Compound semiconductors impurities

At the same time, it was demonstrated that hydrogen neutralization of dopant impurities also occurs in compound semiconductors. This was first achieved with n-type dopants in GaAs (Chevallier etal., 1985) and then with p-type dopants in GaAs (Johnson et al., 1986b). 

Several practical applications of hydrogen neutralization of impurities in compound semiconductors are described, including waveguiding, the lateral confinement of carriers for injection lasers, and the generation of resistive regions. Intentional hydrogenation has also been used to fine tune the properties of field-effect transistors. Finally, some remaining problems are identified. 

For nonstoichiometric compounds, the general rule is that when there is an excess of cations or a deficiency of anions, the compound is an n-type semiconductor. Conversely, an excess of anions or deficiency of cations creates a / -type semiconductor. There are some compounds that may exhibit either p- or n-type behavior, depending on what kind of ions are in excess. Lead sulfide, PbS, is an example. An excess of Pb + ions creates an n-type semiconductor, whereas an excess of ion creates a /7-type semiconductor. Similarly, many binary oxide ceramics owe their electronic conductivity to deviations from stoichiometric compositions. For example, CU2O is a well-known / -type semiconductor, whereas ZnO with an excess of cations as interstitial atoms is an n-type semiconductor. A partial list of some impurity-controlled compound semiconductors is given in Table 6.9. 

Dietl et al. 2001c). There exists another mechanism by which strain may affect 7c. It is presently well known that the upper limit of the achievable carrier concentration is controlled by pinning of the Fermi level by impurity or defect states in virtually all compound semiconductors. Since the energies of such states in respect to bands vary strongly with the bond length, the hole concentration and thus 7c will depend on strain. 

This process describes the scattering of free carriers by the screened Coulomb potential of charged impurities (dopants) or defects theoretically treated already in 1946 by Conwell [74,75], later by Shockley [10] and Brooks and Herring [76,77]. In 1969, Fistul gave an overview on heavily-doped semiconductors [78]. A comprehensive review of the different theories and a comparison to the experimental data of elemental and compound semiconductors was performed by Chattopadhyay and Queisser in 1980 [79]. For nondegenerate semiconductors the ionized impurity mobility is given by [79] ... 

An understanding of gas-phase and surface chemistry is particularly important to the next generation of MOVPE processes involving selective epitaxy [18] and atomic layer epitaxy (ALE) [19]. In the first process, the compound semiconductor is deposited selectively on substrate areas opened in a suitable masking material (e.g., SiOz). This is achieved by operating under conditions where nucleation occurs only on the substrates. Slight variations in processing environment and the presence of impurities can cause nucleation on the mask and result in loss of selectivity. 

Generally speaking most of the shallow impurity levels which we shall encounter are based on substitution by an impurity atom for one of the host atoms. An atom must also occupy an interstitial site to be a shallow impurity. In fact, interstitial lithium in silicon has been reported to act as a shallow donor level. All of the impurities associated with shallow impurity levels are not always located at the substitutional sites, but a part of the impurities are at interstitial sites. Indeed, about 90% of group-VA elements and boron implanted into Si almost certainly take up substitutional sites i.e., they replace atoms of the host lattice, but the remaining atoms of 10% are at interstitial sites. About 30% of the implanted atoms of group-IIIA elements except boron are located at either a substitutional site or an interstitial site, and the other 40% atoms exist at unspecified sites in Si [3]. The location of the impurity atoms in the semiconductors substitutional, interstitial, or other site, is a matter of considerable concern to us, because the electric property depends on whether they are at the substitutional, interstitial, or other sites. The number of possible impurity configurations is doubled when we consider even substitutional impurities in a compound semiconductor such as ZnO and gallium arsenide instead of an elemental semiconductor such as Si [4],... 

Although it has been found that even isovalent atoms may act as electrically active impurities in compound semiconductors such as GaP, isovalent atoms of group-VIA such as S, Se, and Te replacing the O atoms in ZnO did not produce any energy levels in the band gap. It is also worth noting that the depth of the energy levels of the impurities in the same row in the periodic table, measured from the upper edge of valence band became smaller as a distance between the columns of O atom sixth column and the column of the impurity atoms decreased. 

Impurity addition, however, is not the only doping mechanism. Nonstoichiometry in compound semiconductors such as CdTe (Table 1) also gives rise to n- or p-type behavior depending on whether Cd or Te is in slight excess, respectively. The defect chemistry in these solid chalcogenides controls their conductivity and doping in a complex manner, a discussion of which is beyond the scope of this chapter. Excellent treatises are available on this topic and on the solid-state chemistry of semiconductors in general [16-22]. 

Figure 4-13e shows a cluster tool configuration. Such configurations are common in Si production lines, but not in compound semiconductors. This type of configuration will continue to gain in importance as compound semiconductors, like Si, try to sequence multiple process steps with minimal exposure of the wafer to impure environments. 

The absorber layer in the thin-fihn solar cells considered here is typically a polycrystalhne p-type compound semiconductor with grain sizes of the order of a micrometer. In order to electrodeposit such a thin film directly, a number of criteria have to be fulfilled. The nucleation density must not be too high so that large grains can be grown uniformly, and secondary nucleation should be avoided if possible. At the same time, overlap of the grains should not leave pinholes that lower the shunt resistance of the device. The layer should also be reasonably flat to minimize recombination losses, and its composition should be as ideal as possible, with low impurity concentrations. 

One final important example of a defect relaxation process should be mentioned in this section. This is the recombination of excess electrons and electron holes m elemental and compound semiconductors following optical excitation (i.e. absorption of light). Very small concentrations of impurities can greatly reduce the relaxation time for this process, since the dissolved impurity atoms (e. g. Ni in Ge) can act as recombination centers. 

To define quickly and exactly if a doped or compound semiconductor containing a fraction y of an impurity B, i.e., A B, is -type or p-type, it is possible to use the following practical rule, called the Grimm-Sommerfeld rule. For this purpose, it is important to introduce the dimensionless physical quantity called the average number of valence electrons, denoted n, and which is calculated by given by the following equation ... 

---