A junction field-effect transistor
By setting a parasitic base region in the junction field-effect transistor and controlling its doping concentration and thickness, the conduction threshold is raised, solving the problem of large gate-source leakage current in traditional junction field-effect transistors and improving the stability and frequency characteristics of the device.
Patent Information
- Application Number
- CN202611637991.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-22
- Publication Date
- 2026-11-17
AI Technical Summary
Traditional junction field-effect transistors (JFETs) suffer from large gate-source leakage current during turn-off, leading to increased static power consumption, decreased high-temperature reliability, and impacting device performance.
By setting a parasitic base region of the second doping type in the junction field-effect transistor, and controlling its doping concentration and thickness, the punch-through voltage of the parasitic transistor is designed to be a preset multiple of the gate-source rated voltage, thereby raising the conduction threshold, suppressing the punch-through of the parasitic transistor, and reducing the blocking leakage current.
It effectively suppresses the punch-through of parasitic transistors, reduces gate-source leakage current, improves the operating stability and reliability of the device, reduces high-temperature blocking leakage current, and improves switching speed and frequency characteristics.
Smart Images

Figure CN123124413A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of semiconductor device technology, and more specifically, to a junction field-effect transistor. Background Technology
[0002] The trench junction field-effect transistor CN120769543B, such as Figure 1 As shown in the attached figures, the reference numerals are as follows: First epitaxial layer 21, second epitaxial layer 22, substrate 23, source region 24, gate region 26, channel shielding region 28, dielectric layer 29, source ohmic contact 210, gate ohmic contact 211, drain ohmic contact 212, polysilicon carbide layer 213, and protrusion on epitaxial layer 214.
[0003] When the device is turned off, the P-type gate region 26, the N-type first epitaxial layer 21, and the P-type channel shielding region 28 constitute a parasitic PNP transistor. When subjected to high voltage during turn-off, the parasitic PNP transistor punches through, and the gate carriers are directly discharged to the source through the punch-through path, ultimately causing an abnormal increase in the gate-source leakage current. The large gate-source leakage current directly results in increased static power consumption, decreased high-temperature reliability, and a heavier burden on the drive circuit.
[0004] The large gate-source leakage current of traditional junction field-effect transistors has a significant impact on device performance and is a technical problem that urgently needs to be solved in this field.
[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this application, and therefore may contain information that is not part of the prior art known to those skilled in the art. Summary of the Invention
[0006] This application provides a junction field-effect transistor to solve the technical problem of large gate-source leakage current in traditional junction field-effect transistors.
[0007] This application provides a junction field-effect transistor, including: Second type of doped substrate; A second doped epitaxial layer is disposed on the substrate; The first type of doped channel shielding region is located within the epitaxial layer; A parasitic base region of the second doping type is formed on the top surface of the channel shielding region, and the doping concentration of the parasitic base region is greater than the doping concentration of the epitaxial layer. A gate region of the first doped type is formed on the parasitic base region; The gate region, the parasitic base region, and the channel shield region constitute a parasitic transistor. The doping concentration and thickness of the parasitic base region are configured such that the punch-through voltage of the parasitic transistor is designed to be a preset multiple of the gate-source rated voltage, wherein the preset multiple is greater than 1.
[0008] The embodiments of this application, by adopting the above technical solutions, have the following technical effects: The junction field-effect transistor (JFET) of this application embodiment has a separately configured parasitic base region. By controlling the doping concentration and thickness of the parasitic base region, the punch-through voltage design value of the parasitic transistor is configured. The punch-through voltage design value of the parasitic transistor is a preset multiple of the gate-source rated voltage, and the preset multiple is greater than 1. That is, the conduction threshold of the parasitic PNP transistor is raised, effectively suppressing the punch-through of the parasitic transistor and reducing the blocking leakage current of the parasitic transistor, thereby suppressing the punch-through effect.
[0009] Since the punch-through voltage of the parasitic transistor is designed to be greater than the gate-source rated voltage, when the gate-source voltage Vgs turns off the junction field-effect transistor, the parasitic transistor will always be suppressed from punching through as long as the absolute value of the gate-source voltage Vgs is less than the punch-through voltage design value.
[0010] When a parasitic transistor punches through, the current will suddenly surge. The simultaneous action of high current and high voltage will cause severe local heating of the device, which may burn out the PN junction. Even if it does not burn out immediately, repeated punching through will accelerate the aging of the device. Attached Figure Description
[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of the existing trench junction field-effect transistor CN120769543B; Figure 2 This is a schematic diagram of a first implementation of the junction field-effect transistor of this application; Figure 3 This is a schematic diagram of a second implementation of the junction field-effect transistor of this application; Figure 4 This is a schematic diagram of a third implementation of the junction field-effect transistor of this application; Figure 5 This is a schematic diagram of the fourth implementation of the junction field-effect transistor of this application; Figure 6 This is a schematic diagram of the fifth implementation of the junction field-effect transistor of this application; Figure 7This is a schematic diagram of the sixth implementation of the junction field-effect transistor of this application; Figure 8 for Figure 5 A schematic diagram of the conduction state; Figure 9 for Figure 5 A schematic diagram of the off state; Figure 10 for Figure 5 The diagram shows a method for fabricating a junction field-effect transistor, including step S51. Figure 11 for Figure 5 The diagram shows a step S52 of the method for fabricating a junction field-effect transistor. Figure 12 for Figure 5 The diagram shows a method for fabricating a junction field-effect transistor, including step S53. Figure 13 for Figure 5 The diagram shows a step S54 of the method for fabricating a junction field-effect transistor. Figure 14 for Figure 6 The diagram shows a method for fabricating a junction field-effect transistor, including step S61. Figure 15 for Figure 6 The diagram shows a method for fabricating a junction field-effect transistor, including step S62. Figure 16 for Figure 6 The diagram shows a method for fabricating a junction field-effect transistor, including step S63. Figure 17 Simulation diagram of gate-source leakage current density distribution after applying gate-source voltage to the existing structure; Figure 18 Simulation diagram of gate-source leakage current density distribution after applying gate-source voltage to this application; Figure 19 The graphs of gate-source leakage current Igss and gate-source voltage Vgs for existing structures and this application are shown. Figure 20 The graphs show the gate-source capacitance Cgs and drain-source voltage Vds of this application and existing structures. Figure 21 The simulation diagram shows the electric field distribution during breakdown of the existing structure excluding the junction gate capacitance control region 6. Figure 22 When the present application, including the junction gate capacitance control region 6, breaks down. Figure 5 Simulation diagram of electric field distribution at the location indicated by the dashed box; Figure 23 for Figure 22 Simulation diagram of doping distribution; Figure 24 The blocking current I-voltage V curves for this application and existing structures; Figure 25 The current density simulation diagram of the existing technology at the breakdown moment at a high temperature of 175℃ is shown. Figure 26 The simulation diagram of the current density of this application at the breakdown moment at a high temperature of 175℃ is shown. Figure 27 The I-V curves of the blocking current and the prior art are shown at the breakdown time at a high temperature of 175°C.
[0012] Figure label: Background technology:
[0014] First epitaxial layer 21, second epitaxial layer 22, substrate 23, source region 24, gate region 26. 28. Channel shielding area; 29. Dielectric layer; 210. Source ohmic contact. Gate ohmic contact 211, drain ohmic contact 212, polycrystalline silicon carbide layer 213. Protrusion 214 on the epitaxial layer; This application: Substrate 1, Epitaxial layer 2, Current spreading layer 3, First parasitic base region 41, second parasitic base region 42. Channel region 5, junction-gate capacitance control region 6, Source region 71, source region trailing region 7, 8. Grid region; 9. Channel shielding region; 10. Channel shielding region metal contact; 11. Channel shielding region dielectric. Gate metal 12, source metal 13, drain metal 14 Current spread layer base 30, The first parasitic base region is based on 410, and the second parasitic base region is based on 420. Source region base 710, source region trailing region base 70. Detailed Implementation
[0015] To make the technical solutions and advantages of the embodiments of this application clearer, the exemplary embodiments of this application will be described in further detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.
[0016] Example 1 like Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 As shown, the junction field-effect transistor of this application embodiment includes: Substrate 1 of the second doping type; A second doped epitaxial layer 2 is disposed on the substrate 1; The first doped type of channel shielding region 9 is disposed within the epitaxial layer 2; The parasitic base region of the second doping type is formed on the top surface of the channel shielding region 9, and the doping concentration of the parasitic base region is greater than the doping concentration of the epitaxial layer 2; the parasitic base region has at least two implementation methods: a first parasitic base region 41 and a second parasitic base region 42; A gate region 8 of the first doped type is formed on the parasitic base region; The gate region 8, the parasitic base region, and the channel shielding region 9 constitute a parasitic transistor. The doping concentration and thickness of the parasitic base region are configured such that the punch-through voltage of the parasitic transistor is designed to be a preset multiple of the gate-source rated voltage, and the preset multiple is greater than 1.
[0017] The gate region 8 serves as the collector of the parasitic transistor, the parasitic base region serves as the base of the parasitic transistor, and the channel shielding region 9 serves as the emitter of the parasitic transistor.
[0018] In practice, the gate-source voltage Vgs turns off the junction field-effect transistor (at this time, the gate-source voltage is the gate-source reverse bias voltage), and when the absolute value of the gate-source voltage Vgs is less than the design value of the punch-through voltage, the parasitic transistor remains off (i.e., the parasitic transistor is suppressed from conducting), and the gate-source blocking leakage current of the device is reduced.
[0019] The punch-through voltage design value of the parasitic transistor is a preset multiple of the gate-source rated voltage. Within the normal operating gate-source voltage range of the device (less than the gate-source rated voltage), the punch-through voltage will necessarily be less than the design value, and the parasitic transistor will always be in the cut-off state, thus not interfering with the normal blocking characteristics of the device.
[0020] That is, when the device is turned off within the gate-source voltage range during normal operation, the parasitic transistor is always in the off state.
[0021] This avoids the problem of parasitic transistors prematurely punching through and forming additional leakage current paths, effectively reducing the gate-source blocking leakage current of the device.
[0022] In practice, the gate-source voltage Vgs turns off the junction field-effect transistor. Only when the absolute value of the gate-source voltage Vgs is greater than or equal to the design value of the punch-through voltage will the current of the parasitic transistor surge due to the punch-through of the gate junction reverse bias depletion layer.
[0023] The problems that arise are as follows: When a parasitic transistor punches through, the current will suddenly surge. The simultaneous action of high current and high voltage will cause severe local heating of the device, which may burn out the PN junction. Even if it does not burn out immediately, repeated punching through will accelerate the aging of the device.
[0024] In practice, the preset multiple is greater than or equal to 1.2.
[0025] The junction field-effect transistor (JFET) of this application embodiment has a separately configured parasitic base region. By controlling the doping concentration and thickness of the parasitic base region, the punch-through voltage design value of the parasitic transistor is configured. The punch-through voltage design value of the parasitic transistor is a preset multiple of the gate-source rated voltage, and the preset multiple is greater than 1. That is, the punch-through threshold of the parasitic PNP transistor is raised, effectively suppressing the punch-through of the parasitic transistor and reducing the blocking leakage current of the parasitic transistor, thereby suppressing the punch-through effect.
[0026] Since the punch-through voltage of the parasitic transistor is designed to be greater than the gate-source rated voltage, when the gate-source voltage Vgs turns off the junction field-effect transistor, the parasitic transistor will always be suppressed from punching through as long as the absolute value of the gate-source voltage Vgs is less than the punch-through voltage design value.
[0027] The mechanism of the junction field-effect transistor in the embodiments of this application will be analyzed in detail below: The normal states of the junction field-effect transistor in this application embodiment include an on-state and an off-state: When the device is in the on-state, the gate-source voltage is greater than the threshold voltage Vth, the depletion region between the gate regions 8 narrows, the channel opens, and the device can carry the main current. In the off-state, the drive circuit needs to continuously apply a reverse gate-source voltage less than the threshold voltage Vth, so that the gate region is pinched off by the depletion region, thus blocking the current.
[0028] The following example, using P-type as the first doping type and N-type as the second doping type, illustrates the mechanism for suppressing the punch-through effect of parasitic transistors: The gate region 8, parasitic base region, and channel shielding region 9 inside the device together constitute a parasitic PNP transistor. The gate region 8 is the collector of the parasitic PNP transistor, the N-type parasitic base region is the base of the parasitic PNP transistor, and the P-type channel shielding region 9 is the emitter of the parasitic PNP transistor.
[0029] When the reverse bias voltage of the collector junction of a parasitic PNP transistor increases to a certain extent, the space charge regions (depletion layers) of the collector junction and emitter junction come into contact with each other. When the reverse bias voltage of the collector junction is large, the depletion region extends into the N-type parasitic base region, which serves as the base. The emitter junction barrier is reduced due to the collector junction voltage. The reduction of the emitter junction barrier will cause a small change in the collector junction voltage to result in a large change in the current. This punch-through phenomenon will cause a sharp increase in the gate-source current, degrading the electrical performance and reliability of the device.
[0030] To this end, this application raises the punch-through threshold of the parasitic PNP transistor by specifically configuring the vertical thickness and doping concentration of the N-type parasitic base region, thereby effectively suppressing the parasitic punch-through effect.
[0031] Specifically, by increasing the vertical thickness and doping concentration of the parasitic base region, when a reverse bias voltage is applied to the collector junction, its depletion layer will primarily extend towards the less doped base (N). As long as the width and doping concentration of the N-type parasitic base region serving as the base are sufficient, this depletion layer will not touch the depletion layer of the emitter junction; The theoretical basis for the above structural optimization can be supported by the transistor punch-through voltage formula, whereby the punch-through voltage of the parasitic transistor satisfies:
[0032] In the formula: is the punch-through voltage of the parasitic transistor, which is the gate negative bias voltage in this device; e is the elementary charge; N represents the base region thickness of the parasitic base region. B The doping concentration of the parasitic base region; N C εs represents the collector concentration of the parasitic transistor; εs represents the semiconductor dielectric constant.
[0033] As can be seen from the above formula, the gate negative bias voltage and the base thickness of the parasitic base region are related. It is proportional to the square of the number of ions and is related to the doping concentration N in the parasitic base region. B Proportional. While keeping other structural parameters constant, increasing the thickness of the parasitic base region... Increasing the doping concentration in the parasitic base region can significantly reduce the collector current of the parasitic transistor and increase the absolute value of the design punch-through voltage of the parasitic transistor.
[0034] This application increases the vertical thickness of the parasitic base region, correspondingly improving the formula... The parameter values directly suppress gate-source leakage current; increasing the doping concentration in the parasitic base region can effectively reduce the minority carrier concentration in the base region. This further weakens the carrier transport capability. The two structural optimization methods work together to effectively raise the punch-through threshold of the parasitic PNP transistor (i.e., the design value of the punch-through voltage of the parasitic transistor) from the physical mechanism level, suppress the punch-through of the parasitic transistor, increase the absolute value of the gate negative bias voltage, reduce the gate-source leakage current of the device, and improve the overall operating stability and reliability of the device.
[0035] During implementation, such as Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 As shown, the junction field-effect transistor also includes: The second doped type of junction gate capacitance control region 6 is disposed on the two spaced gate regions 8; A second doped current spreading layer 3 is formed between and below the channel shielding region 9; The junction gate capacitance control region 6, the gate region 8, and the parasitic base region 41 constitute the first JFET region; the parasitic base region, the channel shielding region 9, the current spreading layer 3, and the epitaxial layer 2 constitute the second JFET region. The withstand voltage of the first JFET region is configured such that the second JFET region is turned off before the first JFET region breaks down.
[0036] That is, the first JFET region and the second JFET region are cascaded vertically. The first JFET region and the second JFET region share a parasitic base region—which is both the drain of the first JFET region and the source of the second JFET region, forming an intermediate node in series.
[0037] The operating mode in which the breakdown voltage of the first JFET region is configured such that the second JFET region is turned off before the first JFET region undergoes a punch-through is as follows: During device shutdown: the second JFET region is shut down first; Steady state after turn-off: The voltage drops mainly to the second JFET region, which bears the high voltage. Final withstand voltage: Before the first JFET region itself breaks down, the second JFET region in the lower layer has been reliably turned off. This is a conventional SiCPN junction avalanche breakdown (PN junction formed by channel shielding region 9 and current spread layer 3). Due to the suppression of impact ionization by phonon scattering, the high-temperature blocking voltage is high.
[0038] In this application, the second JFET region is turned off first, and the channel of the first JFET region does not carry a large current, thus reducing the high-temperature blocking leakage current.
[0039] The gate capacitance control region 6 is prepared by ion implantation / epitaxy, with doping continuously transitioning from the gate capacitance control region 6 to the gate region 8. Therefore, the impurity distribution of the first PN junction formed by the gate capacitance control region 6 and the gate region 8 approximately satisfies the linearly graded junction model, and its gate capacitance can be analyzed using the barrier capacitance formula per unit area of the linearly graded PN junction.
[0040] The formula for the barrier capacitance per unit area of a linearly graded PN junction is as follows: ; in, The barrier capacitance per unit area of the PN junction is the gate capacitance per unit area of this device. Impurity concentration gradient at the PN junction; : Coordinates, along the thickness direction perpendicular to the metallurgical structure; They're gone. The change in net doping over this short distance; The built-in potential of the PN junction is determined by the doping of both the gate capacitance control region 6 and the gate region 8.
[0041] When the device is working, a reverse bias voltage is applied to the PN junction at the gate electrode.
[0042] : is the dielectric constant of this semiconductor material (Si or SiC).
[0043] Therefore, this application reduces the net doping concentration of the junction gate capacitance control region 6 while ensuring voltage withstand capability, thereby reducing the impurity gradient and lowering the gate capacitance. A smaller gate capacitance results in increased device switching speed and faster device operation. The device can support higher circuit operating frequencies. In power conversion circuit systems, the size of passive components such as inductors, transformers, and filter capacitors can be significantly reduced as the switching frequency increases; therefore, low-gate capacitance junction field-effect transistors are beneficial for achieving overall circuit miniaturization.
[0044] Furthermore, the PN junction formed by the gate capacitance control region and the gate region satisfies the PN junction electric field expression: ; Within the range e is the electron charge. The doping concentration of the junction gate capacitance control region 6 is... The dielectric constant of a semiconductor is _____. This is the distance from the outer boundary of the junction gate capacitance control region 6 to the PN junction metallurgical junction. For the PN junction metallurgical junction position to Any position between them. From this electric field relationship, it can be seen that the lower the doping concentration of the junction gate capacitance control region 6, the weaker the internal electric field of the PN junction formed by the junction gate capacitance control region 6 and the gate region 8, and the less likely the PN junction formed by the junction gate capacitance control region 6 and the gate region 8 is to undergo reverse breakdown.
[0045] In practice, the first PN junction formed by the junction gate capacitance control region 6 and the gate region 8, and the second PN junction formed by the gate region 8 and the parasitic base region; The first PN junction and the second PN junction are connected in parallel, and their equivalent capacitance is the gate-source capacitance.
[0046] The first and second PN junctions connected in parallel have an equivalent capacitance that is the gate-source capacitance. Reducing the capacitance of one of the parallel PN junctions lowers the overall capacitance, which is the fundamental reason for the reduction in gate-source capacitance. In other words, the junction-gate capacitance control region 6 optimizes the PN junction / junction capacitance structure between the gate and source, reducing the gate-source capacitance. Optimizing the gate-source capacitance helps improve the device's switching speed and high-frequency characteristics.
[0047] In practice, the gate region 8 is a P-type region and the junction gate capacitance control region 6 is an N-type region, which together form a parasitic PN junction diode; The junction gate capacitance control region 6 constitutes the series body resistance of the parasitic PN junction diode. When the gate-source voltage is applied, it can generate a voltage divider, reduce the actual bias voltage of the parasitic PN junction diode, and suppress the forward gate-source leakage current. When the gate-source voltage Vgs is applied and controlled to be greater than the threshold voltage Vth, a conducting channel is formed in channel region 5. The depletion lines between the shielded trenches 9 contract and do not overlap, forming a conductive channel; The forward gate-source leakage current path of the parasitic PN junction diode passes sequentially through gate region 8, junction-gate capacitance control region 6, source region tailing region 7, and source region 71.
[0048] During implementation, such as Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 As shown, the junction field-effect transistor also includes: A second doped channel region 5 is disposed between the two gate regions 8; The source region 71 and the source region tail region 7 are arranged from top to bottom; the source region tail region 7 is located above the junction gate capacitance control region 6. The parameters of the channel region 5 are determined by the spacing, doping concentration and threshold voltage of the gate region 8.
[0049] If the junction gate capacitance control region 6 is not set, the source tail region 7 will interfere with the doping distribution of the channel region 5, resulting in large high-temperature blocking leakage current and deterioration of device withstand voltage.
[0050] This application adds a junction-gate capacitance control region 6, which acts as an isolation and buffer, reducing the interference of the source tail region 7 on the channel region 5. Thus, after the doping of the channel region 5 is no longer interfered with by the source tail region 7, the blocking voltage of the first JFET region itself is improved.
[0051] Thus, this application solves the technical problems of traditional junction field-effect transistors, such as low gate-source reverse bias voltage, large gate-source leakage current, large high-temperature blocking leakage current, low high-temperature blocking voltage, and large gate-source capacitance.
[0052] Example 2 The junction field-effect transistor of this application embodiment, based on Embodiment 1, also has the following features.
[0053] Parasitic base regions can be implemented in several ways, as detailed below: As the first optional method, such as Figure 2 , Figure 5 As shown, the parasitic base region is the first parasitic base region 41; Each of the channel shielding regions 9 is provided with a first parasitic base region 41, which is formed above the junction of the corresponding channel shielding region 9 and the epitaxial layer 2.
[0054] A portion of the first parasitic base region 41 is located on the top surface of the channel shielding region 9, and another portion is located on the top surface of the epitaxial layer 2.
[0055] As a second alternative, such as Figure 3 , Figure 4 , Figure 6 , Figure 7 As shown, each cell contains two spaced-apart channel shielding regions 9; The parasitic base region is the second parasitic base region 42; The second parasitic base region 42 is located across the top of two adjacent channel shielding regions 9 and covers the region in the epitaxial layer located between the two channel shielding regions 9.
[0056] The second parasitic base region 42 simultaneously overlaps the two side channel shielding regions 9 and the epitaxial layer 2 between the two channel shielding regions. The second parasitic base region 42, the two side channel shielding regions 9, and the gate region 8 can each form a parasitic transistor structure, which simplifies the manufacturing process.
[0057] In practice, the parasitic base region is formed epitaxially; Alternatively, the parasitic base region can be formed by injection.
[0058] Epitaxial growth of the parasitic base region yields a uniform doping distribution with good controllability of doping concentration and thickness. The minority carrier recombination efficiency within the parasitic base region is stable, which is beneficial for stabilizing the parameters of the parasitic transistor and improving the suppression of gate-source leakage current. Figure 4 As shown, the second parasitic base region 42 is formed by epitaxy.
[0059] The parasitic base region is formed by ion implantation, which allows for the use of photolithography to define the implantation area. This provides strong patterning capabilities and enables precise definition of the lateral position and range of the parasitic base region, thus meeting the structural requirements of the alignment between the parasitic base region and the channel shielding region, as well as the epitaxial layer. Figure 2 As shown, the first parasitic base region 41 is formed by implantation. Figure 3 As shown, the second parasitic base region 42 is formed by injection.
[0060] Both fabrication methods can achieve the preparation of parasitic base regions, and the choice can be made according to the actual device design requirements.
[0061] In practice, the doping concentration of the parasitic base region ranges from greater than or equal to 1 × 10⁻⁶. 17 cm -3 And less than or equal to 5 × 10 18 cm -3 ; The thickness of the parasitic base region is greater than or equal to 0.01 μm and less than or equal to 0.3 μm.
[0062] In the parasitic base region that meets the above requirements, after the junction field-effect transistor of this application enters the turn-off operating range, the absolute value of the gate-source reverse bias voltage can reach greater than or equal to 10V within the range that does not reach the avalanche breakdown voltage.
[0063] Example 3 The junction field-effect transistor of this application embodiment, based on Embodiment 2, also has the following features.
[0064] In practice, the doping concentration of the junction gate capacitance control region 6 is greater than or equal to 1 × 10⁻⁶. 15 cm -3 And less than or equal to 1×10 18 cm -3 The thickness of the junction gate capacitance control region 6 is greater than or equal to 0.001 μm and less than or equal to 0.3 μm.
[0065] This application reduces the doping concentration of the junction-gate capacitance control region to decrease the gate capacitance. This process must be implemented while ensuring the breakdown voltage. The breakdown voltage consists of two parts: first, the reverse breakdown voltage of the PN junction formed by the junction-gate capacitance control region and the gate region; and second, the overall breakdown voltage between the source and drain of the device. For the junction-gate PN junction breakdown voltage: the lower limit of doping in the junction-gate capacitance control region is limited to 1×10⁻⁶. 15 cm -3 The thickness is limited to 0.001μm to 0.3μm to prevent premature breakdown of the PN junction due to excessive impurity gradient.
[0066] A thickness greater than 0.3 μm increases the on-resistance. Therefore, the above parameter range is a set of mutually constraining parameter intervals, not arbitrarily selected. The above parameter range takes into account low gate capacitance, junction-gate breakdown voltage, and process feasibility, and has upper and lower limits due to physical mechanisms and process constraints.
[0067] In practice, the doping concentration of the channel region 5 is greater than or equal to 5 × 10⁻⁶. 16 cm -3 And less than or equal to 5 × 10 18 cm -3 The thickness of the channel region 5 is greater than or equal to 0.01 μm and less than or equal to 1 μm.
[0068] The range of doping concentration in channel region 5 meets the above requirements, which can ensure that the channel region has sufficient doping compensation capability to achieve effective threshold voltage regulation, while avoiding excessive compensation and excessive absolute value of threshold voltage caused by excessive doping, which leads to difficulty in device turn-off; at the same time, it avoids the problem of insufficient regulation effect caused by excessively low doping.
[0069] The thickness range of the channel region 5 meets the above requirements, taking into account the threshold voltage regulation effect, cell size and process fabrication window, avoiding the sensitivity of process fluctuations caused by excessively small thickness, and also preventing excessively large thickness from increasing the cell area of the device.
[0070] In practice, the doping concentration of the current spreading layer 3 is greater than that of the epitaxial layer.
[0071] The current spreading layer 3 is disposed between and below the channel shielding regions 9, and its doping concentration is higher than that of the epitaxial layer 2. This narrows the width 31 of the depletion region within the current spreading layer 3; simultaneously, the current spreading layer 3 compensates for the semiconductor doping below the channel shielding regions 9, shortening the resistance length along the current conduction path. This reduces the device's on-resistance, improves device conduction performance, and does not increase the device's cell size.
[0072] Shortening the resistance length in the current conduction path means reducing the distance the current travels through the high-resistivity epitaxial layer when the device is turned on. A highly doped current spread layer has stronger conductivity, allowing current flowing out of the channel shield to preferentially travel through the current spread layer, reducing the current's travel distance within the high-resistivity epitaxial layer and thus lowering the device's on-resistance.
[0073] In practice, junction field-effect transistors also include: The metal contact 10 of the channel shielding area is formed on the channel shielding area 9; The channel shielding medium 11 is formed on the channel shielding region 9 and the channel shielding region metal contact 10. The top surface of the channel shielding region 9 and the top surface of the channel shielding region metal contact 10 are completely covered by the parasitic base region and the channel shielding medium 11 to isolate them from the gate region 8. The source region 71 and the source region tail region 7 are arranged from top to bottom; the source region tail region 7 is located above the junction gate capacitance control region 6. A cap-shaped source metal 13 is formed on the top of the source region 71, with the inner top of the lateral portion of the source metal 13 closely attached to the top of the source region 71, and the inner wall of the vertical portion of the source metal 13 closely attached to the side wall of the source region 71. The channel shielding area metal contact 10 and the source metal 13 are electrically connected.
[0074] In practice, junction field-effect transistors also include: Gate metal 12 is formed on the top and outer wall of the gate region 8; Drain metal 14 is formed on the back side of the substrate 1.
[0075] Figure 8 for Figure 5 The corresponding device conduction state diagram is shown. Gate region 8 is a P-type doped region, and junction-gate capacitance control region 6 is an N-type doped region. Together, they form a parasitic PN junction diode. Among them, junction-gate capacitance control region 6 constitutes the series body resistance of this parasitic PN junction diode. When a gate-source voltage is applied, it can generate a voltage divider, reducing the actual bias voltage of the parasitic PN junction diode, thereby effectively suppressing the forward gate-source leakage current.
[0076] When the device is working, the gate-source voltage can be applied and controlled through the gate metal 12, so that the gate-source voltage Vgs is greater than the threshold voltage Vth, and the channel region 5 forms a conducting channel.
[0077] The depletion lines (dashed lines) between the channel shielding areas 9 are not connected to each other, forming a conductive path for the device, and the main power conduction current flows through the conductive channel. For example... Figure 8 As shown, the main power conduction current corresponding to the solid arrow can flow from the drain metal 14 to the source metal 13.
[0078] Meanwhile, the forward gate-source leakage current generated by the parasitic PN junction diode is relatively small. The current path of the forward gate-source leakage current is shown by the dashed arrow. The forward gate-source leakage current flows sequentially through the gate region 8, the junction-gate capacitance control region 6, the source tail region 7, and finally flows to the source region 71. Figure 8 As shown, the dashed arrows represent the current paths of the forward gate-source leakage current.
[0079] The principle behind the smaller forward gate-source leakage current generated by parasitic PN junction diodes is as follows: Gate region 8 and junction gate capacitance control region 6 form a parasitic PN junction diode, wherein gate region 8 is a P-type region and junction gate capacitance control region 6 is an N-type region. The total applied voltage Vgs (referred to as gate-source voltage) with fixed gate and source electrodes satisfies equations (1) and (2): Equation (1) Ideal current-voltage relationship of a PN junction: Equation (2).
[0080] In equation (1), Vgs is the total applied voltage at the fixed gate and source electrodes, and R N It is the resistance of the gate capacitance control region 6 outside the depletion region of the PN junction, and Va is the bias voltage across the PN junction.
[0081] In equation (2), e is the electron charge, and D p and D n These are the hole and electron diffusion coefficients, respectively, L p and L n These represent the hole and electron diffusion lengths, respectively, p n0 and n p0 Here, V represents the thermal equilibrium concentrations of minority electrons and holes, Va is the bias voltage across the PN junction (i.e., the parasitic PN junction diode), k is the Boltzmann constant, and T is the temperature.
[0082] From equation (1), it can be seen that when the doping concentration of the junction gate capacitance control region 6 decreases, the resistance R of this region increases. N Increase; Under the same current conditions, the voltage drop across the N region increases, which reduces the actual bias voltage Va across the PN junction; Combined with equation (2), the PN junction current changes exponentially with Va, so the PN junction current decreases exponentially.
[0083] By controlling the doping concentration and thickness of the parasitic base region, the conduction of the parasitic transistor is delayed, and the blocking leakage current of the parasitic transistor is reduced.
[0084] Figure 9 for Figure 5A schematic diagram of the off state shows that when a reverse bias voltage is applied to the gate metal 12, the depletion lines (dashed lines) between the channel shielding regions 9 widen and connect together, blocking the conduction current. For example... Figure 9 As shown, the gate region 8, the first parasitic base region 41, and the channel shielding region 9 constitute a parasitic transistor with weak conduction. The current path is channel shielding region 9 → first parasitic base region 41 → gate region 8, indicated by a dashed arrow. By controlling the doping concentration and thickness of the parasitic base region, the punch-through voltage of the parasitic transistor can be increased, and the blocking leakage current of the parasitic transistor can be reduced.
[0085] Figure 5 The method for fabricating the junction field-effect transistor shown includes the following steps: like Figure 10 As shown, step S51: an epitaxial layer 2 is formed on the substrate 1; an N-type current spreading layer base 30 is formed on the epitaxial layer 2 by ion implantation.
[0086] like Figure 11 As shown, step S52: through epitaxy, an N-type first parasitic base region foundation 410 is formed on the N-type current extension layer 3; The N-type high-concentration source region 71 and the source region tail region 7 are formed by ion implantation, or by epitaxy.
[0087] Grooves 81 are formed by etching under the obstruction of mask 45.
[0088] like Figure 12 As shown, in step S53: a P-type gate region 8 is formed by ion implantation, and an N-type first parasitic base region 41 is formed simultaneously; under the obstruction of the mask 45, a sidewall is formed by thin film deposition, a trench 92 is formed by etching, and a P-type channel shielding region 9 is formed by ion implantation.
[0089] like Figure 13 As shown, in step S54: through ohmic contact process, thin film deposition, photolithography, and etching, the channel shielding region metal contact 10, the channel shielding region dielectric 11, the gate metal 12, the source metal 13, and the drain metal 14 are formed.
[0090] Figure 6 The method for fabricating the junction field-effect transistor shown includes the following steps: like Figure 14 As shown, step S61: an epitaxial layer 2 is formed on the substrate 1; an N-type current spreading layer base 30 is formed on the epitaxial layer 2 by ion implantation.
[0091] like Figure 15As shown, in step S62: an N-type second parasitic base region foundation 420 is formed on the N-type current extension layer 3 by epitaxy; an N-type highly doped source region foundation 710 and a source region tailing region foundation 70 are formed by ion implantation, or an N-type highly doped source region foundation 710 is formed by epitaxy, while simultaneously forming the N-type second parasitic base region foundation 420.
[0092] like Figure 16 As shown, in step S63: under the obstruction of the mask, a trench is formed by etching; a P-type gate region 8 is formed by ion implantation; under the obstruction of the mask, a sidewall is formed by thin film deposition, a trench is formed by etching, and a P-type channel shielding region 9 is formed by ion implantation. Through ohmic contact process, thin film deposition, photolithography, and etching, the channel shielding region metal contact 10, the channel shielding region dielectric 11, the gate metal 12, the source metal 13, and the drain metal 14 are formed.
[0093] Figure 17 Simulation diagram of gate-source leakage current density distribution after applying gate-source voltage to the existing structure.
[0094] Figure 18 For this application Figure 5 Simulation diagram of gate-source leakage current density distribution after applying gate-source voltage to a junction field-effect transistor.
[0095] Figure 19 The graph shows the gate-source leakage current Igss versus gate-source voltage Vgs for this application and existing structures.
[0096] like Figure 17 , Figure 18 As shown, the gate-source current density of this application is lower than that of the prior art.
[0097] like Figure 19 As shown, the reverse gate-source voltage of this application is greater than that of the prior art; in other words, the gate-source leakage current is lower under the same reverse-biased gate-source voltage.
[0098] Figure 20 The graph shows the gate-source capacitance Cgs and drain-source voltage Vds of this application and existing structures.
[0099] like Figure 20 As shown, the gate-source capacitance Cgs of this application is lower than that of the prior art. This is because the doping concentration of the gate-junction capacitance control region 6 in this application is relatively low. The gate-source capacitance Cgs is part of the gate capacitance, which also includes the gate-drain capacitance Cgd and the gate-substrate capacitance Cgb.
[0100] Figure 21 The above is a simulation diagram of the electric field distribution during breakdown of the existing structure excluding the junction gate capacitance control region 6.
[0101] Figure 22When the present application, including the junction gate capacitance control region 6, breaks down. Figure 5 Simulation diagram of electric field distribution at the location indicated by the dashed box.
[0102] Figure 23 for Figure 22 Simulation diagram of doping distribution; Figure 24 The blocking current I-voltage V curves are for this application and existing structures.
[0103] Figure 21 and Figure 22 The simulation conditions are: a bias voltage less than the threshold voltage is applied to the gate metal 12 and the source metal 13 is grounded, and the drain voltage is scanned.
[0104] Figure 21 and Figure 22 Simulation results show that the electric field strength of the junction gate capacitance control region 6 and the gate region 8 in this application is lower, and the maximum electric field strength is transferred to the PN junction of the gate region 8 and the first parasitic base region 41, which brings the advantage of a larger breakdown voltage in this application compared with the prior art.
[0105] like Figure 24 As shown, the existing technology in V DS At approximately 20V, the drain current rises sharply (breakdown characteristic); the rising inflection point of this application is significantly later (higher voltage) and the rise is slower—indicating that in the blocking state, the voltage at which the drain current suddenly increases is higher in this application.
[0106] Figure 25 The current density simulation diagram of the existing technology at the breakdown moment at a high temperature of 175℃ is shown.
[0107] Figure 26 The current density simulation diagram of this application at the breakdown moment at a high temperature of 175℃ is shown.
[0108] Figure 27 The I-V curves of the blocking current and the prior art are shown at the breakdown time at a high temperature of 175°C.
[0109] At a high temperature of 175°C, the leakage current is small and the current path is redirected from a portion of the channel to the PN junction formed by the channel shielding region 9 and the current extension layer 3.
[0110] The following technical problems also exist in the background technology: 1) The channel shielding area 8 is part of the epitaxial layer 2, which causes some devices to be unable to conduct or increases the device cell size.
[0111] 2) Epitaxial layer 1 is only an epitaxial layer and cannot simultaneously meet multiple electrical parameters such as threshold voltage, on-resistance, gate-source leakage current, and gate-source capacitance.
[0112] 3) If no electrode is set above the channel shielding area 8, the potential of the channel shielding area 8 will be floating, resulting in unstable device performance and abnormal phenomena such as inability to turn off.
[0113] This application incorporates a current spreading layer 3, which has a higher doping concentration than the epitaxial layer 2, ensuring that the device in this application can conduct. This solves the aforementioned problem 1).
[0114] Gate-source leakage current is divided into forward gate-source leakage current and reverse gate-source leakage current. In this application, the reverse gate-source leakage current is suppressed by a parasitic transistor formed by the gate region 8, the parasitic base region, and the channel shielding region 9. Similarly, the forward gate-source leakage current is suppressed by a parasitic PN junction diode formed by the junction-gate capacitance control region 6 and the gate region 8. This solves the technical problem of large gate-source leakage current caused by the limited doping concentration of the epitaxial layer mentioned above (problem 2).
[0115] In this application, the first PN junction formed by the gate capacitance control region 6 and the gate region 8, and the second PN junction formed by the gate region 8 and the parasitic base region, are connected in parallel, and their equivalent capacitance is the gate-source capacitance. The parallel equivalent capacitance means that reducing the capacitance of one of the PN junctions will reduce the overall capacitance, which is the fundamental reason for the reduction in gate-source capacitance. This solves the technical problem of the relatively large gate-source capacitance caused by the limited doping concentration of the epitaxial layer mentioned above.
[0116] The channel region 5 and current spreading layer 3 of this application reduce the on-resistance by controlling the doping concentration of the channel region 5 and current spreading layer 3. This solves the technical problem of high on-resistance caused by the limited doping concentration of the epitaxial layer in problem 2) above.
[0117] The channel region 5 of this application allows for the configuration of the threshold voltage by controlling the doping concentration and thickness of the channel region 5. This solves the technical problem of low threshold voltage caused by the limited doping concentration of the epitaxial layer mentioned above (problem 2).
[0118] This application solves the technical problem of the above-mentioned problem 3) by setting the channel shielding area metal contact 10 to be electrically connected to the source metal 13.
[0119] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.
[0120] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.
Claims
1. A junction field-effect transistor, characterized in that, include: Second type of doped substrate (1); A second doped epitaxial layer (2) is disposed on the substrate (1); The first type of doped channel shielding region (9) is located within the epitaxial layer; A parasitic base region of the second doping type is formed on the top surface of the channel shielding region (9), and the doping concentration of the parasitic base region is greater than the doping concentration of the epitaxial layer (2); A gate region (8) of the first doped type is formed on the parasitic base region; The gate region (8), the parasitic base region, and the channel shield region (9) constitute a parasitic transistor. The doping concentration and thickness of the parasitic base region are configured such that the punch-through voltage of the parasitic transistor is a preset multiple of the gate-source rated voltage, and the preset multiple is greater than 1.
2. The junction field-effect transistor according to claim 1, characterized in that, The gate-source voltage Vgs turns off the junction field-effect transistor, and when the absolute value of the gate-source voltage Vgs is less than the design value of the punch-through voltage, the parasitic transistor remains off, and the blocking leakage current of the device is reduced.
3. The junction field-effect transistor according to claim 2, characterized in that, The gate-source voltage Vgs turns off the junction field-effect transistor. Only when the absolute value of the gate-source voltage Vgs is greater than or equal to the design value of the punch-through voltage will the current of the parasitic transistor surge due to the punch-through of the gate junction reverse bias depletion layer.
4. The junction field-effect transistor according to claim 2, characterized in that, The preset multiple is greater than or equal to 1.
2.
5. The junction field-effect transistor according to claim 1, characterized in that, Also includes: The second doped type of junction gate capacitance control region (6) is disposed on the two spaced gate regions (8); A second doped type current spreading layer (3) is formed between and below the channel shielding region (9); The junction gate capacitance control region (6), the gate region (8), and the parasitic base region (41) constitute the first JFET region; the parasitic base region, the channel shielding region (9), the current spreading layer (3), and the epitaxial layer (2) constitute the second JFET region. The withstand voltage of the first JFET region is configured such that the second JFET region is turned off before the first JFET region breaks down.
6. The junction field-effect transistor according to claim 5, characterized in that, The first PN junction is formed by the gate capacitance control region (6) and the gate region (8), and the second PN junction is formed by the gate region (8) and the parasitic base region (41). The first PN junction and the second PN junction are connected in parallel, and their equivalent capacitance is the gate-source capacitance.
7. The junction field-effect transistor according to claim 6, characterized in that, The junction gate capacitance control region (6) and the gate region (8) constitute the series body resistance of the parasitic PN junction diode. When the gate source voltage is applied, a voltage divider is generated to reduce the actual bias voltage of the parasitic PN junction diode and suppress the forward gate source leakage current. When the gate-source voltage Vgs is applied and controlled to be greater than the threshold voltage Vth, a conducting channel is formed in the channel region (5); The depletion lines between the shielded areas (9) of the trench do not overlap, forming a conductive channel; The forward gate-source leakage current path of the parasitic PN junction diode passes through the gate region (8), the junction gate capacitance control region (6), the source region tail region (7), and the source region (71) in sequence.
8. The junction field-effect transistor according to claim 5, characterized in that, Also includes: A second doped channel region (5) is disposed between the two gate regions (8); The source region (71) and the source region tail region (7) are arranged from top to bottom; the source region tail region (7) is located above the junction gate capacitance control region (6); The parameters of the channel region (5) are determined by the spacing, doping concentration and threshold voltage of the gate region (8).
9. The junction field-effect transistor according to claim 1, characterized in that, The parasitic base region is the first parasitic base region (41); Each gate region (8) is provided with a first parasitic base region (41), which is formed between the corresponding channel shield region (9) and the gate region (8).
10. The junction field-effect transistor according to claim 1, characterized in that, Each cell contains two spaced-apart channel shielding areas (9); The parasitic base region is the second parasitic base region (42); The second parasitic base region (42) spans the bottom of two adjacent gate regions (8) and covers the area below the region between the two gate regions (8).
11. The junction field-effect transistor according to claim 1, characterized in that, The parasitic base region is formed epitaxially; Alternatively, the parasitic base region can be formed by injection.
12. The junction field-effect transistor according to claim 1, characterized in that, The doping concentration of the parasitic base region is greater than or equal to 1 × 10⁻⁶. 17 cm -3 And less than or equal to 5 × 10 18 cm -3 ; The thickness of the parasitic base region is greater than or equal to 0.01 μm and less than or equal to 0.3 μm.
13. The junction field-effect transistor according to claim 5, characterized in that, The doping concentration of the gate capacitance control region (6) is greater than or equal to 1×10⁻⁶. 15 cm -3 And less than or equal to 1×10 18 cm -3 The thickness of the junction gate capacitance control region (6) is greater than or equal to 0.001 μm and less than or equal to 0.3 μm.
14. The junction field-effect transistor according to claim 8, characterized in that, The doping concentration of the channel region (5) is greater than or equal to 5 × 10⁻⁶. 16 cm -3 And less than or equal to 5 × 10 18 cm -3 The thickness of the channel region (5) is greater than or equal to 0.01 μm and less than or equal to 1 μm.
15. The junction field-effect transistor according to claim 7, characterized in that, Also includes: The channel shielding area metal contact (10) is formed on the channel shielding area (9); The channel shielding medium (11) is formed on the channel shielding area (9) and the channel shielding metal contact (10). The top surface of the channel shielding area (9) and the top surface of the channel shielding metal contact (10) are completely covered by the channel shielding medium (11) to isolate them from the gate area (8). A cap-shaped source metal (13) is formed on the top of the source region (71). The inner side of the lateral portion of the source metal (13) is in close contact with the top of the source region (71), and the inner wall of the vertical portion of the source metal (13) is in close contact with the side wall of the source region (71). The channel shielding area metal contact (10) and the source metal (13) are electrically connected. Gate metal (12) is formed on the top and outer sidewalls of the gate region (8); Drain metal (14) is formed on the back side of the substrate (1).
Citation Information
Patent Citations
A trench junction field effect transistor and a method of manufacturing the same
CN120769543B
