Semiconductor equipment

Abstract

To ensure that the protection of a power MOSFET is fully exercised by a free-wheel diode formed on the same semiconductor substrate as the power MOSFET.SOLUTION: A structure associated with trenches 10A, 10B functions as a MOSFET. A gate oxide film 14D and a second gate electrode 15B are disposed on the trench 10C side. The gate oxide film 14D is formed in smaller thickness than a gate oxide film 14A. An electrode 19 for second gate control is formed on the trench 10C, the second gate electrode 15B and an electrode 19 for second gate control are connected together, and the electrode 19 for second gate control and a source electrode 18 are connected together via a resistive element R. When a source potential is higher than a drain potential, a pseudo-MOSFET formed by the trench 10C goes on, and the absolute value of a second gate potential becomes smaller than the source potential, enabling to raise the breakdown voltage of a free-wheel diode.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a semiconductor device structure in which control electrodes are provided in trenches (grooves) formed on the surface of a semiconductor substrate. [Background technology]

[0002] Semiconductor devices (power semiconductor elements: power MOSFETs, IGBTs, etc.) are used in which the on / off state of the current between the front and back surfaces of a semiconductor substrate is controlled by the potential of the gate electrode. In such semiconductor devices, the potential of the gate electrode (control electrode) causes a channel to form in the semiconductor layer opposite the gate electrode, which becomes the current path. By controlling the on / off state of this channel, the on / off state of the current is controlled. Furthermore, trench-type elements, in which a trench is formed on the front surface of the semiconductor substrate and the gate electrode is provided within this trench, are particularly preferred because they facilitate cell miniaturization and can reduce on-resistance. In trench-type elements, a thin gate oxide film (gate insulating film) is formed on the inner wall of the trench. The semiconductor layer constituting the inner wall of the trench and the gate electrode face each other via this gate oxide film, and the on / off state of the channel in this portion of the semiconductor layer is controlled by the potential of the gate electrode (gate potential).

[0003] For example, in an n-channel power MOSFET, a p-type body layer is formed on top of an n-type drift layer, and a high-density n-type source region is formed above the body layer. A trench is formed to penetrate both the source region and the body region. When the MOSFET is on, the current flows through the channel formed in the body layer, between the source region and the drift layer, and then through the drift layer in the thickness direction to the drain. Here, a pn junction is formed between the body layer and the drift layer, and this part functions as a diode (body diode). During normal operation of the power MOSFET, the potential of the drain (main electrode on the drift layer side) is higher than that of the source (main electrode on the body layer side), so the body diode becomes reverse-biased, the pn junction between the body layer and the drift layer does not conduct, and current flows only through the channel formed in the body layer. Therefore, the on / off switching of the channel by the gate potential can be controlled to control the on / off switching of the current flowing between the source and the drain.

[0004] On the other hand, when an inductor such as a coil is connected as a load to such a power MOSFET and its on / off switching is controlled, a situation may occur where the source potential becomes significantly higher than the drain potential transiently immediately after the power MOSFET switches from the on state to the off state. In such cases, in order to prevent the power MOSFET from being destroyed by overcurrent, it is preferable to connect a diode (freewheeling diode) that becomes forward-biased when the source potential becomes higher than the drain potential, and in such cases, current flows through the freewheeling diode to bypass the power MOSFET. During normal operation as described above, the drain potential becomes high, so the freewheeling diode becomes reverse-biased, and no current flows through the freewheeling diode, and the operation of the power MOSFET is not affected.

[0005] Since the characteristics (polarity) of such a freewheeling diode are the same as those of the body diode described above, the body diode can also be used as a freewheeling diode. However, since the body diode is formed by a pn junction of the body layer and drift layer in a power MOSFET, and both the body layer and drift layer are set to suit the normal on / off operation of the power MOSFET, it is generally difficult to configure the body layer and drift layer to be optimal for use as a freewheeling diode. For this reason, it is particularly preferable to provide a separate portion on the same semiconductor substrate that has the same polarity as the body diode, functions similarly as a diode, and has characteristics that are more favorable for a freewheeling diode, so that when the source potential becomes high, current flows through both the body diode and this separate portion.

[0006] In this context, it is preferable for such a freewheeling diode to be able to conduct sufficient current in the forward direction. However, generally, even in the forward direction, if the voltage is below VF (forward voltage), only a small current flows, similar to the case in the reverse direction. In the Si pn junction that constitutes the body diode, VF is generally around 0.7V, whereas for a freewheeling diode, a smaller VF is desirable.

[0007] Therefore, Patent Document 1 describes a semiconductor device in which a portion that functions as a diode with a small VF is formed on the same chip as the power MOSFET, separate from the body diode in the power MOSFET. In this device, multiple trenches are formed, with the power MOSFET in some of the trenches and the portion that functions as such a diode in other trenches.

[0008] Figure 5 is a cross-sectional view showing the structure of the semiconductor device 9. This semiconductor device 9 combines an n-channel power MOSFET with a diode component as described above. Both the power MOSFET and the diode component are formed using trenches formed in a common semiconductor substrate. In Figure 1, in the semiconductor substrate 10 made of semiconductor material (Si), three trenches 10A to 10C are formed sequentially from right to left, excavated from the surface (top) side. The rightmost trenches 10A and 10B function as MOSFETs, and the leftmost trench 10C functions as a diode. Figure 5 shows a cross-section perpendicular to the direction of extension of these trenches.

[0009] In this semiconductor substrate 10, a p-type (second conductivity type) body layer (second semiconductor layer) 12 is laminated on top of an n-type (first conductivity type) drift layer (first semiconductor layer) 11. Trenches 10A to 10C are formed to penetrate the body layer 12 from the surface side of the semiconductor substrate 10, with their bottom surfaces located within the drift layer 11. On the surface of the semiconductor substrate 10, adjacent to each of the trenches 10A to 10C, there are high-concentration n-type layers (n + A source region (first source region) 13A, which will become a layer, is selectively formed. The body layer 12 can be formed by epitaxial growth on the drift layer 11, or by ion implantation on the surface of the semiconductor substrate 10 with the body layer 12 already provided on its surface. The source region 13A can be formed by locally implanting ions on the surface of the body layer 12.

[0010] In Figure 5, only the components related to trench 10B are labeled, and the structure related to trench 10A is identical to the structure related to trench 10. Trenches 10A and 10B (first grooves) are embedded by an upper first gate electrode (control electrode) 15A and a lower shield electrode 16, with a gate oxide film (first gate insulating film) 14A formed on their inner surfaces. Both the first gate electrode 15A and the shield electrode 16 are made of a conductive metal material or highly conductive polycrystalline silicon with a high concentration of impurities added. The first gate electrode 15A is set so that there is a source region 13 and a body layer 12 to its side, and its bottom surface is at the height of the drift layer 11. The shield electrode 16 is located below the first gate electrode 15A and at the height of the drift layer 11. Since the first gate electrode 15A and the shield electrode 16 are insulated by the oxide film 14B, their potentials are independent. The shield electrode 16 and the drift layer 11 are insulated by the oxide film 14C, and the above structure also insulates the shield electrode 16 from the body layer 12.

[0011] Furthermore, an interlayer insulating layer 17A is locally thickened above the trenches 10A and 10B to cover the first gate electrode 15A from above. However, the portion of the source region 13A that is separated from the trenches 10A and 10B is not covered by the interlayer insulating layer 17A. The gate oxide film 14A is composed of SiO2 formed by thermal oxidation of the semiconductor substrate 10 that constitutes the inner surface of the trench 10A. The oxide films 14B, 14C, and the interlayer insulating layer 17A are also composed of SiO2, but these are formed by deposition to a thickness greater than the gate oxide film 14A using methods such as CVD.

[0012] The surface side (upper side in the figure) of the semiconductor substrate 10 is covered over its entire surface by a source electrode (first main electrode) 28 made of a low-resistivity metallic material (such as Al). As described above, an interlayer insulating layer 17A is provided, so the source electrode 28 is in contact with the body layer 12 and the first source region 13A on the surface of the semiconductor substrate 10, and is insulated from the first gate electrode 15A. The shield electrode 16 is connected to the source electrode 28 outside the shown area.

[0013] On the other hand, the back side (lower side in the figure) of the semiconductor substrate 10 is completely covered by a drain electrode (second main electrode) 20 made of a metal material that makes ohmic contact with the drift layer 11 (n-type layer). A high-density n-type layer may be provided between the drain electrode 20 and the drift layer 11.

[0014] In the above structure, the potentials of the source electrode 28, the drain electrode 20, and the first gate electrode 15A are controlled independently, and the structure involving the trenches 10A and 10B as described above functions as a MOSFET (power MOSFET) in which the on / off switching of the current flowing between the source electrode 28 and the drain electrode 20 is controlled by the voltage (gate potential) applied to the first gate electrode 15A. That is, the gate potential controls the on / off switching (presence or absence) of channels in the body layer 12 that constitutes the inner surface of the trenches 10A and 10B, thereby controlling the on / off switching of the electron flow between the first source region 13A and the drift layer 11, and the on / off switching of the current between the source electrode 28 and the drain electrode 20. In this case, during normal operation, the source electrode 28 is at ground potential, and by providing a shield electrode 16, which is also at ground potential, below the first gate electrode 15 in the trench 10A, the feedback capacitance Crss (capacitance between gate and drain) can be reduced, and this power MOSFET can be operated at a higher speed. However, if the operating speed is sufficient even without the shield electrode 16, the shield electrode 16 may be omitted, and only the first gate electrode 15A may be provided in the trenches 10A and 10B via the gate oxide film 14A. In this case, the trench 10A can be made shallower than in the configuration shown in Figure 1.

[0015] In the left trench 10C, similar to the gate oxide film 14A, first gate electrode 15A, shield electrode 16, oxide films 14B and 14C, and interlayer insulating layer 17A in the right trenches 10A and 10B, a gate oxide film (second gate insulating film) 14D, second gate electrode 15B, shield electrode 16, oxide films 14B and 14C, and interlayer insulating layer 17B are provided. On the surface side, similar to the first source region 13A, n+ A second source region 13B, which is a layer, is also provided. The source electrode 28 is formed to cover the entire region from trench 10A to trench 10C, and is in contact with the body layer 12 and the second source region 13B near trench 10C, and is also connected to the shield electrode 16 in trench 10C outside the shown area.

[0016] However, the gate oxide film 14D is formed thinner than the gate oxide film 14A. Also, in trenches 10A and 10B, the first gate electrode 15A and the source electrode 28 were insulated by the interlayer insulating layer 17A, whereas an opening is formed in the interlayer insulating layer 17B formed on trench 10C, and the second gate electrode 15B and the source electrode 28 are connected by this opening. For this reason, in trench 10C, a pseudo-MOSFET is formed that is different from that in trenches 10A and 10B, with the second gate electrode 15B acting as a pseudo-gate, and this MOSFET operates with the potential of the source electrode 28 (source potential) as its pseudo-gate potential.

[0017] As described above, in normal operation where the drain potential is higher than the source potential, the parts related to trenches 10A and 10B operate as MOSFETs, and the pn junction (body diode) composed of the body layer 12 and drift layer 11 is reverse-biased. On the other hand, trench 10C does not have a control electrode to which the gate potential is applied, and the source potential (≤ drain potential) becomes the pseudo-gate potential of this pseudo-MOSFET. Therefore, regardless of the gate potential, this pseudo-MOSFET is in the off state, and the operation of the parts related to trenches 10A and 10B is not affected by the parts related to trench 10C.

[0018] On the other hand, when the source potential becomes higher than the drain potential, as described above, the pn junction (body diode) composed of the body layer 12 and the drift layer 11 becomes forward-biased, and current flows between the source electrode 28 and the drain electrode 20 regardless of the gate potential. This point is common to the parts related to trenches 10A and 10B and the part related to trench 10C. However, in trench 10C, the pseudo gate potential is the actual source potential, so in this case, the pseudo MOSFET turns on. At this time, the gate oxide film 14D is formed thinly, so the threshold voltage Vt of this pseudo MOSFET is low. That is, this pseudo MOSFET turns on even when the source potential is slightly positive relative to the drain potential. For this reason, the diode composed of the part related to trench 10C has a substantially small VF. In other words, a freewheeling diode with a substantially small VF is formed in this semiconductor device 9. At this time, many common parts are used in the structure within these two types of trenches, so this semiconductor device can be manufactured using a simple manufacturing process.

[0019] Furthermore, in this structure, the power MOSFET and the diode-functioning portion are formed on the same semiconductor substrate (chip), and long wiring is not used to connect them, thus suppressing the formation of unwanted inductance components. In addition, the current that can flow through the power MOSFET and the diode-functioning portion can be adjusted by the ratio of the power MOSFET components and the diode-functioning portion distributed in the multiple trenches, as well as the spacing between the trenches.

[0020] From the same perspective, a semiconductor device in which the portion functioning as a diode has a configuration different from the above is described in Patent Document 2. Also in this semiconductor device, as in the above case, a plurality of trenches are divided into a portion functioning as a MOSFET and a portion functioning as a diode as described above. Different from the semiconductor device described in Patent Document 1, in the latter trenches, a gate oxide film (MOS structure) is not formed, and a Schottky electrode is formed so as to directly contact the trench side surface. Therefore, a Schottky diode is formed in this portion.

[0021] Generally, since the VF of a Schottky diode is smaller than that of a pn junction, by using this structure, a freewheel diode with a substantially small VF can also be obtained. Also in this case, since the commonly formed trenches can be used, this semiconductor device can be manufactured by a simple manufacturing process.

Prior Art Documents

Patent Documents

[0022]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0023] The VF of the freewheel diode formed in the semiconductor device described in Patent Document 2 is determined by the VF of the Schottky diode formed. As described above, although the VF of the Schottky diode is smaller than that of the pn junction, a smaller VF is required for the freewheel diode. That is, in this case, it was difficult to sufficiently reduce the VF.

[0024] In contrast, in the semiconductor device described in Patent Document 1, the VF can be made sufficiently small by making the gate oxide film 14D in the trench 10C, which functions as a diode, sufficiently thinner than the gate oxide film 14A in the trenches 10A and 10B, which function as MOSFETs. However, in this structure, when the source potential becomes high, the electric field strength in the gate oxide film 14D increases, making the gate oxide film 14D more susceptible to dielectric breakdown. In other words, in this case, although the VF becomes small, the breakdown voltage against the source potential becomes low.

[0025] Therefore, it was difficult to adequately protect the power MOSFET using a freewheeling diode formed on the same semiconductor substrate as the power MOSFET.

[0026] This invention has been made in view of the above-mentioned problems, and aims to provide an invention that solves the above-mentioned problems. [Means for solving the problem]

[0027] In order to solve the above problems, the present invention has the following configuration. The semiconductor device of the present invention comprises a first semiconductor layer made of a semiconductor material having a first conductivity type, and a second semiconductor layer having a second conductivity type opposite to the first conductivity type and formed on the first semiconductor layer, wherein a first groove is provided that penetrates the second semiconductor layer from the surface side and reaches the first semiconductor layer, and a first source region having the first conductivity type is formed on the surface side so as to be in contact with the first groove and partially along the extending direction of the first groove, and the semiconductor device of the present invention comprises a first main electrode made of metal connected to the second semiconductor layer and the first source region on the surface side of the semiconductor substrate, a second main electrode connected to the first semiconductor layer on the back side of the semiconductor substrate, and a first gate electrode facing the inner surface of the first groove with a first gate insulating film in between, A semiconductor device in which the current flowing between a first main electrode and a second main electrode is controlled by the potential of the first gate electrode, comprising: a second groove dug from the surface side at a location spaced apart from the first groove in the semiconductor substrate; a second source region having the first conductivity type, partially formed along the extending direction of the second groove so as to be in contact with the second groove on the surface side; a second gate control electrode made of metal, connected to the second semiconductor layer and the second source region on the surface side of the semiconductor substrate, and separated from the first main electrode; and a second gate electrode facing the inner surface of the second groove within the second groove, with a second gate insulating film thinner than the first gate insulating film in between, wherein the second gate control electrode is connected to the second gate electrode and connected to the first main electrode via a resistive element. The semiconductor device of the present invention is characterized in that the first main electrode contacts only a portion of the first source region in the direction of extension of the first groove in a plan view, and the resistive element is formed using a region of the first source region that does not contact the first main electrode, or the second gate control electrode contacts only a portion of the second source region in the direction of extension of the second groove in a plan view, and the resistive element is formed using a region of the second source region that does not contact the second gate control electrode. The semiconductor device of the present invention is characterized in that, in the thickness direction of the semiconductor substrate, a third semiconductor layer having a first conductivity type, a lower impurity concentration than the first semiconductor layer, and being thinner than the first semiconductor layer is inserted between the first semiconductor layer and the second semiconductor layer. The semiconductor device of the present invention is characterized in that, in the first groove, a shield electrode is provided below the first gate electrode, which is insulated from the first semiconductor layer, the second semiconductor layer, and the first gate electrode and electrically connected to the first main electrode. [Effects of the Invention]

[0028] As described above, the present invention can adequately protect the power MOSFET by using a freewheeling diode formed on the same semiconductor substrate as the power MOSFET. [Brief explanation of the drawing]

[0029] [Figure 1] This is a cross-sectional view of a semiconductor device according to the first embodiment of the present invention. [Figure 2] This is a perspective cross-sectional view of a semiconductor device according to an embodiment of the present invention. [Figure 3] These are the results of simulations calculating the characteristics of the freewheeling diode under forward bias in the example and the conventional example. [Figure 4] This is a cross-sectional view of a modified semiconductor device according to the first embodiment of the present invention. [Figure 5] This is a cross-sectional view of a conventional semiconductor device. [Modes for carrying out the invention]

[0030] Hereinafter, a semiconductor device according to an embodiment of the present invention will be described. In this semiconductor device 1 as well, a portion that functions as a power MOSFET and a portion that functions as a freewheel diode are each formed using trenches. The basic structure of each of these portions is the same as that of the semiconductor device described in Patent Document 1, but is particularly characterized by the connection between these portions.

[0031] FIG. 1 is a cross-sectional view showing the structure of a semiconductor device 1 according to an embodiment of the present invention. Here, a part is shown by a circuit symbol instead of a structure. Similar to the semiconductor device 9 described above, trenches 10A to 10C are formed in the semiconductor substrate 10 in this semiconductor device 1. Correspondingly, an n-type (first conductivity type) drift layer (first semiconductor layer) 11, a p-type (second conductivity type) body layer (second semiconductor layer) 12, a high-concentration n-type layer (n + layer) serving as a first source region 13A and a second source region 13B are formed in the semiconductor substrate 10. The impurity concentrations of the drift layer 11 (n-type), the body layer 12 (p-type), the first source region 13A, and the second source region 13B (n-type) are, for example, 10 16 cm -3 , 10 15 cm -3 , 10 19 cm -3 respectively.

[0032] Furthermore, within the trenches 10A and 10B (first grooves), a gate oxide film (first gate insulating film) 14A, a first gate electrode (control electrode) 15A, a shield electrode 16, and oxide films 14B and 14C are formed. Above the trenches 10A and 10B, a source electrode (first main electrode) 18 is formed in contact with the interlayer insulating layer 17A, the first source region 13A, and the body layer 12. A drain electrode (second main electrode) 20 is also formed on the back side of the semiconductor substrate 10. Therefore, similar to the semiconductor device 9 described above, the structure related to the trenches 10A and 10B functions as a power MOSFET. Similarly, a diode (body diode) that functions as a freewheel diode is formed between the body layer 12 and the drift layer 11. Also similarly, the shield electrode 16 may be unnecessary in some cases.

[0033] Therefore, the operation related to trenches 10A and 10B is no different from that of the semiconductor device 9 described above. That is, in normal operation, the source potential is set to ground potential and the drain potential is set to positive. The gate potential (potential of the first gate electrode 15A) controls the on / off state of the current between the source electrode 18 and the drain electrode 20. Also, if the source potential becomes higher than that of the drain electrode, the body diode formed between the body layer 12 and the drift layer 11 becomes forward-biased, so current flows between the source electrode 18 and the drain electrode 20 regardless of the gate potential.

[0034] On the other hand, on the trench 10C side, similar to the semiconductor device 9, a gate oxide film (second gate insulating film) 14D, a second gate electrode 15B, a shield electrode 16, oxide films 14B and 14C, and an interlayer insulating layer 17B are provided. Here, the gate oxide film 14D is formed thinner than the gate oxide film 14A, which is also the same. Therefore, the structure related to the trench 10C constitutes the part that functions as a freewheel diode, which is also the same.

[0035] However, in the semiconductor device 9 described above, the source electrode 28 was formed to cover all of the trenches 10A to 10C, whereas in this semiconductor device 1, the source electrode 18 (the electrode to which the source potential is applied) is formed only on trenches 10A and 10B. On the other hand, a second gate control electrode 19, separated from the source electrode 18, is formed on trench 10C in the same way as the source electrode 18. The second gate control electrode 19 is electrically connected to the body layer 12 and the second source region 13B adjacent to the trench 10C, similar to the source electrode 28 in the semiconductor device 9 described above. The second gate control electrode 19 is formed of metal, similar to the source electrodes 18 and 28. For this reason, in practice, the source electrode 28 in Figure 5 can be patterned to separate it into the source electrode 18 and the second gate control electrode 19 in Figure 1, and these can be formed accordingly.

[0036] Furthermore, similar to the semiconductor device 9 described above, the second gate electrode 15B and the second gate control electrode 19 are connected through an opening formed in the interlayer insulating layer 17B. Therefore, in trench 10C, a pseudo-MOSFET is formed with the second gate electrode 15B as the gate, which is different from that on trenches 10A and 10B. However, in semiconductor device 9, the pseudo-gate potential of this pseudo-MOSFET was the source potential (the potential of the source electrode 28), whereas in semiconductor device 1, this pseudo-gate potential is the potential of the second gate control electrode 19 (the second gate potential), which is different from the source potential (the potential of the source electrode 18).

[0037] Here, as shown by the circuit symbol in Figure 1, the second gate control electrode 19 on the left and the source electrode 18 on the right are connected via a resistive element R. Therefore, when the source potential fluctuates, the second gate potential and the source potential change in conjunction, but when the source potential is higher than the drain potential, the second gate potential becomes lower than the source potential by the amount of the voltage drop across the resistive element R. On the other hand, in normal operation where the source potential ≤ drain potential, this pseudo-MOSFET is always off, and the body diode is also reverse-biased, so the part related to trench 10C does not affect the operation of the power MOSFET.

[0038] Thus, when the source potential is higher than the drain potential, the second gate potential is lower than the source potential, but its polarity is the same. Therefore, the pseudo-MOSFET formed here turns on, and, as in the case of semiconductor device 9, the VF of the diode formed in the trench 10C can be reduced. However, the absolute value of the second gate potential is smaller than the source potential by the amount of the voltage drop due to the resistive element R, so the maximum electric field strength in the gate oxide film 14D can be reduced compared to semiconductor device 9. For this reason, the breakdown voltage of the freewheel diode formed in this case can be increased when forward biased.

[0039] The specific form of the resistive element R on the semiconductor substrate 10 will now be described. Figure 2 is a perspective cross-sectional view taken from diagonally above, showing a partial structure of the semiconductor device 1 on which the resistive element R is formed. The portion shown as a cross-section here corresponds to Figure 1. For convenience, hatching other than that of the first source region 13A and the second source region 13B has been omitted. In addition, the perspective view of the upper surface structure shown is a perspective view of the source electrode 18, the second gate control electrode 19, the first source region 13A, and the second source region 13B as shown in Figure 1.

[0040] As described above, the source electrode 18 (right side) and the second gate control electrode 19 (left side) formed on the upper side of the semiconductor substrate 10 are formed separately. Also, as shown here, both the source electrode 18 and the second gate control electrode 19 are formed only in the area L1 on the near side of the paper, and not in the area further back from the paper. In contrast, the trenches 10A to 10C are formed continuously from the near side to the far side of the paper, and the first source region 13A and the second source region 13B also extend in multiple parallel and continuous directions along each trench toward the far side of the paper in an elongated shape when viewed in plan. This configuration is similar to that of a commonly known trench-type power MOSFET.

[0041] The source electrode 18 and the second gate control electrode 19 are made of a metal with low resistivity, whereas the first source region 13A and the second source region 13B are made of, for example, 10 19 cm -3 n + This is a layer (semiconductor layer). Therefore, while the resistance of the source electrode 18 and the second gate control electrode 19 can be ignored, the resistance of the elongated source regions 13A and 13B, especially along the longitudinal direction (extension direction), cannot be ignored, and a resistive element R can be formed using this. In the region L1 on the near side of the paper, the source electrode 18 and the second gate control electrode 19 are present, so the potential of the first source region 13A and the second source region 13B is uniform, and the first source region 13A and the second source region 13B cannot be used as a resistive element R. However, in the region further back from L1 in Figure 2, the first source region 13A and the second source region 13B can be used as such a resistive element R.

[0042] Therefore, in Figure 2, if connection point P1, which is spaced apart across the region L2 behind the source electrode 18 in the first source region 13A on the left side of trench 10B, and connection point P2, which is spaced apart across the region L2 behind the second gate control electrode 19 in the second source region 13B on the right side of trench 10C, are connected by wiring W, then the resistance value corresponding to the region L2 in the longitudinal direction of source regions 13A and 13B (R in the figure) will be obtained. 01A resistive element R with twice the resistance value of ) can be substantially formed. In Figure 2, the wiring W is schematically shown, but in reality, similar to the interlayer insulating layer 17B described above, an insulating layer is formed in the region including the connection points, and then openings are formed in this insulating layer at locations corresponding to the connection points P1 and P2, and the wiring W is patterned with the same metal material as the source electrode 18 and the second gate control electrode 19. In reality, this wiring W can also be formed simultaneously with the source electrode 18 and the second gate control electrode 19 by patterning the source electrode 28 in Figure 5.

[0043] In other words, for a trench-type power MOSFET, a configuration can be easily realized in which a source electrode 18 and a second gate control electrode 19 are provided as shown in Figure 1, and a resistive element R is connected between a first source region 13A and a second source region 13B. In the above example, a resistor R01 is formed using the first source region 13A and the second source region 13B respectively, and the resistive element R is a series connection of two resistors R01. The resistance value of the resistive element R can be adjusted by adjusting the position of the connection point in Figure 2. Alternatively, one end of the wiring W may be connected to the second gate control electrode 19 or the source electrode 18, in which case the resistive element R is formed using only one of the first source region 13A and the second source region 13B.

[0044] Furthermore, similar to the semiconductor device 9 described above, the shield electrode 16 is irrelevant to the above operation. Therefore, if the operating speed of the semiconductor device 1 is sufficient for the intended use, it is not necessary to provide a shield electrode 16 in each trench.

[0045] Figure 3 shows examples of simulations calculating the source-drain voltage VSD and the source-drain current IS (forward current of the freewheel diode) when the drain electrode is grounded and the source electrode is positive, for both a conventional example using a body diode formed only from a pn junction and the structure shown in Figure 1 (an example). Here, the impurity concentrations of the drift layer and the body layer are 10 each. 16 ~1017 cm -3 , 10 17 cm -3 In Figure 1, the resistive element R is 10 3 It was identified as Omega.

[0046] In this result, in the conventional example, since the freewheeling diode is formed only from a pn junction, the VF is about 0.7V, and when VDS exceeds 0.7V, a current flows rapidly, while when VDS is 0.7V or less, only a low current similar to the reverse current flows. In contrast, in the embodiment, current flows from about 0.4V, which is smaller than the aforementioned VF. Therefore, in the embodiment, when the freewheeling diode is forward-biased, a large current can be passed, especially when VDS is small, compared to the technology described in Patent Document 1. As mentioned above, in the technology described in Patent Document 2, a Schottky diode is used as the freewheeling diode, and although the VF of a Schottky diode is generally smaller than 0.7V in the case of a pn junction, it is about 0.5V. In the embodiment, since current flows from about 0.4V, current can be passed with a smaller VDS than in the technology described in Patent Document 2. In particular, in the semiconductor device 1, although the voltage applied to the part that functions as a freewheeling diode (forward bias) is lower than that of the semiconductor device 9 due to the presence of the resistive element R, a large current can be passed, especially when VSD is small, compared to the conventional example.

[0047] Next, a modified example of the semiconductor device 1 described above will be explained. As described above, in this semiconductor device 1, the maximum electric field strength in the gate oxide film (second gate insulating film) 14D within the trench 10C can be reduced when the source potential is high. However, the location where the maximum electric field strength is highest in this gate oxide film 14D is region B (the lower ends on both the left and right sides of the second gate electrode 15B) enclosed by the dotted line in Figure 2.

[0048] When the source potential becomes higher than the drain potential, the body layer 12, which is opposite the second gate electrode 15B across the gate oxide film 14D, becomes depleted, and an (n-type) channel is formed in this depletion layer, thereby turning on this pseudo-MOSFET. On the other hand, this depletion layer is coupled with the depletion layer formed between the drift layer 11 (n-type layer) and the body layer 12 (p-type layer). The electric field (potential) applied between the second gate electrode 15B and the drain electrode 20 is mainly distributed between the gate oxide film 14D and this depletion layer, so widening this depletion layer is effective in reducing the maximum electric field strength in the gate oxide film 14D. In order to widen the width of the depletion layer that extends particularly towards the drift layer 11 from the boundary between the body layer 12 and the drift layer 11, it is preferable to lower the impurity concentration of the drift layer 11.

[0049] On the other hand, when the power MOSFET is turned on in trenches 10A and 10B, the on-current flows vertically through the drift layer 11. Therefore, if the impurity concentration of the drift layer 11 is reduced, the on-resistance of the power MOSFET will increase. For this reason, simply reducing the impurity concentration of the drift layer 11 is not desirable. For this reason, the impurity concentration of the drift layer 11 is 10 16 cm -3 It is considered to be to that extent.

[0050] Figure 4 is a cross-sectional view corresponding to Figure 1 of the semiconductor device 2, in which the maximum electric field strength in the gate oxide film 14D has been reduced taking this point into consideration. In the semiconductor substrate 40 used here, an n-type layer (n) with a lower impurity concentration than the drift layer 11 is placed between the drift layer 11 and the body layer 12. - A low impurity concentration layer (third semiconductor layer) 11A is formed. Except for the presence of the low impurity concentration layer 11A, this structure is no different from the semiconductor device 1 in Figure 1. The impurity concentration of the drift layer 11 is 10 16 cm -3 If the situation is such that the impurity concentration of the low impurity concentration layer 11A is lower than this, for example, 10 15 cm -3This is considered to be the extent to which the depletion layer tends to spread in the low impurity concentration layer 11A, which is closer to the body layer 12 than to the drift layer 11, and the maximum electric field strength in the gate oxide film 14D can be further reduced than that of the semiconductor device 1 in Figure 1.

[0051] On the other hand, particularly on the trench 10A and 10B sides, the presence of the low-impurity layer 11A, which has high resistivity, increases the on-resistance of the power MOSFET. However, since this on-resistance is approximately the sum of the component due to the impurity layer 11A and the component due to the drift layer 11, as shown in Figure 4, if the low-impurity layer 11A is made sufficiently thin compared to the drift layer 11, this increase in on-resistance becomes negligible. On the other hand, even if the low-impurity layer 11A is thin in this way, if the low-impurity layer 11A is in direct contact with the body layer 12, the effect of reducing the maximum electric field strength in the gate oxide film 14D as described above is significant.

[0052] The low-impurity layer 11A can be formed by epitaxial growth, similar to the body layer 12, and the impurity concentration can also be set during this growth process. Therefore, the semiconductor device 2 can be manufactured simply by adding the step of inserting the low-impurity layer 11A to the semiconductor device 1. In this semiconductor device 2, as described above, a large current can be passed when the freewheel diode is forward-biased, and the breakdown voltage at this time can be further increased compared to the semiconductor device 1.

[0053] In the above example, the semiconductor material constituting the semiconductor substrate was assumed to be Si, but this semiconductor material is arbitrary as long as a power MOSFET can be constructed in the same way. For example, SiC can be used instead of Si as described above.

[0054] Furthermore, while the above example described an n-channel power MOSFET in which the drift layer is n-type (first conductivity type) and the body layer is p-type (second conductivity type), it is also clear that a p-type power MOSFET with the conductivity types reversed can be constructed in a similar manner. [Explanation of Symbols]

[0055] 1, 2, 9 Semiconductor Equipment 10, 40 Semiconductor substrates 10A, 10B Trench (First groove) 10C Trench (Second Groove) 11. Drift layer (first semiconductor layer) 11A Low impurity concentration layer (third semiconductor layer) 12. Body layer (second semiconductor layer) 13A First Source Area 13B Second Source Area 14A Gate oxide film (first gate insulating film) 14B, 14C oxide film 14D gate oxide film (second gate insulating film) 15A 1st Terminal 15B Second Grid Terminal 16 Shielding electrodes 17A, 17B Interlayer insulating layer 18, 28 Source electrode (first main electrode) 19. Electrode for controlling the second gate 20 Drain electrode (second main electrode) R Resistance element W wiring

Claims

1. A semiconductor substrate is used, comprising: a first semiconductor layer made of a semiconductor material having a first conductivity type; and a second semiconductor layer having a second conductivity type opposite to the first conductivity type and formed on the first semiconductor layer, wherein a first groove is provided, which is dug out from the surface side so as to penetrate the second semiconductor layer and reach the first semiconductor layer, and a first source region having the first conductivity type is formed on the surface side so as to be in contact with the first groove and along the extending direction of the first groove. On the surface side of the semiconductor substrate, a first main electrode made of metal is connected to the second semiconductor layer and the first source region, A second main electrode connected to the first semiconductor layer on the back side of the semiconductor substrate, Within the first groove, a first gate electrode facing the inner surface of the first groove, with the first gate insulating film in between, It is equipped with, A semiconductor device in which the current flowing between the first main electrode and the second main electrode is controlled by the potential of the first gate electrode, In the semiconductor substrate, a second groove is dug out from the surface side at a location spaced apart from the first groove, A second source region having the first conductivity type, which is partially formed along the extending direction of the second groove so as to be in contact with the second groove on the surface side, A second gate control electrode, made of metal and separated from the first main electrode, is connected to the second semiconductor layer and the second source region on the surface side of the semiconductor substrate. Within the second groove, a second gate electrode faces the inner surface of the second groove, with a second gate insulating film thinner than the first gate insulating film in between. It is equipped with, The semiconductor device is characterized in that the second gate control electrode is connected to the second gate electrode and to the first main electrode via a resistive element.

2. The first main electrode contacts only a portion of the first source region in the direction of extension of the first groove in a plan view, and the resistive element is formed using a region of the first source region that does not contact the first main electrode. or The second gate control electrode contacts only a portion of the second source region in the direction of extension of the second groove in a plan view, and the resistive element is formed using a region of the second source region that does not contact the second gate control electrode. The semiconductor device according to feature 1.

3. In the thickness direction of the semiconductor substrate, The semiconductor device according to claim 1 or 2, characterized in that a third semiconductor layer having the first conductivity type, a lower impurity concentration than the first semiconductor layer, and being thinner than the first semiconductor layer is inserted between the first semiconductor layer and the second semiconductor layer.

4. The semiconductor device according to claim 1 or 2, characterized in that, in the first groove, a shield electrode that is insulated from the first semiconductor layer, the second semiconductor layer, and the first gate electrode and electrically connected to the first main electrode is provided below the first gate electrode.

Images