Classifications

• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10B—ELECTRONIC MEMORY DEVICES
  • H10B43/00—EEPROM devices comprising charge-trapping gate insulators
  • H10B43/30—EEPROM devices comprising charge-trapping gate insulators characterised by the memory core region
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D30/00—Field-effect transistors [FET]
  • H10D30/01—Manufacture or treatment
  • H10D30/021—Manufacture or treatment of FETs having insulated gates [IGFET]
  • H10D30/0411—Manufacture or treatment of FETs having insulated gates [IGFET] of FETs having floating gates
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D30/00—Field-effect transistors [FET]
  • H10D30/01—Manufacture or treatment
  • H10D30/021—Manufacture or treatment of FETs having insulated gates [IGFET]
  • H10D30/0413—Manufacture or treatment of FETs having insulated gates [IGFET] of FETs having charge-trapping gate insulators, e.g. MNOS transistors
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D30/00—Field-effect transistors [FET]
  • H10D30/60—Insulated-gate field-effect transistors [IGFET]
  • H10D30/69—IGFETs having charge trapping gate insulators, e.g. MNOS transistors

Definitions

• This invention relates generally to semiconductor devices, and more specifically, to semiconductor devices for use in memory cells.
• HCI hot-carrier electron injection
• the Nt may increase to a point where the state of the memory cell may change from a low Nt state to a high Nt state, thus resulting in a reliability failure of the memory cell. Therefore, a need exists for a memory cell with increased reliability during repeated reads.
• FIG. 1 illustrates a cross-sectional view of a semiconductor substrate having well implants and channel implants formed therein in accordance with an embodiment of the present invention
• FIG. 2 illustrates a cross-sectional in view of the semiconductor substrate of FIG. 1 having a gate stack formed over the semiconductor substrate in accordance with an embodiment of the present invention
• FIG. 3 illustrates a cross-sectional view of the gate stack of FIG. 2 after formation of a halo implant in accordance with an embodiment of the present invention
• FIG. 4 illustrates the semiconductor device of FIG. 3 after forming source and drain regions and extension regions within the semiconductor substrate and sidewall spacers along the sidewalls of the gate stack in accordance with an embodiment of the present invention
• FIG. 5 illustrates a cross-sectional view of a semiconductor substrate having well implants formed therein in accordance with an alternate embodiment of the present invention
• FIG. 6 illustrates a cross-sectional view of the semiconductor substrate of FIG. 5 having a first oxide layer, a nitride layer, and a second oxide layer formed over the semiconductor substrate and a channel implant in accordance with an embodiment of the present invention
• FIG. 7 illustrates a cross-sectional view of the semiconductor substrate of FIG. 6 after formation of a gate stack in accordance with an embodiment of the present invention.
• FIG. 8 illustrates the semiconductor device of FIG. 7 after forming source and drain regions and extension regions within the semiconductor substrate and sidewall spacers along the sidewalls of the gate stack in accordance with an embodiment of the present invention.
• a semiconductor device which may be used as a NNM memory cell is formed having an anti-punch through (APT) region and an optional drain side highly doped region (halo).
• the halo region if present, results in an increased dopant gradient between a channel region and a drain region of the semiconductor device.
• the APT region allows for the channel region to have a relatively low dopant concentration or be counter doped with respect to the APT region which minimizes read disturb (i.e. threshold voltage drift during a read cycle) by lowering the natural Nt. Therefore, use of the halo region and APT regions allows for efficient hot carrier injection programming of the semiconductor device to be maintained while reducing the read disturb.
• FIG. 1 illustrates a semiconductor device 10 including a semiconductor substrate 12 having isolation trenches 22 and 24, surrounding ⁇ -type wells 14 and 18, isolating ⁇ -type well 16 between isolation trenches 22 and 24, and a masking layer 30.
• isolation trenches 22 and 24, surrounding N-type wells 14 and 18, isolating N- type well 16, and masking layer 30 are known in the art and will only briefly be described herein.
• Isolation trenches 22 and 24 are formed in substrate 12, and afterwards, surrounding N-type wells 14 and 18 are formed.
• Isolation trenches 22 and 24 may include any type of insulating material, such as, for example, oxide, nitride, etc., or any combination thereof.
• a patterned masking layer 30 is used to define an opening between isolation trenches 22 and 24.
• patterned masking layer 30 can be any type of masking layer, such as, for example, a photo resist layer, a hard mask, etc.
• Isolating N-type well 16 is then formed within substrate 12.
• an isolated P-type well 20 is formed within isolating N-type well 16, such that P-type well 20 is isolated from substrate 12.
• an anti-punch through (APT) region 26 and channel region 28 are formed between isolation trenches 22 and 24.
• APT region 26 and channel region 28 may be formed in any order.
• Channel region 28 and APT region 26 are formed such that channel region 28 is located between a top surface of substrate 12 and APT region 26, and APT region 26 is located between channel region 28 and isolated P-type well 20.
• APT region 26 may also be referred to as highly doped region 26.
• a dopant used in the formation of APT region 26 is chosen such that it does not significantly diffuse into channel region 28. Arrows 31 illustrate that the dopant is applied uniformly to substrate 12.
• the direction of the implant for both APT region 26 and channel region 28 is substantially perpendicular to substrate 12. That is, the direction is no greater than approximately 10 degrees from vertical. Also note that the dopant concentration of APT region 26 is greater than the dopant concentration of isolated P-type well 20.
• APT region 26 and channel region 28 are formed such that the dopant concentration of channel region 28 is less than the dopant concentration of APT region 26.
• APT region 26 and channel region 28 are formed using P- type dopants, such as, for example, boron or indium.
• the dopant concentration of channel region 28 may be ten to fifty times lower than the dopant concentration of APT region 26.
• APT region 26 may therefore be implanted with an energy in a range of approximately 30 to 50 kilo electron-volts (keN) and a dosage in a range of approximately lx 10 12 /cm 2 to 1 x 10 14 /cm 2
• channel region 28 may be implanted with an energy in a range of approximately 5 to 30 keN and a dosage in a range of approximately lx 10 ⁇ /cm 2 to 1 x 10 13 /cm 2
• different P-type dopants may be used for channel region 28 and APT region 26, such as, for example, boron for channel region 28 and indium for APT region 26.
• a same P-type dopant may be used for both regions.
• the semiconductor substrate 12 is a bulk substrate.
• substrate 12 is a semiconductor-containing substrate and may include silicon, gallium arsenide, silicon germanium, etc., or any combination thereof.
• substrate 12 may be a silicon on insulator (SOI) substrate (not shown) having a bottom semiconductor layer, a buried insulating layer overlying the bottom semiconductor layer, and a top semiconductor layer.
• SOI silicon on insulator
• surrounding N-type wells 14 and 18 and isolating N-type well 16 are not needed. That is, isolated P-type well 20 would correspond to the top semiconductor layer of the SOI substrate.
• buried insulating layer can be a silicon oxide layer and top and bottom semiconductor layers may be formed of silicon, germanium, gallium arsenide, or the like.
• FIG. 2 illustrates semiconductor device 10 after removal of masking layer 30 and formation of a SONOS gate stack 32 over channel region 28, between isolation wells 22 and
• SONOS gate stack 32 includes a first oxide 40 formed over channel region 28, a nitride 38 formed over first oxide 40, a second oxide 36 formed over nitride 38, and a gate 34 formed over second oxide 36.
• first oxide 40, nitride 38, and second oxide 36 may be referred to as an oxide-nitride-oxide structure.
• Masking layer 30 can be removed using conventional processing.
• a first oxide layer is blanket deposited or grown over semiconductor substrate 12 using chemical vapor deposition (CVD) or a thermal oxidation process, respectively.
• the first oxide layer may be formed by physical vapor deposition (PND), atomic layer deposition (ALD), thermal oxidation, the like or combinations of the above.
• a nitride layer is deposited over the first oxide layer.
• the nitride layer may formed by CND, PND, ALD, the like or combinations thereof.
• a second oxide layer is blanket deposited on the nitride layer using chemical vapor deposition (CND) or a thermal oxidation process, respectively.
• the second oxide layer may be formed by physical vapor deposition (PND), atomic layer deposition (ALD), thermal oxidation, the like or combinations of the above.
• a gate layer is blanket deposited over the second oxide layer formed by CND, PND, ALD, the like or combinations thereof.
• the first oxide layer, nitride layer, second oxide layer, and gate layer may then be patterned and etched to form the resulting gate stack 32. (Note than in alternate embodiments, each layer of the stack may be patterned and etched individually to form the resulting gate stack 32.)
• the resulting gate stack 32 (and likewise, the portion of channel region 28 below gate stack 32) has a length in a range of approximately of 0.35 microns to 0.06 microns.
• Gate 34 of gate stack 32 may be any conductive material, such as polysilicon or a metal-containing material, and may be referred to as a control gate.
• First oxide 40 and second oxide 36 can be any dielectric, such as, for example, an insulating material or stack of insulating materials, such as, for example, silicon oxide, oxynitride, metal-oxide, nitride, etc., or any combination thereof.
• Nitride 38 may be a silicon nitride, oxynitride, or any other material known to have charge traps such that the charges can be stored therein.
• first oxide 40 and second oxide 36 may also be referred to as first and second insulating layers, respectively, or bottom and top dielectrics, respectively, and nitride 38 may be referred to as a charge storing layer, a storage element, or a dielectric.
• gate stack 32 is illustrated as a SONOS stack, in alternate embodiments, gate stack 32 may be any type of NVM gate stack.
• gate stack 32 may be replaced by a floating gate stack (not shown) having a tunnel dielectric formed over channel region 28, between isolation trenches 22 and 24, a floating gate formed over the tunnel dielectric, a control dielectric formed over the floating gate, and a control gate over the control dielectric.
• a tunnel dielectric layer is formed overlying semiconductor substrate 12 by CVD, PND, ALD, thermal oxidation, the like, or combination thereof.
• the tunnel dielectric layer can be any insulating material, such as an oxide (e.g.
• tunnel dielectric layer is then patterned and etched using conventional processing to form the tunnel dielectric of the floating gate stack overlying channel region 28 (where the tunnel dielectric is located in a similar location as oxide 40 of gate stack 32 illustrated in FIG. 2).
• a floating gate layer is then formed over the semiconductor substrate 12 and the tunnel dielectric by CND, PND, ALD, the like, or combinations thereof.
• the floating gate layer may be any conductive material, such as polysilicon, metal, or the like.
• floating gate layer may be a plurality of nanocrystals (i.e. discrete storage elements) such as in a nanocrystal ⁇ NM device. The floating gate layer is then patterned and etched using conventional processing to form the floating gate of the floating gate stack overlying the tunnel dielectric.
• control dielectric layer is then formed over the semiconductor substrate 12 and the floating gate by CND, PND, ALD, thermal oxidation, the like, or combinations thereof.
• the control dielectric layer is then patterned and etched using conventional processing to form the control dielectric of the floating gate stack overlying the floating gate.
• the control dielectric is optional and may not be formed in all floating gate devices.
• the control dielectric layer can be any insulating material, such as an oxide (e.g. silicon dioxide), nitride, metal oxide, high dielectric constant material (i.e. a material having a dielectric constant of greater than approximately 4 and less than approximately 15), the like, or combinations thereof.
• control gate layer is then formed over the semiconductor substrate 12 and the control dielectric by CND, PND, ALD, the like, or combinations thereof.
• Control gate layer may be any conductive material, such as polysilicon or a metal-containing material.
• the control gate layer is patterned and etched to form the control gate of the floating gate stack overlying the control dielectric. (Note that in alternate embodiments, rather than patterning and etching each layer of the floating stack separately, combination of layers or all the layers may be patterned and etched using a same pattern and etch process in order to reduce processing steps required to form the resulting floating gate stack.)
• a patterned masking layer 42 is formed using conventional masldng processes.
• masldng layer 42 can be any type of masking layer, such as, for example, photo resist or a hard mask.
• Patterned masking layer 42 (also referred to as an implant mask) masks a source side of semiconductor device 10 (at a first side of gate stack 32, in which a source region will later be formed) while exposing a drain side of semiconductor device 10 (at a second side of gate stack 32, opposite the first side, in which a drain region will later be formed).
• an angled implant 44 is used to form a halo region 46 which extends beneath gate stack 32 by a distance 47 as measured from a first edge of gate stack 32.
• distance 47 is at most approximately 500 Angstroms.
• Angled implant 44 has a corresponding angle of implant ⁇ , where ⁇ is measured from vertical. In one embodiment, ⁇ is in a range of approximately 20 to 60 degrees, and more preferably, approximately 30 to 40 degrees. The angle of implant 44 is therefore sufficient to increase the dopant concentration in halo region 46 at a region 45 beneath gate stack 32 such that it is greater than the dopant concentration of channel region 28.
• halo region 46 is implanted using a P-type dopant, such as, for example, boron or indium, at an energy in a range of approximately 10 to 50 keN having a dosage in a range of approximately lxl0 12 /cm 2 to lxl0 14 /cm 2 .
• a P-type dopant such as, for example, boron or indium
• halo region 46 may be referred to as angled halo 46 or as a highly or heavily doped region 46.
• the dopant concentration of halo region 46 is generally greater than the dopant concentration of isolated P-type well 20.
• FIG. 4 illustrates semiconductor device 10 after removal of masking layer 42 and the formation of sidewall spacers 48 and 50, source and drain extensions 51 and 53, and source and drain regions 52 and 54.
• Masldng layer 42 can be removed using conventional processing steps.
• source extension 51 and drain extension 53 are formed using conventional masldng and implanting processes. Note that extensions 51 and 53 extend into channel region 28 and each underlie a portion of gate stack 32.
• an N-type dopant such as arsenic, phosphorous, or antimony, is implanted at an energy in a range of approximately 30 to 70 keN having a dosage in a range of approximately lxl0 14 /cm 2 to lxl0 15 /cm 2 to form extensions 51 and 53.
• Drain extension 53 is formed such that it does not extend beyond halo region 46. Note that after formation of drain extension 53, an increasing dopant gradient results from channel region 28 to drain extension 53. Although an increasing dopant gradient exists from channel region 28 to drain extension 53 without halo region 46, the presence of halo region 46 further increases this dopant gradient. Also, the presence of halo region 46 allows for a relatively low dopant concentration within channel region 28.
• spacers 48 and 50 are formed along the sidewalls of gate stack 32 using conventional processing steps. These spacers, for example, may include any insulating material, such as, for example, oxide or nitride. Alternatively, spacers 48 and 50 may not be present. If spacers 48 and 50 are not present, then source and drain regions 52 and 54 may not be formed such that extensions 51 and 53 are used as the source and drain regions, respectively. However, with the presence of spacers 48 and 50, source and drain regions may be formed using another implant step.
• an N-type dopant such as arsenic, phosphorous, or antimony
• an energy in a range of approximately 10 to 30 keN having a dosage in a range of approximately 1 10 /cm to 5xl0 16 /cm 2 to form source region 52 and drain region 54.
• drain and source regions 52 and 54 do not extend below isolation trenches 22 and 24.
• the depth of APT 26 is selected such that it does not extend below the depth of source and drain regions 52 and 54.
• further conventional processing may be used to complete semiconductor device 10. For example, contacts may be formed to the source region 52, gate 34, drain region 54, and isolated P-type well 20. Also, other semiconductor device levels may be formed underneath or above semiconductor device 10.
• Vw 60 corresponds to the voltage applied to isolated P-type well 20
• Vs 62 corresponds to the voltage applied to source region 52
• Ng 64 corresponds to the voltage applied to gate 34
• Nd 66 corresponds to the voltage applied to drain region 54.
• semiconductor device 10 may be used as an ⁇ NM memory cell within an ⁇ NM memory (not shown).
• a high Nt state corresponds to a program state of the memory cell
• a low Nt state corresponds to an erase state of the memory cell. (Note, however, that in alternate embodiments, the program and erase states may be reversed.)
• Semiconductor device 10 is erased by removing electrons from nitride 38 which results in semiconductor device 10 having a low Nt (such as, for example, below approximately 2 volts).
• a low Nt such as, for example, below approximately 2 volts.
• Many known methods may be used to place semiconductor device 10 into a low Nt state, such as, for example, Fowler- ⁇ ordheim tunneling, hot hole injection, direct tunneling, etc.
• semiconductor device 10 is programmed by storing electrons within nitride 38 which results in semiconductor device 10 having a high Nt (such as, for example, above approximately 4 volts). Therefore, semiconductor device 10 may be programmed by applying a drain voltage (Nd) and a source voltage (Ns) where Nd is approximately 3 to 5 volts greater than Ns. For example, in one embodiment, a Ns of 1 volt and a Nd of 4 volts may be used. In this embodiment, a gate voltage (Ng) of approximately 5 to 10 volts and a well voltage (Vw) of approximately 0 to -3 volts is applied.
• Nd drain voltage
• Ns source voltage
• Ng gate voltage
• Vw well voltage
• the natural Nt of semiconductor device 10 refers to the threshold voltage prior to placing any charge into nitride 38.
• the read disturb is degraded.
• read disturb describes the gradual increase in threshold voltage (Nt) as the low Nt memory cell is continuously read, i.e. the threshold voltage drift during a read cycle.
• the time to failure of the memory cell decreases. That is, as natural Nt increases, a smaller number of reads to the memory cell results in failure due to the drift from a low Nt to a high Nt. Therefore, by decreasing the natural Nt, read disturb of the low Nt state is improved (i.e. threshold voltage drift is reduced). For example, referring back to FIG.
• a read of semiconductor device 10 may be performed by applying a Nd that is approximately 0.5 to 1.5 volts greater than Ns.
• Ns may be 0 volts and Nd may be 1 volt.
• a Ng and Nw sufficient to produce approximately 10 to 30 microamperes of current in channel region 28 is applied.
• a Ng of 2 volts and a Nw of 0 volts may be used.
• Ns is increased by 1 volt
• Nd, Ng, and Nw are also increased by 1 volt.
• an inversion layer is formed in channel region 28 and a depletion region (not shown) is formed around drain region 54 and drain extension 53.
• This depletion region substantially masks the dopant gradient created in halo region 46 thereby preventing the higher dopant of halo region 46 from increasing the Nt of semiconductor 10. In this manner, the Nt remains in a low Nt state, thus improving the read disturb by reducing Vt drift.
• gate stack 32 being in a range of approximately 0.35 to 0.06 microns as was described above, a short channel leakage may result during programming of semiconductor device 10.
• highly doped APT region 26 also functions to reduce this short channel leakage, thereby reducing power consumption and improving programming efficiency.
• FIGs. 5-8 illustrates an alternate embodiment of the present invention where rather than forming channel region 28 and APT region 26 using dopants of the same conductivity type, two implant steps using dopants of different conductivity types may be used to form a channel region 86 and an APT region 74 instead. That is, in this alternate embodiment, channel region 28 and APT region 26 can be replaced with channel region 86 and APT region 74, respectively, which function in a similar manner to channel region 28 and APT region 26 described above to allow for efficient hot carrier injection programming of the semiconductor device while reducing the read disturb. Also, as will be described below, in this alternate embodiment, halo region 46 may not be present. (Note that in the following descriptions of FIGs. 5-8, reference numerals which are the same as reference numerals used in the description of FIGs. 1-4 indicate like or similar elements.)
• FIG. 5 illustrates a semiconductor device 70 including a semiconductor substrate 12 having isolation trenches 22 and 24, surrounding N-type wells 14 and 18, isolating N-type well 16 between isolation trenches 22 and 24, and patterned masking layer 30.
• isolation trenches 22 and 24, surrounding N-type wells 14 and 18, isolating N- type well 16, and masking layer 30 are the same as was described in reference to FIG. 1 above, and therefore will not be described again here in reference to FIG. 5.
• an APT region 74 is formed between isolation trenches 22 and 24 in isolated P-type well 20. (Note that APT region 74 may also be referred to as highly doped region 74.)
• Arrows 72 illustrate that the dopant is applied uniformly to substrate 12.
• the direction of the implant for APT region 74 is substantially perpendicular to substrate 12. That is, the direction is no greater than approximately 10 degrees from vertical.
• APT region 74 is formed using a P-type dopant, such as, for example, boron or indium.
• APT region 74 may be implanted with an energy in a range of approximately 30 to 50 keV and a dosage in a range of approximately lx 10 12 /cm 2 to 1 x 10 14 /cm 2 .
• the dopant of APT region 74 and isolated P-type well 20 are of the same conductivity type and the dopant concentration of APT region 74 is greater than the dopant concentration of isolated P-type well 20.
• the dopant concentration of APT region 74 is approximately 2 to 100 times greater than the dopant concentration of isolated P-type well 20.
• the dopant concentration of APT region 74 may be in a range of approximately 5xl0 17 cm “3 to 5xl0 18 cm “3
• the dopant concentration of isolated P-type well 20 may be in a range of approximately 5xl0 16 cm ⁇ 3 to 5xl0 17 cm "3 .
• first oxide layer 80 is blanket deposited or grown over semiconductor substrate 12 using chemical vapor deposition (CVD) or a thermal oxidation process, respectively.
• first oxide layer may be formed by physical vapor deposition (PND), atomic layer deposition (ALD), thermal oxidation, the like or combinations of the above.
• nitride layer 82 is deposited over first oxide layer 80.
• Nitride layer 82 may formed by CND, PND, ALD, the like or combinations thereof.
• Second oxide layer 84 is then blanket deposited over nitride layer 82 using chemical vapor deposition (CND) or a thermal oxidation process, respectively.
• CND chemical vapor deposition
• second oxide layer 84 may be formed by physical vapor deposition (PND), atomic layer deposition (ALD), thermal oxidation, the like or combinations of the above.
• a patterned masking layer 76 is used to define an opening between isolation trenches 22 and 24.
• patterned masking layer 76 can be any type of masking layer, such as, for example, a photo resist layer, a hard mask, etc.
• channel region 86 is formed in isolated P-type well 20.
• channel region 86 is formed using an N-type dopant, such as, for example, arsenic, phosphorous, or antimony.
• This N-type dopant may be implanted with an energy in a range of approximately 5 to 70 keN and a dosage in a range of approximately lx 10 ⁇ /cm 2 to 5 x 10 13 /cm 2 .
• N-type dopant compensates a portion of the existing P-type dopant of APT region 74 to form channel region 86.
• channel region 86 has a first conductivity type (such as N-type in this embodiment) and is located between a top surface of substrate 12 and APT region 74, and APT region 74 has a second conductivity type (such as P-type in this embodiment) and is located between channel region 86 and isolated P-type well 20.
• the N-type dopant concentration in channel region 86 should be higher than the P-type dopant concentration in APT region 74.
• the net doping concentration refers to the absolute difference between dopants of one conductivity type and dopants of another conductivity type.
• the net doping concentrations provided for channel region 86 refers to the absolute value of the difference between the P-type dopants of APT region 74 and N-type dopants of channel region 86.
• the concentration of P-type dopants in channel region 86 minus the concentration of N-type dopants in channel region 86 is less than or equal to the net doping concentration in isolated P-type well 20.
• the concentration of P-type dopants in channel region 86 minus the concentration of N-type dopants in channel region 86 may provide a negative number having an absolute value greater than the net doping concentration in isolated P-type well 20.
• the concentration of P-type dopants in channel region 86 minus the concentration of N-type dopants in channel region 86 may provide a negative number having an absolute value less than the net doping concentration in isolated P-type well 20.
• channel region 86 is formed after formation of first oxide layer 80, nitride layer 82, and second oxide layer 84.
• channel region 86 may be formed prior to formation of these layers. That is, after formation of APT region 74 described in reference to FIG. 5, a subsequent implant step can be used to form channel region 86 using the same patterned masking layer 30. Therefore, in this embodiment, patterned masking layer 76 would not be needed.
• FIG. 7 illustrates semiconductor device 70 after formation of gate stack 32.
• patterned masking layer 76 is removed (for example, using conventional processing).
• a gate layer is then blanket deposited over second oxide layer 84 formed by CVD, PVD, ALD, the like or combinations thereof.
• first oxide layer 80 results in first oxide 40
• etching of nitride layer 82 results in nitride 38
• the etching of second oxide layer 84 results in second oxide 36
• the etching of the gate layer results in gate 34.
• each layer of the stack may be patterned and etched individually to form the resulting gate stack 32.
• oxide layers 80 and 84 and nitride layer 82 can be patterned and etched prior to formation of channel region 86.
• the resulting gate stack 32 (and likewise, the portion of channel region 86 below gate stack 32) has a length in a range of approximately of 0.35 microns to 0.06 microns.
• gate stack 32 is illustrated as a SONOS stack in FIG. 7, in alternate embodiments, gate stack 32 may be any type of NVM gate stack, as was described above in reference to FIG. 3. Therefore, all the descriptions provided for gate stack 32 above apply to this embodiment as well. That is, all methods of formation, materials, and alternatives described above in reference to gate stack 32 of FIG. 3 apply again here to gate stack 32.
• gate stack 32 may be replaced by a floating gate stack (not shown) as was described above. However, note that if gate stack 32 is replaced by a floating gate stack, the floating gate may to be too thick to allow the proper penetration of implants for forming channel region 86. Therefore, in an embodiment using a floating gate stack, channel region 86 may be formed after forming APT region 74 and prior to forming any portion of the floating gate stack.
• a halo region such as halo region 46
• a halo region may be formed in isolated P-type well 20 as was described above in reference to FIG. 3. That is, after formation of gate stack 32, patterned masking layer 42 may be used to form halo region 46, as was described above in reference to FIG. 3.
• halo region 46 (not shown in FIGs. 7 and 8) would be adjacent to channel region 86 and APT region 74 (rather than channel region 28 and APT region 26).
• the same methods of formation, materials, and alternatives described for halo region 46 and angled implant 44 in reference to FIG. 3 can be applied to the current embodiment having channel region 86 and APT region 74 in place of channel region 28 and APT region 26. Note that in the current embodiment of FIGs. 5-8, halo region 46 may not be necessary due to the counter doping methods used to form channel region 86 and APT region 74.
• FIG. 8 illustrates semiconductor device 70 after removal of masldng layer 76, formation of gate stack 32, formation of halo region 46, and the formation of sidewall spacers 48 and 50, source and drain extensions 51 and 53, and source and drain regions 52 and 54.
• halo region 46, sidewall spacers 48 and 50, source and drain extensions 51 and 53, and source and drain regions 52 and 54 apply here in reference to FIG. 8. That is, the same methods of formation, materials, and alternatives described in reference to FIG. 4 apply to FIG. 8.
• halo region 46 is shown and hence, semiconductor device 70 of FIG. 8 is similar to semiconductor device 10 of FIG. 4, except that channel region 28 and APT region 26 of FIG.
• halo region 46 is adjacent to channel region 86 and APT region 74.
• halo region 46 may not be present.
• channel region 86 and APT region 74 would be adjacent to drain extension 53 and drain region 54.
• Vw 60 corresponds to the voltage applied to isolated P-type well 20
• Vs 62 corresponds to the voltage applied to source region 52
• Vg 64 corresponds to the voltage applied to gate 34
• Vd 66 corresponds to the voltage applied to drain region 54.
• semiconductor device 70 may be used as an NVM memory cell within an NVM memory (not shown).
• a high Vt state corresponds to a program state of the memory cell
• a low Vt state corresponds to an erase state of the memory cell. (Note, however, that in alternate embodiments, the program and erase states may be reversed.)
• Program and erase operations for semiconductor device 70 are the same as described above with reference to semiconductor device 10 of FIG. 4.
• hot carriers are generated in the drain depletion region, some of which are injected through oxide 40 into nitride 38. This results in increasing the Vt of semiconductor device 70.
• the dopant gradient that is created by halo region 46 and drain extension 53 amplifies this hot carrier injection thus maintaining efficient hot carrier programming of semiconductor device 70. This efficiency is maintained even with channel region 86 being counter doped relative to APT region 74.
• the counter doping of channel region 86 reduces the natural Vt of semiconductor device 70 thereby improving the read disturb, as will be described below.
• the natural Vt of semiconductor device 70 refers to the threshold voltage prior to placing any charge into nitride 38. As with semiconductor device 10, for a higher natural Vt of semiconductor device 70, the read disturb is degraded. Therefore, by decreasing the natural Vt, read disturb of the low Vt state is improved (i.e. threshold voltage drift is reduced).
• One of the ways that a lower natural Vt reduces read disturb is by enabling a lower
• Vt for the low Vt state.
• a gate bias (Vg) that exceeds the Vt of the low Vt state by a predetermined amount (typically referred to as gate overdrive) is necessary.
• the reduced Vt of the low Vt state allows for the reduction of the absolute gate bias (Vg) during a read operation while maintaining a constant gate overdrive.
• a reduced absolute gate bias (Vg) reduces the electric field across gate stack
• Vt of the low Vt state is too low (due to the counter doping of channel region 86)
• a source to drain leakage current can occur in unselected devices in a memory array containing semiconductor device 70.
• Unselected devices are those devices in the memory array which are not intended to be read during the read operation of semiconductor device 70.
• a reverse well to source bias increases the Vt of the low Vt state. Therefore the source to drain leakage current may be prevented by applying a reverse well to source bias to the unselected devices in the memory array during the read operation of semiconductor device 70.
• the reverse well to source bias should be sufficient to reduce the source to drain leakage current caused by the low Vt of the low Vt state. For example, referring back to FIG.
• a read of semiconductor device 70 may be performed by applying a Vd that is approximately 0.5 to 1.5 volts greater than Vs.
• Vs may be 0 volts and Vd may be 1 volt.
• a Vg and Vw sufficient to produce approximately 10 to 30 microamperes of current in channel region 28 is applied.
• Vs source voltage
• erased semiconductor device 70 i.e. semiconductor device 70 in a low Vt state
• an inversion layer is formed in channel region 86 and a depletion region (not shown) is formed around drain region 54 and drain extension 53.
• This depletion region substantially masks the dopant gradient created in halo region 46 thereby preventing the higher dopant of halo region 46 from increasing the Vt of semiconductor 70.
• the Vt remains in a low Vt state, thus improving the read disturb by reducing Vt drift.
• a short channel leakage may result during programming of semiconductor device 70.
• highly doped APT region 74 also functions to reduce this short channel leakage, thereby reducing power consumption and improving programming efficiency.
• the source and drains and extensions may be p-type or n-type, depending on the polarity of the isolated well, in order to form either p-type or n-type semiconductor devices. Therefore, isolated well
• Source and drain regions 52 and 54 and extensions 51 and 53 may be P-type.
• other materials and processing steps may be used to form semiconductor device 10; those described above have only been provided as examples.

Landscapes

• Non-Volatile Memory (AREA)
• Semiconductor Memories (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (2)

Family

ID=32068349

Family Applications (1)

Country Status (7)

Cited By (5)

Families Citing this family (60)

Family Cites Families (29)

• 2002
  • 2002-10-09 US US10/267,153 patent/US6887758B2/en not_active Expired - Lifetime
• 2003
  • 2003-09-23 KR KR1020057006192A patent/KR20050055003A/ko not_active Ceased
  • 2003-09-23 WO PCT/US2003/030588 patent/WO2004034426A2/en not_active Ceased
  • 2003-09-23 AU AU2003277017A patent/AU2003277017A1/en not_active Abandoned
  • 2003-09-23 JP JP2004543036A patent/JP2006502581A/ja active Pending
  • 2003-09-23 CN CNB038240033A patent/CN100420036C/zh not_active Expired - Lifetime
  • 2003-09-23 CN CN2007101962649A patent/CN101197292B/zh not_active Expired - Lifetime
  • 2003-10-07 TW TW092127818A patent/TWI322498B/zh not_active IP Right Cessation

Also Published As

Similar Documents

Legal Events