WO1997002605A1 - Method of fabricating a fast programming flash e2prom cell - Google Patents

Method of fabricating a fast programming flash e2prom cell Download PDF

Info

Publication number
WO1997002605A1
WO1997002605A1 PCT/CA1996/000446 CA9600446W WO9702605A1 WO 1997002605 A1 WO1997002605 A1 WO 1997002605A1 CA 9600446 W CA9600446 W CA 9600446W WO 9702605 A1 WO9702605 A1 WO 9702605A1
Authority
WO
WIPO (PCT)
Prior art keywords
cell
drain
source
floating gate
voltage
Prior art date
Application number
PCT/CA1996/000446
Other languages
French (fr)
Inventor
Jeewika Chandanie Ranaweera
Ivan Kalastirsky
Elvira Gulerson
Wai Tung Ng
Clement Andre T. Salama
Original Assignee
Jeewika Chandanie Ranaweera
Ivan Kalastirsky
Elvira Gulerson
Wai Tung Ng
Salama Clement Andre T
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jeewika Chandanie Ranaweera, Ivan Kalastirsky, Elvira Gulerson, Wai Tung Ng, Salama Clement Andre T filed Critical Jeewika Chandanie Ranaweera
Priority to AU61851/96A priority Critical patent/AU6185196A/en
Priority to US08/981,745 priority patent/US6034896A/en
Publication of WO1997002605A1 publication Critical patent/WO1997002605A1/en

Links

Classifications

    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/66007Multistep manufacturing processes
    • H01L29/66075Multistep manufacturing processes of devices having semiconductor bodies comprising group 14 or group 13/15 materials
    • H01L29/66227Multistep manufacturing processes of devices having semiconductor bodies comprising group 14 or group 13/15 materials the devices being controllable only by the electric current supplied or the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched, e.g. three-terminal devices
    • H01L29/66409Unipolar field-effect transistors
    • H01L29/66477Unipolar field-effect transistors with an insulated gate, i.e. MISFET
    • H01L29/66825Unipolar field-effect transistors with an insulated gate, i.e. MISFET with a floating gate
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/02Semiconductor bodies ; Multistep manufacturing processes therefor
    • H01L29/06Semiconductor bodies ; Multistep manufacturing processes therefor characterised by their shape; characterised by the shapes, relative sizes, or dispositions of the semiconductor regions ; characterised by the concentration or distribution of impurities within semiconductor regions
    • H01L29/10Semiconductor bodies ; Multistep manufacturing processes therefor characterised by their shape; characterised by the shapes, relative sizes, or dispositions of the semiconductor regions ; characterised by the concentration or distribution of impurities within semiconductor regions with semiconductor regions connected to an electrode not carrying current to be rectified, amplified or switched and such electrode being part of a semiconductor device which comprises three or more electrodes
    • H01L29/1025Channel region of field-effect devices
    • H01L29/1029Channel region of field-effect devices of field-effect transistors
    • H01L29/1033Channel region of field-effect devices of field-effect transistors with insulated gate, e.g. characterised by the length, the width, the geometric contour or the doping structure
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/76Unipolar devices, e.g. field effect transistors
    • H01L29/772Field effect transistors
    • H01L29/78Field effect transistors with field effect produced by an insulated gate
    • H01L29/788Field effect transistors with field effect produced by an insulated gate with floating gate
    • H01L29/7881Programmable transistors with only two possible levels of programmation
    • H01L29/7884Programmable transistors with only two possible levels of programmation charging by hot carrier injection
    • H01L29/7886Hot carrier produced by avalanche breakdown of a PN junction, e.g. FAMOS

Definitions

  • the present invention relates in general to Electrically Erasable and Programmable Read-Only Memory (E 2 PROM) , and more particularly to a method of fabricating a flash E 2 PROM with fast programming speed and low operating voltages.
  • E 2 PROM Electrically Erasable and Programmable Read-Only Memory
  • Flash memory is a modified form of E 2 PROM which can be erased one block at a time and can be programmed one bit at a time.
  • flash memory chips are available in densities of up to 32M bits. Large data storage capacity with complete nonvolatility, results in numerous applications for such chips, ranging from cellular telephones, solid-state disks and memory cards. Such applications have traditionally been the domain of ROM, conventional E 2 PROM, battery-backed RAM, static RAM (SRAM) and magnetic storage.
  • a flash E 2 PROM cell resembles an ordinary MOS (Metal- Oxide-Semiconductor) transistor, except for the addition of a floating gate, which is buried in the insulator between the substrate and the conventional control gate. Charge stored on the floating gate alters the threshold voltage (V th ) of the device as measured at the control gate. Since the floating gate and the control gate are both stacked directly above the transistor channel, very high densities can be achieved. Even higher densities can be obtained by self aligning the control and floating gates to the source and drain regions.
  • MOS Metal- Oxide-Semiconductor
  • the cell is programmed and erased by adding electrons to and removing electrons from the floating gate, respectively.
  • a cell with a high threshold voltage (V th ) s in its "0" state. After erasure has been completed, the threshold voltage is reduced resulting in a "1" state.
  • the conductivity of the channel determines the information stored in the memory cell (i.e. current flowing through the channel is detected by sense- amplifier circuitry as a "1", while the absence of current is detected as a "0”) .
  • Flash E 2 PR0M technology has received industry-wide attention recently. Due to its simple single transistor cell architecture, flash memory may eventually cost less to make than DRAM (Dynamic Random Access Memory) .
  • DRAM Dynamic Random Access Memory
  • the channel hot electron programming method used in conventional flash E 2 PROM cells, requires biasing the device at high drain voltage (6 to 8V) to generate hot electrons. This results in additional circuit complexity and cost (ie. an additional external voltage supply) , which is a particular disadvantage in mobile applications such as cellular telephones, etc.
  • the channel hot electron injection method used for programming conventional flash E 2 PROM cells generates very high lateral source-to-drain currents (in the range of milliamperes) . These high currents limit the number of cells that can be programmed at one time.
  • FCAT-A Low-Voltage High Speed Alternative n-Channel Nonvolatile Memory Device is described in Masatada Horiuchi and Hisao Katto,"FCAT-A Low-Voltage High Speed Alternative n-Channel Nonvolatile Memory Device”; Transactions on Electron Devices, Vol. ED-26, No. 6, June 1979, pp. 914-918.
  • the FCAT cell has the p + regions placed outside the channel area corners adjacent to the diffused n + source and/or drain regions. This method of positioning the p + regions results in a corner point injection for the hot electrons which contributes to acute stress in the tunnel oxide. Also, the FCAT cell uses hot hole injection into the thin gate oxide for erasure, which can lead to poor reliability and endurance characteristics.
  • a low voltage flash E 2 PROM cell which overcomes the above- identified programming speed limitations of conventional prior art flash memory cells, and in addition allows the cell to operate at low voltages.
  • the flash E PR0M cell is comprised of two sections which are butted together.
  • One portion is covered by a highly doped p + pocket implant adjacent to at least one of the n + source and drain regions. This portion is referred to as the program section.
  • the remaining portion is not covered by the highly doped p + pocket implant and resembles a conventional E 2 PROM cell. This portion is referred to as the sense section.
  • the highly doped p + pocket implant and the n + drain and/or source regions create a junction having narrow depletion width such that when the junction is reverse biased an electric field is created for generating hot electrons for storage on the floating gate, thereby programming the flash E 2 PROM.
  • the programming speed of the improved flash E 2 PROM cell of the present invention is in general an order of magnitude faster than a conventional flash E 2 PROM cell of the same dimension.
  • the doping concentration of the p + pocket implant is chosen such that the junction will breakdown at low voltages (typically in the range of 2.5 to 5V) , thus enabling the drain and/or source regions to be driven directly by the logic level supply voltages. This eliminates the need for large charge pump circuits with high current capacity as used in conventional flash E 2 PROMs.
  • the enhancement of hot electron generation results in smaller drain and/or source currents (typically in the range of 50 to 200 ⁇ A/ ⁇ m of cell width) than prior art devices during the programming operation.
  • the flash E 2 PROM cell of the present invention consumes much less power, and is more suitable for battery operated portable equipment than similar prior art devices.
  • High positive and negative voltages are needed only to drive the control gate. Consequently, they can be generated on-chip using simple charge pump circuitry with small current driving capability.
  • the voltages required at the source and drain in all modes of operation can be obtained directly from a low power supply rail.
  • the flash E 2 PROM cell of the present inventi'on is compatible with CMOS and BiCMOS processes. More particularly, according to one aspect of the present invention, a method of fabricating an E 2 PROM cell is provided in which the fabrication steps are modular and compatible with conventional CMOS and BiCMOS process flows. Cell fabrication can be integrated into existing CMOS and BiCMOS processes with minimal changes. The resulting flash E 2 PROM cell is a self aligned structure.
  • the p + pocket implant is carried out in the same way as the n-LDD implant commonly used in modern CMOS and BiCMOS processes. The generation of hot electrons is dependent on the doping level of the p + /n + junction rather than the size of the channel length.
  • Figure 1 is a schematic representation of a prior art flash E 2 PROM cell
  • Figure 2 is a schematic representation of hot electron injection programming of the prior art flash memory cell depicted in Figure 1;
  • Figure 3 is a schematic representation of erasure of the prior art flash memory cell depicted in Figure 1 using Fowler-Nordheim tunnelling across the thin gate oxide;
  • Figure 4 is a schematic representation of the read operation of the prior art flash memory cell depicted in Figure 1;
  • Figure 5 is a three-dimensional schematic representation of a flash E 2 PR0M cell according to a first embodiment of the present invention
  • Figure 6 is a three-dimensional schematic representation of Zener and/or avalanche breakdown-based electron injection programming of the flash E 2 PROM cell of Figure 5;
  • Figure 7 is a three-dimensional schematic representation of the read operation of the flash E 2 PROM cell of Figure 5;
  • Figure 8 is a three-dimensional schematic representation of the erasure operation of the flash E 2 PROM cell of Figure 5;
  • Figure 9 is a three-dimensional schematic representation of the preferred embodiment of flash E 2 PROM cell according to the present invention.
  • Figure 10 is a plan view of the device shown in Figure 9;
  • FIGS 11(a) through 11(r) depict the process flow steps for fabricating the flash E 2 PROM cell according to the present invention
  • Figures 12A and 12B show a circuit diagram and cross-sectional diagram, respectively, of the flash E 2 PROM cell of the preferred embodiment configured in a cross point array architecture;
  • Figure 13 shows a simulation of programming time as a function of channel length for different p + doping concentrations for the preferred embodiment of Figure 9;
  • Figure 14 shows simulated and measured programming characteristics for the preferred embodiment of Figure 9; and Figure 15 shows the experimental effect of the p + /n + junction on the disturb characteristics of a half- selected programmed and erased E 2 PR0M cell according to the preferred embodiment of Figure 9.
  • a flash E 2 PROM cell is shown (ETOXTM cell by Intel Corp.), having a floating gate 1 and a control gate 3 stacked vertically above the MOFSET channel 5 in substrate 6 intermediate the source 7 and drain 9.
  • a tunnel oxide 11 is provided intermediate the floating gate l and channel 5, and an inter-poly oxide 13 is provided intermediate the floating gate 1 and control gate 3.
  • a voltage of +12V is applied to the control gate 3
  • the source 7 is grounded, and approximately +6V is applied between source 7 and drain 9 as shown in Figure 2.
  • the control gate 3 is capacitively coupled to the floating gate 1.
  • the source-drain voltage (+6V) generates hot electrons that are swept across the channel 5 from source 7 to drain 9. These hot electrons collide with atoms in the channel 5 thereby creating additional hot electrons.
  • the high voltage (+12V) on control gate 3 attracts the hot electrons across the thin tunnel oxide 11 where they accumulate on the floating gate 1. When enough electrons have accumulated, the cell switches from its "1" (erased) state to its "0" (programmed) state.
  • the voltage applied to the drain 9 does not, in general, have a significant effect on the programmed threshold voltage of the EEPROM cell. However, it does affect the speed of programming of the device.
  • Fowler-Nordheim tunnelling is used to remove electrons from the floating gate 1. More particularly, by floating the drain 9, grounding the control gate 3 and applying +12V to the source 7, an electric field is generated across the thin tunnel oxide 11 between the floating gate 1 and the source 7. This electric field attracts electrons off of the floating gate 1 toward the source 7, as shown in Figure 3.
  • the threshold voltage (V th ) returns to its initial value thereby resetting the cell to a "1" state.
  • the prior art ETOXTM memory cell operates like an ordinary transistor.
  • address inputs select specific transistors within the memory cell array.
  • Supply-voltage levels are applied to the control gate 3 and drain 9, while the source 7 is grounded as shown in Figure 4.
  • FIG. 5 a flash E 2 PROM cell is shown for achieving simultaneous reduction of programming time and operating voltages over the prior art cell of Figure 1.
  • the cell according to the present invention comprises a substrate 15, drain and source regions 17 and 19, respectively, a floating gate 21 and control gate 23, in the usual manner. Tunnel and inter-poly dielectric layers 22 and 24 are also provided in the usual manner.
  • additional highly doped regions 25 and 27 are provided adjoining the drain and/or source regions 17 and/or 19, respectively.
  • the additional highly doped regions 25 and 27 (occasionally referred to herein as pocket implants) extend along a portion of the width of the cell structure identified as the program section 29, the remaining sense section 31 being provided for the read operation of the cell, as discussed in greater detail below.
  • the substrate is made of p-type material
  • the drain and source regions are fabricated from n + -type material
  • the additional highly doped regions 25 and 27 comprise p + -type regions.
  • programming is accomplished via low voltage Zener and/or avalanche breakdown at the reverse biased p + /n + junctions in the program section 29.
  • High doping concentrations at these junctions results in a narrow depletion width so that only a small voltage is required to generate a sufficiently strong electric field to create hot electrons for storage on the floating gate 21.
  • the breakdown mechanism which dominates and the applied voltage required for breakdown depend on the doping of the p + regions.
  • Computer simulation results have shown that p + doping concentrations in the range of lxl0 18 cm “3 to 5xl0 19 cm "3 are suitable for the implementation of this structure.
  • the cell is programmed and read from the program and sense sections 29 and 31, respectively, and erased by Fowler-Nordheim tunnelling of electrons from the floating gate 21 to the source 19.
  • low positive voltages (less than 5V) are applied to the drain 17 and source 19 while a high programming voltage (in the range of 10 to 15 Volts) is applied to control gate 23.
  • the substrate 15 is grounded.
  • Low voltage Zener and/or avalanche breakdown at the reversed biased p + /n + junctions results in the generation of hot electrons which are drawn to the floating gate 21 as a result of the high voltage applied to control gate 23.
  • FIG. 7 schematically shows the read mode operation of the flash memory cell according to the present invention.
  • a low read voltage e.g. +5 Volts
  • a voltage lower than the breakdown voltage of the p + /n + junction is applied to the drain 17 (e.g. 13 volts), and the source 19 and substrate 15 are grounded.
  • Erasure is performed by applying a high negative voltage (in the range of -12 to -15 Volts) to the control gate 23, applying a voltage, lower than the breakdown voltage of the p + /n + junction (e.g. +3.5 volts), to the source 17, while grounding the substrate 15 and leaving the drain in an open circuit condition, as shown in Figure 8.
  • the voltage difference between the control gate 23 and the source 19, creates a strong electric field, across the thin oxide 22 between the floating gate 21 and the source 19, which in turn causes the removal of electrons from the floating gate 21 by Fowler-Nordheim tunnelling.
  • the flash E 2 PROM memory cell according to the present invention results in extremely fast programming of the cell (in the order of nanoseconds) at low supply voltages, making the cell particularly applicable in the field of portable electronics.
  • the preferred cell structure according to the present invention uses a single p + /n + junction in the programming section of the drain region of the transistor, as illustrated in Figures 9 and 10.
  • This embodiment offers the advantage that the erasing voltage is not limited by the p + /n + junction breakdown voltage, whereby the erasing speed can be improved by applying higher voltages to the source.
  • the cell is programmed from the drain side of the program section, erased from the source side, and read from the sense section of the cell, according to the principles discussed above in connection with Figures 5-8.
  • An improved method of reading the preferred embodiment of flash E 2 PROM cell shown in Figures 9 and 10 comprises the steps of grounding the substrate 15 and drain 17, applying a positive logic level voltage to the control gate 23, and applying a voltage of no greater than logic level to the source region 19.
  • the voltage applied to the control gate 23 allows current to flow through the channel region in the sense section, from the source region to the drain region in the event that little or no charge is stored on the floating gate.
  • the voltage applied to the source region 19 can be higher than the breakdown voltage (typically 3V or less) of the p + pocket implant/ n + drain junction, resulting in a higher current flow through the channel during read operation. This will allow faster read time and lower sensing error rate.
  • the cross-section after this step is shown in Figure 11(c) .
  • amorphous silicon by low pressure chemical vapor deposition (LPCVD) , the target thickness is 3600 A.
  • LPCVD low pressure chemical vapor deposition
  • the oxide-nitride-oxide (ONO) interpoly dielectric is grown in this step.
  • the target equivalent thickness is 200 A.
  • the cross- section after this step is shown in Figure
  • control gate 23 Deposit amorphous silicon by LPCVD to form the control gate 23.
  • the target thickness is 3600 A.
  • This ion implantation step is used for setting the doping of the polysilicon. The cross-section after this step is shown in
  • LPCVD oxide The target thickness is "3500 A. This oxide serves as a mask in Reactive Ion Etching (RIE) of the polysilicon layers.
  • RIE Reactive Ion Etching
  • the target oxide thickness is ⁇ 500 A. This oxide is used as a screen layer during the ion implantation to form the substrate contacts and the P + region of the Zener injector.
  • This photolithography process step opens a window through which the p + region of the Zener junction and the substrate contacts are formed.
  • This step defines the program section of the cell.
  • This ion implantation step is used for implanting the p + regions of the Zener junction in order to form a junction at the drain side of the gate and also forms the substrate contacts.
  • the cross-section after this step is shown in Figure 11(j).
  • the target oxide thickness is " 3500 A. This oxide layer is used for the formation of the side wall spacers (SWS) .
  • the SWS are very important for the location of the p + /n + junction underneath the gate oxide 22. Experiments show that for this particular stacked gate structure and this range of LPCVD oxide thickness (" 3500 A) , the SWS width should be approximately 58% of the LPCVD oxide layer thickness. The cross-section after this step is shown in Figure 11(1).
  • the target oxide thickness is ⁇ 500 A. This oxide is used as a screen layer during the subsequent ion implantation which forms the source and the drain regions.
  • This photolithography process step defines windows through which n + source and drain regions are implanted.
  • the photoresist covers p + diffusion regions designed to serve as contacts to the substrate. The windows are open only over the device active area.
  • This ion implantation step is used for implanting the n + source and drain regions 19 and 17. The cross-section after this step is shown in Figure 11(m) .
  • the target oxide thickness is ⁇ 8000 A. This oxide is used as an isolation layer between the devices and the metal layer.
  • the p + implant step used to form the injector in the fabrication of this device is similar to the step used in the implementation of the n- LDD implant in a standard CMOS/BiCMOS process flow.
  • the flash E 2 PROM cell of the preferred embodiment can be configured in a cross point array architecture, as shown in Figures 12A and 12B.
  • the p + pocket implant 25 does not necessarily result in a larger cell size since it can be accommodated within the minimum device width. However, the cell exhibits a lower read current.
  • the p- buried layer 15A provides sufficient substrate current extraction to minimize interference with on-chip logic circuits.
  • the programming time of the flash E 2 PR0M of the present invention is dependent on the p + region doping levels, and has been found by computer simulation to be at least an order of magnitude less than that of an equivalent dimension conventional flash E 2 PROM cell operating under the same conditions, as shown in Figure 13.
  • the decrease in programming time with decreased channel length is mainly due to the reduction in gate area since the channel length has no effect on the generation of hot electrons.
  • a heavily doped p + /n + junction between the drain 17 and pocket implant 25 is preferred since breakdown occurs at lower voltages and higher electric field strength, resulting in faster programming speeds.
  • the programming time is on the order of 150 ns for 3.3V operation.
  • the device gate length was established as 3 ⁇ m
  • the tunnel oxide 22 thickness was 100 Angstroms
  • the interpoly oxide 24 was 300 Angstroms.
  • a heavy boron implant was used to form the p + region 25 with an effective doping concentration of approximately 1.8xl0 18 cm ⁇ 3 .
  • the measured and simulated programming times plotted in Figure 14 are seen to be in good agreement.
  • the programming time was measured to be 6 ⁇ s at a drain bias of 6.5V which is an order of magnitude smaller than that of an equivalent dimension prior art conventional flash E 2 PR0M cell operating under the same conditions.
  • the drain-to-substrate current was limited to 120 ⁇ A per ⁇ m of channel width, which is comparable to currents observed in conventional flash E 2 PR0M cells.
  • the erase time was measured at 100 milliseconds which is consistent with prior art device characteristics. Shorter erase times may be obtained by increasing the coupling ratio between the control and floating gates.
  • a measured read current of 100 ⁇ A per ⁇ m of device width was obtained from the memory cell in the erased state with +3 Volts and +5 Volts applied to the drain 17 and control gate 23, respectively, which is also consistent with prior art operating characteristics.
  • the pocket p + implant 25 may be located adjacent the source region 19, with minor modifications being necessary to the applied voltage in various modes of operation.
  • an improved version of these devices can be fabricated by using a p + buried layer in the substrate to collect the substrate current generated during the programming operation. All such modifications and alternatives are believed to be within the sphere and scope as defined by the claims appended hereto.

Abstract

In a flash E2PROM cell having source and drain regions disposed in a substrate, a channel region intermediate to the source and drain regions, a tunnel dielectric layer overlying the channel region, a floating gate overlying the tunnel dielectric layer, an inter-poly dielectric layer overlying the floating gate and a control gate overlying the inter-poly dielectric layer, the improvement comprising a highly doped p+ pocket implant covering a portion of the cell width and adjacent to least one of the source and drain regions. The flash E2PROM cell is comprised of two sections butted together. The portion (width-wise) covered by the highly doped p+ pocket implant is referred to as the program section and the remaining portion (width-wise) not covered by the highly doped p+ pocket implant resembles a conventional E2PROM cell and is referred to as the sense section. The highly doped p+ pocket implanted and the n+ drain and/or source regions create a junction having narrow depletion width such that in the event the junction is reverse biased, an electric field is created for generating hot electrons for storage on the floating gate, thereby programming the flash E2PROM cell when a high positive potential is applied to the control gate. The cell according to the present invention provides short programming time and low operating voltages as compared to prior art devices.

Description

METHOD OF FABRICATING A FAST PROGRAMMING FLASH E2PROM CELL
Field of the Invention
The present invention relates in general to Electrically Erasable and Programmable Read-Only Memory (E2PROM) , and more particularly to a method of fabricating a flash E2PROM with fast programming speed and low operating voltages.
Background of the Invention
Flash memory is a modified form of E2PROM which can be erased one block at a time and can be programmed one bit at a time. At the time of filing this application, flash memory chips are available in densities of up to 32M bits. Large data storage capacity with complete nonvolatility, results in numerous applications for such chips, ranging from cellular telephones, solid-state disks and memory cards. Such applications have traditionally been the domain of ROM, conventional E2PROM, battery-backed RAM, static RAM (SRAM) and magnetic storage.
A flash E2PROM cell resembles an ordinary MOS (Metal- Oxide-Semiconductor) transistor, except for the addition of a floating gate, which is buried in the insulator between the substrate and the conventional control gate. Charge stored on the floating gate alters the threshold voltage (Vth) of the device as measured at the control gate. Since the floating gate and the control gate are both stacked directly above the transistor channel, very high densities can be achieved. Even higher densities can be obtained by self aligning the control and floating gates to the source and drain regions.
The cell is programmed and erased by adding electrons to and removing electrons from the floating gate, respectively. A cell with a high threshold voltage (Vth) s in its "0" state. After erasure has been completed, the threshold voltage is reduced resulting in a "1" state. The conductivity of the channel determines the information stored in the memory cell (i.e. current flowing through the channel is detected by sense- amplifier circuitry as a "1", while the absence of current is detected as a "0") .
Flash E2PR0M technology has received industry-wide attention recently. Due to its simple single transistor cell architecture, flash memory may eventually cost less to make than DRAM (Dynamic Random Access Memory) .
The channel hot electron programming method, used in conventional flash E2PROM cells, requires biasing the device at high drain voltage (6 to 8V) to generate hot electrons. This results in additional circuit complexity and cost (ie. an additional external voltage supply) , which is a particular disadvantage in mobile applications such as cellular telephones, etc.
Furthermore, the channel hot electron injection method used for programming conventional flash E2PROM cells, generates very high lateral source-to-drain currents (in the range of milliamperes) . These high currents limit the number of cells that can be programmed at one time.
Existing flash E2PR0M cells also suffer from slow programming speed (~10/zs) which prevents their widespread application as a replacement for RAMs and electronic hard disks. If the device gate length is scaled down in order to reduce the programming time, then punchthrough between the source and drain is likely to occur.
Discussion of Prior Art A well known early prior art predecessor to modern day flash memory devices is the FAMOS (floating-gate avalanche-injection MOS) memory. This device is described in a number of publications, including: Dov Frohman-Bentchowsky, "A Fully Decoded 2048-Bit
Electrically Programmable FAMOS Read-Only Memory"; IEEE Journal of Solid state Circuits, Vol. SC-6, No. 5, October 1971, pp 301-306; Ron D. Katznelson and Dov Frohman-Bentchowsky, "An Erase Model for FAMOS EPROM Devices"; IEEE Transactions on Electron Devices, Vol. ED- 27, No. 9, September 1980, pp 1744-1752; and U.S. Patent No. 4,203,158 (Frohman-Bentchkowsky, et al) .
Another prior art flash memory device is described in Masatada Horiuchi and Hisao Katto,"FCAT-A Low-Voltage High Speed Alternative n-Channel Nonvolatile Memory Device"; Transactions on Electron Devices, Vol. ED-26, No. 6, June 1979, pp. 914-918. The FCAT cell has the p+ regions placed outside the channel area corners adjacent to the diffused n+ source and/or drain regions. This method of positioning the p+ regions results in a corner point injection for the hot electrons which contributes to acute stress in the tunnel oxide. Also, the FCAT cell uses hot hole injection into the thin gate oxide for erasure, which can lead to poor reliability and endurance characteristics.
U.S. Patent 5,464,785 entitled "Method of Making A Flash EPROM Device Having A Drain Edge P+ Implant", (Hong et al.), describes a flash EPROM structure with a drain edge p+ implant that utilizes the electric field at the junction between the drain and implant to enhance the generation and injection of hot channel electrons during programming operation. According to Hong et al., the drain edge p+ implant extends across the entire width of the EPROM cell, thereby preventing lateral current flow through the MOS channel between the source and drain. Therefore, the data stored in the cell cannot be sensed readily.
The fabrication steps suggested by Hong et. al. are impractical since it is very difficult to etch the narrow vertical trenches (reported typically to be lOOOA wide by 500θA high) used for the drain edge p+ implant. Also, the fabrication steps imply a dedicated process which is not CMOS compatible.
No read, write, and erase schemes are described in the Hong et al. patent, nor is any evidence of experimental performance indicated.
Summary of the invention
According to the present invention, a low voltage flash E2PROM cell is provided which overcomes the above- identified programming speed limitations of conventional prior art flash memory cells, and in addition allows the cell to operate at low voltages.
More particularly, the flash E PR0M cell is comprised of two sections which are butted together. One portion (width-wise) is covered by a highly doped p+ pocket implant adjacent to at least one of the n+ source and drain regions. This portion is referred to as the program section. The remaining portion (width-wise) is not covered by the highly doped p+ pocket implant and resembles a conventional E2PROM cell. This portion is referred to as the sense section.
The highly doped p+ pocket implant and the n+ drain and/or source regions create a junction having narrow depletion width such that when the junction is reverse biased an electric field is created for generating hot electrons for storage on the floating gate, thereby programming the flash E2PROM. The programming speed of the improved flash E2PROM cell of the present invention is in general an order of magnitude faster than a conventional flash E2PROM cell of the same dimension. The doping concentration of the p+ pocket implant is chosen such that the junction will breakdown at low voltages (typically in the range of 2.5 to 5V) , thus enabling the drain and/or source regions to be driven directly by the logic level supply voltages. This eliminates the need for large charge pump circuits with high current capacity as used in conventional flash E2PROMs. The enhancement of hot electron generation results in smaller drain and/or source currents (typically in the range of 50 to 200μA/μm of cell width) than prior art devices during the programming operation. Thus, the flash E2PROM cell of the present invention consumes much less power, and is more suitable for battery operated portable equipment than similar prior art devices.
High positive and negative voltages are needed only to drive the control gate. Consequently, they can be generated on-chip using simple charge pump circuitry with small current driving capability. The voltages required at the source and drain in all modes of operation can be obtained directly from a low power supply rail.
The flash E2PROM cell of the present inventi'on is compatible with CMOS and BiCMOS processes. More particularly, according to one aspect of the present invention, a method of fabricating an E2PROM cell is provided in which the fabrication steps are modular and compatible with conventional CMOS and BiCMOS process flows. Cell fabrication can be integrated into existing CMOS and BiCMOS processes with minimal changes. The resulting flash E2PROM cell is a self aligned structure. The p+ pocket implant is carried out in the same way as the n-LDD implant commonly used in modern CMOS and BiCMOS processes. The generation of hot electrons is dependent on the doping level of the p+/n+ junction rather than the size of the channel length. The electric field at this junction ( " 106 V/cm) significantly exceeds the maximum lateral electric field ( " IO5 V/cm) available in conventional flash E2PROM cells with short channel lengths (0.3 to 0.5μm), thus ensuring very efficient injection of hot electrons into the floating gate.
Brief Introduction to the Drawings
Detailed descriptions of one prior art flash memory cell and of flash E2PROM cells according to a first embodiment and a preferred embodiment of the present invention are provided herein below, with reference to the following drawings, in which:
Figure 1 is a schematic representation of a prior art flash E2PROM cell;
Figure 2 is a schematic representation of hot electron injection programming of the prior art flash memory cell depicted in Figure 1;
Figure 3 is a schematic representation of erasure of the prior art flash memory cell depicted in Figure 1 using Fowler-Nordheim tunnelling across the thin gate oxide;
Figure 4 is a schematic representation of the read operation of the prior art flash memory cell depicted in Figure 1;
Figure 5 is a three-dimensional schematic representation of a flash E2PR0M cell according to a first embodiment of the present invention; Figure 6 is a three-dimensional schematic representation of Zener and/or avalanche breakdown-based electron injection programming of the flash E2PROM cell of Figure 5;
Figure 7 is a three-dimensional schematic representation of the read operation of the flash E2PROM cell of Figure 5;
Figure 8 is a three-dimensional schematic representation of the erasure operation of the flash E2PROM cell of Figure 5;
Figure 9 is a three-dimensional schematic representation of the preferred embodiment of flash E2PROM cell according to the present invention;
Figure 10 is a plan view of the device shown in Figure 9;
Figures 11(a) through 11(r) depict the process flow steps for fabricating the flash E2PROM cell according to the present invention;
Figures 12A and 12B show a circuit diagram and cross-sectional diagram, respectively, of the flash E2PROM cell of the preferred embodiment configured in a cross point array architecture;
Figure 13 shows a simulation of programming time as a function of channel length for different p+ doping concentrations for the preferred embodiment of Figure 9;
Figure 14 shows simulated and measured programming characteristics for the preferred embodiment of Figure 9; and Figure 15 shows the experimental effect of the p+/n+ junction on the disturb characteristics of a half- selected programmed and erased E2PR0M cell according to the preferred embodiment of Figure 9.
Detailed Description of the Prior Art ETOX™ Memory Cell
Turning to Figure 1, a flash E2PROM cell is shown (ETOX™ cell by Intel Corp.), having a floating gate 1 and a control gate 3 stacked vertically above the MOFSET channel 5 in substrate 6 intermediate the source 7 and drain 9. A tunnel oxide 11 is provided intermediate the floating gate l and channel 5, and an inter-poly oxide 13 is provided intermediate the floating gate 1 and control gate 3.
In order to program the ETOX™ flash memory cell of Figure 1, a voltage of +12V is applied to the control gate 3, the source 7 is grounded, and approximately +6V is applied between source 7 and drain 9 as shown in Figure 2. As discussed above, the control gate 3 is capacitively coupled to the floating gate 1.
The source-drain voltage (+6V) generates hot electrons that are swept across the channel 5 from source 7 to drain 9. These hot electrons collide with atoms in the channel 5 thereby creating additional hot electrons. The high voltage (+12V) on control gate 3 attracts the hot electrons across the thin tunnel oxide 11 where they accumulate on the floating gate 1. When enough electrons have accumulated, the cell switches from its "1" (erased) state to its "0" (programmed) state.
As the floating gate 1 becomes fully charged, the current across the oxide 11 reduces almost to zero as a result of the electric field in the oxide 11 becoming repulsive to additional electron injection from the high electric field region at the drain 9. Therefore, electron injection during programming is a self limiting process.
The voltage applied to the drain 9 does not, in general, have a significant effect on the programmed threshold voltage of the EEPROM cell. However, it does affect the speed of programming of the device.
To erase a flash memory cell, Fowler-Nordheim tunnelling is used to remove electrons from the floating gate 1. More particularly, by floating the drain 9, grounding the control gate 3 and applying +12V to the source 7, an electric field is generated across the thin tunnel oxide 11 between the floating gate 1 and the source 7. This electric field attracts electrons off of the floating gate 1 toward the source 7, as shown in Figure 3.
When all of the excess electrons have been removed from the floating gate 1, the threshold voltage (Vth) returns to its initial value thereby resetting the cell to a "1" state.
During reading, the prior art ETOX™ memory cell operates like an ordinary transistor. When the flash memory is read, address inputs select specific transistors within the memory cell array. Supply-voltage levels are applied to the control gate 3 and drain 9, while the source 7 is grounded as shown in Figure 4. By sensing the amount of current that is conducted in the channel 5 between the source 7 and the drain 9, it is possible to determine if the cell has been programmed or erased, as discussed above.
Detailed Description of a First Embodiment of Flash E2PROM Cell
Turning now to Figure 5, a flash E2PROM cell is shown for achieving simultaneous reduction of programming time and operating voltages over the prior art cell of Figure 1.
The cell according to the present invention comprises a substrate 15, drain and source regions 17 and 19, respectively, a floating gate 21 and control gate 23, in the usual manner. Tunnel and inter-poly dielectric layers 22 and 24 are also provided in the usual manner. However, in contrast with the prior art ETOX™ flash memory cell, additional highly doped regions 25 and 27 are provided adjoining the drain and/or source regions 17 and/or 19, respectively. The additional highly doped regions 25 and 27 (occasionally referred to herein as pocket implants) extend along a portion of the width of the cell structure identified as the program section 29, the remaining sense section 31 being provided for the read operation of the cell, as discussed in greater detail below.
In the illustrated embodiment, the substrate is made of p-type material, the drain and source regions are fabricated from n+-type material, and the additional highly doped regions 25 and 27 comprise p+-type regions.
According to the present invention, programming is accomplished via low voltage Zener and/or avalanche breakdown at the reverse biased p+/n+ junctions in the program section 29. High doping concentrations at these junctions results in a narrow depletion width so that only a small voltage is required to generate a sufficiently strong electric field to create hot electrons for storage on the floating gate 21. The breakdown mechanism which dominates and the applied voltage required for breakdown depend on the doping of the p+ regions. Computer simulation results have shown that p+ doping concentrations in the range of lxl018cm"3 to 5xl019cm"3 are suitable for the implementation of this structure.
As indicated above, the cell is programmed and read from the program and sense sections 29 and 31, respectively, and erased by Fowler-Nordheim tunnelling of electrons from the floating gate 21 to the source 19.
With reference to Figure 6, a schematic representation of the programming mode of operation is provided. In order to program a logic "0" into the cell, sufficient electrons must be stored on the floating gate 21 to raise the transistor threshold voltage (Vth) so that the normal read voltage (e.g. 5 Volts) applied to control gate 23 is insufficient to turn the transistor on.
Accordingly, low positive voltages (less than 5V) are applied to the drain 17 and source 19 while a high programming voltage (in the range of 10 to 15 Volts) is applied to control gate 23. The substrate 15 is grounded. Low voltage Zener and/or avalanche breakdown at the reversed biased p+/n+ junctions, results in the generation of hot electrons which are drawn to the floating gate 21 as a result of the high voltage applied to control gate 23.
In contrast with the prior art ETOX™ memory cell, the generation of hot electrons necessary to program the flash memory cell of the present invention does not require a large drain-to-source current flow. This is a result of the high electric field (e.g. 106V/cm) generated in the narrow depletion regions 26 associated with the heavily doped p+/n+ junctions. Figure 7 schematically shows the read mode operation of the flash memory cell according to the present invention. In operation, a low read voltage (e.g. +5 Volts) is applied to control gate 23; a voltage lower than the breakdown voltage of the p+/n+ junction is applied to the drain 17 (e.g. 13 volts), and the source 19 and substrate 15 are grounded. Operation in the read mode is essentially the same as discussed above with respect to the ETOX™ prior art memory cell. In particular, in an erased cell, the voltage applied to control gate 23 is sufficient to overcome the transistor turn-on threshold voltage (Vth) , and the drain-to-source current is detected in the sense section 31 by sense amplifier circuitry (not shown) and translated into a logic "1". Conversely, in a programmed cell, the added electrons stored on the floating gate 21 raise the transistor turn-on threshold voltage (Vth) so that the read voltage applied to control gate 23 is insufficient to turn on the transistor. The absence of current is detected in the sense section 31 as a logic "0".
Erasure is performed by applying a high negative voltage (in the range of -12 to -15 Volts) to the control gate 23, applying a voltage, lower than the breakdown voltage of the p+/n+ junction (e.g. +3.5 volts), to the source 17, while grounding the substrate 15 and leaving the drain in an open circuit condition, as shown in Figure 8. The voltage difference between the control gate 23 and the source 19, creates a strong electric field, across the thin oxide 22 between the floating gate 21 and the source 19, which in turn causes the removal of electrons from the floating gate 21 by Fowler-Nordheim tunnelling.
In summary, the flash E2PROM memory cell according to the present invention, with appropriate p+ dopings and a short gate length, results in extremely fast programming of the cell (in the order of nanoseconds) at low supply voltages, making the cell particularly applicable in the field of portable electronics.
Description of a Preferred Embodiment of Flash E2PROM Cell
The preferred cell structure according to the present invention uses a single p+/n+ junction in the programming section of the drain region of the transistor, as illustrated in Figures 9 and 10. This embodiment offers the advantage that the erasing voltage is not limited by the p+/n+ junction breakdown voltage, whereby the erasing speed can be improved by applying higher voltages to the source. The cell is programmed from the drain side of the program section, erased from the source side, and read from the sense section of the cell, according to the principles discussed above in connection with Figures 5-8.
An improved method of reading the preferred embodiment of flash E2PROM cell shown in Figures 9 and 10, comprises the steps of grounding the substrate 15 and drain 17, applying a positive logic level voltage to the control gate 23, and applying a voltage of no greater than logic level to the source region 19. The voltage applied to the control gate 23 allows current to flow through the channel region in the sense section, from the source region to the drain region in the event that little or no charge is stored on the floating gate.
This eliminates possible band to band tunnelling leakage current at the junction of the p+ pocket implant 25 and the n+ drain 17, thus reducing the probability of sensing error on the stored data. The voltage applied to the source region 19 can be higher than the breakdown voltage (typically 3V or less) of the p+ pocket implant/ n+ drain junction, resulting in a higher current flow through the channel during read operation. This will allow faster read time and lower sensing error rate.
Process Flow for Fabricating Flash E2PROM
A detailed description of the flash EEPROM cell fabrication process is given below. All of the steps described are optimized to achieve the appropriate specifications for the flash cells according to the present invention. The wafers used in the fabrication process are p-type (boron doped) with resistivity of 1- 1.5 Ohm-cm and orientation <100>. The steps are as follows, with reference to Figures 11(a) through 11(r):
1. Grow field oxide on the silicon wafers, to a target oxide thickness of "5000 - 700θA.
2. Define the device active areas using photolithography (mask #1) . The cross-section after this step is shown in Figure 11(a).
3. Etch the oxide to form the device active area.
4. Remove the photoresist. The cross-section after this step is shown in Figure 11(b).
5. Grow the tunnelling or gate oxide 22, to a target oxide thickness of lOOA.
Implant boron (energy E=30 keV, dose QI=4xl012 cm-2) for threshold voltage adjustment. The cross-section after this step is shown in Figure 11(c) .
Deposit amorphous silicon by low pressure chemical vapor deposition (LPCVD) , the target thickness is 3600 A. The reason for using amorphous silicon is that it results in a smooth surface which is important for the interpoly dielectric growth discussed in greater detail below.
8. Oxidize the amorphous silicon layer, to a target oxide thickness of " 200 A. During this step, the amorphous silicon is converted to polysilicon. The thin oxide is used as a screen oxide during the subsequent ion implantation of phosphorus.
9. Implant phosphorus (energy E=30 keV, dose QI=8xl015 cm-2) . This ion implantation step is used for setting the doping of the polysilicon. The cross-section after this step is shown in Figure 11(d) .
10. Perform a rapid thermal anneal to activate the phosphorus and redistribute in the polysilicon after ion implantation.
11. Define the first polysilicon layer using photolithography (mask #2) . The cross-section after this step is shown in Figure 11(e).
12. Etch the thin screen oxide and polysilicon, remove the photoresist and etch the remaining areas of thin oxide.
13. The oxide-nitride-oxide (ONO) interpoly dielectric is grown in this step. The target equivalent thickness is 200 A. The cross- section after this step is shown in Figure
11(f). 14. Deposit amorphous silicon by LPCVD to form the control gate 23. The target thickness is 3600 A.
15. Oxidize the deposited amorphous silicon layer, to a target oxide thickness of "200 A. During this step, the amorphous silicon is converted to polysilicon. The thin oxide is used as a screen oxide during the subsequent ion implantation of phosphorus.
16. Implant phosphorus (energy E=30 keV, dose QI=8xl015 cm-2) . This ion implantation step is used for setting the doping of the polysilicon. The cross-section after this step is shown in
Figure 11(g) .
17. Deposit LPCVD oxide. The target thickness is "3500 A. This oxide serves as a mask in Reactive Ion Etching (RIE) of the polysilicon layers.
18. Perform a rapid thermal anneal step to activate the phosphorus implanted in the polysilicon and densify the LPCVD oxide.
19. Define the stacked gate area over LPCVD oxide using photolithography (mask #3) . The cross- section after this step is shown in Figure 11(h).
20. Etch the LPCVD oxide using RIE.
21. Remove the photoresist.
22. Perform a RIE of the second polysilicon layer. The control gate 23 is formed in this step. 23. Perform a RIE of the undesired areas of interpoly oxide 24 and a RIE of the undesired areas of the first polysilicon layer. The floating gate is formed in this step. The remaining undesired areas of the first polysilicon are removed from the source and drain regions. The cross-section after this step is shown in Figure 11(i).
24. Perform a dry oxidation to ensure that the floating gate is encapsulated by a high quality dielectric layer.
25. Deposit thin LPCVD oxide. The target oxide thickness is ~ 500 A. This oxide is used as a screen layer during the ion implantation to form the substrate contacts and the P+ region of the Zener injector.
26. Define the windows for boron implantation using photolithography (mask #4) . This photolithography process step opens a window through which the p+ region of the Zener junction and the substrate contacts are formed.
This step defines the program section of the cell.
27. Implant boron (energy E=40keV, dose QI=2xl015cm-2) . This ion implantation step is used for implanting the p+ regions of the Zener junction in order to form a junction at the drain side of the gate and also forms the substrate contacts. The cross-section after this step is shown in Figure 11(j).
28. Remove the photoresist. 29. Etch the thin screen oxide. A wet oxide etch is used in this step.
30. Deposit LPCVD oxide. The target oxide thickness is " 3500 A. This oxide layer is used for the formation of the side wall spacers (SWS) .
31. Anneal the wafers to density the LPCVD oxide and to drive-in boron after the implantation.
The cross-section after this step is shown in Figure 11(k) .
32. Etch the deposited LPCVD oxide layer. The side wall spacers (SWS) are formed in this step.
Perform a RIE of oxide in the process. The SWS are very important for the location of the p+/n+ junction underneath the gate oxide 22. Experiments show that for this particular stacked gate structure and this range of LPCVD oxide thickness (" 3500 A) , the SWS width should be approximately 58% of the LPCVD oxide layer thickness. The cross-section after this step is shown in Figure 11(1).
33. Deposit thin LPCVD oxide. The target oxide thickness is ~ 500 A. This oxide is used as a screen layer during the subsequent ion implantation which forms the source and the drain regions.
34. Define windows for phosphorus implantation using photolithography (mask #5) . This photolithography process step defines windows through which n+ source and drain regions are implanted. In this step the photoresist covers p+ diffusion regions designed to serve as contacts to the substrate. The windows are open only over the device active area.
35. Implant phosphorus (energy E=80 keV, dose QI=8xl015 cm-2) . This ion implantation step is used for implanting the n+ source and drain regions 19 and 17. The cross-section after this step is shown in Figure 11(m) .
36. Remove the photoresist, and etch the thin screen oxide.
37. Deposit LPCVD oxide. The target oxide thickness is ~ 8000 A. This oxide is used as an isolation layer between the devices and the metal layer.
38. Perform a thermal drive-in to form the metallurgical junction of the p+/n+ injector to be directly underneath the floating gate electrode (on the drain side) .
39. Define the contact windows before aluminum deposition using photolithography (mask #6) . This photolithography process step defines metal to diffusion and polysilicon contacts.
40. Etch oxide to define contact windows. The cross-section after this step is shown in Figure 11(n) .
41. Remove the photoresist.
42. Sputter aluminum on the wafers. The cross- section after this step is shown in Figure ll(o).
43. Define the aluminum using photolithography (mask #7) . The cross-section after this step is shown in Figure 11(p) .
44. Etch the aluminum.
45. Remove the photoresist.
46. Metal sintering is done in forming gas. The final cross-section of the cell's program section and sense section are shown in Figures
11(q) and in Figure 11(r) respectively.
It should be noted that the p+ implant step used to form the injector in the fabrication of this device is similar to the step used in the implementation of the n- LDD implant in a standard CMOS/BiCMOS process flow.
Application of the Flash E2PROM of the Present Invention
The flash E2PROM cell of the preferred embodiment can be configured in a cross point array architecture, as shown in Figures 12A and 12B. The p+ pocket implant 25 does not necessarily result in a larger cell size since it can be accommodated within the minimum device width. However, the cell exhibits a lower read current. The p- buried layer 15A provides sufficient substrate current extraction to minimize interference with on-chip logic circuits.
Experimental and Test Results
The programming time of the flash E2PR0M of the present invention is dependent on the p+ region doping levels, and has been found by computer simulation to be at least an order of magnitude less than that of an equivalent dimension conventional flash E2PROM cell operating under the same conditions, as shown in Figure 13. The decrease in programming time with decreased channel length is mainly due to the reduction in gate area since the channel length has no effect on the generation of hot electrons. A heavily doped p+/n+ junction between the drain 17 and pocket implant 25 is preferred since breakdown occurs at lower voltages and higher electric field strength, resulting in faster programming speeds. With an appropriate p+ doping concentration of >1019cm"3 and a short channel length (e.g. <«0.5 μm) , the programming time is on the order of 150 ns for 3.3V operation.
Successful prototypes of the flash E2PROM memory cell in accordance with the preferred embodiment have been fabricated using a conventional CMOS process flow as discussed in detail above with reference to Figures 11(a) through 11(r). According to the successful prototypes, the device gate length was established as 3μm, the tunnel oxide 22 thickness was 100 Angstroms and the interpoly oxide 24 was 300 Angstroms. A heavy boron implant was used to form the p+ region 25 with an effective doping concentration of approximately 1.8xl018cm~3.
The measured and simulated programming times plotted in Figure 14 are seen to be in good agreement. For a 3μm channel length with p+ doping concentration of 1.8xl018cm"3, the programming time was measured to be 6μs at a drain bias of 6.5V which is an order of magnitude smaller than that of an equivalent dimension prior art conventional flash E2PR0M cell operating under the same conditions. The drain-to-substrate current was limited to 120 μA per μm of channel width, which is comparable to currents observed in conventional flash E2PR0M cells. The erase time was measured at 100 milliseconds which is consistent with prior art device characteristics. Shorter erase times may be obtained by increasing the coupling ratio between the control and floating gates. Furthermore, a measured read current of 100 μA per μm of device width was obtained from the memory cell in the erased state with +3 Volts and +5 Volts applied to the drain 17 and control gate 23, respectively, which is also consistent with prior art operating characteristics.
The effect of the p+/n+ junction on the disturbed characteristics of half-selected cells is illustrated in Figure 15. The drain disturbance for programmed cells becomes apparent if the bit-lines are raised high for more than one second. This represents a significant safety margin for sequential programming in a large array. The write/erase endurance characteristics of the cells were also measured to be larger than 10,000 cycles.
Other embodiments and modifications of the invention are possible. For example, as a further alternative to the preferred embodiment of Figure 9, the pocket p+ implant 25 may be located adjacent the source region 19, with minor modifications being necessary to the applied voltage in various modes of operation. Also, an improved version of these devices can be fabricated by using a p+ buried layer in the substrate to collect the substrate current generated during the programming operation. All such modifications and alternatives are believed to be within the sphere and scope as defined by the claims appended hereto.

Claims

WE CLAIM
1. In a flash E2PROM cell having n+ source and drain regions disposed in a p-type substrate, a channel region intermediate to said source and drain regions, a tunnel dielectric layer overlying said channel region, a floating gate overlying said tunnel dielectric layer, an inter-poly dielectric layer overlying said floating gate and a control gate overlying said inter-poly dielectric layer, the improvement comprising a highly doped p+ pocket implant adjoining one of said drain and source regions along a portion of the width of said cell for creating a metalurgical junction having narrow depletion width, said portion of the width of said cell defining a program section of said cell and the remaining width of said cell defining a sense section thereof.
2. The improvement of claim 1, wherein said p+-type material has a doping concentration in the range of from lxlO18 cm"3 to lxlO20 cm"3.
3. The improvement of claim 1 or 2, wherein said metallurgical junction is driven in so as to be positioned directly beneath said floating gate.
4. A method of programming the improved flash E2PROM cell of claim 1, comprising the steps of: a) grounding said substrate; b) applying a low positive voltage to said one of said drain and source regions adjoining said p+ pocket implant, whereby said junction becomes reverse biased and one of either low voltage Zener or avalanche breakdown occurs resulting in generation of hot electrons; and c) applying a high positive voltage to said control gate, whereby said hot electrons are attracted to said floating gate.
5. A method of erasing the improved flash E2PROM cell of claim 1, comprising the steps of: a) grounding said substrate; b) applying a logic level voltage to the other of said drain and source regions; c) leaving said one of said drain and source regions adjoining the p+pocket implant in an open circuit condition; and d) applying a high negative voltage to said control gate, whereby a strong electric field is generated between said floating gate and said other of said drain and source regions resulting in removal of electrons from said floating gate to said other of said drain and source regions by Fowler-Nordheim tunnelling.
6. A method of reading the improved flash E2PROM cell of claim 1, comprising the steps of: a) grounding said substrate and said source region; b) applying a low voltage to said control gate; and c) applying a voltage less than breakdown voltage of said junction to said drain region, whereby said voltage applied to the control gate causes current to flow through said channel region in said sense section from said drain region to said source region.
7. A method of reading the improved flash E2PROM cell of claim 1, comprising the steps of: a) grounding said substrate and said one of said drain and source regions adjoining said p+ pocket implant; b) applying a positive logic level voltage to said control gate; c) applying a voltage of no greater than logic level to said other one of said source and drain regions, whereby said voltage applied to the control gate allows current to flow through said channel region in said sense section, from said source region to said drain region in the event that little or no charge is stored on said floating gate.
8. In a flash E2PROM cell having n+ source and drain regions disposed in a p-type substrate, a channel region intermediate to said source and drain regions, a tunnel dielectric layer overlying said channel region, a floating gate overlying said tunnel dielectric layer, an inter-poly dielectric layer overlying said floating gate and a control gate overlying said inter-poly dielectric layer, the improvement comprising a pair of highly doped p+ pocket implants adjoining respective ones of said drain and source regions along a portion of the width of said cell for creating a pair of metalurgical junctions each having narrow depletion width, said portion of the width of said cell defining a program section of said cell and the remaining width of said cell defining a sense section thereof.
9. The improvement of claim 8, wherein said p+-type material has a doping concentration in the range of from lxlO18 cm"3 to lxlO20 cm"3.
10. The improvement of claim 8 or 9, wherein said metallurgical junctions are driven in so as to be positioned directly beneath said floating gate.
11. A method of programming the improved flash E2PROM cell of claim 8, comprising the steps of: a) grounding said substrate; b) applying a low positive voltage to each of said drain and source regions, whereby said junctions become reverse biased and one of either low voltage Zener or avalanche breakdown occurs resulting in generation of hot electrons; and c) applying a high positive voltage to said control gate, whereby said hot electrons are attracted to said floating gate.
12. A method of erasing the improved flash E2PROM cell of claim 8, comprising the steps of: a) grounding said substrate; b) applying a voltage less than breakdown voltage of said junction to one of said drain and source regions; c) leaving the other of said drain and source regions in an open circuit condition; and d) applying a high negative voltage to said control gate, whereby a strong electric field is generated between said floating gate and said one of said drain and source regions resulting in removal of electrons from said floating gate to said one of said drain and source regions by Fowler-Nordheim tunnelling.
13. A method of reading the improved flash E2PROM cell of claim 8, comprising the steps of: a) grounding said substrate and said source region; b) applying a low voltage to said control gate; and c) applying a voltage less than breakdown voltage of said junction to said drain region, whereby said voltage applied to the control gate causes current to flow through said channel region in said sense section from said drain region to said source region.
14. A method of fabricating a flash E2PR0M cell, comprising the steps of: a) growing a gate oxide layer on a p-type substrate; b) depositing amorphous silicon on said gate oxide layer; c) oxidizing said amorphous silicon to create polysilicon; d) doping said polysilicon to create a floating gate layer ; e) growing an interpoly dielectric layer over said floating gate layer; f) depositing a second amorphous silicon layer on said interpoly dielectric layer; g) oxidizing said second amorphous silicon layer to create a second polysilicon layer; h) doping said second polysilicon layer to create a control gate layer; i) masking and etching said gate oxide layer, floating gate layer, interpoly dielectyric layer and control gate layer to form a defined stack of gate oxide, floating gate, interpoly dielectric and control gate, respectively; j) doping said substrate adjacent one side of said stack with a p+ impurity, thereby creating a pocket implant region along a program region forming a first portion of the width of said cell, a sense region forming the remainder of the width of said cell; k) extending an oxide side wall spacer from both sides of said stack;
1) doping said substrate with an n+ impurity, for creating source and drain regions, such that a narrow depletion width metallurgical p+/n+ junction is created between said pocket implant region and said drain region within said program region of said cell upon application of a reverse bias voltage; m) performing a thermal drive-in for locating said metallurgical p+/n+ junction entirely beneath said stack; and n) depositing and patterning metal contacts for said substrate, said control gate and said source and drain regions.
15. The method of claim 14, further comprising the step of doping a buried layer in said substrate with an impurity of said first type, for collecting substrate current during programming of said cell.
PCT/CA1996/000446 1995-07-03 1996-07-03 Method of fabricating a fast programming flash e2prom cell WO1997002605A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
AU61851/96A AU6185196A (en) 1995-07-03 1996-07-03 Method of fabricating a fast programming flash e2prm cell
US08/981,745 US6034896A (en) 1995-07-03 1996-07-03 Method of fabricating a fast programmable flash E2 PROM cell

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US104695P 1995-07-03 1995-07-03
US60/001,046 1995-07-03

Publications (1)

Publication Number Publication Date
WO1997002605A1 true WO1997002605A1 (en) 1997-01-23

Family

ID=21694126

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CA1996/000446 WO1997002605A1 (en) 1995-07-03 1996-07-03 Method of fabricating a fast programming flash e2prom cell

Country Status (4)

Country Link
US (1) US6034896A (en)
AU (1) AU6185196A (en)
CA (1) CA2226015A1 (en)
WO (1) WO1997002605A1 (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000036642A1 (en) * 1998-12-18 2000-06-22 Lattice Semiconductor Corporation Method of forming a non-volatile memory device
WO2000038240A1 (en) * 1998-12-21 2000-06-29 Lattice Semiconductor Corporation Floating gate memory cell structure with programming mechanism outside the read path
WO2000038245A1 (en) * 1998-12-21 2000-06-29 Lattice Semiconductor Corporation Dual pocket, two sided program/erase non-volatile memory cell
WO2000039805A1 (en) * 1998-12-23 2000-07-06 Lattice Semiconductor Corporation Floating gate memory apparatus and method for selected programming thereof
WO2000041216A2 (en) * 1999-01-07 2000-07-13 Lattice Semiconductor Corporation Pmos avalanche programmed floating gate memory cell structure
US6294809B1 (en) 1998-12-28 2001-09-25 Vantis Corporation Avalanche programmed floating gate memory cell structure with program element in polysilicon
US6326663B1 (en) 1999-03-26 2001-12-04 Vantis Corporation Avalanche injection EEPROM memory cell with P-type control gate
WO2002037550A1 (en) * 2000-10-30 2002-05-10 Advanced Micro Devices, Inc. Non-volatile memory with source side boron implantation
US6424000B1 (en) 1999-05-11 2002-07-23 Vantis Corporation Floating gate memory apparatus and method for selected programming thereof
EP1145279B1 (en) * 1999-01-14 2008-04-09 Infineon Technologies AG Semiconductor element with a tungsten oxide layer and method for its production
US7535053B2 (en) * 1997-11-18 2009-05-19 Semiconductor Energy Laboratory Co., Ltd. Nonvolatile memory and electronic apparatus

Families Citing this family (75)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
IL125604A (en) * 1997-07-30 2004-03-28 Saifun Semiconductors Ltd Non-volatile electrically erasable and programmble semiconductor memory cell utilizing asymmetrical charge
US6768165B1 (en) * 1997-08-01 2004-07-27 Saifun Semiconductors Ltd. Two bit non-volatile electrically erasable and programmable semiconductor memory cell utilizing asymmetrical charge trapping
JP4236722B2 (en) * 1998-02-05 2009-03-11 株式会社半導体エネルギー研究所 Method for manufacturing semiconductor device
US6348711B1 (en) * 1998-05-20 2002-02-19 Saifun Semiconductors Ltd. NROM cell with self-aligned programming and erasure areas
US6215148B1 (en) * 1998-05-20 2001-04-10 Saifun Semiconductors Ltd. NROM cell with improved programming, erasing and cycling
US6281089B1 (en) * 1999-09-14 2001-08-28 Worldwide Semiconductor Manufacturing Corp. Method for fabricating an embedded flash memory cell
US6429063B1 (en) 1999-10-26 2002-08-06 Saifun Semiconductors Ltd. NROM cell with generally decoupled primary and secondary injection
US6396741B1 (en) 2000-05-04 2002-05-28 Saifun Semiconductors Ltd. Programming of nonvolatile memory cells
US6928001B2 (en) * 2000-12-07 2005-08-09 Saifun Semiconductors Ltd. Programming and erasing methods for a non-volatile memory cell
US6490204B2 (en) 2000-05-04 2002-12-03 Saifun Semiconductors Ltd. Programming and erasing methods for a reference cell of an NROM array
US6614692B2 (en) * 2001-01-18 2003-09-02 Saifun Semiconductors Ltd. EEPROM array and method for operation thereof
US6584017B2 (en) 2001-04-05 2003-06-24 Saifun Semiconductors Ltd. Method for programming a reference cell
US7098107B2 (en) * 2001-11-19 2006-08-29 Saifun Semiconductor Ltd. Protective layer in memory device and method therefor
US6583007B1 (en) 2001-12-20 2003-06-24 Saifun Semiconductors Ltd. Reducing secondary injection effects
TW518725B (en) * 2002-01-23 2003-01-21 Macronix Int Co Ltd Virtual ground flash memory
US6975536B2 (en) * 2002-01-31 2005-12-13 Saifun Semiconductors Ltd. Mass storage array and methods for operation thereof
US6700818B2 (en) * 2002-01-31 2004-03-02 Saifun Semiconductors Ltd. Method for operating a memory device
US7190620B2 (en) * 2002-01-31 2007-03-13 Saifun Semiconductors Ltd. Method for operating a memory device
US6614694B1 (en) * 2002-04-02 2003-09-02 Macronix International Co., Ltd. Erase scheme for non-volatile memory
US6917544B2 (en) 2002-07-10 2005-07-12 Saifun Semiconductors Ltd. Multiple use memory chip
US6826107B2 (en) * 2002-08-01 2004-11-30 Saifun Semiconductors Ltd. High voltage insertion in flash memory cards
US7136304B2 (en) 2002-10-29 2006-11-14 Saifun Semiconductor Ltd Method, system and circuit for programming a non-volatile memory array
US6963505B2 (en) * 2002-10-29 2005-11-08 Aifun Semiconductors Ltd. Method circuit and system for determining a reference voltage
US6992932B2 (en) 2002-10-29 2006-01-31 Saifun Semiconductors Ltd Method circuit and system for read error detection in a non-volatile memory array
US6967896B2 (en) * 2003-01-30 2005-11-22 Saifun Semiconductors Ltd Address scramble
US7178004B2 (en) 2003-01-31 2007-02-13 Yan Polansky Memory array programming circuit and a method for using the circuit
US7142464B2 (en) * 2003-04-29 2006-11-28 Saifun Semiconductors Ltd. Apparatus and methods for multi-level sensing in a memory array
US6954393B2 (en) * 2003-09-16 2005-10-11 Saifun Semiconductors Ltd. Reading array cell with matched reference cell
US7123532B2 (en) 2003-09-16 2006-10-17 Saifun Semiconductors Ltd. Operating array cells with matched reference cells
US7151692B2 (en) * 2004-01-27 2006-12-19 Macronix International Co., Ltd. Operation scheme for programming charge trapping non-volatile memory
WO2005094178A2 (en) * 2004-04-01 2005-10-13 Saifun Semiconductors Ltd. Method, circuit and systems for erasing one or more non-volatile memory cells
US7133313B2 (en) * 2004-04-26 2006-11-07 Macronix International Co., Ltd. Operation scheme with charge balancing for charge trapping non-volatile memory
US7164603B2 (en) * 2004-04-26 2007-01-16 Yen-Hao Shih Operation scheme with high work function gate and charge balancing for charge trapping non-volatile memory
US7187590B2 (en) * 2004-04-26 2007-03-06 Macronix International Co., Ltd. Method and system for self-convergent erase in charge trapping memory cells
US7075828B2 (en) * 2004-04-26 2006-07-11 Macronix International Co., Intl. Operation scheme with charge balancing erase for charge trapping non-volatile memory
US7209390B2 (en) * 2004-04-26 2007-04-24 Macronix International Co., Ltd. Operation scheme for spectrum shift in charge trapping non-volatile memory
US7166876B2 (en) * 2004-04-28 2007-01-23 Taiwan Semiconductor Manufacturing Company, Ltd. MOSFET with electrostatic discharge protection structure and method of fabrication
US7366025B2 (en) * 2004-06-10 2008-04-29 Saifun Semiconductors Ltd. Reduced power programming of non-volatile cells
US7190614B2 (en) * 2004-06-17 2007-03-13 Macronix International Co., Ltd. Operation scheme for programming charge trapping non-volatile memory
US20060007732A1 (en) * 2004-07-06 2006-01-12 Macronix International Co., Ltd. Charge trapping non-volatile memory and method for operating same
US7106625B2 (en) * 2004-07-06 2006-09-12 Macronix International Co, Td Charge trapping non-volatile memory with two trapping locations per gate, and method for operating same
US7095655B2 (en) * 2004-08-12 2006-08-22 Saifun Semiconductors Ltd. Dynamic matching of signal path and reference path for sensing
US20060068551A1 (en) * 2004-09-27 2006-03-30 Saifun Semiconductors, Ltd. Method for embedding NROM
US7638850B2 (en) 2004-10-14 2009-12-29 Saifun Semiconductors Ltd. Non-volatile memory structure and method of fabrication
US7133317B2 (en) * 2004-11-19 2006-11-07 Macronix International Co., Ltd. Method and apparatus for programming nonvolatile memory
US20060113586A1 (en) * 2004-11-29 2006-06-01 Macronix International Co., Ltd. Charge trapping dielectric structure for non-volatile memory
US7242618B2 (en) * 2004-12-09 2007-07-10 Saifun Semiconductors Ltd. Method for reading non-volatile memory cells
US8482052B2 (en) 2005-01-03 2013-07-09 Macronix International Co., Ltd. Silicon on insulator and thin film transistor bandgap engineered split gate memory
US7473589B2 (en) * 2005-12-09 2009-01-06 Macronix International Co., Ltd. Stacked thin film transistor, non-volatile memory devices and methods for fabricating the same
US7315474B2 (en) * 2005-01-03 2008-01-01 Macronix International Co., Ltd Non-volatile memory cells, memory arrays including the same and methods of operating cells and arrays
CN1838323A (en) * 2005-01-19 2006-09-27 赛芬半导体有限公司 Methods for preventing fixed pattern programming
US8053812B2 (en) 2005-03-17 2011-11-08 Spansion Israel Ltd Contact in planar NROM technology
US7158420B2 (en) * 2005-04-29 2007-01-02 Macronix International Co., Ltd. Inversion bit line, charge trapping non-volatile memory and method of operating same
US20070141788A1 (en) * 2005-05-25 2007-06-21 Ilan Bloom Method for embedding non-volatile memory with logic circuitry
US8400841B2 (en) * 2005-06-15 2013-03-19 Spansion Israel Ltd. Device to program adjacent storage cells of different NROM cells
US7184313B2 (en) * 2005-06-17 2007-02-27 Saifun Semiconductors Ltd. Method circuit and system for compensating for temperature induced margin loss in non-volatile memory cells
JP2007027760A (en) 2005-07-18 2007-02-01 Saifun Semiconductors Ltd High density nonvolatile memory array and manufacturing method
US7763927B2 (en) * 2005-12-15 2010-07-27 Macronix International Co., Ltd. Non-volatile memory device having a nitride-oxide dielectric layer
US20070036007A1 (en) * 2005-08-09 2007-02-15 Saifun Semiconductors, Ltd. Sticky bit buffer
US7668017B2 (en) 2005-08-17 2010-02-23 Saifun Semiconductors Ltd. Method of erasing non-volatile memory cells
US8116142B2 (en) * 2005-09-06 2012-02-14 Infineon Technologies Ag Method and circuit for erasing a non-volatile memory cell
US7352627B2 (en) * 2006-01-03 2008-04-01 Saifon Semiconductors Ltd. Method, system, and circuit for operating a non-volatile memory array
US7808818B2 (en) 2006-01-12 2010-10-05 Saifun Semiconductors Ltd. Secondary injection for NROM
US20070173017A1 (en) * 2006-01-20 2007-07-26 Saifun Semiconductors, Ltd. Advanced non-volatile memory array and method of fabrication thereof
US7692961B2 (en) 2006-02-21 2010-04-06 Saifun Semiconductors Ltd. Method, circuit and device for disturb-control of programming nonvolatile memory cells by hot-hole injection (HHI) and by channel hot-electron (CHE) injection
US8253452B2 (en) * 2006-02-21 2012-08-28 Spansion Israel Ltd Circuit and method for powering up an integrated circuit and an integrated circuit utilizing same
US7760554B2 (en) 2006-02-21 2010-07-20 Saifun Semiconductors Ltd. NROM non-volatile memory and mode of operation
US7701779B2 (en) 2006-04-27 2010-04-20 Sajfun Semiconductors Ltd. Method for programming a reference cell
US7907450B2 (en) * 2006-05-08 2011-03-15 Macronix International Co., Ltd. Methods and apparatus for implementing bit-by-bit erase of a flash memory device
US8772858B2 (en) * 2006-10-11 2014-07-08 Macronix International Co., Ltd. Vertical channel memory and manufacturing method thereof and operating method using the same
US7811890B2 (en) * 2006-10-11 2010-10-12 Macronix International Co., Ltd. Vertical channel transistor structure and manufacturing method thereof
US8223540B2 (en) 2007-02-02 2012-07-17 Macronix International Co., Ltd. Method and apparatus for double-sided biasing of nonvolatile memory
US7737488B2 (en) * 2007-08-09 2010-06-15 Macronix International Co., Ltd. Blocking dielectric engineered charge trapping memory cell with high speed erase
US8320191B2 (en) 2007-08-30 2012-11-27 Infineon Technologies Ag Memory cell arrangement, method for controlling a memory cell, memory array and electronic device
US9240405B2 (en) 2011-04-19 2016-01-19 Macronix International Co., Ltd. Memory with off-chip controller

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5130769A (en) * 1991-05-16 1992-07-14 Motorola, Inc. Nonvolatile memory cell
EP0642172A1 (en) * 1993-09-06 1995-03-08 Koninklijke Philips Electronics N.V. Semiconductor device having a non-volatile memory and method of manufacturing such a semiconductor device

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4203158A (en) * 1978-02-24 1980-05-13 Intel Corporation Electrically programmable and erasable MOS floating gate memory device employing tunneling and method of fabricating same
JPH05110114A (en) * 1991-10-17 1993-04-30 Rohm Co Ltd Nonvolatile semiconductor memory device
US5464785A (en) * 1994-11-30 1995-11-07 United Microelectronics Corporation Method of making a flash EPROM device having a drain edge P+ implant
US5719427A (en) * 1997-01-14 1998-02-17 Pericom Semiconductor Corp. Avalanche-enhanced CMOS transistor for EPROM/EEPROM and ESD-protection structures

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5130769A (en) * 1991-05-16 1992-07-14 Motorola, Inc. Nonvolatile memory cell
EP0642172A1 (en) * 1993-09-06 1995-03-08 Koninklijke Philips Electronics N.V. Semiconductor device having a non-volatile memory and method of manufacturing such a semiconductor device

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7535053B2 (en) * 1997-11-18 2009-05-19 Semiconductor Energy Laboratory Co., Ltd. Nonvolatile memory and electronic apparatus
WO2000036642A1 (en) * 1998-12-18 2000-06-22 Lattice Semiconductor Corporation Method of forming a non-volatile memory device
US6214666B1 (en) 1998-12-18 2001-04-10 Vantis Corporation Method of forming a non-volatile memory device
WO2000038240A1 (en) * 1998-12-21 2000-06-29 Lattice Semiconductor Corporation Floating gate memory cell structure with programming mechanism outside the read path
WO2000038245A1 (en) * 1998-12-21 2000-06-29 Lattice Semiconductor Corporation Dual pocket, two sided program/erase non-volatile memory cell
US6232631B1 (en) 1998-12-21 2001-05-15 Vantis Corporation Floating gate memory cell structure with programming mechanism outside the read path
US6282123B1 (en) 1998-12-21 2001-08-28 Lattice Semiconductor Corporation Method of fabricating, programming, and erasing a dual pocket two sided program/erase non-volatile memory cell
WO2000039805A1 (en) * 1998-12-23 2000-07-06 Lattice Semiconductor Corporation Floating gate memory apparatus and method for selected programming thereof
US6294809B1 (en) 1998-12-28 2001-09-25 Vantis Corporation Avalanche programmed floating gate memory cell structure with program element in polysilicon
WO2000041216A2 (en) * 1999-01-07 2000-07-13 Lattice Semiconductor Corporation Pmos avalanche programmed floating gate memory cell structure
WO2000041216A3 (en) * 1999-01-07 2002-09-26 Lattice Semiconductor Corp Pmos avalanche programmed floating gate memory cell structure
US6215700B1 (en) 1999-01-07 2001-04-10 Vantis Corporation PMOS avalanche programmed floating gate memory cell structure
EP1145279B1 (en) * 1999-01-14 2008-04-09 Infineon Technologies AG Semiconductor element with a tungsten oxide layer and method for its production
US6326663B1 (en) 1999-03-26 2001-12-04 Vantis Corporation Avalanche injection EEPROM memory cell with P-type control gate
US6424000B1 (en) 1999-05-11 2002-07-23 Vantis Corporation Floating gate memory apparatus and method for selected programming thereof
WO2002037550A1 (en) * 2000-10-30 2002-05-10 Advanced Micro Devices, Inc. Non-volatile memory with source side boron implantation
US6524914B1 (en) 2000-10-30 2003-02-25 Advanced Micro Devices, Inc. Source side boron implanting and diffusing device architecture for deep sub 0.18 micron flash memory
KR100810709B1 (en) 2000-10-30 2008-03-07 어드밴스드 마이크로 디바이시즈, 인코포레이티드 Non-volatile memory with source side boron implantation

Also Published As

Publication number Publication date
CA2226015A1 (en) 1997-01-23
AU6185196A (en) 1997-02-05
US6034896A (en) 2000-03-07

Similar Documents

Publication Publication Date Title
US6034896A (en) Method of fabricating a fast programmable flash E2 PROM cell
EP0360504B1 (en) One transistor flash eprom cell
US6949788B2 (en) Nonvolatile semiconductor memory device and method for operating the same
US7056785B2 (en) Non-volatile memory cell fabricated with slight modification to a conventional logic process and methods of operating same
US6211011B1 (en) Method for fabricating asymmetric virtual ground P-channel flash cell
KR100357644B1 (en) Non-volatile semiconductor memory and methods of driving operating, and manufacturing this memory
EP0928485B1 (en) Two-transistor flash eprom cell
JP2848223B2 (en) Erasing method and manufacturing method for nonvolatile semiconductor memory device
EP1181692A1 (en) Device with embedded flash and eeprom memories
US6025625A (en) Single-poly EEPROM cell structure operations and array architecture
US6265266B1 (en) Method of forming a two transistor flash EPROM cell
JPH0997849A (en) Semiconductor device
US6801456B1 (en) Method for programming, erasing and reading a flash memory cell
US6111286A (en) Low voltage low power n-channel flash memory cell using gate induced drain leakage current
JP3630491B2 (en) Semiconductor device
US7136306B2 (en) Single bit nonvolatile memory cell and methods for programming and erasing thereof
US6914826B2 (en) Flash memory structure and operating method thereof
US20060197139A1 (en) Non-volatile memory device having improved band-to-band tunneling induced hot electron injection efficiency and manufacturing method thereof
JP2003224215A (en) Transistor, semiconductor memory comprising it and method for driving transistor
EP0642172B1 (en) Semiconductor device having a non-volatile memory and method of manufacturing such a semiconductor device
KR100273705B1 (en) Method for fabricating of nonvolatile memory device and well structure thereof
US20020055228A1 (en) Sidewall process to improve the flash memory cell performance
EP0612108B1 (en) Double polysilicon EEPROM cell and corresponding manufacturing process
US20060226467A1 (en) P-channel charge trapping memory device with sub-gate
KR100745030B1 (en) Flash memory device, method for manufacturing the same and driving method for the same

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AL AM AT AU AZ BB BG BR BY CA CH CN CZ DE DK EE ES FI GB GE HU IL IS JP KE KG KP KR KZ LK LR LS LT LU LV MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK TJ TM TR TT UA UG US UZ VN AM AZ BY KG KZ MD RU TJ TM

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): KE LS MW SD SZ UG AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
ENP Entry into the national phase

Ref document number: 2226015

Country of ref document: CA

Kind code of ref document: A

Ref document number: 2226015

Country of ref document: CA

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

WWE Wipo information: entry into national phase

Ref document number: 08981745

Country of ref document: US

122 Ep: pct application non-entry in european phase