US20020011625A1 - Scalable tunnel oxide window with no isolation edges - Google Patents
Scalable tunnel oxide window with no isolation edges Download PDFInfo
- Publication number
- US20020011625A1 US20020011625A1 US09/235,198 US23519899A US2002011625A1 US 20020011625 A1 US20020011625 A1 US 20020011625A1 US 23519899 A US23519899 A US 23519899A US 2002011625 A1 US2002011625 A1 US 2002011625A1
- Authority
- US
- United States
- Prior art keywords
- floating gate
- tunnel
- oxide
- tunnel oxide
- eeprom
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
- 238000002955 isolation Methods 0.000 title description 2
- 238000009792 diffusion process Methods 0.000 claims abstract description 64
- 230000015654 memory Effects 0.000 claims abstract description 57
- 239000000758 substrate Substances 0.000 claims abstract description 44
- 239000004065 semiconductor Substances 0.000 claims abstract description 15
- 230000008859 change Effects 0.000 claims abstract description 6
- 229910021420 polycrystalline silicon Inorganic materials 0.000 claims description 32
- 229920005591 polysilicon Polymers 0.000 claims description 32
- 238000000034 method Methods 0.000 claims description 27
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 claims description 14
- 229910052710 silicon Inorganic materials 0.000 claims description 14
- 239000010703 silicon Substances 0.000 claims description 14
- 239000007943 implant Substances 0.000 claims description 12
- 239000000463 material Substances 0.000 claims description 11
- 239000000377 silicon dioxide Substances 0.000 claims description 7
- 238000005530 etching Methods 0.000 claims description 3
- 235000012239 silicon dioxide Nutrition 0.000 claims description 3
- 230000000873 masking effect Effects 0.000 claims description 2
- 238000000151 deposition Methods 0.000 claims 1
- 239000007772 electrode material Substances 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract description 15
- 238000004519 manufacturing process Methods 0.000 abstract description 7
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 18
- 239000010410 layer Substances 0.000 description 13
- 230000008569 process Effects 0.000 description 7
- 230000005641 tunneling Effects 0.000 description 7
- 239000002019 doping agent Substances 0.000 description 6
- 150000002500 ions Chemical class 0.000 description 6
- 229920002120 photoresistant polymer Polymers 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000009977 dual effect Effects 0.000 description 5
- 229910052681 coesite Inorganic materials 0.000 description 4
- 229910052906 cristobalite Inorganic materials 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 229910052682 stishovite Inorganic materials 0.000 description 4
- 229910052905 tridymite Inorganic materials 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000003647 oxidation Effects 0.000 description 3
- 238000007254 oxidation reaction Methods 0.000 description 3
- 229910021332 silicide Inorganic materials 0.000 description 3
- FVBUAEGBCNSCDD-UHFFFAOYSA-N silicide(4-) Chemical compound [Si-4] FVBUAEGBCNSCDD-UHFFFAOYSA-N 0.000 description 3
- ZOXJGFHDIHLPTG-UHFFFAOYSA-N Boron Chemical compound [B] ZOXJGFHDIHLPTG-UHFFFAOYSA-N 0.000 description 2
- 229910052796 boron Inorganic materials 0.000 description 2
- 239000003990 capacitor Substances 0.000 description 2
- 239000002800 charge carrier Substances 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 230000000149 penetrating effect Effects 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 239000002356 single layer Substances 0.000 description 2
- 230000005689 Fowler Nordheim tunneling Effects 0.000 description 1
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- 229910052581 Si3N4 Inorganic materials 0.000 description 1
- 229910008814 WSi2 Inorganic materials 0.000 description 1
- 229910052785 arsenic Inorganic materials 0.000 description 1
- RQNWIZPPADIBDY-UHFFFAOYSA-N arsenic atom Chemical compound [As] RQNWIZPPADIBDY-UHFFFAOYSA-N 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 210000003323 beak Anatomy 0.000 description 1
- 238000005229 chemical vapour deposition Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000001627 detrimental effect Effects 0.000 description 1
- 239000003989 dielectric material Substances 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005468 ion implantation Methods 0.000 description 1
- 229910021421 monocrystalline silicon Inorganic materials 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 239000011574 phosphorus Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000004549 pulsed laser deposition Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- HQVNEWCFYHHQES-UHFFFAOYSA-N silicon nitride Chemical compound N12[Si]34N5[Si]62N3[Si]51N64 HQVNEWCFYHHQES-UHFFFAOYSA-N 0.000 description 1
- 235000012431 wafers Nutrition 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B41/00—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates
- H10B41/30—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates characterised by the memory core region
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/66—Types of semiconductor device ; Multistep manufacturing processes therefor
- H01L29/66007—Multistep manufacturing processes
- H01L29/66075—Multistep manufacturing processes of devices having semiconductor bodies comprising group 14 or group 13/15 materials
- H01L29/66227—Multistep 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
- H01L29/66409—Unipolar field-effect transistors
- H01L29/66477—Unipolar field-effect transistors with an insulated gate, i.e. MISFET
- H01L29/66825—Unipolar field-effect transistors with an insulated gate, i.e. MISFET with a floating gate
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L29/00—Semiconductor devices specially adapted for rectifying, amplifying, oscillating or switching and having potential barriers; Capacitors or resistors having potential barriers, e.g. a PN-junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof ; Multistep manufacturing processes therefor
- H01L29/66—Types of semiconductor device ; Multistep manufacturing processes therefor
- H01L29/68—Types 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
- H01L29/76—Unipolar devices, e.g. field effect transistors
- H01L29/772—Field effect transistors
- H01L29/78—Field effect transistors with field effect produced by an insulated gate
- H01L29/788—Field effect transistors with field effect produced by an insulated gate with floating gate
- H01L29/7881—Programmable transistors with only two possible levels of programmation
- H01L29/7883—Programmable transistors with only two possible levels of programmation charging by tunnelling of carriers, e.g. Fowler-Nordheim tunnelling
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B41/00—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates
- H10B41/60—Electrically erasable-and-programmable ROM [EEPROM] devices comprising floating gates the control gate being a doped region, e.g. single-poly memory cell
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10B—ELECTRONIC MEMORY DEVICES
- H10B69/00—Erasable-and-programmable ROM [EPROM] devices not provided for in groups H10B41/00 - H10B63/00, e.g. ultraviolet erasable-and-programmable ROM [UVEPROM] devices
Definitions
- the present invention relates generally to electrically erasable/programmable read only memories (“EEPROMs”) and methods for their fabrication. More specifically, the invention relates to improved tunnel oxide windows on EEPROMs and methods of their fabrication.
- EEPROMs electrically erasable/programmable read only memories
- FIG. 1 depicts a basic two electrode EEPROM structure employed to control programming and erasing.
- a floating gate 2 forms one electrode and a heavily doped n-type silicon memory diffusion 4 within the silicon substrate 1 forms the other.
- These two electrodes are separated by a tunnel oxide 6 of 80 ⁇ in thickness, for example.
- a thick field oxide 8 of approximately 5000 ⁇ in thickness, for example, exists on either edge of the tunnel oxide 6 in order to isolate this device from other devices on the chip.
- the EEPROM cell is programmed (turned “off”) or erased (turned “on”) by charging or discharging, respectively, the floating gate 2 .
- Charge is moved on or off the floating gate 2 by tunneling electrons between the floating gate 2 and a heavily doped memory diffusion 4 in the semiconductor substrate 1 .
- high voltages are applied to the highly doped diffusion 4 or the floating gate 2 , electrons tunnel through tunnel oxide 6 separating the memory diffusion 4 from the floating gate 2 .
- the floating gate 2 is charged negatively to provide a very high threshold voltage which effectively prevents mobile charge carriers from flowing in a channel region of a read transistor (not shown) having as its gate electrode the floating gate 2 .
- the positively charged floating gate 2 provides a negative threshold voltage, allowing charge carriers to flow freely in the channel region.
- the tunnel oxide or more precisely, that portion of the tunnel oxide through which electrons flow—sometimes referred to as a tunnel window.
- the tunnel oxide acts as a dielectric and does not conduct electricity.
- electron tunneling occurs through this oxide by a process known as Fowler-Nordheim tunneling.
- the tunnel window contains defects (e.g., vacancies, charge trapping centers and dangling bonds), it may allow electrons to leak from a charged floating gate such that the EEPROM rapidly becomes inadvertently erased.
- the defects provide intermediate energy states allowing the electrons to tunnel more easily across the tunnel window. EEPROM devices having such defects are, of course, unacceptable.
- FIG. 2A A top view of one commonly used tunnel window design is shown in FIG. 2A.
- the five features defining the tunnel window 16 are represented here by the outlines of photolithographic masks employed during their formation. Reference numbers designate both the mask regions and their respective on-chip features.
- the five components are an N + memory diffusion 12 , a gate oxide 13 , a field oxide 14 , a tunnel oxide mask 11 , and a polysilicon floating gate 15 .
- a tunnel oxide is grown in the region defined by the tunnel oxide mask 11 over the memory diffusion (MD) 12 .
- the polysilicon floating gate 15 covers all of the tunnel oxide region.
- the tunnel window 16 is defined by that portion of the tunnel oxide mask 11 lying beyond the field oxide 14 and intersecting the memory diffusion (MD) 12 .
- FIG. 2B shows a cross section A-A of FIG. 2A.
- FIG. 2C A top view of another commonly used tunnel window design is shown in FIG. 2C.
- the features defining the tunnel window 26 are represented in part by the outlines of photolithographic masks employed during their formation, with the reference numbers designating both the mask regions and their respective on-chip features.
- a tunnel oxide 21 is grown within a region of gate oxide 22 , over a memory diffusion (MD) 23 , but removed from the edges of a field oxide 24 .
- a polysilicon floating gate 25 covers the entire tunnel oxide region and a large part of the surrounding area.
- the tunnel window 26 is determined by the size of the hole in the tunnel oxide mask (the mask applied prior to etching the gate oxide in preparation for the growth of the tunnel oxide). Therefore the tunnel oxide is entirely coincident with the tunnel window, which separates the MD 23 , below, and floating gate 25 , above.
- FIG. 2D shows a cross section B-B of FIG. 2C.
- the present invention provides an EEPROM device, and a method of making such a device, which incorporates a tunnel window having acceptably low gate capacitance at the tunnel oxide node, and which avoids the defects caused by field oxide induced stresses in the tunnel oxide.
- the EEPROM of the present invention includes a semiconductor substrate with a doped memory diffusion region serving as an electrical connection under the floating gate. Overlying at least a portion of the memory diffusion is a tunnel oxide. Overlying at least a portion of the tunnel oxide is a floating gate extension.
- the tunnel window of the EEPROM of the present invention is self-aligned, being defined within at least a portion of the tunnel oxide overlying the memory diffusion and having at least two edges defined by the floating gate extension, so that when a defined voltage is applied to the memory diffusion a tunnel current sufficient to change the state (on/off) of the EEPROM flows between the memory diffusion and the floating gate structure.
- the invention further provides a method of forming an EEPROM, including forming a doped memory diffusion region in a semiconductor substrate, and forming a tunnel oxide on the substrate overlying at least a portion of the memory diffusion.
- a floating gate structure including an extension overlying at least a portion of the tunnel oxide overlying the memory diffusion and having at least two edges defined by the floating gate extension is then formed. When a defined voltage is applied to the memory diffusion a tunnel current sufficient to change the state of the EEPROM flows between the memory diffusion and the floating gate structure.
- the tunnel window is substantially separated from the field oxide-induced stress region.
- tunnel oxide and polysilicon masks can be precisely scaled to produce smaller tunnel windows for future applications without encountering the field oxide area defects or the capacitance problems of the prior art designs.
- FIG. 1 is a cross-sectional schematic view of the tunnel oxide node of a basic EEPROM device, including defects.
- FIG. 2A is a top view of one commonly used EEPROM tunnel window design.
- FIG. 2B is a cross-sectional view (A-A) of the EEPROM tunnel window design of FIG. 2A.
- FIG. 2C is a top view of another commonly used EEPROM tunnel window design.
- FIG. 2D is a cross-sectional view (B-B) of the EEPROM tunnel window design of FIG. 2C.
- FIG. 3 is a cross-sectional perspective view of a preferred embodiment of the present invention, a single polysilicon EEPROM cell.
- FIGS. 4 A-D are cross-sectional side views of stages in the preparation of a substrate for processing according to the present invention.
- FIGS. 5 A-E are cross-sectional perspective views of stages in the making the scalable tunnel oxide windows according to the present invention.
- FIG. 6 is a top view of a scalable tunnel oxide window according to the present invention.
- FIGS. 7 A-D are top views of representative EEPROM implementations of the tunnel oxide window of the present invention.
- a single polysilicon EEPROM cell 30 is shown.
- the cell includes a single polysilicon floating gate structure 32 which performs three functions.
- a tunnel extension 34 of floating gate 32 acts as an electrode in the two terminal device used for tunneling electrons from a heavily doped N + implant 35 (also referred to as a Memory Diffusion or MD) through a tunnel oxide 36 (often about 80 ⁇ thick) onto floating gate structure 32 .
- a wide area plate 38 is employed as one electrode of a capacitor enabling the floating gate 32 to be raised to a high voltage (e.g., about 6 to 11 volts) by capacitively coupling a programming voltage (e.g., about 9 to 13 volts) from a second electrode 40 (heavily doped N+ silicon, referred to herein as a control gate memory diffusion) through an oxide 42 (often about 180 ⁇ thick). Between these two ends is a section of polysilicon that forms the gate 44 of a read transistor.
- a high voltage e.g., about 6 to 11 volts
- a programming voltage e.g., about 9 to 13 volts
- oxide 42 often about 180 ⁇ thick
- the read transistor is connected in series with a word line transistor having a gate 46 forming part of a word line 47 .
- the read and word line transistors separate a sense amp negative ( ⁇ ) input 48 from a sense amp positive (+) input 50 .
- Charging the floating gate 32 by tunneling electrons onto it (through tunnel oxide 36 ) raises the threshold voltage of the read transistor (EEPROM cell 30 is programmed). This shuts off the channel between the sense amp inputs, even when the adjacent word line transistor is turned on. Tunneling electrons off the floating gate 32 reduces the read transistor threshold voltage to negative values, effectively turning this device on (EEPROM cell 30 is erased).
- the word line transistor in series then controls the signal path between the two sense amp inputs 48 and 50 .
- the EEPROM cell is programmed or erased by charging or discharging, respectively, the floating gate 32 .
- a high voltage must be applied to the control gate memory diffusion 40 .
- the write column 56 is grounded and the write column select line 61 is selected with, for example, 5 volts.
- the sense amp ( ⁇ ) input 48 can be biased from 5 volts to a high voltage to assist tunneling electrons onto the floating gate 32 .
- the voltage on the control gate memory diffusion 40 is capacitively coupled to the floating gate 32 as is the sense amp ( ⁇ ) input 48 voltage. The resulting positive voltage on floating gate 32 is sufficient to cause tunneling onto floating gate 32 through the tunnel oxide 36 , thereby programming the EEPROM cell 30 .
- a high voltage In order to tunnel electrons off floating gate 32 , a high voltage must be applied to memory diffusion 35 while ground is applied to second heavily doped N+ implant 40 which underlies and is capacitively coupled to the wide area plate 38 . During this process, ground is also applied to sense amp ( ⁇ ) input 48 .
- the application of high voltage to memory diffusion 35 is accomplished through a write column 56 and a write column select transistor 52 including (i) a diffusion region 54 conductively connected to write column 56 for data input, (ii) a source/drain diffusion 58 electrically connected to memory diffusion 35 , and (iii) a gate electrode 60 forming part of write column select transistor 61 .
- FIGS. 4 A-D and 5 A-E illustrate some of the important steps in the fabrication of a self-aligned scalable tunnel oxide window in accordance with a preferred embodiment of the present invention. It should be understood that numerous variations of this process are permissible and within the scope of this invention. Further, it should be understood that the conditions recited herein are appropriate for devices having feature sizes in the neighborhood of less than about 0.5 ⁇ m. The process conditions may have to be modified somewhat for devices in other size regimes, according to design criteria known in the art. In addition, while the invention is mainly intended for use with single polysilicon EEPROM cells, it is also applicable to double polysilicon EEPROM cells and other designs.
- the method for producing a self-aligned scalable tunnel oxide window in accordance with this invention forms the window over a N + memory diffusion of an otherwise conventional EEPROM programmable logic device (PLD).
- PLD EEPROM programmable logic device
- the semiconductor substrate may be prepared for processing according to the present invention as follows.
- a p-type ion implant is performed over the substrate 211 , penetrating an upper portion of the substrate (the “p-well portion”).
- the p-type dopant is provided at a dose and energy sufficient to provide a very lightly doped well region 221 .
- this implant might be conducted with boron implanted at about 5 ⁇ 10 12 to 5 ⁇ 10 13 cm ⁇ 2 at an energy of between about 35 and 70 keV.
- a subsequent diffusion anneal may then be conducted at a temperature and for a time sufficient to smooth out the p-type dopant concentration over well region 221 to give a relatively uniform overall light p-doping level.
- the well formation conditions should be chosen such that the p-well region has a dopant concentration at the substrate surface (i.e., the channel region) of between about 2 ⁇ 10 16 and 1 ⁇ 10 17 cm ⁇ 3 (as discussed above). It should be noted that, in many systems, the shape and extent of the p-well is defined by providing the semiconductor material with an ion implant mask of photoresist or other suitable masking material prior to ion implantation. Further alternative embodiments include silicon wafers cut from single crystal silicon grown with a specific dopant concentration, and growing a layer of silicon of the desired bulk concentration epitaxially on top of a heavily doped substrate.
- a thick SiN and thin SiO 2 stack mask 224 is provided to protect the active areas (i.e., regions where electrodes, tunnel regions, channels, etc. are to be formed).
- a thick field oxide 229 (preferably of silicon dioxide) is grown at a temperature of between about 900 and 1050° C. to a thickness of between about 0.2 and 0.5 micrometers to produce the structure shown in FIG. 4C.
- the SiN/SiO 2 mask 224 is removed to produce the structure shown in FIG. 4D.
- the field oxide 229 functions to electrically isolate the various devices and some device elements from one another on the substrate 211 . In alternative embodiments, conventional isolation trenches could be used in place of the field oxide to isolate devices on the substrate 211 .
- FIGS. 5 A-F the fabrication of the scalable tunnel oxide windows according to the present invention will be described.
- FIG. 5A after the bare silicon substrate has been exposed following the field oxide growing step, another photoresist mask 330 is provided such that an area of the silicon substrate surface bounded by the field oxide 229 is left exposed. Then an n-type ion implant is performed over the substrate 211 , penetrating the substrate's exposed p-well region 221 to create a region underlying the substrate surface that has an n-type conductivity (the “memory diffusion” or “MD” region 335 ).
- MD memory diffusion
- the n-type dopants phosphorus and arsenic may be provided at doses of between about 1 ⁇ 10 14 to 1 ⁇ 10 15 and 7 ⁇ 10 14 to 7 ⁇ 10 15 , respectively, at an energy of about 80 to 120 keV and 60 to 80 keV, respectively, which is sufficient to provide a highly doped MD region 335 .
- This implant produces an n-type dopant concentration of about 1 ⁇ 10 18 to 1 ⁇ 10 20 cm ⁇ 3 in an MD region about 0.4 and 0.8 ⁇ m deep immediately under the substrate surface (an area which will act as tunnel oxide connection under the polysilicon floating gate in the completed device—e.g., MD 35 in FIG. 3).
- the photoresist is removed.
- threshold voltage adjustment implants may be performed on the device substrate in order to set appropriate transistor threshold voltages.
- the implant is conducted with Boron (B) ions at a dose of between about 1 ⁇ 10 12 and 6 ⁇ 10 12 and an energy of between about 20 and 40 keV. Methods for generating these implants are well known in the art.
- a thin gate oxide 340 is grown on the bare silicon surface in an oxygen-containing atmosphere at a temperature of between about 800 and 950° C. for about 6 to 10 minutes to produce a thickness of between about 145 and 190 ⁇ over the MD region 335 . Because the gate oxide grows more quickly over the highly doped MD region 335 than the well 221 , the thickness of the gate oxide outside of the MD 335 will be slightly less, about 125 to 170 ⁇ .
- the gate oxide 340 will be a single layer of silicon dioxide, but it may also be a multi-layer structure including, for instance, silicon nitride. (Technically, such multi-layer structures should be referred to as “gate dielectrics” rather than merely “gate oxides”).
- a photoresist mask 345 is applied, leaving only a slot 350 whose width is selected to be a first dimension of the desired tunnel oxide window.
- this slot may have the following range of dimensions: width, about 0.25 to 0.5 ⁇ m by length, about 0.75 to 1.5 ⁇ m.
- the gate oxide exposed in the slot area 350 is removed, for instance, with a wet SiO 2 etch using 1% HF or 10:1 buffered HF solution. The wet etch will generally make the final window dimension about 0.05 to 0.1 ⁇ m per side larger than the mask 345 dimensions.
- the exposed silicon surface should be as smooth as possible. Additional surface smoothening may be conducted by further oxidation and/or by other processing methods known in the art, such as RCA cleans.
- FIG. 5D The structure of FIG. 5D is produced by removing the photoresist used during the most recent etch step. This leaves a region of bare silicon 355 within the gate oxide region 340 on substrate 211 overlying at least a portion of the MD 335 . Note that the region 355 may extend into the field oxide region 229 . Then, as shown in FIG. 5E, a tunnel oxide layer 358 is grown over the slot of bare silicon surface 355 created by the etch. A tunnel oxide 358 is generated by thermal oxidation at 850 to 950° C. for about 7 minutes in a dry O 2 /Cl atmosphere so that the bare silicon surface in the slot 355 is converted to an oxide layer of about 70 to 100 ⁇ in thickness. This thermal oxidation will further increase the thickness of the gate oxide layer to about 160 to 200 ⁇ over the MD region 335 , and 145 to 190 ⁇ over the well 221 .
- a blanket of polysilicon 360 approximately 1000 to 2000 ⁇ in thickness is deposited over the device surface by, for example, chemical vapor deposition.
- the polysilicon layer may be followed by a silicide (e.g., WSi 2 ) layer (not shown) of about 1000 to 2000 ⁇ .
- a strip of photoresist mask 365 is formed on the surface of polysilicon (or polysilicon/silicide) layer 360 and oriented 90 degrees from the longer axis of the tunnel oxide slot 355 .
- the mask 365 defines two edges of the desired tunnel oxide window (with the other edges being defined by the tunnel oxide mask).
- an etch is used to remove the unmasked polysilicon while minimizing damage to the tunnel window edge. Since charging and tunnel oxide removal along the polysilicon edge may degrade the tunnel oxide reliability and endurance, the etch should be highly selective to polysilicon over SiO 2 , approximately 20-100:1, so as to minimize the amount of tunnel oxide removed from along the edges of the polysilicon.
- the etch should also minimize the charge on the polysilicon after the floating gates are isolated during the etch.
- An example of a suitable etch is a reactive ion etch (RIE) using SF 6 /Cl 2 .
- a tunnel window 475 is defined by the intersection of the tunnel oxide 358 and the polysilicon floating gate extension 370 .
- the tunnel window is self-aligned with the polysilicon strip 370 thereby facilitating its precise placement over the memory diffusion and its precise scaling.
- no edge of the tunnel window is defined by field oxide. Any field oxide stress region-associated defects in the tunnel window are thereby avoided. Also, the difficulty of cutting small scale tunnel oxide “holes” in a gate oxide region is not faced. Further, the design does not result in excessive MD/gate oxide/floating gate capacitance.
- FIGS. 7 A-D show cell layouts for preferred implementations of the tunnel window design of the present invention.
- FIG. 7A is a top view of an EEPROM cell 500 having active region diffusions 508 , 511 and 512 , and having a single layer of polysilicon forming a floating gate 502 straddling all three diffusions.
- the tunnel window 503 is shown as the intersection of the floating gate extension 504 and a tunnel oxide region 506 overlying a heavily doped memory diffusion 509 .
- a wide area plate 510 is employed as one electrode of a capacitor enabling the floating gate 502 to be raised to a high voltage (about 6.5 to 10.5 volts) by capacitively coupling from a second electrode 512 (heavily doped N+ silicon) through a gate oxide 514 (often about 200 ⁇ thick). Between these two extremes is a section of polysilicon that forms the gate 516 of a read transistor. This layout also includes a polysilicon single row-line select 518 (word line) for controlling read and write operations.
- FIG. 7B is a top view of a second single poly EEPROM layout having the same elements as FIG.
- This dual row-line select 520 a and 520 b allows independent control of the read operation ( 520 a ) and the write operation ( 520 b ).
- FIG. 7C is a top view of an EEPROM cell 530 having a double layer polysilicon stack 532 which includes a floating gate (lower layer) and a control gate 540 (upper layer). Double layer polysilicon stack 532 straddles diffusions 538 and 541 as shown.
- the tunnel window 533 is again shown as the intersection of a floating gate extension 534 and a tunnel oxide region 536 overlying a heavily doped memory diffusion 539 .
- This layout also includes a polysilicon single row-line select 542 as in the embodiment of FIG. 7A.
- FIG. 7D is a top view of a second double poly EEPROM layout having the same elements as FIG. 7C, but with a dual row-line select 544 , rather than a single. This dual row-line select 544 is analogous to dual row-line select 520 a and 520 b in FIG. 7B.
Landscapes
- Engineering & Computer Science (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- Ceramic Engineering (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Manufacturing & Machinery (AREA)
- Non-Volatile Memory (AREA)
- Semiconductor Memories (AREA)
Abstract
Disclosed is an EEPROM device, and a method of making such a device, which incorporates a self-aligned tunnel window having acceptably low gate capacitance at the tunnel oxide node, and which avoids the defects caused by field oxide induced stresses in the tunnel oxide. The EEPROM of the present invention includes a semiconductor substrate with a doped memory diffusion region. Overlying at least a portion of the memory diffusion is a tunnel oxide. Overlying at least a portion of the tunnel oxide is a floating gate structure including an extension. The tunnel window of the EEPROM of the present invention is defined within at least a portion of the tunnel oxide and having at least two edges defined by the floating gate extension, so that when a defined voltage is applied to the memory diffusion a tunnel current sufficient to change the state of the EEPROM flows between the memory diffusion and the floating gate structure.
Description
- This application claims the benefit of the filing date of Provisional Application Serial No. 60/023,725, filed Aug. 8, 1996.
- The present invention relates generally to electrically erasable/programmable read only memories (“EEPROMs”) and methods for their fabrication. More specifically, the invention relates to improved tunnel oxide windows on EEPROMs and methods of their fabrication.
- As the name implies, electrically erasable/programmable devices can be both erased and programmed electrically without the need for exposure to ultraviolet light or other external factors. This has many advantages, of which the most obvious is the ability to erase or program EEPROM devices in window-less packages (unlike EPROM devices). FIG. 1 depicts a basic two electrode EEPROM structure employed to control programming and erasing. A floating gate2 forms one electrode and a heavily doped n-type silicon memory diffusion 4 within the
silicon substrate 1 forms the other. These two electrodes are separated by atunnel oxide 6 of 80 Å in thickness, for example. A thick field oxide 8 of approximately 5000 Å in thickness, for example, exists on either edge of thetunnel oxide 6 in order to isolate this device from other devices on the chip. - The EEPROM cell is programmed (turned “off”) or erased (turned “on”) by charging or discharging, respectively, the floating gate2. Charge is moved on or off the floating gate 2 by tunneling electrons between the floating gate 2 and a heavily doped memory diffusion 4 in the
semiconductor substrate 1. When high voltages are applied to the highly doped diffusion 4 or the floating gate 2, electrons tunnel throughtunnel oxide 6 separating the memory diffusion 4 from the floating gate 2. In the EEPROM's “off” state, the floating gate 2 is charged negatively to provide a very high threshold voltage which effectively prevents mobile charge carriers from flowing in a channel region of a read transistor (not shown) having as its gate electrode the floating gate 2. In the “on” state, the positively charged floating gate 2 provides a negative threshold voltage, allowing charge carriers to flow freely in the channel region. - Key to the programming and erasing processes is the tunnel oxide, or more precisely, that portion of the tunnel oxide through which electrons flow—sometimes referred to as a tunnel window. Under normal conditions, the tunnel oxide acts as a dielectric and does not conduct electricity. However, under high electric fields of, for example, 8 to 13 volts across the oxide, electron tunneling occurs through this oxide by a process known as Fowler-Nordheim tunneling. If the tunnel window contains defects (e.g., vacancies, charge trapping centers and dangling bonds), it may allow electrons to leak from a charged floating gate such that the EEPROM rapidly becomes inadvertently erased. The defects provide intermediate energy states allowing the electrons to tunnel more easily across the tunnel window. EEPROM devices having such defects are, of course, unacceptable.
- Defects commonly result at the interface of a field oxide region and the tunnel oxide. The thick field oxide introduces severe stress in the materials proximate its “bird beak” protrusion. These stresses in turn can create defects in those portions of the tunnel window abutting the field oxide. The locus of this problem is illustrated in FIG. 1 (depicted as black dots) near the field oxide8. In a related problem, when two field oxide regions come into close proximity (e.g., less than 0.8 μm separation), their combined stresses may bow the substrate surface together with the overlying tunnel oxide as shown, in an exaggerated fashion, in FIG. 1. This bowing may in its own right introduce defects into the tunnel oxide. While such a bowing may not be evident in relatively large devices, as EEPROMs scale to smaller dimensions (having for example tunnel windows of 0.5×0.5 μm) it can become a major concern.
- A top view of one commonly used tunnel window design is shown in FIG. 2A. The five features defining the
tunnel window 16 are represented here by the outlines of photolithographic masks employed during their formation. Reference numbers designate both the mask regions and their respective on-chip features. The five components are an N+ memory diffusion 12, agate oxide 13, afield oxide 14, atunnel oxide mask 11, and apolysilicon floating gate 15. In this design, a tunnel oxide is grown in the region defined by thetunnel oxide mask 11 over the memory diffusion (MD) 12. Thepolysilicon floating gate 15 covers all of the tunnel oxide region. Thetunnel window 16 is defined by that portion of thetunnel oxide mask 11 lying beyond thefield oxide 14 and intersecting the memory diffusion (MD) 12. Thetunnel window 16 is therefore bordered byfield oxide 14 on two opposing sides. As noted above, the stresses in the tunnel oxide at the field oxide boundary areas can produce defects which may be detrimental to EEPROM function. As further noted above, these defects are increasingly likely with this design as device sizes decrease since smaller tunnel windows require more closely spaced field oxide regions. FIG. 2B shows a cross section A-A of FIG. 2A. - A top view of another commonly used tunnel window design is shown in FIG. 2C. Again, the features defining the
tunnel window 26 are represented in part by the outlines of photolithographic masks employed during their formation, with the reference numbers designating both the mask regions and their respective on-chip features. In this design, atunnel oxide 21 is grown within a region ofgate oxide 22, over a memory diffusion (MD) 23, but removed from the edges of afield oxide 24. Apolysilicon floating gate 25 covers the entire tunnel oxide region and a large part of the surrounding area. In this design, thetunnel window 26 is determined by the size of the hole in the tunnel oxide mask (the mask applied prior to etching the gate oxide in preparation for the growth of the tunnel oxide). Therefore the tunnel oxide is entirely coincident with the tunnel window, which separates the MD 23, below, and floatinggate 25, above. FIG. 2D shows a cross section B-B of FIG. 2C. - While this design avoids the problems associated with tunnel oxide defects caused at the field oxide edges, it presents similar scaling problems since it becomes increasingly difficult to define and etch smaller square openings for the tunnel oxide mask. This design approach suffers from the additional problem that it produces a large overlap of polysilicon floating gate and MD separated by gate (or MD) oxide. The result is that there is a relatively high gate capacitance at the tunnel oxide node of the EEPROM. This gate capacitance will degrade the performance of the cell.
- Thus, there is a need for an improved tunnel window design which does not have an unacceptably high capacitance and which can scale to smaller device sizes without introducing unacceptable defects in active portions of the tunnel oxide.
- The present invention provides an EEPROM device, and a method of making such a device, which incorporates a tunnel window having acceptably low gate capacitance at the tunnel oxide node, and which avoids the defects caused by field oxide induced stresses in the tunnel oxide.
- The EEPROM of the present invention includes a semiconductor substrate with a doped memory diffusion region serving as an electrical connection under the floating gate. Overlying at least a portion of the memory diffusion is a tunnel oxide. Overlying at least a portion of the tunnel oxide is a floating gate extension. The tunnel window of the EEPROM of the present invention is self-aligned, being defined within at least a portion of the tunnel oxide overlying the memory diffusion and having at least two edges defined by the floating gate extension, so that when a defined voltage is applied to the memory diffusion a tunnel current sufficient to change the state (on/off) of the EEPROM flows between the memory diffusion and the floating gate structure.
- The invention further provides a method of forming an EEPROM, including forming a doped memory diffusion region in a semiconductor substrate, and forming a tunnel oxide on the substrate overlying at least a portion of the memory diffusion. A floating gate structure including an extension overlying at least a portion of the tunnel oxide overlying the memory diffusion and having at least two edges defined by the floating gate extension is then formed. When a defined voltage is applied to the memory diffusion a tunnel current sufficient to change the state of the EEPROM flows between the memory diffusion and the floating gate structure.
- In a preferred embodiment of the present invention, the tunnel window is substantially separated from the field oxide-induced stress region.
- Among the advantages of the present invention is that the tunnel oxide and polysilicon masks can be precisely scaled to produce smaller tunnel windows for future applications without encountering the field oxide area defects or the capacitance problems of the prior art designs.
- These and other features and advantages of the invention will be described in more detail below with reference to the drawings.
- FIG. 1 is a cross-sectional schematic view of the tunnel oxide node of a basic EEPROM device, including defects.
- FIG. 2A is a top view of one commonly used EEPROM tunnel window design.
- FIG. 2B is a cross-sectional view (A-A) of the EEPROM tunnel window design of FIG. 2A.
- FIG. 2C is a top view of another commonly used EEPROM tunnel window design.
- FIG. 2D is a cross-sectional view (B-B) of the EEPROM tunnel window design of FIG. 2C.
- FIG. 3 is a cross-sectional perspective view of a preferred embodiment of the present invention, a single polysilicon EEPROM cell.
- FIGS.4A-D are cross-sectional side views of stages in the preparation of a substrate for processing according to the present invention.
- FIGS.5A-E are cross-sectional perspective views of stages in the making the scalable tunnel oxide windows according to the present invention.
- FIG. 6 is a top view of a scalable tunnel oxide window according to the present invention.
- FIGS.7A-D are top views of representative EEPROM implementations of the tunnel oxide window of the present invention
- The present invention will be described in terms of a preferred embodiment. It will be understood by those skilled in the art that variations of the described method and apparatus may be made without departing from the scope of the present invention.
- In a preferred embodiment illustrated in FIG. 3, a single
polysilicon EEPROM cell 30 is shown. The cell includes a single polysilicon floatinggate structure 32 which performs three functions. At a first end, atunnel extension 34 of floatinggate 32 acts as an electrode in the two terminal device used for tunneling electrons from a heavily doped N+ implant 35 (also referred to as a Memory Diffusion or MD) through a tunnel oxide 36 (often about 80 Å thick) onto floatinggate structure 32. At the other end of this floating gate, awide area plate 38 is employed as one electrode of a capacitor enabling the floatinggate 32 to be raised to a high voltage (e.g., about 6 to 11 volts) by capacitively coupling a programming voltage (e.g., about 9 to 13 volts) from a second electrode 40 (heavily doped N+ silicon, referred to herein as a control gate memory diffusion) through an oxide 42 (often about 180 Å thick). Between these two ends is a section of polysilicon that forms thegate 44 of a read transistor. - The read transistor is connected in series with a word line transistor having a
gate 46 forming part of a word line 47. The read and word line transistors separate a sense amp negative (−)input 48 from a sense amp positive (+)input 50. Charging the floatinggate 32 by tunneling electrons onto it (through tunnel oxide 36) raises the threshold voltage of the read transistor (EEPROM cell 30 is programmed). This shuts off the channel between the sense amp inputs, even when the adjacent word line transistor is turned on. Tunneling electrons off the floatinggate 32 reduces the read transistor threshold voltage to negative values, effectively turning this device on (EEPROM cell 30 is erased). The word line transistor in series then controls the signal path between the twosense amp inputs - The EEPROM cell is programmed or erased by charging or discharging, respectively, the floating
gate 32. In order to tunnel electrons onto floatinggate 32, a high voltage must be applied to the controlgate memory diffusion 40. At the same time, thewrite column 56 is grounded and the write columnselect line 61 is selected with, for example, 5 volts. The sense amp (−)input 48 can be biased from 5 volts to a high voltage to assist tunneling electrons onto the floatinggate 32. The voltage on the controlgate memory diffusion 40 is capacitively coupled to the floatinggate 32 as is the sense amp (−)input 48 voltage. The resulting positive voltage on floatinggate 32 is sufficient to cause tunneling onto floatinggate 32 through thetunnel oxide 36, thereby programming theEEPROM cell 30. - In order to tunnel electrons off floating
gate 32, a high voltage must be applied to memory diffusion 35 while ground is applied to second heavily dopedN+ implant 40 which underlies and is capacitively coupled to thewide area plate 38. During this process, ground is also applied to sense amp (−)input 48. The application of high voltage to memory diffusion 35 is accomplished through awrite column 56 and a write columnselect transistor 52 including (i) a diffusion region 54 conductively connected to writecolumn 56 for data input, (ii) a source/drain diffusion 58 electrically connected to memory diffusion 35, and (iii) agate electrode 60 forming part of write columnselect transistor 61. When a sufficient potential is applied to write columnselect transistor 61 while a write signal is applied throughwrite column 56, electrons can tunnel off of floatinggate 32 to erase the EEPROM cell. - A further description of a typical EEPROM cell and its functional elements is available the publication “EPM7032 Process, Assembly, and Reliability Information Package” available from Altera Corporation of San Jose Calif. That document is incorporated herein by reference for all purposes.
- FIGS.4A-D and 5A-E illustrate some of the important steps in the fabrication of a self-aligned scalable tunnel oxide window in accordance with a preferred embodiment of the present invention. It should be understood that numerous variations of this process are permissible and within the scope of this invention. Further, it should be understood that the conditions recited herein are appropriate for devices having feature sizes in the neighborhood of less than about 0.5 μm. The process conditions may have to be modified somewhat for devices in other size regimes, according to design criteria known in the art. In addition, while the invention is mainly intended for use with single polysilicon EEPROM cells, it is also applicable to double polysilicon EEPROM cells and other designs.
- Preferably, the method for producing a self-aligned scalable tunnel oxide window in accordance with this invention forms the window over a N+ memory diffusion of an otherwise conventional EEPROM programmable logic device (PLD). The design disclosed improves reliability and performance of PLDs presently available by allowing placement of a tunnel window away from the diffusion edge without detrimentally increasing gate capacitance at the tunnel window node of the EEPROM cell.
- The semiconductor substrate may be prepared for processing according to the present invention as follows. In FIG. 4A, a p-type ion implant is performed over the
substrate 211, penetrating an upper portion of the substrate (the “p-well portion”). The p-type dopant is provided at a dose and energy sufficient to provide a very lightlydoped well region 221. For example, this implant might be conducted with boron implanted at about 5×1012 to 5×1013 cm−2 at an energy of between about 35 and 70 keV. A subsequent diffusion anneal may then be conducted at a temperature and for a time sufficient to smooth out the p-type dopant concentration overwell region 221 to give a relatively uniform overall light p-doping level. The well formation conditions should be chosen such that the p-well region has a dopant concentration at the substrate surface (i.e., the channel region) of between about 2×1016 and 1×1017 cm−3 (as discussed above). It should be noted that, in many systems, the shape and extent of the p-well is defined by providing the semiconductor material with an ion implant mask of photoresist or other suitable masking material prior to ion implantation. Further alternative embodiments include silicon wafers cut from single crystal silicon grown with a specific dopant concentration, and growing a layer of silicon of the desired bulk concentration epitaxially on top of a heavily doped substrate. - To obtain the structure shown in FIG. 4B, a thick SiN and thin SiO2 stack mask 224 is provided to protect the active areas (i.e., regions where electrodes, tunnel regions, channels, etc. are to be formed). Next, a thick field oxide 229 (preferably of silicon dioxide) is grown at a temperature of between about 900 and 1050° C. to a thickness of between about 0.2 and 0.5 micrometers to produce the structure shown in FIG. 4C. Thereafter, the SiN/SiO2 mask 224 is removed to produce the structure shown in FIG. 4D. The
field oxide 229 functions to electrically isolate the various devices and some device elements from one another on thesubstrate 211. In alternative embodiments, conventional isolation trenches could be used in place of the field oxide to isolate devices on thesubstrate 211. - Turning to FIGS.5A-F, the fabrication of the scalable tunnel oxide windows according to the present invention will be described. As shown in FIG. 5A, after the bare silicon substrate has been exposed following the field oxide growing step, another
photoresist mask 330 is provided such that an area of the silicon substrate surface bounded by thefield oxide 229 is left exposed. Then an n-type ion implant is performed over thesubstrate 211, penetrating the substrate's exposed p-well region 221 to create a region underlying the substrate surface that has an n-type conductivity (the “memory diffusion” or “MD” region 335). For example, the n-type dopants phosphorus and arsenic may be provided at doses of between about 1×1014 to 1×1015 and 7×1014 to 7×1015, respectively, at an energy of about 80 to 120 keV and 60 to 80 keV, respectively, which is sufficient to provide a highly dopedMD region 335. This implant produces an n-type dopant concentration of about 1×1018 to 1×1020 cm−3 in an MD region about 0.4 and 0.8 μm deep immediately under the substrate surface (an area which will act as tunnel oxide connection under the polysilicon floating gate in the completed device—e.g., MD 35 in FIG. 3). Following this n-type ion implant step, the photoresist is removed. - At this stage, threshold voltage adjustment implants (not shown) may be performed on the device substrate in order to set appropriate transistor threshold voltages. In a preferred embodiment, the implant is conducted with Boron (B) ions at a dose of between about 1×1012 and 6×1012 and an energy of between about 20 and 40 keV. Methods for generating these implants are well known in the art.
- Next, as shown in FIG. 5B, a
thin gate oxide 340 is grown on the bare silicon surface in an oxygen-containing atmosphere at a temperature of between about 800 and 950° C. for about 6 to 10 minutes to produce a thickness of between about 145 and 190 Å over theMD region 335. Because the gate oxide grows more quickly over the highly dopedMD region 335 than the well 221, the thickness of the gate oxide outside of theMD 335 will be slightly less, about 125 to 170 Å. Typically, thegate oxide 340 will be a single layer of silicon dioxide, but it may also be a multi-layer structure including, for instance, silicon nitride. (Technically, such multi-layer structures should be referred to as “gate dielectrics” rather than merely “gate oxides”). - As shown in FIG. 5C, after the gate oxide or gate dielectric has been formed, a
photoresist mask 345 is applied, leaving only aslot 350 whose width is selected to be a first dimension of the desired tunnel oxide window. In one embodiment, this slot may have the following range of dimensions: width, about 0.25 to 0.5 μm by length, about 0.75 to 1.5 μm. After themask 345 is formed, the gate oxide exposed in theslot area 350 is removed, for instance, with a wet SiO2 etch using 1% HF or 10:1 buffered HF solution. The wet etch will generally make the final window dimension about 0.05 to 0.1 μm per side larger than themask 345 dimensions. - The exposed silicon surface should be as smooth as possible. Additional surface smoothening may be conducted by further oxidation and/or by other processing methods known in the art, such as RCA cleans.
- The structure of FIG. 5D is produced by removing the photoresist used during the most recent etch step. This leaves a region of
bare silicon 355 within thegate oxide region 340 onsubstrate 211 overlying at least a portion of theMD 335. Note that theregion 355 may extend into thefield oxide region 229. Then, as shown in FIG. 5E, atunnel oxide layer 358 is grown over the slot ofbare silicon surface 355 created by the etch. Atunnel oxide 358 is generated by thermal oxidation at 850 to 950° C. for about 7 minutes in a dry O2/Cl atmosphere so that the bare silicon surface in theslot 355 is converted to an oxide layer of about 70 to 100 Å in thickness. This thermal oxidation will further increase the thickness of the gate oxide layer to about 160 to 200 Å over theMD region 335, and 145 to 190 Å over the well 221. - Next, in order to form the floating gate, a blanket of polysilicon360 approximately 1000 to 2000 Å in thickness is deposited over the device surface by, for example, chemical vapor deposition. The polysilicon layer may be followed by a silicide (e.g., WSi2) layer (not shown) of about 1000 to 2000 Å.
- To produce the structure of FIG. 5E, a strip of
photoresist mask 365 is formed on the surface of polysilicon (or polysilicon/silicide) layer 360 and oriented 90 degrees from the longer axis of thetunnel oxide slot 355. Themask 365 defines two edges of the desired tunnel oxide window (with the other edges being defined by the tunnel oxide mask). Once themask 365 is in place, an etch is used to remove the unmasked polysilicon while minimizing damage to the tunnel window edge. Since charging and tunnel oxide removal along the polysilicon edge may degrade the tunnel oxide reliability and endurance, the etch should be highly selective to polysilicon over SiO2, approximately 20-100:1, so as to minimize the amount of tunnel oxide removed from along the edges of the polysilicon. The etch should also minimize the charge on the polysilicon after the floating gates are isolated during the etch. An example of a suitable etch is a reactive ion etch (RIE) using SF6/Cl2. Once themask 365 is removed, a strip of polysilicon (or polysilicon/silicide) 370 (the floating gate extension) intersecting thetunnel oxide 358 remains. - As shown in FIG. 6, a
tunnel window 475 is defined by the intersection of thetunnel oxide 358 and the polysilicon floatinggate extension 370. As such, the tunnel window is self-aligned with thepolysilicon strip 370 thereby facilitating its precise placement over the memory diffusion and its precise scaling. Among the advantages of this design are that no edge of the tunnel window is defined by field oxide. Any field oxide stress region-associated defects in the tunnel window are thereby avoided. Also, the difficulty of cutting small scale tunnel oxide “holes” in a gate oxide region is not faced. Further, the design does not result in excessive MD/gate oxide/floating gate capacitance. - FIGS.7A-D show cell layouts for preferred implementations of the tunnel window design of the present invention. FIG. 7A is a top view of an
EEPROM cell 500 having active region diffusions 508, 511 and 512, and having a single layer of polysilicon forming a floatinggate 502 straddling all three diffusions. The tunnel window 503 is shown as the intersection of the floatinggate extension 504 and atunnel oxide region 506 overlying a heavily dopedmemory diffusion 509. At the other end of the floatinggate 502, a wide area plate 510 is employed as one electrode of a capacitor enabling the floatinggate 502 to be raised to a high voltage (about 6.5 to 10.5 volts) by capacitively coupling from a second electrode 512 (heavily doped N+ silicon) through a gate oxide 514 (often about 200 Å thick). Between these two extremes is a section of polysilicon that forms the gate 516 of a read transistor. This layout also includes a polysilicon single row-line select 518 (word line) for controlling read and write operations. FIG. 7B is a top view of a second single poly EEPROM layout having the same elements as FIG. 7A, but with a dual row-line select 520 a and 520 b, rather than a single. This dual row-line select 520 a and 520 b allows independent control of the read operation (520 a) and the write operation (520 b). - FIG. 7C is a top view of an EEPROM cell530 having a double layer polysilicon stack 532 which includes a floating gate (lower layer) and a control gate 540 (upper layer). Double layer polysilicon stack 532 straddles diffusions 538 and 541 as shown. The
tunnel window 533 is again shown as the intersection of a floatinggate extension 534 and a tunnel oxide region 536 overlying a heavily dopedmemory diffusion 539. This layout also includes a polysilicon single row-line select 542 as in the embodiment of FIG. 7A. FIG. 7D is a top view of a second double poly EEPROM layout having the same elements as FIG. 7C, but with a dual row-line select 544, rather than a single. This dual row-line select 544 is analogous to dual row-line select 520 a and 520 b in FIG. 7B. - Note that all four layouts use separate transistors for the read and write (program) paths. This allows for the read current to be high without affecting the programming function of the cell.
- Although a specific embodiment of the present invention has been described in detail, it should be understood that the present invention may be embodied in many other specific forms without departing from the spirit or scope of the invention as recited in the claims. For example, other suitable materials, well known in the art, may be substituted for the substrate, oxides and floating gate materials used in the described embodiment.
Claims (21)
1. An EEPROM comprising:
(a) a semiconductor substrate having a surface;
(b) a doped memory diffusion region in the semiconductor substrate;
(c) a tunnel oxide on said substrate surface and overlying at least a portion of said memory diffusion;
(d) a floating gate structure including an extension overlying at least a portion of the tunnel oxide overlying said memory diffusion; and
(e) a self-aligned tunnel window defined within at least a portion of the tunnel oxide and having at least two edges defined by the floating gate extension, whereby when a defined voltage is applied to said memory diffusion a tunnel current sufficient to change the state of said EEPROM flows between said memory diffusion and said floating gate structure.
2. The EEPROM of claim 1 , wherein said tunnel window has two edges defined by edges of the floating gate extension.
3. The EEPROM of claim 1 , further comprising one or more field oxide regions separated from the tunnel window.
4. The EEPROM of claim 3 , further comprising a gate oxide located between said one or more field oxide regions and said tunnel window.
5. The EEPROM of claim 1 , wherein the tunnel window is defined by the intersection of the floating gate extension and the tunnel oxide.
6. The EEPROM of claim 5 , wherein the floating gate extension and the tunnel oxide are both rectangular shaped.
7. The EEPROM of claim 1 , wherein said memory diffusion is n-doped.
8. An EEPROM comprising:
(a) a semiconductor substrate having a surface;
(b) a n-doped memory diffusion region in the semiconductor substrate;
(c) a tunnel oxide on said substrate surface, surrounded by an area of gate oxide, and overlying at least a portion of said memory diffusion;
(d) a floating gate structure including an extension overlying at least a portion of the tunnel oxide overlying said memory diffusion;
(e) a write column electrically connected with a write transistor having as one source/drain region said memory diffusion;
(f) a read column electrically connected to a read transistor having as its gate electrode said floating gate; and
(g) a self-aligned tunnel window defined by the intersection of the floating gate extension and the tunnel oxide, whereby when a defined voltage is applied to said memory diffusion through said write column a tunnel current sufficient to change the state of said EEPROM flows between said memory diffusion and said floating gate structure and may be read through said read column.
9. A method of forming an EEPROM, the method of comprising:
(a) forming a doped memory diffusion region in a semiconductor substrate;
(b) forming a tunnel oxide on said substrate and overlying at least a portion of the memory diffusion; and
(c) forming a floating gate structure including an extension overlying at least a portion of the tunnel oxide such that the floating gate extension defines at least two edges of a tunnel window within at least a portion of the tunnel oxide, whereby when a defined voltage is applied to said memory diffusion a tunnel current sufficient to change the state of said EEPROM flows between said memory diffusion and said floating gate structure.
10. The method of claim 9 , wherein forming the floating gate structure includes
(i) depositing a blanket layer of gate electrode material over the semiconductor substrate and tunnel oxide;
(ii) masking the blanket layer to define boundaries of the floating gate structure; and
(iii) etching the blanket layer to form the floating gate structure.
11. The method of claim 9 , wherein the etching is performed by a reactive ion etch including a plasma formed from SF6 and Cl2, which selectively etches the blanket layer rather than the tunnel oxide located beyond the edges of the floating gate extension.
12. The method of claim 9 , wherein forming the floating gate forms two edges of the floating gate extension which define two edges of the tunnel window.
13. The method of claim 9 , wherein forming the tunnel oxide forms a rectangular shaped tunnel oxide and forming the floating gate structure forms a rectangular shaped floating gate extension intersecting the tunnel oxide, and wherein the intersection of the tunnel oxide and the floating gate extension defines the tunnel window.
14. A method of forming an EEPROM, the method of comprising:
(a) generating a p-doped region at a semiconductor substrate surface; and
(b) providing a first mask on the substrate surface, said first mask defining boundaries of a field oxide region; and
(c) generating a field oxide over the area of the substrate surface unmasked by said first mask; and
(d) providing a second mask on said semiconductor substrate, said second mask defining boundaries of a memory diffusion region in at least a portion of the substrate surface free of field oxide; and
(e) performing an n-type ion implant over the substrate to create said memory diffusion in the area of the substrate surface unmasked by said second mask; and
(f) providing a third mask on the semiconductor substrate, said third mask defining boundaries of a tunnel oxide region over at least a portion of the memory diffusion; and
(g) generating a tunnel oxide over the area of the substrate surface unmasked by said third mask; and
(h) generating a blanket of a suitable floating gate material over at least a portion of the area defining the intersection of the tunnel oxide and the memory diffusion; and
(i) providing a fourth mask on the floating gate material, said fourth mask defining the boundaries of a floating gate, said floating gate comprising an extension overlying at least a portion of the tunnel oxide such that the floating gate defines at least two edges of a tunnel window within at least a portion of the intersection of the tunnel oxide and the memory diffusion; and
(j) removing that part of the floating gate material unmasked by the fourth mask.
15. The method of claim 14 , further comprising removing the masks provided in steps (b), (d), (f) and (i) following steps (c), (e), (g) and (j), respectively.
16. The method of claim 14 , wherein the memory diffusion is bordered on at least one side by field oxide.
17. The method of claim 14 , wherein the memory diffusion has a depth of between about 0.4 and 0.8 μm and a concentration of about 1×1018 to 1×1020 cm−3.
18. The method of claim 14 , further comprising:
(i) generating a gate oxide on the exposed substrate surface following step (e); and
(ii) providing a fifth mask, said fifth mask defining the boundaries of a tunnel oxide region; and
(iii) removing the gate oxide in the area unmasked by said fifth mask.
19. The method of claim 14 , wherein the substrate material is silicon, the field oxide and tunnel oxide are silicon dioxide, and the floating gate material is polysilicon.
20. The method of claim 14 , wherein the tunnel window is separated from the field oxide-induced stress region.
21. The method of claim 14 , wherein the removal of the blanket of floating gate material in step (j) is performed by a reactive ion etch including a plasma formed from SF6 and Cl2, which selectively etches the blanket layer rather than the tunnel oxide located beyond the edges of the floating gate extension.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/235,198 US6365929B2 (en) | 1996-08-08 | 1999-01-22 | Scalable tunnel oxide window with no isolation edges |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US2372596P | 1996-08-08 | 1996-08-08 | |
US08/810,642 US5904524A (en) | 1996-08-08 | 1997-03-03 | Method of making scalable tunnel oxide window with no isolation edges |
US09/235,198 US6365929B2 (en) | 1996-08-08 | 1999-01-22 | Scalable tunnel oxide window with no isolation edges |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US08/810,642 Division US5904524A (en) | 1996-08-08 | 1997-03-03 | Method of making scalable tunnel oxide window with no isolation edges |
Publications (2)
Publication Number | Publication Date |
---|---|
US20020011625A1 true US20020011625A1 (en) | 2002-01-31 |
US6365929B2 US6365929B2 (en) | 2002-04-02 |
Family
ID=26697524
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US08/810,642 Expired - Lifetime US5904524A (en) | 1996-08-08 | 1997-03-03 | Method of making scalable tunnel oxide window with no isolation edges |
US09/235,198 Expired - Fee Related US6365929B2 (en) | 1996-08-08 | 1999-01-22 | Scalable tunnel oxide window with no isolation edges |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US08/810,642 Expired - Lifetime US5904524A (en) | 1996-08-08 | 1997-03-03 | Method of making scalable tunnel oxide window with no isolation edges |
Country Status (1)
Country | Link |
---|---|
US (2) | US5904524A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030023743A1 (en) * | 2001-07-26 | 2003-01-30 | Raphel Jose Kolencheril | System, method and computer program product to maximize server throughput while avoiding server overload by controlling the rate of establishing server-side net work connections |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6187634B1 (en) * | 1997-11-19 | 2001-02-13 | Altera Corporation | Process for making an EEPROM active area castling |
US6624467B1 (en) | 1997-11-19 | 2003-09-23 | Altera Corporation | EEPROM active area castling |
US6757196B1 (en) | 2001-03-22 | 2004-06-29 | Aplus Flash Technology, Inc. | Two transistor flash memory cell for use in EEPROM arrays with a programmable logic device |
US6369422B1 (en) | 2001-05-01 | 2002-04-09 | Atmel Corporation | Eeprom cell with asymmetric thin window |
US7144775B2 (en) * | 2004-05-18 | 2006-12-05 | Atmel Corporation | Low-voltage single-layer polysilicon eeprom memory cell |
WO2012036752A2 (en) * | 2010-09-17 | 2012-03-22 | Aplus Flash Technology, Inc. | An eeprom-based, data-oriented combo nvm design |
US8629032B2 (en) * | 2011-01-18 | 2014-01-14 | Systems On Silicon Manufacturing Co. Pte. Ltd. | Non-volatile memory cell structure and a method of fabricating the same |
US9312014B2 (en) * | 2013-04-01 | 2016-04-12 | SK Hynix Inc. | Single-layer gate EEPROM cell, cell array including the same, and method of operating the cell array |
US9236453B2 (en) | 2013-09-27 | 2016-01-12 | Ememory Technology Inc. | Nonvolatile memory structure and fabrication method thereof |
TWI546903B (en) * | 2015-01-15 | 2016-08-21 | 聯笙電子股份有限公司 | Non-volatile memory cell |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4822750A (en) * | 1983-08-29 | 1989-04-18 | Seeq Technology, Inc. | MOS floating gate memory cell containing tunneling diffusion region in contact with drain and extending under edges of field oxide |
JPS61222175A (en) * | 1985-03-01 | 1986-10-02 | Fujitsu Ltd | Manufacture of semiconductor memory device |
IT1232354B (en) * | 1989-09-04 | 1992-01-28 | Sgs Thomson Microelectronics | PROCEDURE FOR THE CREATION OF EEPROM MEMORY CELLS AT A SINGLE LEVEL OF POLYSILICIO AND THIN OXIDE USING DIFFERENTIAL OXIDATION. |
US5108939A (en) * | 1990-10-16 | 1992-04-28 | National Semiconductor Corp. | Method of making a non-volatile memory cell utilizing polycrystalline silicon spacer tunnel region |
US5411904A (en) | 1990-11-19 | 1995-05-02 | Sharp Kabushiki Kaisha | Process for fabricating nonvolatile random access memory having a tunnel oxide film |
US5379253A (en) * | 1992-06-01 | 1995-01-03 | National Semiconductor Corporation | High density EEPROM cell array with novel programming scheme and method of manufacture |
US5225362A (en) * | 1992-06-01 | 1993-07-06 | National Semiconductor Corporation | Method of manufacturing a full feature high density EEPROM cell with poly tunnel spacer |
US5293331A (en) * | 1992-06-01 | 1994-03-08 | National Semiconductor Corporation | High density EEPROM cell with tunnel oxide stripe |
US5412238A (en) * | 1992-09-08 | 1995-05-02 | National Semiconductor Corporation | Source-coupling, split-gate, virtual ground flash EEPROM array |
US5404037A (en) * | 1994-03-17 | 1995-04-04 | National Semiconductor Corporation | EEPROM cell with the drain diffusion region self-aligned to the tunnel oxide region |
US5501996A (en) * | 1994-12-14 | 1996-03-26 | United Microelectronics Corporation | Method of manufacture of high coupling ratio single polysilicon floating gate EPROM or EEPROM cell |
US5570314A (en) * | 1994-12-28 | 1996-10-29 | National Semiconductor Corporation | EEPROM devices with smaller cell size |
-
1997
- 1997-03-03 US US08/810,642 patent/US5904524A/en not_active Expired - Lifetime
-
1999
- 1999-01-22 US US09/235,198 patent/US6365929B2/en not_active Expired - Fee Related
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030023743A1 (en) * | 2001-07-26 | 2003-01-30 | Raphel Jose Kolencheril | System, method and computer program product to maximize server throughput while avoiding server overload by controlling the rate of establishing server-side net work connections |
Also Published As
Publication number | Publication date |
---|---|
US6365929B2 (en) | 2002-04-02 |
US5904524A (en) | 1999-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US4822750A (en) | MOS floating gate memory cell containing tunneling diffusion region in contact with drain and extending under edges of field oxide | |
US4701776A (en) | MOS floating gate memory cell and process for fabricating same | |
KR100274491B1 (en) | Spacer Flash Cell Process | |
US6351017B1 (en) | High voltage transistor with modified field implant mask | |
KR100231964B1 (en) | Method of making a non-volatile memory cell utilizing polycrystalline silicon spacer tunnel region | |
KR100372151B1 (en) | EEPROM cell having self-aligned drain diffusion region in tunnel oxide region and its manufacturing method | |
US5674762A (en) | Method of fabricating an EPROM with high voltage transistors | |
EP0389721A2 (en) | Flash EPROM cell and method of making such cell | |
KR100316089B1 (en) | " EEPROM " omitted | |
US6261903B1 (en) | Floating gate method and device | |
US6586806B1 (en) | Method and structure for a single-sided non-self-aligned transistor | |
US6184554B1 (en) | Memory cell with self-aligned floating gate and separate select gate, and fabrication process | |
EP0160003B1 (en) | Mos floating gate memory cell and process for fabricating same | |
US5904524A (en) | Method of making scalable tunnel oxide window with no isolation edges | |
US6313498B1 (en) | Flash memory cell with thin floating gate with rounded side wall, and fabrication process | |
EP1506573B1 (en) | Manufacturing method for ultra small thin windows in floating gate transistors | |
EP0560435B1 (en) | Method of manufacturing a floating gate transistor with tunnel oxide | |
US6573142B1 (en) | Method to fabricate self-aligned source and drain in split gate flash | |
US6806143B2 (en) | Self-aligned source pocket for flash memory cells | |
US20020055228A1 (en) | Sidewall process to improve the flash memory cell performance | |
US20060006453A1 (en) | Nonvolatile semiconductor memory device and method of fabricating the same | |
EP0450401B1 (en) | Method of manufacturing non-volatile semiconductor memory device | |
US6534364B1 (en) | Tunnel diode layout for an EEPROM cell for protecting the tunnel diode region | |
JPH06237000A (en) | Nonvolatile memory cell and manufacture thereof | |
US4683640A (en) | Method of making a floating gate memory cell |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FPAY | Fee payment |
Year of fee payment: 4 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
REMI | Maintenance fee reminder mailed | ||
LAPS | Lapse for failure to pay maintenance fees | ||
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20140402 |