scispace - formally typeset
Search or ask a question

Showing papers on "Metal gate published in 1981"


Book ChapterDOI
John R. Brews1
01 Jan 1981

78 citations


Patent
16 Oct 1981
TL;DR: In this article, a current-controlling MOS transistor is connected between a power source and an MOS circuit, and a control voltage which has a level related to temperature is applied to the gate electrode in order to compensate for current reduction at high temperatures due to the lowering of the mobility of minority carriers.
Abstract: A current-controlling MOS transistor is connected between a power source and an MOS circuit. A control voltage which has a level related to temperature is applied to the gate electrode of the control MOS transistor in order to compensate for current reduction at high temperatures due to the lowering of the mobility of minority carriers. The response time of the MOS circuit is made less dependent on temperature as a result of the current compensation.

68 citations


Patent
23 Nov 1981
TL;DR: In this paper, a self-refreshing non-volatile memory cell with two cross-coupled transistors includes a first floating gate formed between the gate and the channel of the first transistor, and a second floating gate overlying by tunnel oxide a portion of the drain of the second transistor.
Abstract: A self-refreshing non-volatile memory cell having two cross-coupled transistors includes a first floating gate formed between the gate and the channel of said first transistor, said first floating gate overlying by means of a tunnel oxide a portion of the drain of said second transistor and a second floating gate formed between the gate and channel of said second transistor, a portion of said second floating gate overlying by tunnel oxide a portion of the drain of the first transistor. Disturbances in the supply voltage and the gate voltage of the device normally enhance rather than degrade the state of data stored in the cell, thereby providing an extremely long storage time for the cell. The cell is capable of operating simultaneously in a volatile and a non-volatile state.

66 citations


Patent
09 Oct 1981
TL;DR: In this article, a novel metaloxide-semiconductor (MOS) field effect transistor has been proposed with enhanced oxide thickness at the edge of the gate electrode and having metal silicide regions in the gate and source and drain areas.
Abstract: A novel metal-oxide-semiconductor (MOS) field effect transistor having enhanced oxide thickness at the edge of the gate electrode and having metal silicide regions in the gate electrode and source and drain areas. The enhanced oxide thickness improves interconnect-to-interconnect breakdown voltage in multilevel interconnect devices as well as minimizing gate overlap of source and drain. The metal silicide regions reduce series resistance and improve device speed and packing density.

56 citations


Patent
28 Jan 1981
TL;DR: In this paper, the output of a memory cell is fed to several flip-flops, which are synchronized with the input ramp, thereby setting the flipflops in accordance with the threshold or state of the cell.
Abstract: A read only memory in which the memory cells are single metal gate or silicon gate field effect transistors. Each FET has one of several different thresholds or states. The size or area of the FET gates at the surface of the semiconductor chip are substantially the same regardless of the cells threshold or state. The input to the gates is a ramp, and the cells are rendered conducting by the amplitude of the ramp at a given instant. The output of a cell is fed to several flip-flops, which are synchronized with the input ramp, thereby setting the flip-flops in accordance with the threshold or state of the cell. An encoder converts the output from the flip-flops to a bit binary signal. This permits a very high density ROM, e.g. 128K on a single chip. In another embodiment the input to the gates is a step, and the cells are all rendered conducting simultaneously. The amount of current drawn by each gate however, depends upon the doping in the gate region. The amplitude of the current being drawn by a selected cell is compared with the reference , this in turn is decoded to indicate the state of that cell.

53 citations


Patent
29 Jun 1981
TL;DR: In this paper, a process for forming MOS circuits which include underlying polysilicon members such as gate members covered with metal is described, and low temperature "rear end" steps are used to prevent deterioration of the underlying metal.
Abstract: A process is described for forming MOS circuits which include underlying polysilicon members such as gate members covered with metal. In one embodiment, a self-aligning tungsten process is used to cover the polysilicon members. Low temperature "rear end" steps are used to prevent deterioration of the underlying metal. For example, a plasma nitride protective layer is used to cover the metal. The polysilicon/metal members provide reduced resistance and increase the speed of the resultant MOS circuits.

49 citations


Patent
02 Oct 1981
TL;DR: In this paper, a method of producing a monolithically integrated two-transistor memory cell including a silicon crystal for accommodating the memory cell, a first MOS field effect transistor having a current-carrying channel and both a control gate and a floating gate disposed between the control and surface of the crystal, and an erase area for the floating gate was proposed.
Abstract: A method of producing a monolithically integrated two-transistor memory cell, including a silicon crystal for accommodating the memory cell, a first MOS field effect transistor having a current-carrying channel and both a control gate and a floating gate disposed between the control gate and surface of the crystal, a second MOS field effect transistor having a current-carrying channel and a control gate, an SiO 2 film supporting the gates, a doped polycrystalline silicon layer deposited on the SiO 2 film, the control gates and the floating gate being formed from the doped polycrystalline silicon layer, and an erase area for the floating gate, the improvement which includes covering a part of the silicon crystal intended for the memory cell with an SiO 2 film, forming a part of the gate oxide of the first MOS field effect transistor, forming a window through the SiO 2 film at a location intended for the erase area, re-oxidizing the exposed area of the surface of the crystal in the window and increasing the remaining areas of the SiO 2 film, depositing a first doped polycrystalline silicon layer forming a base of the floating gate, covering the polycrystalline silicon layer with another SiO 2 film, depositing a second doped polycrystalline silicon layer, and forming the control gate of the first MOS field effect transistor from the second doped polycrystalline silicon layer and producing the source and drain zone of the two MOS field effect transistors.

39 citations


Journal ArticleDOI
S. Tanaka1, M. Ishikawa
TL;DR: In this article, a physical writing model for an n-channel floating gate ionization-injection MOS (FIMOS) is described, and the strength of accelerating field for the injection is calculated, taking the two-dimensional components near the drain junction into account.
Abstract: A physical writing model for an n-channel floating gate ionization-injection MOS (FIMOS) is described. Strength of accelerating field for the injection is calculated, taking the two-dimensional components near the drain junction into account. A new expression for the channel hot electron injection efficiency is also derived, using Baraff's analytic electron distribution function. The gate current, which shows a complicated dependence on the floating gate and drain voltages, has been reasonably formulated by the model. Integrating the gate current, time behavior of the threshold shift is predicted. The result is in reasonable agreement with experimental results.

37 citations


Patent
03 Jun 1981
TL;DR: In this paper, a storage cell of a nonvolatile electrically alterable MOS memory (EAROM) comprises a p-type silicon substrate with n-doped drain and source areas interlinked by an n-channel which is partly overlain by a floating gate extending over part of the drain area.
Abstract: A storage cell of a nonvolatile electrically alterable MOS memory (EAROM) comprises a p-type silicon substrate with n-doped drain and source areas interlinked by an n-channel which is partly overlain by a floating gate extending over part of the drain area. An accessible gate overlaps the floating gate and has an extension overlying a gap between the latter gate and the source area to act as a common control electrode for two series IGFETs defined by the source and gate areas, namely a main or storage transistor and an ancillary or switching transistor. The capacitance of the floating gate relative to the drain area accounts for about half the overall capacitance of that gate relative to the entire semiconductor structure.

36 citations


Patent
24 Dec 1981
TL;DR: In this article, a self-aligned gate process using anisotropic etch to self-align the gate and source/drain is described. But the vertical etch is not used in this paper.
Abstract: A MESFET is fabricated using a self-aligned gate process. This process uses a vertical (anisotropic) etch to self-align the gate and source/drain. The vertical etch, in conjunction with a two-level insulator, creates a barrier between the gate and source/drain, so that when metal is deposited and reacted, and any excess removed, the gate is self-aligned with the source/drain, and contacts to the source/drain and gate are well isolated. The alignment obtained by this process is advantageous in that series channel resistance is reduced, and a more compact structure is attained for improvement in packing density.

34 citations


Patent
19 Jan 1981
TL;DR: In this paper, a silicide pattern is formed on polysilicon by a lift-off technique, and the patterned silicide is utilized as a mask for anisotropic etching of the underlying poly-silicon.
Abstract: It is known to deposit a refractory metal silicide on a polysilicon gate layer to form a low-resistivity composite structure. For VLSI MOS devices, very-high-resolution patterning of the composite structure is required. In accordance with this invention, a silicide pattern is formed on polysilicon by a lift-off technique. In turn, the patterned silicide is utilized as a mask for anisotropic etching of the underlying polysilicon. High-conductivity composite silicide-on-polysilicon gate structures for VLSI MOS devices are thereby achieved.

Patent
12 Jun 1981
TL;DR: In this article, the depletion mode MOS transistor's gate receives a control signal only when power is on, and the gate is nonconductive when the power is off, but it is conductive when power consumption is low.
Abstract: A protected MOS transistor circuit includes an input MOS transistor and a depletion mode MOS transistor having a drain-source current path connected between ground and the gate of the input MOS transistor of obviating rupture of the gate oxide of the input MOS transistor when power is off. The depletion mode MOS transistor's gate receives a control signal only when power is on which renders the depletion mode MOS transistor nonconductive when power is on. The depletion mode MOS transistor is conductive when power is off.

Patent
16 Dec 1981
TL;DR: In this article, a method for fabricating MOS devices of the type found in very large scale integrated circuits is described, where various gate oxides and insulating layers are fabricated independently of each other in order to improve isolation between gate electrodes and interconnects, and independently controllable operating characteristics for multiple gate electrode structures.
Abstract: A method is described for fabricating MOS devices of the type found in very large scale integrated circuits. According to the method described herein, various gate oxides and insulating layers are fabricated independently of each other in order to independently tailor their thicknesses and thereby provide improved isolation between gate electrodes and interconnects, and independently controllable operating characteristics for multiple gate electrode structures. The fabrication of a dynamic RAM memory cell, an overlapping gate CCD device and a self-aligned MNOS transistor cell are described using the disclosed method.

Patent
12 Nov 1981
TL;DR: In this article, a method for fabricating a gate-source structure for a recessed-gate static induction transistor is described by use of doped polysilicon to fill the recessed gate grooves after the gate groove have been etched and diffused.
Abstract: A method for fabricating a gate-source structure for a recessed-gate static induction transistor. The method is characterized by use of doped polysilicon to fill the recessed gate grooves after the gate grooves have been etched and diffused. The gate grooves have depth greater than width and therefore the surface of the polysilicon layer deposit is substantially planar. The planar surface allows photolithographic techniques to be used for formation of gate contact regions and for depositing of metal gate and source electrodes.

Journal ArticleDOI
TL;DR: In this article, a new technique for the electrical monitoring of polymerization reactions such as resin cure is described, which is based on the charge-flow transistor, which resembles a conventional metal-oxide-semiconductor field effect transistor (MOSFET), but with a portion of the metal gate replaced by the resin under study.
Abstract: A new technique for the electrical monitoring of polymerization reactions such as resin cure is described. The technique is based on the charge-flow transistor, which resembles a conventional metal-oxide-semiconductor field-effect transistor (MOSFET), but with a portion of the metal gate replaced by the resin under study. Electrical signals obtained from several resins undergoing cure are presented, along with an electrical circuit model that can account for the principal features of these signals. The dramatic change in signal shape during cure can be related to corresponding changes in both the real and imaginary parts of the dielectric constant.

Patent
29 Jan 1981
TL;DR: In this article, a zero threshold mode transistor is connected between an enhancement-mode MOS driver transistor and a depletion-mode load transistor to provide a power-down function for the MOS transistor circuit.
Abstract: An MOS transistor circuit contains at least one "zero" threshold mode transistor to provide a power-down function for the circuit. The "zero" threshold mode transistor is connected between an enhancement-mode MOS driver transistor and a depletion-mode MOS load transistor.

Book
01 Jan 1981

Patent
09 Sep 1981
TL;DR: In this article, a high power VMOS semiconductor device and fabrication method was described, which uses a doped polysilicon gate electrode in the V groove and an overlying metal electrode located over an insulation layer protecting the DPG.
Abstract: This disclosure relates to a high power VMOS semiconductor device and fabrication method therefor This VMOS semiconductor device uses a doped polysilicon gate electrode in the V groove and an overlying metal electrode located over an insulation layer protecting the doped polysilicon gate electrode This overlying metal electrode layer covers substantially the entire surface area (except for a small area where electrical contact is made to the doped polysilicon gate electrode) of one surface of the device Another embodiment discloses the use of a self-aligned metal contact to the source or drain region of the VMOS device between adjacent V groov

Journal ArticleDOI
TL;DR: Metal gate MOS-transistors with channel lengths down to approximately 0.5 μm and gate oxide thicknesses of 19 nm and 34 nm with good agreement between measured and calculated values was found regarding the geometry dependence.
Abstract: Metal gate MOS-transistors with channel lengths down to approximately 0.5 μm and with gate oxide thicknesses of 19 nm and 34 nm have been fabricated and evaluated. For devices shorter than 1 μm we have found significant short channel effects on threshold voltage, transconductance and subthreshold current. The experimental results have been compared with computer model calculations. A good agreement between measured and calculated values was found regarding the geometry dependence.

Journal ArticleDOI
TL;DR: In this article, measured data of two different types of dynamic amplifier which have been integrated in a standard CMOS metal gate technology were presented with a power dissipation of 58 μW and 500 μW at clock frequencies of 10 kHz and 100 kHz.
Abstract: The letter presents measured data of two different types of dynamic amplifier which have been integrated in a standard CMOS metal gate technology. For a supply voltage of ±5 V, gains of 70 dB were obtained with a power dissipation of 58 μW and 500 μW at clock frequencies of 10 kHz and 100 kHz, respectively. The dynamic range exceeds 100 dB.

Patent
19 Jan 1981
TL;DR: In this article, an MOS read only memory or ROM is formed by a process compatible with standard P or N channel metal gate manufacturing methods, and the ROM is programmed at a late stage of the process after the metal level of contacts and interconnections has been deposited and patterned.
Abstract: An MOS read only memory or ROM is formed by a process compatible with standard P or N channel metal gate manufacturing methods. The ROM is programmed at a late stage of the process after the metal level of contacts and interconnections has been deposited and patterned. Address lines and gates are polysilicon with an overlying patterned metal layer and output and ground lines are defined by elongated heavily doped regions. Thin gate oxide is formed for every gate position, rather than for only the selected gates as in the prior standard programming method. Each potential MOS transistor in the array is programmed to be a logic "1" or "0" by ion implanting through the polysilicon gates where metal has been removed, using photoresist as a mask.

Patent
06 Oct 1981
TL;DR: In this article, a JFET with the top gate isolated from the bottom gate by an annulus source region and thin channel region and a top gate ohmic contact region is described.
Abstract: A JFET having the top gate isolated from the bottom gate by an annulus source region and thin channel region and a top gate ohmic contact region isolated from the bottom gate by a deep isolation region. The isolation region and the top gate contact region are exterior the active channel region.

Patent
14 Apr 1981
TL;DR: In this article, a negative resistance device utilizing a substrate bias effect is comprised of two MOS transistors of n-channel and p-channel type, connected at the sources and the gates.
Abstract: A negative resistance device utilizing a substrate bias effect is comprised of two MOS transistors of n-channel type and p-channel type. The two transistors are connected at the sources and the gates. The drain of the n-channel MOS transistor is connected to the substrate of the p-channel MOS transistor. The drain of the p-channel MOS transistor is connected to the substrate of the n-channel MOS transistor.

Patent
17 Jun 1981
TL;DR: In this article, a depletion mode MOS transistor (TrP) with a drain-source current path connected between ground and the gate of the TrP was used to prevent breakdown of the gate oxide of the trP when power is off as a consequence of stray voltages.
Abstract: A protected MOS transistor circuit includes an input (or output) MOS transistor (1) and a protective circuit (10) including a depletion mode MOS transistor (TrP) having a drain-source current path connected between ground and the gate of the MOS transistor and arranged to prevent breakdown of the gate oxide of the protected MOS transistor when power is off as a consequence of stray voltages. When power is on, the depletion mode MOS transistor's gate receives a control signal which renders the depletion mode MOS transistor non-conductive but, when power is off, the depletion mode transistor is conductive so as to ground currents caused by the stray voltages.

Patent
11 Dec 1981
TL;DR: In this article, the gate electrodes (28, 38) are formed from a polysilicon layer (128), which is then covered by a mask including a silicon oxide layer (130) and a silicon nitride layer (132).
Abstract: In a process for forming a CMOS integrated circuit (10) having polysilicon gate electrodes (28, 38) the polysilicon gate electrodes (28, 38) are simultaneously doped with impurities of a single conductivity type, independently of the semiconductor substrate (18). The invention enables the avoidance of the penetration problem which arises when boron is utilized to dope polysilicon. After forming the gate electrodes (28, 38) from a polysilicon layer (128), they are covered by a mask including a silicon oxide layer (130) and a silicon nitride layer (132). Then sources (22, 34) and drains (24, 32) of the n-channel and p-channel transistors (12, 14) are formed and an implantation or diffusion barrier (148) is grown over the sources (22, 34) and drains (24, 32). The mask (130, 132) is removed over the gate electrodes (28, 38) which are then doped with an n-type impurity. Polysilicon resistors (50) may be formed by initially doping the polysilicon layer (128) to a low level of conductivity and protecting the resistor areas (50) by a further mask (138), which may be of polysilicon or silicon nitride, during subsequent doping to a high level of conductivity.

Patent
Sheng T. Hsu1
16 Jul 1981
TL;DR: In this article, a nonvolatile memory structure of the floating gate type is described, where current carriers are injected onto a floating gate from the control gate as distinguished from the prior art which injects current carriers into the floating-gate from the substrate.
Abstract: A non-volatile memory structure of the floating gate type is described wherein current carriers are injected onto the floating gate from the control gate as distinguished from the prior art which injects current carriers into the floating gate from the substrate This invention teaches that by tailoring the capacitance between the control gate and the floating gate and the capacitance between the floating gate and the substrate different field intensities are created in the region between the floating gate and the control gate and in the region between the substrate and the floating gate When the field intensity across the capacitor formed between the control gate and the floating gate is greater than the field intensity across the capacitor formed between the floating gate and the substrate, current carriers will be injected onto the floating gate from the control gate

Patent
Yoshibumi Ando1, Sakamoto Takashi1, Kanji Yoh1, Hisahiro Moriuchi1, Takei Sumiaki1 
06 Aug 1981
TL;DR: In this article, a resistor is connected between the gate of the output MOS transistor and a drive circuit for driving that output transistor to prevent the gate insulating film from being broken down.
Abstract: An abnormal surge voltage such as frictional static electricity is often applied to the external terminals of a MOSIC. In the past, the output MOS transistor in the MOSIC during normal handling of the device frequently will have its gate insulating film broken down by the application of such an abnormal surge voltage to the drain thereof. In order to prevent the gate insulating film from being broken down, in this manner a resistor is connected between the gate of the output MOS transistor and a drive circuit for driving that output MOS transistor. This construction using a resistor is superior to the construction in which the voltage to be applied to the drain of the output MOS transistor is clamped by the use of suitable clamp means only because, with the resistor arrangement, the output characteristics of the MOSIC are not restricted.

Patent
05 Mar 1981
TL;DR: In this article, a method of electrically isolating a plurality of semiconductor integrated circuit components and for forming gate elements for silicon gate transistors is disclosed whereby extremely narrow line widths can be formed which heretofore have been unattainable by practicing conventional photolithography.
Abstract: A method of electrically isolating a plurality of semiconductor integrated circuit components and for forming gate elements for silicon gate transistors is disclosed whereby extremely narrow line widths can be formed which heretofore have been unattainable by practicing conventional photolithography.

Journal ArticleDOI
TL;DR: In this paper, a new Si MOSFET was proposed, featuring an insulated gate structure, channel doping, and finite spacing between gate and source and between gate between drain and drain, and two-dimensional numerical analysis showed that punchthrough is suppressed and that minimum gate length, limited bypunchthrough or V T shift, is extended into the submicrometer range.
Abstract: A new device structure is proposed for Si MOSFET, featuring an insulated gate structure, channel doping, and finite spacing between gate and source and between gate and drain. Two-dimensional numerical analysis shows that punchthrough is suppressed and that minimum gate length, limited bypunchthrough or V T shift, is extended into the submicrometer range.

Patent
22 Oct 1981
TL;DR: In this paper, a MOS transistor is formed and the gate and underlying gate oxide layer removed to expose the substrate, and a conducting layer through the aperture formed makes contact to the substrate.
Abstract: Method of forming buried contact in recessed gate MOSTEK technology is provided. When the buried contact is desired, a MOS transistor is formed and the gate and underlying gate oxide layer removed to expose the substrate. A conducting layer through the aperture formed makes contact to the substrate.