Abstract: The invention relates to a charge-coupled device. Such devices comprise at least one insulated conducting gate (3) connecting two semiconductor zones. According to the invention, each insulated conducting gate (3) has a width progressively increasing from the first semiconductor zone (1) to the second semiconductor zone (2). The width of each gate (3) is sufficiently narrow for the potential well created by the application of a voltage V to the gate to have a depth increasing progressively from the first zone (1) to the second zone (2), thus enabling the charges to be driven away. The invention applies to any type of charge-coupled device and particularly to photodiodes.

Abstract: A protection circuit against the electrostatic discharges that could appear at the terminals of a circuit, wherein said protection circuit comprises a first transistor made in a well whose potential is a floating potential and enabling the value of the discharge voltage to be limited to a value equal to minus the value of the threshold voltage of said first transistor and a second transistor made in a well whose potential is a floating potential and enabling the value of the discharge voltage to be limited to a value equal to the value of the threshold voltage of said second transistor. The disclosure can be applied to MOS technology integrated circuits.

Abstract: The disclosed photosensitive matrix comprises, in a standard way, a P type semiconductor substrate, an N type channel layer separated by narrow insulating zones into a plurality of columns and, on a thin layer of insulating oxide placed on the channel layer, a network of transfer grids extending perpendicularly to the insulating zones, dividing the columns into a large number of "pixels". According to the invention, the matrix has, between the substrate and the channel layer, a weakly doped P type base layer, in which are buried anti-blooming diodes consisting of a narrow, strong doped N type drain extending in a direction parallel to the insulation zones. Beneath the drain, there is a strongly doped, P type protective screen. The arrangement gives an optical aperture of the matrix close to unity and a spectral response that is improved towards the red side of the spectrum owing to the thickness of the base layer.

Abstract: The disclosed photosensitive matrix comprises, in a standard way, a P type semiconductor substrate, an N type channel layer separated by narrow insulating zones into a plurality of columns and, on a thin layer of insulating oxide placed on the channel layer, a network of transfer gates extending perpendicularly to the insulating zones, dividing the columns into a large number of "pixels". According to the invention, the matrix has, between the substrate and the channel layer, a weakly doped P type base layer, in which are buried anti-blooming diodes consisting of a narrow, strongly doped N type drain extending in a direction parallel to the insulation zones. Beneath the drain, there is a strongly doped, P type protective screen. The arrangement gives an optical aperture of the matrix close to unity and a spectral response that is improved towards the red side of the spectrum owing to the thickness of the base layer.

Abstract: A charge transfer memory and its fabrication method are disclosed. The memory has charge transfer shift registers, with four phases and two level of electrodes, and a reading register with two phases and three levels of electrodes. At one end of each shift register, there is a final electrode contiguous with a reading storage electrode of the reading register, which is itself contiguous to a reading transfer electrode. These electrodes are made in a layer, with a second type of doping, of a semiconductor substrate with a first type of doping. Zones with a third type of doping are made facing the transfer electrodes of the reading register. According to the invention, facing the final electrode of each shift register, a zone with a fourth type of doping is made. This zone with a fourth type of doping prevents charges flowing in the reading register from returning to a shift register.

Abstract: Disclosed are a charge-coupled device with lowering of output potential as well as a method for the fabrication of this device. In a known way, the device comprises, upstream on a semiconducting substrate with a first type of doping (P), a semiconducting layer with a second type of doping (N) and an insulating layer covering the former layer. Pairs of electrodes are formed on the insulating layer. Each pair has a transfer electrode and a storage electrode. Zones with a third type of doping N.sup.+) are made in the layer of a second type (N). A layer with a third type of doping (N.sup.-) is made downstream, in the layer with a second type of doping, and, downstream, there is formed at least one other pair of additional transfer and storage electrodes. A zone with a fourth type of doping (N.sup.--) is made beneath the additional transfer electrode in the layer with a third type of doping (N.sup.-).

Abstract: A new method of manufacturing a two-phase charge-transfer device operating by the charge-coupled technique, in which the asymmetry means associated with each group of electrodes are constituted by impurity barriers implanted in the semiconductor substrate and are automatically positioned in relation to the electrodes, in particular by carrying out implantation operations through masks constituted by films of silicon nitride and by the electrodes themselves, the films of silicon nitride subsequently being completely removed.This method makes it possible to create linear charge-transfer registers and photo-sensitive matrices.