A VDMOSFET with high short-circuit resistance and preparation method thereof
Patent Information
- Application Number
- CN202411056005.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-02
AI Technical Summary
Existing SiC MOSFET devices are difficult to effectively deal with short circuit failures in high voltage, high temperature and high power density scenarios, and existing solutions to improve short circuit capabilities will lead to a decrease in specific on-resistance.
Using VDMOSFET with a square asymmetric semicellular structure, the JFET region is formed by forming an N-epitaxial layer, PWELL, P+ source region and N+ source region on the substrate layer to form a JFET region and optimize its width and depth to improve the short circuit capability of the device. At the same time, the specific on-resistance is maintained low.
Without increasing the production cost and changing the manufacturing process, the short circuit capability of the device is improved, the specific on-resistance is reduced, and the current peak and device temperature are reduced during the short circuit.
Smart Images

Figure CN118571948B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of semiconductor technology, and in particular relates to a VDMOSFET (Vertical Double-diffused Metal Oxide Semiconductor Field Effect Transistor) with high short-circuit resistance and a preparation method thereof. Background Art
[0002] In the overall positive environment of the power electronics industry, power semiconductor devices, which play a decisive role, have become a direct factor affecting the cost and efficiency of power electronic equipment. Although silicon (Si)-based power devices are already very mature at this stage, as power semiconductors gradually develop towards high power, high frequency and low power consumption, silicon-based devices are limited by their own physical characteristics and are beginning to be difficult to apply to some high-voltage, high-temperature, high-efficiency and high-power density application scenarios.
[0003] Silicon carbide (SiC) materials have begun to attract widespread attention from practitioners due to their superior physical properties. Therefore, SiCMOSFET (Silicon Carbide Metal Oxide Semiconductor Field Effect Transistor) technology has developed accordingly. Compared with silicon-based devices, the high thermal conductivity and large bandgap of silicon carbide materials determine its application scenarios with high current density, high breakdown field strength and high operating temperature. Compared with Si MOSFETs of the same level, SiC MOSFETs have lower specific on-resistance and switching losses, making them suitable for higher operating frequencies, and their high thermal conductivity greatly improves high-temperature stability. Among the many failure mechanisms, short circuit is one of the failure modes that lead to permanent failure of switching devices. Before the protection circuit intervenes, SiC MOSFET must withstand high electrothermal stress during the short-circuit transient without failure, which puts forward requirements for the reinforcement of short-circuit-related devices. In addition, a reasonable structural design is required. It is also very important not to reduce the specific on-resistance of the device while protecting the short-circuit capability of the device. Currently, the solutions to the short-circuit problem include: improving short-circuit capability through compromise design based on JFET region width and doping; and improving short-circuit capability through deep PWELL injection design to limit the current path.
[0004] However, the compromise design of JFET region width and doping is difficult to significantly improve the short-circuit capability while keeping the specific on-resistance unchanged; the design of deep PWELL injection has too high process requirements and will bring about the problem of decreasing the specific on-resistance. Summary of the invention
[0005] In order to solve the above problems existing in the prior art, the present invention provides a VDMOSFET with high short-circuit resistance and a preparation method. The technical problem to be solved by the present invention is achieved by the following technical solutions:
[0006] In a first aspect, an embodiment of the present invention provides a VDMOSFET with high short-circuit resistance, wherein the VDMOSFET includes a plurality of square asymmetric half-cell structures; a structure formed by two adjacent square asymmetric half-cell structures includes:
[0007] substrate layer;
[0008] An N-epitaxial layer, located on the upper surface of the substrate layer;
[0009] A first PWELL, located at one end of the N-epitaxial layer;
[0010] A first P+ source region, located in the N-epitaxial layer and adjacent to the first PWELL;
[0011] A second P+ source region, located at the other end of the N- epitaxial layer;
[0012] A second PWELL is located in the N-epitaxial layer and adjacent to the second P+ source region; wherein a JFET region is formed between the first P+ source region and the second PWELL;
[0013] a first N+ source region, located in the first PWELL and adjacent to the first P+ source region;
[0014] A second N+ source region, located in the second PWELL and adjacent to the second P+ source region;
[0015] A first gate electrode, located on the first PWELL and a portion of the first N+ source region;
[0016] A second gate electrode, located on the second PWELL, a portion of the second N+ source region and a portion of the JFET region;
[0017] A first source electrode, located on a portion of the first N+ source region and a portion of the first P+ source region;
[0018] A second source electrode is located on a portion of the second N+ source region and a portion of the second P+ source region;
[0019] A drain electrode, located on the lower surface of the substrate layer;
[0020] Among them, viewed from the top: with the center of the first N+ source region as the axis of symmetry, the half-cell structure including the first N+ source region, the first PWELL and the first P+ source region is an asymmetric structure up and down and left and right; with the center of the second N+ source region as the axis of symmetry, the half-cell structure including the second N+ source region, the second PWELL and the second P+ source region is an asymmetric structure up and down and left and right.
[0021] In one embodiment of the present invention, the width of the JFET region is 0.6 μm to 3 μm.
[0022] In one embodiment of the present invention, the width of the first P+ source region is the same as the width of the second P+ source region, and the width of the first P+ source region is smaller than the width of the first N+ source region and the width of the second N+ source region.
[0023] In one embodiment of the present invention, viewed from a top view: the first N+ source region and the second N+ source region are in a square structure; the first P+ source region and the first N+ source region form a square structure, the second P+ source region and the second N+ source region form a square structure, and the first P+ source region and the second P+ source region are in an L-shaped structure respectively; the first PWELL, the first P+ source region and the first N+ source region form a square structure, the second PWELL, the second P+ source region and the second N+ source region form a square structure.
[0024] In one embodiment of the present invention, the maximum dimension of each square asymmetric half-cell structure is 4.5 μm.
[0025] In a second aspect, an embodiment of the present invention provides a method for preparing a VDMOSFET with high short-circuit resistance, wherein the VDMOSFET includes a plurality of square asymmetric half-cell structures; the method for preparing two adjacent square asymmetric half-cell structures includes:
[0026] Obtaining a substrate layer;
[0027] epitaxially growing an N-epitaxial layer on the upper surface of the substrate layer;
[0028] Performing ion implantation on the N-epitaxial layer to form a JFET region;
[0029] Forming a PWELL injection mask on the JFET region, and performing ion implantation on the surface of the JFET region outside the PWELL injection mask to form a first PWELL at one end in the JFET region and a second PWELL at the other end near the JFET region;
[0030] Forming an N+ source region injection mask on the PWELL injection mask, part of the first PWELL and part of the second PWELL by a self-alignment process, and performing ion implantation on the first PWELL surface and the second PWELL surface outside the N+ source region injection mask to form a first N+ source region in the first PWELL and a second N+ source region in the second PWELL;
[0031] The PWELL injection mask and the N+ source injection mask are removed, and a P+ source injection mask is formed on part of the first PWELL, part of the JFET region, part of the second PWELL, the first N+ source region, and the second N+ source region. Ion implantation is performed on the first PWELL and the JFET region outside the P+ source injection mask to form a first P+ source region, and ion implantation is performed on the second PWELL and the JFET region outside the P+ source injection mask to form a second P+ source region.
[0032] Removing the P+ source region implantation mask, forming a first gate electrode on the first PWELL and a portion of the first N+ source region, and forming a second gate electrode on the second PWELL, a portion of the second N+ source region, and a portion of the JFET region;
[0033] forming a first source electrode on a portion of the first N+ source region and a portion of the first P+ source region, and forming a second source electrode on a portion of the second N+ source region and a portion of the second P+ source region;
[0034] forming a drain electrode on the lower surface of the substrate layer;
[0035] Among them, viewed from the top: with the center of the first N+ source region as the axis of symmetry, the half-cell structure including the first N+ source region, the first PWELL and the first P+ source region is an asymmetric structure up and down and left and right; with the center of the second N+ source region as the axis of symmetry, the half-cell structure including the second N+ source region, the second PWELL and the second P+ source region is an asymmetric structure up and down and left and right.
[0036] In one embodiment of the present invention, the width of the JFET region after the first P+ source region is formed is 0.6 μm to 3 μm.
[0037] In one embodiment of the present invention, the prepared VDMOSFET, viewed from a top view: the first N+ source region and the second N+ source region are in a square structure; the first P+ source region and the first N+ source region form a square structure, the second P+ source region and the second N+ source region form a square structure, and the first P+ source region and the second P+ source region are in an L-shaped structure respectively; the first PWELL, the first P+ source region and the first N+ source region form a square structure, and the second PWELL, the second P+ source region and the second N+ source region form a square structure.
[0038] In one embodiment of the present invention, the maximum process size of each square asymmetric half-cell structure in the fabricated VDMOSFET is 4.5 μm.
[0039] Beneficial effects of the present invention:
[0040] The present invention proposes a VDMOSFET with high short-circuit resistance, and the VDMOSFET is composed of a plurality of square asymmetric half-cell structures. The square asymmetric half-cell structure has the square cell characteristic of the highest channel density. Compared with the existing traditional square cell structure, the square asymmetric half-cell structure narrows the JFET region, so that during the short-circuit process, the current path narrowing caused by the expansion of the depletion region under high voltage is more obvious, thereby making the device resistance increase effect under high voltage more obvious than the traditional structure, all current peaks are lower, and the device heats up more slowly; the square asymmetric half-cell structure proposed by the present invention can be well adapted to the sub-alignment process without increasing any preparation cost. Without changing any manufacturing process, the short-circuit capability of the device can be improved and the specific on-resistance of the device can be reduced by changing the structure of the device.
[0041] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 It is a schematic diagram of the device structure of a traditional VDMOSFET;
[0043] Figure 2 It is a top view schematic diagram of a conventional VDMOSFET;
[0044] Figure 3 1 is a schematic diagram of a device structure of a VDMOSFET with high short-circuit resistance provided by an embodiment of the present invention;
[0045] Figure 4 is a schematic top view of the VDMOSFET provided by the present invention;
[0046] Figure 5It is a top view schematic diagram of a half-cell structure of a conventional VDMOSFET and a VDMOSFET proposed in the present invention based on a minimum process size design;
[0047] Figure 6 Schematic diagram of the specific on-resistance performance of the VDMOSFET proposed in the present invention;
[0048] Figure 7 It is a schematic diagram of the change of short-circuit peak current and short-circuit temperature of the VDMOSFET proposed in the present invention during the short-circuit process;
[0049] Figure 8 It is a schematic diagram of a method for preparing two adjacent asymmetric half-cell structures in a VDMOSFET with high short-circuit resistance provided by an embodiment of the present invention;
[0050] Figure 9(a) to Figure 9(h) It is a schematic structural diagram corresponding to each step in the preparation process of two adjacent asymmetric half-cell structures in the VDMOSFET proposed in the present invention.
[0051] Description of reference numerals:
[0052] 1-substrate layer; 2-N-epitaxial layer; 3-JFET region 3; 41-first PWELL; 42-second PWELL; 51-first N+ source region; 52-second N+ source region; 61-first P+ source region; 62-second P+ source region; 71-first gate electrode; 72-second gate electrode; 81-first source electrode; 82-second source electrode; 9-drain electrode; 10-PWELL injection mask; 11-N+ source region injection mask; 12-P+ source region injection mask. DETAILED DESCRIPTION
[0053] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.
[0054] Figure 1 The figure shows the structure of a traditional VDMOSFET device. In order to reduce the device specific on-resistance, a square half-cell structure with the highest channel density is usually used. Figure 2 As shown in the top view. With the advancement of technology, asymmetric structures have been mentioned for some reliability requirements. These asymmetric structures have significant advantages in blocking ability and gate oxide protection ability. However, the inventors have found that for asymmetric structures, due to the asymmetric relationship, the device pattern size will be sharply reduced compared to the traditional square cell structure. For the lithography machine, it is easier to expose long narrow lines than short narrow lines. For example, for a lithography machine with an accuracy of 0.5μm, it can generally expose , but for The graphics will have blurred edges and unclear corners. In this case, the size of the graphics should generally be increased, which means an increase in the size of the cells.
[0055] In this case, in order to meet the requirements of lithography, the asymmetric structure is more often made into a strip structure, so as to ensure that the cell size is limited by lithography and the cell size will not be too large. At the same time, for the strip structure, since it is necessary to perform injection in steps to form structures such as PWELL and P+ source area, the process control problem is more serious due to the lithography overlay deviation, and the device size is further increased due to the overlay deviation. Multiple lithography also means that additional process steps need to be added, which ultimately seriously affects both the manufacturing cost and the device's specific conduction characteristics. It can be seen that although the asymmetric structure has a certain improvement in the short-circuit capability of the device, the process treatment will seriously deteriorate the specific conduction characteristics. Therefore, in order to better apply the asymmetric structure to VDMOSFET devices, it is necessary to change the device structure to reduce the device's specific on-resistance.
[0056] Based on the above research, see Figure 3 The embodiment of the present invention provides a VDMOSFET with high short-circuit resistance, the VDMOSFET comprising a plurality of novel square asymmetric half-cell structures; the structure formed by two adjacent square asymmetric half-cell structures comprises:
[0057] Substrate layer 1;
[0058] N-epitaxial layer 2, located on the upper surface of substrate layer 1;
[0059] A first PWELL 41 is located at one end of the N-epitaxial layer 2;
[0060] A first P+ source region 61 , located in the N- epitaxial layer 2 and adjacent to the first PWELL 41 ;
[0061] A second P+ source region 62, located at the other end of the N- epitaxial layer 2;
[0062] The second PWELL 42 is located in the N-epitaxial layer 2 and adjacent to the second P+ source region 62 ; wherein a JFET region 3 is formed between the first P+ source region 61 and the second PWELL 42 ;
[0063] A first N+ source region 51 , located in the first PWELL 41 and adjacent to the first P+ source region 61 ;
[0064] A second N+ source region 52 , located in the second PWELL 42 and adjacent to the second P+ source region 62 ;
[0065] The first gate electrode 71 is located on the first PWELL 41 and a portion of the first N+ source region 51. It should be noted that the first gate electrode 71 is also located on a portion of the JFET region. Figure 3 The JFET region on the left side of the first PWELL 41 is not fully illustrated. For example, there is a partial JFET region below the first PWELL 41 , similar to the second PWELL 42 ;
[0066] A second gate electrode 72 , located on the second PWELL 42 , a portion of the second N+ source region 52 , and a portion of the JFET region;
[0067] A first source electrode 81 is located on a portion of the first N+ source region 51 and a portion of the first P+ source region 61;
[0068] A second source electrode 82 is located on a portion of the second N+ source region 52 and a portion of the second P+ source region 62;
[0069] A drain electrode 9 is located on the lower surface of the substrate layer 1;
[0070] Among them, Figure 4 As shown in the top view: with the center of the first N+ source region 51 as the axis of symmetry, the half-cell structure including the first N+ source region 51, the first PWELL41 and the first P+ source region 61 is an asymmetric structure up and down and left and right; with the center of the second N+ source region 52 as the axis of symmetry, the half-cell structure including the second N+ source region 52, the second PWELL42 and the second P+ source region 62 is an asymmetric structure up and down and left and right.
[0071] In the embodiment of the present invention, the substrate layer 1 may be an N-type SiC substrate with a thickness of 360 μm and a doping concentration of 5e18 cm -3 ~1e20cm -3 .
[0072] In the embodiment of the present invention, the width of the JFET region is 0.6 μm to 3 μm.
[0073] In the embodiment of the present invention, the width of the first P+ source region 61 is the same as the width of the second P+ source region 62, and the width of the first P+ source region 61 is smaller than the width of the first N+ source region 51 and the second N+ source region 52, thereby ensuring the forward current flow capacity of the device; the injection depth of the first P+ source region 61 is the same as the injection depth of the second P+ source region 62, and the injection depth of the first P+ source region 61 is greater than the injection depth of the first PWELL41 and the second PWELL42, so that its depletion effect is stronger, thereby providing better protection during the short circuit process and improving the blocking capability of the device.
[0074] In the embodiment of the present invention, Figure 4As shown, viewed from the top: the first N+ source region 51 and the second N+ source region 52 are in a square structure; the first P+ source region 61 and the first N+ source region 51 form a square structure, the second P+ source region 62 and the second N+ source region 52 form a square structure, and the first P+ source region 61 and the second P+ source region 62 are respectively in an L-shaped structure; the first PWELL41, the first P+ source region 61 and the first N+ source region 51 form a square structure, the second PWELL42, the second P+ source region 62 and the second N+ source region 52 form a square structure.
[0075] It should be noted that the top-view direction refers to the direction perpendicular to the plane where the substrate layer 1 is located; the first PWELL 41 and the second PWELL 42 are in a square structure. Figure 4 Part of the structure is shown schematically in FIG. 1 , and the rest is blocked by the first P+ source region 61 and the first N+ source region 51 , the second P+ source region 62 and the second N+ source region 52 .
[0076] In an embodiment of the present invention, the maximum process size of each square asymmetric half-cell structure is half of the minimum process size of the square symmetric half-cell structure in the traditional VDMOSFET. At present, based on the same minimum process size requirements such as lithography, etching, isolation, etc., the minimum process size of the square symmetric half-cell structure in the traditional VDMOSFET has reached 9μm, then the process size of each square asymmetric half-cell structure in the embodiment of the present invention can be 4.5μm, and even the process size of each square asymmetric half-cell structure can be less than 4.5μm, that is, the maximum process size of each asymmetric half-cell structure proposed in the present invention is 4.5μm, and such a conflict has a huge impact on semiconductor process manufacturing. Among them, the process size of 4.5μm means that the length and width of a half-cell structure are 4.5μm respectively.
[0077] Figure 5 The advantages of the structure proposed in the embodiment of the present invention can be more intuitively explained. Figure 5 The left side is the half-cell structure proposed by the present invention, and the right side is the traditional half-cell structure. For VDMOSFET devices, while reducing the size of the cell, it is also necessary to ensure a certain contact area between the N+ source region and the P+ source region, because the current is conducted through the N+ source region structure. If the contact area is too small, the resistance of the ohmic contact area will be too high, causing the VDMOSFET device to fail due to electrode heating before reaching its conduction capability. At the same time, the ohmic contact window opening is also limited by the lithography accuracy (generally greater than 1μm) and the gate-source isolation spacing (generally greater than 0.8μm). In order to show the conduction advantage of the square asymmetric structure, such as Figure 5 The figure shows a top view of the square asymmetric half-cell structure proposed in the present invention and a traditional square symmetric half-cell structure at the same process size. For the convenience of observation, the gate electrode and the source electrode are omitted here. Figure 5 All dimensions are designed according to the current minimum process dimensions, and the design rules of the two are consistent. Calculation shows that: under the same cell area, the channel density of the traditional square symmetrical half-cell structure and the square asymmetric half-cell structure proposed in the present invention is 16.05%, which means that the square asymmetric half-cell structure proposed in the present invention retains the advantages of high channel density and low specific on-resistance of the traditional square symmetrical half-cell while changing the overall structure of the cell. On this basis, compared with the traditional square symmetrical half-cell structure, the square asymmetric half-cell structure proposed in the present invention can perform additional JFET region doping while ensuring the same blocking capability, so that the specific on-resistance can be further reduced, such as Figure 6 As shown, Figure 6 The horizontal axis represents the gate voltage (Gate Voltage), the unit is volt (V), and the vertical axis represents the drain total current (Drain Total Current), the unit is ampere (A). Figure 6 The novel structure is the VDMOSFET structure mentioned in the present invention.
[0078] For the short-circuit process, the heat concentration in the JFET region under large current is the main reason for the device to burn out. The existing short-circuit optimization solutions, such as inserting a P+ source region structure into the JFET region and modulating the doping of the JFET region, do not solve the problem of high power density in the JFET region. Although deep PWELL injection improves the short-circuit problem, it also reduces the device's conduction capability. For the square asymmetric structure proposed in the present invention, first of all, it changes the traditional dual-channel structure into a single-channel structure. This change greatly reduces the current density in the JFET region of the VDMOSFET device without changing the cell size. In addition, Figure 5 The distance between adjacent JFET regions of the traditional square symmetrical half-cell structure shown is as long as 7μm, and this part cannot participate well in the heat dissipation process. However, the distance of the square asymmetric half-cell structure proposed in the present invention is only 3.5μm, so that the place where the short circuit process does not occur can participate well in the heat dissipation process of the short circuit, that is, under the same short circuit conditions, the heat generation of the device proposed in the present invention is lower. Finally, because the square asymmetric half-cell structure proposed in the present invention narrows the JFET region, the current path narrowing caused by the expansion of the depletion region under high voltage is more obvious during the short circuit process, so the device resistance rise effect under high voltage is more obvious than that of the traditional square symmetrical half-cell structure, all current peaks are lower, and the device heats up more slowly, such as Figure 7 As shown, Figure 7 The horizontal axis in the middle represents time (Time), in microseconds (μs), the vertical axis on the left represents the total drain current (Drain Total Current), in amperes (A), and the vertical axis on the right represents the device temperature, in Kelvin (K). Figure 7The novel structure is the VDMOSFET structure mentioned in the present invention, which is composed of Figure 7 It can be seen that the short-circuit peak current of the VDMOSFET device proposed in the present invention is lower during the short-circuit process, which is about 50% lower than that of the traditional square symmetrical half-cell structure, and the short-circuit maximum temperature is reduced by about 270K. This reduction is the most obvious among the existing short-circuit reinforcement methods.
[0079] Taking the above advantages into consideration, it can be seen that the square asymmetric half-cell structure proposed in the present invention not only maintains the low specific on-resistance advantage of the traditional square symmetric half-cell structure, but also improves the short-circuit resistance of the device, which is not achievable by existing short-circuit solutions.
[0080] In summary, the VDMOSFET with high short-circuit resistance provided by the embodiment of the present invention is composed of several new square asymmetric half-cell structures. The square asymmetric half-cell structure has the characteristics of a square cell with the highest channel density, and compared with the existing traditional square cell structure, the square asymmetric half-cell structure narrows the JFET area, so that during the short-circuit process, the current path narrowing caused by the expansion of the depletion region under high voltage is more obvious, thereby making the device resistance increase effect under high voltage more obvious than the traditional structure, all current peaks are lower, and the device heats up more slowly; the square asymmetric half-cell structure proposed in the present invention can be well adapted to the sub-alignment process without increasing any preparation cost. Without changing any manufacturing process, the short-circuit capability of the device can be improved and the specific on-resistance of the device can be reduced by changing the structure of the device.
[0081] Second, see Figure 8 The embodiment of the present invention provides a method for preparing a VDMOSFET with high short-circuit resistance, wherein the VDMOSFET includes a plurality of square asymmetric half-cell structures; the method for preparing two adjacent square asymmetric half-cell structures includes:
[0082] S10, obtaining substrate layer 1.
[0083] The material of the substrate layer 1 in the embodiment of the present invention can be SiC, but is not limited to SiC. Preferably, the substrate layer 1 is an N-type SiC substrate layer with a thickness of 360 μm and a doping concentration of 5e18 cm -3 -1e20cm -3 .
[0084] S20 , epitaxially growing an N-epitaxial layer 2 on the upper surface of the substrate layer 1 .
[0085] As shown in FIG. 9( a ), in the embodiment of the present invention, an N-epitaxial layer 2 is epitaxially grown on the substrate layer 1 using a MOCVD (Metal-Organic Chemical Vapor Deposition) process, and the doping concentration of the N-epitaxial layer 2 is preferably 1e15 cm -3 ~5e16cm -3 .
[0086] S30 , performing ion implantation on the N-epitaxial layer 2 to form a JFET region 3 .
[0087] As shown in FIG. 9( b ), in the embodiment of the present invention, an ion implantation process is used to implant ions into the N-epitaxial layer 2 to form a JFET region 3. Preferably, the JFET region 3 is doped with N-type doping at a doping concentration of 5e15 cm -3 ~5e17cm -3 .
[0088] S40 , forming a PWELL injection mask 10 on the JFET region 3 , and performing ion injection on the surface of the JFET region 3 outside the PWELL injection mask 10 , forming a first PWELL 41 at one end in the JFET region 3 , and forming a second PWELL 42 near the other end in the JFET region 3 .
[0089] As shown in FIG. 9( c ), in the embodiment of the present invention, a SiO2 mask is grown on the JFET region 3 to form a PWELL injection mask 10, and an ion injection process is used to perform ion injection on the surface of the JFET region 3 not covered by the PWELL injection mask 10, thereby forming a first PWELL 41 at one end of the JFET region 3 and a second PWELL 42 at the other end close to the JFET region 3. Preferably, the first PWELL 41 and the second PWELL 42 are doped with P-type doping and the doping concentration is 1e16 cm -3 ~1e19cm -3 .
[0090] S50, forming an N+ source region injection mask 11 on the PWELL injection mask 10, on part of the first PWELL 41 and on part of the second PWELL 42 by a self-alignment process, and performing ion implantation on the surface of the first PWELL 41 and the surface of the second PWELL 42 outside the N+ source region injection mask 11, forming a first N+ source region 51 in the first PWELL 41, and forming a second N+ source region 52 in the second PWELL 42.
[0091] As shown in FIG9(d), the embodiment of the present invention uses a self-aligned process to continue growing a SiO2 mask on the PWELL injection mask 10, on a portion of the first PWELL 41, and on a portion of the second PWELL 42 to form an N+ source region injection mask 11. For example, a 0.5 μm SiO2 mask can be grown again because the channel length is generally 0.5 μm. Then, an ion implantation process is used to perform ion implantation on the area not covered by the N+ source region injection mask 11 to form a first N+ source region 51 in the first PWELL 41 and a second N+ source region 52 in the second PWELL 42. Preferably, the first N+ source region 51 and the second N+ source region 52 are doped with N-type doping and the doping concentration is 5e18 cm -3 ~1e20cm -3 .
[0092] Here is an explanation of the self-alignment process, that is, the ion implantation window is narrowed by deposition and etching instead of photolithography. The advantage of this method is that the narrowing distance can be precisely controlled, and one less photolithography step can reduce costs. However, the self-alignment process requires that the first PWELL41 and the second PWELL42 cannot be too small, and must be at least much larger than 2 times the channel length, which is four times the length of the self-aligned narrowing. Otherwise, the narrowed areas will be connected together and cannot be etched to open windows. At the same time, a certain length is required to form an ohmic contact, which needs to be considered during the process.
[0093] S60, remove the PWELL injection mask 10 and the N+ source region injection mask 11, form a P+ source region injection mask 12 on part of the first PWELL 41, part of the JFET region, part of the second PWELL 42, the first N+ source region 51, and the second N+ source region 52, and perform ion implantation on the first PWELL 41 and the JFET region outside the P+ source region injection mask 12 to form a first P+ source region 61, and perform ion implantation on the second PWELL 42 and the JFET region 3 outside the P+ source region injection mask 12 to form a second P+ source region 62.
[0094] In the embodiment of the present invention, the PWELL injection mask 10 and the N+ source region injection mask 11 are first removed, and then a SiO2 mask is grown on part of the first PWELL 41, part of the JFET region, part of the second PWELL 42, the first N+ source region 51, and the second N+ source region 52 to form a P+ source region injection mask 12. An ion implantation process is used to perform ion implantation on the area not covered by the P+ source region injection mask 12. As shown in FIG. 9(e), a first P+ source region 61 is implanted on the first PWELL 41 and the JFET region, and a second P+ source region 62 is implanted on the second PWELL 42 and the JFET region 3. At this time, the width of the JFET region after the first P+ source region 61 is formed is half of the width of the JFET region before the first P+ source region 61 is formed. Preferably, the doping of the first P+ source region 61 and the second P+ source region 62 is P-type doping with a doping concentration of 1e18cm -3 ~1e20cm -3 , the width of the JFET region after forming the first P+ source region 61 is 0.6μm~3μm; after injection, the injection depth of the first P+ source region 61 and the second P+ source region 62 is preferably greater than the injection depth of the first PWELL41 and the second PWELL42, so that the depletion effect is stronger, thereby providing better protection during the short circuit process and improving the device blocking capability.
[0095] S70 , remove the P+ source region implantation mask 12 , form a first gate electrode 71 on the first PWELL 41 and a portion of the first N+ source region 51 , and form a second gate electrode 72 on the second PWELL 42 , a portion of the second N+ source region 52 and a portion of the JFET region.
[0096] In the embodiment of the present invention, the P+ source region injection mask 12 is first removed, and then, as shown in FIG. 9( f), a gate metal is deposited on the first PWELL 41 and a portion of the first N+ source region 51 using an electron beam evaporation process to form a first gate electrode 71, and a gate metal is deposited on the second PWELL 42, a portion of the second N+ source region 52 and a portion of the JFET region to form a second gate electrode 72, wherein the gate metal is preferably polysilicon.
[0097] S80 , forming a first source electrode 81 on a portion of the first N+ source region 51 and a portion of the first P+ source region 61 , and forming a second source electrode 82 on a portion of the second N+ source region 52 and a portion of the second P+ source region 62 .
[0098] As shown in Figure 9(g), an embodiment of the present invention uses an electron beam evaporation process to deposit source metal on part of the first N+ source region 51 and part of the first P+ source region 61 to form a first source electrode 81, and deposits source metal on part of the second N+ source region 52 and part of the second P+ source region 62 to form a second source electrode 82, and the source metal is preferably Al.
[0099] S90, forming a drain electrode on the lower surface of the substrate layer.
[0100] As shown in FIG. 9( h ), in the embodiment of the present invention, a drain metal is deposited on the lower surface of the substrate layer by an electron beam evaporation process to form a drain electrode 9 , and preferably the drain metal is Al.
[0101] The VDMOSFET prepared by S10~S90 in the embodiment of the present invention has a first N+ source region 51 and a second N+ source region 52 in a square structure when viewed from above; the first P+ source region 61 and the first N+ source region 51 form a square structure, the second P+ source region 62 and the second N+ source region 52 form a square structure, and the first P+ source region 61 and the second P+ source region 62 are L-shaped structures respectively; the first PWELL41, the first P+ source region 61 and the first N+ source region 51 form a square structure, and the second PWELL42, the second P+ source region 62 and the second N+ source region 52 form a square structure. Seen from a top-down direction, with the center of the first N+ source region 51 as the axis of symmetry, the half-cell structure including the first N+ source region 51, the first PWELL 41 and the first P+ source region 61 is an asymmetric structure both vertically and horizontally; with the center of the second N+ source region 52 as the axis of symmetry, the half-cell structure including the second N+ source region 52, the second PWELL 42 and the second P+ source region 62 is an asymmetric structure both vertically and horizontally.
[0102] In summary, the method for preparing a VDMOSFET with high short-circuit resistance proposed in an embodiment of the present invention is compatible with the existing general process of SiC power MOSFET. Compared with the traditional asymmetric structure, it reduces the process steps and reduces the production cost. It uses simpler process steps to reduce the current density in the JFET area and increase the heat dissipation area during the short-circuit process, thereby achieving the purpose of suppressing the short-circuit current and reducing short-circuit thermal burnout, thereby improving the device's specific conduction performance and increasing the short-circuit time of the device.
[0103] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0104] Although the present invention is described herein in conjunction with various embodiments, in the process of implementing the claimed invention, those skilled in the art may understand and implement other variations of the disclosed embodiments by viewing the specification and its drawings. In the specification, the word "comprising" does not exclude other components or steps, and "a" or "one" does not exclude multiple situations. Certain measures are recorded in different embodiments, but this does not mean that these measures cannot be combined to produce good results.
[0105] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.
Claims
1. A VDMOSFET with high short-circuit resistance, characterized in that: The VDMOSFET includes a plurality of square asymmetric half-cell structures; a structure formed by two adjacent square asymmetric half-cell structures includes: A substrate layer (1); An N-epitaxial layer (2), located on the upper surface of the substrate layer (1); A first PWELL (41), located at one end of the N-epitaxial layer (2); A first P+ source region (61), located in the N-epitaxial layer (2) and adjacent to the first PWELL (41); A second P+ source region (62) located at the other end of the N- epitaxial layer (2); A second PWELL (42) is located in the N-epitaxial layer (2) and adjacent to the second P+ source region (62); wherein a JFET region (3) is formed between the first P+ source region (61) and the second PWELL (42); A first N+ source region (51), located in the first PWELL (41) and adjacent to the first P+ source region (61); A second N+ source region (52), located in the second PWELL (42) and adjacent to the second P+ source region (62); A first gate electrode (71), located on the first PWELL (41) and a portion of the first N+ source region (51); A second gate electrode (72) located on the second PWELL (42), a portion of the second N+ source region (52) and a portion of the JFET region (3); A first source electrode (81), located on a portion of the first N+ source region (51) and a portion of the first P+ source region (61); A second source electrode (82) is located on a portion of the second N+ source region (52) and a portion of the second P+ source region (62); A drain electrode (9) located on the lower surface of the substrate layer (1); Wherein, viewed from the top, with the center of the first N+ source region (51) as the axis of symmetry, the half-cell structure including the first N+ source region (51), the first PWELL (41) and the first P+ source region (61) is an asymmetric structure in both the top and bottom and in both the left and right directions; with the center of the second N+ source region (52) as the axis of symmetry, the half-cell structure including the second N+ source region (52), the second PWELL (42) and the second P+ source region (62) is an asymmetric structure in both the top and bottom and in both the left and right directions; the first N+ source region (51) and the second N+ source region (52) are ) is a square structure; the first P+ source region (61) and the first N+ source region (51) form a square structure, the second P+ source region (62) and the second N+ source region (52) form a square structure, and the first P+ source region (61) and the second P+ source region (62) are L-shaped structures respectively; the first PWELL (41), the first P+ source region (61) and the first N+ source region (51) form a square structure, and the second PWELL (42), the second P+ source region (62) and the second N+ source region (52) form a square structure.
2. The VDMOSFET with high short-circuit resistance according to claim 1, characterized in that: The width of the JFET region (3) is 0.6 μm to 3 μm.
3. The VDMOSFET with high short-circuit resistance according to claim 1, characterized in that: The width of the first P+ source region (61) is the same as the width of the second P+ source region (62), and the width of the first P+ source region (61) is smaller than the width of the first N+ source region (51) and the width of the second N+ source region (52).
4. The VDMOSFET with high short-circuit resistance according to claim 1, characterized in that: The maximum process size of each square asymmetric half-cell structure is 4.5μm.
5. A method for preparing a VDMOSFET with high short-circuit resistance, characterized in that: The VDMOSFET comprises a plurality of square asymmetric half-cell structures; a method for preparing two adjacent square asymmetric half-cell structures comprises: Obtaining a substrate layer (1); Epitaxially growing an N-epitaxial layer (2) on the upper surface of the substrate layer (1); Performing ion implantation on the N-epitaxial layer (2) to form a JFET region (3); Forming a PWELL injection mask on the JFET region (3), and performing ion implantation on the surface of the JFET region (3) outside the PWELL injection mask, forming a first PWELL (41) at one end in the JFET region (3), and forming a second PWELL (42) at the other end close to the JFET region (3); Forming an N+ source region injection mask on the PWELL injection mask, on a portion of the first PWELL (41) and on a portion of the second PWELL (42) using a self-alignment process, and performing ion implantation on the surface of the first PWELL (41) and the surface of the second PWELL (42) outside the N+ source region injection mask to form a first N+ source region (51) in the first PWELL (41) and a second N+ source region (52) in the second PWELL (42); The PWELL injection mask and the N+ source region injection mask are removed, and a P+ source region injection mask is formed on part of the first PWELL (41), part of the JFET region (3), part of the second PWELL (42), the first N+ source region (51), and the second N+ source region (52); ion implantation is performed on the first PWELL (41) and the JFET region (3) outside the P+ source region injection mask to form a first P+ source region (61); and ion implantation is performed on the second PWELL (42) and the JFET region (3) outside the P+ source region injection mask to form a second P+ source region (62); removing the P+ source region implantation mask, forming a first gate electrode (71) on the first PWELL (41) and a portion of the first N+ source region (51), and forming a second gate electrode (72) on the second PWELL (42), a portion of the second N+ source region (52) and a portion of the JFET region (3); Forming a first source electrode (81) on part of the first N+ source region (51) and part of the first P+ source region (61), and forming a second source electrode (82) on part of the second N+ source region (52) and part of the second P+ source region (62); forming a drain electrode (9) on the lower surface of the substrate layer (1); Wherein, viewed from the top, with the center of the first N+ source region (51) as the axis of symmetry, the half-cell structure including the first N+ source region (51), the first PWELL (41) and the first P+ source region (61) is an asymmetric structure in both the top and bottom and in both the left and right directions; with the center of the second N+ source region (52) as the axis of symmetry, the half-cell structure including the second N+ source region (52), the second PWELL (42) and the second P+ source region (62) is an asymmetric structure in both the top and bottom and in both the left and right directions; the first N+ source region (51) and the second N+ source region (52) are ) is a square structure; the first P+ source region (61) and the first N+ source region (51) form a square structure, the second P+ source region (62) and the second N+ source region (52) form a square structure, and the first P+ source region (61) and the second P+ source region (62) are L-shaped structures respectively; the first PWELL (41), the first P+ source region (61) and the first N+ source region (51) form a square structure, and the second PWELL (42), the second P+ source region (62) and the second N+ source region (52) form a square structure.
6. The method for preparing a VDMOSFET with high short-circuit resistance according to claim 5, characterized in that: After the first P+ source region (61) is formed, the width of the JFET region (3) is 0.6 μm to 3 μm.
7. The method for preparing a VDMOSFET with high short-circuit resistance according to claim 5, characterized in that: The maximum process size of each square asymmetric half-cell structure in the prepared VDMOSFET is 4.5μm.
