IGBT process to produce platinum lifetime control

Claim 2

An improved fabrication process for making a MOS-type insulated gate controlled four-layer power switching device, the process comprising: forming a semiconductor substrate having a first layer of a first dopant type defining a device anode and second layer of a second, opposite-polarity dopant type defining a drain region extending from an upper surface of the substrate toward the first layer; forming an insulative layer on the upper surface of the second layer of the substrate and an insulated gate contact layer on the insulative layer; forming double diffused regions including a body region of the first dopant type and a source region of the second dopant type within the body region, the body region forming two PN junctions with the drain and source regions, respectively spaced apart so as to define a channel region in the body region subjacent the insulated gate contact; forming a source contact alongside the gate contact but spaced insulatively therefrom, the source contact forming an electrical connection to the source region and the body region and a short therebetween and defining a cathode contact for the device; forming a anode contact on the opposite side of the substrate in electrical connection to the first layer; the step of forming the second layer including: forming a first portion contacting the first layer and having a first thickness and a first doping concentration; forming a second portion contacting the second layer and extending to said upper surface to receive said double diffused regions; sizing and doping the second portion to a second thickness and a second doping concentration sufficient to block a predetermined maximum reverse bias voltage; and sizing and doping the first portion to produce a predetermined output impedance (R 0 ) sufficient to resist current flow during forward conduction when a high voltage (V ce ) is across the cathode and anode contacts.

Claim 3

A process according to claim 1 in which the first portion of the second layer is sized and doped to provide an output impedance such that the short circuit withstand current at the end of a short circuit pulse does not exceed 3× the room-temperature-rated continuous anode-to-cathode current of the device.

Claim 4

A process according to claim 1 in which the first portion of the second layer is sized and doped to provide an output impedance such that the short circuit withstand current at the end of a short circuit pulse is in the range of 2× to 3× the room temperature-rated continuous anode to cathode current of the device.

Claim 5

A process according to claim 1 in which the first portion of the second layer is sized and doped to provide an effective output impedance of at least 370 ohm-cm. at 80% of rated anode-to-cathode voltage.

Claim 6

A process according to claim 1 in which the first portion of the second layer is sized and doped so that the output impedance of the device is raised sufficiently to pass a short-circuit withstand test requirement of at least 10 microseconds.

Claim 7

A process according to claim 1 in which the first portion of the second layer is sized to a thickness of at least 20 micrometers.

Claim 8

A process according to claim 6 in which the first portion of the second layer is sized to a thickness in the range of 20-50 micrometers.

Claim 9

A process according to claim 6 in which the first portion of the second layer is doped to a doping concentration greater than the doping concentration of the second portion.

Claim 10

A process according to claim 1 in which the first portion of the second layer is doped to a doping concentration greater than the doping concentration of the second portion but less than 5×10 17 atoms/cm 3 and sized to a thickness of at least 20 micrometers.

Claim 11

A process according to claim 9 in which the first portion of the second layer is doped to a doping concentration in the range of 5×10 15 to 9×10 16 atoms/cm 3 .

Claim 12

A process according to claim 9 in which the second portion of the second layer is doped and sized to a doping concentration of about 10 14 atoms/cm 3 and a thickness just sufficient to block the maximum reverse bias voltages for which the device is designed.

Claim 13

A process according to claim 1 in which the first portion of the second layer is doped and sized to a doping concentration in a range of 5 to 9×10 16 atoms/cm 3 and a thickness in the range of 30 micrometers plus or minus 20%.

Claim 14

A process according to claim 1 in which the first portion of the second layer is formed in two subportions including a first subportion and a second subportion each having a doping concentration greater than the doping concentration of said second portion, the first subportion having a greater doping concentration than the second subportion.

Claim 15

A process according to claim 13 in which the second subportion has a greater thickness than the first subportion.

Claim 16

A process according to claim 14 in which the first subportion has a thickness of about 5 micrometers and the second subportion has a thickness of 20 to 50 micrometers.

Claim 17

A process according to claim 13 in which the first subportion has a doping concentration of at least 10 17 atoms/cm 3 and the second subportion has a doping concentration in the range of 1×10 15 to 5×10 16 atoms/cm 3 .

Claim 18

A process according to claim 16 in which the first subportion has a thickness less than 10 micrometers and the second subportion has a thickness of at least 20 micrometers.

Claim 19

A process according to claim 1 including depositing a predetermined dose of transition metal suitable for minority carrier lifetime control and then diffusing the transition metal atoms throughout the substrate.

Claim 20

A process according to claim 18 including controlling the dose of the transition metal to produce a concentration sufficient to effect lifetime control without substantially increasing leakage current of the device.

Claim 21

A process according to claim 18 in which the transition metal is platinum and the depositing step includes controlling the deposition of transition metal to deposit a maximum dose of 1×10 16 atoms/cm 2 or less.

Claim 22

An improved fabrication process for making a MOS-type insulated gate controlled four-layer power switching device, the process comprising: forming a semiconductor substrate having a first layer of a first dopant type defining a device anode and second layer of a second, opposite-polarity dopant type defining a drain region extending from an upper surface of the substrate toward the first layer; forming an insulative layer on the upper surface of the second layer of the substrate and an insulated gate contact layer on the insulative layer; forming double diffused regions including a body region of the first dopant type and a source region of the second dopant type within the body region, the body region forming two PN junctions with the drain and source regions, respectively spaced apart so as to define a channel region in the body region subjacent the insulated gate contact; forming a source contact alongside the gate contact but spaced insulatively therefrom, the source contact forming an electrical connection to the source region and the body region and a short therebetween and defining a cathode contact for the device; forming a anode contact on the opposite side of the substrate in electrical connection to the first layer; and performing a minority carrier lifetime control procedure including: providing a transition metal having a deep level in silicon suitable for recombination; determining a maximum dose of the selected transition metal that can be fully dissolved into the substrate at a temperature in a range between a eutectic temperature of the substrate and an annealing temperature of the substrate; depositing a predetermined dose of the transition metal less than the maximum dose and then diffusing the metal atoms throughout the substrate at a temperature within said range.

Claim 23

An improved fabrication process for making a MOS-type insulated gate controlled four-layer power switching device, the process comprising: forming a semiconductor substrate having a first layer of a first dopant type defining a device anode and second layer of a second, opposite-polarity dopant type defining a drain region extending from an upper surface of the substrate toward the first layer; forming an insulative layer on the upper surface of the second layer of the substrate and an insulated gate contact layer on the insulative layer; forming double diffused regions including a body region of the first dopant type and a source region of the second dopant type within the body region, the body region forming two PN junctions with the drain and source regions, respectively spaced apart so as to define a channel region in the body region subjacent the insulated gate contact; forming a source contact alongside the gate contact but spaced insulatively therefrom, the source contact forming an electrical connection to the source region and the body region and a short therebetween and defining a cathode contact for the device; forming a anode contact on the opposite side of the substrate in electrical connection to the first layer; and performing a minority carrier lifetime control procedure including: providing a transition metal having a deep level in silicon suitable for recombination; determining a maximum dose of the selected transition metal that can be fully dissolved into the substrate at a temperature in a range between a eutectic temperature of the substrate and an annealing temperature of the substrate; depositing a predetermined dose of the transition metal less than the maximum dose and then diffusing the metal atoms throughout the substrate at a temperature within said range.

Claim 24

A process according to claim 21 including determining and depositing a dose of the transition metal less than the maximum dose sufficient to effect lifetime control without substantially increasing leakage current of the device.

Claim 25

A process according to claim 21 in which the depositing step includes ion implantation of a controlled dose of the transition metal less than the maximum dose.

Claim 26

A process according to claim 21 in which the depositing step includes forming a silicide layer of the transition metal on the substrate and stripping away excess deposited transition metal to produce a dose of the transition metal less than the maximum dose.

Claim 27

A process according to claim 21 in which the diffusion temperature is sufficiently high that solid solubility of the transition metal exceeds the implanted dose of the metal in silicon.

Claim 28

A process according to claim 25 in which the transition metal is platinum, the maximum dose is 1×10 16 atoms/cm 2 , and the diffusion temperature is in a range of 830° C. to 1000° C.

Claim 29

A process according to claim 26 in which the deposited dose of platinum is in the range of 1×10 13 to 1×10 16 atoms/cm 2 .

Claim 30

A process according to claim 26 in which the diffusion temperature is 870° C. to 900° C.

Claim 31

In a fabrication process for making a semiconductor power device having at least one PN junction, an improved minority carrier lifetime control process comprising: selecting a transition metal having a deep level in silicon suitable for recombination; determining a maximum dose of the selected transition metal that can be fully dissolved into the substrate at a temperature in a range between a eutectic temperature of the substrate and an annealing temperature of the substrate; determining and depositing a dose of the transition metal not exceeding the maximum dose sufficient to effect lifetime control without substantially increasing leakage current of the device; and diffusing the metal atoms in the substrate at a temperature within said range; the depositing step including forming a silicide layer of the transition metal on the substrate and stripping away excess deposited transition metal to produce a dose of the transition metal less than the maximum dose.

Claim 32

In a fabrication process for making a semiconductor power device having at least one PN junction, an improved minority carrier lifetime control process comprising: selecting a transition metal having a deep level in silicon suitable for recombination; determining a maximum dose of the selected transition metal that can be fully dissolved into the substrate at a temperature in a range between a eutectic temperature of the substrate and an annealing temperature of the substrate; determining and depositing a dose of the transition metal not exceeding the maximum dose sufficient to effect lifetime control without substantially increasing leakage current of the device; and diffusing the metal atoms in the substrate at a temperature within said range; the depositing step including forming a silicide layer of the transition metal on the substrate and stripping away excess deposited transition metal to produce a dose of the transition metal less than the maximum dose.

Claim 33

A process according to claim 29 in which the diffusion temperature is sufficiently high that solid solubility of the transition metal exceeds the implanted dose of the metal in silicon.

Claim 34

A process according to claim 29 in which the transition metal is platinum.

Claim 35

A process according to claim 31 in which the maximum dose is 1×10 16 atoms/cm 2 , and the diffusion temperature is in a range of 830° C. to 1000° C.

Claim 36

A process according to claim 31 in which the deposited dose of platinum is in the range of 1×10 13 to 1×10 16 atoms/cm 2 .

Claim 37

A process according to claim 33 in which the diffusion temperature is 870° C. to 900° C.

Claim 38

A process according to claim 29 in which the transition metal is gold.

Claim 39

A process according to claim 35 in which the deposited dose of gold is in the range of 1×10 13 to 1×10 16 atoms/cm 2 .

Claim 40

A process according to claim 29 in which the device is a MOS-type insulated gate controlled power switching device.

Claim 41

A process according to claim 29 in which the device is a three-layer MOS-type insulated gate controlled power switching device.

Claim 42

A process according to claim 29 in which the device is a four layer MOS-type insulated gate controlled power switching device.

Claim 43

A process according to claim 29 in which the device is a MOS-type insulated gate controlled PN diode or P-i-N rectifier.

Claim 44

A process according to claim 29 in which the device is a power-type PN diode or P-i-N rectifier

Claim 45

A process according to claim 29 in which the device is a power-type bipolar transistor.

Claim 46

A process according to claim 29 in which the device is a thyristor.

Claim 47

An improved fabrication process for making a MOS-type insulated gate controlled power switching device, the process comprising: forming a semiconductor substrate having an upper surface and a lower surface and including one or more layers defining a drain region of a first dopant type extending from the upper surface toward the lower surface of the substrate; forming an insulative layer on the upper surface of the substrate and an insulated gate contact layer on the insulative layer; forming double diffused regions including a body region of a second, opposite dopant type and a source region of the first dopant type within the body region, the body region forming two PN junctions with the drain and source regions, respectively spaced apart so as to define a channel region in the body region subjacent the insulated gate contact; forming a source contact layer alongside the gate contact but spaced insulatively therefrom, the source contact forming an electrical connection to the source region and the body region and a short therebetween and defining a source or cathode contact for the device; and forming a drain or anode contact on the lower surface of the substrate; the insulated gate contact layer and the source contact layer being patterned to form a plurality of complementary, parallel interdigitated gate and source fingers and a gate bus interconnecting the gate fingers; the body and source regions extending lengthwise along opposite margins of the gate fingers and gate bus, the body regions being spaced at a first lateral spacing L 1 beneath the gate fingers across the drain region of a first dopant type and being spaced at a second spacing L 2 beneath the gate bus greater than the first spacing L 1 and forming a first breakdown prone region; the process further including doping at least a portion of the first breakdown prone region with dopant of the second dopant type over a lateral extent effective to interconnect the source regions beneath the gate bus and neutralize a portion of the channel region adjoining the interconnection.

Claim 48

An improved fabrication process for making a MOS-type insulated gate controlled power switching device, the process comprising: forming a semiconductor substrate having an upper surface and a lower surface and including one or more layers defining a drain region of a first dopant type extending from the upper surface toward the lower surface of the substrate; forming an insulative layer on the upper surface of the substrate and an insulated gate contact layer on the insulative layer; forming double diffused regions including a body region of a second, opposite dopant type and a source region of the first dopant type within the body region, the body region forming two PN junctions with the drain and source regions, respectively spaced apart so as to define a channel region in the body region subjacent the insulated gate contact; forming a source contact layer alongside the gate contact but spaced insulatively therefrom, the source contact forming an electrical connection to the source region and the body region and a short therebetween and defining a source or cathode contact for the device; and forming a drain or anode contact on the lower surface of the substrate; the insulated gate contact layer and the source contact layer being patterned to form a plurality of complementary, parallel interdigitated gate and source fingers and a gate bus interconnecting the gate fingers; the body and source regions extending lengthwise along opposite margins of the gate fingers and gate bus, the body regions being spaced at a first lateral spacing L 1 beneath the gate fingers across the drain region of a first dopant type and being spaced at a second spacing L 2 beneath the gate bus greater than the first spacing L 1 and forming a first breakdown prone region; the process further including doping at least a portion of the first breakdown prone region with dopant of the second dopant type over a lateral extent effective to interconnect the source regions beneath the gate bus and neutralize a portion of the channel region adjoining the interconnection.

Claim 49

A process according to claim 44 in which the gate bus includes a first bus transversely interconnecting proximal ends of the gate fingers and a second bus extending parallel to and spaced between two of the gate fingers, the second bus having a width greater than a width of the gate fingers and thereby establishing the second spacing L 2 and forming said first breakdown prone region between adjacent underlying body regions, the step of doping the first breakdown prone region including forming a central doped stripe spaced between the adjacent underlying body regions and sized to a width effective to reduce the spacing between source regions beneath the gate bus from the second spacing L 2 to a spacing not greater than the first spacing L 1 .

Claim 50

A process according to claim 45 in which doping the first breakdown prone region includes forming a series of transverse doped shorting bars interconnecting the source regions at intervals spaced along the central doped stripe.

Claim 51

A process according to claim 44 in which doping the first breakdown prone region includes forming a series of transverse doped shorting bars interconnecting the source regions at intervals spaced beneath the gate bus.

Claim 52

A process according to claim 47 in which shorting bars are spaced to retain a substantial proportion of channel region between the shorting bars.

Claim 53

A process according to claim 44 in which the gate bus includes a first bus transversely interconnecting proximal ends of the gate fingers, the body and source regions extending in a convex shape along a side of the first gate bus and forming a second breakdown prone region, the process further including doping at least a margin alongside the first bus of a width encompassing the second breakdown prone region with dopant of the second dopant type so as to interconnect adjacent body regions beneath the proximal ends of the gate fingers.

Claim 54

A process according to claim 49 in which the first bus has a width greater than a width of the gate fingers, thereby establishing the second spacing L 2 and forming said first breakdown prone region between adjacent underlying body regions, the step of doping the first breakdown prone region including doping at least a portion of the first breakdown prone region underlying the first bus with dopant of the second dopant type over a lateral extent effective to reduce the spacing between source regions beneath the gate bus from the second spacing L 2 to a spacing not greater than the first spacing L 1 .

Claim 55

A process according to claim 49 in which the first bus has a width greater than a width of the gate fingers, thereby establishing the second spacing L 2 and forming said first breakdown prone region between adjacent underlying body regions, the step of doping the first breakdown prone region including doping the entire first breakdown prone region underlying the first bus and including said margins with dopant of the second dopant type.

Claim 56

A process according to claim 44 in which the source and gate finger structures are tapered in a complementary fashion such that the width of the source metal fingers increasing from distal ends thereof proceeding toward proximal ends thereof connected to a source bus and pad to a width having sufficient current-carrying capacity adjacent the source bus and pad to minimize current crowding.

Claim 57

A process according to claim 52 including depositing a second layer of metal in the source pad and bus regions of the source contact layer.

Claim 58

A process according to claim 53 in which the second layer is patterned to interconnect all the source pad regions in isolation from the gate pads over a passivation layer.

Claim 59

A process according to claim 53 in which the second layer is deposited with a thickness sufficient to withstand pressure contact in a compression-bonded package.

Claim 60

A process according to claim 53 in which the second layer is deposited with a thickness of at least 8 micrometers.

Claim 61

A process according to claim 52 including depositing a second layer of metal in the gate pad regions of the gate contact layer.

Claim 62

A process according to claim 57 in which the second layer is patterned to interconnect all the gate pad regions in isolation from the source pads over a passivation layer.

Claim 63

A process according to claim 57 in which the second layer is deposited with a thickness sufficient to withstand pressure contact in a compression bonded package.

Claim 64

A process according to claim 57 in which the second layer is deposited with a thickness of at least 8 micrometers.

Claim 65

An improved fabrication process for making a MOS-type insulated gate controlled power switching device, the process comprising: forming a semiconductor substrate having an upper surface and a lower surface and including one or more layers defining a drain region of a first dopant type extending from the upper surface toward the lower surface of the substrate; forming an insulative layer on the upper surface of the substrate and an insulated gate contact layer on the insulative layer; forming double diffused regions including a body region of a second, opposite dopant type and a source region of the first dopant type within the body region, the body region forming two PN junctions with the drain and source regions, respectively spaced apart so as to define a channel region in the body region subjacent the insulated gate contact; forming a source contact layer alongside the gate contact but spaced insulatively therefrom, the source contact forming an electrical connection to the source region and the body region and a short therebetween and defining a source or cathode contact for the device; forming a drain or anode contact on the lower surface of the substrate; and depositing a dielectric film on the upper surface of the substrate, including selecting a dielectric composition and controlling deposition thereof to produce a film that is substantially nonpolarizable under high voltage conditions and forms an effective mobile ion and moisture barrier.

Claim 66

An improved fabrication process for making a MOS-type insulated gate controlled power switching device, the process comprising: forming a semiconductor substrate having an upper surface and a lower surface and including one or more layers defining a drain region of a first dopant type extending from the upper surface toward the lower surface of the substrate; forming an insulative layer on the upper surface of the substrate and an insulated gate contact layer on the insulative layer; forming double diffused regions including a body region of a second, opposite dopant type and a source region of the first dopant type within the body region, the body region forming two PN junctions with the drain and source regions, respectively spaced apart so as to define a channel region in the body region subjacent the insulated gate contact; forming a source contact layer alongside the gate contact but spaced insulatively therefrom, the source contact forming an electrical connection to the source region and the body region and a short therebetween and defining a source or cathode contact for the device; forming a drain or anode contact on the lower surface of the substrate; and depositing a dielectric film on the upper surface of the substrate, including selecting a dielectric composition and controlling deposition thereof to produce a film that is substantially nonpolarizable under high voltage conditions and forms an effective mobile ion and moisture barrier.

Claim 67

A process according to claim 61 in which the source contact layer is recessed relative to the gate contact layer into a trench in the upper surface of the substrate, the step of depositing a dielectric film including forming the film with sufficient thickness to provide step coverage between the gate and source contacts.

Claim 68

A process according to claim 61 including forming a perimetral guardring surrounding an area occupied by the source and gate contact layers, the guardring including a conductive field plate and a conductive channel stop plate on the upper surface of the substrate, separated by a recessed opening defining a high field polarization-prone region, the step of depositing a dielectric film including forming the film with sufficient thickness to provide step coverage between the field plate and channel stop plate.

Claim 69

A process according to claim 63 in which the step of depositing a dielectric film includes depositing a conformal first layer of phosphosilicate glass (PSG) with a phosphorus content of less than 4% in such a way as to be nonpolarizable even under very high voltage conditions and depositing a second layer that is composed of a dielectric suitable to form a moisture and mobile ion resistant barrier atop the first, PSG layer.

Claim 70

A process according to claim 64 in which the first layer of phosphosilicate glass (PSG) has a phosphorus content of 3.5% or less.

Claim 71

A process according to claim 64 in which the first layer is made sufficiently thick that the second layer can be of a polarizable material but is spaced from the substrate by the first layer to reduce net charge creation due to polarization in the second layer while retaining mobile ion and moisture resistance.

Claim 72

A process according to claim 64 in which the first, PSG layer is made with a thickness of 10,000 to 35,000 Å.

Claim 73

A process according to claim 67 in which the second layer is formed of silicon nitride or oxynitride.

Claim 74

A process according to claim 64 in which the second layer is formed by depositing a layer of silicon nitride or oxynitride in such a way as to be nonpolarizable even under very high voltage conditions.

Claim 75

A process according to claim 61 in which the step of depositing a dielectric film includes depositing a conformal layer of oxynitride in such a way as to form a nonpolarizable film.

Claim 76

A process according to claim 61 in which the step of depositing a nonpolarizable dielectric film includes depositing one of an oxide film, a nitride film, an oxynitride film and a low-phosphorus phosphosilicate glass (PSG) film by plasma enhanced vapor chemical deposition (PEVCD) using silane (SiH 4 ) based gases and controlling the deposition conditions so as to minimize the presence of ionized gases including ions of hydrogen compounds during deposition to minimize Si-H molecule incorporation into the film.

Claim 77

A process according to claim 71 in which PEVCD is controlled to deposit a nonpolarizable dielectric film on a wafer in a deposition chamber by limiting the residence time of ionized gas molecules in close proximity to the wafer before exiting the deposition chamber and by controlling the partial pressure of silane in the gas mixture to minimize the presence of Si-H molecules.

Claim 78

A process according to claim 72 in which the gas mixture comprises silane (SiH 4 ) and ammonia (NH 3 ) in the ratio of 1:14.

Claim 79

A process according to claim 73 in which the gas mixture further includes nitrous oxide (N 2 O) in the amount of >0.1 in ratio to silane.

Claim 80

A process according to claim 72 in which the gas mixture comprises silane (SiH 4 ), ammonia (NH 3 ) and nitrous oxide (N 2 O) in the ratio of 1:14:14.

Claim 81

A process according to claim 71 in which the PEVCD method is implemented in a system with long parallel electrodes with gases flowing down the length of a furnace tube, the residence time of ionized gases is controlled by a pulsed-type deposition with short on-cycles interspersed between long off-cycles to allow ionized gases sufficient time to exit the system before a next deposition on-cycle.

Claim 82

A process according to claim 76 in which the gas mixture comprises silane (SiH 4 ), ammonia (NH 3 ) in the ratio of 1:14 and nitrous oxide (N 2 O) in the amount of >0.1 in ratio to silane and the deposition cycle is about 12 mS on/200 mS off using a chamber pressure of about 2.5 torr and a deposition temperature of about 430° C.

Claim 83

A process according to claim 71 in which the PEVCD method is implemented in a system which has a short interaction distance from gas inlets to outlets, the residence time is controlled by maintaining a high gas flow with a low chamber pressure.

Claim 84

A process according to claim 71 in which the PEVCD method is implemented in a system which has a pulsed deposition mode of operation which allows for chopping of the RF power with discrete on/off cycles, the deposition method being controlled by introducing a period of off-time is after a short deposition period sufficient for hydrogen atoms to desorb and remove themselves from the deposited film surface leaving a film less populated with Si-H molecules

Claim 85

A process according to claim 79 including adding a small amount of oxygen sufficient to bind the hydrogen atoms to form Si-OH bonds.

Claim 86

A process according to claim 80 including balancing the amount of Si-H incorporation into the film so as not to exceed the Si-OH formation rate to avoid producing excessive Si-H bonds to produce polarization.

Claim 87

A process according to claim 71 in which the PEVCD method is implemented in a system which has a CW mode of operation using continuous deposition with single or multiple frequency RF source, the deposition method being controlled by controlling the gas ratio to provide a sufficiently reduced partial pressure of silane to minimize absorption of hydrogen atoms into the deposited film surface leaving a film less populated with Si-H molecules.

Claim 88

A process according to claim 82 including adding a small amount of oxygen sufficient to bind the hydrogen atoms to form Si-OH bonds.

Claim 89

A process according to claim 83 including balancing the amount of Si-H incorporation into the film so as not to exceed the Si-OH formation rate to avoid producing excessive Si-H bonds to produce polarization.