Transfer inefficiency

Another potential image sensor is the CCD discussed in Chapter 9 by Matsumura. The major problem in using a-Si H for CCD applications is the high density of traps or localized states in the band gap. A theoretical analysis assuming an exponential distribution of gap states permits us to predict the transfer characteristics (residual electron density as a function of time). With state-of-the-art material it should be possible to make usable image sensors. Early experimental results with novel test structures have yielded transfer inefficiencies of less than 1% at clock frequencies between 1 and 200 kHz. 

They also fabricated three-phase-clock a-Si H CCDs with coplanar transfer electrodes. The n+-a-Si H resistive layer is set up on and between the electrodes to produce a uniform electric field in the charge-transfer direction. This results in 0.7% transfer inefficiency, at as high as 200-kHz clock frequency (Kishida et al., 1983.)... 

Figure 3 shows the analytical results obtained from (15) and (20). For reference purposes, numerical results are shown by dotted curves. It can be seen that even the coarsest approximation [i.e., (20)] agrees with the numerical results in the region n t) N0, which is a necessary condition for CCD operation. The important feature of (20) is that ln( ,) forms a straight line. The slope of the line is determined by a function of a (i.e., Ta/T) and is independent of other material constants or of the initial density N0 of signal electrons. The transfer inefficiency e of the CCDs is given by... 

Fig. 8. Input gate voltage (upper trace) and output current (lower trace) waveforms of a-Si H CCD with 24 transfer gates with 10-//m pitch. Clock frequency was 100 kHz transfer inefficiency was less than 0.7%.

Fig. 9. Transfer inefficiency as a function ofclock frequency for the improved a-Si H CCD.

When analyzing the operation of a-Si H devices, the localized states must be taken into account. In this chapter the transfer inefficiency of a-Si H CCDs is evaluated numerically and analytically on the basis of the assumption that the localized states in a-Si are distributed exponentially in energy. We have clarified that ln(e) versus ln(/) is linear and that its coefficient is determined by Ta and T. This feature agreed with our experimental results. And the rate equation and the conservation equation used in this analysis can be applied to a-Si H FETs. 

However, the performance is adversely affected by fouling, pore wetting, mass transfer inefficiency due to entrapment of air within the membrane pores, poor heat transfer, heat losses and h h energy consumption [47]. Under the best case scenario the energy consumption is 15 kW h/m , which is very high vis-a-vis seawater RO desalination [48],... 

Fig. 6.5. Modulation transfer function (MTF) as a function of Ne where N is the total number of elementary transfers and E is the transfer inefficiency per elementary transfer. The curves are drawn for different spatial frequencies [/ is the Nyquist frequency (JJ2) ] and different array con-iigurations. The solid curves are for a staring array and the dashed curves for an array which uses series-parallel scan with TDl...

The use of a fat zero will help transfer efficiency however the electrical introduction of the fat zero can introduce an additional source of temporal and pattern noise depending upon the input circuit. With infrared focal planes the fat zero is often automatically provided by the uniform background illumination. The low contrast situation thus automatically reduces the influence of surface interface states on transfer inefficiency. The transfer efficiency of buried channel CCDs is not influenced by trapping at interface states. Trapping at bulk defects plays the same role although the density of bulk traps is usually smaller than typical interface state densities [6.3]. 

When transfer inefficiency is limited by free charge transfer processes, a shot noise point of view can be used [6.14] such that... 

---