Theoretical Results on Charge Transfer

In a-Si H, it is difficult to operate CCDs in the same mode as conventional SCCDs. This is because the ideal inversion at the p-type a-Si interface has not been observed as yet (Sugiura and Matsumura, 1982), and as for H-type a-Si H, whose interface can be inverted, the drift mobility of the signal charge (i.e., hole) is very low. However, the band gap of a-Si H is about 1.7 eV, which is much wider than that of a single-crystal silicon. This indicates that the carrier concentration in undoped a-Si H under thermal equilibrium condition is so low that the accumulation mode can store transient signals sufficiently long for CCD operation, as described later. 

we assume that the localized state density distribution is approximated by 

For the purpose of grasping intuitively the physical phenomena, the simple approximate analysis is more suitable than the cumbersome numerical calculation. Therefore in this section we deal with approximate analysis. 

Before the transfer starts, the energy distribution of electrons takes the form of a Fermi-Dirac distribution function. While the number of electrons is decreasing steadily with time, the distribution of electrons keep the form of a Fermi-Dirac distribution function. This constancy of the distribution is due to the fact that the capture rate of free electrons by the localized states is much faster than the loss of free electrons caused by the transfer when the occupation probability of localized states is not approximately one. Therefore, electrons are considered to be in their quasi-thermal equilibrium condition i.e., the energy distribution of electrons is described by quasi-Fermi energy EF. Then the total density t of electrons captured by the localized states per unit volume can be written as 

A detailed derivation was described elsewhere (Kishida et al., 1983). On the other hand, the free electron density nc in the conduction band is also given 

---