WO2012017227A1 - Semiconductor device - Google Patents

Semiconductor device Download PDF

Info

Publication number
WO2012017227A1
WO2012017227A1 PCT/GB2011/051413 GB2011051413W WO2012017227A1 WO 2012017227 A1 WO2012017227 A1 WO 2012017227A1 GB 2011051413 W GB2011051413 W GB 2011051413W WO 2012017227 A1 WO2012017227 A1 WO 2012017227A1
Authority
WO
WIPO (PCT)
Prior art keywords
region
semiconductor
conductivity type
injector
drift
Prior art date
Application number
PCT/GB2011/051413
Other languages
French (fr)
Inventor
Peter Ward
Philip Mawby
Martin Westmoreland
Original Assignee
The University Of Warwick
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 The University Of Warwick filed Critical The University Of Warwick
Publication of WO2012017227A1 publication Critical patent/WO2012017227A1/en

Links

Classifications

    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/76Unipolar devices, e.g. field effect transistors
    • H01L29/772Field effect transistors
    • H01L29/78Field effect transistors with field effect produced by an insulated gate
    • H01L29/7801DMOS transistors, i.e. MISFETs with a channel accommodating body or base region adjoining a drain drift region
    • H01L29/7802Vertical DMOS transistors, i.e. VDMOS transistors
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/70Bipolar devices
    • H01L29/72Transistor-type devices, i.e. able to continuously respond to applied control signals
    • H01L29/739Transistor-type devices, i.e. able to continuously respond to applied control signals controlled by field-effect, e.g. bipolar static induction transistors [BSIT]
    • H01L29/7393Insulated gate bipolar mode transistors, i.e. IGBT; IGT; COMFET
    • H01L29/7395Vertical transistors, e.g. vertical IGBT
    • H01L29/7398Vertical transistors, e.g. vertical IGBT with both emitter and collector contacts in the same substrate side
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/02Semiconductor bodies ; Multistep manufacturing processes therefor
    • H01L29/06Semiconductor bodies ; Multistep manufacturing processes therefor characterised by their shape; characterised by the shapes, relative sizes, or dispositions of the semiconductor regions ; characterised by the concentration or distribution of impurities within semiconductor regions
    • H01L29/0603Semiconductor 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 characterised by particular constructional design considerations, e.g. for preventing surface leakage, for controlling electric field concentration or for internal isolations regions
    • H01L29/0607Semiconductor 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 characterised by particular constructional design considerations, e.g. for preventing surface leakage, for controlling electric field concentration or for internal isolations regions for preventing surface leakage or controlling electric field concentration
    • H01L29/0611Semiconductor 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 characterised by particular constructional design considerations, e.g. for preventing surface leakage, for controlling electric field concentration or for internal isolations regions for preventing surface leakage or controlling electric field concentration for increasing or controlling the breakdown voltage of reverse biased devices
    • H01L29/0615Semiconductor 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 characterised by particular constructional design considerations, e.g. for preventing surface leakage, for controlling electric field concentration or for internal isolations regions for preventing surface leakage or controlling electric field concentration for increasing or controlling the breakdown voltage of reverse biased devices by the doping profile or the shape or the arrangement of the PN junction, or with supplementary regions, e.g. junction termination extension [JTE]
    • H01L29/063Reduced surface field [RESURF] pn-junction structures
    • H01L29/0634Multiple reduced surface field (multi-RESURF) structures, e.g. double RESURF, charge compensation, cool, superjunction (SJ), 3D-RESURF, composite buffer (CB) structures
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/66007Multistep manufacturing processes
    • H01L29/66075Multistep manufacturing processes of devices having semiconductor bodies comprising group 14 or group 13/15 materials
    • H01L29/66227Multistep manufacturing processes of devices having semiconductor bodies comprising group 14 or group 13/15 materials the devices being controllable only by the electric current supplied or the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched, e.g. three-terminal devices
    • H01L29/66234Bipolar junction transistors [BJT]
    • H01L29/66325Bipolar junction transistors [BJT] controlled by field-effect, e.g. insulated gate bipolar transistors [IGBT]
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/70Bipolar devices
    • H01L29/72Transistor-type devices, i.e. able to continuously respond to applied control signals
    • H01L29/739Transistor-type devices, i.e. able to continuously respond to applied control signals controlled by field-effect, e.g. bipolar static induction transistors [BSIT]
    • H01L29/7393Insulated gate bipolar mode transistors, i.e. IGBT; IGT; COMFET
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/76Unipolar devices, e.g. field effect transistors
    • H01L29/772Field effect transistors
    • H01L29/78Field effect transistors with field effect produced by an insulated gate
    • H01L29/7801DMOS transistors, i.e. MISFETs with a channel accommodating body or base region adjoining a drain drift region
    • H01L29/7802Vertical DMOS transistors, i.e. VDMOS transistors
    • H01L29/7803Vertical DMOS transistors, i.e. VDMOS transistors structurally associated with at least one other device
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/76Unipolar devices, e.g. field effect transistors
    • H01L29/772Field effect transistors
    • H01L29/78Field effect transistors with field effect produced by an insulated gate
    • H01L29/7801DMOS transistors, i.e. MISFETs with a channel accommodating body or base region adjoining a drain drift region
    • H01L29/7802Vertical DMOS transistors, i.e. VDMOS transistors
    • H01L29/7803Vertical DMOS transistors, i.e. VDMOS transistors structurally associated with at least one other device
    • H01L29/7806Vertical DMOS transistors, i.e. VDMOS transistors structurally associated with at least one other device the other device being a Schottky barrier diode
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/40Electrodes ; Multistep manufacturing processes therefor
    • H01L29/43Electrodes ; Multistep manufacturing processes therefor characterised by the materials of which they are formed
    • H01L29/47Schottky barrier electrodes
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/66007Multistep manufacturing processes
    • H01L29/66053Multistep manufacturing processes of devices having a semiconductor body comprising crystalline silicon carbide
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/66Types of semiconductor device ; Multistep manufacturing processes therefor
    • H01L29/68Types of semiconductor device ; Multistep manufacturing processes therefor controllable by only the electric current supplied, or only the electric potential applied, to an electrode which does not carry the current to be rectified, amplified or switched
    • H01L29/70Bipolar devices
    • H01L29/72Transistor-type devices, i.e. able to continuously respond to applied control signals
    • H01L29/739Transistor-type devices, i.e. able to continuously respond to applied control signals controlled by field-effect, e.g. bipolar static induction transistors [BSIT]
    • H01L29/7393Insulated gate bipolar mode transistors, i.e. IGBT; IGT; COMFET
    • H01L29/7395Vertical transistors, e.g. vertical IGBT

Definitions

  • the present invention relates to a semiconductor device and more specifically to a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor (MOSFET) and a vertical or lateral insulated gate bipolar transistor (IGBT).
  • MOSFET vertical metal oxide semiconductor field effect transistor
  • IGBT vertical or lateral insulated gate bipolar transistor
  • the present invention also relates to a semiconductor device having an integrated vertical and lateral charge balance structure.
  • MOSFET operates based on the flow of only majority carriers through the drift region.
  • An IGBT has a similar structure to a vertical MOSFET, but differs in that the drain is of the opposite conductivity type. Thus, both majority and minority carriers are injected into the drift region. This has the effect of reducing the ON resistance of the IGBT compared to the MOSFET, but also has the effect of increasing the gate switching speed.
  • a form of vertical MOSFET device has been proposed in which a MOSFET is provided with a minority carrier injector in the form of a p-type region which forms a p-n junction with the n-type drift region so as to increase switching speed and reduce the ON resistance.
  • a novel trench-injector power device with low ON resistance and high switching speed David K. Y. Liu and James D. Plummer, IEEE Electron Device Letters, volume 9, Issue 7, pages 321 - 323 (1988) and "Design and analysis of a new conductivity-modulated power MOSFET", David K. Y. Liu and James D. Plummer, IEEE Transactions on Electron Devices, volume 40, Issue 2, pages 428 - 438 (1993).
  • the device has a drawback of needing to control separately an additional terminal.
  • this can be avoided by connecting the injector to the drain.
  • this arrangement presents a problem, namely that when the same bias is applied to the injector and drain, no voltage difference develops across the p-n junction and so no minority carrier injection occurs.
  • a small voltage difference initially develops due to differing resistances between the source and the drain and the source and the injector.
  • the difference in resistance is soon compensated by low-level minority carrier injection and so the effect stops.
  • the number of injected minority carriers is low.
  • conductivity can be modulated using minority carriers, the device falls well short of achieving full IGBT-like bipolar action.
  • the present invention seeks to provide an improved semiconductor device.
  • semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor having a Schottky- biased p-n junction.
  • the Schottky biasing helps to ensure that the p-n junction is forward biased and continues to inject minority carriers. Therefore, the device provides an integrated vertical metal oxide semiconductor field effect transistor and vertical or lateral insulated gate bipolar transistor.
  • the semiconductor device may comprise a minority-carrier injector which comprises the p-n junction and a Schottky diode, wherein the Schottky diode is arranged in parallel with the p-n junction so as to provide the Schottky-biased p-n junction.
  • the semiconductor device may comprise a semiconductor region (for example, an epitaxial region) disposed between first and second opposite surfaces, at least a first portion of the semiconductor region comprising a semiconductor of a first conductivity type providing a drift region for majority carriers in a direction between the first and second surfaces, at least a second portion of the semiconductor region (for example, an epitaxial region) disposed between first and second opposite surfaces, at least a first portion of the semiconductor region comprising a semiconductor of a first conductivity type providing a drift region for majority carriers in a direction between the first and second surfaces, at least a second portion of the
  • semiconductor region providing a body region comprising a semiconductor of a second, opposite conductivity type disposed at the first surface of the
  • the semiconductor region may comprise the same semiconductor, such as silicon or silicon carbide, but have regions of different conductivity types, i.e. n-type and p- type regions.
  • the minority-carrier injector may be disposed at the first surface (e.g. top surface) of the semiconductor region laterally separated from the body region. Thus, a lateral insulated gate bipolar transistor can be formed. However, the minority- carrier injector may be disposed at the second surface (e.g. bottom surface) of the semiconductor region so as to form a vertical insulated gate bipolar transistor.
  • At least a fourth portion of the semiconductor region may comprise a
  • the Schottky diode may comprise a metallization in contact with the first surface of semiconductor region.
  • the Schottky diode may have a barrier height equal to or greater than 0.7 eV. If the minority-carrier injector is disposed at the second surface, then the metallization may be in contact with the first surface of semiconductor region. A different metallization and/ or process may be used to form a drain contact to the semiconductor region.
  • the semiconductor device may further comprise a gate structure disposed on the first surface of the semiconductor region and configured to control flow of carriers from the source region into drift region.
  • the semiconductor device may further comprise a drain region disposed on the second surface of the semiconductor region.
  • the drain region may include a semiconductor substrate, e.g. highly-doped substrate such as a wafer.
  • the drain region may include a metallization, which preferably forms an ohmic contact.
  • a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor, the device comprising a semiconductor substrate drain of a first conductivity type, a drift region or epitaxial region of the first conductivity type disposed on the semiconductor substrate drain, a body region of a second, opposite conductivity type disposed within the drift region or epitaxial region, a source region of the first conductivity type disposed within the body region.
  • the device further comprises a minority-carrier injector which comprises an injector region of the second conductivity type disposed within the drift region or epitaxial region and laterally separated from the body region, the injector region and the drift region or epitaxial region forming a p-n junction, and a metallic electrode in contact with the drift region or epitaxial region laterally adjacent to and electrically connected to the injector region, the metallic electrode and the drift region or epitaxial region forming a Schottky diode in parallel with the p-n junction.
  • a minority-carrier injector which comprises an injector region of the second conductivity type disposed within the drift region or epitaxial region and laterally separated from the body region, the injector region and the drift region or epitaxial region forming a p-n junction, and a metallic electrode in contact with the drift region or epitaxial region laterally adjacent to and electrically connected to the injector region, the metallic electrode and the drift region or epitaxial region forming a Schottky diode in parallel with the p
  • the Schottky diode helps to ensure that the p-n junction is forward biased and continues to inject minority carriers into the drift region or epitaxial region.
  • the device provides an integrated vertical metal oxide semiconductor field effect transistor and lateral insulated gate bipolar transistor.
  • the first and second conductivity types may be n-type and p-type respectively.
  • the body region and the substrate may be vertically separated by a first given distance and the body region and the injector region are laterally separated by a second given distance exceeding equal to or greater than the first given distance.
  • the second given distance may be about 6 to 12 ⁇ .
  • the device may further comprise a depletion reducing region comprising a semiconductor of the first conductivity type disposed within the drift region, more highly doped than the drift region and disposed between the body region and the injector region. This can help reduce the lateral size of the device.
  • the body region and the substrate may be vertically separated by a first given distance and the body region and the injector region are laterally separated by a second given distance less than the first given distance.
  • the Schottky diode may have a barrier height equal to or greater than 0.7 eV.
  • the metallic electrode may include a layer of platinum silicide forming the Schottky diode.
  • the metallic electrode and the semiconductor substrate may be coupled so as to be at substantially the same potential.
  • the device may comprise a further, different metallic electrode in contact with the source region.
  • the drift region or epitaxial region may comprise silicon.
  • the drift region or epitaxial region may comprise silicon carbide semiconductor.
  • the drift region, the body region, the source region and the injector region may comprise the same semiconductor.
  • the device may further comprise a charge balance structure.
  • the charge balance structure may comprise first and second charge balance regions comprising a semiconductor of the second conductivity type, the first charge balance region extending vertically between the substrate (or bottom surface of a semiconductor region which would have been formed on a substrate) and the source region and the second charge balance region laterally spaced between the p-n junction and the first charge balance region, such that a vertical drift region comprising a semiconductor of the first conductivity type is disposed between the first and second charge balance regions.
  • the charge balance structure may further comprise a third charge balance region comprising a semiconductor of the second conductivity type disposed under the p-n region and extending laterally to the second charge balance region, wherein a lateral drift region comprising a
  • the charge balance structure may further comprise a region comprising a semiconductor of the second conductivity type linking the first and second charge balance regions, the region configured to leave a region comprising a semiconductor of the first conductivity type linking the lateral and vertical drift regions.
  • semiconductor device having an integrated vertical and lateral charge balance structure.
  • the semiconductor device may comprise a planar substrate, a first conductive region of a first conductivity type extending perpendicularly to the substrate between first and second ends and having at least first and second sides, first and second charge balance regions of second conductivity type disposed on the first and second sides respectively of the first conductive region, a second conductive region of the first conductivity type extending parallel to the substrate between first and second ends and having at least a first side, a third charge balance region of the second conductivity type disposed on the first side of the second conductive region, and the device may further comprise at least one region of the first conductivity type linking the first and second conductive regions so as to form a continuous region of the first conductivity type between the first end of the first conductive region to the second end of the second conductive region and at least one region of the second conductivity type linking the first, second and third charge balance regions so as to form an integrated charge balance structure.
  • an integrated circuit comprising at least one semiconductor device.
  • a method of fabricating a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor comprising providing the insulated gate bipolar transistor with a Schottky- biased p-n junction.
  • the method may comprise providing a minority-carrier injector which comprises the p-n junction and a Schottky diode, wherein the Schottky diode is arranged in parallel with the p-n junction so as to provide the Schottky-biased p-n junction.
  • the method may comprise providing a semiconductor region, such as an epitaxial layer of semiconductor material(s), between first and second opposite surfaces (or top and bottom surfaces or faces).
  • At least a third portion of the semiconductor region may provide a source region comprising a semiconductor of the first conductivity type disposed at the first surface of semiconductor region and within the body region.
  • the semiconductor region may comprise the same semiconductor, such as silicon or silicon carbide.
  • the minority-carrier injector may be disposed at the first surface of the semiconductor region laterally separated from the body region so as to form a lateral insulated gate bipolar transistor or disposed at the second surface of the semiconductor region vertically separated from body region so as to form a vertical insulated gate bipolar transistor.
  • At least a fourth portion of the semiconductor region may comprise a
  • the Schottky diode may comprise a metallization in contact with the first surface of semiconductor region.
  • the method may further comprise providing a gate structure on the first surface of the semiconductor region.
  • the method may further comprise providing a drain region on the second surface of the semiconductor region. This may comprise growing the semiconductor region on a semiconductor substrate. Additionally or alternatively, the drain region or contact may include a metallization.
  • the method may comprise providing a semiconductor substrate drain of a first conductivity type, a drift region of the first conductivity type disposed on the semiconductor substrate drain, a body region of a second, opposite conductivity type, for example formed in or adjacent to the drift region, and a source region of the first conductivity type disposed within the body region.
  • the method may comprise providing a drift region of first conductivity type on a semiconductor substrate of the first conductivity type, forming a body region of a second, opposite conductivity type within the drift region, forming a source region of the first conductivity type within the body region, and providing the Schottky- biased p-n junction may comprise forming an injector region of the second conductivity type within the drift region, laterally separated from the body region, wherein the injector region and the drift region provide a p-n junction and providing a metallic electrode in contact with the drift region laterally adjacent and electrically connected to the injector region, wherein the metallic electrode and the drain region provide a Schottky diode in parallel with the p-n junction.
  • a method of fabricating a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor comprising providing a drift region of first conductivity type on a semiconductor substrate of the first conductivity type, forming a body region of a second, opposite conductivity type within the drift region, forming a source region of the first conductivity type within the body region, forming an injector region of the second conductivity type within the drift region, laterally separated from the body region, wherein the injector region and the drift region provide a p-n junction; and providing a metallic electrode in contact with the drift region laterally adjacent and electrically connected to the injector region, wherein the metallic electrode and the drain region provide a Schottky diode in parallel with the p-n junction.
  • the body region and the substrate may be vertically separated by a first given distance and wherein forming the injector region within the drift region comprises laterally separating the injector region from the body region by a second
  • the device may further comprise a depletion reducing region comprising a semiconductor of the first conductivity type disposed within the drift region, more highly doped than the drift region and disposed between the body region and the injector region.
  • the body region and the substrate may be vertically separated by a first given distance and the body region and the injector region are laterally separated by a second given distance less than the first given distance.
  • the method may further comprise providing a charge balance structure.
  • a seventh aspect of the present invention there is provided a method of fabricating a semiconductor device, the method comprising providing an integrated vertical and lateral charge balance structure.
  • Providing the integrated vertical and lateral charge balance structure may comprise providing a planar substrate, providing a first conductive region of a first conductivity type extending perpendicularly to the substrate between first and second ends and having at least first and second sides, providing first and second charge balance regions of second conductivity type disposed on the first and second sides respectively of the first conductive region, providing a second conductive region of the first conductivity type extending parallel to the substrate between first and second ends and having at least a first side, providing a third charge balance region of the second conductivity type disposed on the first side of the second conductive region, providing at least one region of the first conductivity type linking the first and second conductive regions so as to form a continuous region of the first conductivity type between the first end of the first conductive region to the second end of the second conductive region, and providing at least one region of the second conductivity type linking the first, second and third charge balance regions so as to form an integrated charge balance structure.
  • the metal oxide semiconductor field effect transistor and insulated gate bipolar transistor sharing a source region (or source/ emitter terminal) and a drain region (or drain terminal) and wherein the Schottky-biased p-n junction has a terminal (i.e. an injector and/or collector terminal), the method comprising grounding the source region and applying given bias(es) to the injector/collector terminal and to the drain region.
  • the same bias is applied to the injector/collector terminal and the drain region.
  • Figure la is more detailed view of the semiconductor device shown in Figure 1 ;
  • Figure 2 illustrates a set of mask layers used to fabricate the semiconductor device shown in Figure 1 ;
  • Figure 3a is a schematic circuit diagram of the semiconductor device shown in Figure 1 before the onset of minority carrier injection;
  • Figure 4 is a vertical section of the semiconductor device shown in Figure 1 illustrating operation
  • Figure 5 schematically illustrates the behaviour of the semiconductor device shown in Figure 1 ;
  • FIGS 6a to 6p are vertical sections through the semiconductor device shown in Figure 1 at different stages during fabrication
  • Figure 8 is a vertical section of another embodiment of a semiconductor device in accordance with the present invention.
  • Figure 9 is a vertical section of yet another embodiment of a semiconductor device in accordance with the present invention.
  • Figure 10 is a vertical section of still another embodiment of a semiconductor device in accordance with the present invention.
  • Figure 11 is a perspective view of a further embodiment of a semiconductor device in accordance with the present invention.
  • Figure 12 is a plan view of the semiconductor device shown in Figure 11 ;
  • Figure 13 is a vertical section of the semiconductor device shown in Figure 12 taken along the line B-B';
  • Figure 14 is a vertical section of the semiconductor device shown in Figure 12 taken along the line C-C;
  • Figure 15 is a vertical section of the semiconductor device shown in Figure 12 taken along the line D-D';
  • Figure 16 is a vertical section of the semiconductor device shown in Figure 13 taken along the line E-E';
  • Figure 17a is a schematic circuit diagram of the semiconductor device shown in Figure 11 before the onset of minority carrier injection
  • Figure 17b is a schematic circuit diagram of the semiconductor device shown in Figure 11 after the onset of minority carrier injection
  • Figure 18 schematically illustrates the behaviour of the semiconductor device shown in Figure 11 ;
  • Figure 19a to 19d are vertical sections through the semiconductor device shown in Figure 11 at different stages during fabrication of a charge balance structure; and Figure 20 is a process flow diagram of a method of fabricating a charge balance structure.
  • the transistor 1 includes a heavily-doped n-type monocrystalline silicon substrate 2 which functions as a drain region.
  • An epitaxial layer 3 of lightly-doped n-type monocrystalline silicon (herein also referred to as the "drift region") is arranged on an upper surface 4 of the substrate 2.
  • a field oxide 5 is located at an upper surface of the epitaxial layer 3 and has first and second windows 6 l5 6 2 defining first and second laterally-separated upper surfaces 7 l5 7 2 of the epitaxial layer 3.
  • the first and second windows 6 l5 6 2 have lengths l l5 ⁇ 2 of about 6 ⁇ and 4.8 ⁇ respectively and are separated by a length, 1 3 , of about 7 ⁇ .
  • a gate oxide 8 is disposed within the first window 6 1 on the upper surface 7 1 of the epitaxial layer 3.
  • the gate oxide 8 runs along the upper surface 7 l of the epitaxial layer 3 and abuts the field oxide 5 thereby forming a step 9.
  • a layer of heavily- doped n-type polycrystalline silicon 10 (which may also be referred to as the "gate poly") is disposed on the gate oxide 8 and runs over the step 9 onto the field oxide 5. Silicon dioxide spacers 11 are formed on the sides of the gate poly 10.
  • First and second p-type injectors 13 l5 13 2 in the form of heavily-doped p-type diffusion wells are located within the epitaxial layer 3 at the second upper surface 7 2 .
  • the p-type injectors 13 l 5 13 2 are spaced apart to leave a region 14 of the n-type epitaxial layer 3 extending up to the second upper surface 7 2 of the epitaxial layer 3.
  • An n-type source region 15 in the form of a heavily-doped n-type diffusion well is disposed within the p-type body 12 at the upper surface 7 l .
  • a p-type body short (not shown) in the form of a heavily-doped p-type diffusion well is also provided at the first upper surface using a body short mask ( Figure 2).
  • the metallization layers 19 l 5 19 2 , 19 3 each comprise a bi-layer comprising a high-barrier metal silicide base layer comprising, for example, platinum silicide (PtSi), and a high-conductivity overlayer comprising, for example, aluminium (Al).
  • a high-barrier metal silicide base layer comprising, for example, platinum silicide (PtSi)
  • a high-conductivity overlayer comprising, for example, aluminium (Al).
  • the third metallisation layer 19 3 is in contact with the first and second p-type injectors 13 l5 13 2 and the n-type region 14 between the p-type injectors 13 l 5 13 2 .
  • the p-type injectors 13 l 5 13 2 form respective p-n junctions 20 l 5 20 2 with the epitaxial layer 3 and the metallisation layer 19 3 forms a Schottky diode 21 with the epitaxial layer 3.
  • the pair of p-n junctions 20 l 5 20 2 and the Schottky diode 21 are arranged in parallel between the metallisation layer 19 3 and the epitaxial layer 3.
  • the high-barrier metal silicide forms a Schottky diode with the epitaxial layer 3, but forms ohmic contacts to the heavily-doped injector and source regions 13 l 5 13 2 , 15.
  • an inner edge 22 of the p-type body 12 and an inner edge 23 of the p-type injector 13 a which is closest to the p-type body 12 are separated by a length, L.
  • a bottom 24 of the p-type body 12 is separated from the highly-doped substrate 2 by a depth,
  • the length, L approximately equals the depth, d l 5 i.e. L ⁇ d, to maintain the breakdown voltage of the device.
  • the separation length, L, and the depth, d are about 8 ⁇ .
  • the length, L can be smaller if a highly-doped field stop region or a charge balance structure is used.
  • a source-drain depletion region 25 having an edge 26 is formed in the epitaxial layer 3 due to the junction between the p-type body 12 and the n-type epitaxial layer 3.
  • the transistor 1 is arranged so that, in blocking mode, the edge 26 of the depletion region 25 does not reach the inner edge 23 of the first injector 13 1 .
  • the transistor 1 includes a MOSFET 28.
  • the drain of the MOSFET 28 is connected to the drain D of the transistor 1 through a first, vertical resistive element 29
  • the drain of the MOSFET 28 is also connected to a local bias point 30 through a second, lateral resistive element 29 2 .
  • the first and second resistive elements 29 l 5 29 2 are joined at node 31 at which electron current branches when flowing towards the drain D and the injector I.
  • the first and second resistive elements 29 l5 29 2 have values R j and R 2 respectively. In this example, Rj and R 2 depend on the depth, d l5 of the epitaxial region 3 and the lateral separation of the p-body region 12 and the Schottky diode 21 respectively. Referring in particular to Figure 3b, after the onset of minority carrier injection, minority carriers (in this example, holes) flow from the injector I through a bipolar transistor 32 to source S.
  • Figure 3b shows the flow of majority carriers and minority carriers through the transistor 1 which, in this example, are electrons and holes respectively.
  • a parasitic bipolar transistor 33 may affect operation of the device.
  • the transistor 1 is arranged so that the source S is grounded and that the drain D and injector I are connected together to a positive voltage V d .
  • the gate G is used to control the transistor 1 and if a voltage V g is applied to the gate G and the voltage exceeds a threshold voltage V th , then current flows through the transistor 1 from the source region 15 into the epitaxial region 3.
  • the transistor 1 can provide an integrated vertical metal oxide semiconductor field effect transistor and lateral insulated gate bipolar transistor.
  • the transistor 1 operates in a similar way to a conventional vertical MOSFET, as shown in Figure 3a.
  • the potential barriers of the p-n junctions 20 l5 20 2 limit or prevent the flow of majority carriers (i.e. electrons) to injector I.
  • V d increases, electron charge builds up in a region 34 beneath the p-type injectors 13 l5 13 2 .
  • charge builds up a depletion region 35 associated with the Schottky diode 21 is reduced.
  • the depletion region 35 of the Schottky diode 21 has a capacitance C dep(Sch) which varies with the thickness of the depletion region 35. As shown in Figure 3a, the capacitance is shown in series with the Schottky diode 21 which has a resistance, R Sc hottky
  • the p-n junctions 20 l5 20 2 also have respective depletion regions 36 l 5 36 2 , each having a capacitance C dep(p _ n) . As V d increases further, the depletion region 35 becomes so small that thermionic emission of electrons through the Schottky diode 21 occurs.
  • the Schottky diode 21 helps to dynamically bias the p-n junctions 20 l5 20 2 and ensure that the p-n junctions 20 l5 20 2 continue to inject minority carriers into the base region provided by the epitaxial region 3 at higher V d unlike, for example, an ohmic resistor. If an ohmic resistor is used instead of the Schottky diode, then the p-n junctions 20 l5 20 2 would initially be forward biased. However, once this occurs, the resistivity of the epitaxial region would drop, allowing the bias in region 34 to fall below that required to forward bias the p-n junctions 20 l5 20 2 resulting in termination of injection of minority carriers.
  • the transistor 1 has a breakdown voltage of about 120 V.
  • An array (not shown) of transistors 1 and bond pads (not shown) can be implemented in a chip having an area of about 3.3 mm 2 and can support 10 A of current comprising majority carriers and 70 A of current comprising majority and minority carriers.
  • the transistor 1 has a breakdown voltage of about 1000 V.
  • the transistor 1 has a further advantage that if the device heats up (e.g. to 400 K or above) due to a surge current, then this will assist the onset of minority carrier injection, thereby reducing the ON state resistance of the device and, thus lowering, power loss.
  • the device 1 is fabricated using a succession of masks including an active area mask for defining the extent of the field oxide 5, a gate poly mask for defining the extent of the gate poly 10, a p-body mask for defining the p-body well 12, an n-source mask for defining the n-source well 15, a body short mask for defining a body short (not shown), a contact mask for defining openings in the silicon dioxide layer 17 and a metal mask for defining metallization 19 l 5 19 2 , 19 3 .
  • monocrystalline silicon is grown epitaxially on a highly-doped n-type
  • the epitaxial layer 3' is doped in-situ during growth with arsenic or antimony to a concentration of about 5 Xl O 15 cirf 3 and has a thickness of about 8 ⁇ .
  • a highly-doped n-type field stop region 37 ( Figure 8) can be formed (step S2).
  • a field oxide 5 is formed at the surface 7' (Figure 6a) of the epitaxial layer 3' ( Figure 6a) by thermal oxidation using a LOCOS process (step S3).
  • the field oxide 5 has thickness of about 1 ⁇ .
  • the field oxide 5 has first and second windows 6 l5 6 2 defining first and second laterally-separated upper surfaces 7 l5 7 2 of the epitaxial layer 3.
  • a layer 8' of silicon dioxide is grown by thermal oxidation on the upper surfaces 7 l , 7 2 of the epitaxial layer 3 (step S4).
  • the silicon dioxide layer 8' has a thickness of about 1,500 A.
  • a layer 10' of heavily-doped n-type is grown by thermal oxidation on the upper surfaces 7 l , 7 2 of the epitaxial layer 3 (step S4).
  • polycrystalline silicon is grown by chemical vapour deposition (CVD) over the silicon dioxide layer 8' and field oxide 5 (step S5).
  • the polycrystalline layer 10' is doped in-situ during growth with phosphorous or arsenic to a concentration equal to or greater than about 1 X 10 20 cirf 3 and has a thickness of about 3,000 A.
  • an undoped polycrystalline layer 10' can be grown and doped by ion implantation after growth.
  • a layer (not shown) of photoresist is applied over the polycrystalline layer 10' and is patterned to form a mask (not shown).
  • the mask pattern is transferred to the underlying polycrystalline layer 10' and silicon dioxide layer 8' by reactive ion etching (RIE) using suitable feed gases (step S6).
  • RIE reactive ion etching
  • the resulting structure includes the gate poly 10 and gate oxide 8.
  • a further layer (not shown) of photoresist is applied and patterned so as to protect the epitaxial layer 3 in the second window 6 2 .
  • Boron ions are implanted at an energy between about 20 to 160 keV.
  • the masks (not shown) are removed and the structure is annealed to activate the implant leaving a p-type well 12 doped to a concentration of between about 1 Xl O 17 cirf 3 and about 5 Xl O 18 cm -3 and having a depth of about 0.7 to 1.2 ⁇ (step S7).
  • a conformal layer (not shown) of silicon dioxide is deposited and globally etched back by RIE using suitable feed gases to form a spacer 11 around the sides of the gate poly 10 (step S8) .
  • a layer (not shown) of photoresist is applied and patterned so as to protect the epitaxial layer 3 in the second window 6 2 .
  • Arsenic ions are implanted into the epitaxial layer 3 in the first window 6 1 at an energy of about 40 keV.
  • the mask (not shown) is removed and the structure is annealed to activate the implant leaving an n-type well 15 doped to a concentration of between about 1 Xl O 20 cirf 3 and about 3 Xl O 20 cm -3 and having a depth of about 1 ,000 A to 5,000 A (step S9).
  • a layer (not shown) of photoresist is applied and patterned to open a small hole over the first window 6 1 and two narrow windows (not shown) over the second window 6 2 .
  • Boron ions are implanted into the epitaxial layer 3 in the second window 6 2 at an energy of about 40 keV.
  • the mask (not shown) is removed and the structure is annealed to activate the implant leaving an p-type body short ( Figure 2) and p-type injectors 13 l5 13 2 doped to a concentration of between about 1 XlO 20 cm “3 and about 3 XlO 20 cm “3 and having a depth of about 1,000 A to 5,000 A (step S10).
  • step Sl l another layer 17' of silicon dioxide having a thickness of about 4,000 A is deposited (step Sl l).
  • a layer (not shown) of photoresist is applied over the silicon dioxide layer 17' and is patterned to form a mask (not shown).
  • the mask pattern is transferred to the underlying silicon dioxide layer 17' by RIE using suitable feed gases (step SI 2).
  • the resulting structure includes a patterned silicon dioxide layer 17 with contact windows 18 l5 18 2 , 18 3 .
  • a layer (not shown) of photoresist is applied over the metallization 19' and is patterned to form a mask (not shown).
  • the mask pattern is transferred to the underlying metallization 19' by RIE using suitable feed gases (step S14).
  • the resulting structure comprises metallization contacts 19 l3 19 2 , 19 3 .
  • layer thicknesses and process parameters can be varied and optimised and that the fabrication process can include other steps, such as substrate thinning and depositing a metallization on the back side of the substrate.
  • FIG 8 another embodiment of a power semiconductor device 1 ' in accordance with the present invention is shown.
  • the device 1 ' is shown in simplified form, for example, as can be used in numerical simulations.
  • the device 1 ' is substantially the same as the device 1 ( Figure 1) hereinbefore described, but differs in that it includes a field stop region 37 and has different dimensions.
  • the field stop region 37 takes the form of a shallow n-type diffusion well disposed at the upper surface of the epitaxial layer 3 under the field oxide 10 between p-type body region 12 and the p-type injector 13 1 .
  • the field stop region helps to restrict the depletion layer width laterally and improve performance of the bipolar transistor 32 ( Figure 3b). This allows the lateral dimension of the device 1 ' to be reduced, in particular, a shorter separation length, L, of about 4 ⁇ .
  • the depth of the device, d 2 can have a value in a range of about 8 ⁇ to 10 ⁇ , although the depth can be higher or lower depending on required operating voltage.
  • the half-cell width of the device, w 2 can be about 7 ⁇ to 11 ⁇ . Again, the width, w 2 , can be larger or smaller depending on operating voltage.
  • FIG. 9 yet another embodiment of a power semiconductor device 1 " in accordance with the present invention is shown.
  • the device 1 " is substantially the same as the device 1 ' hereinbefore described, but differs in that the position of the first injector 13 a is moved away from the -type body region 12 and does not extend to the edge of the oxide 10, 17.
  • the second injector 13 2 may be omitted.
  • the positions of the p-n junction 2Q i and the Schottky diode 21 are swapped.
  • the device 1 " is fabricated using a modified mask for the p-type implant at step S10 above. Operation the device 1 " is substantially the same as device 1 '. In particular, there is no detrimental effect on blocking voltage. However, the onset of injection occurs at a slightly higher source-drain voltage. For example, for the device 1 ' shown in Figure 8, simulation shows that hole injection starts at voltage of about 1.7 V at a temperature of 300 K and, for the device 1 " shown in Figure 9, simulation shows that hole injection starts at voltage of about 1.9 V at a temperature of 300 K. This is thought to be due to the slightly increased distance from the channel to the injector Referring to Figure 10, another embodiment of a power semiconductor device 1 "' in accordance with the present invention is shown.
  • the Schottky diode 20' comprises a pad 39 directly in contact with the epitaxial layer 3 adjacent to or between p-type well(s) 13/, 13 2 '.
  • the pad 39 is formed by a layer of a first metallization, such as platinum silicide or other metallization providing a barrier height of at least 0.7 eV.
  • the drain contact comprises a layer 40 of a metallization in contact with the epitaxial layer 3 which forms an ohmic contact to the epitaxial layer.
  • the metallization is preferably alloyed or annealed.
  • the metallization may comprise aluminium or aluminium silicide.
  • the layer 40 may run over the pad 39.
  • the layers 39, 40 may not join or overlap, but may be connected in some other way, e.g. by a bond wire or high-temperature solder.
  • the device 1 "' having a vertical injector operates in a similar way to the devices having a lateral injector. However, one difference is that the injector uses vertical biasing. The biasing depends on the amount of electron charge in the area immediately above the Schottky diode 21 '.
  • the blocking voltage performance is set by the drift region thickness, d 3 , from the bottom of the p-body region 12 to the drain.
  • d 3 drift region thickness
  • Neighbouring p-body regions 12 provide mutual protection to the radius of curvature, thus minimising the electric field in that area.
  • the device 1 "' has the advantage that, due to the position of the injector, no separate bond wire or connection is required from front to back face. It may also allow a heat sink (not shown) to be placed more easily on top of the source metallization to help dissipate heat.
  • the placement of the injector can allow a smaller cell width to be used and so reduce the drain-source specific resistance per unit area in a pre-injection state and in an injection state compared to the embodiments using a lateral injector.
  • injector elements can be evenly distributed across the underside of the wafer and so help to ensure an even current share across the resultant device area once injection begins.
  • the use of electrons available from two voltage-controlled, gated MOS channels per injector, instead of a single channel per injector, as in transistor 1 ( Figure 1) allows for an increased current saturation level.
  • a Schottky-biased injector can be used to reduce the ON resistance of a vertical MOSFET.
  • the blocking voltage of the hybrid device depends on the thickness and doping concentration of the drift region 3, as well as the radius of curvature of the p-type body 12.
  • One way to increase the blocking voltage in a conventional power MOSFET is to increase the doping density of the n-type drift region and include charge
  • a modified form of charge compensation structure can be used not only to allow the flow of majority carriers to the injector, but also to direct their flow.
  • the modified charge compensation structure can be adapted to control the voltage at which IGBT-like action occurs.
  • FIGs 11 to 16 a further embodiment of a power semiconductor device 101 in accordance with the present invention is shown.
  • the semiconductor device 101 takes the form of a hybrid silicon n-channel vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor device comprising a combined vertical and lateral charge balance structure 131.
  • the device 101 is also hereinreferred to as a "super junction device".
  • the super junction device 101 includes a heavily-doped n-type monocrystalline silicon substrate 102 which functions as a drain region.
  • An epitaxial layer 103 of silicon is arranged on an upper surface 104 of the substrate 102.
  • the epitaxial layer 103 comprises p-type and n-type regions which provide the charge balance structure 131.
  • the structure of super junction device 101 at the upper surface 107 of the epitaxial layer 103 is generally the same as that of the device 1 shown in Figures 1 and l a.
  • a field oxide 105 l 5 05 2 is located at the upper surface 107 of the epitaxial layer 103 and has first, second and third windows (not shown) defining first, second and third laterally-separated portions of upper surfaces 107 of the epitaxial layer 103.
  • a first gate oxide 108 t is provided on the first portion of the upper surface 107 of the epitaxial layer 103.
  • the first gate oxide 108 t runs along the first portion of the upper surface 107 of the epitaxial layer 103 and abuts the first field oxide 105 1 .
  • a first layer of heavily-doped n-type polycrystalline silicon 1 10 a is disposed on the first gate oxide 108 1 .
  • a first p-type body 1 12 t comprising a lightly-doped p-type diffusion well is disposed within the epitaxial layer 103 at the first portion of the upper surface 107.
  • the first p-type body 1 12 t extends laterally under a first spacer and the first gate oxide 108 ! .
  • First and second p-type injectors 1 13 l 5 1 3 2 in the form of heavily-doped p-type diffusion wells are located within the epitaxial layer 103 at the second portion of the upper surface 107.
  • the p-type injectors 1 13 l 5 1 3 2 are spaced apart to leave a region 1 14 of n-type epitaxial layer 103 extending up to the second portion of the upper surface 107 of the epitaxial layer 103.
  • the injectors 1 13 l 5 1 13 2 straddle the centre of the cell.
  • a first n-type source region 1 15 l in the form of a heavily-doped n-type diffusion well is disposed within the first p-type body 1 12 l at the upper surface 107.
  • a p-type body short (not shown) in the form of a heavily-doped p-type diffusion well is also provided at the first upper surface.
  • a layer (not shown) of silicon dioxide runs over the first gate poly 10 and the first field oxide 105 and has windows (not shown).
  • a first metallisation layer 119 provides a first source/emitter terminal SI
  • a second metallisation layer (not shown) provides a first gate terminal Gl
  • a third metallisation layer 1 9 3 provides a collector/injector terminal I.
  • the metallization layers 9,, 119 3 each comprise a bi- layer comprising a high-barrier metal silicide base layer comprising, for example, platinum silicide, and a high-conductivity overlayer comprising, for example, aluminium. Similar to the device 1 shown in Figures 1 and la, the third metallisation layer 1 9 3 is in contact with the first and second p-type injectors 3,, 1 3 2 and the n-type region 114 between the p-type injectors 3,, 1 3 2 .
  • the p-type injectors 3,, 113 2 form respective p-n junctions 145,, 145 2 ( Figures 17a and 17b) with the epitaxial layer 103 and the metallisation layer 1 9 3 forms a Schottky diode 146 ( Figures 17a and 17b) with the epitaxial layer 103.
  • the pair of p-n junctions (not shown) and the Schottky diode (not shown) are arranged in parallel between the
  • the high-barrier metal silicide forms a Schottky diode with the epitaxial layer 103, but forms ohmic contacts to the heavily-doped injector and source regions 3,, 113 2 , 115,.
  • the second half of the cell is a mirror image of the first half and includes a second field oxide 105 2 , a second gate oxide 108 2 , a second p-type body 2 2 , a second source region 115 2 , a second source/emitter terminal S2 and a second gate terminal G2.
  • the first and second injectors 3,, 1 3 2 straddle the centre of the cell and injector I is shared by both halves of the cell.
  • first and second, outer p-type regions 132j, 132 2 are disposed under the p-type body regions 112 l5 112 2 .
  • the p-type regions 132 l5 132 2 extend from the substrate 102 and reach the bottoms of the respective p-type body regions 112 l5 112 2 .
  • Third and fourth inner p-type regions 134 l5 134 2 are disposed between the outer p- type regions 132 l5 132 2 .
  • a third n-type region 135 is formed between the third and fourth p-type regions 134 j , 134 2 . This forms a voltage sustaining region.
  • the p-type regions 132 l5 132 2 , 134 l5 134 2 are interconnected by lateral, bridge-like, p-type linking regions 136 l5 36 2 and by a buried p-type cross layer 137 (which collectively may be referred to as "p-type ground links" as the p-type body regions 112 l5 112 2 are connected to ground).
  • the buried p-type cross layer 137 forms part of a second, lateral charge balance structure
  • the third and fourth p-type regions 134 l 5 134 2 and the p-type cross layer 137 form a core structure generally having the appearance of an inverted parallel flange channel, i.e. an elongate structure which is 'n'-shaped in cross section.
  • the outer p-type walls 132 l5 132 2 flank the core structure 134 l 5 134 2 , 137 and are joined to the top of the core structure at intervals along its length by the linking regions 136 l5 36 2 which extend laterally from the p- type cross layer 137 and so form a unitary p-type region, i.e. a single continuous region of p-type semiconductor.
  • the column-like, n-type linking regions 141 l5 141 2 pass between the p-type linking regions 136 l5 36 2 such that the p-type and n-type unitary regions interlock or interlace.
  • majority carriers i.e. electrons
  • Majority carriers can then flow through the n-type linking drift regions 141 l5 141 2 and into the lower drift regions 133 l5 133 2 towards the drain 102.
  • Majority carriers can also flow through the upper drift region 142 towards the injector 113 l5 113 2 .
  • this arrangement allows a hybrid silicon n-channel vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor device to benefit from a charge balanced structure.
  • the charge balanced structure allows the thickness of the device (i.e. the dimension along the z-axis) to be reduced because the electric field generated by applying a source-drain voltage (with gate-source voltage of 0 V) is distributed laterally (along the x-direction), thereby driving electrons towards the n-type columns 133 l5 33 2 and holes towards the p-type columns 132 l 5 132 2 , 134 l5 134 2 .
  • a space charge region (not shown) is formed which spreads out along a vertical p-n junction (not shown).
  • the lower drift regions 133 l5 133 2 are depleted and act like the voltage sustaining layers of a PIN diode structure.
  • the source-drain voltage required to reach a critical electric field is now much higher than a similar device without a charge balance structure.
  • the length of the drift region can be reduced, which reduces the resistance of the drift region length.
  • use of the vertical charge balance structure allows the doping of the n-type drift regions to be increased by about an order of magnitude or more, for example, from about 5xl0 15 cm -3 to about 5x10 16 cm -3 .
  • the higher doping concentration results in a further reduction in drift region resistance.
  • the reduction in the thickness of the device (i.e. along the z-axis) and the lower resistivity can help to significantly reduce the resistance of the lower drift regions, while maintaining or even increasing blocking voltage.
  • the charge balance structure allows the device to break the so-called “silicon limit” in one dimension defined by: sRon « 3.7x10 "9 . (BV) 2'6 (1) where sRon is the specific ON resistance (in ⁇ . ⁇ 2 ) and BV is the blocking voltage (in V). Reference is made to "On the Specific On-Resistance of High-Voltage and Power Devices", by R.
  • the super junction device 101 has a lateral charge balance structure configured such that a lateral blocking voltage for the lateral insulated-gate bipolar transistor is the same or similar to the vertical blocking voltage for vertical metal-oxide
  • the lateral charge balance structure helps to shield the radius of curvature of the p-type body regions 112 l5 112 2 .
  • FIG 17a a schematic circuit diagram of the super junction device 101 prior to the onset of injection is shown.
  • the device 101 includes a MOSFET 143, a bipolar transistor 144 which includes p-n junctions 145 l5 145 2 (forming the emitter to base junction) and a Schottky diode 146.
  • the bipolar transistor 144 is a pnp transistor.
  • the drain 147 of the MOSFET 143 is connected to the drain D of the transistor 101 through a first, vertical resistive element 148 t and to the base 149 of the bipolar transistor 144 via a second, lateral resistive element 148 2 .
  • the first and second resistive elements 148 l5 148 2 are joined at node 150 at which electron current branches.
  • the first and second resistive elements 148 l 5 148 2 have values R vertical and ⁇ -lateral respectively.
  • the Schottky diode 146 has a depletion region 151 having a capacitance C dep(Sch) which varies with the thickness of the depletion region 151.
  • the capacitance is shown in series with the Schottky diode 146 which has a resistance, R Schottk .
  • the p-n junctions 145 l5 145 2 also have respective depletion regions 152 l 5 152 2 , each having a capacitance C dep(p _ n) .
  • Figure 17b shows the flow of majority carriers and minority carriers through the transistor 101 which, in this example, are electrons and holes respectively.
  • a parasitic bipolar transistor 153 may affect operation of the device.
  • V d As the drain voltage, V d , increases further, the potential barrier is overcome and thermionic emission of electrons via the Schottky diode 146 starts, thereby biasing the region 149 beneath the p-type injectors 113 l5 1 3 2 below V d and, thus, forward biasing the p-n junctions 145 l 5 145 2 within the base-emitter region of the bipolar transistor 144.
  • the Schottky diode 146 acts to maintain that bias thereby ensuring that any change in resistance of the second resistive element 149 2 due to increased electron
  • the charge balance structure offers a way of controlling the ratio of electrons flowing to the substrate 102 and injector 113 l , 3 2 and, thus, set the onset of IGBT-action, as well providing other benefits and features, as will now be described in more detail.
  • the charge balance structure in particular the p-type cross layer 137 and upper drift region 142, provides a reduced surface field (RESURF)-like distribution of electric field resulting in a reduction in electric field at the surface between the oxide 105 l5 105 2 and the upper drift region 142.
  • This helps to maintain the one- dimensional (that is, doping-dependent) blocking voltage for the interface between the p-type body region 112 l 5 112 2 and the upper drift region 142.
  • the lateral distance, L, between the channel and injector can be reduced without the risk of punch-through, as the extent of the depletion region (not shown) from the p-body regions 112 l5 112 2 can be reduced by increasing the doping concentration in the upper drift region 142.
  • the doping concentration in the upper drift region 142 and the lower drift regions 133 l5 133 2 can be independently controlled (that is, they can have different doping concentrations).
  • the flow of electrons from the n-type source regionsl l S j , 1 5 2 to the injectors 113 l5 113 2 and the substrate 102 can be independently controlled.
  • the ratio, R, of the areas (in the x-y plane) of the p-type linking regions 136 l5 136 2 and the n-type linking regions 141 l5 141 2 can be set to control the proportion of electrons flowing laterally towards the injectors 113 l5 113 2 through the upper drift region 142, i.e. set R j and R 2 respectively.
  • minority carriers i.e. holes
  • the ratio, R has a value of about 1 :10.
  • the ratio, R can have a value, for example, between about 1 :5 to about 1 :20.
  • the p-type linking regions 136 l5 36 2 do not appreciably inhibit electron flow to the drain.
  • the ability to control electrons flowing to the injector and to the base 148 of the pnp transistor 144 provides another parameter which can be controlled.
  • automatic de-biasing of the pnp transistor 144 can occur which can help to avoid localisation of current and ensure that emitter current is evenly distributed.
  • the base region 148 of the pnp transistor 144 within the charge balanced structure is fixed by the gap having a thickness, b.
  • the thickness of the gap can be optimised to ensure adequate effective base thickness of the pnp transistor 144 to provide improved levels of pnp emitter current gain for the limited levels of base drive available via the metal-oxide- semiconductor channel.
  • the base thickness, b is relatively unaffected by any growing depletion width from the p-type body regions 1 12 l 5 112 2 as the source- drain voltage increases.
  • the base thickness is reduced and so pnp current gain, hFE, increases.
  • the behaviour of pnp gain in the transistor 1 shown in Figure 1 differs from a conventional pnp device as a conventional device is fabricated with minimum base width, whereas base region within the transistor 1 ( Figure 1) is relatively large and results in low gain.
  • the effective base width is reduces as source depletion region extends laterally towards region 13 1 ( Figure 1) and so gain improves with source- drain voltage.
  • the base width is set by fabrication to have a high gain and the doping in the base region 149 is preferably sufficiently high so as to avoid collector to emitter punch-through.
  • the thickness of the p-type cross layer 137 can be adjusted for both doping density and vertical thickness, s, along the z-axis.
  • the doping in the p-type cross layer 137 is chosen to balance the doping in the top n-type layer 142.
  • the ability to alter the thickness of the p-type cross layer 137 allows optimisation of the charge balance structure and extent of the depletion from the p-type body regions 112 l 5 112 2 into n-type regions 141 l 5 141 2 , 142. Furthermore this can be achieved without
  • the minimum lateral resistance is limited by punch-through thereby setting a minimum distance, L, between the edge 22 of the p-body 12 and the edge 23 of the injector region 13 1 .
  • the minimum lateral resistance also sets a minimum length for the drift region.
  • the distance from the edge of the channel 122 to the edge of the injector 1 3 a effectively controls the value of source-drain voltage needed bring about the onset of conductivity modulation of the drift region by the injector.
  • the lateral distance can be greatly reduced and so provide way to further reduce the injection start-up voltage.
  • a wider range of injection start-up voltages can be used.
  • the charge balanced structure allows all of the channel area to participate in injecting electrons into the n-type regions 133 l 5 133 2 to the drain 102. Holes are restricted to the upper layers bounded by the p-type cross layer 137. This has the effect of reducing the switch off time as the holes can easily reach the p- type body 112 l 5 i.e. the collector of the bipolar transistor 144, with minimal travel. Unlike a conventional IGBT, there is no need to wait for recombination since recombination is not the only means of charge removal.
  • the device 1 shown in Figure 1 has been simulated in an un-clamped inductive switching test circuit and has shown to be free of latch-up.
  • the transistor 101 also has the potential to provide latch-up-free operation as the mechanism for charge removal from the drift region is via the p-type grounding channels 136 l5 36 2 , 137 for holes to the p-type body regions 112 l5 112 2 and via the drift region to the drain 102 for electrons. Even if a parasitic npn bipolar transistor 152 started conducting, then electrons collected in the drift region (which serves as the npn collector) simply find their way to the drain 102 thereby contributing to the MOSFET current.
  • the parasitic npn transistor 152 will therefore switch off when the channel restricts the electron base drive to the pnp transistor 144. This in turn removes any hole current to the source S via the p-type body regions 112 l5 H2 2 . Thus, there is no sustained base bias to the bipolar transistor 144 to cause latch-up
  • the devices 1 ( Figure 1), 101 ( Figure 11) also provide a mechanism to "self clamp” when switching inductive loads, thus preventing the device going into avalanche from which recovery times can be long.
  • the device 101 is fabricated in a similar to way to the device 1 shown in Figures 1 and la. However, instead of growing an epitaxial layer of lightly-doped n-type
  • monocrystalline silicon having appropriate n-type and p-type regions for providing the charge balance structure is formed.
  • Figure 19a shows a highly-doped n-type monocrystalline substrate 103.
  • P-type and n-type columns 132/, 133/, 134/, 135, 132/, 133/, 134/ are formed (step S1A).
  • the p-type regions are doped to a net concentration of about 5xl0 16 cm -3 .
  • the columns can be formed by growing layers of lightly doped p-type silicon may be deposited and, for each epitaxial layer, doping regions of the epitaxial layer with donors.
  • the columns can be formed by growing undoped layers and, for each epitaxial layer, selectively doping respective regions of the epitaxial layer with acceptors and donors. Doping can be performed by ion implantation.
  • the structure is annealed to activate the implants. The process can be repeated for each epitaxial layer until the columns can have a suitable thickness, d 3 , e.g. between about 10 and 50 ⁇ . Between 2 and 10 or more epitaxial layers can be used.
  • the columns 132 , 133 , 134 , 135, 132 2 ', 133 2 ', 134/ can be formed by growing an epitaxial layer of lightly-doped n-type monocrystalline silicon.
  • the epitaxial layer (not shown) can be doped in-situ during growth with a donor to a concentration of about 5xl0 16 cirf 3 and has a thickness of between about 10 and 50 ⁇ .
  • a dielectric layer (not shown) is formed on the epitaxial layer (not shown) and it patterned to form an etch mask.
  • trenches are formed by dry etching which reach the substrate 102.
  • the trenches (not shown) can be filled by epitaxially growing lightly-doped p-type monocrystalline silicon.
  • more than one epitaxial layer can be deposited, etched and filled.
  • the p-type cross layer 137 and the rest of the outer p-type columns 132 l5 132 2 , the p-type linking regions 136 l 5 136 2 ( Figure 14), the n-type linking regions 141 l5 141 2 ( Figure 13) are formed (step SIB).
  • a similar process to one of those described earlier for forming the columns can be used.
  • the regions are doped to a concentration of about 5x10 16 cirf 3 .
  • the thickness, s, of the p-type cross layer 137 can be between about 0.5 ⁇ to about 5 ⁇ .
  • the epitaxial layer 154 can be doped in-situ during growth with a donor to a concentration of about 5x10 16 cm and can have a thickness, t, of between about 1 and 2 ⁇ .
  • the values of lengths q l 5 q 2 , q 3 , q 4 and values of thicknesses d 3 , s, b and t can be found by routine experiment and/ or by computer simulation using, for example, a bespoke or commercially-available Poisson solver.
  • Different p-type and n-type doping concentrations may be used.
  • the doping concentrations can differ between the lower drift regions 133 l5 133 2 and the upper drift region 142.
  • the doping concentrations can differ between the inner p-type columns 134 l5 134 2 and the buried p-type cross layer 137.
  • the semiconductor device may be a p-channel device.
  • the drift region, the body region, the source region and injector regions may take the form of a lightly - doped p-type semiconductor material, lightly-doped n-type semiconductor material, heavily-doped p-type semiconductor material and heavily-doped n-type
  • the semiconductor device may be based on silicon carbide.
  • the substrate may comprise 4H or 3C polytypes.
  • the drift region and the body, source and injector regions may comprise silicon carbide.
  • Other elemental or compound semiconductors, such as GaN, can be used.
  • Other metallizations can be used to provide the high Schottky barrier.
  • a single continuous metallization forming a Schottky contact and an ohmic contact to the p-n junctions need not be used.
  • Separate layers of metallization (which may comprise different materials) can be used and may be connected, for example, by an overlayer or by a bond wire.
  • Processing steps and process parameters may be altered and optimised. For example, doping concentrations, dopant ions, ion-implant energies can be varied. Methods of deposition and methods of etching can also be varied. Also, dielectric materials can be altered. For example, silicon dioxide need not be used and can be replaced, for example, by silicon oxynitride or high-k dielectrics. A gate metallization may be used instead of gate poly, for example, comprising tungsten (W).

Abstract

A semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor is disclosed. The device has a Schottky-biased p-n junction.

Description

Semiconductor device
Field of the Invention
The present invention relates to a semiconductor device and more specifically to a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor (MOSFET) and a vertical or lateral insulated gate bipolar transistor (IGBT). The present invention also relates to a semiconductor device having an integrated vertical and lateral charge balance structure. Background
In a vertical MOSFET, current flows vertically between a source having one conductivity type and a substrate drain of the same conductivity type. The
MOSFET operates based on the flow of only majority carriers through the drift region.
An IGBT has a similar structure to a vertical MOSFET, but differs in that the drain is of the opposite conductivity type. Thus, both majority and minority carriers are injected into the drift region. This has the effect of reducing the ON resistance of the IGBT compared to the MOSFET, but also has the effect of increasing the gate switching speed.
A form of vertical MOSFET device has been proposed in which a MOSFET is provided with a minority carrier injector in the form of a p-type region which forms a p-n junction with the n-type drift region so as to increase switching speed and reduce the ON resistance. Reference is made to "A novel trench-injector power device with low ON resistance and high switching speed", David K. Y. Liu and James D. Plummer, IEEE Electron Device Letters, volume 9, Issue 7, pages 321 - 323 (1988) and "Design and analysis of a new conductivity-modulated power MOSFET", David K. Y. Liu and James D. Plummer, IEEE Transactions on Electron Devices, volume 40, Issue 2, pages 428 - 438 (1993).
The device has a drawback of needing to control separately an additional terminal. However, this can be avoided by connecting the injector to the drain. However, this arrangement presents a problem, namely that when the same bias is applied to the injector and drain, no voltage difference develops across the p-n junction and so no minority carrier injection occurs. In particular, a small voltage difference initially develops due to differing resistances between the source and the drain and the source and the injector. However, the difference in resistance is soon compensated by low-level minority carrier injection and so the effect stops.
An alternative approach is to form a hybrid vertical DMOS/lateral IGBT device and reference is made to "A new hybrid VDMOS-LIGBT transistor" T. Paul Chow & B. Jayant Baliga, IEEE Electron Device Letters, volume 9, pages 473 - 475 (1988).
This device comprises two devices in parallel, namely a p-channel vertical MOSFET DMOSFET and a p-channel lateral IGBT. However, the hybrid device needs external resistors to bias the device and, since the resistors are fixed, bipolar action is not optimized.
In lateral MOSFET devices, it has been proposed to inject minority carriers using a Schottky contact and reference is made to "The SINFET - A Schottky injection MOS-gated power transistor", Johnny K. O. Sin, C. Andre T. Salama and Li-Zhang Hou, IEEE Transactions on Electron Devices, volume ED-33, page 1940-1947 (1986).
However, the number of injected minority carriers is low. Thus, although conductivity can be modulated using minority carriers, the device falls well short of achieving full IGBT-like bipolar action.
Summary
The present invention seeks to provide an improved semiconductor device.
According to a first aspect of the present invention there is provided a
semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor having a Schottky- biased p-n junction.
Thus, the Schottky biasing helps to ensure that the p-n junction is forward biased and continues to inject minority carriers. Therefore, the device provides an integrated vertical metal oxide semiconductor field effect transistor and vertical or lateral insulated gate bipolar transistor.
The semiconductor device may comprise a minority-carrier injector which comprises the p-n junction and a Schottky diode, wherein the Schottky diode is arranged in parallel with the p-n junction so as to provide the Schottky-biased p-n junction.
The semiconductor device may comprise a semiconductor region (for example, an epitaxial region) disposed between first and second opposite surfaces, at least a first portion of the semiconductor region comprising a semiconductor of a first conductivity type providing a drift region for majority carriers in a direction between the first and second surfaces, at least a second portion of the
semiconductor region providing a body region comprising a semiconductor of a second, opposite conductivity type disposed at the first surface of the
semiconductor region and at least a third portion of the semiconductor region providing a source region comprising a semiconductor of the first conductivity type disposed at the first surface of semiconductor region and within the body region. The semiconductor region may comprise the same semiconductor, such as silicon or silicon carbide, but have regions of different conductivity types, i.e. n-type and p- type regions. The minority-carrier injector may be disposed at the first surface (e.g. top surface) of the semiconductor region laterally separated from the body region. Thus, a lateral insulated gate bipolar transistor can be formed. However, the minority- carrier injector may be disposed at the second surface (e.g. bottom surface) of the semiconductor region so as to form a vertical insulated gate bipolar transistor.
At least a fourth portion of the semiconductor region may comprise a
semiconductor of the second conductivity type disposed at the first surface of the semiconductor region with a semiconductor of the first conductivity type so as to form the p-n junction. The Schottky diode may comprise a metallization in contact with the first surface of semiconductor region. The Schottky diode may have a barrier height equal to or greater than 0.7 eV. If the minority-carrier injector is disposed at the second surface, then the metallization may be in contact with the first surface of semiconductor region. A different metallization and/ or process may be used to form a drain contact to the semiconductor region.
The semiconductor device may further comprise a gate structure disposed on the first surface of the semiconductor region and configured to control flow of carriers from the source region into drift region. The semiconductor device may further comprise a drain region disposed on the second surface of the semiconductor region. The drain region may include a semiconductor substrate, e.g. highly-doped substrate such as a wafer. The drain region may include a metallization, which preferably forms an ohmic contact. The device may comprise a drain region comprising a semiconductor substrate (or layer) of a first conductivity type or a metallic layer, a drift region of the first conductivity type disposed on the drain region, a body region of a second, opposite conductivity type, for example formed in or adjacent to the drift region, and a source region of the first conductivity type disposed within the body region.
According to a second aspect of the present invention there is provided a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor, the device comprising a semiconductor substrate drain of a first conductivity type, a drift region or epitaxial region of the first conductivity type disposed on the semiconductor substrate drain, a body region of a second, opposite conductivity type disposed within the drift region or epitaxial region, a source region of the first conductivity type disposed within the body region. The device further comprises a minority-carrier injector which comprises an injector region of the second conductivity type disposed within the drift region or epitaxial region and laterally separated from the body region, the injector region and the drift region or epitaxial region forming a p-n junction, and a metallic electrode in contact with the drift region or epitaxial region laterally adjacent to and electrically connected to the injector region, the metallic electrode and the drift region or epitaxial region forming a Schottky diode in parallel with the p-n junction.
Thus, the Schottky diode helps to ensure that the p-n junction is forward biased and continues to inject minority carriers into the drift region or epitaxial region.
Therefore, the device provides an integrated vertical metal oxide semiconductor field effect transistor and lateral insulated gate bipolar transistor.
The first and second conductivity types may be n-type and p-type respectively.
The body region and the substrate (or, if no substrate is present, for example, because it has been removed or a portion of it has been removed, the second surface of the semiconductor region) may be vertically separated by a first given distance and the body region and the injector region are laterally separated by a second given distance exceeding equal to or greater than the first given distance. The second given distance may be about 6 to 12 μπι.
The device may further comprise a depletion reducing region comprising a semiconductor of the first conductivity type disposed within the drift region, more highly doped than the drift region and disposed between the body region and the injector region. This can help reduce the lateral size of the device. The body region and the substrate may be vertically separated by a first given distance and the body region and the injector region are laterally separated by a second given distance less than the first given distance. The Schottky diode may have a barrier height equal to or greater than 0.7 eV. The metallic electrode may include a layer of platinum silicide forming the Schottky diode.
The metallic electrode and the semiconductor substrate may be coupled so as to be at substantially the same potential. The device may comprise a further, different metallic electrode in contact with the source region.
The drift region or epitaxial region may comprise silicon. Alternatively, the drift region or epitaxial region may comprise silicon carbide semiconductor. The drift region, the body region, the source region and the injector region may comprise the same semiconductor.
The device may further comprise a charge balance structure. The charge balance structure may comprise first and second charge balance regions comprising a semiconductor of the second conductivity type, the first charge balance region extending vertically between the substrate (or bottom surface of a semiconductor region which would have been formed on a substrate) and the source region and the second charge balance region laterally spaced between the p-n junction and the first charge balance region, such that a vertical drift region comprising a semiconductor of the first conductivity type is disposed between the first and second charge balance regions. The charge balance structure may further comprise a third charge balance region comprising a semiconductor of the second conductivity type disposed under the p-n region and extending laterally to the second charge balance region, wherein a lateral drift region comprising a
semiconductor of the first conductivity type is disposed over the third charge balance region. The charge balance structure may further comprise a region comprising a semiconductor of the second conductivity type linking the first and second charge balance regions, the region configured to leave a region comprising a semiconductor of the first conductivity type linking the lateral and vertical drift regions. According to a third aspect of the present invention there is provided a
semiconductor device having an integrated vertical and lateral charge balance structure.
The semiconductor device may comprise a planar substrate, a first conductive region of a first conductivity type extending perpendicularly to the substrate between first and second ends and having at least first and second sides, first and second charge balance regions of second conductivity type disposed on the first and second sides respectively of the first conductive region, a second conductive region of the first conductivity type extending parallel to the substrate between first and second ends and having at least a first side, a third charge balance region of the second conductivity type disposed on the first side of the second conductive region, and the device may further comprise at least one region of the first conductivity type linking the first and second conductive regions so as to form a continuous region of the first conductivity type between the first end of the first conductive region to the second end of the second conductive region and at least one region of the second conductivity type linking the first, second and third charge balance regions so as to form an integrated charge balance structure.
According to a fourth aspect of the present invention there is provided an integrated circuit comprising at least one semiconductor device.
According to a fifth aspect of the present invention there is provided a method of fabricating a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor, the method comprising providing the insulated gate bipolar transistor with a Schottky- biased p-n junction. The method may comprise providing a minority-carrier injector which comprises the p-n junction and a Schottky diode, wherein the Schottky diode is arranged in parallel with the p-n junction so as to provide the Schottky-biased p-n junction. The method may comprise providing a semiconductor region, such as an epitaxial layer of semiconductor material(s), between first and second opposite surfaces (or top and bottom surfaces or faces). The semiconductor region may be formed in stages, e.g. by growing or depositing a first layer of semiconductor material and growing or depositing one or more successive layers of semiconductor material on the first layer. At least a first portion of the semiconductor region may comprise a semiconductor of a first conductivity type, preferably n-type, providing a drift region for majority carriers in a direction between the first and second surfaces. A lateral drift region may also be provided. At least a second portion of the semiconductor region may provide a body region comprising a semiconductor of a second, opposite conductivity type, preferably p-type, disposed at the first surface of the semiconductor region. At least a third portion of the semiconductor region may provide a source region comprising a semiconductor of the first conductivity type disposed at the first surface of semiconductor region and within the body region. The semiconductor region may comprise the same semiconductor, such as silicon or silicon carbide. The minority-carrier injector may be disposed at the first surface of the semiconductor region laterally separated from the body region so as to form a lateral insulated gate bipolar transistor or disposed at the second surface of the semiconductor region vertically separated from body region so as to form a vertical insulated gate bipolar transistor.
At least a fourth portion of the semiconductor region may comprise a
semiconductor of the second conductivity type disposed at the first surface of the semiconductor region with a semiconductor of the first conductivity type so as to form the p-n junction. The Schottky diode may comprise a metallization in contact with the first surface of semiconductor region.
The method may further comprise providing a gate structure on the first surface of the semiconductor region. The method may further comprise providing a drain region on the second surface of the semiconductor region. This may comprise growing the semiconductor region on a semiconductor substrate. Additionally or alternatively, the drain region or contact may include a metallization. The method may comprise providing a semiconductor substrate drain of a first conductivity type, a drift region of the first conductivity type disposed on the semiconductor substrate drain, a body region of a second, opposite conductivity type, for example formed in or adjacent to the drift region, and a source region of the first conductivity type disposed within the body region.
The method may comprise providing a drift region of first conductivity type on a semiconductor substrate of the first conductivity type, forming a body region of a second, opposite conductivity type within the drift region, forming a source region of the first conductivity type within the body region, and providing the Schottky- biased p-n junction may comprise forming an injector region of the second conductivity type within the drift region, laterally separated from the body region, wherein the injector region and the drift region provide a p-n junction and providing a metallic electrode in contact with the drift region laterally adjacent and electrically connected to the injector region, wherein the metallic electrode and the drain region provide a Schottky diode in parallel with the p-n junction.
According to a sixth aspect of the present invention there is provided a method of fabricating a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor, the method comprising providing a drift region of first conductivity type on a semiconductor substrate of the first conductivity type, forming a body region of a second, opposite conductivity type within the drift region, forming a source region of the first conductivity type within the body region, forming an injector region of the second conductivity type within the drift region, laterally separated from the body region, wherein the injector region and the drift region provide a p-n junction; and providing a metallic electrode in contact with the drift region laterally adjacent and electrically connected to the injector region, wherein the metallic electrode and the drain region provide a Schottky diode in parallel with the p-n junction. The body region and the substrate may be vertically separated by a first given distance and wherein forming the injector region within the drift region comprises laterally separating the injector region from the body region by a second given distance exceeding the first given distance.
The device may further comprise a depletion reducing region comprising a semiconductor of the first conductivity type disposed within the drift region, more highly doped than the drift region and disposed between the body region and the injector region.
The body region and the substrate may be vertically separated by a first given distance and the body region and the injector region are laterally separated by a second given distance less than the first given distance.
The method may further comprise providing a charge balance structure.
According to a seventh aspect of the present invention there is provided a method of fabricating a semiconductor device, the method comprising providing an integrated vertical and lateral charge balance structure.
Providing the integrated vertical and lateral charge balance structure may comprise providing a planar substrate, providing a first conductive region of a first conductivity type extending perpendicularly to the substrate between first and second ends and having at least first and second sides, providing first and second charge balance regions of second conductivity type disposed on the first and second sides respectively of the first conductive region, providing a second conductive region of the first conductivity type extending parallel to the substrate between first and second ends and having at least a first side, providing a third charge balance region of the second conductivity type disposed on the first side of the second conductive region, providing at least one region of the first conductivity type linking the first and second conductive regions so as to form a continuous region of the first conductivity type between the first end of the first conductive region to the second end of the second conductive region, and providing at least one region of the second conductivity type linking the first, second and third charge balance regions so as to form an integrated charge balance structure. According to an eighth aspect of the present invention there is provided a method of operating a semiconductor device operable as a vertical metal oxide
semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor having a Schottky-biased p-n junction, the metal oxide semiconductor field effect transistor and insulated gate bipolar transistor sharing a source region (or source/ emitter terminal) and a drain region (or drain terminal) and wherein the Schottky-biased p-n junction has a terminal (i.e. an injector and/or collector terminal), the method comprising grounding the source region and applying given bias(es) to the injector/collector terminal and to the drain region. Preferably the same bias is applied to the injector/collector terminal and the drain region.
Brief Description of the Drawings
Certain embodiments of the present invention will now be described, by way of example, with reference to the accompanying drawings in which:
Figure 1 is a vertical section of an embodiment of a semiconductor device in accordance with the present invention;
Figure la is more detailed view of the semiconductor device shown in Figure 1 ; Figure 2 illustrates a set of mask layers used to fabricate the semiconductor device shown in Figure 1 ;
Figure 3a is a schematic circuit diagram of the semiconductor device shown in Figure 1 before the onset of minority carrier injection;
Figure 3b is a schematic circuit diagram of the semiconductor device shown in Figure 1 after the onset of minority carrier injection;
Figure 4 is a vertical section of the semiconductor device shown in Figure 1 illustrating operation;
Figure 5 schematically illustrates the behaviour of the semiconductor device shown in Figure 1 ;
Figures 6a to 6p are vertical sections through the semiconductor device shown in Figure 1 at different stages during fabrication;
Figure 7 is a process flow diagram of a method of fabricating the device shown in Figure 1 and part of the device shown in Figure 11 ;
Figure 8 is a vertical section of another embodiment of a semiconductor device in accordance with the present invention;
Figure 9 is a vertical section of yet another embodiment of a semiconductor device in accordance with the present invention;
Figure 10 is a vertical section of still another embodiment of a semiconductor device in accordance with the present invention;
Figure 11 is a perspective view of a further embodiment of a semiconductor device in accordance with the present invention;
Figure 12 is a plan view of the semiconductor device shown in Figure 11 ;
Figure 13 is a vertical section of the semiconductor device shown in Figure 12 taken along the line B-B';
Figure 14 is a vertical section of the semiconductor device shown in Figure 12 taken along the line C-C; Figure 15 is a vertical section of the semiconductor device shown in Figure 12 taken along the line D-D';
Figure 16 is a vertical section of the semiconductor device shown in Figure 13 taken along the line E-E';
Figure 17a is a schematic circuit diagram of the semiconductor device shown in Figure 11 before the onset of minority carrier injection;
Figure 17b is a schematic circuit diagram of the semiconductor device shown in Figure 11 after the onset of minority carrier injection;
Figure 18 schematically illustrates the behaviour of the semiconductor device shown in Figure 11 ;
Figure 19a to 19d are vertical sections through the semiconductor device shown in Figure 11 at different stages during fabrication of a charge balance structure; and Figure 20 is a process flow diagram of a method of fabricating a charge balance structure.
Detailed Description of Certain Embodiments
Referring to Figures 1, la and 2, an embodiment of a power semiconductor device in accordance with the present invention is shown. The semiconductor device takes the form of a hybrid silicon n-channel vertical metal oxide semiconductor field effect transistor and insulated gate bipolar transistor device 1. The transistor 1 is configured to permit injection of both majority and minority carriers into a drift region.
The transistor 1 includes a heavily-doped n-type monocrystalline silicon substrate 2 which functions as a drain region. An epitaxial layer 3 of lightly-doped n-type monocrystalline silicon (herein also referred to as the "drift region") is arranged on an upper surface 4 of the substrate 2. A field oxide 5 is located at an upper surface of the epitaxial layer 3 and has first and second windows 6l5 62 defining first and second laterally-separated upper surfaces 7l5 72 of the epitaxial layer 3. The first and second windows 6l5 62 have lengths ll5 \2 of about 6 μηι and 4.8 μηι respectively and are separated by a length, 13, of about 7 μηι. A gate oxide 8 is disposed within the first window 61 on the upper surface 71 of the epitaxial layer 3. The gate oxide 8 runs along the upper surface 7l of the epitaxial layer 3 and abuts the field oxide 5 thereby forming a step 9. A layer of heavily- doped n-type polycrystalline silicon 10 (which may also be referred to as the "gate poly") is disposed on the gate oxide 8 and runs over the step 9 onto the field oxide 5. Silicon dioxide spacers 11 are formed on the sides of the gate poly 10.
A p-type body 12 comprising a lightly-doped p-type diffusion well is disposed within the epitaxial layer 3 at the first upper surface 7l. As shown in Figure la, the p-type body 12 extends laterally under the spacer 11 and the gate oxide 8.
First and second p-type injectors 13l5 132 in the form of heavily-doped p-type diffusion wells are located within the epitaxial layer 3 at the second upper surface 72. The p-type injectors 13l 5 132 are spaced apart to leave a region 14 of the n-type epitaxial layer 3 extending up to the second upper surface 72 of the epitaxial layer 3.
An n-type source region 15 in the form of a heavily-doped n-type diffusion well is disposed within the p-type body 12 at the upper surface 7l. A p-type body short (not shown) in the form of a heavily-doped p-type diffusion well is also provided at the first upper surface using a body short mask (Figure 2).
A layer 17 of silicon dioxide runs over the gate poly 10 and the field oxide 5, and has windows 18l5 182, 183. Layers of 19l5 192, 193 of metallization are disposed on the silicon dioxide layer 17 covering windows 18l5 182, 183. The first metallisation layer 19a provides a source/emitter terminal S, the second metallisation layer 192 provides a gate terminal G and the third metallisation layer 19a provides a collector/injector terminal I. The metallization layers 19l 5 192, 193 each comprise a bi-layer comprising a high-barrier metal silicide base layer comprising, for example, platinum silicide (PtSi), and a high-conductivity overlayer comprising, for example, aluminium (Al).
As shown in Figure la, the third metallisation layer 193 is in contact with the first and second p-type injectors 13l5 132 and the n-type region 14 between the p-type injectors 13l 5 132. The p-type injectors 13l 5 132 form respective p-n junctions 20l 5 202 with the epitaxial layer 3 and the metallisation layer 193 forms a Schottky diode 21 with the epitaxial layer 3. Thus, the pair of p-n junctions 20l 5 202 and the Schottky diode 21 are arranged in parallel between the metallisation layer 193 and the epitaxial layer 3.
The high-barrier metal silicide forms a Schottky diode with the epitaxial layer 3, but forms ohmic contacts to the heavily-doped injector and source regions 13l 5 132, 15. As shown in Figure 1 , an inner edge 22 of the p-type body 12 and an inner edge 23 of the p-type injector 13a which is closest to the p-type body 12 are separated by a length, L. A bottom 24 of the p-type body 12 is separated from the highly-doped substrate 2 by a depth, The length, L, approximately equals the depth, dl 5 i.e. L ~ d, to maintain the breakdown voltage of the device. The separation length, L, and the depth, d, are about 8 μπι. As will be explained later, the length, L, can be smaller if a highly-doped field stop region or a charge balance structure is used.
Also as shown in Figure 1 , a source-drain depletion region 25 having an edge 26 is formed in the epitaxial layer 3 due to the junction between the p-type body 12 and the n-type epitaxial layer 3. As will be explained in more detail later, the transistor 1 is arranged so that, in blocking mode, the edge 26 of the depletion region 25 does not reach the inner edge 23 of the first injector 131.
Referring to Figures 3a and 3b, schematic circuit diagrams of the transistor 1 before and after the onset of minority carrier injection are shown.
Referring in particular to Figure 3a, the transistor 1 includes a MOSFET 28.
The drain of the MOSFET 28 is connected to the drain D of the transistor 1 through a first, vertical resistive element 29 The drain of the MOSFET 28 is also connected to a local bias point 30 through a second, lateral resistive element 292. The first and second resistive elements 29l 5 292 are joined at node 31 at which electron current branches when flowing towards the drain D and the injector I. The first and second resistive elements 29l5 292 have values Rj and R2 respectively. In this example, Rj and R2 depend on the depth, dl5 of the epitaxial region 3 and the lateral separation of the p-body region 12 and the Schottky diode 21 respectively. Referring in particular to Figure 3b, after the onset of minority carrier injection, minority carriers (in this example, holes) flow from the injector I through a bipolar transistor 32 to source S.
Figure 3b shows the flow of majority carriers and minority carriers through the transistor 1 which, in this example, are electrons and holes respectively. When a large minority carrier current flows through bipolar transistor 32 between emitter and collector, a parasitic bipolar transistor 33 may affect operation of the device.
Referring also to Figures 3a, 3b, 4 and 5, operation of the transistor 1 will now be described.
The transistor 1 is arranged so that the source S is grounded and that the drain D and injector I are connected together to a positive voltage Vd. The gate G is used to control the transistor 1 and if a voltage Vg is applied to the gate G and the voltage exceeds a threshold voltage Vth, then current flows through the transistor 1 from the source region 15 into the epitaxial region 3.
The transistor 1 can provide an integrated vertical metal oxide semiconductor field effect transistor and lateral insulated gate bipolar transistor.
At low Vd and high V , most of the current (which comprises majority carriers, i.e. electrons) flows to the drain D. Thus, the transistor 1 operates in a similar way to a conventional vertical MOSFET, as shown in Figure 3a. Initially, the potential barriers of the p-n junctions 20l5 202 limit or prevent the flow of majority carriers (i.e. electrons) to injector I. As Vd increases, electron charge builds up in a region 34 beneath the p-type injectors 13l5 132. As charge builds up a depletion region 35 associated with the Schottky diode 21 is reduced. The depletion region 35 of the Schottky diode 21 has a capacitance Cdep(Sch) which varies with the thickness of the depletion region 35. As shown in Figure 3a, the capacitance is shown in series with the Schottky diode 21 which has a resistance, RSchottky The p-n junctions 20l5 202 also have respective depletion regions 36l 5 362, each having a capacitance Cdep(p_n). As Vd increases further, the depletion region 35 becomes so small that thermionic emission of electrons through the Schottky diode 21 occurs. This biases the region 34 beneath the p-type injectors 13l5 132 below Vd and, thus, forward-biases the p-n junctions 20l5 202. The p-n junction 20j forms the emitter- base junction of the bipolar transistor 32. Thus, minority carriers (i.e. holes) are injected into the portion of the epitaxial layer 3 which provides base region.
Majority carriers flow through the MOSFET via node 31 to provide a base current to drive the now activated pnp bipolar transistor 32 which passes minority carriers from the injector I to the source S thereby reducing the ON-resistance. Thus, the transistor 1 starts to imitate operation of a conventional IGBT. However, the transistor 1 benefits from low loss conduction with a high saturation current level, which in turn depends on channel geometry.
The Schottky diode 21 helps to dynamically bias the p-n junctions 20l5 202 and ensure that the p-n junctions 20l5 202 continue to inject minority carriers into the base region provided by the epitaxial region 3 at higher Vd unlike, for example, an ohmic resistor. If an ohmic resistor is used instead of the Schottky diode, then the p-n junctions 20l5 202 would initially be forward biased. However, once this occurs, the resistivity of the epitaxial region would drop, allowing the bias in region 34 to fall below that required to forward bias the p-n junctions 20l5 202 resulting in termination of injection of minority carriers.
The bias level within region 34 relative to Vd can be altered depending on the choice of metallisation (which varies the barrier height), the area of the Schottky diode and the operating point of Schottky diode 21. A barrier height of at least 0.7 eV is preferred. For example, the metallization may be platinum silicide with an aluminium overlayer. Platinum silicide has a work function of about 5.2 to 5.8 eV, resulting in a barrier height of around 0.84 eV. Referring also to Figure 1 , the geometry and dimensions of the transistor 1 are chosen so that the depletion layer 25 arising from junction between the p-type body region 12 and the epitaxial layer 3 does not touch the depletion layer arising from the closest p-n junction 20l. If the depletion layer should touch, then there is no space charge or neutral region between the source (which is at ground) and the injector (which is at Vd) leading to punch through and uncontrolled drain to source current flow. Thus, although injection start up voltage is reduced (in terms of Vd), the closer the injector is to the p-type body region 12, the lateral extent of the depletion layer 25 in blocking mode sets a minimum lateral separation, L.
For silicon, for the dimensions given earlier, the transistor 1 has a breakdown voltage of about 120 V. An array (not shown) of transistors 1 and bond pads (not shown) can be implemented in a chip having an area of about 3.3 mm2 and can support 10 A of current comprising majority carriers and 70 A of current comprising majority and minority carriers. For silicon carbide, for the dimensions given earlier, the transistor 1 has a breakdown voltage of about 1000 V.
The transistor 1 has a further advantage that if the device heats up (e.g. to 400 K or above) due to a surge current, then this will assist the onset of minority carrier injection, thereby reducing the ON state resistance of the device and, thus lowering, power loss.
Referring to Figures 1 , l a, 2, 6a to 6p and 7, a method of fabricating the
semiconductor device 1 will now be described.
Referring in particular to Figures 1 and 2, the device 1 is fabricated using a succession of masks including an active area mask for defining the extent of the field oxide 5, a gate poly mask for defining the extent of the gate poly 10, a p-body mask for defining the p-body well 12, an n-source mask for defining the n-source well 15, a body short mask for defining a body short (not shown), a contact mask for defining openings in the silicon dioxide layer 17 and a metal mask for defining metallization 19l 5 192, 193. Referring in particular to Figure 6a, a layer 3' of lightly-doped n-type
monocrystalline silicon is grown epitaxially on a highly-doped n-type
monocrystalline substrate 2 (step SI). The epitaxial layer 3' is doped in-situ during growth with arsenic or antimony to a concentration of about 5 Xl O15 cirf3 and has a thickness of about 8 μπι. In some embodiments, a highly-doped n-type field stop region 37 (Figure 8) can be formed (step S2).
Referring in particular to Figure 6b, a field oxide 5 is formed at the surface 7' (Figure 6a) of the epitaxial layer 3' (Figure 6a) by thermal oxidation using a LOCOS process (step S3). The field oxide 5 has thickness of about 1 μπι.
As shown in Figure 6b, the field oxide 5 has first and second windows 6l5 62 defining first and second laterally-separated upper surfaces 7l5 72 of the epitaxial layer 3.
Referring in particular to Figure 6c, a layer 8' of silicon dioxide is grown by thermal oxidation on the upper surfaces 7l, 72 of the epitaxial layer 3 (step S4). The silicon dioxide layer 8' has a thickness of about 1,500 A. Referring in particular to Figure 6d, a layer 10' of heavily-doped n-type
polycrystalline silicon is grown by chemical vapour deposition (CVD) over the silicon dioxide layer 8' and field oxide 5 (step S5). The polycrystalline layer 10' is doped in-situ during growth with phosphorous or arsenic to a concentration equal to or greater than about 1 X 1020 cirf3 and has a thickness of about 3,000 A. Instead of in-situ doping, an undoped polycrystalline layer 10' can be grown and doped by ion implantation after growth.
A layer (not shown) of photoresist is applied over the polycrystalline layer 10' and is patterned to form a mask (not shown). The mask pattern is transferred to the underlying polycrystalline layer 10' and silicon dioxide layer 8' by reactive ion etching (RIE) using suitable feed gases (step S6). Referring in particular to Figure 6e, the resulting structure includes the gate poly 10 and gate oxide 8.
Referring in particular to Figure 6f, with the mask (not shown) still in place, a further layer (not shown) of photoresist is applied and patterned so as to protect the epitaxial layer 3 in the second window 62. Boron ions are implanted at an energy between about 20 to 160 keV.
Referring to Figure 6g, the masks (not shown) are removed and the structure is annealed to activate the implant leaving a p-type well 12 doped to a concentration of between about 1 Xl O17 cirf3 and about 5 Xl O18 cm-3 and having a depth of about 0.7 to 1.2 μπι (step S7).
Referring to Figure 6h, a conformal layer (not shown) of silicon dioxide is deposited and globally etched back by RIE using suitable feed gases to form a spacer 11 around the sides of the gate poly 10 (step S8) .
Referring in particular to Figure 6i, a layer (not shown) of photoresist is applied and patterned so as to protect the epitaxial layer 3 in the second window 62. Arsenic ions are implanted into the epitaxial layer 3 in the first window 61 at an energy of about 40 keV.
Referring to Figure 6j, the mask (not shown) is removed and the structure is annealed to activate the implant leaving an n-type well 15 doped to a concentration of between about 1 Xl O20 cirf3 and about 3 Xl O20 cm-3 and having a depth of about 1 ,000 A to 5,000 A (step S9).
Referring to Figure 6k, a layer (not shown) of photoresist is applied and patterned to open a small hole over the first window 61 and two narrow windows (not shown) over the second window 62. Boron ions are implanted into the epitaxial layer 3 in the second window 62 at an energy of about 40 keV. Referring to Figure 61, the mask (not shown) is removed and the structure is annealed to activate the implant leaving an p-type body short (Figure 2) and p-type injectors 13l5 132 doped to a concentration of between about 1 XlO20 cm"3 and about 3 XlO20 cm"3 and having a depth of about 1,000 A to 5,000 A (step S10).
Referring to Figure 6m, another layer 17' of silicon dioxide having a thickness of about 4,000 A is deposited (step Sl l).
A layer (not shown) of photoresist is applied over the silicon dioxide layer 17' and is patterned to form a mask (not shown). The mask pattern is transferred to the underlying silicon dioxide layer 17' by RIE using suitable feed gases (step SI 2).
Referring in particular to Figure 6n, the resulting structure includes a patterned silicon dioxide layer 17 with contact windows 18l5 182, 183.
Referring to Figure 6o, the mask (not shown) is removed and a layer 19' of a first metallization layer comprising platinum is deposited by physical vapour deposition (PVD) (step SI 3). The metallization is annealed using one or more rapid thermal anneals (RTA). Unreacted platinum is removed using aqua regia. A second metallization layer comprising aluminium is deposited by PVD having a thickness of about 10,000 A.
A layer (not shown) of photoresist is applied over the metallization 19' and is patterned to form a mask (not shown). The mask pattern is transferred to the underlying metallization 19' by RIE using suitable feed gases (step S14).
Referring in particular to Figure 6p, the resulting structure comprises metallization contacts 19l3 192, 193. It will be appreciated that layer thicknesses and process parameters can be varied and optimised and that the fabrication process can include other steps, such as substrate thinning and depositing a metallization on the back side of the substrate. Referring to Figure 8, another embodiment of a power semiconductor device 1 ' in accordance with the present invention is shown. The device 1 ' is shown in simplified form, for example, as can be used in numerical simulations. The device 1 ' is substantially the same as the device 1 (Figure 1) hereinbefore described, but differs in that it includes a field stop region 37 and has different dimensions.
The field stop region 37 takes the form of a shallow n-type diffusion well disposed at the upper surface of the epitaxial layer 3 under the field oxide 10 between p-type body region 12 and the p-type injector 131. The field stop region helps to restrict the depletion layer width laterally and improve performance of the bipolar transistor 32 (Figure 3b). This allows the lateral dimension of the device 1 ' to be reduced, in particular, a shorter separation length, L, of about 4 μπι. The depth of the device, d2, can have a value in a range of about 8 μπι to 10 μπι, although the depth can be higher or lower depending on required operating voltage. The half-cell width of the device, w2, can be about 7 μπι to 11 μπι. Again, the width, w2, can be larger or smaller depending on operating voltage.
Referring to Figure 9, yet another embodiment of a power semiconductor device 1 " in accordance with the present invention is shown. Referring to Figures 8 and 9, the device 1 " is substantially the same as the device 1 ' hereinbefore described, but differs in that the position of the first injector 13a is moved away from the -type body region 12 and does not extend to the edge of the oxide 10, 17. The second injector 132 may be omitted. Thus, the positions of the p-n junction 2Qi and the Schottky diode 21 are swapped.
The device 1 " is fabricated using a modified mask for the p-type implant at step S10 above. Operation the device 1 " is substantially the same as device 1 '. In particular, there is no detrimental effect on blocking voltage. However, the onset of injection occurs at a slightly higher source-drain voltage. For example, for the device 1 ' shown in Figure 8, simulation shows that hole injection starts at voltage of about 1.7 V at a temperature of 300 K and, for the device 1 " shown in Figure 9, simulation shows that hole injection starts at voltage of about 1.9 V at a temperature of 300 K. This is thought to be due to the slightly increased distance from the channel to the injector Referring to Figure 10, another embodiment of a power semiconductor device 1 "' in accordance with the present invention is shown.
The device 1 "' is similar to the devices hereinbefore described except that the device 1 "' employs a vertical injector rather than a lateral injector. Thus, the injectors 13 , 132' and Schottky diode 20' can be placed at a lower surface 38 of the epitaxial layer 3 on the underside of a thinned substrate (not shown).
The Schottky diode 20' comprises a pad 39 directly in contact with the epitaxial layer 3 adjacent to or between p-type well(s) 13/, 132'. The pad 39 is formed by a layer of a first metallization, such as platinum silicide or other metallization providing a barrier height of at least 0.7 eV.
The drain contact comprises a layer 40 of a metallization in contact with the epitaxial layer 3 which forms an ohmic contact to the epitaxial layer. The metallization is preferably alloyed or annealed. The metallization may comprise aluminium or aluminium silicide. The layer 40 may run over the pad 39.
Alternatively, it may laterally abut the pad 38. In some embodiments, the layers 39, 40 may not join or overlap, but may be connected in some other way, e.g. by a bond wire or high-temperature solder.
The device 1 "' having a vertical injector operates in a similar way to the devices having a lateral injector. However, one difference is that the injector uses vertical biasing. The biasing depends on the amount of electron charge in the area immediately above the Schottky diode 21 '.
Using the vertical injector, there is no detrimental affect to one-dimensional (that is, doping-dependent) blocking voltage.
The blocking voltage performance is set by the drift region thickness, d3, from the bottom of the p-body region 12 to the drain. In some embodiments, d3 =
Neighbouring p-body regions 12 provide mutual protection to the radius of curvature, thus minimising the electric field in that area.
The device 1 "' has the advantage that, due to the position of the injector, no separate bond wire or connection is required from front to back face. It may also allow a heat sink (not shown) to be placed more easily on top of the source metallization to help dissipate heat.
The placement of the injector can allow a smaller cell width to be used and so reduce the drain-source specific resistance per unit area in a pre-injection state and in an injection state compared to the embodiments using a lateral injector.
Furthermore, injector elements can be evenly distributed across the underside of the wafer and so help to ensure an even current share across the resultant device area once injection begins. Moreover, the use of electrons available from two voltage-controlled, gated MOS channels per injector, instead of a single channel per injector, as in transistor 1 (Figure 1) allows for an increased current saturation level.
As described earlier, a Schottky-biased injector can be used to reduce the ON resistance of a vertical MOSFET. The blocking voltage of the hybrid device depends on the thickness and doping concentration of the drift region 3, as well as the radius of curvature of the p-type body 12. One way to increase the blocking voltage in a conventional power MOSFET is to increase the doping density of the n-type drift region and include charge
compensating structure comprising regions of p-type material on either side of the drift region as described, for example, in "COOLMOS™-a new milestone in high voltage power MOS" by L. Lorenz, G. Deboy, A. Knapp and M. Marz, Proceedings of the 11th International Symposium on Power Semiconductor Devises and ICs, pages 3 to 10 (1999). Reference is also made to "Theory of Semiconductor
Superjunction Devices" by T. Fujihara, Japanese Journal of Applied Physics, volume 36, part 1, number 10, pages 6254 to 6262 (1997) which introduces the use of alternatively stacked p- and n-type, heavily doped thin semiconductor layers.
However, such charge compensation structures are unsuitable for a hybrid device since the charge compensation structure would restrict lateral flow of majority carriers to the injector. However, a modified form of charge compensation structure can be used not only to allow the flow of majority carriers to the injector, but also to direct their flow. Thus, the modified charge compensation structure can be adapted to control the voltage at which IGBT-like action occurs. Referring to Figures 11 to 16, a further embodiment of a power semiconductor device 101 in accordance with the present invention is shown. The semiconductor device 101 takes the form of a hybrid silicon n-channel vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor device comprising a combined vertical and lateral charge balance structure 131. The device 101 is also hereinreferred to as a "super junction device".
The super junction device 101 includes a heavily-doped n-type monocrystalline silicon substrate 102 which functions as a drain region. An epitaxial layer 103 of silicon is arranged on an upper surface 104 of the substrate 102. As will be explained later, the epitaxial layer 103 comprises p-type and n-type regions which provide the charge balance structure 131. The structure of super junction device 101 at the upper surface 107 of the epitaxial layer 103 is generally the same as that of the device 1 shown in Figures 1 and l a.
A field oxide 105l 5 052 is located at the upper surface 107 of the epitaxial layer 103 and has first, second and third windows (not shown) defining first, second and third laterally-separated portions of upper surfaces 107 of the epitaxial layer 103.
In the following description, a first half of a full cell is described. A first gate oxide 108t is provided on the first portion of the upper surface 107 of the epitaxial layer 103. The first gate oxide 108t runs along the first portion of the upper surface 107 of the epitaxial layer 103 and abuts the first field oxide 1051. A first layer of heavily-doped n-type polycrystalline silicon 1 10a is disposed on the first gate oxide 1081.
A first p-type body 1 12t comprising a lightly-doped p-type diffusion well is disposed within the epitaxial layer 103 at the first portion of the upper surface 107. The first p-type body 1 12t extends laterally under a first spacer and the first gate oxide 108! .
First and second p-type injectors 1 13l 5 1 32 in the form of heavily-doped p-type diffusion wells are located within the epitaxial layer 103 at the second portion of the upper surface 107. The p-type injectors 1 13l 5 1 32 are spaced apart to leave a region 1 14 of n-type epitaxial layer 103 extending up to the second portion of the upper surface 107 of the epitaxial layer 103. The injectors 1 13l 5 1 132 straddle the centre of the cell.
A first n-type source region 1 15l in the form of a heavily-doped n-type diffusion well is disposed within the first p-type body 1 12l at the upper surface 107. A p-type body short (not shown) in the form of a heavily-doped p-type diffusion well is also provided at the first upper surface. A layer (not shown) of silicon dioxide runs over the first gate poly 10 and the first field oxide 105 and has windows (not shown). A first metallisation layer 119, provides a first source/emitter terminal SI, a second metallisation layer (not shown) provides a first gate terminal Gl and a third metallisation layer 1 93 provides a collector/injector terminal I. The metallization layers 9,, 1193 each comprise a bi- layer comprising a high-barrier metal silicide base layer comprising, for example, platinum silicide, and a high-conductivity overlayer comprising, for example, aluminium. Similar to the device 1 shown in Figures 1 and la, the third metallisation layer 1 93 is in contact with the first and second p-type injectors 3,, 1 32 and the n-type region 114 between the p-type injectors 3,, 1 32. The p-type injectors 3,, 1132 form respective p-n junctions 145,, 1452 (Figures 17a and 17b) with the epitaxial layer 103 and the metallisation layer 1 93 forms a Schottky diode 146 (Figures 17a and 17b) with the epitaxial layer 103. Thus, the pair of p-n junctions (not shown) and the Schottky diode (not shown) are arranged in parallel between the
metallisation layer 1 93 and the epitaxial layer 103.
The high-barrier metal silicide forms a Schottky diode with the epitaxial layer 103, but forms ohmic contacts to the heavily-doped injector and source regions 3,, 1132, 115,.
The second half of the cell is a mirror image of the first half and includes a second field oxide 1052, a second gate oxide 1082, a second p-type body 22, a second source region 1152, a second source/emitter terminal S2 and a second gate terminal G2. However, as explained earlier, the first and second injectors 3,, 1 32 straddle the centre of the cell and injector I is shared by both halves of the cell.
The epitaxial layer 103 comprises alternating stripes (or "walls" or "columns") of p- type and n-type regions 132,, 1322, 133,, 1332, 34,, 1342, 135 extending up from the substrate 102 (i.e. along the z-axis) and in a direction along the length of the device (i.e. along the y-axis). As will be explained later, these alternating n-type and p-type regions 132l5 1322, 133l5 1332, 134l5 1342, 135 provide a first, vertical charge balance structure.
As shown particularly in Figures 13 and 14, first and second, outer p-type regions 132j, 1322 are disposed under the p-type body regions 112l5 1122. The p-type regions 132l5 1322 extend from the substrate 102 and reach the bottoms of the respective p-type body regions 112l5 1122.
Third and fourth inner p-type regions 134l5 1342 are disposed between the outer p- type regions 132l5 1322.
A first n-type region 133t is formed between the first and third p-type regions 132l5 34j . Likewise, a second n-type region 1332 is formed between the second and fourth p-type regions 1322, 1342. As will be explained later, the first and second n- type regions 133l5 1332 form a vertical path through which majority carriers (i.e. electrons) from the source regions 115l5 H 52 can flow to reach the substrate 102.
A third n-type region 135 is formed between the third and fourth p-type regions 134j, 1342. This forms a voltage sustaining region.
Referring in particular to Figures 11, 13 and 14, the p-type regions 132l5 1322, 134l5 1342 are interconnected by lateral, bridge-like, p-type linking regions 136l5 362 and by a buried p-type cross layer 137 (which collectively may be referred to as "p-type ground links" as the p-type body regions 112l5 1122 are connected to ground). The buried p-type cross layer 137 forms part of a second, lateral charge balance structure
Referring in particular to Figure 13, the third and fourth p-type regions 134l 5 1342 and the p-type cross layer 137 form a core structure generally having the appearance of an inverted parallel flange channel, i.e. an elongate structure which is 'n'-shaped in cross section.
Referring also to Figures 11 and 14, the outer p-type walls 132l5 1322 flank the core structure 134l 5 1342, 137 and are joined to the top of the core structure at intervals along its length by the linking regions 136l5 362 which extend laterally from the p- type cross layer 137 and so form a unitary p-type region, i.e. a single continuous region of p-type semiconductor. The first and second n-type regions 133l5 1332 (herein referred to as the "lower drift regions") are connected by short, vertical, column-line, n-type linking regions 141l5 1412 (which are also referred to as "linking drift regions") to a n-type surface layer 142 (herein referred to as the "upper drift region") and so form a unitary n-type region. The third n-type region 135 does not form part of the unitary n-type region.
The column-like, n-type linking regions 141l5 1412 pass between the p-type linking regions 136l5 362 such that the p-type and n-type unitary regions interlock or interlace. Referring also to Figure 13, in the interlocking arrangement, majority carriers (i.e. electrons) from the source regions 115l5 H 52 can flow into the upper drift region 142. Majority carriers can then flow through the n-type linking drift regions 141 l5 1412 and into the lower drift regions 133l5 1332 towards the drain 102. Majority carriers can also flow through the upper drift region 142 towards the injector 113l5 1132.
Thus, this arrangement allows a hybrid silicon n-channel vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor device to benefit from a charge balanced structure.
In particular, the charge balanced structure allows the thickness of the device (i.e. the dimension along the z-axis) to be reduced because the electric field generated by applying a source-drain voltage (with gate-source voltage of 0 V) is distributed laterally (along the x-direction), thereby driving electrons towards the n-type columns 133l5 332 and holes towards the p-type columns 132l 5 1322, 134l5 1342. A space charge region (not shown) is formed which spreads out along a vertical p-n junction (not shown). The lower drift regions 133l5 1332 are depleted and act like the voltage sustaining layers of a PIN diode structure. The source-drain voltage required to reach a critical electric field is now much higher than a similar device without a charge balance structure. Thus, to obtain the same blocking voltage, the length of the drift region can be reduced, which reduces the resistance of the drift region length. In addition, use of the vertical charge balance structure allows the doping of the n-type drift regions to be increased by about an order of magnitude or more, for example, from about 5xl015 cm-3 to about 5x1016 cm-3. The higher doping concentration results in a further reduction in drift region resistance.
The reduction in the thickness of the device (i.e. along the z-axis) and the lower resistivity can help to significantly reduce the resistance of the lower drift regions, while maintaining or even increasing blocking voltage. The charge balance structure allows the device to break the so-called "silicon limit" in one dimension defined by: sRon « 3.7x10"9. (BV)2'6 (1) where sRon is the specific ON resistance (in Ω.ΟΉ2) and BV is the blocking voltage (in V). Reference is made to "On the Specific On-Resistance of High-Voltage and Power Devices", by R. Zingg, IEEE Transactions on Electron Devices, volume 51, number 3, page 492 (2004) which describes that, for the same level of current and the same active area, forward voltage drop across a charge balanced MOSFET is lower than that across a normal device, thus improving transconductance.
The super junction device 101 has a lateral charge balance structure configured such that a lateral blocking voltage for the lateral insulated-gate bipolar transistor is the same or similar to the vertical blocking voltage for vertical metal-oxide
semiconductor field effect transistor. The lateral charge balance structure helps to shield the radius of curvature of the p-type body regions 112l5 1122. Referring to Figure 17a, a schematic circuit diagram of the super junction device 101 prior to the onset of injection is shown. The device 101 includes a MOSFET 143, a bipolar transistor 144 which includes p-n junctions 145l5 1452 (forming the emitter to base junction) and a Schottky diode 146. In this example, the bipolar transistor 144 is a pnp transistor.
The drain 147 of the MOSFET 143 is connected to the drain D of the transistor 101 through a first, vertical resistive element 148t and to the base 149 of the bipolar transistor 144 via a second, lateral resistive element 1482. The first and second resistive elements 148l5 1482 are joined at node 150 at which electron current branches. The first and second resistive elements 148l 5 1482 have values Rvertical and ^-lateral respectively. As shown in Figure 17a, the Schottky diode 146 has a depletion region 151 having a capacitance Cdep(Sch) which varies with the thickness of the depletion region 151. The capacitance is shown in series with the Schottky diode 146 which has a resistance, RSchottk . The p-n junctions 145l5 1452 also have respective depletion regions 152l 5 1522, each having a capacitance Cdep(p_n).
Figure 17b shows the flow of majority carriers and minority carriers through the transistor 101 which, in this example, are electrons and holes respectively. When a large minority carrier current flows through bipolar transistor 144, a parasitic bipolar transistor 153 may affect operation of the device.
Referring to Figures 14, 17a and 18, as explained earlier, at a low drain voltage (Vj) and a high gate voltage (Vg), the device operates principally as a MOSFET, electrons flow mainly from the source S to the drain D. However, enabled by the p-type ground links 136l5 1362, a small proportion of the electrons flow to the injector 113j, 1132 via a lateral resistance 1482, which are used to bring about IGBT-action.
These electrons are unable to flow to the injector, I, as they are prevented by the potential barriers of the p-n junction within the base-emitter region 145l5 1452, 149 of the bipolar transistor 144 and by the parallel Schottky diode 146. As the drain voltage, Vd, increases, the electron charge (and the potential) in the base 149 is increased. The width of the depletion region 151 associated with the Schottky diode 146 is reduced. As the drain voltage, Vd, increases further, the potential barrier is overcome and thermionic emission of electrons via the Schottky diode 146 starts, thereby biasing the region 149 beneath the p-type injectors 113l5 1 32 below Vd and, thus, forward biasing the p-n junctions 145l 5 1452 within the base-emitter region of the bipolar transistor 144.
Referring to Figure 17b, once the p-n junctions 145l5 1452 are forward-biased, the Schottky diode 146 acts to maintain that bias thereby ensuring that any change in resistance of the second resistive element 1492 due to increased electron
concentration does not de-bias the emitter to base junction. Minority carriers flow via the bipolar transistor 144, p-body regions \ \ 2i,\ \ 22 and body short (not shown) to the source S.
By varying the number and dimensions of p-type ground links 136l5 1362 (Figure 15) to n-type linking regions 141l5 1412 (Figure 15), the charge balance structure offers a way of controlling the ratio of electrons flowing to the substrate 102 and injector 113l, 32 and, thus, set the onset of IGBT-action, as well providing other benefits and features, as will now be described in more detail.
Firstly, the charge balance structure, in particular the p-type cross layer 137 and upper drift region 142, provides a reduced surface field (RESURF)-like distribution of electric field resulting in a reduction in electric field at the surface between the oxide 105l5 1052 and the upper drift region 142. This helps to maintain the one- dimensional (that is, doping-dependent) blocking voltage for the interface between the p-type body region 112l 5 1122 and the upper drift region 142.
Secondly, the lateral distance, L, between the channel and injector can be reduced without the risk of punch-through, as the extent of the depletion region (not shown) from the p-body regions 112l5 1122 can be reduced by increasing the doping concentration in the upper drift region 142. Furthermore, the doping concentration in the upper drift region 142 and the lower drift regions 133l5 1332 can be independently controlled (that is, they can have different doping concentrations). The flow of electrons from the n-type source regionsl l Sj, 1 52 to the injectors 113l5 1132 and the substrate 102 can be independently controlled.
Referring also to Figures 12 and 15, the ratio, R, of the areas (in the x-y plane) of the p-type linking regions 136l5 1362 and the n-type linking regions 141 l5 1412, i.e. (kXm):(kXn), can be set to control the proportion of electrons flowing laterally towards the injectors 113l5 1132 through the upper drift region 142, i.e. set Rj and R2 respectively.
This can be used to tune the device 101 to ensure that the injectors 113l5 1 32 turn on at a given source-drain voltage.
At the onset of injection, minority carriers (i.e. holes) travel to the sources 112l5 1122 via the p-type linking regions 136l5 1362.
In this example, the ratio, R, has a value of about 1 :10. However, the ratio, R, can have a value, for example, between about 1 :5 to about 1 :20. For a value of about 1 :10, the p-type linking regions 136l5 362 do not appreciably inhibit electron flow to the drain.
Referring still to Figures 13, 14, 17a, 17b and 18, since injected holes flowing to the source S via p-type body regions 112l5 1122 mainly take a different path to electrons flowing to the substrate 102, they do not interfere and do not recombine in the region adjacent to the channel. This leads to improved transconductance of the pnp transistor 144 characteristic once conductance begins when the electron-limited current saturated state is removed. In the devices hereinbefore described, this state tends to require further electron charge build up next to the channel to provide a potential barrier to holes before the device fully conducts via the pnp transistor 144. In the charge balanced structure, the ability to control electrons flowing to the injector and to the base 148 of the pnp transistor 144 provides another parameter which can be controlled. Using a low contact resistance for the injector, automatic de-biasing of the pnp transistor 144 can occur which can help to avoid localisation of current and ensure that emitter current is evenly distributed.
Referring in particular to Figures 13, 14, 17a and 17b, the base region 148 of the pnp transistor 144 within the charge balanced structure is fixed by the gap having a thickness, b. The thickness of the gap can be optimised to ensure adequate effective base thickness of the pnp transistor 144 to provide improved levels of pnp emitter current gain for the limited levels of base drive available via the metal-oxide- semiconductor channel. The base thickness, b, is relatively unaffected by any growing depletion width from the p-type body regions 1 12l 5 1122 as the source- drain voltage increases. In contrast, in the device 1 shown in Figures 1 and l a, as the source-drain voltage increases, the base thickness is reduced and so pnp current gain, hFE, increases.
The behaviour of pnp gain in the transistor 1 shown in Figure 1 differs from a conventional pnp device as a conventional device is fabricated with minimum base width, whereas base region within the transistor 1 (Figure 1) is relatively large and results in low gain. The effective base width is reduces as source depletion region extends laterally towards region 131 (Figure 1) and so gain improves with source- drain voltage. In the super junction device 101 , the base width is set by fabrication to have a high gain and the doping in the base region 149 is preferably sufficiently high so as to avoid collector to emitter punch-through.
The thickness of the p-type cross layer 137 can be adjusted for both doping density and vertical thickness, s, along the z-axis. The doping in the p-type cross layer 137 is chosen to balance the doping in the top n-type layer 142. The ability to alter the thickness of the p-type cross layer 137 allows optimisation of the charge balance structure and extent of the depletion from the p-type body regions 112l 5 1122 into n-type regions 141 l 5 1412, 142. Furthermore this can be achieved without
compromising MOSFET performance as the drift length (not shown) need not be increased as the p-type cross layer 137 thickness, s, is increased. Any such increase in thickness, s, reduces the z-dimension of the voltage sustaining, n-type region 135, which is not critical. In the device 1 shown in Figures 1 and l a, the minimum lateral resistance is limited by punch-through thereby setting a minimum distance, L, between the edge 22 of the p-body 12 and the edge 23 of the injector region 131. The minimum lateral resistance also sets a minimum length for the drift region. This is because, if the vertical resistance is reduced, the ratio of lateral resistance to vertical resistance would be reduced below a minimum value for a sufficient proportion of electrons to reach the injector region 131. This can make it difficult to improve the ON state resistance prior to the onset of injection.
Assuming a fixed work function for the Schottky contact, the distance from the edge of the channel 122 to the edge of the injector 1 3a effectively controls the value of source-drain voltage needed bring about the onset of conductivity modulation of the drift region by the injector. Using the charge balanced structure, the lateral distance can be greatly reduced and so provide way to further reduce the injection start-up voltage. Thus, a wider range of injection start-up voltages can be used.
The charge balanced structure allows all of the channel area to participate in injecting electrons into the n-type regions 133l 5 1332 to the drain 102. Holes are restricted to the upper layers bounded by the p-type cross layer 137. This has the effect of reducing the switch off time as the holes can easily reach the p- type body 112l 5 i.e. the collector of the bipolar transistor 144, with minimal travel. Unlike a conventional IGBT, there is no need to wait for recombination since recombination is not the only means of charge removal.
The device 1 shown in Figure 1 has been simulated in an un-clamped inductive switching test circuit and has shown to be free of latch-up. The transistor 101 also has the potential to provide latch-up-free operation as the mechanism for charge removal from the drift region is via the p-type grounding channels 136l5 362, 137 for holes to the p-type body regions 112l5 1122 and via the drift region to the drain 102 for electrons. Even if a parasitic npn bipolar transistor 152 started conducting, then electrons collected in the drift region (which serves as the npn collector) simply find their way to the drain 102 thereby contributing to the MOSFET current. The parasitic npn transistor 152 will therefore switch off when the channel restricts the electron base drive to the pnp transistor 144. This in turn removes any hole current to the source S via the p-type body regions 112l5 H22. Thus, there is no sustained base bias to the bipolar transistor 144 to cause latch-up
The devices 1 (Figure 1), 101 (Figure 11) benefit from greatly improved switch off time similar to that achieved by a conventional unipolar MOSFET. As hereinbefore described, charge storage does not occur in the devices 1 (Figure 1), 101 (Figure 11), unlike conventional IGBTs, which would otherwise have the effect of greatly increasing switch off time. The devices 1 (Figure 1), 101 (Figure 11) therefore benefit from low ON state loss, available from a source-drain voltage of≥ 0.1 V with fast switching capability.
The devices 1 (Figure 1), 101 (Figure 11) also provide a mechanism to "self clamp" when switching inductive loads, thus preventing the device going into avalanche from which recovery times can be long.
The devices 1 (Figure 1), 101 (Figure 11) do not suffer from the so-called "snap- back" phenomenon associated with other devices capable of MOSFET-IGBT hybrid operation, for example, as described in "The Bi-mode Insulated Gate
Transistor (BIGT) a potential technology for higher power applications" by M. Rahimo, A. Kopta, U. Schlapbach, J. Vobecky, R. Schnell and S.Klaka, Proceedings of the 22nd International Symposium on Power Semiconductor Devises and ICs, pages 391 to 394 (2010). In the devices 1 (Figure 1), 101 (Figure 11), the Schottky- biased injector ensures a smooth transition between MOSFET-like and IGBT-like conduction states. Finally, with increasing ambient heat-sink temperatures, for example up to 400 K, the potential barrier between the p-type and n-type regions will decrease. This is because, as the device heats up, bipolar conduction is activated at a lower source- drain voltage. A reduction in ON state voltage with increasing temperature is helpful in a device experiencing surge current.
The device 101 is fabricated in a similar to way to the device 1 shown in Figures 1 and la. However, instead of growing an epitaxial layer of lightly-doped n-type
monocrystalline silicon (step SI, Figures 6a and 7), an epitaxial layer of
monocrystalline silicon having appropriate n-type and p-type regions for providing the charge balance structure is formed.
Referring to Figures 19a to 19d and 20, a method of fabricating an epitaxial layer for providing the charge balance structure will now be described.
Figure 19a shows a highly-doped n-type monocrystalline substrate 103.
P-type and n-type columns 132/, 133/, 134/, 135, 132/, 133/, 134/ are formed (step S1A).
The columns can be formed 132/, 133/, 134/, 135, 132/, 133/, 134/ by growing a plurality of epitaxial layers (not shown) of lightly doped n-type, monocrystalline silicon and, for each epitaxial layer, doping regions of the epitaxial layer with acceptors to compensate for the n-type doping and to provide excess acceptors to dope the regions p-type. For example, this may be achieved by masking regions of an epitaxial layer (not shown), doping unmasked regions of the using an acceptor, such as boron. The n-type layers (not shown) are doped to a concentration of about 5x1016 cm-3. The p-type regions are doped to a net concentration of about 5xl016 cm-3. Alternatively, the columns can be formed by growing layers of lightly doped p-type silicon may be deposited and, for each epitaxial layer, doping regions of the epitaxial layer with donors. In some embodiments, the columns can be formed by growing undoped layers and, for each epitaxial layer, selectively doping respective regions of the epitaxial layer with acceptors and donors. Doping can be performed by ion implantation. The structure is annealed to activate the implants. The process can be repeated for each epitaxial layer until the columns can have a suitable thickness, d3, e.g. between about 10 and 50 μπι. Between 2 and 10 or more epitaxial layers can be used.
Alternatively, the columns 132 , 133 , 134 , 135, 1322', 1332', 134/ can be formed by growing an epitaxial layer of lightly-doped n-type monocrystalline silicon. The epitaxial layer (not shown) can be doped in-situ during growth with a donor to a concentration of about 5xl016 cirf3 and has a thickness of between about 10 and 50 μιη. A dielectric layer (not shown) is formed on the epitaxial layer (not shown) and it patterned to form an etch mask. Then, trenches (not shown) are formed by dry etching which reach the substrate 102. The trenches (not shown) can be filled by epitaxially growing lightly-doped p-type monocrystalline silicon. In some embodiments, more than one epitaxial layer can be deposited, etched and filled.
As shown in Figure 19b, the outer, p-type columns 132/, 132/ have length (along the x-axis), ql5 of between about 0.5 μπι to about 5 μπι. The n-type columns 133/, 133/ can have a length, q2, of between about 0.5 μπι to about 5 μπι. The inner p- type columns 134/, 134/ can have a length, q3, of between about 0.5 μπι to about 5 μιη. The central, n-type column 135 can have a length of about 0.5 μπι to about 30 μπι. Referring in particular to Figure 19c, the p-type cross layer 137 and the rest of the outer p-type columns 132l5 1322, the p-type linking regions 136l 5 1362 (Figure 14), the n-type linking regions 141 l5 1412 (Figure 13) are formed (step SIB). A similar process to one of those described earlier for forming the columns can be used. The regions are doped to a concentration of about 5x1016 cirf3. The thickness, s, of the p-type cross layer 137 can be between about 0.5 μπι to about 5 μπι.
Referring in particular to Figure 19d, a layer 154 of lightly doped n-type
monocrystalline silicon is epitaxially grown (step SI C). The epitaxial layer 154 can be doped in-situ during growth with a donor to a concentration of about 5x1016 cm and can have a thickness, t, of between about 1 and 2 μπι.
The values of lengths ql 5 q2, q3, q4 and values of thicknesses d3, s, b and t can be found by routine experiment and/ or by computer simulation using, for example, a bespoke or commercially-available Poisson solver. Different p-type and n-type doping concentrations may be used. The doping concentrations can differ between the lower drift regions 133l5 1332 and the upper drift region 142. The doping concentrations can differ between the inner p-type columns 134l5 1342 and the buried p-type cross layer 137.
The rest of the device can be fabricated in substantially the same way as that described earlier using steps S3 to S14 (Figure 7). It will be appreciated that many modifications may be made to the embodiments hereinbefore described.
The semiconductor device may be a p-channel device. Thus, the drift region, the body region, the source region and injector regions may take the form of a lightly - doped p-type semiconductor material, lightly-doped n-type semiconductor material, heavily-doped p-type semiconductor material and heavily-doped n-type
semiconductor material respectively. Thus, the majority carriers may be holes instead of electrons and the minority carriers may be electrons instead of holes. The semiconductor device may be based on silicon carbide. For example, the substrate may comprise 4H or 3C polytypes. The drift region and the body, source and injector regions may comprise silicon carbide. Other elemental or compound semiconductors, such as GaN, can be used. Other metallizations can be used to provide the high Schottky barrier.
A single continuous metallization forming a Schottky contact and an ohmic contact to the p-n junctions need not be used. Separate layers of metallization (which may comprise different materials) can be used and may be connected, for example, by an overlayer or by a bond wire.
Processing steps and process parameters may be altered and optimised. For example, doping concentrations, dopant ions, ion-implant energies can be varied. Methods of deposition and methods of etching can also be varied. Also, dielectric materials can be altered. For example, silicon dioxide need not be used and can be replaced, for example, by silicon oxynitride or high-k dielectrics. A gate metallization may be used instead of gate poly, for example, comprising tungsten (W).

Claims

Claims
1. A semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor having a Schottky-biased p-n junction (20^ 45^.
2. A semiconductor device according to claim 1, comprising:
a minority-carrier injector which comprises:
the p-n junction (20^ 145 ; and
a Schottky diode (21 ; 146);
wherein the Schottky diode is arranged in parallel with the p-n junction so as to provide the Schottky-biased p-n junction.
3. A semiconductor device according to claim 2, comprising:
a semiconductor region (3; 133t 141l5 142) disposed between first and second opposite surfaces,
at least a first portion of the semiconductor region comprising a semiconductor of a first conductivity type providing a drift region for majority carriers in a direction between the first and second surfaces,
at least a second portion of the semiconductor region providing a body region (12; 112a) comprising a semiconductor of a second, opposite conductivity type disposed at the first surface of semiconductor region; and
at least a third portion of the semiconductor region providing a source region (15; 5^ comprising a semiconductor of the first conductivity type disposed at the first surface of semiconductor region and within the body region.
4. A semiconductor device according to claim 3 wherein the minority-carrier injector is disposed at the first surface of the semiconductor region laterally separated from the body region.
5. A semiconductor device according to claim 4, wherein at least a fourth portion of the semiconductor region comprises a semiconductor of the second conductivity type disposed at the first surface of the semiconductor region with a semiconductor of the first conductivity type so as to form a p-n junction.
6. A semiconductor device according to claims 4 or 5, wherein the Schottky diode (21 ; 146) comprises a metallization in contact with the first surface of semiconductor region.
7. A semiconductor device according to claim 6, wherein the Schottky diode (21 ; 146) has a barrier height equal to or greater than 0.7 eV
8. A semiconductor device according to any one of claims 3 to 7, further comprising a gate structure disposed on the first surface of the semiconductor region and configured to control flow of carriers from the source region into drift region.
9. A device according to claim 1, comprising:
a semiconductor substrate (2; 102) of a first conductivity type;
a drift region (3; 133t 141l5 142) comprising a semiconductor of the first conductivity type disposed on the semiconductor substrate;
a body region (12; 121) comprising a semiconductor of a second, opposite conductivity type;
a source region (15; 5 comprising a semiconductor of the first conductivity type disposed within the body region;
the device further comprising:
a minority-carrier injector which comprises:
an injector region (13^ H3j) comprising a semiconductor of the second conductivity type disposed within the drift region and laterally separated from the body region, the injector region and the drift region forming a p-n junction (20^ 145j); and
a metallic electrode ( 93; 1193) in contact with the drift region laterally adjacent and electrically connected to the injector region, the metallic electrode and the drift region forming a Schottky diode (21 ; 146) in parallel with the p-n junction to provide the Schottky-biased p-n junction.
10. A semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor, the device comprising:
a semiconductor substrate (2; 102) of a first conductivity type;
a drift region (3; 1331 1411, 142) comprising a semiconductor of the first conductivity type disposed on the semiconductor substrate;
a body region (12; 121) comprising a semiconductor of a second, opposite conductivity type;
a source region (15; 115^ comprising a semiconductor of the first
conductivity type disposed within the body region;
the device further comprising:
a minority-carrier injector which comprises:
an injector region (13^ 113 comprising a semiconductor of the second conductivity type disposed within the drift region and laterally separated from the body region, the injector region and the drift region forming a p-n junction (20^ 145 ; and
a metallic electrode ( 93; 1193) in contact with the drift region laterally adjacent and electrically connected to the injector region, the metallic electrode and the drift region forming a Schottky diode (21) in parallel with the p-n junction.
11. A device according to claim 9 or 10, wherein the body region (12) and the substrate (2) are vertically separated by a first given distance and the body region (12) and the injector region (13a) are laterally separated by a second given distance equal to or greater than the first given distance.
12. A device according to claim 11 , wherein the second given distance is about 6
13. A device according to claim 9 or 10, further comprising:
a depletion reducing region (35) comprising a semiconductor of the first conductivity type disposed within the drift region (3), more highly doped than the drift region and disposed between the body region (12) and the injector region (13 .
14. A device according to claim 13, wherein the body region (12) and the substrate (2) are vertically separated by a first given distance and the body region (12) and the injector region (13 are laterally separated by a second given distance less than the first given distance.
15. A device according to any one of claims 9 to 14, wherein the Schottky diode (21 ; 146) has a barrier height equal to or greater than 0.7 eV.
16. A device according to any one of claims 9 to 15, wherein the metallic electrode (193; 1193) includes a layer of platinum silicide forming the Schottky diode (21 ; 146).
17. A device according to any one of claims 9 to 16, wherein the metallic electrode (193; 1193) and the semiconductor substrate (2) are coupled so as to be at the same potential.
18. A device according to any one of claims 9 to 17, wherein the drift region (3; 133j 14 17 142) comprises silicon.
19. A device according to any one of claims 9 to 17, wherein the drift region (3; 133j 141 j, 142) comprises silicon carbide semiconductor.
20. A device according to any preceding claim, further comprising:
a charge balance structure (131).
21. A device according to claim 20, wherein the charge balance structure (131) comprises:
first and second charge balance regions (132l5 134t) comprising a semiconductor of the second conductivity type, the first charge balance region
(132j) extending vertically between the substrate and the body region ( 121) and the second charge balance region (134j) laterally spaced between the p-n junction (113a) and the first charge balance region (132 , such that a vertical drift region (133 comprising a semiconductor of the first conductivity type is disposed between the first and second charge balance regions.
22. A device according to claim 21, wherein the charge balance structure (131) further comprises:
a third charge balance region (137) comprising a semiconductor of the second conductivity type disposed under the p-n junction (145j) and extending laterally to the second charge balance region ( 34a), wherein a lateral drift region (142) comprising a semiconductor of the first conductivity type is disposed over the third charge balance region.
23. A device according to claim 22, wherein the charge balance structure (131) further comprises:
a region ( 36 comprising a semiconductor of the second conductivity type linking the first and second charge balance regions (132l5134j), the region configured to leave a region (141 j) comprising a semiconductor of the first conductivity type linking the lateral and vertical drift regions.
24. A semiconductor device having an integrated vertical and lateral charge balance structure.
25. A semiconductor device according to claim 24, wherein the semiconductor device comprises:
a planar substrate ( 02);
a first conductive region (133 of a first conductivity type extending perpendicularly to the substrate between first and second ends and having at least first and second sides;
first and second charge balance regions (132l51341) of second conductivity type disposed on the first and second sides respectively of the first conductive region;
a second conductive region (142) of the first conductivity type extending parallel to the substrate between first and second ends and having at least a first side; a third charge balance region (137) of the second conductivity type disposed on the first side of the second conductive region;
the device further comprising:
at least one region (141 of the first conductivity type linking the first and second conductive regions so as to form a continuous region of the first
conductivity type between the first end of the first conductive region to the second end of the second conductive region; and
at least one region ( 36^ of the second conductivity type linking the first, second and third charge balance regions so as to form an integrated charge balance structure.
26. An integrated circuit comprising at least one semiconductor device according to any preceding claim.
27. A method of fabricating a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor, the method comprising:
providing a Schottky-biased p-n junction.
28. A method according to claim 27, further comprising:
providing a drift region of first conductivity type on a semiconductor substrate of the first conductivity type;
forming a body region of a second, opposite conductivity type within the drift region;
forming a source region of the first conductivity type within the body region; wherein providing the Schottky-biased p-n junction comprises:
forming an injector region of the second conductivity type within the drift region, laterally separated from the body region, wherein the injector region and the drift region provide a p-n junction; and
providing a metallic electrode in contact with the drift region laterally adjacent and electrically connected to the injector region, wherein the metallic electrode and the drift region provide a Schottky diode in parallel with the p-n junction.
29. A method of fabricating a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a lateral insulated gate bipolar transistor, the method comprising:
providing a drift region of first conductivity type on a semiconductor substrate of the first conductivity type;
forming a body region of a second, opposite conductivity type within the drift region;
forming a source region of the first conductivity type within the body region; forming an injector region of the second conductivity type within the drift region, laterally separated from the body region, wherein the injector region and the drift region provide a p-n junction; and
providing a metallic electrode in contact with the drift region laterally adjacent and electrically connected to the injector region, wherein the metallic electrode and the drift region provide a Schottky diode in parallel with the p-n junction.
30. A method according to claim 29, wherein the body region and the substrate are vertically separated by a first given distance and wherein forming the injector region within the drift region comprises laterally separating the injector region from the body region by a second given distance exceeding the first given distance.
31. A device according to claim 29, further comprising:
a depletion reducing region comprising a semiconductor of the first conductivity type disposed within the drift region, more highly doped than the drift region and disposed between the body region and the injector region.
32. A method according to claim 31, wherein the body region and the substrate are vertically separated by a first given distance and the body region and the injector region are laterally separated by a second given distance less than the first given distance.
33. A method according to any one of claims 27 to 32, further comprising:
providing a charge balance structure.
34. A method of fabricating a semiconductor device, the method comprising: providing an integrated vertical and lateral charge balance structure.
35. A method according to claim 34, wherein providing the integrated vertical and lateral charge balance structure comprises:
providing a planar substrate;
providing a first conductive region of a first conductivity type extending perpendicularly to the substrate between first and second ends and having at least first and second sides;
providing first and second charge balance regions of second conductivity type disposed on the first and second sides respectively of the first conductive region;
providing a second conductive region of the first conductivity type extending parallel to the substrate between first and second ends and having at least a first side;
providing a third charge balance region of the second conductivity type disposed on the first side of the second conductive region;
providing at least one region of the first conductivity type linking the first and second conductive regions so as to form a continuous region of the first conductivity type between the first end of the first conductive region to the second end of the second conductive region; and
providing at least one region of the second conductivity type linking the first, second and third charge balance regions so as to form an integrated charge balance structure
36. A method of operating a semiconductor device operable as a vertical metal oxide semiconductor field effect transistor and a vertical or lateral insulated gate bipolar transistor having a Schottky-biased p-n junction, the metal oxide
semiconductor field effect transistor and insulated gate bipolar transistor sharing a source region and a drain region and wherein the Schottky-biased p-n junction has a terminal, the method comprising grounding the source region and applying given bias(es) to terminal and the drain region.
37. A method according to claim 36, wherein the drain region comprises a semiconductor substrate of a first conductivity type or metallic region, and the device further comprises a drift region of the first conductivity type disposed on the semiconductor substrate or metallic region, a body region of a second, opposite conductivity type disposed within the drift region, a source region of the first conductivity type disposed within the body region, an injector region of the second conductivity type disposed within the drift region and laterally separated from the body region, the injector region and the drift region forming a p-n junction and a metallic electrode in contact with the drift region laterally adjacent and electrically connected to the injector region, the metallic electrode and the drift region forming a Schottky diode in parallel with the p-n junction.
PCT/GB2011/051413 2010-08-02 2011-07-25 Semiconductor device WO2012017227A1 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
GB1012951.8A GB2482479B (en) 2010-08-02 2010-08-02 Semiconductor device
GB1012951.8 2010-08-02
GBGB1107979.5A GB201107979D0 (en) 2010-08-02 2011-05-13 Semiconductor device
GB1107979.5 2011-05-13

Publications (1)

Publication Number Publication Date
WO2012017227A1 true WO2012017227A1 (en) 2012-02-09

Family

ID=42799452

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2011/051413 WO2012017227A1 (en) 2010-08-02 2011-07-25 Semiconductor device

Country Status (2)

Country Link
GB (2) GB2482479B (en)
WO (1) WO2012017227A1 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2016129192A (en) * 2015-01-09 2016-07-14 株式会社デンソー Semiconductor device
CN111403385A (en) * 2020-03-02 2020-07-10 电子科技大学 RC-L IGBT device with embedded Schottky diode
GB202019454D0 (en) 2020-12-10 2021-01-27 Chancellor Masters And Scholars Of The Univ Of Oxford Method for purifying virus
CN113270476A (en) * 2021-04-08 2021-08-17 西安电子科技大学 Lateral insulated gate bipolar transistor with electronic control gate region and Schottky anode and manufacturing method thereof
CN113451296A (en) * 2020-03-24 2021-09-28 立锜科技股份有限公司 Power element with lateral insulated gate bipolar transistor and manufacturing method thereof
WO2024027574A1 (en) * 2022-08-03 2024-02-08 无锡华润上华科技有限公司 Shorted anode lateral insulated gate bipolar transistor model and modeling method therefor

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6362925B2 (en) * 2014-05-30 2018-07-25 三菱電機株式会社 Silicon carbide semiconductor device and method for manufacturing silicon carbide semiconductor device
CN104810283B (en) * 2015-05-13 2019-03-12 国网智能电网研究院 A kind of igbt chip manufacturing method suitable for compression joint type encapsulation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030011039A1 (en) * 2001-07-03 2003-01-16 Dirk Ahlers Semiconductor component with a charge compensation structure and associated fabrication methods
US20040212011A1 (en) * 2003-04-24 2004-10-28 Sei-Hyung Ryu Silicon carbide mosfets with integrated antiparallel junction barrier schottky free wheeling diodes and methods of fabricating the same
US20060211189A1 (en) * 2005-02-28 2006-09-21 Infineon Technologies Austria Ag Method for producing a buried semiconductor layer
US20060284248A1 (en) * 2005-06-20 2006-12-21 Kabushiki Kaisha Toshiba Semiconductor device and method of fabricating the same
WO2008035134A1 (en) * 2006-09-22 2008-03-27 Freescale Semiconductor, Inc. Semiconductor device and method of forming a semiconductor device

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4989058A (en) * 1985-11-27 1991-01-29 North American Philips Corp. Fast switching lateral insulated gate transistors
US4901127A (en) * 1988-10-07 1990-02-13 General Electric Company Circuit including a combined insulated gate bipolar transistor/MOSFET
JPH0716009B2 (en) * 1988-12-02 1995-02-22 株式会社日立製作所 Lateral insulated gate bipolar transistor
JPH042169A (en) * 1990-04-19 1992-01-07 Fuji Electric Co Ltd Horizontal type conductivity modulation semiconductor device
JPH06104443A (en) * 1992-09-22 1994-04-15 Fuji Electric Co Ltd Semiconductor device
KR0152345B1 (en) * 1995-06-14 1998-10-01 김광호 Hybrid schottky injection field effect transistor

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030011039A1 (en) * 2001-07-03 2003-01-16 Dirk Ahlers Semiconductor component with a charge compensation structure and associated fabrication methods
US20040212011A1 (en) * 2003-04-24 2004-10-28 Sei-Hyung Ryu Silicon carbide mosfets with integrated antiparallel junction barrier schottky free wheeling diodes and methods of fabricating the same
US20060211189A1 (en) * 2005-02-28 2006-09-21 Infineon Technologies Austria Ag Method for producing a buried semiconductor layer
US20060284248A1 (en) * 2005-06-20 2006-12-21 Kabushiki Kaisha Toshiba Semiconductor device and method of fabricating the same
WO2008035134A1 (en) * 2006-09-22 2008-03-27 Freescale Semiconductor, Inc. Semiconductor device and method of forming a semiconductor device

Non-Patent Citations (9)

* Cited by examiner, † Cited by third party
Title
DAVID K. Y. LIU, JAMES D. PLUMMER: "A novel trench-injector power device with low ON resistance and high switching speed", IEEE ELECTRON DEVICE LETTERS, vol. 9, no. 7, 1988, pages 321 - 323, XP001293268
DAVID K. Y. LIU, JAMES D. PLUMMER: "Design and analysis of a new conductivity-modulated power MOSFET", IEEE TRANSACTIONS ON ELECTRON DEVICES, vol. 40, no. 2, pages 428 - 438, XP000335348, DOI: doi:10.1109/16.182524
JOHNNY K. O. SIN, C. ANDRE T. SALAMA, LI-ZHANG HOU: "The SINFET - A Schottky injection MOS-gated power transistor", IEEE TRANSACTIONS ON ELECTRON DEVICES, vol. ED-33, 1986, pages 1940 - 1947, XP001293181
L. LORENZ, G. DEBOY, A. KNAPP, M. MARZ, PROCEEDINGS OF THE I I INTERNATIONAL SYMPOSIUM ON POWER SEMICONDUCTOR DEVISES AND TCS, 1999, pages 3 - 10
M. RAHIMO, A. KOPTA, U. SCHLAPBACH, J. VOBECKY, R. SCHNELL, S.KLAKA: "The Bi-mode Insulated Gate Transistor (BIGT) a potential technology for higher power applications", PROCEEDINGS OF THE 22ND INTERNATIONAL SYMPOSIUM ON POWER SEMICONDUCTOR DEVISES AND ICS, 2010, pages 391 - 394
R. ZINGG: "On the Specific On-Resistance of High-Voltage and Power Devices", IEEE TRANSACTIONS ON ELECTRON DEVICES, vol. 51, no. 3, 2004, pages 492, XP011107833, DOI: doi:10.1109/TED.2003.822948
SIN J K O ET AL: "Hybrid Schottky injection MOS-gated power transistor", ELECTRONICS LETTERS, IEE STEVENAGE, GB, vol. 22, no. 19, 11 September 1986 (1986-09-11), pages 1003 - 1005, XP009151680, ISSN: 0013-5194, DOI: 10.1049/EL:19860686 *
T. FUJIHARA: "Theory of Semiconductor Superjunction Devices", JAPANESE JOURNAL OF APPLIED PHYSICS, vol. 36, no. 10, 1997, pages 6254 - 6262, XP000918328, DOI: doi:10.1143/JJAP.36.6254
T. PAUL CHOW, B. JAYANT BALIGA: "A new hybrid VDMOS-LIGBT transistor", IEEE ELECTRON DEVICE LETTERS, vol. 9, 1988, pages 473 - 475

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2016129192A (en) * 2015-01-09 2016-07-14 株式会社デンソー Semiconductor device
CN111403385A (en) * 2020-03-02 2020-07-10 电子科技大学 RC-L IGBT device with embedded Schottky diode
CN113451296A (en) * 2020-03-24 2021-09-28 立锜科技股份有限公司 Power element with lateral insulated gate bipolar transistor and manufacturing method thereof
CN113451296B (en) * 2020-03-24 2023-10-27 立锜科技股份有限公司 Power device with lateral insulated gate bipolar transistor and method of manufacturing the same
GB202019454D0 (en) 2020-12-10 2021-01-27 Chancellor Masters And Scholars Of The Univ Of Oxford Method for purifying virus
CN113270476A (en) * 2021-04-08 2021-08-17 西安电子科技大学 Lateral insulated gate bipolar transistor with electronic control gate region and Schottky anode and manufacturing method thereof
WO2024027574A1 (en) * 2022-08-03 2024-02-08 无锡华润上华科技有限公司 Shorted anode lateral insulated gate bipolar transistor model and modeling method therefor

Also Published As

Publication number Publication date
GB201012951D0 (en) 2010-09-15
GB2482479B (en) 2015-02-18
GB2482479A (en) 2012-02-08
GB201107979D0 (en) 2011-06-29

Similar Documents

Publication Publication Date Title
US10720511B2 (en) Trench transistors and methods with low-voltage-drop shunt to body diode
CN111463277B (en) Semiconductor device with a semiconductor layer having a plurality of semiconductor layers
JP5520215B2 (en) Improved power switching transistor
US6239463B1 (en) Low resistance power MOSFET or other device containing silicon-germanium layer
US7238576B2 (en) Semiconductor device and method of manufacturing the same
US20140264477A1 (en) Vjfet devices
US20220376107A1 (en) Mosfet in sic with self-aligned lateral mos channel
WO2012017227A1 (en) Semiconductor device
US20130313570A1 (en) Monolithically integrated sic mosfet and schottky barrier diode
WO2018237355A1 (en) Vertical rectifier with added intermediate region
JP5687582B2 (en) Semiconductor device and manufacturing method thereof
JP2018152426A (en) Semiconductor device
US6180959B1 (en) Static induction semiconductor device, and driving method and drive circuit thereof
EP1703566A1 (en) MOS device having at least two channel regions
CN113130650A (en) Power semiconductor device and preparation process thereof
CN101512738B (en) Semiconductor device and method of forming the same
US8039906B2 (en) High-voltage metal oxide semiconductor device and fabrication method thereof
CN111584481A (en) Transistor structure for electrostatic protection and manufacturing method thereof
KR100360079B1 (en) Manufacturing method of insulated gate semiconductor device to improve robustness
US20230163167A1 (en) Semiconductor device including a trench gate structure
CN114566537A (en) Semiconductor device with a plurality of semiconductor chips
WO2008099229A1 (en) Semiconductor device and method of forming a semiconductor device

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11741483

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11741483

Country of ref document: EP

Kind code of ref document: A1