Abstract:
An electrically programmable and electrically erasable MOS memory device having a floating gate which is separated from the semiconductor substrate by a thin oxide layer, the memory device also having an impurity implant in the substrate which extends under an edge of the floating gate beneath the thin oxide layer. In one embodiment the thin oxide layer underlies the entire floating gate while in another embodiment only a portion of a small thin side window extends under the floating gate's edge. Also disclosed is a fabrication process in which the one embodiment is formed by first forming the floating gate over the thin oxide layer and then implanting the impurity near an edge of the floating gate. Later steps with heating cause the implanted impurity to diffuse under the floating gate edge. An alternative process first forms a window in the gate oxide layer and implants the impurity through the window. The window is filled with a thin oxide layer and the floating gate is formed so that its edge lies over a portion of the window. Control gates, sources and drains are formed last.
Abstract:
An electrically programmable and electrically erasable MOS memory device having a floating gate which is separated from the semiconductor substrate by a thin oxide layer, the memory device also having an impurity implant in the substrate which extends under an edge of the floating gate beneath the thin oxide layer. In one embodiment the thin oxide layer underlies the entire floating gate while in another embodiment only a portion of a small thin side window extends under the floating gate's edge. Also disclosed is a fabrication process in which the one embodiment is formed by first forming the floating gate over the thin oxide layer and then implanting the impurity near an edge of the floating gate. Later steps with heating cause the implanted impurity to diffuse under the floating gate edge. An alternative process first forms a window in the gate oxide layer and implants the impurity through the window. The window is filled with a thin oxide layer and the floating gate is formed so that its edge lies over a portion of the window. Control gates, sources and drains are formed last.
Abstract:
An electrically programmable and electrically erasable MOS memory device having a floating gate which is separated from the semiconductor substrate by a thin oxide layer, the memory device also having an impurity implant in the substrate which extends under an edge of the floating gate beneath the thin oxide layer. In one embodiment the thin oxide layer underlies the entire floating gate while in another embodiment only a portion of a small thin side window extends under the floating gate's edge. Also disclosed is a fabrication process in which the one embodiment is formed by first forming the floating gate over the thin oxide layer and then implanting the impurity near an edge of the floating gate. Later steps with heating cause the implanted impurity to diffuse under the floating gate edge. An alternative process first forms a window in the gate oxide layer and implants the impurity through the window. The window is filled with a thin oxide layer and the floating gate is formed so that its edge lies over a portion of the window. Control gates, sources and drains are formed last.
Abstract:
A process of fabricating high performance EPROMs in which memory cell devices and high voltage circuit devices are formed in p-type tub regions of high threshold voltage. The tub regions are formed by implanting boron ions in photolithographically defined memory cell and high voltage device areas of a p-type wafer substrate, then subjecting the substrate to a high temperature drive-in. The N-channel isolation field is formed separately and has a lower threshold voltage than the tub regions. The isolation field is formed by implanting boron ions around all device areas, including low voltage device areas, using a nitride mask and a low implantation energy. The wafer is then subjected to an anneal step followed by a field oxidation step. The memory cell and other MOS devices are finally formed in the appropriate defined regions. Since the isolation field's threshold voltage can be adjusted separately from the tub regions, the threshold voltage of the field can be reduced making it possible to reduce the isolation spacing of low voltage devices, reduce parasitic capacitance and increase device speed.
Abstract:
A two-terminal power diode has improved reverse bias breakdown voltage and on resistance includes a semiconductor body having two opposing surfaces and a superjunction structure therebetween, the superjunction structure including a plurality of alternating P and N doped regions aligned generally perpendicular to the two surfaces. The P and N doped regions can be parallel stripes or a mesh with each region being surrounded by doped material of opposite conductivity type. A diode junction associated with one surface can be an anode region with a gate controlled channel region connecting the anode region to the superjunction structure. Alternatively, the diode junction can comprise a metal forming a Schottky junction with the one surface. The superjunction structure is within the cathode and spaced from the anode. The spacing can be varied during device fabrication.
Abstract:
A method for manufacturing a discrete power rectifier device having a VLSI multi-cell design employs a two spacer approach to defining a P/N junction profile having good breakdown voltage characteristics. The method provides highly repeatable device characteristics at reduced cost. The active channel regions of the device are also defined using the same two spacers. The method is a self-aligned process and channel dimensions and doping characteristics may be precisely controlled despite inevitable process variations in spacer formation. Only two masking steps are required, and additional spacers for defining the body region profile can be avoided, reducing processing costs.
Abstract:
A vertical semiconductor rectifier device includes a semiconductor substrate of first conductivity type and having a plurality of gates insulatively formed on a first major surface and a plurality of source/drain regions of the first conductivity type formed in surface regions of second conductivity type in the first major surface adjacent to the gates. A plurality of channels of the second conductivity type each abuts a source/drain region and extends under a gate.
Abstract:
A method of fabricating a semiconductor integrated circuit including a power diode includes providing a semiconductor substrate of first conductivity type, fabricating a integrated circuit such as a CMOS transistor circuit in a first region of the substrate, and fabricating a power diode in a second region in the semiconductor substrate. Dielectric material is formed between the first region and the second regions thereby providing electrical isolation between the integrated circuit in the first region and the power diode in the second region. The power diode can comprise a plurality of MOS source/drain elements and associated gate elements all connected together by one electrode of the diode, and a semiconductor layer in the second region can function as another source/drain of the power diode.
Abstract:
A method of fabricating a semiconductor integrated circuit including a power diode includes providing a semiconductor substrate of first conductivity type, fabricating a integrated circuit such as a CMOS transistor circuit in a first region of the substrate, and fabricating a power diode in a second region in the semiconductor substrate. Dielectric material is formed between the first region and the second regions thereby providing electrical isolation between the integrated circuit in the first region and the power diode in the second region. The power diode can comprise a plurality of MOS source/drain elements and associated gate elements all connected together by one electrode of the diode, and a semiconductor layer in the second region can function as another source/drain of the power diode.
Abstract:
A semiconductor memory device and method for making the same, where a memory cell and high voltage MOS transistor are formed on the same substrate. An insulating layer is formed having a first portion that insulates the control and floating gates of the memory cell from each other, and a second portion that insulates the poly gate from the substrate in the MOS transistor. The insulating layer is formed so that its first portion has a smaller thickness than that of its second portion.