Fumihiko Andoh

A CMOS imager hybridized to an avalanche multiplied film

A highly sensitive solid-state imager has been made by connecting an avalanche multiplier film to a MOS readout circuit through microbump electrodes. Optimization of the vapor-deposition conditions for the indium bump material made it possible for microbumps 5 /spl mu/m in diameter and 5 /spl mu/m in height to be formed into a 2/3-in matrix array of 380 000 pixels. A prototype imager was constructed with a 0.5-/spl mu/m thick avalanche photoconductive film. Clear avalanche multiplication of about ten times was observed at an applied voltage of 75 V. The imager had a good resolution and no recognizable afterimages.

A pixel size shrinkage of amplified MOS imager with two-line mixing

This paper describes a pixel size shrinkage of an amplified MOS image sensor (AMI). We have developed a new circuit technique to achieve the reduction of a pixel size while realizing vertical two-line mixing and high sensitivity. A 1/4-in format 250-k pixel image sensor was developed using a 0.8-/spl mu/m CMOS process. The difference from the conventional CMOS process is an additional layer of ion-implantation process. The power supply voltages of this imager are 4 and 6 V. The dynamic range of 75 dB, the sensitivity of 1.8 /spl mu/A/Ix, and the smear noise of less than -120 dB have been attained for the pixel size of 7.2 (H)/spl times/5.6 (V) /spl mu/m/sup 2/. Although the measured fixed pattern noise ratio (FPN) of this imager is -55 dB, analysis with a test chip shows that FPN can be improved by 2 dB by adopting a suitable gate length for amplifier and resetting MOSFET, respectively.

Analyses of Noise in a Highly Sensitive Image Device

A dipole charge-discharge (i.e., DCD) model was proposed to explain the excess noise factor (i.e., θ) of an amorphous selenium APD target below unity beyond the theoretical limit of shot noise. If the dipole distribution between stored holes and electrons is proper, the electric field in the target is formed between storage periods without interfering with the reading-writing mechanism for the signal, and can control the number of stored carriers so that θ becomes smaller than unity. The theoretical relationship between the signal (i.e., Is) and was obtained and could explain the experimental value well. The theoretical photoelectric conversion characteristics were obtained, and agreed with the experimental values obtained at a high illumination intensity. The magnitude of the negative lag was calculated and was in good agreement with the experimental value.

Studies of multiplication process of a novel image intensifier of an amplified metal-oxide-semiconductor imager overlaid with electron-bombarded amorphous silicon

We studied the multiplication process of a novel image intensifier (II) of an amplified metal-oxide-semiconductor (MOS) imager (AMI) overlaid with electron-bombarded amorphous silicon (a-Si), by Monte Carlo (MC) simulation. The electron bombardment gains (EB-gains) of MC simulation for acceleration voltages between 2 and 10 kV coincide well with the measured values. The threshold voltage of 2 kV is well explained in terms of Bethes electron beam energy losses of the Al and Si/sub x/N/sub 1-x/ layers. The penetration depth of an electron beam of 10 kV is 0.83 /spl mu/m and supports an experimentally safe optimal target thickness (1.2-1.3 /spl mu/m). The standard deviation of lateral spread is 2198 /spl Aring/. The theoretical excess noise coefficient is 1.2 between 7 and 1.3 kV, which coincides exactly with the measured value.

The noise reduction effect of the amplified metal oxide semiconductor imager

It is predicted that the amplified metal oxide semiconductor imager (AMI) exhibits carrier movements between adjacent pixels. Consequently, it is considered that a spurious resolution reduction effect or a random noise reduction effect occurs. The authors have clarified these phenomena theoretically and have measured the sensitivity characteristic between pixels and the random noise using devices. As a result, carrier movements below 5% between adjacent pixels were clearly observed, and it was confirmed that 0.89-times random noise reduction already occurred at an intensity of incidence of 1/1000 of saturation, and 0.79-times reduction at intensity of 1/100 of saturation.

Analysis of the correlative characteristics between adjacent pixels of amplified metal oxide semiconductor imager by a new-type least squares method

The magnitude and distribution of spurious response of the pattern of a circular zone plate of the amplified metal oxide semiconductor (MOS) imager are different from those for charge-coupled devices (CCD). When using the trial-and-error or the least squares method, these features could not be explained by any special shape of the sensing aperture or the sensing distribution, but were explained only by the correlation of accumulated charges between adjacent pixels. This correlation causes less spurious resolution and improves the characteristics of the S/N. (1) The shape of the aperture can be shown by a Gaussian distribution in which the standard deviations in the horizontal and vertical directions are 0.266, and 0.231, respectively. (2) The correlative coefficients in the longitudinal, lateral and diagonal directions each stood at 0.0175. The reliability factor R is 0.6%. (3) The measured signal leakage in adjacent pixels by the microoptical spot is 0.02, which fits the calculated value well. (4) The calculated coefficient of noise reduction is 0.65, which agrees with the obtained value of 0.621.