Silicon carbide MOSFET device and method for manufacturing the same

Abstract

The present disclosure discloses a self-aligned silicon carbide MOSFET device with an optimized P.sup.+ region and a manufacturing method thereof. The self-aligned silicon carbide MOSFET device is formed by a plurality of silicon carbide MOSFET device cells connected in parallel, and these silicon carbide MOSFET device cells are arranged evenly. The silicon carbide MOSFET device cell comprises two source electrodes, one gate electrode, one gate oxide layer, two N.sup.+ source regions, two P.sup.+ contact regions, two P wells, one N.sup. drift layer, one buffer layer, one N.sup.+ substrate, one drain electrode and one isolation dielectric layer. By optimizing the P.sup.+ region, the present disclosure forms a good source ohmic contact, reduces the on-resistance, and also shorts the source electrode and the P well to prevent the parasitic transistor effect of the parasitic NPN and PiN, which may take both conduction characteristics and the breakdown characteristics of the device into consideration, and may be applied to a high voltage, high frequency silicon carbide MOSFET device. The self-aligned manufacturing method used in the present disclosure simplifies the process, controls a size of a channel accurately, and may produce a lateral and vertical power MOSFET.

Claims

1. A self-aligned silicon carbide MOSFET device, wherein the self-aligned silicon carbide MOSFET device is formed by a plurality of silicon carbide MOSFET device cells connected in parallel, and these silicon carbide MOSFET device cells are arranged evenly, wherein the silicon carbide MOSFET device cell comprises two source electrodes, one gate electrode, one gate oxide layer, two N.sup.+ source regions, two P.sup.+ contact regions, two P wells, one N.sup. drift layer, one buffer layer, one N.sup.+ substrate, one drain electrode and one isolation dielectric layer, wherein the gate electrode and the source electrode are arranged on the same plane and located at an upper part of the device, and the drain electrode is located at a bottom of the device; the buffer layer and the N.sup. drift layer are sequentially formed on the N.sup.+ substrate; the two P wells are respectively arranged on an upper part of a left and a right ends of the N.sup. drift layer; one N.sup.+ source region and one P.sup.+ contact region are formed on each of the P wells, a depth of the N.sup.+ source region being less than that of the P.sup.+ contact region; one source electrode is formed on each of the P wells; the gate electrode is arranged between the two source electrodes, and is isolated from the two source electrodes by the isolation dielectric layer; the gate oxide layer is formed under the gate electrode, and is arranged on the N.sup. drift layer and the two P wells; and the drain electrode is formed on a bottom surface of the N.sup.+ substrate, wherein the P.sup.+ contact region includes a P.sup.+.sub.1 region, a P.sup.+.sub.2 region and a P.sup.+.sub.2 diffusion region, wherein both of the P.sup.+.sub.1 region and the P.sup.+.sub.2 region are formed with a heavy doping in an ion implantation way, the doping concentration(s) of the P.sup.+.sub.1 region and the P.sup.+.sub.2 region being higher than that of the P well in no less than an IE19 cm.sup.3 order; and the P.sup.+.sub.2 diffusion region is formed in a diffusion way, which is diffused till a bottom of the P well or even lower than the bottom of the P well, wherein Al ions, which have a lower ion activation energy and are not prone to be diffused in a high temperature activation annealing process, are selected as ions doped in the P.sup.+.sub.1 region; and B ions, which are prone to be diffused in the high temperature activation annealing process and have a deeper implantation depth, are selected as ions doped in the P.sup.+.sub.2 region, wherein a temperature of the high temperature activation annealing is ranged between 1500 C. and 1900 C.

2. A method of manufacturing the silicon carbide MOSFET device according to claim 1, wherein the method comprising: cleaning a SiC wafer; forming the P well on a surface of the SiC wafer; forming a self-aligned channel in the P well; forming the P+ contact region at outside of the self-aligned channel; high temperature activation annealing, so as to perform substitutional activation on the ions which are implanted into the P+ contact region, the N+ source region and the P well, and forming the P+2 diffusion region; forming the gate oxide layer on a surface of the N-drift layer; forming the gate electrode on the gate oxide layer; forming the isolation dielectric layer at two sides of the gate electrode and on the surface of the gate electrode; and forming the source electrodes at two sides of the isolation dielectric layer, and forming the drain electrode on the bottom surface of the N+ substrate.

3. The method of manufacturing the silicon carbide MOSFET device according to claim 2, wherein the step of cleaning a SiC wafer comprises: washing the surface of the SiC wafer with acetone, ethanol and deionized water sequentially, and drying the surface of the SiC wafer with N.sub.2, the surface of the SiC wafer being dried for ten minutes in a N.sub.2 atmosphere, wherein the SiC wafer has three layers from top to bottom, which are the N+ substrate, the buffer layer and the N drift layer subsequently.

4. The method of manufacturing the silicon carbide MOSFET device according to claim 3, wherein the step of forming the P well on a surface of the SiC wafer comprises: depositing SiO.sub.2 with a thickness of 2 m and polysilicon (Poly-Si) with a thickness of 5000 m subsequently on the N drift layer as mask layer material, and opening a window on the mask layer material by photolithography, the mask layer material on the two P wells being etched and the mask layer material in other regions being remained; forming two P well implantation windows on the two P wells and leaving the mask layer material between the two P wells as a P well implantation mask layer; and performing an ion implantation of Al ions on the P wells from the two P well implantation windows at a high temperature of 500 C., the implantation energies of the ions being 300 kev, 410 kev, 500 kev respectively and a total dose thereof being 4.615E13 cm-2.

5. The method of manufacturing the silicon carbide MOSFET device according to claim 4, wherein the step of forming a self-aligned channel in the P well comprises: depositing a SiO.sub.2 layer with a thickness of 1 m on the P well implantation mask layer and the ion-implanted P wells, performing full vertical etching on the SiO.sub.2 layer using an ICP dry etching process and stopping the etching till the surfaces of the P wells are reached, the SiO.sub.2 layer on the P well implantation mask layer and the SiO.sub.2 layer on the P wells being etched and the SiO.sub.2 at two sides of the P well implantation mask layer being remained, and SiO.sub.2 dielectric side walls at the two sides of the P well implantation mask layer between the two P wells being formed, the SiO.sub.2 dielectric side walls and the P well implantation mask layer between the two P wells together being used as an N+ source region mask layer; performing an ion implantation of Nitrogen ions on the N+ source region at a high temperature of 500 C., the implantation energies of the ions being 50 kev, 90 kev, 150 kev respectively and a total dose thereof being 9.84E13 cm-2, so that the self-aligned channels are formed under the SiO.sub.2 dielectric side walls within the two P wells; and removing the N+ source region implantation mask layer after the implantation is completed.

6. The method of manufacturing the silicon carbide MOSFET device according to claim 5, wherein the step of forming the P+ contact region at outside of the self-aligned channel comprises: depositing a SiO.sub.2 layer with a thickness of 2 m and a Poly-Si with a thickness of 5000 m subsequently on the N drift layer for which the ion implantation on the P well and the ion implantation on the N+ source region have been performed, as a mask layer; opening a window on the mask layer material by photolithography, the mask layer material on the two P+ contact regions being etched and the mask layer material in other regions being remained; forming two P+ contact region implantation windows on the two P+ contact regions and leaving the mask layer material between the two P+ contact regions as a P+ contact region implantation mask layer; performing high temperature ion implantation from the two P+ contact region implantation windows, wherein Al ions are implanted into the P.sup.+1 region at 500 C., implantation energies of the ions being 50 kev, 90 kev, 150 kev respectively and a total dose thereof being 3.9E15 cm-2, and B ions are implanted into the P.sup.+2 region at 500 C., implantation energies of the ions being 160 kev, 270 kev respectively and a total dose thereof being 2.5E15 cm-2, so that an implantation concentration of the P+ region is up to 2E20 cm-3; and removing the P+ contact region implantation mask layer after the implantation is completed.

7. The method of manufacturing the silicon carbide MOSFET device according to claim 6, wherein the step of high temperature activation annealing comprises: performing RCA cleaning on the surface of the SiC wafer for which the ion implantation on the P well, the ion implantation on the N+ source region and the ion implantation on the P+ contact region have been implemented; performing carbon film protection on the surface of the SiC wafer after it is dried; performing activation annealing for 15 minutes in an Ar atmosphere at a temperature of 1750 C., so that the ions implanted into the P+ contact region including the P.sup.+1 region and the P.sup.+2 region, the N+ source region, the P well are subjected to the substitutional activation to have electrical characteristics, while the ions in the P.sup.+2 region are diffused to form the P.sup.+ 2 diffusion region; and removing the carbon film protection layer after the activation annealing is completed.

8. The method of manufacturing the silicon carbide MOSFET device according to claim 7, wherein the step of forming the gate oxide layer on a surface of the N-drift layer comprises: performing dry oxidation at 1300 C. by an oxidation furnace on the surface of the N drift layer for which the high temperature activation annealing on the P well, the N+ source region and the P+ contact region has been performed, so as to form the gate oxide layer with a thickness of 60 nm; performing annealing at 1300 C. in the N.sub.2 atmosphere and annealing at 1300 C. in a NO atmosphere to improve a quality of the gate oxide layer; and etching the gate oxide layer on the two source electrodes by wet etching, with only the gate oxide layer right under the gate electrode and within 0.5 pm away from the gate electrode between the gate electrode and the source electrode being remained.

9. The method of manufacturing the silicon carbide MOSFET device according to claim 8, wherein the step of forming the gate electrode on the gate oxide layer comprises: depositing Poly-Si with a thickness of 5000 on the SiC wafer on which the gate oxide layer has been formed, and performing implantation doping and annealing activation; and performing dry-etching on the Poly-Si to remove the Poly-Si on the two P+ contact regions and on a part of the N+ source region and leave the Poly-Si on the gate oxide layer, so as to form a Poly-Si gate electrode, a width of the gate electrode being less than a width of the gate oxide layer.

10. The method of manufacturing the silicon carbide MOSFET device according to claim 9, wherein the step of forming the isolation dielectric layer at two sides of the gate electrode comprises: depositing SiO.sub.2 with a thickness of 1.2 m on the SiC wafer on which the gate electrode has been formed, by using a LPCVD or PECVD method; and then etching the SiO.sub.2 on the source electrode to form a window shape for the required source electrode.

11. The method of manufacturing the silicon carbide MOSFET device according to claim 10, wherein the step of forming the source electrodes and the drain electrode comprises: evaporating Ni metal with a thickness of 2000 on the bottom surface of the N+ substrate as the drain electrode.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) In order to further illustrate the technical solutions of embodiments of the present disclosure, various embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In the illustrated drawings,

(2) FIG. 1a is a structure schematic diagram of a single self-aligned silicon carbide MOSFET device cell according to an embodiment of the present disclosure;

(3) FIG. 1b is a layout for manufacturing a single self-aligned silicon carbide MOSFET device according to an embodiment of the present disclosure, FIG. 1c is a schematic diagram of one active region in the layout of FIG. 1b, in which one cell is defined in a solid line block, and a cross-section view of the cell as shown in FIG. 1a may be obtained along a tangent, wherein an active region is also under a source electrode Pad in FIG. 1b, and the active region covers an area excluding a gate electrode Pad and a peripheral terminal;

(4) FIG. 2a is an illustration diagram of a conventional self-aligning process;

(5) FIG. 2b is an illustration diagram of a self-aligning process according to an embodiment of the present disclosure;

(6) FIGS. 3a to 3i are process flowcharts of manufacturing a self-aligned silicon carbide MOSFET device cell according to an embodiment of the present disclosure, in which

(7) FIG. 3a is a process schematic diagram of cleaning a silicon carbide wafer;

(8) FIG. 3b is a process schematic diagram of forming a P well;

(9) FIG. 3c is a process schematic diagram of forming a N.sup.+ source region and a self-aligned channel;

(10) FIG. 3d is a process schematic diagram of forming a P.sup.+ contact region;

(11) FIG. 3e is a process schematic diagram of high temperature activation annealing;

(12) FIG. 3f is a process schematic diagram of forming a gate oxide layer;

(13) FIG. 3g is a process schematic diagram of forming a gate electrode;

(14) FIG. 3h is a process schematic diagram of forming an isolation dielectric layer; and

(15) FIG. 3i is a process schematic diagram of forming a drain electrode and a source electrode.

DETAILED DESCRIPTION

(16) In order to clarify the object, technical solutions and advantages of the present disclosure better, embodiments of the present disclosure will be described in detail in connection with the drawings and the detailed description.

(17) The self-aligned silicon carbide MOSFET device provided by the present disclosure is formed by a plurality of silicon carbide MOSFET device cells connected in parallel, and these silicon carbide MOSFET device cells are arranged evenly. FIG. 1a shows a schematic block diagram of a single self-aligned silicon carbide MOSFET device cell according to an embodiment of the present disclosure, FIG. 1b is a layout for manufacturing a single self-aligned silicon carbide MOSFET device according to an embodiment of the present disclosure, FIG. 1c is a schematic diagram of one active region in the layout of FIG. 1b, in which one cell is defined in a solid line block, and a cross-section view of the cell as shown in FIG. 1a may be obtained along a tangent. The self-aligned silicon carbide MOSFET device cell in FIG. 1c is a square structure, but the self-aligned silicon carbide MOSFET device provided by the present disclosure is not limited to such as square structure.

(18) With reference to FIG. 1, the single self-aligned silicon carbide MOSFET device cell according to the embodiment of the present disclosure comprises two source electrodes 1, one gate electrode 2, one gate oxide layer 3, two N.sup.+ source regions 4, two P.sup.+ contact regions 5, two P wells 6, one N.sup. drift layer 7, one buffer layer 8, one N.sup.+ substrate 9, one drain electrode 10 and one isolation dielectric layer 11, wherein the gate electrode 2 and the source electrode 1 are arranged on the same plane and located at an upper part of the device, and the drain electrode is located at a bottom of the device; the buffer layer 8 and the N.sup. drift layer 7 are sequentially formed on the N.sup.+ substrate 9; the two P wells 6 are respectively arranged on an upper part of a left and a right ends of the N.sup. drift layer 7; one N.sup.+ source region 4 and one P.sup.+ contact region 5 are formed on each of the P wells 6, a depth of the N.sup.+ source region 4 being less than a depth of the P.sup.+ contact region 5; one source electrode 1 is formed on each of the P wells 6; the gate electrode 2 is arranged between the two source electrodes 1, and is isolated from the two source electrodes 1 by the isolation dielectric layer 11; and the gate oxide layer 3 is formed under the gate electrode 2, and is arranged on the N.sup. drift layer 7 and the two P wells 6.

(19) In a preferred embodiment, the P.sup.+ contact region 5 includes a P.sup.+.sub.1 region, a P.sup.+.sub.2 region and a P.sup.+.sub.2 diffusion region, wherein both the P.sup.+.sub.1 region and the P.sup.+.sub.2 region are formed with heavy doping in an ion implantation way, doping concentration(s) of the P.sup.+.sub.1 region and the P.sup.+.sub.2 region being higher than that of the P well 6 in no less than an IE19 cm.sup.3 order; and the P.sup.+.sub.2 diffusion region is formed in a diffusion way, which is diffused till a bottom of the P well or even lower than the bottom of the P well.

(20) In a preferred embodiment, Al ions, which have lower ion activation energy and are not prone to be diffused in high temperature activation annealing, are selected as ions doped in the P.sup.+.sub.1 region; and B ions, which are prone to be diffused in the high temperature activation annealing and have a deeper implantation depth, are selected as ions doped in the P.sup.+.sub.2 region. A temperature of the high temperature activation annealing is ranged between 1500 C. and 1900 C.

(21) Based on the structure schematic diagram of the single self-aligned silicon carbide MOSFET device cell according to an embodiment of the present disclosure as shown in FIGS. 1a to 1c, FIGS. 3a to 3i show process flowcharts of manufacturing a self-aligned silicon carbide MOSFET device cell according to an embodiment of the present disclosure, which particularly comprises the following steps:

(22) Step 1 of cleaning a SiC wafer.

(23) In such as step as shown in FIG. 3a, the surface of the SiC wafer is washed with acetone, ethanol and deionized water sequentially, and the surface of the SiC wafer is dried with N.sub.2, the surface of the SiC wafer being dried for ten minutes in a N.sub.2 atmosphere, wherein the SiC wafer has three layers from top to bottom, which are the N.sup.+ substrate 9, the buffer layer 8 and the N.sup. drift layer 7 subsequently.

(24) Step 2 of forming the P well on a surface of the SiC wafer.

(25) in which, as shown in FIG. 3b, SiO.sub.2 with a thickness of 2 m and polysilicon (Poly-Si) with a thickness of 5000 m are deposited subsequently on the N.sup. drift layer 7 as mask layer material, and a window is opened on the mask layer material by photolithography, the mask layer material on the two P wells 6 being etched and the mask layer material in other regions being remained; two P well implantation windows are formed on the two P wells 6 and the mask layer material between the two P wells 6 is remained as a P well implantation mask layer; then an ion implantation of Al ion at a high temperature of 500 C. is performed on the P wells from the two P well implantation windows, the implantation energies of the ions being 300 kev, 410 kev, 500 kev respectively and a total dose thereof being 4.615E13 cm.sup.2.

(26) Step 3 of forming a self-aligned channel in the P well.

(27) In such a step, as shown in FIG. 3c, a SiO.sub.2 layer with a thickness of 1 m is deposited on the P well implantation mask layer and the ion-implanted P wells, full vertical etching is performed on the SiO.sub.2 layer using an ICP dry etching process and the etching is stopped till the surfaces of the P wells are reached, the SiO.sub.2 layer on the P well implantation mask layer and the SiO.sub.2 layer on the P wells 6 being etched and the SiO.sub.2 at two sides of the P well implantation mask layer being remained, and SiO.sub.2 dielectric side walls at the two sides of the P well implantation mask layer between the two P wells 6 being formed, the SiO.sub.2 dielectric side walls and the P well implantation mask layer between the two P wells 6 together being used as an N.sup.+ source region 4 implantation mask layer; an ion implantation of N ions is performed on the N.sup.+ source region 4 at a high temperature of 500 C., implantation energies of the ions being 50 kev, 90 kev, 150 kev respectively and a total dose thereof being 9.84E13 cm.sup.2, so that the self-aligned channels are formed under the SiO.sub.2 dielectric side walls within the two P wells by the above steps; and removing the N.sup.+ source region 4 implantation mask layer after the implantation is completed.

(28) Step 4 of forming the P.sup.+ contact region out of the self-aligned channel;

(29) in which, as shown in FIG. 3d, a SiO.sub.2 layer with a thickness of 2 m and a Poly-Si with a thickness of 5000 m are deposited subsequently on the N.sup. drift layer 7 for which the ion implantation on the P well 6 and the ion implantation on the N.sup.+ source region 4 have been performed, as a mask layer; the window is photoetched on the mask layer material, the mask layer material on the two P.sup.+ contact regions 5 being etched and the mask layer material in other regions being remained; two P.sup.+ contact region implantation windows are formed on the two P.sup.+ contact regions 5 and the mask layer material between the two P.sup.+ contact regions 5 is remained as a P.sup.+ contact region implantation mask layer; then high temperature ion implantation is performed from the two P.sup.+ contact region implantation windows, wherein Al ions are implanted into the P.sup.+.sub.1 region at 500 C., the implantation energies of the ions being 50 kev, 90 kev, 150 kev respectively and a total dose thereof being 3.9E15 cm.sup.2; and B ions are implanted into the P.sup.+.sub.2 region at 500 C., the implantation energies of the ions being 160 kev, 270 kev respectively and a total dose thereof being 2.5E15 cm.sup.2, so that an implantation concentration of the P.sup.+ region is up to 2E20 cm.sup.3; and the implantation mask layer for the P.sup.+ contact region 5 is removed after the implantation is completed.

(30) Step 5 of high temperature activation annealing, so as to perform substitutional activation on the ions which are implanted into the P.sup.+ contact region, the N.sup.+ source region and the P well, and form the P.sup.+.sub.2 diffusion region.

(31) In such a step as shown in FIG. 3e, RCA cleaning is performed on the surface of the SiC wafer for which the ion implantation on the P well 6, the ion implantation on the N.sup.+ source region 4 and the ion implantation on the P.sup.+ contact region 5 have been formed; a carbon film protection is performed on the surface of the SiC wafer after it is dried; an activation annealing is performed for 15 minutes in an Ar atmosphere at a temperature of 1750 C., so that the ions implanted into the P.sup.+ contact region 5 including the P.sup.+.sub.1 region and the P.sup.+.sub.2 region, the N.sup.+ source region 4, the P well 6 are subjected to the substitutional activation to have electrical characteristics, while the ions in the P.sup.+.sub.2 region are diffused to form the P.sup.+.sub.2 diffusion region; and the carbon film protection layer is removed after the activation annealing is completed.

(32) Step 6 of forming the gate oxide layer on a surface of the N.sup. drift layer.

(33) In such a step as shown in FIG. 3f, dry-oxidation at 1300 C. is performed by an oxidation furnace on the surface of the N.sup. drift layer 7 for which the high temperature activation annealing on the P well 6, the N.sup.+ source region 4 and the P.sup.+ contact region 5 has been performed, so as to form the gate oxide layer with a thickness of 60 nm; annealing at 1300 C. in the N.sub.2 atmosphere and annealing at 1300 C. in a NO atmosphere are performed to improve a quality of the gate oxide layer; then etching the gate oxide layer on the two source electrodes 1 by wet etching, with only the gate oxide layer right under the gate electrode 2 and within 0.5 m away from the gate electrode 2 between the gate electrode 2 and the source electrode 1 being remained.

(34) Step 7 of forming the gate electrode on the gate oxide layer.

(35) In such a step as shown in FIG. 3g, Poly-Si with a thick ness of 5000 A is deposited on the SiC wafer on which the gate oxide layer 3 has been formed, and implantation doping and annealing activation are performed; and dry-etching is performed on the Poly-Si to remove the Poly-Si on the two P.sup.+ contact regions 5 and on a part of the N.sup.+ source region 4 and leave the Poly-Si on the gate oxide layer 3, so as to form a Poly-Si gate electrode 2, a width of the gate electrode 2 being less than a width of the gate oxide layer 3.

(36) Step 8 of forming the isolation dielectric layer at two sides of the gate electrode and on the surface of the gate electrode.

(37) In such a step as shown in FIG. 3h, SiO.sub.2 with a thickness of 1.2 m is deposited on the SiC wafer on which the gate electrode (2) has been formed, by using a LPCVD or PECVD method; and then the SiO.sub.2 on the source electrode 1 is etched out to form a window shape for the required source electrode 1.

(38) Step 9 of forming the source electrodes at two sides of the isolation dielectric layer, and forming the drain electrode on the bottom surface of the N.sup.+ substrate.

(39) In such a step as shown in FIG. 3i, Ni metal with a thickness of 2000 A is evaporated on the bottom surface of the N.sup.+ substrate 9 as the drain electrode 10; on the SiC wafer on which the isolation dielectric layer 11 has been formed, negative photoresist is spin-coated and photolithography and development are performed, the negative photoresist on the P.sup.+ contact region 5 and the part of the N.sup.+ source region 4 is removed, and the negative photoresist on the gate oxide layer 3 is remained as a source electrode 1 metallic ohmic contact region; then Ni metal with a thickness of 800 A is deposited, the metal on the negative photoresist on the gate oxide layer 3 is peeled off to keep the metal on the source electrode 1 metallic ohmic contact region without the negative photoresist as source electrode metal; and annealing is performed on alloy at 970 C. for 2 minutes in the N.sub.2 atmosphere to form the ohmic contact of the source electrode 1 and the drain electrode 10.

(40) Finally, the manufacture of one silicon carbide MOSFET device cell is completed.

(41) In practical applications, the silicon carbide MOSFET device is consisted of many identical cells, and the above manufacturing process may form a plurality of cells simultaneously. The manufacturing process may be spread to the production process of the whole device which thus will not be described here for simplicity.

(42) It may be understood that the above implementations are only exemplary implementations for illustrating the principles of the present disclosure, but the present disclosure is not limited to these. For the skilled in the art, various variations and improvements may be made without being apart from the sprit and substance of the present disclosure, which also fall into the protection scope of the present disclosure.