and germanium all behave in this way, forming alloys with a-Si H. Most of the phosphorus and boron atoms which are added as dopants, are in three-fold coordinated inactive sites 

---

Defect states, in amorphous semiconductors, 22 128-129 Defense applications, for high performance fibers, 13 397-398... 

As in the case of NMR, ESR measurements often provide a detailed, local probe of bonding at paramagnetic sites such as defects, impurities, or band tails in amorphous semiconductors. The important terms in the Hamiltonian for most situations of interest in ESR experiments in a-Si and a-Si H are... 

Figure 9.4 Electron energy states of semiconductors with a large concentration of defects (or amorphous semiconductors) and surface states. energy limit of free mobility in the conduction band, energy limit of free mobility in the valence band, E mobility gap, (a) and (b) gap states, and (c) surface states.

Unsatisfied bonds introduced accidentally during the formation of the amorphous phase might form the analogue to point defects like vacancies and interstitials in a crystal lattice. Such defects might be linked to the relatively few unpaired electrons or acceptor and, perhaps, donor levels found in amorphous semiconductors. 

This book is divided into two parts. Part A deals with hydrogen in amorphous semiconductors. After a review, by several speakers, of the theoretical understanding of electronic and transport properties in these systems, there follow lectures dealing with the distribution of hydrogen in external and internal surfaces and its effect on defect structure. Finally, attention is given to the formation and trapping of molecular hydrogen in voids in the host matrix. 

The way in which a semiconductor material is made has significant implications for NMR spectroscopy in several ways it governs the amount of sample available for analysis, it determines whether the material will be single-crystal or polycrystalline (or even amorphous), and it controls the nature and amounts of dopants, intentional or otherwise, and defects. In categories most relevant to the NMR spectroscopist, the ways in which most semiconductors are made can be classified as follows ... 

---