Abstract: Numerous embodiments for performing tuning of a page or a word of non-volatile memory cells in an analog neural memory are disclosed. High voltage circuits used to generate high voltages applied to terminals of the non-volatile memory cells during the precision tuning process are also disclosed. Programming sequences for the application of the voltages to the terminals to minimize the occurrence of disturbances during tuning are also disclosed.

Abstract: The present subject matter relates to accessing memory units in a memory bank. In an example implementation, a bank select transistor is common to a plurality of memory units in a memory bank. The bank select transistor facilitates accessing a memory unit of the plurality of memory units based on a bank select signal.

Abstract: An apparatus is provided which comprises: a source and a drain with a channel region therebetween, the channel region comprising a semiconductor material, and a gate dielectric layer over at least a portion of the channel region, wherein the gate dielectric layer comprises a first thickness proximate to the source and a second thickness proximate to the drain, wherein the second thickness is greater than the first thickness, and wherein at least a portion of the gate dielectric layer comprises a linearly varying thickness over the channel region. Other embodiments are also disclosed and claimed.

Abstract: A structure of a memory device and a fabrication method thereof are provided. The structure of the memory device includes a tunneling layer disposed on a substrate. A first oxide/nitride/oxide (ONO) layer is disposed on the substrate abutting to the tunneling layer. A floating gate is disposed on the tunneling layer, wherein a side portion of the floating gate is also disposed on the first ONO layer. A second ONO layer is disposed on the floating gate. A control gate is disposed on the second ONO layer. An isolation layer is disposed on first sidewalls of the floating gate and sidewalls of the control gate. An erase gate is disposed on the first ONO layer, wherein the erase gate is isolated from the floating gate and the control gate by the isolation layer.

Abstract: A method includes forming a tunneling dielectric layer on a semiconductor substrate, a first portion of the tunneling dielectric layer is directly above a channel region in the semiconductor substrate and a second portion of the tunneling dielectric layer is directly above source-drain regions located on opposing sides of the channel region, the second portion of the tunneling dielectric layer is thicker than the first portion of the tunneling dielectric layer, forming a floating gate directly above the first portion of the tunneling dielectric layer and the second portion of the tunneling dielectric layer, and forming a control dielectric layer directly above the floating gate.

Abstract: A method includes forming a tunneling dielectric layer on a semiconductor substrate, a first portion of the tunneling dielectric layer is directly above a channel region in the semiconductor substrate and a second portion of the tunneling dielectric layer is directly above source-drain regions located on opposing sides of the channel region, the second portion of the tunneling dielectric layer is thicker than the first portion of the tunneling dielectric layer, forming a floating gate directly above the first portion of the tunneling dielectric layer and the second portion of the tunneling dielectric layer, and forming a control dielectric layer directly above the floating gate.

Abstract: A non-volatile semiconductor memory device is proposed whereby voltage can be more flexibly set in accumulating electric charges into a selected memory cell transistor in comparison with a conventional device. In a non-volatile semiconductor memory device (1), when a selected memory cell transistor (115) is caused to accumulate electric charges, high voltage as writing prevention voltage is applied from a PMOS transistor (9b) while low voltage as writing voltage is applied from an NMOS transistor (15a). Thus, a role of applying voltage to either the selected memory cell transistor (115) or a non-selected memory cell transistor (116) is shared by the PMOS transistor (9b) and the NMOS transistor (15a). Therefore, the gate voltage and the source voltage of the PMOS transistor (9b) and those of the NMOS transistor (15a) can be separately adjusted, and gate-to-substrate voltage thereof can be finally set to be, for instance, 4[V] or etc.

Abstract: A thin-film transistor device includes: a gate electrode above a substrate; a gate insulating film on the gate electrode; a crystalline silicon thin film including a channel region which is provided on the gate insulating film; semiconductor films on at least the channel region; an insulating film made of an organic material which is provided over the channel region and above the semiconductor films; a source electrode over at least an end portion of the insulating film; and a drain electrode over at least the other end portion of the insulating film and facing the source electrode. The semiconductor films include at least a first semiconductor film and a second semiconductor film provided on the first semiconductor film. A relationship ECP<EC1 is satisfied where ECP and EC1 denote energy levels at lower ends of conduction bands of the crystalline silicon thin film and the first semiconductor film, respectively.

Abstract: The present disclosure provides a semiconductor device which includes a semiconductor substrate, a first gate structure disposed over the substrate, the first gate structure including a first gate electrode of a first conductivity type, a second gate structure disposed over the substrate and proximate the first gate structure, the second gate structure including a second gate electrode of a second conductivity type different from the first conductivity type, a first doped region of the first conductivity type disposed in the substrate, the first doped region including a first lightly doped region aligned with a side of the first gate structure, and a second doped region of the second conductivity type disposed in the substrate, the second doped region including a second lightly doped region aligned with a side of the second gate structure.

Abstract: A non-volatile memory is formed on a substrate. The non-volatile memory includes an isolation structure, a floating gate, and a gate dielectric layer. The isolation structure is disposed in the substrate to define an active area. The floating gate is disposed on the substrate and crosses over the active area. The gate dielectric layer is disposed between the floating gate and the substrate. The floating gate includes a first region and a second region. An energy band of the second region is lower than an energy band of the first region, so that charges stored in the floating gate are away from an overlap region of the floating gate and the gate dielectric layer.

Abstract: According to one embodiment, a semiconductor memory element includes a semiconductor layer, a tunnel insulator provided on the semiconductor layer, a charge accumulation film provided on the tunnel insulator having a film thickness of 0.9 nm or more and 2.8 nm or less and the charge accumulation film containing cubic HfO2 particles, a block insulator provided on the charge accumulation film, and a control electrode provided on the block insulator.

Abstract: A three terminal non-volatile memory (NVM) cell for a CMOS IC is formed by either a standard CMOS process flow or a slightly modified CMOS process flow. The NVM cell includes read and injection transistors that share a common floating gate. The floating gate includes a portion disposed over the channel region of the read transistor, a portion disposed over the channel region of the injection transistor, and a portion extending into an enlarged drain diffusion area away from the channel regions, whereby the gate-to-drain capacitance is higher than the gate-to-source capacitances. The source/drain of the injection transistor are formed using different LDD implants to achieve faster program/erase. Alternatively, an optional CHE enhancing implant is added to the source/drain of the injection transistor to enhance CHE programming. Both HV LDD and LV LDD implants are introduced together enabling LDD implant merging under the floating gate extension.

Abstract: A stacked non-volatile memory device comprises a plurality of bit line and word line layers stacked on top of each other. The bit line layers comprise a plurality of bit lines that can be formed using advanced processing techniques making fabrication of the device efficient and cost effective. The device can be configured for NAND operation.

Abstract: A semiconductor device that includes multiple logic circuit cells having respective logic circuits formed therein and multiple interconnects connected to the corresponding logic circuit cells. At least one of the interconnects has an opening formed therein so as to have an opening ratio different from one or more of the opening ratios of the remaining interconnects.

Abstract: A semiconductor memory device includes: a semiconductor substrate; a first impurity region; a second impurity region; a channel region; a first gate formed on a main surface on a side of the first impurity region; a second gate formed on the main surface on a side of the second impurity region, with a second insulating film being interposed; and a third insulating film formed on a side surface of the first gate. An interface between the third insulating film and the semiconductor substrate directly under the third insulating film is located above an interface between the second insulating film and the main surface of the semiconductor substrate directly under the second insulating film. The total number of steps can thus be reduced, and lower cost is achieved.

Abstract: The gate electrode of a metal oxide semiconductor field effect transistor (MOSFET) comprises a source side gate electrode and a drain side gate electrode that abut each other near the middle of the channel. In one embodiment, the source side gate electrode comprises a silicon oxide based gate dielectric and the drain side gate electrode comprises a high-k gate dielectric. The source side gate electrode provides high carrier mobility, while the drain side gate electrode provides good short channel effect and reduced gate leakage. In another embodiment, the source gate electrode and drain gate electrode comprises different high-k gate dielectric stacks and different gate conductor materials, wherein the source side gate electrode has a first work function a quarter band gap away from a band gap edge and the drain side gate electrode has a second work function near the band gap edge.

Abstract: A non-volatile memory device implements self-convergence during the normal erase cycle through control of physical aspects, such as thickness, width, area, etc., of the dielectric layers in the gate structure as well as of the overall gate structure. Self-convergence can also be aided during the normal erase cycle by ramping the erase voltage applied to the control gate during the erase cycle.

Abstract: Method and device embodiments are described for fabricating MOSFET transistors in a semiconductor also containing non-volatile floating gate transistors. MOSFET transistor gate dielectric smiling, or bird's beaks, are adjustable by re-oxidation processing. An additional re-oxidation process is performed by opening a poly-silicon layer prior to forming an inter-poly oxide dielectric provided for the floating gate transistors.

Abstract: A semiconductor memory device has: a substrate; a memory cell transistor of a split-gate type formed on the substrate; and a reference transistor formed on the substrate and used for generating a reference current that is used in sensing data stored in the memory cell transistor. The memory cell transistor has a floating gate and a control gate, while the reference transistor is a MIS (Metal Insulator Semiconductor) transistor having a single gate electrode.

Abstract: A semiconductor device manufacturing method includes forming a first insulating film on a semiconductor substrate containing silicon, the first insulating film having a first dielectric constant and constituting a part of a tunnel insulating film, forming a floating gate electrode film on the first insulating film, the floating gate electrode film being formed of a semiconductor film containing silicon, patterning the floating gate electrode film, the first insulating film, and the semiconductor substrate to form a first structure having a first side surface, exposing the first structure to an atmosphere containing an oxidizing agent, oxidizing that part of the floating gate electrode film which corresponds to a boundary between the first insulating film and the floating gate electrode film using the oxidizing agent, to form a second insulating film having a second dielectric constant smaller than the first dielectric constant and constituting a part of the tunnel insulating film.

Abstract: For enhancing the high performance of a non-volatile semiconductor memory device having an MONOS type transistor, a non-volatile semiconductor memory device is provided with MONOS type transistors having improved performance in which the memory cell of an MONOS non-volatile memory comprises a control transistor and a memory transistor. A control gate of the control transistor comprises an n-type polycrystal silicon film and is formed over a gate insulative film comprising a silicon oxide film. A memory gate of the memory transistor comprises an n-type polycrystal silicon film and is disposed on one of the side walls of the control gate. The memory gate comprises a doped polycrystal silicon film with a sheet resistance lower than that of the control gate comprising a polycrystal silicon film formed by ion implantation of impurities to the undoped silicon film.

Abstract: A method to form a strain-inducing semiconductor region is described. In one embodiment, formation of a strain-inducing semiconductor region laterally adjacent to a crystalline substrate results in a uniaxial strain imparted to the crystalline substrate, providing a strained crystalline substrate. In another embodiment, a semiconductor region with a crystalline lattice of one or more species of charge-neutral lattice-forming atoms imparts a strain to a crystalline substrate, wherein the lattice constant of the semiconductor region is different from that of the crystalline substrate, and wherein all species of charge-neutral lattice-forming atoms of the semiconductor region are contained in the crystalline substrate.

Abstract: A non-volatile memory device comprises a substrate with the dielectric layer formed thereon. A control gate and a floating gate are then formed on top of the dielectric layer. Accordingly, a non-volatile memory device can be constructed using a single poly process that is compatible with conventional CMOS processes. In addition, an assist gate, or assist gates are formed on the dielectric layer next to and between the control gate and floating gate respectively. The assist gates are used to form inversion diffusion regions in the substrate. By using the assist gates to form inversion diffusion regions, the overall size of the device can be reduced, which can improve device density.

Abstract: A non-volatile semiconductor memory device with good write/erase characteristics is provided. A selection gate is formed on a p-type well of a semiconductor substrate via a gate insulator, and a memory gate is formed on the p-type well via a laminated film composed of a silicon oxide film, a silicon nitride film, and a silicon oxide film. The memory gate is adjacent to the selection gate via the laminated film. In the regions on both sides of the selection gate and the memory gate in the p-type well, n-type impurity diffusion layers serving as the source and drain are formed. The region controlled by the selection gate and the region controlled by the memory gate located in the channel region between said impurity diffusion layers have the different charge densities of the impurity from each other.

Abstract: Nonvolatile memory cells and array are provided. The memory cell comprises a body, a source, a drain, and a charge storage region. The body comprises an n-type conductivity and is formed in a well of the n-type conductivity. The source and the drain have p-type conductivity and are formed in the well with a channel of the body defined therebetween. The charge storage region is disposed over and insulated from the channel by a channel insulator. Each cell further comprises a bias setting having a source voltage applied to the source, a well voltage applied to the well, and a drain voltage applied to the drain. A bias configuration for an erase operation of the memory cell is further provided, wherein the source voltage is sufficiently more negative with respect to the well voltage and is sufficiently more positive with respect to the drain voltage to inject hot holes onto the charge storage region. The cells can be arranged in row and column to form memory arrays and memory device.

Abstract: In one embodiment, a semiconductor device includes a semiconductor substrate having a first junction region and a second junction region. An insulated floating gate is disposed on the substrate. The floating gate at least partially overlaps the first junction region. An insulated program gate is disposed on the floating gate. The program gate has a curved upper surface. The semiconductor device further includes an insulated erase gate disposed on the substrate and adjacent the floating gate. The erase gate partially overlaps the second junction region.

Abstract: An active region and an opposite conductivity active region are formed in a semiconductor substrate. The opposite conductivity active region is covered with a resist pattern. Impurities are implanted into a surface layer of the active region. An angle ?0 is defined as a tilt angle obtained by tilting a virtual plane perpendicular to the substrate and including an edge of the active region, toward the resist pattern by using as a fulcrum a point on the substrate nearest to the resist pattern, until the virtual plane contacts the resist pattern. The ion implantation is performed in a direction having a tilt angle larger than ?0 and allowing ions passed through the uppermost edge of the resist pattern to be incident upon an area between the resist pattern and the active region, and is not performed along a direction allowing the ions to be incident upon the active region.

Abstract: Methods of forming a gate structure for an integrated circuit memory device include forming a first dielectric layer having a dielectric constant of under 7 on an integrated circuit substrate. Ions of a selected element from group 4 of the periodic table and having a thermal diffusivity of less than about 0.5 centimeters per second (cm2/s) are injected into the first dielectric layer to form a charge storing region in the first dielectric layer with a tunnel dielectric layer under the charge storing region. A metal oxide second dielectric layer is formed on the first dielectric layer, the second dielectric layer. The substrate including the first and second dielectric layers is thermally treated to form a plurality of discrete charge storing nano crystals in the charge storing region and a gate electrode layer is formed on the second dielectric layer. Gate structures for integrated circuit devices and memory cells are also provided.

Abstract: A method of fabricating a silicon nitride layer is described. First, a substrate is provided. Then, a silicon nitride layer is formed on the substrate. The silicon nitride layer is UV-cured in an atmosphere lower than the standard atmospheric pressure. Through the UV curing treatment, the tensile stress of the silicon nitride layer is increased.

Abstract: A semiconductor device has: an isolation region formed on a semiconductor substrate and defining a continuous active region including a select transistor region and a direct tunnel element region; a gate insulating film formed on a channel region of the select transistor region; a tunnel insulating film formed on a partial area of the direct tunnel element region and having a thickness different from a thickness of the gate insulating film; a continuous floating gate electrode formed above the gate insulating film and the tunnel insulating film; an inter-electrode insulating film formed on a surface of the floating gate electrode; a control gate electrode facing the floating gate electrode via the inter-electrode insulating film; and a pair of source/drain regions formed on both sides of the channel region of the select transistor region and not overlapping the tunnel insulating film.

Abstract: Novel memory cells utilize source-side injection, allowing very small programming currents. If desired, to-be-programmed cells are programmed simultaneously while not requiring an unacceptably large programming current for any given programming operation. In one embodiment, memory arrays are organized in sectors with each sector being formed of a single column or a group of columns having their control gates connected in common. In one embodiment, a high speed shift register is used in place of a row decoder to serially shift in data for the word lines, with all data for each word line of a sector being contained in the shift register on completion of its serial loading. In one embodiment, speed is improved by utilizing a parallel loaded buffer register which receives parallel data from the high speed shift register and holds that data during the write operation, allowing the shift register to receive serial loaded data during the write operation for use in a subsequent write operation.

Abstract: A method of forming an array of floating gate memory cells, and an array formed thereby, that includes source and drain regions formed in a substrate, and a conductive block of material disposed over the source region. The floating gate is formed as a thin, L-shaped layer of conductive material having a first portion disposed over the channel region and a second portion extending vertically along the conductive block. The control gate includes a first portion disposed adjacent to and insulated from a distal end of the floating gate first portion, and a second portion disposed adjacent to the channel region. A portion of the control gate could extend into a trench formed into the substrate, wherein the drain region is formed underneath the trench, and the channel region has a first portion extending along the trench sidewall and a second portion extending along the substrate surface.

Abstract: Structures and methods for write once read only memory employing charge trapping are provided. The write once read only memory cell includes a metal oxide semiconductor field effect transistor (MOSFET) in a substrate. The MOSFET has a first source/drain region, a second source/drain region, and a channel region between the first and the second source/drain regions. A gate insulator is formed opposing the channel region. The gate insulator includes a number of high work function nanoparticles. A gate is formed on the gate insulator. A plug is coupled to the first source/drain region and couples the first source/drain region to an array plate. A transmission line is coupled to the second source/drain region. The MOSFET is a programmed MOSFET having a charge trapped in the number of high work function nanoparticles in the gate insulator adjacent to the first source/drain region.

Abstract: A nitride read only memory cell comprising a silicon-germanium layer with a pair of source/drain regions. A strained silicon layer is formed overlying the silicon-germanium layer such that the pair of source/drain regions is linked by a channel that is generated in the strained silicon layer during operation of the cell. A nitride layer is formed overlying the substrate. The nitride layer has at least one charge storage region. The nitride layer may be a planar layer, a planar split gate nitride layer, or a vertical split nitride layer. A control gate is formed overlying the nitride layer. Ballistic direct injection is used to program the memory cell. A first charge storage region of the nitride layer establishes a virtual source/drain region in the channel. The virtual source/drain region has a lower threshold voltage than the remaining portion of the channel.