Abstract: Systems and methods are provided for nonvolatile memory devices that incorporate a band-gap engineered gate stack with asymmetric tunnel barriers. One embodiment of a memory device includes first and second source/drain regions separated by a channel region in a substrate, a control gate, and a gate stack between the control gate and the channel region. The gate stack includes a first insulator region in contact with the channel region, a floating charge-storage region in contact with the first insulator region, and a second insulator region in contact with the floating charge-storage region and the control gate. The gate stack includes selected material, in conjunction with control gate metallurgy, for providing desired asymmetric energy barriers that are adapted to primarily restrict carrier flow during programming to a selected carrier between the control gate and the floating charge-storage region, and to retain a programmed charge in the floating charge-storage region. Other aspects are provided herein.

Abstract: Methods are disclosed for erasing a flash memory cell including: (a) a semiconductor substrate, (b) a gate, (c) a source, (d) a drain, (e) a well, the gate including: (1) a tunnel oxide film, (2) a floating gate, (3) a dielectric film and (4) a control gate stacked on the semiconductor substrate. In one of the disclosed methods, a positive bias voltage is applied to the control gate, the source and drain are floated, a negative bias voltage is applied to the well, a ground voltage is then applied to the well while maintaining the positive bias voltage at the control gate, and subsequently a ground voltage is applied to the control gate.

Abstract: A module board has trace impedances that are matched at trace junctions. An input line that drives a signal to a junction has its impedance adjusted to match the equivalent impedance of branch traces output from the junction. Since input and output impedances match, reflections caused by the junction are minimized or eliminated. The input impedance can match by being within 20% of the equivalent impedance of the branch lines. The equivalent impedance of branches is the reciprocal of the sum of the individual branch lines' reciprocal impedance. Termination can be eliminated when such junctions are impedance-matched. Secondary junctions can also be impedance-matched, allowing for a variety of trace topologies. Such trace-impedance matching is especially useful for memory modules.

Abstract: A static random access memory (SRAM) device and a method of manufacturing the same. In one embodiment, the SRAM device includes: (1) a first bias voltage contact biasable to a first potential, (2) a second bias voltage contact biasable to a second potential that differs from the first potential and (3) a well having channels formed therein and connected to one of said first and second bias voltage contacts based on a transistor characteristic of said SRAM device that bears on static noise margin (SNM) and write trip voltage Vtrip.

Abstract: A method for writing a memory module includes providing a plurality of memory cells. Each memory cell includes a substrate, a P-type drain and source, a gate, and a stack dielectric layer which stores 2-bit data. Memory cells are arranged in a matrix with gates and sources on the same row connected respectively to the same word line and same source line, and drains on the same column connected to the same bit line. Each line receives a respective voltage with the word line of the memory cell to be written receiving voltage to turn on its P-type channel, the word line of the memory cell not to be written receiving voltage to turn off its P-type channel, and the bit line of the memory cell to be written receiving voltage so that a hot hole in its P-type channel induces hot electron injection into its stack dielectric layer.

Abstract: A memory transistor having a pair of separate floating gates overlying end regions of a channel and a control gate that overlies the floating gates and a central region of the channel effectively operates as a pair of floating gate transistors with an intervening select transistor. Each floating gate can be charged to store a distinct binary, analog, or multi-bit value. An erase operation can use a negative voltage on the control and a positive voltage on an underlying well or source/drain region to cause tunneling that discharges one or both floating gates. Applying a limited current to a source/drain region during an erase operation can cause the source/drain region and a floating gate to rise together and avoid band-to-band tunneling and resulting hole injection into the floating gate.

Abstract: Disclosed is a block select circuit in a flash memory device. The block select circuit comprises a select unit including a flash memory cell that is programmable and erasable by a given voltage condition, for outputting a block select signal depending on address signals and a state of the flash memory cell, a high-voltage pumping unit for outputting a signal to keep a given high voltage according to the block select signal and the clock signal, and a switching unit for applying a given bias to a gate select line, a word line and a source select line of a flash memory cell block according to the output signal of the high-voltage pumping unit. A given voltage is not applied to a fail block by only the operation of programming the flash memory cell of an erase state. Therefore, it is possible to process a fail block even after being packaged.

Abstract: The multiple bit, vertical memory cell includes a vertical metal oxide semiconductor field effect transistor (MOSFET) extending horizontally outward from a substrate. The MOSFET has a first source/drain region, a second source/drain region, a channel region between the first and the second source/drain regions, and a gate separated from the channel region by a gate insulator. The gate insulator may be a composite of oxide-nitride-aluminum oxide. The MOSFET is operated with either the first source/drain region or the second source/drain region serving as the source region, depending on the voltages applied to these regions. A negative substrate bias is applied during programming and erasing operations to enhance hot carrier injection.

Abstract: At the time of burn-in test, substrate voltages of transistors in a sense amplifier are switched by a PMOS substrate voltage generating portion and an NMOS substrate voltage generating portion. Specifically, the substrate voltage of a P channel MOS transistor is increased during the test than in a normal operation, whereas the substrate voltage of an N channel MOS transistor is decreased during the test than in the normal operation. Consequently, the threshold voltages of the P channel and N channel MOS transistors can be increased upon the test. Leakage currents in the turned-off states can be reduced, and thus, power consumption during the burn-in test can be decreased.

Abstract: A flash memory device is disclosed in which includes a plurality of flash memory transistors disposed within a doped well. The transistors are coupled at respective sources to an array ground node via a plurality of array ground lines. A plurality of switching devices distributed throughout the doped well switchingly couple the array ground lines to the doped well to reduce an elevated voltage otherwise present on the array ground lines when the array ground lines are heavily loaded.

Abstract: A nonvolatile semiconductor memory apparatus suitable to logic incorporation, by which a charge injection efficiency is high and hot electrons (HE) can be effectively injected at a low voltage is provided. A memory transistor (M) comprises first and second source/drain regions (S, SSL, D, SBL) formed on a semiconductor substrate (SUB, W), a charge storage film (GD) having a charge storage faculty and a gate electrode (WL). Memory peripheral circuits (2a to 9) generate a first voltage (Vd) and a second voltage (Vg-Vwell), apply the first voltage (Vd) to the second source/drain region (D, SBL) by using potential (0V) of the first source/drain region (S, SSL) as reference, apply the second voltage (Vg-Vwell) to the gate electrode (WL), generate hot electrons (HE) by ionization collision on the second source/drain region (D, SBL) side, and inject the hot electrons (HE) to the charge storage film (GD) from the second source/drain region (D, SBL) side at the time of writing data.

Abstract: A cell array is configured by arranging a plurality of electrically writable erasable nonvolatile memory cells on a semiconductor substrate. Each of the memory cells has a charge accumulation layer formed via a first gate insulating film and a gate electrode formed on the charge accumulation layer via a second gate insulating film. A control circuit controls the sequence of writing and erasing the data into and from a memory cell selected in the memory cell. In writing the data into the memory cell, a first write operation is to apply a write pulse voltage with a first step-up voltage between the gate electrode and the semiconductor substrate. A second write operation is to apply a write pulse voltage with a second step-up voltage lower than the first step-up voltage.

Abstract: A flash memory cell programming system and method that facilitate efficient and quick operation of a flash memory cell by providing a biasable well (e.g., substrate) is presented. The biasable well flash memory cell enables increases in electrical field strengths in a manner that eases resistance to charge penetration of a dielectric barrier (e.g., oxide) around a charge trapping region (e.g., a floating gate). The present biasable well system and method also create a self convergence point that increase control during programming operations and reduces the chances of excessive correction for over erased memory cells. The biasing can assist hard programming to store information and/or soft programming to correct the effects of over-erasing. The biasing can also reduce stress on a drain voltage pump, reduce leakage current and reduce programming durations. Some implementations also include a biasable control gate component, biasable source component and biasable drain component.

Abstract: A memory cell has a trench formed into a surface of a semiconductor substrate, and spaced apart source and drain regions with a channel region formed therebetween. The source region is formed underneath the trench, and the channel region includes a first portion extending vertically along a sidewall of the trench and a second portion extending horizontally along the substrate surface. An electrically conductive floating gate is disposed in the trench adjacent to and insulated from the channel region first portion. An electrically conductive control gate is disposed over and insulated from the channel region second portion. A block of conductive material has at least a lower portion thereof disposed in the trench adjacent to and insulated from the floating gate, and can be electrically connected to the source region. A method of programming the cell comprises the steps of creating an inversion layer in the second portion of the channel.

Abstract: For particularly flexible and space-saving information storage, a charge trapping memory cell and a corresponding semiconductor memory device include a charge trapping gate configuration provided with a plurality of charge trapping gates each configured for substantially independent information storage. As a result, a plurality of information units can be stored independently of one another in the memory cell. Also provided is a method for producing such a memory cell.

Abstract: A novel structure of nonvolatile memory is disclosed. The non-volatile memory includes two serially connected PMOS transistors. The characteristic of the devices is that bias is not necessary to apply to the floating gate during the programming mode. Thus, the control gate is omitted for the structure or layout, thereby saving the space for making the control gate. The carrier may be “automatically injected” into floating gate for programming the status of the devices.

Abstract: A non-volatile memory device includes floating gate memory cells, a pulse counter and voltage pump control circuitry. The control circuitry selectively activates pumps in response to a count output of the counter. In one embodiment, the pump output current is increased as the counter output increases. The memory allows for erase operations that reduce leakage current during initiation of an erase operation.

Abstract: A method of erasing a non-volatile memory includes applying a first potential of first polarity to a control gate; applying a second potential of second polarity to a bulk region, the second potential being an N magnitude; and applying a third potential of second polarity to a source region, the third potential being an M magnitude, wherein the N and M are substantially the same.

Abstract: P-channel MOSFET devices are used as reprogrammable fuse or antifuse elements in a memory decode circuit by utilizing anomalous hole generation. An applied negative gate bias voltage is sufficiently large to cause tunnel electrons to gain enough energy to exceed the band gap energy of the oxide. This causes energetic hole-electron pairs to be generated in the silicon substrate. The holes are then injected from the substrate into the oxide, where they remain trapped. A large shift in the threshold voltage of the p-channel MOSFET results. The device can subsequently be reset by applying a positive gate bias voltage. Various circuits incorporating such fuse or antifuse elements are also disclosed.

Abstract: In a nonvolatile memory in which a load on a boosting circuit changes according to the number of rewrite bytes, the boosting circuit is configured so as to perform voltage boosting at a relatively slow predetermined speed regardless of the number of rewrite bytes, whereby stress applied to each storage element is reduced and rewrite resistance is enhanced.

Abstract: A method of operating a non-volatile memory device includes providing the non-volatile memory device with a body of first conductivity, a source region of second conductivity, a drain region of second conductivity on the body, and a control gate over the body adjacent to the source and drain regions. A first voltage of first polarity is applied to the control gate. A second voltage of first polarity is applied to the drain region, the second voltage being less than about 5.6 volts. A third voltage of second polarity is applied to the source region.

Abstract: A non-volatile semiconductor memory device includes: a memory cell array having NAND strings arranged therein, each NAND string having a plurality of electrically rewritable and non-volatile memory transistors connected in series; and an erase/write/read control circuit configured to perform erasing, writing and reading of the memory cell array, wherein at least one memory transistor within each NAND string of the memory cell array is controlled as a block separation transistor for dividing the memory cell array into a plurality of blocks each serving as a unit of data erasure.

Abstract: The present invention provides a method and apparatus for erasing memory blocks. The apparatus includes a first plurality of memory cells formed in a substrate and a second plurality of memory cells formed in the substrate. The apparatus further includes a bias circuit adapted to provide an erasing voltage differential to the first plurality of memory cells and a compensating voltage differential to the second plurality of memory cells, wherein the erasing voltage differential is larger than the compensating voltage differential.

Abstract: A sector structure of a flash memory device minimizes a layout area in a chip without deteriorating a high-speed operation. The sector structure of the flash memory device includes a plurality of sectors, each sector including memory cell transistors in a cell array block sharing a common bulk region with transistors in a column decoder block.

Abstract: A semiconductor storage device in the invention comprises a first control transistor which is connected between a bit line and a first node and whose control terminal is connected to a word line, a data retention circuit which includes a first transistor that is connected between the first node and a second reference voltage terminal, as well as a first inverter that includes a second transistor connected between a second node and the second reference voltage terminal, and a substrate potential control circuit which selectively alters a substrate potential of the first transistor so as to make a threshold voltage of the first transistor higher as compared with threshold voltages of the first control transistor and the second transistor. Thus, it is permitted to provide the semiconductor storage device of static type which realizes a reduced layout area and a lower-dissipation-power operation while ensuring the reliability and high operating speed of the write and read of data.

Abstract: A memory transistor has a pair of separate floating gates overlying end regions of a channel and a control gate that overlies the floating gates and a central region of the channel. The memory transistor effectively operates as a pair of floating gate transistors with an intervening select transistor. Each floating gate can be charged to store a distinct binary, analog, or multi-bit value. The direction of the channel current controls which floating gate receives channel hot electron injection during programming and which floating gate state is sensed during reading. A read operation biases the word line higher that the threshold voltage used to store data and compares the resulting channel to reference currents to identify a stored binary, analog, or multi-bit value. The threshold voltage range can include negative threshold voltages, which increases the available range for multi-bit-per-floating gate storage.

Abstract: A method to suppress bit-line leakage in a nonvolatile memory cell is achieved. The method comprises providing an array of nonvolatile memory cells comprising source and bulk terminals. The array comprises a plurality of subarrays. The sources of all the nonvolatile cells in each subarray are coupled together to form a common subarray source. Bulks of all the nonvolatile cells in the array are coupled together to form a common array bulk. A first, non-zero voltage is forced between the common subarray source and the common array bulk for a first subarray that is selected for an access operation. A second, non-zero voltage is forced between the common subarray source and the common array bulk for a second subarray that is not selected for an access operation. The second, non-zero voltage inhibits bit line leakage in the second subarray.

Abstract: A three-dimensional memory array includes a plurality of rail-stacks on each of several levels forming alternating levels of X-lines and Y-lines for the array. Memory cells are formed at the intersection of each X-line and Y-line. The memory cells of each memory plane are all oriented in the same direction relative to the substrate, forming a serial chain diode stack. In certain embodiments, row and column circuits for the array are arranged to interchange function depending upon the directionality of memory cells in the selected memory plane. High-voltage drivers for the X-lines and Y-lines are each capable of passing a write current in either direction depending on the direction of the selected memory cell. A preferred bias arrangement reverse biases only unselected memory cells within the selected memory plane, totaling approximately N2 memory cells, rather than approximately 3N2 memory cells as with prior arrays.

Abstract: A method of discharging a charge storage location of a transistor of a non-volatile memory includes applying first and second voltages to a control gate and a well region, respectively, of the transistor. The first voltage is applied to the control gate of the transistor, wherein the control gate has at least a portion located adjacent to a select gate of the transistor. The transistor includes a charge storage location having nanoclusters disposed within dielectric material of a structure of the transistor located below the control gate. Lastly, a second voltage is applied to the well region located below the control gate. Applying the first voltage and the second voltage generates a voltage differential across the structure for discharging electrons from the nanoclusters of the charge storage location.

Abstract: A non-volatile memory cell (10) includes a charge-storing node (16). An electrically insulating first layer (76) is coupled between the node and a source of a first voltage (22). An electrically insulating second layer (66) is coupled between the node and a source of a second voltage (20-21). The area of the first layer is smaller than the area of the second layer. A controller (90) is arranged to cause the first voltage to be greater than the second voltage so that charge is extracted from the node and is arranged to cause the second voltage to be greater than the first voltage so that charge is injected into the node.

Abstract: System for using a dynamic reference cell in a double-bit cell memory. A method is provided for reading and verifying a double bit core cell in a memory device. The memory device includes a dynamic reference cell and a fixed reference cell. The method comprises the steps of programming the dynamic reference cell using the fixed reference cell, and programming the double bit core cell using the dynamic reference cell. When the dynamic reference cell is located along with the core cell on the same word line, a constant current source added to a core cell data line operates during program verify to provide a current difference between the core cell and the dynamic reference cell.

Abstract: A method for programming PMOS single transistor flash memory cells through channel hot carrier induced hot electron injection mechanism is disclosed. The PMOS single transistor flash memory cell includes an ONO stack layer situated on an N-well of a semiconductor substrate, a P+ poly gate formed on the ONO stack layer, a P+ doped source region disposed in the N-well at one side of the gate, and a P+ doped drain region disposed in the N-well at the other side of the gate. The method includes the steps of: applying a word line voltage VWL on the P+ poly gate, applying a source line voltage VSL on the source, wherein the source line voltage VSL is greater than the word line voltage VWL, thereby providing adequate bias to turn on the P channel thereof.

Abstract: In a nonvolatile memory in which a load on a boosting circuit changes according to the number of rewrite bytes, the boosting circuit is configured so as to perform voltage boosting at a relatively slow predetermined speed regardless of the number of rewrite bytes, whereby stress applied to each storage element is reduced and rewrite resistance is enhanced.

Abstract: An nonvolatile memory device having improved endurance is comprised of an array of nonvolatile memory cells arranged in rows and columns. Each memory cell of each row is connected to a word line and a source select line, and each memory cell of each column connected to a first bit line and a second bit line. Each memory cell is composed of a first transistor and second transistor. The first and second transistors have control gate connected to the word line receive a word line voltage, a source connected the source select line to receive a source line voltage, and a floating gate onto which an electronic charge is placed representing a data bit stored within the nonvolatile memory device. The first transistor has a drain connected the first bit line to receive a first bit line voltage and the second transistor a drain connected to the second bit line to receive a second bit line voltage.

Abstract: There are provided a boosted voltage generating circuit and a semiconductor memory device having the boosted voltage generating circuit which includes a booster circuit for outputting high voltage obtained by boosting the power supply voltage, a regulator circuit supplied with the high voltage, for generating voltage whose voltage value is smaller than the value of the high voltage and which is variably set to at least two values based on the high voltage at the operating time, and a equalizer circuit connected to the booster circuit and regulator circuit, for short-circuiting an output node of the booster circuit and an output node of the regulator circuit in response to a first control signal, wherein the operative period of the regulator circuit and the short-circuiting operation period of the equalizer circuit do not overlap each other.

Abstract: The present invention presents a non-volatile memory having a plurality of erase units or blocks, where each block is divided into a plurality of parts sharing the same word lines to save on the row decoder area, but which can be read or programmed independently. An exemplary embodiment is a Flash EEPROM memory with a NAND architecture that has blocks composed of a left half and a right half, where each part will accommodate one or more standard page (data transfer unit) sizes of 512 bytes of data. In the exemplary embodiment, the left and right portions of a block each have separate source lines, and separate sets of source and drain select lines. During the programming or reading of the left side, as an example, the right side can be biased to produce channel boosting to reduce data disturbs. In an alternate set of embodiments, the parts can have separate well structures.

Abstract: In this invention, by offering specific array-end structures and their fabrication method, the three resistive layers of diffusion bit line, control gate and word gate polysilicons, where control gate polysilicon can run on top of the diffusion bit line, are most effectively stitched with only three layers of metal lines keeping minimum metal pitches. The stitching method can also incorporate a bit diffusion select transistor and/or a control gate line select transistor. The purpose of the select transistors may be to reduce the overall capacitance of the bit line or control gate line, or to limit the disturb conditions that a grouped sub-array of cells may be subjected to during program and/or erase.

Abstract: In the present method of programming a selected flash EEPROM memory cell of a pair thereof in series, a positive voltage is applied to the drain of the selected cell to be programmed, a voltage lower than the voltage applied to the drain is applied to the source of the selected cell, a negative voltage is applied to the substrate, and a positive voltage is applied to the control gate sufficient to induce hot electron injection from the drain to the floating gate of the selected cell.

Abstract: A method to suppress bit-line leakage in a nonvolatile memory cell is achieved. The method comprises providing an array of nonvolatile memory cells comprising source and bulk terminals. The array comprises a plurality of subarrays. The sources of all the nonvolatile cells in each subarray are coupled together to form a common subarray source. Bulks of all the nonvolatile cells in the array are coupled together to form a common array bulk. A first, non-zero voltage is forced between the common subarray source and the common array bulk for a first subarray that is selected for an access operation. A second, non-zero voltage is forced between the common subarray source and the common array bulk for a second subarray that is not selected for an access operation. The second, non-zero voltage inhibits bit line leakage in the second subarray.

Abstract: An error recovery technique is used on marginal nonvolatile memory cells. A marginal memory cell is unreadable because it has a voltage threshold (VT) of less than zero volts. By biasing adjacent memory cells, this will shift the voltage threshold of the marginal memory cells, so that it is a positive value. Then the VT of the marginal memory cell can be determined. The technique is applicable to both binary and multistate memory cells.

Abstract: A technique for reducing the read gate voltage in a memory array including memory cells having a transistor for storing charge indicative of the value stored in the cell. In one example, a voltage greater than the substrate voltage is applied to the sources of the transistors of the memory cells of the array to increase the threshold voltage of a transistor due the body effect. The read gate voltage is greater than the source voltage which is greater than the substrate voltage. A non read voltage of less than the source voltage is applied to the gates of the transistors of the unselected rows to reduce leakage current. With this embodiment, the threshold voltages of transistors having an erased state can be less than 0V. With some embodiments, the read disturb caused by a gate voltage can be reduced due to the reduction in the gate voltage. In other examples, a negative voltage is applied to the gates of unselected rows to prevent leakage current.

Abstract: A charge pump circuit includes inverters INV1 and INV2. The inverter INV1 receives a clock signal CLK2, and applies a voltage waveform at an immediately previous node to a second end of a capacitor connected to a transistor and to the p-well thereof. The voltage of the capacitor on the side of the control terminal of the transistor, and the voltage waveform at the node, are raised with the same phase timing as the clock signal CLK1. The inverter INV2 receives a clock signal CLK1, and applies a voltage waveform at an immediately previous node to a second end of another capacitor connected to another transistor and to the p-well thereof. The voltage of the another capacitor on the side of the control terminal of the another transistor, and the voltage waveform at the node, are raised with the same phase timing as the clock signal CLK2.

Abstract: The memory cell is formed in a body of a P-type semiconductor material forming a channel region and housing N-type drain and source regions at two opposite sides of the channel region. A floating gate region extends above the channel region. A P-type charge injection region extends in the body contiguously to the drain region, at least in part between the channel region and the drain region. An N-type base region extends between the drain region, the charge injection region, and the channel region. The charge injection region and the drain region are biased by special contact regions so as to forward bias the PN junction formed by the charge injection region and the base region. The holes thus generated in the charge injection region are directly injected through the base region into the body, where they generate, by impact, electrons that are injected towards the floating gate region.

Abstract: An EEPROM memory cell comprising a transistor on a first conductivity type semiconductor substrate and a capacitor formed on a second conductivity type semiconductor substrate. The capacitor comprises first and second injector regions of third conductivity type, a channel region of second conductivity type separating the first and second injector regions and a first electrically floating structure disposed above the channel region, wherein a first edge portion of the floating structure overlaps a portion of the first injector region and a second edge portion of the first floating structure overlaps a portion of the second injector region, and a control gate region of fourth conductivity type located within the second conductivity type semiconductor substrate region. The gate structure and first floating structure are electrically connected together. In different aspects of the present invention, the EEPROM memory cell may also include a second capacitor.

Abstract: An electronic device, such as an opto-electronic device and an integrated semiconductor memory device, includes at least one integrated memory point structure including a quantum well semiconductor area buried in the substrate of the structure and disposed under the insulated gate of a transistor. A biasing voltage source is adapted to bias the structure to enable charging or discharging of charges in the quantum well or outside the quantum well.

Abstract: The invention advantageously provides a nonvolatile memory device and associated methods therefore, and, more particularly, an optimally designed nonvolatile memory device and methods therefor that advantageously prevent data loss in its trapping layer. A preferred embodiment of the method for operating a nonvolatile memory cell according to the invention advantageously comprises the steps of programming the memory cell, injecting electrons into a trapping layer of the memory cell from a semiconductor substrate, erasing the memory cell, detrapping the memory cell, and repeating the erasing and detrapping steps until a threshold voltage of the memory cell reaches a predetermined value. For the detrapping step, electrons can be detrapped from the trapping layer to a channel region of the memory cell, or to a gate of the memory cell.

Abstract: A method for operating a P-channel SONOS memory device that has a charge trapping layer located on a substrate, a gate electrode located on the trapping layer, two doped regions located in the substrate at each side of the charge trapping layer. The two doped regions are set to be a drain region and a source region. When a programming action is intended, the gate electrode and the drain region are applied with a first negative high-level bias, and the source region and the substrate are applied with a grounded voltage. When an erasing action is intended, the gate electrode is a second negative bias which is smaller than the first negative voltage in absolute value. In the mean time, the drain region is applied with the third negative bias and the substrate is applied with a grounded voltage. The third negative voltage is larger than the second negative bias in absolute value.

Abstract: The present invention relates to a method of erasing flash memory cells. In the flash memory cell having a substrate, a source, a drain, a tunnel oxide film, a floating gate, a dielectric film and a control gate, the method of erasing the flash memory cell comprises the steps of performing an erase operation for the cell, by applying a negative voltage being an erase voltage to the control gate and a positive voltage being an erase voltage to the substrate, discharging the control gate by making the control gate grounded, discharging the source by making the source grounded, and simultaneously performing a discharge operation and a recovery operation by making the substrate grounded. Therefore, the threshold voltages of the cells can be converted to have a constant voltage even though additional recovery operation is not performed.

Abstract: A flash memory device is disclosed in which includes a plurality of flash memory transistors disposed within a doped well. The transistors are coupled at respective sources to an array ground node via a plurality of array ground lines. A plurality of switching devices distributed throughout the doped well switchingly couple the array ground lines to the doped well to reduce an elevated voltage otherwise present on the array ground lines when the array ground lines are heavily loaded.

Abstract: An erasing method for a p-channel nitride read only memory. The method is used for a p-channel nitride read only memory having charges stored in a charge-trapping layer. A positive voltage is applied to the control gate and a negative voltage to the drain; also, the source is floating and the n-well is grounded. The voltage difference between the positive voltage applied to the control gate and the negative voltage to the drain is sufficient to trigger a band-to-band induced hot electron injection to erase the p-channel nitride read only memory.