WO2000051165A2 - Misfet with narrow bandgap source - Google Patents

Misfet with narrow bandgap source Download PDF

Info

Publication number
WO2000051165A2
WO2000051165A2 PCT/IB2000/000235 IB0000235W WO0051165A2 WO 2000051165 A2 WO2000051165 A2 WO 2000051165A2 IB 0000235 W IB0000235 W IB 0000235W WO 0051165 A2 WO0051165 A2 WO 0051165A2
Authority
WO
WIPO (PCT)
Prior art keywords
source
channel
fermi
layer
level
Prior art date
Application number
PCT/IB2000/000235
Other languages
French (fr)
Other versions
WO2000051165A3 (en
Inventor
Carlos J. R. P. Augusto
Original Assignee
Augusto Carlos J R P
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Augusto Carlos J R P filed Critical Augusto Carlos J R P
Priority to DE20022706U priority Critical patent/DE20022706U1/en
Priority to JP2000601674A priority patent/JP2003523615A/en
Priority to US09/889,815 priority patent/US6674099B1/en
Priority to EP00905235A priority patent/EP1166347A2/en
Priority to AU26857/00A priority patent/AU2685700A/en
Publication of WO2000051165A2 publication Critical patent/WO2000051165A2/en
Publication of WO2000051165A3 publication Critical patent/WO2000051165A3/en
Priority to US10/725,830 priority patent/US7023030B2/en

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66—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
    • H01L29/66666—Vertical transistors
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/70—Manufacture or treatment of devices consisting of a plurality of solid state components formed in or on a common substrate or of parts thereof; Manufacture of integrated circuit devices or of parts thereof
    • H01L21/77—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate
    • H01L21/78—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices
    • H01L21/82—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components
    • H01L21/822—Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components the substrate being a semiconductor, using silicon technology
    • H01L21/8232—Field-effect technology
    • H01L21/8234—MIS technology, i.e. integration processes of field effect transistors of the conductor-insulator-semiconductor type
    • H01L21/8238—Complementary field-effect transistors, e.g. CMOS
    • H01L21/823885—Complementary field-effect transistors, e.g. CMOS with a particular manufacturing method of vertical transistor structures, i.e. with channel vertical to the substrate surface
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/02—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having at least one potential-jump barrier or surface barrier; including integrated passive circuit elements with at least one potential-jump barrier or surface barrier
    • H01L27/04—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having at least one potential-jump barrier or surface barrier; including integrated passive circuit elements with at least one potential-jump barrier or surface barrier the substrate being a semiconductor body
    • H01L27/08—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having at least one potential-jump barrier or surface barrier; including integrated passive circuit elements with at least one potential-jump barrier or surface barrier the substrate being a semiconductor body including only semiconductor components of a single kind
    • H01L27/085—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having at least one potential-jump barrier or surface barrier; including integrated passive circuit elements with at least one potential-jump barrier or surface barrier the substrate being a semiconductor body including only semiconductor components of a single kind including field-effect components only
    • H01L27/088—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having at least one potential-jump barrier or surface barrier; including integrated passive circuit elements with at least one potential-jump barrier or surface barrier the substrate being a semiconductor body including only semiconductor components of a single kind including field-effect components only the components being field-effect transistors with insulated gate
    • H01L27/092—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having at least one potential-jump barrier or surface barrier; including integrated passive circuit elements with at least one potential-jump barrier or surface barrier the substrate being a semiconductor body including only semiconductor components of a single kind including field-effect components only the components being field-effect transistors with insulated gate complementary MIS field-effect transistors
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/02—Semiconductor bodies ; Multistep manufacturing processes therefor
    • H01L29/06—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by their shape; characterised by the shapes, relative sizes, or dispositions of the semiconductor regions ; characterised by the concentration or distribution of impurities within semiconductor regions
    • H01L29/08—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by their shape; characterised by the shapes, relative sizes, or dispositions of the semiconductor regions ; characterised by the concentration or distribution of impurities within semiconductor regions with semiconductor regions connected to an electrode carrying current to be rectified, amplified or switched and such electrode being part of a semiconductor device which comprises three or more electrodes
    • H01L29/0895—Tunnel injectors
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/02—Semiconductor bodies ; Multistep manufacturing processes therefor
    • H01L29/12—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by the materials of which they are formed
    • H01L29/16—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by the materials of which they are formed including, apart from doping materials or other impurities, only elements of Group IV of the Periodic System
    • H01L29/161—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by the materials of which they are formed including, apart from doping materials or other impurities, only elements of Group IV of the Periodic System including two or more of the elements provided for in group H01L29/16, e.g. alloys
    • H01L29/165—Semiconductor bodies ; Multistep manufacturing processes therefor characterised by the materials of which they are formed including, apart from doping materials or other impurities, only elements of Group IV of the Periodic System including two or more of the elements provided for in group H01L29/16, e.g. alloys in different semiconductor regions, e.g. heterojunctions
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/40—Electrodes ; Multistep manufacturing processes therefor
    • H01L29/43—Electrodes ; Multistep manufacturing processes therefor characterised by the materials of which they are formed
    • H01L29/45—Ohmic electrodes
    • H01L29/456—Ohmic electrodes on silicon
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66—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
    • H01L29/66568—Lateral single gate silicon transistors
    • H01L29/66643—Lateral single gate silicon transistors with source or drain regions formed by a Schottky barrier or a conductor-insulator-semiconductor structure
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66—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
    • H01L29/7827—Vertical transistors
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66—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
    • H01L29/786—Thin film transistors, i.e. transistors with a channel being at least partly a thin film
    • H01L29/78642—Vertical transistors
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00—Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66—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
    • H01L29/7839—Field effect transistors with field effect produced by an insulated gate with Schottky drain or source contact

Definitions

  • the present invention relates to metal insulator semiconductor field effect transistors (MISFET).
  • MISFET metal insulator semiconductor field effect transistors
  • CMOS Complementary Metal-Oxide Semiconductor
  • Inverter Complementary Metal-Oxide Semiconductor
  • CMOS complementary metal-oxide-semiconductor
  • parasitic capacitances For deep sub-micron CMOS, as the gate lengths get shorter, leakage current tends to get higher, and the overall process technology becomes more complex. Not only the number of processing steps increases, but the complexity and difficulty of some of those steps is also increased. Since to make CMOS circuits, NMOS and PMOS devices are needed, many Front-End processing steps have to be made twice, separately for each device type.
  • CMOS circuits can also be made with other MOSFET architectures, like Vertical MOSFETs (see reference [1]).
  • the perspectives opened by Vertical MOSFETs are very attractive. That is especially true when considering the technological and fundamental physical limitations facing conventional (Planar) MOSFETs for gate lengths below 100nm.
  • the channel length is defined by the doping and/or heterojunction profiles, made by low temperature epitaxy. Lithography defines the cross section of the devices (channel width), and therefore the density of integration.
  • the present invention pertains to the field of Complementary Metal-Insulator- Semiconductor Field-Effect Transistors (C-MISFETs). Since the most common insulator is an oxide (silicon dioxide), these devices are almost always designated by Complemetary Metal-Oxide-Semiconductor Field-Effect Transistors (C-MOSFETs). More specifically, it pertains to CMOS circuits made with a new kind of Vertical MOSFETs.
  • C-MISFETs Complementary Metal-Insulator- Semiconductor Field-Effect Transistors
  • the present invention introduces a MOSFET device, that behaves as N- or P-type transistor, depending only on the applied bias.
  • Setting of the source voltage supply determines if the device will behave as a NMOS or as a PMOS.
  • Vps positive drain to source
  • VDS negative drain to source
  • VGS gate to source
  • the same device acts like a PMOS. Therefore, with the device of the present invention it is possible to make complementary circuits (CMOS), even though only a single device type is fabricated, which "a priori" is neither N- or P-type.
  • the subject of this invention will hereafter be designated by "Single Device Complementary Metal Oxide Semiconductor Field Effect Transistor", or SD- CMOS.
  • VH-MOSFETs Vertical Heterojunction MOSFETs
  • the device type (NMOS or PMOS) is defined by what type of dopant is incorporated in the source and drain regions.
  • Numerical simulations of Double-Gate SOI CMOS with 30 nm gate/channel lengths (see reference [4]), predict extraordinary performance levels.
  • a very illustrative parameter is the CMOS ring oscillator delay being less than 1 picosecond. Equal or better performance levels should be expected for the VH-MOSFET with channel lengths like 20 nm for example.
  • CMOS integration schemes have been proposed (see reference [3]), where the device layers of one device type are stacked on the device layers of the other device type, thereby enabling a single epitaxial growth step, and a common gate stack (gate insulator and gate electrode).
  • Such integration schemes offer the perspective of significant overall front-end process simplification, and area gains, over configurations where NMOS and PMOS transistors would be made "side by side”.
  • MISFET metal insulator semiconductor field effect transistor
  • a source layer being made with a material having a source band gap (EG2) and a source mid-gap value (EGM2), said source layer having a source
  • a channel layer between the source layer and the drain layer said channel layer being made with a material having a channel band gap (EG3) and a channel mid-gap value (EGM3), said channel layer having a channel Fermi level (EF3); a source contact layer connected to the source layer opposite the channel layer, said source contact layer having a source contact Fermi-Level (EF1 ); and
  • said source band gap is substantially narrower (EG2) than said channel band gap (EG3); • said source contact Fermi-Level (EF1), said source Fermi-Level (EF2), said channel Fermi-Level (EF3), said drain Fermi-Level (EF4) and said gate electrode Fermi-Level (EF6) are equal to said source mid-gap value (EGM2) and said channel mid-gap value (EGM3), within a predetermined tolerance value, when no voltage is applied to the device.
  • EG2 source mid-gap value
  • ECM3 gate electrode Fermi-Level
  • the Fermi-Levels are substantially equal to the source and the channel mid-gap values, symmetric paths from source to drain for electrons and for holes are created. This allows the device to behave as an NMOS or PMOS, depending on the voltage applied. It substantially improves the manufacturing process of MISFET's, since it is no longer necessary to decide, contrary to known devices hitherto, upon fabrication if the device should behave as NMOS or as PMOS.
  • Figure 1 is a schematic cross section through the layers of one preferred embodiment of the device according to the invention.
  • Figures 2a, 2b and 2c are schematic band alignments, along a vertical section from Source to Drain in the device of Figure 1 , near the interface with the Gate insulator, for different Drain and Gate bias conditions, when the device is used as an N-MOS.
  • Figures 3a, 3b and 3c are schematic band alignments, along a vertical section from Source to Drain in the device of Figure 1 , near the interface with the Gate insulator, for different Drain and Gate bias conditions, when the device is used as a P-MOS.
  • Figures 4a, 4b, 4c and 4d are schematic band alignments of two identical SD- CMOS devices, along a vertical section from Source to Drain, near the interface with the Gate insulator, for different Drain and Gate bias conditions.
  • Figures 5a, 5b, 5c, 5d, 5e and 5f are schematic band alignments of one SD- CMOS device, along a vertical section from Source to Drain, near the interface with the Gate insulator, for different Drain and Gate bias conditions.
  • Figure 6 illustrates a 3-dimensional perspective of a "Gate All Around” device.
  • Figure 7a illustrates a 3-dimensional perspective of one possible implementation of the "Edge-Gate" arrangement.
  • Figure 7b illustrates a 3-dimensional perspective of a second possible implementation of the "Edge-Gate" arrangement.
  • Figure 7c illustrates a 3-dimensional perspective of third possible implementation of the "Edge-Gate" arrangement.
  • Figure 8 illustrates a 3-dimensional perspective of a possible implementation of the "Inner-Gate” together with “Edge-Gate” arragements.
  • Figure 9a illustrates an implementation of a "NOR" logic gate, where SD- CMOS devices to behave as NMOS and PMOS are made on opposite sides of the "Drain Contact".
  • Figure 9b illustrates an implementation of a "NOR" logic gate, where a SD- CMOS devices will behave alternatively as NMOS or PMOS by changing the bias at the source contact.
  • Figures 10A to 10N illustrate a process flow for manufacturing a device according to the invention according to a first preferred embodiment.
  • Figures 11A to 11M illustrate a process flow for manufacturing a device according to the invention according to a second preferred embodiment.
  • Figures 12A to 12Q illustrate a process flow for manufacturing a device according to the invention according to a third preferred embodiment.
  • Figure 13a is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the channel, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
  • Figure 13b is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the source, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
  • Figure 13c is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the source contact, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
  • Figure 13d is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the drain, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
  • Layer 1 is the contact to the Source: metal with workfunction or Fermi-Level in the middle of the band-gap of the Source material.
  • Layer 2 is the Source: "narrow" band-gap material, with its mid-gap point aligned with the mid-gap point of the channel material, resulting in similar offsets in the conduction and valence bands, with respect to the channel material.
  • Layer 3 is the channel: "wide" band-gap material.
  • Layer 4 is the Drain: metal with a workfunction or Fermi-Level in the middle of the gap of the channel material.
  • Layer 5 is the Gate insulator.
  • Layer 6 is the Gate electrode: conductor with Fermi-Level in the middle of the gap of the channel material.
  • Layers 1, 4, and 6 can be of the same material.
  • FIG. 2a a band diagram of the device is shown when no voltage is applied, in other words the drain to source voltage and the gate to source voltage are both equal to zero.
  • EF(1) is the Fermi-Level in material 1.
  • EF(2) is the Fermi-Level in material 2.
  • EC(2) is the conduction band edge of material 2.
  • EV(2) is the valence band edge of material 2.
  • EF(3) is the Fermi-Level in material 3.
  • EC(3) is the conduction band edge of material 3.
  • EV(3) is the valence band edge of material 3.
  • EF(4) is the Fermi-Level in material 4.
  • VS is the potential at the Source.
  • VD is the potential at the Drain.
  • VDS>0, VGS 0
  • EC (2) is the conduction band edge of material 2
  • EV(2) is the valence band edge of material 2
  • EF(3) is the Fermi-Level in material 3
  • EC(3) is the conduction band edge of material 3
  • EV(3) is the valence band edge of material 3
  • EF(4) is the Fermi-Level in material 4 VS is the potential at the Source.
  • VD is the potential at the Drain.
  • VDS>0, VGS>0 EF(1) is the Fermi-Level in material 1
  • EC(2) is the conduction band edge of material 2
  • EV(2) is the valence band edge of material 2
  • EF(3) is the Fermi-Level in material 3
  • EC(3) is the conduction band edge of material 3
  • EV(3) is the valence band edge of material 3
  • VS is the potential at the Source.
  • VD is the potential at the Drain.
  • ECn(2) is the region of the conduction band of material 2, that is below the
  • EC(2) is the conduction band edge of material 2
  • EV(2) is the valence band edge of material 2
  • EF(3) is the Fermi-Level in material 3
  • EC(3) is the conduction band edge of material 3
  • EV(3) is the valence band edge of material 3
  • VD is the potential at the Drain.
  • VGS (VS-VG) is the difference of potential between Source and Gate.
  • VDS 0
  • VGS 0.
  • EF(2) is the Fermi-Level in material 2
  • EC(2) is the conduction band edge of material 2
  • EV(2) is the valence band edge of material 2
  • EF(3) is the Fermi-Level in material 3
  • EC(3) is the conduction band edge of material 3
  • EV(3) is the valence band edge of material 3
  • EF(4) is the Fermi-Level in material 4
  • VS is the potential at the Source.
  • VD is the potential at the Drain.
  • EC(2) is the conduction band edge of material 2
  • EV(2) is the valence band edge of material 2
  • EF(3) is the Fermi-Level in material 3
  • EC(3) is the conduction band edge of material 3
  • EV(3) is the valence band edge of material 3
  • EF(4) is the Fermi-Level in material 4 VS is the potential at the Source.
  • VD is the potential at the Drain.
  • EVn(2) is the region of the valence band of material 2, that is above the EF(2), as an effect of negative Gate to Source voltage (VGS ⁇ 0).
  • Figures 2a and 3a When comparing Figures 2a and 3a, it can be noted that the device is identical for both cases. The device will however behave as an N-MOS ( Figures 2b and 2c) or a P-MOS ( Figures 3b and 3c) in function of the voltage applied. It should be noted that the vertical axis in Figures 2 and 3 is expressed as potential in Volts. It could also be expressed as potential energy in electron-volts. The same applies to other figures where potential is indicated.
  • the devices are connected together in a "CMOS Inverter" arrangement: The Drains are connected together. The Gates are connected together. The Source of the device on the left hand side of the figure, is connected to the ground potential. This device will behave as a NMOS. The Source of the device on the right hand side of the figure, is connected to a negative potential. This device will behave as a PMOS.
  • VD GND.
  • the device on the left has just been switched “Off 1 .
  • the device on the right has just been switched “On”, and current starts to flow.
  • FIG. 4b Steady state.
  • VG GND.
  • VD -VSS.
  • the device on the left is still “Off”.
  • the device on the left has just been turned On. VDS>0, current flows.
  • the device on the right has just been turned "Off'.
  • the device on the right is "Off”.
  • Source to Drain near the interface with the Gate insulator, for different Drain and Gate bias conditions.
  • the Source of the device is going to be changed between GND and -VSS potentials.
  • the device behaves as a NMOS transistor.
  • the device behaves as a PMOS transistor.
  • the Gate of the device switches between GND and -VSS potentials.
  • VDS -VSS.
  • VGS -VSS.
  • the device has just been turned “On” as PMOS, and hole current starts to flow.
  • VS -VSS.
  • VG GND.
  • VD -VSS.
  • VGS -VSS.
  • the device has just been switched from a "On-state” PMOS, to a "Off-state” NMOS. No electron current flows.
  • VS GND.
  • VG -VSS.
  • VD -VSS.
  • VDS +VSS.
  • VGS +VSS.
  • the device has just been turned “On” as NMOS, and “electron current” starts to flow.
  • VD GND.
  • the device concept of the present invention is independent of any particular implementation. It can be implemented in different materials systems, like Si- based and GaAs-based alloys, for example. Independently of the materials system, it can also be implemented with different "Process Flows” or "Process Integration Architectures”.
  • the shape of the conduction band should mirror the shape of the valence band (or vice-versa), the middle of the band-gap being the mirror line.
  • the Source material is made of an undoped semiconductor, with a very narrow band-gap, which is much narrower, in particular 9 to 10 times narrower, than the band-gap of the channel material.
  • the band-alignment must be such that the band-gap of the source material is fully nested in the band-gap of the channel material.
  • the offsets in the conduction and valence bands should be of the same magnitude. Actually, due to the differences in effective masses, the conduction and valence band offsets may in fact be slightly different.
  • the Channel material is made of an undoped semiconductor, with a band- gap wide enough to comprise the barrier height for electrons and for holes, and the very small band-gap of the source material.
  • An elemental semiconductor might also have the advantage of no alloy scattering.
  • the barrier height for electrons and holes determines off-state current (for electrons and holes respectively), and therefore should be large enough to enable room-temperature operation with negligible "off-state" currents.
  • the barrier heights ought to be engineering parameters, which should be possible to continuously vary across a wide range of values (for example by variation of alloy compositions of the source layer).
  • the Drain is defined by a Schottky junction, between the channel and a metal with a workfunction or Fermi-Level in the middle of the band-gap of the channel material.
  • the Gate electrode also needs to have a Fermi-Level in the middle of the band-gap of the channel material: metal with workfucntion in the middle of the band-gap of the source and channel materials.
  • the metal insulator semiconductor field effect transistor comprises a source layer 2 being made with a material having a source band gap (EG2) and a source mid-gap value (EGM2), said source layer having a source Fermi-Level (EF2).
  • a drain layer 4 has a drain Fermi-Level (EF4).
  • a channel layer 3 is provided between the source layer and the drain layer.
  • the channel layer is made with a material having a channel band gap (EG3) and a channel mid- gap value (EGM3).
  • the channel layer further has a channel Fermi level (EF3).
  • a source contact layer 1 is connected to the source layer opposite the channel layer, said source contact layer having a source contact Fermi-Level (EF1).
  • a gate electrode 6 has a gate electrode Fermi-Level (EF6).
  • the source band gap is substantially narrower (EG2) than said channel band gap (EG3), in particular at least 9 to 10 times.
  • the source contact Fermi-Level (EF1), the source Fermi-Level (EF2), the channel Fermi- Level (EF3), the drain Fermi-Level (EF4) and the gate electrode Fermi-Level (EF6) are equal to the source mid-gap value (EGM2) and the channel mid- gap value (EGM3), within a predetermined tolerance value, when no voltage is applied to the device.
  • the source band gap (EG2) is the difference between the conduction band edge (EC2) and the valence band edge (EV2) for the source. It can be equated as follows :
  • the channel band gap (EG3) is the difference between the conduction band edge (EC3) and the valence band edge (EV3) for the channel. It can be equated as follows :
  • the source mid-gap value can be equated as follows :
  • EGM2 (EC2 - EV2)/2;
  • the band-gap of the source material EG2 could be for example around 0.11 eV, with a tolerance of plus or minus ( ⁇ ) 5% on this value (total of 10%), resulting in the range of 0.1 to 0.12 eV.
  • the band-gap of the channel material EG3 should be around 1.1 eV, with a tolerance of plus or minus ( ⁇ ) 5% on this value (total of 10%), resulting in the range of 1.0 to 1.2 eV.
  • EC3 - EC2 0.5 eV ( ⁇ ) 5% (total of 10%), resulting in a range from 0.475 eV to 0.525 eV
  • EV3 - EV2 0.5 eV ( ⁇ ) 5% (total of 10%), resulting in a range from 0.475 eV to 0.525 eV
  • the tolerance value as indicated in claim 1 can be expressed in different ways.
  • the Fermi-Levels in the source and channel regions should be close to the following values:
  • EF3 EV3 + EGM3, plus or minus ( ⁇ ) 5% (total of 10%)
  • EF3 0 ( ⁇ ) 0.05 eV or will range from -0.05 eV to + 0.05 eV.
  • the 0.05 eV tolerance is approximately 0.05/1.1 or +/- 5% of the channel band gap (EG3).
  • the tolerance value can be expressed diffenently.
  • the device according to the invention has the following layers with the following characteristics: • layer 1 is the contact to the Source: metal with Fermi-Level in the middle of the band-gap of the Source material, and therefore also in the middle of the band-gap of the channel material;
  • layer 2 is the Source: "narrow" band-gap material, with its mid-gap point aligned with the mid-gap point of the channel material, resulting in similar offsets in the conduction and valence bands, with respect to the channel material;
  • layer 3 is the channel: "wide" band-gap material
  • layer 4 is the Drain: metal with a Fermi-Level in the middle of the gap of the channel material; • layer 5 is the Gate insulator; and
  • layer 6 is the Gate electrode: conductor with Fermi-Level in the middle of the gap of the channel material.
  • Layers 1 , 4, and 6 can be the same material.
  • Figures 2a, 2b, 2c show schematics of the band-diagrams for positive bias conditions, when the device behaves as a NMOS transistor.
  • Figures 3a, 3b, 3c show schematics of the band-diagrams for negative bias conditions, when the device behaves as a PMOS transistor.
  • silicon-based technology is very important.
  • the embodiment of the SD-CMOS with silicon-compatible materials will make use of silicon-based alloys like Sii- ⁇ Ge ⁇ , Sh-yCy, Si - ⁇ - yGe x Cy, etc. Due to layer formation difficulties, it is less likely that alloys with Sn can be used. However if the perceived/ anticipated technological issues are overcome, compounds with this element could also be used.
  • 5 Typically can be SiO2, or SiON/Si3N4, etc.
  • Titanium Nitride TiN
  • the Source is a very narrow (for example 5 KT, which at room temperature is about 130 mili-electron-Volts) band-gap material pseudomorphically grown on silicon.
  • the band-alignment with silicon is such that the band offsets in the conduction and valence bands must be symmetric (for example 0.5 volts for each band discontinuity).
  • Examples of possible materials providing such requirements are a combination of Sh- ⁇ Ge ⁇ , Sh-yCy, Sii- ⁇ -yGe x Cy, either as random alloys or as short-period superlattices of alternating layers of Sii- yCy, and Sii- ⁇ Ge ⁇ , for example.
  • the exact composition and thickness of these layers is an engineering question, not a conceptual one. Enough data is already known about these alloys (see reference [8]), to be able to predict that some combination will deliver the band-alignment necessary for this concept.
  • the Source is contacted by a metal electrode with a Fermi-Level in the middle of the silicon bad-gap. Since the band-gap of the source material is very narrow, and centered (with equal discontinuities for the conduction and valence bands) in the band-gap of silicon, it means that the Fermi-Level of the metal at the source is also in the middle its band-gap. Therefore it is possible to have good ohmic contacts without any rectifying properties (for both electrons and holes) between the source metal and the very narrow semiconductor at the source, even though there is no doping involved. TiN (titanium nitride) is an example of a metal with such properties (see reference [9]).
  • the Channel is made of non-doped pure silicon.
  • the Channel/Drain interface is a Schottky junction, between the silicon channel, and a metal drain with a Fermi-Level near the mid-gap of silicon.
  • TiN titanium nitride
  • TiN titanium nitride
  • the Gate electrode is a conductor with the Fermi-Level in the middle of the band-gap of silicon.
  • TiN titanium nitride
  • the Fermi-Level (or Chemical Potential) will be in the middle of the band-gaps of the source (narrow band-gap material) and channel regions.
  • a "flat band condition" exists across the gate to channel interface. So, from the shape of the electrostatic potential (band edges), the physical picture for electrons and holes is very symmetric. However, in real space, there is an asymmetry between the source/channel and the channel/drain interfaces.
  • the reverse "Off-state” current is the thermionic current over the barrier.
  • that current is indeed very low. Therefore, the drain of these devices is not able to inject current over the barrier, thereby preventing it from acting as the source of the complementary device-type.
  • the very narrow band-gap material is positioned between the metal contact and the silicon channel, in order to enable a "switch-On/Off" effect mentioned before. If the metal contact was positioned directly on the silicon channel (Schottky junction), that switch On/Off" mechanism would not be possible.
  • the presence of the very narrow band-gap film in the source layer is the key enabler of this mechanism, which makes possible the lowering of the effective barrier height (distance between band edge in the channel and the Fermi- Level in the source), for either electrons or holes.
  • the presence of that film also breaks the symmetry between source and drain interfaces with the channel material.
  • That film enables drift-diffusion or ballistic current across the source to channel heterojunction. If a Schottky junction was made directly on silicon (at the source), the barrier height could not be changed, and the only turn-on current mechanism possible would be tunneling (see references [10, 11]).
  • CMOS complementary metal-oxide-semiconductor
  • CMOS Inverter could be made either by having “conventional” “Static Source Voltage Supply”, in which case it would require two transistors, or by having a “Dynamic Source Voltage Supply”, in which case only one transistor is required.
  • the devices will be "separated" into NMOS- or PMOS-like devices, by the metallization scheme, in which power supply voltages for NMOS and PMOS transistors are "hardwired".
  • the power supply can be alternated between “positive” and “negative” voltages, the same device acts like “NMOS” AND “PMOS” sequentially in time.
  • CMOS Inverter By changing the bias of the source, and maintaining the gate bias constant while the source voltage is changed, a "CMOS Inverter" with just one transistor is possible.
  • Figure 4a, 4b, 4c, and 4d show schematics of the band diagrams of two identical SD-CMOS devices with common Gates and Drains, for different bias conditions.
  • the potentials applied to the sources are such that the device on the left will behave as a NMOS, and the device on the right will behave as a PMOS.
  • VD GND.
  • the device on the right has just been switched “On”, and current starts to flow.
  • the device on the left is still “Off”.
  • the device on the left has just been turned On. VDS>0, current flows.
  • the device on the right has just been turned "Off".
  • the device on the right is "Off".
  • Source to Drain near the interface with the Gate insulator, for different Drain and Gate bias conditions.
  • the Source of the device is going to be changed between GND and -VSS potentials.
  • the device behaves as a NMOS transistor.
  • the device behaves as a PMOS transistor.
  • the Gate of the device switches between GND and -VSS potentials.
  • Drain voltage is the output of the Inverter.
  • Gate Voltage is the input of the Inverter.
  • the source voltage is switched between “0” and “-1", in which case the device "behaves” as NMOS and PMOS respectively.
  • the device is "Off" as NMOS. No electron current flows.
  • VS -VSS.
  • VG GND.
  • VD GND.
  • VDS -VSS.
  • VGS -VSS.
  • the device has just been turned “On” as PMOS, and hole current starts to flow.
  • VGS -VSS.
  • VS GND.
  • VG GND.
  • VD -VSS.
  • VDS +VSS.
  • VGS 0.
  • the device has just been switched from a "On-state” PMOS, to a "Off-state"
  • VDS +VSS.
  • VGS +VSS.
  • the device has just been turned “On” as NMOS, and "electron current” starts to flow.
  • VS GND.
  • VG -VSS.
  • VD GND.
  • VGS +VSS.
  • the topology of choice for logic gates is the “NAND” configuration.
  • the logic inputs are the gate terminals of a series of n-type MOSFETs, connected in series with a PMOS (the load transistor). Each additional logic input, requires an additional NMOS device to be inserted in the series with all of them.
  • the sources of several NMOS are shunted together, and the same is done for the drains.
  • the set of parallel NMOS devices is connected in series with a PMOS device (the load transistor).
  • Each additional logic input requires an extra NMOS device to be connected in parallel with the other NMOS transistors.
  • NANDs being the configuration of choice for "Planar MOSFETs are: 1) For "Planar MOSFETs", the series connection enables area savings, because for devices of the same type, the source of one transistor can be the drain of another. However, for bulk CMOS, and because of the "body effect", the number of number of logic inputs is typically reduced to two. A larger number of inputs is possible, only if Silicon- On-lnsulator (SOI) technology is used.
  • SOI Silicon- On-lnsulator
  • the total “Off-state” current is the “Off-state” current of the least leaky device.
  • the total “Off-state” current is the sum of the "Off- state” currents of ail individual NMOS devices.
  • the step which defines "Mesas” exposing the device layers where the gate stack is formed simultaneously provides “isolation” between devices.
  • the contacts to the several device layers are made inside the perimeter of the gate stack (which as its name suggests, surrounds the device layers).
  • Figure 6 is a schematic of 3-dimensional perspective of a "Gate All Around” device. "EDGE-GATE"
  • the "Isolation" and “Gate Stack” formation steps are performed separately.
  • the gate is not surrounding the “Mesa” of the device layers, but it is placed on only one crystalline plane.
  • the other sides of the “device layer Mesa” face the "Field Isolation” for example.
  • Figure 7a is a schematic of 3-dimensional perspective of one possible implementation of the "Edge-Gate" arrangement.
  • Figure 7b is a schematic of 3-dimensional perspective of a second possible implementation of the "Edge-Gate" arrangement.
  • Figure 7c is a schematic of 3-dimensional perspective of third possible implementation of the "Edge-Gate" arrangement.
  • the gate is surrounded by the device layers, which at the edges face the "Field Isolation".
  • the "Field Isolation” and “Gate Stack” formation steps are performed separately. This configuration enables not only single drain contact, but also single source contact, for several independent gates. This is an ideal configuration for "NOR" logic gates.
  • Figure 8 is a schematic of 3-dimensional perspective of a possible implementation of the "Inner-Gate” together with “Edge-Gate” arrangements.
  • Figure 9 is a top view of a possible implementation of the "Inner-Gate” together with “Edge-Gate” arrangements.
  • Figure 9a shows an implementation where SD-CMOS devices to behave as NMOS and PMOS are made on opposite sides of the "Drain Contact”.
  • Figure 9b shows an implementation where a single SD-CMOS device will behave alternatively as NMOS or PMOS by changing the bias at the source contact.
  • SD-CMOS does not suffer from “corner effect”, due to the intrinsics of the device physics, and therefore none of the possible implementations or process integration architectures, is affected by it.
  • the reason for the immunity to “corner effect”, is that the “corner effect” is a geometric effect that re-enforces the "zero-bias" electric field across a MOS interface.
  • references [17] and [18] demonstrate the feasibility of epitaxial insulators like SrTiO3 and BaTiO3 (the later a Ferroelectric) using epitaxial TiN on silicon as a buffer material.
  • CMOS Integration schemes for Vertical MOSFETs have been proposed before, but for SD-CMOS only one device structure needs to be fabricated, and that carries important consequences.
  • SD-CMOS is an asymmetric Vertical MOSFET, it means that source and drain are not interchangeable, and therefore, choosing which one is at the bottom and at the top of the layer stack, carries consequences.
  • Source as the top layer implies one of the following options:
  • Another set of options regarding the device layers has to do with the formation of the device Mesas: blanket growth followed by patterning of the epitaxial layers, or pre-patterning of hard mask followed by selective epitaxial growth.
  • Inner-Gate and Edge-Gate can be implemented simultaneously, without extra masks.
  • the circuit configuration chosen is a "NOR Gate” with 5 inputs (4 “inner- gates” and 1 "edge-gate”).
  • the capacitor layers are not used (are in fact transparent to the functionality of simple logic gates). In an small change to this flow with an extra mask, the capacitor flms could have been removed from the "logic-only” areas.
  • gate stack (gate insulator & gate electrode). 23) CMP of gate stack, stopping on Si3N4 and SiO2.
  • MISD Metal-Insulator-Semiconductor Field-Effect Transistor
  • MI-FET Metal-lnsulator-Semiconductor Field-Effect Transistor
  • a Metal-lnsulator-Semiconductor Field-Effect Transistor is disclosed, composed of the following active regions: a) A channel layer, made with a "wider" band-gap undoped semiconductor. b) A drain layer, made with a metal directly interfaced to the channel material (Schottky junction), having a Fermi-Level value in the middle of the band-gap of the channel material. c) A source layer, made with a semiconductor with a "narrower" band-gap, centered with the band-gap of the channel material (equal offsets in the conduction and valence bands between the source and channel layers). d) A source contact metal with a Fermi-Level in the middle of the band-gap of the channel material. e) A gate electrode with a Fermi-Level in the middle of the band-gap of the channel material.
  • a MISFET is disclosed with specific embodiment in the silicon materials system, having the following device layers: a) Channel material: undoped silicon (Si). b) Drain material: epitaxial titanium nitride (TiN). c) Source material: Sii- ⁇ -yG ⁇ Cy, either as random alloy or short-period superlattice of alternating Sii-yCy/S - ⁇ Ge ⁇ layers for example. d) Source contact metal: epitaxial titanium nitride (TiN). e) Gate electrode: titanium nitride (TiN).
  • MISD Metal-lnsulator-Semiconductor Field-Effect Transistor
  • CMOS circuits having configurations where "inverters”, “logic gates”, memory cells, are fabricated with a single device which sequentially behaves as NMOS and PMOS, by appropriately sequentially changing the voltage at the source terminal.
  • the MISFET can be used for logic applications, for Random Memory (Dynamic, Static, Flash, Ferroelectric) applications, for logic applications with embedded Random Memory (Dynamic, Static, Flash, Ferroelectric) elements, for Random Memory (Dynamic, Static, Flash, Ferroelectric) applications with embedded logic elements, for co-integration with image sensors, like CCD and CMOS imagers, or any other kind, for co-integration with any kind of Micro-Electronic-Mechanical Systems (MEMS), or Micro-Optical or Optoelectronic integrated systems.
  • MEMS Micro-Electronic-Mechanical Systems
  • Process Integration Architecture depicted in Figure 7a can be achieved, where the Drain layer is at the bottom of the device layer stack (and therefore the Source at the top), and where a Gate Stack is positioned in such a way, as to be common to two devices, each one with its separate drain and source layers and respective contacts.
  • Process Integration Architecture depicted in Figure 7b can be achieved, where the Source layer is at the bottom of the device layer stack (and therefore the Drain at the top), and where a Gate Stack is positioned in such a way, as to be common to two devices, each one with its separate drain and source layers and respective contacts.
  • Process Integration Architecture depicted in Figure 7a can be achieved, where the Drain layer is at the bottom of the device layer stack (and therefore the Source at the top), and where a Drain Contact is positioned in such a way, as to be common to two devices, each one with its separate Gate Stack at the extreme edges of the device layer stack, and with its separate source layers and respective contacts.
  • Process Integration Architecture depicted in Figures 8 and 9a can be achieved where a single device layer stack, has single source and drain contacts, but several parallel Gates, for a compact "NOR-gate” arrangement.
  • the arrangement as depicted, represents a solution with "Dynamic Source Voltage Supply”.
  • Process Integration Architecture depicted in Figure 9b can be achieved, where a single device layer stack, has a single drain contact, two source contacts, and several parallel Gates, for a compact "NOR-gate” arrangement.
  • the arrangement as depicted, represents a solution with "Fixed Source Voltage Supply”.
  • the process flow according to Figures 10A to 10N can be used for CMOS logic, with very reduced number of extra processing steps for the inclusion of embedded memory (possibly Ferroelectric).
  • the process flow according to Figures 11A to 11M can be used for CMOS logic, with very reduced number of extra processing steps for the inclusion of embedded memory (possibly Ferroelectric).
  • VFT Vertical F-Shape, Transistor
  • Insulator Substrate e g : Q Quu;artz or Sapphire

Abstract

A metal insulator semiconductor field effect transistor (MISFET) is disclosed comprising a source layer being made with a material having a source band-gap (EG2) and a source mid-gap value (EGM2), the source layer having a source Fermi-Level (EF2). A drain layer has a drain Fermi-Level (EF4). A channel layer is provided between the source layer and the drain layer, the channel layer being made with a material having a channel band-gap (EG3) and a channel mid-gap value (EGM3), the channel layer having a channel Fermi-Level (EF3). A source contact layer is connected to the source layer opposite the channel layer, the source contact layer having a source contact Fermi-Level (EF1). A gate electrode has a gate electrode Fermi-Level (EF6). The source band-gap is substantially narrower (EG2) than the channel band-gap (EG3). The source contact Fermi-Level (EF1), the source Fermi-Level (EF2), the channel Fermi-Level (EF3), the drain Fermi-Level (EF4) and the gate electrode Fermi-Level (EF6) are equal to the source mid-gap value (EGM2) and the channel mid-gap value (EGM3), within a predetermined tolerance value, when no voltage is applied to the device.

Description

"MISFET"
BACKGOUND OF THE INVENTION
The present invention relates to metal insulator semiconductor field effect transistors (MISFET).
Complementary Metal-Oxide Semiconductor (CMOS), has been the semiconductor technology of choice since the late 1970's, and in 1998 the 0.25 micron CMOS technology generation is in production. There are many reasons to choose CMOS over other technologies. The most important is the reduced power consumption, because the basic building block of circuits for binary logic, the CMOS "Inverter", only consumes power when changing logic states.
The fundamental factors determining the performance of standard "Planar Technology", are the channel length of the MOSFETs, and the parasitic capacitances. For deep sub-micron CMOS, as the gate lengths get shorter, leakage current tends to get higher, and the overall process technology becomes more complex. Not only the number of processing steps increases, but the complexity and difficulty of some of those steps is also increased. Since to make CMOS circuits, NMOS and PMOS devices are needed, many Front-End processing steps have to be made twice, separately for each device type.
However, CMOS circuits can also be made with other MOSFET architectures, like Vertical MOSFETs (see reference [1]). The perspectives opened by Vertical MOSFETs are very attractive. That is especially true when considering the technological and fundamental physical limitations facing conventional (Planar) MOSFETs for gate lengths below 100nm. For Vertical MOSFETs the channel length is defined by the doping and/or heterojunction profiles, made by low temperature epitaxy. Lithography defines the cross section of the devices (channel width), and therefore the density of integration.
The present invention pertains to the field of Complementary Metal-Insulator- Semiconductor Field-Effect Transistors (C-MISFETs). Since the most common insulator is an oxide (silicon dioxide), these devices are almost always designated by Complemetary Metal-Oxide-Semiconductor Field-Effect Transistors (C-MOSFETs). More specifically, it pertains to CMOS circuits made with a new kind of Vertical MOSFETs.
The present invention, introduces a MOSFET device, that behaves as N- or P-type transistor, depending only on the applied bias. Setting of the source voltage supply, determines if the device will behave as a NMOS or as a PMOS. For positive drain to source (Vps) and gate to source (VGS) voltages, the device behaves as NMOS. For negative drain to source (VDS) and gate to source (VGS) voltages, the same device acts like a PMOS. Therefore, with the device of the present invention it is possible to make complementary circuits (CMOS), even though only a single device type is fabricated, which "a priori" is neither N- or P-type.
The subject of this invention will hereafter be designated by "Single Device Complementary Metal Oxide Semiconductor Field Effect Transistor", or SD- CMOS.
The independence of channel length from lithography, and the kind of doping/ heterojunction profiles possible with low temperature epitaxy, enables the fabrication of Vertical MOSFETs with channels only tens of nanometers long, and with atomic layer control across the entire wafer. The limitations to make Vertical MOSFETs with very short channel lengths, will no longer be technological, but related to device physics.
Vertical MOSFETs, have an intrinsic advantage over horizontal MOSFETs: it is straightforward to build an asymmetry of the source to channel junction, versus the channel to drain junction. With horizontal MOSFETs, it is also possible to introduce asymmetry, but that comes at a cost in terms of process complexity (extra masks to differentiate source from drain for each type of device). And in any case, the doping and/or heterojunction profiles (made by ion implantation) can never come close to what low temperature epitaxy has demonstrated.
Like with Horizontal homojunction MOSFETs, Vertical homojunction MOSFETs suffer from Short Channel Effect (SCE), albeit at shorter channel lengths due to the ability to have sharper doping profiles, and therefore reduced depletion widths. Numerical simulations of a "Planar-Doped" Vertical MOSFET with a 50 nm channel length, predict very high performance levels (see reference [2]). But as channel length is further reduced, higher doping levels are needed to keep the electrostatic barrier between source and channel. In the limit, even without any bias (at the drain or gate), the built-in electric field, induces band to band tunneling through the source to channel barrier. Naturally, the necessity of applying drain bias, leads to even stronger scaling limitations due to Drain Induced Barrier Lowering (DIBL). For these reasons, it has been predicted (see reference [1]), that due to SCE and DIBL, the practical limit for how short channels can be, is around 80 nm.
An alternative type of Vertical MOSFETs, the Vertical Heterojunction MOSFETs (VH-MOSFETs), uses heterojunctions instead of homojunctions to build the source to channel electrostatic barrier (see reference [3]). Because the potential barrier is originated by a heterojunction, there is no need to introduce doping in the channel to make the barrier, and therefore the device is, by definition "fully depleted". Also the heterojunction barrier exists across the entire channel thickness, and therefore removes any restriction on the distance between gates. With this device architecture, simulations show that ultra-short channels are possible (down to 10 nm), without suffering from SCE or DIBL (see reference [3]). The device type (NMOS or PMOS) is defined by what type of dopant is incorporated in the source and drain regions. Numerical simulations of Double-Gate SOI CMOS with 30 nm gate/channel lengths (see reference [4]), predict extraordinary performance levels. A very illustrative parameter is the CMOS ring oscillator delay being less than 1 picosecond. Equal or better performance levels should be expected for the VH-MOSFET with channel lengths like 20 nm for example.
CMOS integration schemes have been proposed (see reference [3]), where the device layers of one device type are stacked on the device layers of the other device type, thereby enabling a single epitaxial growth step, and a common gate stack (gate insulator and gate electrode). Such integration schemes offer the perspective of significant overall front-end process simplification, and area gains, over configurations where NMOS and PMOS transistors would be made "side by side".
Vertical MOSFETs have yet other attractive features. It has been shown how Vertical MOSFETs make possible memory cells, with a quarter of the area of cells made with planar MOSFETs, for the same generation of lithography equipment (see references [5, 6, 7]). For decades, DRAM has driven the progress in process technology. When optical lithography finds its ultimate limitations (believed to be around 100 nm), it is very likely that cells made with Vertical MOSFETs will be seriously considered as viable alternatives to increase the bit density.
However, and assuming that the capability of making Vertical MOSFETs with very short channels is to be fully exploited, it is required to have very low temperature processing (typically, below the temperature at which dopants will start to significantly diffuse and/or strained layers relax). Vertical MOSFETs, regardless of their channel length and particular device layer composition/profile, different device regions like gate, have source and drain lying on different planes. Therefore contacts to these regions (and to gate electrode) must be made by separate sequences of contact hole formation and contact hole filling with a metal. SUMMARY OF THE INVENTION
It is an object of the present invention to improve the manufacturing process of MISFET's.
This object is achieved by providing a metal insulator semiconductor field effect transistor (MISFET) comprising: a source layer being made with a material having a source band gap (EG2) and a source mid-gap value (EGM2), said source layer having a source
Fermi-Level (EF2);
• a drain layer having a drain Fermi-Level (EF4);
■ a channel layer between the source layer and the drain layer, said channel layer being made with a material having a channel band gap (EG3) and a channel mid-gap value (EGM3), said channel layer having a channel Fermi level (EF3); a source contact layer connected to the source layer opposite the channel layer, said source contact layer having a source contact Fermi-Level (EF1 ); and
• a gate electrode having a gate electrode Fermi-Level (EF6) wherein:
• said source band gap is substantially narrower (EG2) than said channel band gap (EG3); • said source contact Fermi-Level (EF1), said source Fermi-Level (EF2), said channel Fermi-Level (EF3), said drain Fermi-Level (EF4) and said gate electrode Fermi-Level (EF6) are equal to said source mid-gap value (EGM2) and said channel mid-gap value (EGM3), within a predetermined tolerance value, when no voltage is applied to the device.
By providing that the Fermi-Levels are substantially equal to the source and the channel mid-gap values, symmetric paths from source to drain for electrons and for holes are created. This allows the device to behave as an NMOS or PMOS, depending on the voltage applied. It substantially improves the manufacturing process of MISFET's, since it is no longer necessary to decide, contrary to known devices hitherto, upon fabrication if the device should behave as NMOS or as PMOS.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a schematic cross section through the layers of one preferred embodiment of the device according to the invention.
Figures 2a, 2b and 2c are schematic band alignments, along a vertical section from Source to Drain in the device of Figure 1 , near the interface with the Gate insulator, for different Drain and Gate bias conditions, when the device is used as an N-MOS.
Figures 3a, 3b and 3c are schematic band alignments, along a vertical section from Source to Drain in the device of Figure 1 , near the interface with the Gate insulator, for different Drain and Gate bias conditions, when the device is used as a P-MOS.
Figures 4a, 4b, 4c and 4d are schematic band alignments of two identical SD- CMOS devices, along a vertical section from Source to Drain, near the interface with the Gate insulator, for different Drain and Gate bias conditions.
Figures 5a, 5b, 5c, 5d, 5e and 5f are schematic band alignments of one SD- CMOS device, along a vertical section from Source to Drain, near the interface with the Gate insulator, for different Drain and Gate bias conditions.
Figure 6 illustrates a 3-dimensional perspective of a "Gate All Around" device.
RECTIFIED SHEET (RULE 91) Figure 7a illustrates a 3-dimensional perspective of one possible implementation of the "Edge-Gate" arrangement.
Figure 7b illustrates a 3-dimensional perspective of a second possible implementation of the "Edge-Gate" arrangement.
Figure 7c illustrates a 3-dimensional perspective of third possible implementation of the "Edge-Gate" arrangement.
Figure 8 illustrates a 3-dimensional perspective of a possible implementation of the "Inner-Gate" together with "Edge-Gate" arragements.
Figure 9a illustrates an implementation of a "NOR" logic gate, where SD- CMOS devices to behave as NMOS and PMOS are made on opposite sides of the "Drain Contact".
Figure 9b illustrates an implementation of a "NOR" logic gate, where a SD- CMOS devices will behave alternatively as NMOS or PMOS by changing the bias at the source contact.
Figures 10A to 10N illustrate a process flow for manufacturing a device according to the invention according to a first preferred embodiment.
Figures 11A to 11M illustrate a process flow for manufacturing a device according to the invention according to a second preferred embodiment.
Figures 12A to 12Q illustrate a process flow for manufacturing a device according to the invention according to a third preferred embodiment.
Figure 13a is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the channel, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
Figure 13b is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the source, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
Figure 13c is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the source contact, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
Figure 13d is a band-diagram, along a horizontal section through the gate electrode, the gate insulator, the drain, the gate insulator and the gate electrode of the device according to Figure 1 in absence of any applied voltage to any terminal of the device.
DETAILED DESCRIPTION OF THE INVENTION
In Figure 1 , showing a schematic cross section of the device, the following layers may be distinguished:
Layer 1 is the contact to the Source: metal with workfunction or Fermi-Level in the middle of the band-gap of the Source material. Layer 2 is the Source: "narrow" band-gap material, with its mid-gap point aligned with the mid-gap point of the channel material, resulting in similar offsets in the conduction and valence bands, with respect to the channel material.
Layer 3 is the channel: "wide" band-gap material. Layer 4 is the Drain: metal with a workfunction or Fermi-Level in the middle of the gap of the channel material.
Layer 5 is the Gate insulator. Layer 6 is the Gate electrode: conductor with Fermi-Level in the middle of the gap of the channel material.
Layers 1, 4, and 6, can be of the same material.
In Figure 2a, a band diagram of the device is shown when no voltage is applied, in other words the drain to source voltage and the gate to source voltage are both equal to zero.
EF(1) is the Fermi-Level in material 1. EF(2) is the Fermi-Level in material 2.
EC(2) is the conduction band edge of material 2.
EV(2) is the valence band edge of material 2.
EF(3) is the Fermi-Level in material 3.
EC(3) is the conduction band edge of material 3. EV(3) is the valence band edge of material 3.
EF(4) is the Fermi-Level in material 4.
VS is the potential at the Source.
VD is the potential at the Drain.
Figure 2b
VDS>0, VGS=0
EF(1 ) is the Fermi-Level in material 1
EF(2) is the Fermi-Level in material 2
EC (2) is the conduction band edge of material 2 EV(2) is the valence band edge of material 2
EF(3) is the Fermi-Level in material 3
EC(3) is the conduction band edge of material 3
EV(3) is the valence band edge of material 3
EF(4) is the Fermi-Level in material 4 VS is the potential at the Source.
VD is the potential at the Drain.
VDS (=VS-VD) is the difference of potential between Source and Drain. VGS (=VS-VG) is the difference of potential between Source and Gate.
Figure 2c
VDS>0, VGS>0 EF(1) is the Fermi-Level in material 1
EF(2) is the Fermi-Level in material 2
EC(2) is the conduction band edge of material 2
EV(2) is the valence band edge of material 2
EF(3) is the Fermi-Level in material 3 EC(3) is the conduction band edge of material 3
EV(3) is the valence band edge of material 3
EF(4) is the Fermi-Level in material 4
VS is the potential at the Source.
VD is the potential at the Drain. VDS (=VS-VD) is the difference of potential between Source and Drain.
VGS (=VS-VG) is the difference of potential between Source and Gate.
ECn(2) is the region of the conduction band of material 2, that is below the
EF(2), as an effect of positive Gate to Source voltage (VGS > 0).
In Figure 3a, VDS=0, VGS=0.
EF(1 ) is the Fermi-Level in material 1
EF(2) is the Fermi-Level in material 2
EC(2) is the conduction band edge of material 2
EV(2) is the valence band edge of material 2 EF(3) is the Fermi-Level in material 3
EC(3) is the conduction band edge of material 3
EV(3) is the valence band edge of material 3
EF(4) is the Fermi-Level in material 4
VS is the potential at the Source. VD is the potential at the Drain.
VDS (=VS-VD) is the difference of potential between Source and Drain.
VGS (=VS-VG) is the difference of potential between Source and Gate. In Figure 3b, VDS<0, VGS=0.
EF(1) is the Fermi-Level in material 1
EF(2) is the Fermi-Level in material 2 EC(2) is the conduction band edge of material 2
EV(2) is the valence band edge of material 2
EF(3) is the Fermi-Level in material 3
EC(3) is the conduction band edge of material 3
EV(3) is the valence band edge of material 3 EF(4) is the Fermi-Level in material 4
VS is the potential at the Source.
VD is the potential at the Drain.
VDS (=VS-VD) is the difference of potential between Source and Drain.
VGS (=VS-VG) is the difference of potential between Source and Gate.
In Figure 3c, VDS<0, VGS<0.
EF(1) is the Fermi-Level in material 1
EF(2) is the Fermi-Level in material 2
EC(2) is the conduction band edge of material 2 EV(2) is the valence band edge of material 2
EF(3) is the Fermi-Level in material 3
EC(3) is the conduction band edge of material 3
EV(3) is the valence band edge of material 3
EF(4) is the Fermi-Level in material 4 VS is the potential at the Source.
VD is the potential at the Drain.
VDS (=VS-VD) is the difference of potential between Source and Drain.
VGS (=VS-VG) is the difference of potential between Source and Gate.
EVn(2) is the region of the valence band of material 2, that is above the EF(2), as an effect of negative Gate to Source voltage (VGS < 0).
When comparing Figures 2a and 3a, it can be noted that the device is identical for both cases. The device will however behave as an N-MOS (Figures 2b and 2c) or a P-MOS (Figures 3b and 3c) in function of the voltage applied. It should be noted that the vertical axis in Figures 2 and 3 is expressed as potential in Volts. It could also be expressed as potential energy in electron-volts. The same applies to other figures where potential is indicated.
Figure 4a, 4b, 4c, 4d
Schematic band alignments of two identical SD-CMOS devices, along a cut from Source to Drain, near the interface with the Gate insulator, for different Drain and Gate bias conditions.
The devices are connected together in a "CMOS Inverter" arrangement: The Drains are connected together. The Gates are connected together. The Source of the device on the left hand side of the figure, is connected to the ground potential. This device will behave as a NMOS. The Source of the device on the right hand side of the figure, is connected to a negative potential. This device will behave as a PMOS.
Figure 4a
Initial condition.
VG=GND.
VD=GND.
The device on the left has just been switched "Off1. The device on the right has just been switched "On", and current starts to flow.
Figure 4b Steady state. VG=GND. VD=-VSS. The device on the left is still "Off". The device on the right is "On", but VDS=0, and therefore current=0.
Figure 4c
Transient when VG is switched to -VSS. VG=-VSS. VD=-VSS.
The device on the left has just been turned On. VDS>0, current flows. The device on the right has just been turned "Off'.
Figure 4d
Steady state.
VG=-VSS.
VG=GND.
The device on the left is "On", but VDS=0, and therefore current=0. The device on the right is "Off".
Figure 5a, 5b, 5c, 5d, 5e, 5f
Schematic band alignments of one SD-CMOS device, along a cut from
Source to Drain, near the interface with the Gate insulator, for different Drain and Gate bias conditions.
The Source of the device is going to be changed between GND and -VSS potentials.
When VS=GND, the device behaves as a NMOS transistor.
When VS=-VSS, the device behaves as a PMOS transistor. The Gate of the device switches between GND and -VSS potentials.
Figure 5a Initial condition: VS=GND. VG=GND. VD=GND. Therefore: VDS=0. VGS=0. The device is "Off" as NMOS. No electron current flows.
Figure 5b
Transient when VS is switched to -VSS (setting the device to work as PMOS), and VG is kept at GND.
VS=-VSS.
VG=GND. VD=GND.
Therefore:
VDS=-VSS.
VGS=-VSS.
The device has just been turned "On" as PMOS, and hole current starts to flow.
Figure 5c
Steady state when VS=-VSS, and VG=GND.
VS=-VSS. VG=GND.
VD=-VSS.
Therefore:
VDS=0.
VGS=-VSS. The device is "On" as PMOS, but VDS=0 and no hole current flows.
Figure 5d
Transient when VS is switched to GND (setting the device to work as NMOS), and VG is kept at GND. VS=GND. VG=GND. VD=-VSS. Therefore: VDS=+VSS. VGS=0.
The device has just been switched from a "On-state" PMOS, to a "Off-state" NMOS. No electron current flows.
Figure 5e
Transient when VS is kept at GND, and VG is switched to -VSS.
VS=GND. VG=-VSS.
VD=-VSS.
Therefore:
VDS=+VSS.
VGS=+VSS. The device has just been turned "On" as NMOS, and "electron current" starts to flow.
Figure 5f
Steady state when VS=GND, and VG=-VSS. VS=GND.
VG=-VSS.
VD=GND.
Therefore:
VDS=0. VGS=+VSS.
The device is "On" as NMOS, but VDS=0 and no electron current flows.
"Single-Device CMOS" concept
The device concept of the present invention, is independent of any particular implementation. It can be implemented in different materials systems, like Si- based and GaAs-based alloys, for example. Independently of the materials system, it can also be implemented with different "Process Flows" or "Process Integration Architectures".
Exploiting the unique ability of defining asymmetric Vertical MOSFETs, it is possible to make an "Universal MOSFET" device, that behaves as NMOS or as PMOS depending only on the applied bias.
It is only possible to conceive and fabricate such device if there is no doping in any region of the device. The fundamental point is to have symmetric paths from source to drain for electrons and for holes. If doping was introduced, this symmetry would brake immediately.
When drawing a straight line along the middle of the band-gap of the several materials/regions of the device (source, channel, drain), the shape of the conduction band should mirror the shape of the valence band (or vice-versa), the middle of the band-gap being the mirror line.
The Source material is made of an undoped semiconductor, with a very narrow band-gap, which is much narrower, in particular 9 to 10 times narrower, than the band-gap of the channel material. The band-alignment must be such that the band-gap of the source material is fully nested in the band-gap of the channel material. The offsets in the conduction and valence bands should be of the same magnitude. Actually, due to the differences in effective masses, the conduction and valence band offsets may in fact be slightly different.
The Channel material is made of an undoped semiconductor, with a band- gap wide enough to comprise the barrier height for electrons and for holes, and the very small band-gap of the source material. An elemental semiconductor might also have the advantage of no alloy scattering. The barrier height for electrons and holes, determines off-state current (for electrons and holes respectively), and therefore should be large enough to enable room-temperature operation with negligible "off-state" currents. Ideally the barrier heights ought to be engineering parameters, which should be possible to continuously vary across a wide range of values (for example by variation of alloy compositions of the source layer).
The Drain is defined by a Schottky junction, between the channel and a metal with a workfunction or Fermi-Level in the middle of the band-gap of the channel material.
The Gate electrode also needs to have a Fermi-Level in the middle of the band-gap of the channel material: metal with workfucntion in the middle of the band-gap of the source and channel materials.
Referring to Figures 1 , 2a, 3a and 13a, the metal insulator semiconductor field effect transistor (MISFET) according to the invention comprises a source layer 2 being made with a material having a source band gap (EG2) and a source mid-gap value (EGM2), said source layer having a source Fermi-Level (EF2). A drain layer 4 has a drain Fermi-Level (EF4). A channel layer 3 is provided between the source layer and the drain layer. The channel layer is made with a material having a channel band gap (EG3) and a channel mid- gap value (EGM3). The channel layer further has a channel Fermi level (EF3). A source contact layer 1 is connected to the source layer opposite the channel layer, said source contact layer having a source contact Fermi-Level (EF1). A gate electrode 6 has a gate electrode Fermi-Level (EF6). According to the invention, the source band gap is substantially narrower (EG2) than said channel band gap (EG3), in particular at least 9 to 10 times. The source contact Fermi-Level (EF1), the source Fermi-Level (EF2), the channel Fermi- Level (EF3), the drain Fermi-Level (EF4) and the gate electrode Fermi-Level (EF6) are equal to the source mid-gap value (EGM2) and the channel mid- gap value (EGM3), within a predetermined tolerance value, when no voltage is applied to the device. The source band gap (EG2) is the difference between the conduction band edge (EC2) and the valence band edge (EV2) for the source. It can be equated as follows :
EG2 = EC2 - EV2
The channel band gap (EG3) is the difference between the conduction band edge (EC3) and the valence band edge (EV3) for the channel. It can be equated as follows :
EG3 = EC3 - EV3
The source mid-gap value can be equated as follows :
EGM2 = (EC2 - EV2)/2;
The channel mid-gap value can be equated as follows : EGM3 = (EC3 - EV3)/2
The band-gap of the source material EG2 could be for example around 0.11 eV, with a tolerance of plus or minus (±) 5% on this value (total of 10%), resulting in the range of 0.1 to 0.12 eV. The band-gap of the channel material EG3 should be around 1.1 eV, with a tolerance of plus or minus (±) 5% on this value (total of 10%), resulting in the range of 1.0 to 1.2 eV.
The tolerance on the barrier heights for electrons and holes (conduction and valence band offsets respectively):
EC3 - EC2 = 0.5 eV (±) 5% (total of 10%), resulting in a range from 0.475 eV to 0.525 eV
EV3 - EV2 = 0.5 eV (±) 5% (total of 10%), resulting in a range from 0.475 eV to 0.525 eV
The tolerance value as indicated in claim 1 can be expressed in different ways. For example, in the absence of applied voltage to any of the terminals of the device, the Fermi-Levels in the source and channel regions should be close to the following values: For the source, EF3 = EV3 + EGM3, plus or minus (±) 5% (total of 10%), with EGM3 = (EC3 - EV3) / 2 = EG3 / 2.
Taking EGM3 to be the reference (i.e. zero value) and EG3 = 1.1eV, then EF3=0 (±) 0.05 eV or will range from -0.05 eV to + 0.05 eV. In that case the 0.05 eV tolerance is approximately 0.05/1.1 or +/- 5% of the channel band gap (EG3). A person skilled in the art will understand that the tolerance value can be expressed diffenently.
In particular, the device according to the invention has the following layers with the following characteristics: • layer 1 is the contact to the Source: metal with Fermi-Level in the middle of the band-gap of the Source material, and therefore also in the middle of the band-gap of the channel material;
• layer 2 is the Source: "narrow" band-gap material, with its mid-gap point aligned with the mid-gap point of the channel material, resulting in similar offsets in the conduction and valence bands, with respect to the channel material;
• layer 3 is the channel: "wide" band-gap material;
• layer 4 is the Drain: metal with a Fermi-Level in the middle of the gap of the channel material; • layer 5 is the Gate insulator; and
• layer 6 is the Gate electrode: conductor with Fermi-Level in the middle of the gap of the channel material.
Layers 1 , 4, and 6, can be the same material.
Figures 2a, 2b, 2c, show schematics of the band-diagrams for positive bias conditions, when the device behaves as a NMOS transistor. Figures 3a, 3b, 3c, show schematics of the band-diagrams for negative bias conditions, when the device behaves as a PMOS transistor.
Embodiment of the invention in the silicon materials system
Because of its economic relevance, silicon-based technology is very important. The embodiment of the SD-CMOS with silicon-compatible materials, will make use of silicon-based alloys like Sii-χGeχ, Sh-yCy, Si -χ- yGexCy, etc. Due to layer formation difficulties, it is less likely that alloys with Sn can be used. However if the perceived/ anticipated technological issues are overcome, compounds with this element could also be used.
In Figure 1 , the layer description for a possible implementation in the silicon materials system, would read:
1 Epitaxial Titanium Nitride (TiN) on silicon.
2 Undoped Sii-χ-yGexCy random alloy or Sii-χGex/Sii-yCy superlattices, grown pseudomorphically on silicon.
3 Undoped Silicon.
4 Epitaxial Titanium Nitride (TiN) on silicon.
5 Typically can be SiO2, or SiON/Si3N4, etc.
6 Titanium Nitride (TiN).
The Source is a very narrow (for example 5 KT, which at room temperature is about 130 mili-electron-Volts) band-gap material pseudomorphically grown on silicon. The band-alignment with silicon is such that the band offsets in the conduction and valence bands must be symmetric (for example 0.5 volts for each band discontinuity). Examples of possible materials providing such requirements, are a combination of Sh-χGeχ, Sh-yCy, Sii-χ-yGexCy, either as random alloys or as short-period superlattices of alternating layers of Sii- yCy, and Sii-χGeχ, for example. The exact composition and thickness of these layers is an engineering question, not a conceptual one. Enough data is already known about these alloys (see reference [8]), to be able to predict that some combination will deliver the band-alignment necessary for this concept.
The Source is contacted by a metal electrode with a Fermi-Level in the middle of the silicon bad-gap. Since the band-gap of the source material is very narrow, and centered (with equal discontinuities for the conduction and valence bands) in the band-gap of silicon, it means that the Fermi-Level of the metal at the source is also in the middle its band-gap. Therefore it is possible to have good ohmic contacts without any rectifying properties (for both electrons and holes) between the source metal and the very narrow semiconductor at the source, even though there is no doping involved. TiN (titanium nitride) is an example of a metal with such properties (see reference [9]).
The Channel, is made of non-doped pure silicon.
The Channel/Drain interface is a Schottky junction, between the silicon channel, and a metal drain with a Fermi-Level near the mid-gap of silicon. Again TiN (titanium nitride) is an example of a metal with such properties.
The Gate electrode is a conductor with the Fermi-Level in the middle of the band-gap of silicon. Once more, TiN (titanium nitride) is an example of a metal with such properties.
Because of the mid-gap values of the Fermi-Level of the metals in the source, drain, and gate electrode, and also because of the symmetry of the band offsets, the Fermi-Level (or Chemical Potential) will be in the middle of the band-gaps of the source (narrow band-gap material) and channel regions. For the same reasons, a "flat band condition" exists across the gate to channel interface. So, from the shape of the electrostatic potential (band edges), the physical picture for electrons and holes is very symmetric. However, in real space, there is an asymmetry between the source/channel and the channel/drain interfaces.
With a Schottky junction at the drain, the reverse "Off-state" current, is the thermionic current over the barrier. For a metal with mid-gap Fermi-Level on non-doped silicon, that current is indeed very low. Therefore, the drain of these devices is not able to inject current over the barrier, thereby preventing it from acting as the source of the complementary device-type.
The concept for the "On/Off switching mechanism", that is the process to modulate the barrier height at the source/channel interface, was introduced with Vertical Heterojunction MOSFETS, and is described in reference [3]. That mechanism was verified by numerical simulation for PMOS devices.
In the present invention, that mechanism was adapted to a device requiring symmetry for electrons and holes. Therefore, for the present invention, the required band-alignments (and consequently alloy composition) of that region in the source, are different than those for the Vertical Heterojunction MOSFET (PMOS or NMOS). Unfortunately, at this moment in time, there are no commercial device simulators available, capable of simulating such structure.
The very narrow band-gap material is positioned between the metal contact and the silicon channel, in order to enable a "switch-On/Off" effect mentioned before. If the metal contact was positioned directly on the silicon channel (Schottky junction), that switch On/Off" mechanism would not be possible.
The presence of the very narrow band-gap film in the source layer is the key enabler of this mechanism, which makes possible the lowering of the effective barrier height (distance between band edge in the channel and the Fermi- Level in the source), for either electrons or holes. The presence of that film, also breaks the symmetry between source and drain interfaces with the channel material.
That film enables drift-diffusion or ballistic current across the source to channel heterojunction. If a Schottky junction was made directly on silicon (at the source), the barrier height could not be changed, and the only turn-on current mechanism possible would be tunneling (see references [10, 11]).
The issue of asymmetry between source to channel and channel to drain interfaces is of utmost importance. If the device was symmetric, CMOS would not be possible, because there would be no difference between a NMOS in the "On-state" and a PMOS in the "Off-state" and vice-versa. Therefore, no transistor would ever block current, and a CMOS inverter could not be built.
"CMOS INVERTER" CONFIGURATIONS
Since the same transistor can behave as "NMOS" or "PMOS" devices (depending only on the applied bias), it is possible to envisage new and very flexible ways of designing circuits, making use of complementary operation.
For example, a "CMOS Inverter" could be made either by having "conventional" "Static Source Voltage Supply", in which case it would require two transistors, or by having a "Dynamic Source Voltage Supply", in which case only one transistor is required.
In the former, the devices will be "separated" into NMOS- or PMOS-like devices, by the metallization scheme, in which power supply voltages for NMOS and PMOS transistors are "hardwired".
In the later, if rather than "hardwiring" the source lines of SD-CMOS transistors, the power supply can be alternated between "positive" and "negative" voltages, the same device acts like "NMOS" AND "PMOS" sequentially in time.
By changing the bias of the source, and maintaining the gate bias constant while the source voltage is changed, a "CMOS Inverter" with just one transistor is possible.
Devices with "hardwired" fixed source voltage supplies are faster, and devices with changing source voltage supply are much more compact.
In fact both options can co-exist in the same integrated circuit, because it is only a matter of interconnect layout. These features were never possible before the SD-CMOS concept of the present invention, and allow for maximum flexibility in optimizing circuit speed, circuit size (number of transistors), power dissipation, etc.
Figure 4a, 4b, 4c, and 4d, show schematics of the band diagrams of two identical SD-CMOS devices with common Gates and Drains, for different bias conditions.
The potentials applied to the sources are such that the device on the left will behave as a NMOS, and the device on the right will behave as a PMOS.
Figure 4a time=0: Initial condition.
VG=GND.
VD=GND.
The device on the left has just been switched "Off".
The device on the right has just been switched "On", and current starts to flow.
Figure 4b time=1 : steady state. VG=GND. VD=-VSS.
The device on the left is still "Off". The device on the right is "On", but VDS=0, and therefore current=0.
Figure 4c time=2: transient when VG is switched to -VSS. VG=-VSS. VD=-VSS.
The device on the left has just been turned On. VDS>0, current flows. The device on the right has just been turned "Off".
Figure 4d time=3: steady state.
VG=-VSS.
VG=GND.
The device on the left is "On", but VDS=0, and therefore current=0.
The device on the right is "Off".
Figures 5a, 5b. 5c. 5d. 5e. 5f
Schematic band alignments of one SD-CMOS device, along a cut from
Source to Drain, near the interface with the Gate insulator, for different Drain and Gate bias conditions. The Source of the device is going to be changed between GND and -VSS potentials.
When VS=GND, the device behaves as a NMOS transistor.
When VS=-VSS, the device behaves as a PMOS transistor.
The Gate of the device switches between GND and -VSS potentials.
These figures demonstrate that with the SD-CMOS concept, a single device can implement, sequentially, the functionality of the CMOS inverter. The type of MOSFET that the SD-CMOS will behave like, is set by the Source potential. For a potential at the source suitable for NMOS devices, the SD- CMOS will behave as NMOS transistor. For a potential at the source suitable for PMOS, the SD-CMOS will behave as a PMOS transistor.
Drain voltage is the output of the Inverter. Gate Voltage is the input of the Inverter.
While maintaining the input (gate voltage), the source voltage is switched between "0" and "-1", in which case the device "behaves" as NMOS and PMOS respectively.
Depending on the gate voltage (the input), there will be electron current, or hole current, or no current at all, depending on the potential at the drain (set by the previous logic state).
Figure 5a
Initial condition:
VS=GND.
VG=GND. VD=GND.
Therefore:
VDS=0.
VGS=0.
The device is "Off" as NMOS. No electron current flows.
Figure 5b
Transient when VS is switched to -VSS (setting the device to work as PMOS), and VG is kept at GND.
VS=-VSS. VG=GND.
VD=GND.
Therefore: VDS=-VSS.
VGS=-VSS.
The device has just been turned "On" as PMOS, and hole current starts to flow.
Figure 5c
Steady state when VS=-VSS, and VG=GND.
VS=-VSS.
VG=GND. VD=-VSS.
Therefore:
VDS=0.
VGS=-VSS.
The device is "On" as PMOS, but VDS=0 and no hole current flows.
Figure 5d
Transient when VS is switched to GND (setting the device to work as NMOS), and VG is kept at GND.
VS=GND. VG=GND.
VD=-VSS.
Therefore:
VDS=+VSS.
VGS=0. The device has just been switched from a "On-state" PMOS, to a "Off-state"
NMOS. No electron current flows.
Figure 5e
Transient when VS is kept at GND, and VG is switched to -VSS. VS=GND. VG=-VSS. VD=-VSS. Therefore: VDS=+VSS. VGS=+VSS.
The device has just been turned "On" as NMOS, and "electron current" starts to flow.
Figure 5f
Steady state when VS=GND, and VG—VSS.
VS=GND. VG=-VSS.
VD=GND.
Therefore:
VDS=0.
VGS=+VSS. The device is "On" as NMOS, but VDS=0 and no electron current flows.
IMPACT ON CMOS CIRCUIT DESIGN
In conventional "Planar CMOS", the topology of choice for logic gates is the "NAND" configuration. In the typical "NAND" configuration, the logic inputs are the gate terminals of a series of n-type MOSFETs, connected in series with a PMOS (the load transistor). Each additional logic input, requires an additional NMOS device to be inserted in the series with all of them.
In the typical "NOR" configuration, the sources of several NMOS are shunted together, and the same is done for the drains. The set of parallel NMOS devices is connected in series with a PMOS device (the load transistor). Each additional logic input requires an extra NMOS device to be connected in parallel with the other NMOS transistors.
The main reasons for NANDs being the configuration of choice for "Planar MOSFETs are: 1) For "Planar MOSFETs", the series connection enables area savings, because for devices of the same type, the source of one transistor can be the drain of another. However, for bulk CMOS, and because of the "body effect", the number of number of logic inputs is typically reduced to two. A larger number of inputs is possible, only if Silicon- On-lnsulator (SOI) technology is used.
2) As the devices are connected in series, the total "Off-state" current, is the "Off-state" current of the least leaky device. For "NOR" gates. The total "Off-state" current is the sum of the "Off- state" currents of ail individual NMOS devices.
For SD-CMOS, "NOR" logic gates are the best choice, for the following reasons: 1 ) Series connection of Vertical MOSFETs is area inefficient
2) Parallel connection of sources and drains can be achieved with very area-efficient process integration schemes.
3) With heterojunctions, "Off-state" current can be well controlled, even for ultra-short channel lengths. 4) Due to the lack of "body-effect", logic gates can have many inputs
(gates), thereby saving area, minimizing interconnect complexity, and decreasing power dissipation.
As already mentioned, the physics of these devices, makes possible channel lengths of 20nm for example. For such short distances transport between source and drain is ballistic even at room temperature. Very low voltage operation (less than 1Volt), very low power consumption, very high current drive (ID 1mA/μm at VGS=VDS=1Volt), and very short ring oscillator delays
(<1 ps), can be expected. On the assumption that 1 ps ring oscillator delays are possible, a conservative estimate would put 200GHz circuit operation well within reach of this CMOS technology. This kind performance level will enable circuits made with these devices to digitize and to synthesize any electrical signal of technological relevance for commercial applications, at the present of in the foreseeable future, thereby eliminating the need for analog signal processing, including signal demodulation in RF/Millimiter-Wave circuits. At 200GHz digital circuit operation, signal demodulation can also be performed by Digital Signal Processing units.
This represents a tremendous breakthrough for silicon-based RF/Millimiter- Wave circuitry, and fundamental shifts in design of such circuits can be expected. With SD-CMOS devices, there is no more a conceptual difference between the "logic" and the "analog RF" transistors. All transistors are digital, and all operate at RF/Millimiter-Wave speeds.
CMOS PROCESS INTEGRATION SCHEMES WITH EMBEDDED MEMORY
Several "Process Integration Architectures" can be envisaged for the new device. Three main alternative exemplary schemes will be described with the disclosure of the present invention.
"GATE ALL AROUND" OR "SURROUNDING GATE"
In this process architecture, the step which defines "Mesas" exposing the device layers where the gate stack is formed, simultaneously provides "isolation" between devices. The contacts to the several device layers are made inside the perimeter of the gate stack (which as its name suggests, surrounds the device layers).
Figure 6, is a schematic of 3-dimensional perspective of a "Gate All Around" device. "EDGE-GATE"
In this process architecture, the "Isolation" and "Gate Stack" formation steps are performed separately. The gate is not surrounding the "Mesa" of the device layers, but it is placed on only one crystalline plane. The other sides of the "device layer Mesa" face the "Field Isolation" for example. With this configuration, it is possible to have a single drain contact for several independent gates.
Figure 7a, is a schematic of 3-dimensional perspective of one possible implementation of the "Edge-Gate" arrangement.
Figure 7b, is a schematic of 3-dimensional perspective of a second possible implementation of the "Edge-Gate" arrangement.
Figure 7c, is a schematic of 3-dimensional perspective of third possible implementation of the "Edge-Gate" arrangement.
"INNER-GATE"
In this process architecture, the gate is surrounded by the device layers, which at the edges face the "Field Isolation". The "Field Isolation" and "Gate Stack" formation steps are performed separately. This configuration enables not only single drain contact, but also single source contact, for several independent gates. This is an ideal configuration for "NOR" logic gates.
Figure 8, is a schematic of 3-dimensional perspective of a possible implementation of the "Inner-Gate" together with "Edge-Gate" arrangements.
Figure 9 is a top view of a possible implementation of the "Inner-Gate" together with "Edge-Gate" arrangements.
Figure 9a, shows an implementation where SD-CMOS devices to behave as NMOS and PMOS are made on opposite sides of the "Drain Contact". Figure 9b, shows an implementation where a single SD-CMOS device will behave alternatively as NMOS or PMOS by changing the bias at the source contact.
The "Edge-Gate" and "Inner-Gate" process architectures differ only in layout, and both types of devices can be simultaneously on the same circuit, as it is shown in Figure 8.
Any of the "SD-CMOS" process integration architectures, has many advantages over "Planar CMOS" as well as over "Vertical Integration of Vertical MOSFETs" where NMOS and PMOS devices are physically different. Some of these advantages are:
1 ) significant reduction in the number of process steps, 2) area savings for a single device,
3) area savings for "many-input" CMOS inverters/logic-gates. This is especially apparent for the "Edge-Gate" and "Inner-Gate" integration architectures, due to the possibility of a single drain contact, and the fact that several gates (NOR) are within the same "Active-Area". 4) Change the "type" of transistor by simply changing bias conditions.
SD-CMOS does not suffer from "corner effect", due to the intrinsics of the device physics, and therefore none of the possible implementations or process integration architectures, is affected by it. The reason for the immunity to "corner effect", is that the "corner effect" is a geometric effect that re-enforces the "zero-bias" electric field across a MOS interface.
In standard "planar technology", it is typical to have a MOS with a n+ poly gate electrode across from a p-type doped potential well. Even when there is no bias applied, there is a built-in electric field, because of the differences in
Fermi-Levels between the n+ poly gate and the p-type well. This built-in eletric field is re-enforced by curvature, as it can be the case at the edge of the gate (in the width direction).
The "zero-bias" electric field across the MOS interface of the SD-CMOS is zero, because of a flat-band condition imposed by the non-doping of the device layers and by the "mid-gap" Fermi-Level of the gate electrode, drain and source contact metals. Therefore and "by definition" the SD-CMOS does not suffer from "corner effect".
PROCESS FLOWS
References [12] to [16] demonstrate the feasibility of epitaxial TiN deposition on silicon, and some of these references also show the feasibility of epitaxial growth of silicon on TiN.
References [17] and [18], demonstrate the feasibility of epitaxial insulators like SrTiO3 and BaTiO3 (the later a Ferroelectric) using epitaxial TiN on silicon as a buffer material.
The process flows described below, give exemplary ways of embodying the several innovations being disclosed in this patent application.
CMOS Integration schemes for Vertical MOSFETs have been proposed before, but for SD-CMOS only one device structure needs to be fabricated, and that carries important consequences.
Three of the four the process flows will include the fabrication steps necessary to have embedded memory, which depending on the materials chosen, can have ferroelectric properties, and therefore provide non-volatile storage of the data. Since the SD-CMOS is an asymmetric Vertical MOSFET, it means that source and drain are not interchangeable, and therefore, choosing which one is at the bottom and at the top of the layer stack, carries consequences.
Having the source as the bottom layer, is the easiest configuration from a technological standpoint. However, this may result in a high series resistance of the source, which is perceived to be a key parameter to minimize for devices with channel lengths below 10Onm.
Source as the top layer implies one of the following options:
1) pseudomorphic growth of the channel (Si) and source (SiGeC) layers to be performed after the epitaxy of the drain metal film (TiN).
2) Pseudomorphic growth of source (SiGeC) and channel (Si), followed by epitaxy of drain metal film (TiN). In order to have the source as the top layer, wafer bonding of the top layer to an insulating substrate, and etchback of the original substrate (where the device layer growth was performed) will expose the source layer at the top of the stack. The new insulating substrate must be compatible with the remaining steps of the process flow. Examples of insulating substrates are glass, quartz, sapphire, etc.
Another set of options regarding the device layers, has to do with the formation of the device Mesas: blanket growth followed by patterning of the epitaxial layers, or pre-patterning of hard mask followed by selective epitaxial growth.
Each of these options has technological advantages and drawbacks over the other.
There are several possible choices regarding the gate architecture:
1 ) "Gate All Around" or "Surrounding Gate"
2) "Edge-Gate" 3) "Inner-Gate"
Inner-Gate and Edge-Gate can be implemented simultaneously, without extra masks.
The combinations of the different options for what device layer is on top of the epitaxial stack, if wafer-bonding or pseudomorphic growth of SiGeC and Si is performed on metals, blanket versus selective epitaxial growth, and the different gate architectures, lead to a very large number of possible process flows.
PROCESS FLOW #1
"Gate All Around" device architecture, with source at the bottom of device layer stack.
Epitaxial Device-Layer Growth
1) Bare silicon wafer (undoped, <100>).
2) Epi of source layer (undoped Sii-χ-yGexCy or Si -xGex/Sii-yCy superlattice)
3) Epi of channel layer (undoped Si) 4) Epi of drain layer (TiN)
5) Epi of Insulator which can be Ferroelectric (for example BaTiO)
6) Epi of capacitor plate (TiN)
7) Deposition of thin SiO2 and thick Si3N4 Figure 10A
Definition of Mesa structures
8) Lithography --> Maskl : definition of mesas.
9) Etch through Si3N4, stopping on SiO2.
10) Resist strip & clean. Figure 10B
11 ) Damage-less removal of oxide (HF dip or vapor for example). Figure 10C
12) Damage-less etch of trenches through the device layers, stopping in the wafer bulk. Figure 10D
13) Deposition (CVD for example) of gate stack (gate insulator & gate electrode).
Figure 10E
Formation of gate contact pads
14) Lithography -> Mask2: patterning of "Gate Stack". Figure 10F
15) Etch through gate stack, stopping on Si3N4 and wafer bulk. Figure 10G
16) Resist strip & clean. Figure 10H
17) Trench filling with deposition of SiO2 (HDP-CVD for example).
18) Planarization by CMP, stopping on Si3N4. Figure 101
Formation of contact holes to top plate of capacitor
19) Lithography — > Mask3: patterning of contact holes.
20) Dry etch of Si3N4 and SiO2, stopping on top plate of capacitor.
21) Resist strip & clean. Figure 10J
22) Deposition (PVD or CVD for example) of metal to fill contact hole.
23) Planarization by CMP, stopping on SiO2 and Si3N4. Figure lOK
Source Contact
24) Lithography -> Mask5: contact to the source. 25) Etch through Si3N4, SiO2, TiN, Insulator, TiN, Si, stopping inside source (SiGeC).
26) Resist strip & clean. Figure 10L
27) Conformal deposition of SiO2 and Si3N4 28) Etchback to make innerwall spacers. Figure 10M
27) Deposition (PVD or CVD for example) of metal (with Fermi-Level in the middle of the band-gap of the source - SiGeC) to fill contact hole.
28) Planarization (CMP for example). Figure 10N
PROCESS FLOW #2
"Edge Gate" device architecture, with embedded capacitor and source at the bottom of device layer stack.
Epitaxial Device-Layer Growth 1 ) Bare silicon wafer (undoped, <100>).
2) Epi of source layer (undoped Sii-χ-yGexCy or Sii-χGex/Sii-yCy superlattice)
3) Epi of channel layer (undoped Si)
4) Epi of drain layer (TiN) 5) Epi of Insulator which can be Ferroelectric (for example BaTiO)
6) Epi of capacitor plate (TiN)
7) Deposition of thin SiO2 and thck Si3N4 Figure 11A
Definition of Trenches
8) Lithography — > Maskl : definition of trenches. 9) Etch through Si3N4, stopping on SiO2.
10) Resist strip & clean. Figure 11 B
11 ) Damage-less removal of oxide (HF dip or vapor for example). Figure 11C
12) Damage-less etch of trenches through the device layers, stopping in the wafer bulk.
Figure 11 D
13) Trench filling with deposition of SiO2 (HDP-CVD for example).
14) CMP of SiO2, stopping on Si3N4. Figure 11E
Formation of "edge-gate"
15) Lithography -> Mask2: trenches on the "Mesa" structures defined by Maskl .
16) Etch through Si3N4, SiO2.
17) Resist strip & clean. Figure 11 F
18) Etch through TiN, Insulator, TiN, Si, SiGeC. Figure 11 G
19) Deposition of gate stack (gate insulator & gate electrode). 20) Planarization by CMP, stopping on SiO2 and Si3N4. Figure 11 H Formation of contact holes to top plate of capacitor
21) Lithography — > Mask3: patterning of contact holes.
22) Dry etch of Si3N4 and SiO2, stopping on top plate of capacitor. 23) Resist strip & clean.
Figure 111
24) Deposition (PVD or CVD for example) of metal to fill contact hole.
25) Planarization (CMP for example). Figure 11 J
Source Contact
26) Lithography — > Mask4: contact to the source.
27) Etch through Si3N4, SiO2, TiN, Insulator, TiN, Si, stopping inside source (SiGeC).
28) Resist strip & clean. Figure 11 K
29) Conformal deposition of SiO2 and Si3N4 30) Etchback to make innerwail spacers.
Figure 11 L
27) Deposition (PVD or CVD for example) of metal to fill contact hole.
28) Planarization (CMP for example). Figure 11M
PROCESS FLOW #3
"Inner- & Edge- Gates" device architecture, with embedded capacitor and source at the top of device layer stack. Process flow for "Inner-Gate" & "Edge-Gate" with "source on top" by wafer bonding, with blanket epitaxial growth, having embedded Ferroelectric capacitor.
The circuit configuration chosen, is a "NOR Gate" with 5 inputs (4 "inner- gates" and 1 "edge-gate"). The capacitor layers are not used (are in fact transparent to the functionality of simple logic gates). In an small change to this flow with an extra mask, the capacitor flms could have been removed from the "logic-only" areas.
Epitaxial Device-Layer Growth
1) Bare silicon wafer (undoped, <100>).
2) Epi of source layer (undoped SiGeC)
3) Epi of channel layer (undoped Si)
4) Epi of drain layer (TiN) 5) Epi of Ferroelectric (for example BaTiO)
6) Epi of capacitor plate (TiN) Figure 12A
7) Wafer bonding to an insulating substrate (Quartz or Sapphire for example). Figure 12B
8) Selective etch of the wafer bulk, stopping on the source layer (SiGeC) Figure 12C
9) Clean & Epi of Metal (for example TiN) Figure 12D
10) CVD of SiO2 / Si3N4 Figure 12E
11 ) Alignment Markers for Lithography tool. Isolation of Source & Channel layers
12) Lithography — > Maskl: definition of mesas.
13) Etch through Si3N4, stopping on SiO2
14) Resist strip & clean, with HF to remove thin SiO2. Figure 12F
15) Selective etch of TiN (for example with H2O2). Figure 12G
16) Low temperature oxidation of SiGeC (Source) and Si (Channel). Figure 12H
17) Trench filling with deposition of SiO2 (HDP-CVD for example).
18) CMP of SiO2, stopping on Si3N4. Figure 121
Formation of "Edge" and "Inner" Gates.
19) Lithography —> Mask2: trenches on the "Mesa" structures defined by Maskl . 20) Etch through Si3N4, SiO2, TiN, SiGeC, Si, TiN, Ferroelectric, TiN.
21) Resist strip & clean. Figure 12J
22) Deposition (CVD) of gate stack (gate insulator & gate electrode). 23) CMP of gate stack, stopping on Si3N4 and SiO2.
Figure 12K
Drain Contact
24) Lithography — > Mask3: contact to the drain. 25) Etch through SiO2, stopping on TiN. 26) Resist strip & clean. Figure 12L 27) Deposition (PVD or CVD for example) of metal to fill contact hole.
28) Planarization (CMP for example). Figure 12M
Source Contact
29) Lithography — > Mask4: contact to the source.
30) Etch through Si3N4, SiO2, stopping on TiN.
31) Resist strip & clean. Figure 12N
32) Deposition (PVD or CVD for example) of metal to fill contact hole.
33) Planarization (CMP for example). Figure 120
Isolation of Drain layers
34) Lithography -> Mask5: patterning of "drain layer (a metal)", which can also be used as "local interconnect".
35) Etch through SiO2 (field isolation), and drain metal film (TiN). 36) Resist strip & clean.
Figure 12P
Ground Plane Contact
37) Lithography -> Mask6: contact to the ground plane. 38) Etch through SiO2 and Ferroelectric Insulator layer, stopping on TiN.
39) Resist strip & clean.
40) Deposition (PVD or CVD for example) of metal to fill contact hole.
41) Planarization (CMP for example), (not shown in the drawings)
A Metal-Insulator-Semiconductor Field-Effect Transistor (MIS-FET) is disclosed where the electrostatic potential in a single device is symmetric for electrons and holes, in the source, channel, drain and gate regions, with the source to channel interface being different from the channel to drain interface.
A Metal-lnsulator-Semiconductor Field-Effect Transistor (MIS-FET) is disclosed where the electrostatic potential in a single device is symmetric for electrons and holes, in the source, channel, drain and gate regions, with the source to channel interface being different from the channel to drain interface, and the source to channel barrier height can be modulated by gate action (field-effect), and the channel to drain barrier height is not affected by gate bias.
A Metal-lnsulator-Semiconductor Field-Effect Transistor (MIS-FET) is disclosed, composed of the following active regions: a) A channel layer, made with a "wider" band-gap undoped semiconductor. b) A drain layer, made with a metal directly interfaced to the channel material (Schottky junction), having a Fermi-Level value in the middle of the band-gap of the channel material. c) A source layer, made with a semiconductor with a "narrower" band-gap, centered with the band-gap of the channel material (equal offsets in the conduction and valence bands between the source and channel layers). d) A source contact metal with a Fermi-Level in the middle of the band-gap of the channel material. e) A gate electrode with a Fermi-Level in the middle of the band-gap of the channel material.
A MISFET is disclosed with specific embodiment in the silicon materials system, having the following device layers: a) Channel material: undoped silicon (Si). b) Drain material: epitaxial titanium nitride (TiN). c) Source material: Sii-χ-yGβχCy, either as random alloy or short-period superlattice of alternating Sii-yCy/S -χGeχ layers for example. d) Source contact metal: epitaxial titanium nitride (TiN). e) Gate electrode: titanium nitride (TiN).
A Metal-lnsulator-Semiconductor Field-Effect Transistor (MIS-FET) concept os disclosed where a single device functions as a N- or P-type, depending only on the applied bias conditions, acting like a NMOS when drain to source (VDS) and gate to source (VGS) voltages are positive, and acts like a PMOS when drain to source (VDS) and gate to source (VDS) voltages are negative.
With the device according to the invention it is possible to build a one- transistor (1T) CMOS circuits having configurations where "inverters", "logic gates", memory cells, are fabricated with a single device which sequentially behaves as NMOS and PMOS, by appropriately sequentially changing the voltage at the source terminal.
It is also possible to build a Two (identical) Transistor (2T) CMOS circuits having configurations where "inverters", "logic gates", memory cells, are fabricated with two indentical devices, which behave as NMOS and PMOS by conecting their source terminals to the appropriate voltage sources.
The MISFET can be used for logic applications, for Random Memory (Dynamic, Static, Flash, Ferroelectric) applications, for logic applications with embedded Random Memory (Dynamic, Static, Flash, Ferroelectric) elements, for Random Memory (Dynamic, Static, Flash, Ferroelectric) applications with embedded logic elements, for co-integration with image sensors, like CCD and CMOS imagers, or any other kind, for co-integration with any kind of Micro-Electronic-Mechanical Systems (MEMS), or Micro-Optical or Optoelectronic integrated systems.
Process Integration Architecture depicted in Figure 7a can be achieved, where the Drain layer is at the bottom of the device layer stack (and therefore the Source at the top), and where a Gate Stack is positioned in such a way, as to be common to two devices, each one with its separate drain and source layers and respective contacts.
Process Integration Architecture depicted in Figure 7b can be achieved, where the Source layer is at the bottom of the device layer stack (and therefore the Drain at the top), and where a Gate Stack is positioned in such a way, as to be common to two devices, each one with its separate drain and source layers and respective contacts.
Process Integration Architecture depicted in Figure 7a can be achieved, where the Drain layer is at the bottom of the device layer stack (and therefore the Source at the top), and where a Drain Contact is positioned in such a way, as to be common to two devices, each one with its separate Gate Stack at the extreme edges of the device layer stack, and with its separate source layers and respective contacts.
Process Integration Architecture depicted in Figures 8 and 9a can be achieved where a single device layer stack, has single source and drain contacts, but several parallel Gates, for a compact "NOR-gate" arrangement. The arrangement as depicted, represents a solution with "Dynamic Source Voltage Supply".
Process Integration Architecture depicted in Figure 9b can be achieved, where a single device layer stack, has a single drain contact, two source contacts, and several parallel Gates, for a compact "NOR-gate" arrangement. The arrangement as depicted, represents a solution with "Fixed Source Voltage Supply".
The process flow according to Figures 10A to 10N can be used for CMOS logic, with very reduced number of extra processing steps for the inclusion of embedded memory (possibly Ferroelectric). The process flow according to Figures 11A to 11M can be used for CMOS logic, with very reduced number of extra processing steps for the inclusion of embedded memory (possibly Ferroelectric).
[The process flow according to Figures 12A to 12Q can be used for CMOS logic, with very reduced number of extra processing steps for the inclusion of embedded memory (possibly Ferroelectric).
REFERENCES
[1] "Vertical MOS Technology with sub 0.1 urn Channel Lengths", H.
Gossner, F. Wittman, I. Eisele, T. Grabolla, D. Behammer; Electronics Letters, 3rd of August 1995, Vol. 31 , No. 16, pp.1394-1395.
[2] "Novel Transport Simulation of Vertically-Grown MOSFETs, by Cellular Automaton Method", A. Rein, G. Zandler, M. Sarantini, P. Lugli, P. Vogl; IEDM 1994, pp. 351-354.
[3] U.S. Provisional Patent Application Number: 60/001 ,022; July 11th, 1995, "Vertical MOSFET Devices, Process of Manufacturing Them", Carlos J. R. P. Augusto.
[4] "Monte Carlo Simulation of a 30 nm Dual-Gate MOSFET: How Short Can Si Go?", D. J. Frank, S. E. Laux, and M. V. Fischetti, IEDM 1992, pp. 553- 556.
[5] U.S. Provisional Patent Application Number: 08/664,874; Jun. 17th, 1996 "DRAM Applications using Vertical MISFET Devices", Carlos J. R. P. Augusto
[6] "Impact of a Vertical F-Shape, Transistor (VFT) Cell for 1 Gbit DRAM and Beyond", S. Maeda, S. Maegawa, T. Ipposhi, H. Nishimura, H. Kuriyama, O. Tanina, Y. Inoue, T. Nishimura, N. Tsubouchi; IEEE Transactions on Electron Devices, Vol. 42, No. 12, December 1995.
[7] "ROS: An Extremely High Density Mask ROM Technology Based On Vertical Transistor Cells", E. Bertagnolli, F. Hofmann, J. Wilier, R. Maly, F. Lau, P.W. von Basse, M. Bollu, R. Thewes, U. Kollmer, U.
Zimmermann, M. Hain, W. H. Krautschneider, A. Rusch, B. Hasler, A. Kohlhase, H. Klose; Symposium on VLSI Technology Digest of
Technical Papers, pp. 58-59, 1996. [8] "SiGeC: Band Gaps, Band Offsets, Optical Properties, and Potential
Applications", K. Brunner, O. G. Schmidt, W. Winter, K. Eberl, M. Glϋck, U. Kόnig; J. Vac. Sci. Technol. B 16(3), May/Jun 1998, pp. 1701-1706.
[9] "Novel Polysilicon/TiN Stacked-Gate Structure for Fully-Depleted
SOI/CMOS"; Jeong-Mo Hwang, Gordon Pollack, IEDM 1992, pp. 45-
348.
[10] "A New Type of Tunnel-Effect Transistor Employing Interna Field
Emission of Schottky Barrier Junction", R. Hattori, A. Nakae, J.
Shirafuji; Jpn. J. Appl. Phys., Vol. 31 (1992), pp. L1467-L1469.
[11] "Numerical Simulation of Tunnel Effect Transistors Employing Internal Field Emission of Schottky Barrier Junction", R. Hattori, J. Shirafuji; SSDM 1993, pp. 258-260.
[12] "Pulsed Laser Deposition of Epitaxial Si/TiN/Si (100) Heterostructures"; R. Chowdhury, X. Chen, J. Narayan, Appl. Phys. Lett. 64(10), 7 March 1994.
[13] "Epitaxial TiN Based Contacts for Silicon Devices"; R. D. Vispute, J. Narayan, Journal of Electronics Materials, Vol. 25, No. 11, 1996, pp.1740- 1747.
[14] "Epitaxial Growth of TiN (100) on Si (100) by Reactive Magnetron
Sputering at Low Temperature"; W-H. Sheu, S-T. Wu, Jpn. J. Appl. Phys. Vol. 37, (1998) pp. 3446-3449, Part I, No. 6A, June 1998.
[15] "Atomic Layer Epitaxy Growth of TiN Thin Films from Til4 and NH3"; M. Ritala, M. Leskela, E. Rauhala, J. Jakinen; J. Electrochem. Soc. Vol. 145, No. 8, August 1998, pp. 29142920. [16] "Atomic Layer Deposition of TiN Films by Alternate Supply of
Tetrakis(ethylmethylamino)-Titanium and Amonia"; J-S. Min, Y-W. Son, W-G. Kang, S-S. Chun, S-W. Kang, Jpn. J. Appl. Phys. Vol. 37, (1998) pp. 4999-5004, Part i, No. 9A, Sept. 1998.
[17] "Study of Dielectric Properties of BaTiθ3 Thin Films on Si (100) with TiN Buffer Layer"; N. Shu, A. Kumar, M. R. Alam, H. L. Chan, Q. You, Applied Surface Science 109/10 (1997), pp. 366-370.
[18] "Structural and Dielectric Properties of Epitaxial SrTiO3 Films Grown on Si (100) Substrate with TiN Buffer Layer", M. B. Lee, H. Koinuma, J. Appl. Phys. 81(5), 1 March, 1997.
IDENTIFICATION OF THE WORDS IN FIGURES 10 A-N, 11 A-M, 12 A-Q
101 Nitride
102 Oxide
103 Plate
104 Insulator
105 Drain
106 Channel
107 Source
108 Undoped <100> Si substrate
109 Gate Insulator
110 Gate Electrode
111 Photo-Resist
112 Metal
113 Capacitor Plate (Metal)
114 Insulator (Ferroelectric)
115 Drain (Metal)
116 Channel (I-Si)
117 Source (I-SiGeC)
118 Insulator Substrate (e g : Q Quu;artz or Sapphire)
119 Source Contact (Metal)
120 SiO2
121 Si3N4

Claims

1. A metal insulator semiconductor field effect transistor (MISFET) comprising: • a source layer being made with a material having a source band-gap (EG2) and a source mid-gap value (EGM2), said source layer having a source Fermi-Level (EF2);
• a drain layer having a drain Fermi-Level (EF4);
■ a channel layer between the source layer and the drain layer, said channel layer being made with a material having a channel band-gap (EG3) and a channel mid-gap value (EGM3), said channel layer having a channel Fermi-Level (EF3);
■ a source contact layer connected to the source layer opposite the channel layer, said source contact layer having a source contact Fermi-Level (EF1); and
• a gate electrode having a gate electrode Fermi-Level (EF6) wherein:
■ said source band-gap (EG2) is substantially narrower than said channel band gap (EG3); • said source contact Fermi-Level (EF1), said source Fermi-Level (EF2), said channel Fermi-Level (EF3), said drain Fermi-Level (EF4) and said gate electrode Fermi-Level (EF6) are equal to said source mid-gap value (EGM2) and said channel mid-gap value (EGM3), within a predetermined tolerance value, when no voltage is applied to the device.
2. The MISFET of claim 1, wherein the first band gap is at least 9 to 10 times narrower than said second band gap.
3. The MISFET of claim 2, wherein the first band gap (EG2) is in the order of 0.1 to 0.12 electron-volts (eV) and the second band gap (EG3) is in the order of 1.0 to 1.2 electron-volts (eV).
4. The MISFET of any one of the preceding claims, wherein the predetermined tolerance value is lower than 10%, respectively, preferably lower than 5% of the channel band gap.
PCT/IB2000/000235 1999-02-24 2000-02-24 Misfet with narrow bandgap source WO2000051165A2 (en)

Priority Applications (6)

Application Number Priority Date Filing Date Title
DE20022706U DE20022706U1 (en) 1999-02-24 2000-02-24 MISFET
JP2000601674A JP2003523615A (en) 1999-02-24 2000-02-24 MISFET
US09/889,815 US6674099B1 (en) 1999-02-24 2000-02-24 Misfet
EP00905235A EP1166347A2 (en) 1999-02-24 2000-02-24 Misfet with narrow bandgap source
AU26857/00A AU2685700A (en) 1999-02-24 2000-02-24 Misfet
US10/725,830 US7023030B2 (en) 1999-02-24 2003-12-01 Misfet

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US12147399P 1999-02-24 1999-02-24
US60/121,473 1999-02-24

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US09889815 A-371-Of-International 2000-02-24
US10/725,830 Continuation-In-Part US7023030B2 (en) 1999-02-24 2003-12-01 Misfet

Publications (2)

Publication Number Publication Date
WO2000051165A2 true WO2000051165A2 (en) 2000-08-31
WO2000051165A3 WO2000051165A3 (en) 2001-05-31

Family

ID=22396952

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2000/000235 WO2000051165A2 (en) 1999-02-24 2000-02-24 Misfet with narrow bandgap source

Country Status (5)

Country Link
EP (1) EP1166347A2 (en)
JP (1) JP2003523615A (en)
AU (1) AU2685700A (en)
DE (1) DE20022706U1 (en)
WO (1) WO2000051165A2 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9905414B2 (en) 2000-09-28 2018-02-27 President And Fellows Of Harvard College Vapor deposition of metal oxides, silicates and phosphates, and silicon dioxide

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7352036B2 (en) * 2004-08-03 2008-04-01 Fairchild Semiconductor Corporation Semiconductor power device having a top-side drain using a sinker trench
KR101448899B1 (en) * 2007-06-12 2014-10-13 가부시키가이샤 한도오따이 에네루기 켄큐쇼 Capacitor-less memory

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4665414A (en) * 1982-07-23 1987-05-12 American Telephone And Telegraph Company, At&T Bell Laboratories Schottky-barrier MOS devices
EP0749162A2 (en) * 1995-06-16 1996-12-18 Interuniversitair Micro-Elektronica Centrum Vzw Vertical MISFET devices, CMOS process integration, RAM applications
US5801398A (en) * 1994-10-18 1998-09-01 Frontec Corporation Field effect transistor

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58188165A (en) * 1982-04-28 1983-11-02 Nec Corp Semiconductor device
JPS62274776A (en) * 1986-05-23 1987-11-28 Hitachi Ltd Semiconductor device
DE69629760T2 (en) * 1995-06-16 2004-07-08 Interuniversitair Micro-Electronica Centrum Vzw Vertical MISFET devices, CMOS process integration, RAM applications

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4665414A (en) * 1982-07-23 1987-05-12 American Telephone And Telegraph Company, At&T Bell Laboratories Schottky-barrier MOS devices
US5801398A (en) * 1994-10-18 1998-09-01 Frontec Corporation Field effect transistor
EP0749162A2 (en) * 1995-06-16 1996-12-18 Interuniversitair Micro-Elektronica Centrum Vzw Vertical MISFET devices, CMOS process integration, RAM applications

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CHOWDHURY R ET AL: "PULSED LASER DEPOSITION OF EPITAXIAL SI/TIN/SI7100) HETEROSTRUCTURES" APPLIED PHYSICS LETTERS,US,AMERICAN INSTITUTE OF PHYSICS. NEW YORK, vol. 64, no. 10, 7 March 1994 (1994-03-07), pages 1236-1238, XP000429868 ISSN: 0003-6951 cited in the application *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9905414B2 (en) 2000-09-28 2018-02-27 President And Fellows Of Harvard College Vapor deposition of metal oxides, silicates and phosphates, and silicon dioxide

Also Published As

Publication number Publication date
EP1166347A2 (en) 2002-01-02
JP2003523615A (en) 2003-08-05
WO2000051165A3 (en) 2001-05-31
DE20022706U1 (en) 2002-01-24
AU2685700A (en) 2000-09-14

Similar Documents

Publication Publication Date Title
US7023030B2 (en) Misfet
US9748145B1 (en) Semiconductor devices with varying threshold voltage and fabrication methods thereof
TWI234283B (en) Novel field effect transistor and method of fabrication
US5920088A (en) Vertical MISFET devices
US6657259B2 (en) Multiple-plane FinFET CMOS
US7687365B2 (en) CMOS structure for body ties in ultra-thin SOI (UTSOI) substrates
EP1760777A2 (en) Transistors and methods of manufacture thereof
US5914504A (en) DRAM applications using vertical MISFET devices
CN102468333B (en) Graphene device, and manufacturing method thereof
KR100333168B1 (en) Soi semiconductor device and method for manufacturing the same
US20060001111A1 (en) Semiconductor device
CN103779226A (en) Quasi-nano-wire transistor and manufacturing method thereof
US11469296B2 (en) Low leakage FET
CN108091698B (en) Field-effect transistor, method of manufacturing field-effect transistor, and electronic device
US20200066604A1 (en) Gate-last process for vertical transport field-effect transistor
CN110416303B (en) Semiconductor structure and method for forming integrated circuit structure
US6674099B1 (en) Misfet
CN107994078B (en) Field effect transistor with source control electrode, manufacturing method and electronic device
EP1166347A2 (en) Misfet with narrow bandgap source
KR20230109666A (en) Field Effect Transistor (FET) Device
CN110957365B (en) Semiconductor structure and semiconductor circuit
CN108054209B (en) Field-effect transistor, method of manufacturing field-effect transistor, and electronic device
CN108140673B (en) Tunneling field effect transistor and manufacturing method thereof
Poiroux et al. Ultra-thin body Silicon On Insulator and nanowire transistors for 22nm technology node and below
CN103779227A (en) Manufacturing method of fin type field effect transistor

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): AE AL AM AT AT AU AZ BA BB BG BR BY CA CH CN CR CU CZ CZ DE DE DK DK DM EE EE ES FI FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SK SL TJ TM TR TT TZ UA UG US UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): GH GM KE LS MW SD SL SZ TZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG

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

Kind code of ref document: A3

Designated state(s): AE AL AM AT AT AU AZ BA BB BG BR BY CA CH CN CR CU CZ CZ DE DE DK DK DM EE EE ES FI FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SK SL TJ TM TR TT TZ UA UG US UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A3

Designated state(s): GH GM KE LS MW SD SL SZ TZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG

ENP Entry into the national phase

Ref document number: 2000 601674

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2000905235

Country of ref document: EP

Ref document number: 09889815

Country of ref document: US

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

WWP Wipo information: published in national office

Ref document number: 2000905235

Country of ref document: EP