JP2009124667A - Bidirectional switch and method for driving the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To achieve a bidirectional switch in which a current is allowed to flow at least unidirectionally and a current flowing bidirectionally is cut off by controlling currents flowing from a first ohmic electrode to a second ohmic electrode and from the second ohmic electrode to the first ohmic electrode in one FET. <P>SOLUTION: A bidirectional switch includes a field-effect transistor 10 having a first ohmic electrode 16, a second ohmic electrode 17 and a gate electrode 18, and a control circuit 20 for controlling between a conduction state and a cut-off state by applying a bias voltage to the gate electrode 18. The control circuit 20 applies the bias voltage with a potential of the first ohmic electrode 16 as a reference when a potential of the second ohmic electrode17 is higher than the potential of the first ohmic electrode 16, and applies the bias voltage with the potential of the second ohmic electrode 17 as a reference when the potential of the second electrode 17 is lower than the potential of the first ohmic electrode 16. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a bidirectional switch and a driving method thereof.

  As semiconductor elements that perform power switching, power MOSFETs (metal oxide semiconductor field effect transistors), IGBTs (insulated gate bipolar transistors), thyristors, and the like are known. When a switching circuit capable of flowing a current bidirectionally using these semiconductor elements is configured, a bidirectional breakdown voltage is required for the semiconductor elements. Bidirectional withstand voltage means having a withstand voltage for both positive and negative voltages.

  Since power MOSFETs and IGBTs generally have a low reverse withstand voltage, for example, in order to realize a bidirectional switch using an IGBT, two IGBTs are connected in parallel in opposite directions as shown in FIG. It is necessary to connect a diode in series with each IGBT. In FIG. 13, the IGBT 201 and the diode 202, and the IGBT 203 and the diode 204 are connected in opposite directions. For this reason, when both the IGBT 201 and the IGBT 202 are turned on, a current flows bidirectionally, and when both the IGBT 201 and the IGBT 202 are turned off, a bidirectional high breakdown voltage can be realized.

  In a semiconductor element that performs such bidirectional switching, switching loss due to the product of a transient voltage and current during switching and conduction loss consumed by the resistance of the semiconductor element (referred to as on-resistance) in the on state are reduced. is important. However, when a bidirectional switching circuit is formed using a semiconductor element made of silicon (Si), it is difficult to reduce the on-resistance due to the material limit of Si.

In order to overcome the material limit and reduce conduction loss, introduction of a semiconductor element using a nitride semiconductor represented by GaN or a wide gap semiconductor such as silicon carbide (SiC) has been studied. A wide-gap semiconductor has a dielectric breakdown electric field about an order of magnitude higher than that of Si. In particular, a charge is generated at the heterojunction interface between aluminum gallium nitride (AlGaN) and gallium nitride (GaN) due to spontaneous polarization and piezoelectric polarization. As a result, a two-dimensional electron gas (2DEG) layer having a sheet carrier concentration of 1 × 10 ¹³ cm ⁻² or more and a high mobility of 1000 cm ² V / sec or more is formed even when undoped. For this reason, the AlGaN / GaN heterojunction field effect transistor (AlGaN / GaN-HFET) is expected as a power switching transistor that realizes low on-resistance and high breakdown voltage.

  However, a normal FET has a lower gate-source breakdown voltage than a gate-drain breakdown voltage. For this reason, even if the FET uses a wide gap semiconductor, two FETs and two protection diodes are required to realize a bidirectional switch.

In order to make the gate-source breakdown voltage equal to the gate-drain breakdown voltage, the distance between the gate electrode and the source electrode and the distance between the gate electrode and the drain electrode may be made equal. In this way, it has been proposed to use an FET having the same gate-drain breakdown voltage and gate-source breakdown voltage in a bidirectional switching circuit (see, for example, Patent Document 1).
US Patent Application Publication No. 2005/0189561

  However, even if the gate-drain breakdown voltage is equal to the gate-source breakdown voltage, a high breakdown voltage bidirectional switch cannot be realized. In a general FET, the current flowing from the drain electrode to the source electrode is controlled by applying a voltage between the gate electrode and the source electrode. However, even if a voltage is applied between the gate electrode and the source electrode, the current flowing from the source electrode to the drain electrode cannot be controlled. For this reason, there is a problem in that a single-FET alone cannot realize a bidirectional switch that must control a current that flows bidirectionally between the source electrode and the drain electrode.

  The present invention solves the above-described conventional problems, and in an FET including a first ohmic electrode and a second ohmic electrode that serve as a source electrode or a drain electrode, the first ohmic electrode to the second ohmic electrode and the second ohmic electrode It is an object of the present invention to realize a bidirectional switch that controls a current from two ohmic electrodes to a first ohmic electrode, allows a current to flow in at least one direction, and interrupts the bidirectional current.

  In order to achieve the above object, according to the present invention, the bidirectional switch is configured such that the bias voltage applied to the gate electrode is set to the first ohmic electrode or the second ohmic electrode according to the potential of the first ohmic electrode and the second ohmic electrode. A configuration is adopted in which the potential of the ohmic electrode is applied as a reference.

  Specifically, the bidirectional switch according to the present invention is a bidirectional switch that controls a conduction state in which a current flows in at least one direction between a first terminal and a second terminal and a cutoff state in which no current flows. A first ohmic electrode and a second ohmic electrode, one of which is a source electrode and the other is a drain electrode, and a gate electrode formed between the first ohmic electrode and the second ohmic electrode A field effect transistor having the first ohmic electrode connected to the first terminal and the second ohmic electrode connected to the second terminal; and applying a bias voltage to the gate electrode, The control circuit controls the potential of the first ohmic electrode when the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode. As a bias voltage is applied, the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, and applying a bias voltage potential of the second ohmic electrode as a reference.

  According to the bidirectional switch of the present invention, the control circuit applies a bias voltage based on the potential of the first ohmic electrode when the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode. When the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, a bias voltage is applied with reference to the potential of the second ohmic electrode. Therefore, it is possible to control both the current from the second ohmic electrode to the first ohmic electrode and the current from the first ohmic electrode to the second ohmic electrode. Therefore, a bidirectional switch that allows current to flow in at least one direction between the first ohmic electrode and the second ohmic electrode and interrupts the bidirectional current can be realized by one FET.

  In the control circuit according to the bidirectional switch of the present invention, the control circuit has a first power supply. When the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, the first power supply is When the first ohmic electrode is electrically connected to the gate electrode, a bias voltage is applied to the gate electrode, and the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, One power supply may be electrically connected between the second ohmic electrode and the gate electrode, and a bias voltage may be applied to the gate electrode. By setting it as such a structure, the bidirectional | two-way electric current between a 1st ohmic electrode and a 2nd ohmic electrode can be controlled reliably.

  In this case, the control circuit includes a first switch connected between the first power source and the first ohmic electrode, and a second switch connected between the first power source and the second ohmic electrode. A switch, and a drive circuit that drives the first switch and the second switch, and the drive circuit compares the potential of the first ohmic electrode with the potential of the second ohmic electrode based on the result of comparison. The on state and the off state of the first switch and the second switch may be switched. By setting it as such a structure, a conduction | electrical_connection state and a interruption | blocking state can be switched reliably.

  In the bidirectional switch of the present invention, the control circuit has a first power source and a second power source, and the first power source is supplied when the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode. Is electrically connected between the first ohmic electrode and the gate electrode, a bias voltage is applied to the gate electrode, and the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode. The bias voltage may be applied to the gate electrode by electrically connecting the second power source between the second ohmic electrode and the gate electrode. With such a configuration, the gate electrode can be reliably driven.

  In this case, the first power supply and the second power supply output a voltage higher than the threshold voltage of the field effect transistor when in the conductive state, and the threshold voltage of the field effect transistor when in the cutoff state. Alternatively, a lower voltage may be output. In addition, the first power supply outputs a voltage higher than the threshold voltage of the field effect transistor, the second power supply outputs a voltage lower than the threshold voltage of the field effect transistor, and the first terminal and the second terminal A current in one direction may be passed between them, and a current in the opposite direction may be cut off.

  In the bidirectional switch of the present invention, the control circuit includes a first switch connected between the first power supply and the gate electrode, and a second switch connected between the second power supply and the gate electrode. , And a driving circuit that drives the first switch and the second switch, and the driving circuit compares the potential of the first ohmic electrode with the potential of the second ohmic electrode based on the result of comparing the first ohmic electrode and the second ohmic electrode. The on state and the off state of the switch and the second switch may be switched.

  In this case, the first switch and the second switch are a first photocoupler and a second photocoupler, respectively, and the drive circuit is applied between the first ohmic electrode and the second ohmic electrode. The differential amplifier may drive the light emitting element of the first photocoupler and the light emitting element of the second photocoupler.

  In addition, the first switch and the second switch are a first photocoupler and a second photocoupler, respectively, and the drive circuit is a voltage applied between the first ohmic electrode and the second ohmic electrode. Is applied to the non-inverting input terminal, and the inverting input terminal is applied between the first ohmic electrode and the second ohmic electrode, with the inverting input terminal connected to the first ohmic electrode. And a second differential amplifier having a non-inverting input terminal connected to the first ohmic electrode, the first differential amplifier including a first differential amplifier having a voltage corresponding to the generated voltage input to the inverting input terminal. The light emitting element of the photocoupler may be driven, and the second differential amplifier may drive the light emitting element of the second photocoupler.

  In the bidirectional switch of the present invention, the control circuit includes a first diode and a second diode in which the anode terminals are connected to each other and are electrically connected between the first ohmic electrode and the second ohmic electrode. And a first power source connected to a connection node where the anode terminal of the first diode and the anode terminal of the second diode are connected to each other, and the first power source is in a conductive state. A voltage higher than the threshold voltage of the field effect transistor may be output, and in the cut-off state, a voltage lower than the threshold voltage of the field effect transistor may be output. With this configuration, a bidirectional switch can be configured with one variable power source.

  In this case, the control circuit may include a third diode electrically connected between the connection node and the gate electrode.

  In the bidirectional switch of the present invention, the field effect transistor is a normally-on type, and the control circuit includes a first diode in which a gate electrode and a cathode terminal are connected, a first ohmic electrode, and an anode terminal of the first diode. Between the second diode connected via the first switch so that the cathode terminal is on the first ohmic electrode side, and the second ohmic electrode and the anode terminal of the first diode A third diode connected via a second switch so that the cathode terminal is on the second ohmic electrode side, and an anode terminal and the first terminal of the first diode are connected. 3, a second switch interposed between the second terminal of the third switch and the first ohmic electrode, and a threshold value of the field effect transistor A threshold voltage of the field-effect transistor, which is connected to the first power supply that outputs a voltage lower than the voltage, and the fifth switch between the second terminal of the third switch and the second ohmic electrode. A second power source that outputs a lower voltage than the first power source. In the conductive state, the third switch is turned off, the first switch and the second switch are turned on, and in the cut-off state, the second power source is output. When the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, the third switch and the fourth switch are turned on, and the first switch, the second switch, and the fifth switch are turned on. When the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, the third switch and the fifth switch are turned on and the first switch, the second switch Switch and fourth switch may be used as the off-state. With this configuration, it is not necessary to use a variable power supply.

  In this case, the control circuit includes a first drive circuit that drives the fourth switch and the fifth switch, and a second drive circuit that drives the first switch, the second switch, and the third switch. And the first drive circuit compares the potential of the first ohmic electrode and the potential of the second ohmic electrode, and based on the comparison result, the fourth switch and the fifth switch are turned on. The second drive circuit switches the third switch to the on state and the first switch and the second switch to the off state in the conductive state, and turns off the third switch in the cutoff state. The switch may be turned off and the first switch and the second switch may be turned on.

  The field effect transistor is a normally-off type, and the control circuit includes a first diode having a gate electrode and a cathode terminal connected to each other, and a cathode between the first ohmic electrode and the anode terminal of the first diode. A cathode terminal is interposed between the second diode connected via the first switch so that the terminal is on the first ohmic electrode side, and the second ohmic electrode and the anode terminal of the first diode. A third diode connected via a second switch so as to be on the second ohmic electrode side, a third switch connected to the anode terminal and the first terminal of the first diode, The third switch is connected between the second terminal of the third switch and the first ohmic electrode, and a voltage higher than the threshold voltage of the field effect transistor is output. The first power source, the second terminal of the third switch, and the second ohmic electrode are connected with a fifth switch interposed therebetween, and outputs a voltage higher than the threshold voltage of the field effect transistor. A second power source, and in the cut-off state, the third switch is turned off, the first switch and the second switch are turned on, and in the conductive state, the potential of the second ohmic electrode is When the potential is higher than the potential of the first ohmic electrode, the third switch and the fourth switch are turned on, the first switch, the second switch, and the fifth switch are turned off. When the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, the third switch and the fifth switch are turned on, and the first switch, the second switch, and the fourth switch are turned on. Chi may be used as the off state.

  In this case, the control circuit includes a first drive circuit that drives the fourth switch and the fifth switch, and a second drive circuit that drives the first switch, the second switch, and the third switch. And the first drive circuit determines whether the fourth switch and the fifth switch are on or off based on the result of comparing the potential of the first ohmic electrode and the potential of the second ohmic electrode. The switching and second drive circuit sets the first switch to the on state and the second switch and the third switch to the off state in the cut-off state, and sets the first switch to the off state in the conductive state. In addition, the second switch and the third switch may be turned on.

  In the bidirectional switch of the present invention, the field effect transistor has a semiconductor layer formed on the substrate, and the first ohmic electrode and the second ohmic electrode are formed on the semiconductor layer at a distance from each other. The gate electrode is formed on the semiconductor layer so that the distance between the first ohmic electrode and the gate electrode is equal to the distance between the second ohmic electrode and the gate electrode. Also good. With such a configuration, it is possible to realize a field effect transistor in which the withstand voltage between the first ohmic electrode and the gate electrode and the withstand voltage between the second ohmic electrode and the gate electrode are equal to each other.

  In this case, the field effect transistor has an insulating film formed between the first ohmic electrode and the gate electrode, and the withstand voltage of the insulating film is higher than the withstand voltage between the first ohmic electrode and the gate electrode. It is preferable.

  In this case, the insulating film may include a first insulating film and a second insulating film that are sequentially stacked.

  The first insulating film may be made of silicon oxide containing phosphorus, polyimide, or benzocyclobutene.

  The second insulating film may be made of silicon nitride or silicon oxide.

  In the bidirectional switch of the present invention, the substrate is conductive, and the field-effect transistor includes a back electrode formed on the back surface of the substrate, a semiconductor layer, and the first ohmic electrode and the back electrode through the conductive substrate. Wiring may be electrically connected to each other.

  In this case, the field effect transistor has a second ohmic electrode pad connected to the second ohmic electrode and a gate electrode pad connected to the gate electrode, and the second ohmic electrode pad and the gate electrode pad are At least one of the above may be formed on a high resistance region in the semiconductor layer.

  In the bidirectional switch of the present invention, the field effect transistor has a p-type semiconductor layer formed between the gate electrode and the semiconductor layer, and the semiconductor layer includes a first layer formed sequentially from the lower side and And the second layer may be made of an n-type semiconductor.

  In the bidirectional switch of the present invention, the semiconductor layer may be made of a nitride semiconductor or silicon carbide.

  The bidirectional switch driving method according to the present invention has a field effect transistor in a conductive state in which current flows in at least one direction between the first ohmic electrode and the second ohmic electrode, and a cut-off state in which no current flows. Targeting a driving method for driving as a bidirectional switch, the step of comparing the potential of the second ohmic electrode with the potential of the first ohmic electrode, and the potential of the second ohmic electrode being higher than the potential of the first ohmic electrode Is higher than the potential of the first ohmic electrode, a bias voltage is applied to the gate electrode of the field effect transistor with reference to the potential of the first ohmic electrode, and the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode. Applying a bias voltage to the gate electrode with respect to the second ohmic electrode as a reference. Voltage is higher than the threshold voltage of the field effect transistor, when the cut-off state, characterized by a bias voltage lower than the threshold voltage.

  In the bidirectional switch driving method according to the present invention, when the potential of the drain electrode is higher than the potential of the first ohmic electrode, the bias voltage is applied to the gate electrode of the field effect transistor with reference to the potential of the first ohmic electrode. When the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, a step of applying a bias voltage to the gate electrode with respect to the second ohmic electrode is provided. Therefore, it is possible to control both the current from the second ohmic electrode to the first ohmic electrode and the current from the first ohmic electrode to the second ohmic electrode. Therefore, it is possible to realize a bidirectional switch that allows a current to flow in at least one direction between the first ohmic electrode and the second ohmic electrode by one FET and interrupts the bidirectional current.

  According to the bidirectional switch of the present invention, the current from the first ohmic electrode to the second ohmic electrode and the current from the second ohmic electrode to the first ohmic electrode is controlled, and the current flows in at least one direction. A bidirectional switch that cuts off the current in the direction can be realized.

(First embodiment)
A first embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows a circuit configuration of a bidirectional switch according to the first embodiment. As shown in FIG. 1, the bidirectional switch of the present embodiment includes a field effect transistor (FET) 10 and a control circuit 20 that controls the FET 10, and includes a first terminal 31 and a second terminal 32. A conduction state in which a current flows in at least one direction therebetween and a cutoff state in which no current flows are controlled. FIG. 1 shows an example in which a load circuit 40 in which a bidirectional power supply 41 and a load 42 are connected in series is connected between a first terminal 31 and a second terminal 32.

  The control circuit 20 is connected to the first ohmic electrode 16, the gate electrode 18, and the second ohmic electrode 17 of the FET 10. The control circuit 20 applies a gate bias voltage to the gate electrode 18 with the potential of the first ohmic electrode 16 as a reference when the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16. The current flowing from the second ohmic electrode 17 to the first ohmic electrode 16 is controlled. Further, when the potential of the second ohmic electrode 17 is lower than the potential of the first ohmic electrode 16, a gate bias voltage is applied to the gate electrode 18 with the potential of the second ohmic electrode 17 as a reference, and the first ohmic electrode 17 The current flowing from the electrode 16 to the second ohmic electrode 17 is controlled. With such a configuration and operation, bidirectional current between the first ohmic electrode 16 and the second ohmic electrode 17 of the FET 10 can be controlled.

  The FET 10 in this embodiment is not particularly limited, but the following may be used. On a substrate 11 made of silicon (Si) having a principal plane orientation of (111), 10 nm thick aluminum nitride (AlN) and 10 nm thick gallium nitride (GaN) are alternately stacked. A buffer layer 12 having a thickness of 1 μm is formed, and a semiconductor layer 13 is formed on the buffer layer 12. The semiconductor layer 13 is formed by laminating an undoped GaN layer 14 having a thickness of 2 μm and an AlGaN layer 15 having a thickness of 20 nm doped with Si. A two-dimensional electron gas (2DEG) layer is generated in the interface region between the GaN layer 14 and the AlGaN layer 15.

  On the semiconductor layer 13, the 1st ohmic electrode 16 and the 2nd ohmic electrode 17 which consist of titanium (Ti) and aluminum (Al) are formed at intervals. One of the first ohmic electrode 16 and the second ohmic electrode 17 serves as a source electrode and the other serves as a drain electrode. The first ohmic electrode 16 and the second ohmic electrode 17 are formed in a portion where the AlGaN layer 15 is removed and the GaN layer 14 is dug down by about 40 nm in order to reduce the contact resistance with the 2DEG layer. A gate electrode 18 made of palladium (Pd) and gold (Au) is formed between the first ohmic electrode 16 and the second ohmic electrode 17 on the semiconductor layer 13.

  With such a configuration, a heterojunction FET (HFET) having a threshold voltage of about −2 V can be realized. Further, by equalizing the distance between the gate electrode 18 and the first ohmic electrode 16 and the distance between the gate electrode 18 and the second ohmic electrode 17, the distance between the gate electrode 18 and the first ohmic electrode 16 is set. And the breakdown voltage between the gate electrode 18 and the second ohmic electrode 17 can be made equal.

  Note that the distance between the second ohmic electrode and the gate electrode and the distance between the first ohmic electrode and the gate electrode are desirably equal, but may not be equal as long as a desired bidirectional breakdown voltage is obtained. In addition, that the distance between the second ohmic electrode and the gate electrode and the distance between the first ohmic electrode and the gate electrode are equal means that they are equal within the range of photolithography alignment accuracy in actual device fabrication. Specifically, the alignment accuracy is about ± 1 μm, and the distance between the second ohmic electrode and the gate electrode and the distance between the first ohmic electrode and the gate electrode may have an error of about ± 1 μm. Absent.

  The first ohmic electrode 16 of the FET 10 is connected to the first terminal 31 and grounded, and the second ohmic electrode 17 is connected to the second terminal 32. The gate electrode 18 is connected to the output of the control circuit 20.

  The control circuit 20 includes a first power supply 22A connected between the first ohmic electrode 16 and the gate electrode 18 via the first switch 21A, and the second ohmic electrode 17 and the gate electrode 18. And a second power source 22B connected via a second switch 21B.

  The first power supply 22 </ b> A and the second power supply 22 </ b> B have a voltage higher than the threshold voltage of the FET 10 in a conductive state in which current flows in both directions between the first ohmic electrode 16 and the second ohmic electrode 17. This is a variable power supply that outputs and outputs a voltage lower than the threshold voltage in a cut-off state where no current flows between the first ohmic electrode 16 and the second ohmic electrode 17.

  The first switch 21A is turned on when the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16, and the potential of the second ohmic electrode 17 is the potential of the first ohmic electrode 16. If it is lower than that, it is turned off. On the other hand, the second switch 21B is turned off when the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16, and the potential of the second ohmic electrode 17 is set to the first ohmic electrode 16. When it is lower than the potential, it is turned on. Therefore, when the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16, the gate electrode 18 is connected to the first power source 22A, and the first ohmic electrode 16 and the gate electrode 18 are connected to each other. A voltage is applied between them. When the potential of the second ohmic electrode 17 is lower than the potential of the first ohmic electrode 16, the gate electrode 18 is connected to the second power source 22B, and the second ohmic electrode 17 and the gate electrode 18 are connected. A voltage is applied.

  The operation of the bidirectional switch of this embodiment will be described below. For example, when the potential of the second ohmic electrode 17 is +100 V with respect to the potential of the first ohmic electrode 16, the first switch 21A is turned on and the second switch 21B is turned off. Therefore, a voltage is applied between the first ohmic electrode 16 and the gate electrode 18 by the first power supply 22A. Accordingly, when the output voltage of the first power supply 22A is higher than the threshold voltage of the FET 10, for example + 1V, a current flows from the second ohmic electrode 17 to the first ohmic electrode 16, and is lower than the threshold voltage, for example − In the case of 5 V, the current flowing from the second ohmic electrode 17 to the first ohmic electrode 16 can be cut off.

  On the other hand, when the potential of the second ohmic electrode 17 is −100 V with respect to the potential of the first ohmic electrode 16, the first switch 21A is turned off and the second switch 21B is turned on. Therefore, a voltage is applied between the second ohmic electrode 17 and the gate electrode 18 by the second power supply 22B. Therefore, when the output voltage of the second power supply 22B is higher than the threshold voltage of the FET 10, for example + 1V, a current flows from the first ohmic electrode 16 to the second ohmic electrode 17 and is lower than the threshold voltage, for example − In the case of 5 V, the current flowing from the first ohmic electrode 16 to the second ohmic electrode 17 can be cut off.

  As described above, the bidirectional switch of the present embodiment controls the conductive state in which current flows in both directions between the first ohmic electrode 16 and the second ohmic electrode 17 and the cutoff state in which no current flows. can do. That is, a bidirectional switch can be realized by a single FET.

  The first switch 21A and the second switch 21B may be switches that can be switched according to the potential of the first ohmic electrode 16 and the potential of the second ohmic electrode 17, and may be electronic even if they are mechanical switches. A simple switch may be used.

  Alternatively, the first power supply 22A may output a voltage higher than the threshold voltage of the FET 10, and the second power supply 22B may output a voltage lower than the threshold voltage of the FET 10. In this case, an operation as a so-called diode that cuts off the current from the first ohmic electrode 16 to the second ohmic electrode 17 by passing a current from the second ohmic electrode 17 to the first ohmic electrode 16 is performed. It becomes possible to make it. Similarly, if the first power supply 22A outputs a voltage lower than the threshold voltage of the FET 10 and the second power supply 22B outputs a voltage higher than the threshold voltage of the FET 10, the first ohmic electrode 16 An operation of passing a current to the second ohmic electrode 17 and interrupting a current from the second ohmic electrode 17 to the first ohmic electrode 16 becomes possible.

  2 and 3 show specific examples of the control circuit 20. The control circuit 20 shown in FIG. 2 uses photocouplers as the first switch 21A and the second switch 21B, and the first photocoupler 51A that is the first switch 21A and the second switch 21B that is the second switch 21B. The photocoupler 51B is driven by the drive circuit 50.

  The drive circuit 50 includes a differential amplifier (op-amp) 52 to which both positive and negative voltages are supplied by a power source 53A and a power source 53B. The output of the operational amplifier 52 is connected to the anode terminal of the light emitting diode (LED) of the first photocoupler 51A and the cathode terminal of the LED of the second photocoupler 51B via the third resistance element R3. The cathode terminal of the LED of the first photocoupler 51 </ b> A and the anode terminal of the LED of the second photocoupler 51 </ b> B are connected to the first ohmic electrode 16. The voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is divided by the first resistance element R1 and the second resistance element R2 at the positive input terminal (non-inverting input terminal) of the operational amplifier 52. The pressed voltage is applied, and the negative input terminal (inverted input terminal) is connected to the first ohmic electrode 16 and is at the ground potential.

  With the above configuration, when the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16, the positive voltage divided by the resistance element R1 and the resistance element R2 is the operational amplifier 52. Is input to the positive input terminal. The operational amplifier 52 compares the potential of the positive input terminal with the potential of the negative input terminal, and outputs a voltage obtained by multiplying the potential of the positive input terminal minus the potential of the negative input terminal by the amplification factor. When the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16, the potential of the positive input terminal is positive and the potential of the negative input terminal is 0V. The voltage obtained by subtracting the negative input terminal potential from the negative potential is positive. Therefore, the operational amplifier 52 outputs a positive voltage. Thereby, a current flows through the LED of the first photocoupler 51A, and the first photocoupler 51A is turned on. On the other hand, since no current flows through the LED of the second photocoupler 51B, the second photocoupler 51B is turned off. Therefore, a voltage is applied between the first ohmic electrode 16 and the gate electrode 18 by the first power supply 22A, and the current flowing from the second ohmic electrode 17 to the first ohmic electrode 16 by the first power supply 22A. Can be controlled.

  Further, when the potential of the second ohmic electrode 17 is lower than the potential of the first ohmic electrode 16, a negative voltage divided by the resistance element R 1 and the resistance element R 2 is input to the operational amplifier 52. The operational amplifier 52 outputs a negative voltage. As a result, the first photocoupler 51A is turned off and the second photocoupler 51B is turned on. Therefore, a voltage is applied between the second ohmic electrode 17 and the gate electrode 18 by the second power supply 22B, and a current flowing from the first ohmic electrode 16 to the second ohmic electrode 17 by the second power supply 22B. Can be controlled.

  The resistance value of the first resistance element R1 and the resistance value of the second resistance element R2 are divided by both resistance elements, and the maximum voltage input to the operational amplifier 52 is equal to or lower than the maximum voltage at which the operational amplifier 52 can operate. For example, when the potential difference between the first ohmic electrode 16 and the second ohmic electrode 17 is 100 V, the resistance value of the first resistance element R1 is set to 50 kΩ, The resistance value of the resistor element R2 may be 950 kΩ. However, when the voltage of the load power supply is equal to or lower than the maximum voltage at which the operational amplifier can operate, the positive input terminal of the operational amplifier is connected to the second ohmic electrode without providing the first resistance element R1 and the second resistance element R2. It is good also as a structure connected to. The third resistance element R3 is a protection resistor for the photocoupler LED. For example, when the output of the operational amplifier 52 is 5 V, and the forward voltage and current of the LED are 1.6 V and 10 mA, respectively, 340Ω is obtained. do it.

  The control circuit 20 may be a circuit as shown in FIG. As shown in FIG. 3, the drive circuit 50 includes a first operational amplifier 54A and a second operational amplifier 54B to which a positive voltage is supplied by a power supply 55. The voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is applied to the positive input terminal of the first operational amplifier 54A and the negative input terminal of the second operational amplifier 54B. A voltage divided by the second resistance element R2 is applied. The negative input terminal of the first operational amplifier 54A and the positive input terminal of the second operational amplifier 54B are connected to the first ohmic electrode and are grounded. The output terminal of the first operational amplifier 54A is connected to the anode terminal of the LED of the first photocoupler 51A via the fourth resistance element R4, and the output terminal of the second operational amplifier 54B is connected to the fifth resistance element R5. Is connected to the anode terminal of the LED of the second photocoupler 51B. The cathode terminal of the LED of the first photocoupler 51A and the cathode terminal of the LED of the second photocoupler 51B are connected to the first ohmic electrode and grounded.

  For example, the resistance value of the first resistance element R1 is 50 kΩ, and the resistance value of the second resistance element R2 is 950 kΩ. When the potential of the second ohmic electrode 17 is +100 V with respect to the potential of the first ohmic electrode 16, the voltage is divided by the first resistance element R1 and the second resistance element R2, and the first operational amplifier 54A. A differential voltage of + 5V is input to the second operational amplifier 54B, and a differential voltage of −5V is input to the second operational amplifier 54B. If the voltage of the power supply 55 is 10V and the amplification factor of the operational amplifier is 1000 times, the circuit is saturated due to the high voltage amplification factor, 10V is output to the output terminal of the first operational amplifier 54A, and the output of the second operational amplifier 54B. 0V is output to the terminal. Therefore, the first photocoupler 51A is turned on, and the second photocoupler 51B is turned off.

  When the potential of the second ohmic electrode 17 is −100 V with respect to the potential of the first ohmic electrode 16, a differential voltage of −5 V is input to the first operational amplifier 54A, and the second operational amplifier 54B receives the differential voltage. Is inputted with a differential voltage of + 5V. Accordingly, the first photocoupler 51A is turned off, and the second photocoupler 51B is turned on.

  In this case as well, when the voltage of the load power supply is equal to or lower than the maximum voltage at which the operational amplifier can operate, the first resistance element R1 and the second resistance element R2 are not provided. The input terminal and the negative input terminal of the second operational amplifier may be connected to the second ohmic electrode. The resistance values of the fourth resistance element R4 and the fifth resistance element R5 may be values that can protect the LED of the photocoupler. For example, the output of the operational amplifier is 10V, and the forward voltage and current of the LED are In the case of 1.6 V and 10 mA, for example, 840Ω may be used.

  In the first embodiment, the case where the FET 10 is a normally-on type whose threshold voltage is a negative voltage has been described. However, a normally-off type FET whose threshold voltage is a positive voltage may be used. When a normally-off type FET is used, for example, an FET having a threshold voltage of +1 V, the output voltages of the first power source and the second power source are higher than the threshold voltage, for example, +5 V in order to turn on the bidirectional switch. Is preferably output, and it is desirable to output, for example, 0 V, which is lower than the threshold voltage, in order to enter the cutoff state.

  Also, by setting one of the voltage of the first power supply 22A and the voltage of the second power supply 22B to a voltage higher than the threshold voltage and setting the other to a value lower than the threshold voltage, the first ohmic electrode 16 and the second ohmic electrode A bidirectional switch that performs a so-called diode operation that allows current to flow only in one direction with respect to 17 and interrupts the other direction can be provided.

  In the first embodiment, an example in which a variable power source is used as the first power source and the second power source for controlling the bidirectional switch has been described. However, a general gate drive circuit may be used instead of the variable power source. However, it is desirable to use an insulated DC voltage converter (DC-DC converter) that is not grounded as a power source connected to the gate drive circuit connected to the second ohmic electrode side. Further, as a power supply that is not grounded, a photocoupler capable of outputting a voltage, a battery, or the like may be used.

(First modification of the first embodiment)
Below, the 1st modification of the 1st Embodiment of this invention is demonstrated with reference to drawings. FIG. 4 shows a configuration of a bidirectional switch according to a first modification of the first embodiment.

  As shown in FIG. 4, in the bidirectional switch of this modification, the third power source 22C connected to the gate electrode 18 is connected to the first ohmic electrode 16 with the first switch 21A interposed therebetween. It is connected to the second ohmic electrode 17 through the two switches 21B.

  By driving the first switch 21A and the second switch 21B in the same manner as in the first embodiment, the bidirectional switch of the present modification performs the same operation as the bidirectional switch in the first embodiment. As the drive circuit for driving the first switch 21A and the second switch 21B, the one shown in the first embodiment can be used. The third power source 22C is an insulated power source, and is configured by an insulated DC / DC converter, a battery, and the like.

  With such a configuration, bidirectional current can be controlled by a single power source, and the circuit can be simplified and reduced in cost.

(Second modification of the first embodiment)
Below, the 2nd modification of the 1st Embodiment of this invention is demonstrated with reference to drawings. 5 and 6 are FETs used in the bidirectional switch according to this modification, FIG. 5 shows a planar configuration, and FIG. 6 shows a cross-sectional configuration taken along line VI-VI in FIG.

  As shown in FIGS. 5 and 6, the FET of this modification is a multi-finger type FET, and the unit 101 including the first ohmic electrode 116, the gate electrode 118, and the second ohmic electrode 117 is a second ohmic electrode. It can be considered that a plurality of electrodes 117 are alternately arranged around the electrode 117. Each first ohmic electrode 116 is electrically connected to a first ohmic electrode pad 131 formed on the surface (back surface) opposite to the surface on which the semiconductor layer 113 of the substrate 111 is formed. Each second ohmic electrode 117 is electrically connected to a second ohmic electrode pad 130 formed on the surface side of the substrate, and each gate electrode 118 is a gate electrode pad 132 formed on the surface side of the substrate 111. And are electrically connected. Thereby, the gate width of the FET can be greatly increased, and an FET capable of operating at a large current can be configured.

  Specifically, on a substrate 111 made of conductive silicon (Si) having a principal plane orientation of (111), 10 nm thick aluminum nitride (AlN) and 10 nm thick gallium nitride (GaN) A buffer layer 112 having a thickness of 1 μm is formed by alternately laminating the layers. A semiconductor layer 113 is formed on the buffer layer 112 by laminating an undoped GaN layer 114 having a thickness of 2 μm and an AlGaN layer 115 doped with Si having a thickness of 20 nm. A two-dimensional electron gas (2DEG) layer is generated in the interface region between the GaN layer 114 and the AlGaN layer 115.

  Impurity ions such as boron are implanted into the semiconductor layer 113 except for the active region 140. Thereby, part of the semiconductor layer 113 selectively becomes the insulating film 142, and the high resistance region 141 is formed. On the active region 140, a first ohmic electrode 116, a second ohmic electrode 117, and a gate electrode 118 are formed at a distance from each other. Focusing on the two units 101, the two units 101 share the second ohmic electrode 117. A first ohmic electrode 116 is formed on both sides of the shared second ohmic electrode 117, and a gate electrode 118 is formed between the second ohmic electrode 117 and the first ohmic electrode 116. With this arrangement, the distance between the second ohmic electrode 117 and the gate electrode 118 is equal to the distance between the first ohmic electrode 116 and the gate electrode 118, and the gate electrode 118 and the second ohmic electrode are equal. The breakdown voltage between the electrode 117 and the breakdown voltage between the first ohmic electrode 116 and the gate electrode 118 can be equalized with a high breakdown voltage. In the FET of this modification, the distance between the second ohmic electrode 117 and the gate electrode 118 and the distance between the first ohmic electrode 116 and the gate electrode 118 are 10 μm.

  The first ohmic electrode 116 and the second ohmic electrode 117 are formed in a portion where the AlGaN layer 115 is removed and the GaN layer 114 is dug down by about 40 nm in order to reduce the contact resistance with the 2DEG layer. Further, each first ohmic electrode 116 is electrically connected to the conductive substrate 111 by an interelectrode wiring 125. The interelectrode wiring 125 is formed in a contact hole formed by selectively removing the semiconductor layer 113 and the buffer layer 112 and a part of the substrate 111. On the other hand, a first ohmic electrode pad 131 is formed on the back surface of the substrate 111. The first ohmic electrode pad 131 is made of, for example, gold (Au) and tin (Sn), forms an ohmic junction with the conductive substrate 111, and is electrically connected. Therefore, the first ohmic electrode 116 is electrically connected to the first ohmic electrode pad 131 via the interelectrode wiring 125 and the substrate 111.

A protective film 121 made of SiN, a first insulating film 122 and a second insulating film 123 are sequentially formed on the semiconductor layer 113. The first insulating film 122 has a flat upper surface. Further, a tapered opening that exposes the second ohmic electrode 117 is formed. The second insulating film 123 is formed so as to cover the upper surface of the first insulating film 122 and the side surface of the opening. The first insulating film 122 may be a film made of SiO ₂ containing phosphorus having a thickness of 6 μm, for example. Alternatively, a polyimide film, a benzocyclobutene (BCB) film, or the like may be used. By using a SiO ₂ film containing phosphorus for the first insulating film 122, the film stress of the first insulating film 122 can be relieved and peeling of the film can be prevented. In addition, since the phosphorus gettering effect can prevent an alkaline impurity from entering the semiconductor layer 113 included in the transistor, the reliability of the transistor can be improved. Further, SiN having a thickness of 0.2 μm may be used for the second insulating film 123.

  Note that the dielectric breakdown electric field strength of the silicon oxide film containing phosphorus, the polyimide film, and the BCB film formed by chemical vapor deposition (CVD) is about 3 MV / cm. However, considering the uneven shape of the nitride semiconductor device and variations in film characteristics, it is necessary to design the nitride semiconductor device assuming that the dielectric breakdown electric field strength is about 1 MV / cm. Therefore, in order to realize a nitride semiconductor device having a withstand voltage of 200 V or higher, the thickness of the first insulating film 122 is preferably 2 μm or more. Further, in order to realize a high breakdown voltage, the thickness may be 5 μm or more, and if it is 10 μm or more, the breakdown voltage can be improved. If the thickness is too thick, there is a problem that side etching is excessively caused by wet etching for opening, so that it is preferably 25 μm or less, more preferably 20 μm or less.

  A second ohmic electrode pad 130 made of Al having a thickness of 4 μm is formed on the second insulating film 123. The second ohmic electrode pad 130 fills the opening and is electrically connected to the second ohmic electrode 117 exposed from the opening. The second ohmic electrode pad 130 may be formed of Al having a thickness of 4 μm, for example.

  Each gate electrode 118 extends to the high resistance region 141 surrounding the active region 140, and the portions formed on the high resistance region 141 are connected to each other. A portion of the gate electrode 118 formed on the high resistance region 141 functions as a gate wiring. A gate electrode pad 132 is formed on the high resistance region 141 with the first insulating film 122 and the second insulating film 123 interposed therebetween. A part of the gate electrode pad 132 is formed in the first insulating film 122 so as to fill an opening exposing the gate electrode 118, and the gate electrode pad 132 and the gate electrode 118 are electrically connected. Has been.

  Further, the gate electrode pad 132 and the second ohmic electrode pad 130 are formed so as to ensure the interval α. By maintaining the distance α, air discharge can be prevented from occurring between the gate electrode pad 132 and the second ohmic electrode pad 130. In this modification, the interval α is set to 100 μm.

  In the FET of this modification, the first ohmic electrode pad 131 is formed on the back surface of the substrate 111. For this reason, the intersection of the gate electrode 118 and the first ohmic electrode 116 does not occur. Also, the distance between the first ohmic electrode pad 131 and the gate electrode 118 can be sufficiently separated. In particular, the resistance of the semiconductor layer 13 is increased in the region outside the active region 140. Therefore, it is possible to sufficiently insulate between the substrate 111 and the portion functioning as the gate wiring of the gate electrode 118. On the other hand, the gate electrode 118 and the second ohmic electrode pad 130 are also sufficiently insulated by the protective film 121, the first insulating film 122, and the second insulating film 123. Therefore, the FET of this modification can secure sufficient withstand voltage between the gate electrode 118 and the first ohmic electrode 116 and between the gate electrode 118 and the second ohmic electrode 117, and can be increased by combining the control circuit. A voltage-proof bidirectional switch can be realized.

  The same control circuit as that of the first embodiment and the first modification of the first embodiment can be used.

  The high resistance region 141 is formed by selectively injecting ions such as boron into the semiconductor layer 113 and selectively forming another insulating film on the semiconductor layer 113 instead of forming the insulating film 142. It may be formed.

  Further, as shown in FIGS. 7 and 8, the second ohmic electrode pad 130 may be formed not on the active region 140 but on the high resistance region 141. With such a structure, a thin insulating film 124 may be formed instead of the thick insulating film that insulates between the second ohmic electrode pad 130 and the gate electrode 118, and the process is performed. Can be reduced. In addition, by forming the second ohmic electrode pad 130 on the high resistance region 141, it is possible to prevent an impact during wire bonding from being applied to the active region 140. Thereby, the advantage that the reliability of FET can be improved is also acquired.

  It is also possible to use a multi-finger type FET as shown in FIG. In this case, the second ohmic electrode pad 160, the first ohmic electrode pad 161, and the gate electrode pad 162 are formed on the surface of the substrate. Therefore, the gate electrode 158 and the second ohmic electrode 157 do not intersect, but the gate electrode 158 and the first ohmic electrode 156 intersect. However, the insulating film (not shown) formed between the gate electrode 158 and the first ohmic electrode 156 is sufficiently thickened at the portion where the gate electrode 158 and the first ohmic electrode 156 intersect. That's fine. In this manner, if the withstand voltage of the insulating film is set to be equal to or higher than the withstand voltage between the gate electrode 158 and the first ohmic electrode 156 and the withstand voltage between the gate electrode 158 and the first ohmic electrode 156 is ensured, FIG. FETs as shown in (2) can also be used for bidirectional switches. In this case, it is preferable to use polyimide, silicon oxide containing phosphorus, BCB, or the like for the insulating film between the gate electrode 158 and the first ohmic electrode 156. Moreover, it is good also as a laminated film.

(Second Embodiment)
The second embodiment of the present invention will be described below with reference to the drawings. FIG. 10 shows a circuit configuration of the bidirectional switch according to the second embodiment. In FIG. 10, the same components as those in FIG.

  In the bidirectional switch of this embodiment, a normally-off type FET is used as the FET 10. In the normally-off type FET, magnesium (Mg) is doped between the first ohmic electrode 16 and the second ohmic electrode 17 on the semiconductor layer 13 in the FET 10 described in the first embodiment. A FET in which a gate electrode 18 made of Pd and Au is formed with a p-type semiconductor layer 19 made of p-type GaN having a thickness of 300 nm interposed therebetween may be used. Moreover, it is good also as a multi-finger type FET like the 2nd modification of 1st Embodiment.

  With such a configuration, a PN junction is formed between the p-type semiconductor layer 19 and the n-type AlGaN layer 15 immediately below the p-type semiconductor layer 19. For this reason, even when the gate voltage is 0 V, a depletion layer extends from the p-type semiconductor layer 19 to the AlGaN layer 15 and the GaN layer 14, and the current flowing through the 2DEG layer is cut off, so that a normally-off FET is obtained. In the example described here, the FET has a threshold voltage of about + 1V.

  In order to ensure a bidirectional breakdown voltage, that is, a breakdown voltage between the first ohmic electrode 16 and the gate electrode 18 and a breakdown voltage between the second ohmic electrode 17 and the gate electrode 18, the first ohmic electrode 16 The distance between the first ohmic electrode 17 and the p-type semiconductor layer 19 is preferably equalized. For example, in the case of a bidirectional switch that performs switching of 100 V, the distance between the first ohmic electrode 16 and the p-type semiconductor layer 19 and the distance between the second ohmic electrode 17 and the p-type semiconductor layer 19. May be 10 μm. In order to obtain a bidirectional breakdown voltage, it is preferable that the distance between the first ohmic electrode 16 and the p-type semiconductor layer 19 and the distance between the second ohmic electrode 17 and the p-type semiconductor layer 19 are equal. As long as the breakdown voltage can be secured in both directions, they are not necessarily equal, and the distance may be equal to or longer than the distance that can ensure the required breakdown voltage.

  The control circuit 20 includes a first diode 61 and a second diode 62 connected between the first ohmic electrode 16 and the second ohmic electrode 17. The cathode terminal of the first diode 61 is connected to the first ohmic electrode 16, and the anode terminal is connected to the anode terminal of the second diode 62. The cathode terminal of the second diode 62 is connected to the second ohmic electrode 17.

  A variable power supply 64 and a third diode 63 are connected in series between the gate electrode 18 and the connection node where the anode terminal of the first diode 61 and the anode terminal of the second diode 62 are connected. Is connected. The variable power source 64 is a power source for applying a gate bias, and is an insulated power source that is not grounded. The variable power source 64 and the third diode 63 are connected to the cathode terminal of the third diode 63 and the gate electrode 18, to the anode terminal and the positive terminal of the variable power source 64, and to the negative terminal of the variable power source 64. The connection node is connected. However, the order of the third diode 63 and the variable power source 64 is changed, the positive terminal of the variable power source 64 is connected to the gate electrode 18, the negative terminal of the variable power source is connected to the cathode terminal of the third diode 63, The anode terminal of the third diode 63 may be connected to the connection node.

  The operation of the bidirectional switch according to the second embodiment will be described below. First, in a cut-off state in which no current flows between the first terminal 31 and the second terminal 32, the output of the variable power source 64 is set to, for example, 0 V that is less than the threshold voltage.

  In this state, for example, when the potential of the second terminal 32 is higher than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is +100 V with respect to the potential of the first terminal 31. Most of the voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is applied to the second diode 62, and the first diode 61 has a voltage corresponding to the rising voltage VF of the diode. Applied. This potential is applied to the gate electrode 18 via the variable power source 64 and the third diode 63. In other words, the potential that has dropped by the VF of the third diode 63 is applied to the gate electrode 18. As a result, the same voltage as that of the variable power source 64 is applied to the gate electrode 18. When the threshold voltage of the FET 10 is +1 V, a voltage lower than the threshold voltage is applied between the gate electrode 18 and the first ohmic electrode 16, and therefore flows from the second ohmic electrode 17 to the first ohmic electrode 16. The current can be cut off.

  Conversely, when the potential of the second terminal 32 is lower than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is −100 V with respect to the potential of the first terminal 31, Most of the voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is applied to the first diode 61. Therefore, a potential equal to the potential of the second ohmic electrode 17 is applied to the gate electrode 18 via the variable power source 64 and the third diode 63. For this reason, the output voltage of the variable power source 64 is applied between the gate electrode 18 and the second ohmic electrode 17, and the output voltage of the variable power source 64 becomes lower than the threshold voltage. The current flowing through the second ohmic electrode 17 can be cut off.

  Next, in the case of a conductive state in which a current flows bidirectionally between the first terminal 31 and the second terminal 32, the output of the variable power supply 64 is set to +5 V, for example, higher than the threshold voltage.

  In this state, for example, when the potential of the second terminal 32 is higher than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is +100 V with respect to the potential of the first terminal 31. Most of the voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is applied to the second diode 62, and the first diode 61 has a voltage corresponding to the rising voltage VF of the diode. Applied. This potential is applied to the gate electrode 18 via the variable power source 64 and the third diode 63. In other words, the potential that has dropped by the VF of the third diode 63 is applied to the gate electrode 18. As a result, the same voltage as that of the variable power source 64 is applied to the gate electrode 18. When the threshold voltage of the FET 10 is +1 V, a voltage higher than the threshold voltage is applied between the gate electrode 18 and the first ohmic electrode 16, so that a current is supplied from the second ohmic electrode 17 to the first ohmic electrode 16. Can flow.

  Conversely, when the potential of the second terminal 32 is lower than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is −100 V with respect to the potential of the first terminal 31, Most of the voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is applied to the first diode 61. Therefore, the output voltage of the variable power source 64 is applied to the gate electrode 18 between the second ohmic electrode 17 and the gate electrode 18 via the variable power source 64 and the third diode 63. Since the output voltage of the variable power source 64 is higher than the threshold voltage, a current can flow from the first ohmic electrode 16 to the second ohmic electrode 17.

  In this embodiment, a bidirectional switch using a normally-off type FET has been described. However, a normally-on type FET as shown in the first embodiment may be used. In this case, when the bidirectional current is interrupted, a negative voltage of, for example, −3 V is output in order to apply a voltage lower than the threshold voltage from the variable power source 64. Further, when energizing the bidirectional current, a voltage higher than the threshold voltage may be applied from the variable power source 64, so that, for example, 0 V is output.

  Although the variable power source 64 is connected to the gate electrode 18 via the third diode 63, the variable power source 64 may be directly connected to the gate electrode 18 as long as bidirectional current can be controlled.

  In the second embodiment, an example in which a variable power source is used as a power source for controlling the bidirectional switch has been described. However, a general gate drive circuit may be used instead of the variable power source. However, it is desirable to use an insulated DC voltage conversion converter (DC-DC converter) that is not grounded as a power source connected to the gate drive circuit. Further, the power supply which is not grounded may be a photocoupler or battery capable of outputting voltage.

(Third embodiment)
The third embodiment of the present invention will be described below with reference to the drawings. FIG. 11 shows a circuit configuration of a bidirectional switch according to the third embodiment. In FIG. 11, the same components as those in FIG.

  In the bidirectional switch of this embodiment, a normally-off type FET is used as the FET 10. Although the normally-off type FET is not particularly limited, for example, the one shown in the second embodiment may be used. A multi-finger FET may be used.

  The control circuit 20 includes a first diode 73A in which the gate electrode 18 and the cathode terminal are connected. Between the anode terminal of the first diode 73A and the first ohmic electrode 16, a second diode 73B is connected via a first switch 71A. The second diode 73B has an anode terminal connected to the first switch 71A and a cathode terminal connected to the first ohmic electrode 16. A third diode 73C is connected between the anode terminal of the first diode 73A and the second ohmic electrode 17 with a second switch 71B interposed therebetween. The third diode 73C has an anode terminal connected to the second switch 71B and a cathode terminal connected to the second ohmic electrode 17.

  The anode terminal of the first diode 73A is connected to one terminal of the third switch 71C, and the fourth switch is provided between the other terminal of the third switch 71C and the first ohmic electrode 16. A first power source 72A is connected via 71D. A second power supply 72B is connected between the other terminal of the third switch 71C and the second ohmic electrode 17 with a fifth switch 71E interposed therebetween. The output voltages of the first power supply 72A and the second power supply 72B may be higher than the threshold voltage of the FET 10, for example, 5V.

  The operation of the bidirectional switch according to the third embodiment will be described below. First, in a cut-off state in which no current flows between the first terminal 31 and the second terminal 32, the third switch 71C is turned off, and the first switch 71A and the second switch 71B. Is turned on. The fourth switch 71D operates in the same manner as the first switch of the first embodiment, and the fifth switch 71E operates in the same manner as the second switch of the first embodiment.

  In this state, for example, when the potential of the second terminal 32 is higher than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is +100 V with respect to the potential of the first terminal 31. Most of the voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is applied to the third diode 73C, and a voltage corresponding to the rising voltage VF of the diode is applied to the second diode 73B. Applied. This potential is applied to the gate electrode 18 through the first diode 73A. That is, the potential that has caused a voltage drop by the VF of the first diode 73A is applied to the gate electrode 18. As a result, the gate electrode 18 is given a potential equal to the potential of the first ohmic electrode 16. When the threshold voltage of the FET 10 is +1 V, a voltage lower than the threshold voltage is applied between the gate electrode 18 and the first ohmic electrode 16, and therefore flows from the second ohmic electrode 17 to the first ohmic electrode 16. The current can be cut off.

  Conversely, when the potential of the second terminal 32 is lower than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is −100 V with respect to the potential of the first terminal 31, Most of the voltage between the first ohmic electrode 16 and the second ohmic electrode 17 is applied to the second diode 73B. Therefore, a potential equal to the potential of the second ohmic electrode 17 is applied to the gate electrode 18 via the first diode. For this reason, since the voltage between the gate electrode 18 and the second ohmic electrode 17 is lower than the threshold voltage, the current flowing from the first ohmic electrode 16 to the second ohmic electrode 17 can be cut off.

Next, in a case where a current flows bidirectionally between the first terminal 31 and the second terminal 32, the third switch 71C is turned on, and the first switch 71A and the second switch The switch 71B is turned off. Further, when the potential of the second terminal 32 is higher than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is +100 V with respect to the potential of the first terminal 31, The switch 71D is turned on, and the fifth switch 71E is turned off. In this case, a voltage of 5 V higher than the threshold voltage of the FET 10 is applied between the first ohmic electrode 16 and the gate electrode 18 by the first power supply 72A. A current can be passed through the ohmic electrode 16.

  On the other hand, when the potential of the second terminal 32 is lower than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is −100 V with respect to the potential of the first terminal 31, The switch 71D is turned off, and the fifth switch 71E is turned on. In this case, since a voltage of 5 V higher than the threshold voltage of the FET 10 is applied between the second ohmic electrode 17 and the gate electrode 18 by the second power source 72B, the second ohmic electrode 16 and the second ohmic electrode 16 A current can be passed through the ohmic electrode 17. Further, since the conductive state is changed from the cut-off state, the voltage of the second ohmic electrode 17 is reduced to the on voltage, and the bidirectional switch is turned on.

  FIG. 12 shows a specific example of the control circuit 20. The control circuit 20 illustrated in FIG. 12 includes a first photocoupler 81A, a second switch 71B, a third switch 71C, a fourth switch 71D, and a fifth switch 71E, respectively. The photocoupler 81B, the third photocoupler 81C, the fourth photocoupler 81D, and the fifth photocoupler 81E are used.

  The drive circuit 50 for driving the fourth photocoupler 81D and the fifth photocoupler 81E is the same as that shown in FIG. 3, and the description of the operation is omitted.

  The LED of the third photocoupler 81C is connected to the gate drive signal source 91, and the first photocoupler 81A and the second photocoupler 81B are connected to the gate drive signal source 91 via the inverter 85. As a result, when the gate drive signal source 91 outputs, for example, a 0V signal that turns off the photocoupler, the third photocoupler 81C is turned off, and the first photocoupler 81A and the second photocoupler 81C are turned off. The coupler 81B is turned on. As a result, the bidirectional switch enters a cut-off state in which no current flows between the first terminal 31 and the second terminal 32.

  When the gate drive signal source 91 outputs, for example, a voltage of 5V that turns on the photocoupler, the third photocoupler 81C is turned on, and the first photocoupler 81A and the second photocoupler 81B are turned on. Turns off. In addition, since the fourth photocoupler 81D and the fifth photocoupler 81E are driven by the drive circuit 50, the bidirectional switch has a bidirectional current between the first terminal 31 and the second terminal 32. Will be in a conducting state.

  With such a configuration, it is not necessary to use the variable power sources of the first power source 72A and the second power source 72B. The sixth resistor element R6 and the seventh resistor element R7 are protective resistors for the photocoupler LED. When the forward voltage of the LED is 3.6 V, the forward current is 20 mA, and the drive voltage is 5 V, for example, the resistance value of the sixth resistance element R6 is 70Ω, and the resistance value of the seventh resistance element R7 is What is necessary is just 35 ohms. The drive circuit 50 may be a circuit as shown in FIG.

(One Modification of Third Embodiment)
Hereinafter, a modification of the third embodiment of the present invention will be described. The circuit configuration of the bidirectional switch of this modification is substantially the same as that of the third embodiment shown in FIG. 11 except that the FET 10 is a normally-on type. In this modification, for example, an FET having a threshold voltage of −2V is used, and the output voltages of the first power supply 72A and the second power supply 72B are less than the threshold voltage, for example −5V in order to turn off the bidirectional switch And

  The operation of the bidirectional switch of this modification is as follows. First, when the bidirectional switch is set to the cutoff state, the first switch 71A and the second switch 71B are turned off, and the third switch 71C is turned on. Further, when the potential of the second terminal 32 is higher than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is +100 V with respect to the potential of the first terminal 31, The switch 71D is turned on, and the fifth switch 71E is turned off. In this case, a voltage of −5 V, which is lower than the threshold voltage of the FET 10, is applied between the first ohmic electrode 16 and the gate electrode 18 by the first power source 72 </ b> A. The current flowing through the ohmic electrode 16 can be cut off.

  On the other hand, when the potential of the second terminal 32 is lower than the potential of the first terminal 31, for example, when the potential of the second terminal 32 is −100 V with respect to the potential of the first terminal 31, The switch 71D is turned off, and the fifth switch 71E is turned on. In this case, a voltage of −5 V, which is lower than the threshold voltage of the FET 10, is applied between the second ohmic electrode 17 and the gate electrode 18 by the second power source 72B. The current flowing through the second ohmic electrode 17 can be cut off.

  Next, when the bidirectional switch is turned on, the first switch 71A and the second switch 71B are turned on, and the third switch 71C is turned off. The fourth switch 71D operates in the same manner as the first switch of the first embodiment, and the fifth switch 71E operates in the same manner as the second switch of the first embodiment.

  When the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16, 0 V higher than the threshold voltage is applied between the first ohmic electrode 16 and the gate electrode 18, and the second ohmic electrode Current can flow from 17 to the first ohmic electrode 16. When the potential of the first ohmic electrode 16 is lower than the potential of the second ohmic electrode 17, 0 V higher than the threshold voltage is applied between the second ohmic electrode 17 and the gate electrode 18, A current can flow from the ohmic electrode 16 to the second ohmic electrode 17.

  In order to apply the normally-on type FET to the circuit of FIG. 12, in FIG. 12, an inverter 85 is connected between the gate drive signal source 91 and the resistor element R6, and the gate drive signal source 91 and the resistor element are connected. Connect R7 directly. In addition, the output voltages of the first power supply 72A and the second power supply 72B are set to, for example, −5V which is lower than the threshold voltage.

  With such a configuration, in order to turn off the bidirectional switch, when the gate drive signal becomes, for example, 0V, the first photocoupler 81A and the second photocoupler 81B are turned off, and the third The photocoupler 81C is turned on. Since the fourth photocoupler 81D and the fifth photocoupler 81E operate in the same manner as the circuit of the third embodiment, the potential of the second ohmic electrode 17 is higher than the potential of the first ohmic electrode 16. A voltage less than the threshold voltage of −5 V is applied between the first ohmic electrode 16 and the gate electrode 18, and the current flowing from the second ohmic electrode 17 to the first ohmic electrode 16 can be cut off. When the potential of the second ohmic electrode 17 is lower than the potential of the first ohmic electrode 16, −5 V less than the threshold voltage is applied between the second ohmic electrode 17 and the gate electrode 18, Current flowing from the ohmic electrode 16 to the second ohmic electrode 17 can be cut off.

  Further, in order to turn on the bidirectional switch, when the gate drive signal becomes 5 V, for example, the first photocoupler 81A and the second photocoupler 81B are turned on, and the third photocoupler 81C is turned off. Become. The fourth photocoupler 81D and the fifth photocoupler 81E perform the same operation as in the third embodiment. In this case, the potential of the first ohmic electrode 16 and the second ohmic electrode 17, the first diode 73A, the second diode 73B, and the third diode, as in the cutoff state of the third embodiment. The gate electrode 18 is given a potential equal to the potential of the first ohmic electrode 16 or the second ohmic electrode 17 by the diode 73C. As a result, it is possible to pass a bidirectional current.

  The third switch 71C is connected to the gate electrode 18 via the third diode 73C. However, if the bidirectional current can be controlled, the third switch 71C is directly connected to the gate electrode 18. May be.

  In each of the embodiments and the modifications thereof, an example in which the first ohmic electrode 16 is grounded is shown. However, if all the grounded terminals in the control unit 20 are connected to the first ohmic electrode, the first ohmic electrode is used. The electrode need not be grounded. In addition, the bidirectional power supply 41 in each embodiment includes a circuit configuration that allows a bidirectional current to flow as well as an AC power supply. For example, a circuit including a capacitance and an inductance may be used.

  When the potential of the first ohmic electrode 16 and the potential of the second ohmic electrode 17 are equal, no current flows between the first ohmic electrode 16 and the second ohmic electrode 17. There is no need to apply a bias. Therefore, when the potential of the first ohmic electrode 16 and the potential of the second ohmic electrode 17 are equal, the state of each switch may be any state.

  Further, a GaN substrate, a sapphire substrate, a SiC substrate, or the like may be used instead of the Si substrate shown in the first to third embodiments and the modifications thereof. In this case, for example, a (0001) plane or the like may be used. It is desirable to form an FET on the representative surface. Further, as long as the transistor characteristics of the desired field effect transistor shown here can be realized, the transistor may be formed in a plane orientation with an off-angle from a representative plane such as the (0001) plane. Furthermore, the composition and the laminated structure of the nitride semiconductor may be arbitrarily changed.

  The bidirectional switch according to the present invention controls current from the first ohmic electrode to the second ohmic electrode and from the second ohmic electrode to the first ohmic electrode in one FET, and allows current to flow in at least one direction. In addition, a bidirectional switch that cuts off bidirectional current can be realized, and a bidirectional switch for a power recovery circuit used for a plasma display, a bidirectional switch for a motor drive circuit using a matrix converter circuit, and a bidirectional switch for power control. Useful as such.

1 is a circuit diagram showing a bidirectional switch according to a first embodiment of the present invention. It is a circuit diagram showing an example of composition of a bidirectional switch concerning a 1st embodiment of the present invention. It is a circuit diagram showing an example of composition of a bidirectional switch concerning a 1st embodiment of the present invention. It is a circuit diagram which shows the bidirectional | two-way switch which concerns on the 1st modification of the 1st Embodiment of this invention. It is a top view which shows FET used for the bidirectional | two-way switch which concerns on the 2nd modification of the 1st Embodiment of this invention. FIG. 6 is a cross-sectional view taken along line VI-VI in FIG. 5, showing an FET used for a bidirectional switch according to a second modification of the first embodiment of the present invention. It is a top view which shows the modification of FET used for the bidirectional switch which concerns on the 2nd modification of the 1st Embodiment of this invention. FIG. 8 is a cross-sectional view taken along line VIII-VIII in FIG. 7, showing a modification of the FET used for the bidirectional switch according to the second modification of the first embodiment of the present invention. It is a top view which shows the modification of FET used for the bidirectional switch which concerns on the 2nd modification of the 1st Embodiment of this invention. It is a circuit diagram which shows the bidirectional | two-way switch which concerns on the 2nd Embodiment of this invention. It is a circuit diagram which shows the bidirectional switch which concerns on the 3rd Embodiment of this invention. It is a circuit diagram which shows an example of a structure of the bidirectional | two-way switch which concerns on the 3rd Embodiment of this invention. It is a circuit diagram which shows the bidirectional switch which concerns on a prior art example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Field effect transistor 11 Board | substrate 12 Buffer layer 13 Semiconductor layer 14 GaN layer 15 AlGaN layer 16 1st ohmic electrode 17 2nd ohmic electrode 18 Gate electrode 20 Control circuit 21A 1st switch 21B 2nd switch 22A 1st Power supply 22B Second power supply 22C Third power supply 31 First terminal 32 Second terminal 40 Load circuit 41 Bidirectional power supply 42 Load 50 Drive circuit 51A First photocoupler 51B Second photocoupler 52 Operational amplifier 53A Power supply 53B Power supply 54A first operational amplifier 54B second operational amplifier 55 power supply 61 first diode 62 second diode 63 third diode 64 variable power supply 71A first switch 71B second switch 71C third switch 71D fourth switch Switch 71E fifth switch 72A first Source 72B Second power source 73A First diode 73B Second diode 73C Third diode 81A First photocoupler 81B Second photocoupler 81C Third photocoupler 81D Fourth photocoupler 81E Fifth photocoupler Coupler 85 inverter 91 gate drive signal source 101 unit 111 substrate 112 buffer layer 113 semiconductor layer 114 GaN layer 115 AlGaN layer 116 first ohmic electrode 117 second ohmic electrode 118 gate electrode 121 protective film 122 first insulating film 123 first 2 Insulating film 124 Insulating film 125 Interelectrode wiring 130 Second ohmic electrode pad 131 First ohmic electrode pad 132 Gate electrode pad 140 Active region 141 High resistance region 142 Insulating film 156 First ohmic electrode 157 Second Ohmi Click electrode 158 gate electrode 160 second ohmic electrode pads 161 first ohmic electrode pad 162 a gate electrode pad

Claims (25)

A bidirectional switch that controls a conductive state in which a current flows in at least one direction between a first terminal and a second terminal and a cut-off state in which no current flows;
A first ohmic electrode and a second ohmic electrode, one of which is a source electrode and the other is a drain electrode, and a gate electrode formed between the first ohmic electrode and the second ohmic electrode, A field effect transistor in which one ohmic electrode is connected to the first terminal and the second ohmic electrode is connected to the second terminal;
A control circuit that controls the conduction state and the cutoff state by applying a bias voltage to the gate electrode;
The control circuit applies the bias voltage based on the potential of the first ohmic electrode when the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, When the potential of the ohmic electrode is lower than the potential of the first ohmic electrode, the bias voltage is applied with reference to the potential of the second ohmic electrode. The control circuit includes:
Having a first power source;
When the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, the first power source is electrically connected between the first ohmic electrode and the gate electrode. Applying the bias voltage to the gate electrode;
When the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, the first power source is electrically connected between the second ohmic electrode and the gate electrode. The bidirectional switch according to claim 1, wherein the bias voltage is applied to the gate electrode. The control circuit includes a first switch connected between the first power source and the first ohmic electrode, and a first switch connected between the first power source and the second ohmic electrode. 2 switches, and a drive circuit for driving the first switch and the second switch,
The drive circuit switches between the on state and the off state of the first switch and the second switch based on a result of comparing the potential of the first ohmic electrode and the potential of the second ohmic electrode. The bidirectional switch according to claim 2. The control circuit includes:
A first power source and a second power source;
When the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, the first power source is electrically connected between the first ohmic electrode and the gate electrode. Applying the bias voltage to the gate electrode;
When the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, the second power source is electrically connected between the second ohmic electrode and the gate electrode. The bidirectional switch according to claim 1, wherein the bias voltage is applied to the gate electrode.   The first power supply and the second power supply output a voltage higher than the threshold voltage of the field effect transistor when in the conductive state, and the threshold voltage of the field effect transistor when in the cutoff state. The bidirectional switch according to claim 4, wherein a lower voltage is output.   The first power supply outputs a voltage higher than the threshold voltage of the field effect transistor, the second power supply outputs a voltage lower than the threshold voltage of the field effect transistor, and the first terminal and the second terminal 5. The bidirectional switch according to claim 4, wherein a current in one direction is passed between and a current in the opposite direction is cut off. The control circuit includes: a first switch connected between the first power supply and the gate electrode; a second switch connected between the second power supply and the gate electrode; A drive circuit for driving the first switch and the second switch;
The drive circuit switches between the on state and the off state of the first switch and the second switch based on a result of comparing the potential of the first ohmic electrode and the potential of the second ohmic electrode. The bidirectional switch according to claim 4 or 5, characterized by the above-mentioned. The first switch and the second switch are a first photocoupler and a second photocoupler, respectively.
The drive circuit includes a differential amplifier in which a voltage corresponding to a voltage applied between the first ohmic electrode and the second ohmic electrode is input to an input terminal;
The bidirectional switch according to claim 7, wherein the differential amplifier drives a light emitting element of the first photocoupler and a light emitting element of the second photocoupler. The first switch and the second switch are a first photocoupler and a second photocoupler, respectively.
In the driving circuit, a voltage corresponding to a voltage applied between the first ohmic electrode and the second ohmic electrode is input to a non-inverting input terminal, and an inverting input terminal is connected to the first ohmic electrode. A voltage corresponding to a voltage applied between the connected first differential amplifier and the first ohmic electrode and the second ohmic electrode is input to the inverting input terminal, and the non-inverting input terminal is A second differential amplifier connected to the first ohmic electrode;
The first differential amplifier drives a light emitting element of the first photocoupler,
The bidirectional switch according to claim 7, wherein the second differential amplifier drives a light emitting element of the second photocoupler. The control circuit includes:
A first diode and a second diode having anode terminals connected to each other and electrically connected between the first ohmic electrode and the second ohmic electrode;
A first power source connected to a connection node where the anode terminal of the first diode and the anode terminal of the second diode are connected to each other;
The first power supply outputs a voltage higher than the threshold voltage of the field effect transistor in the conductive state, and outputs a voltage lower than the threshold voltage of the field effect transistor in the cutoff state. The bidirectional switch according to claim 1, wherein the bidirectional switch is output.   The bidirectional switch according to claim 10, wherein the control circuit includes a third diode electrically connected between the connection node and the gate electrode. The field effect transistor is a normally-on type,
The control circuit includes:
A first diode in which the gate electrode and the cathode terminal are connected;
A second diode connected between the first ohmic electrode and the anode terminal of the first diode via a first switch so that a cathode terminal is on the first ohmic electrode side; ,
A third diode connected between the second ohmic electrode and the anode terminal of the first diode via a second switch so that a cathode terminal is on the second ohmic electrode side; ,
A third switch in which an anode terminal of the first diode and a first terminal are connected;
A first power source connected between a second terminal of the third switch and the first ohmic electrode with a fourth switch interposed therebetween and outputting a voltage lower than a threshold voltage of the field effect transistor; When,
A second switch is connected between the second terminal of the third switch and the second ohmic electrode with a fifth switch interposed therebetween, and outputs a voltage lower than a threshold voltage of the field effect transistor. Power supply,
In the conductive state, the third switch is turned off, the first switch and the second switch are turned on,
In the cut-off state, when the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, the third switch and the fourth switch are turned on and the first switch When the second switch and the fifth switch are turned off, and the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, the third switch and the fifth switch 2. The bidirectional switch according to claim 1, wherein a switch is turned on and the first switch, the second switch, and the fourth switch are turned off. The control circuit includes a first drive circuit that drives the fourth switch and the fifth switch;
A second drive circuit for driving the first switch, the second switch, and the third switch;
The first drive circuit compares the potential of the first ohmic electrode with the potential of the second ohmic electrode, and based on the comparison result, the fourth switch and the fifth switch are turned on. And toggle between off and
The second drive circuit turns on the third switch and turns off the first switch and the second switch in the conductive state, and turns off the third switch in the cut-off state. The bidirectional switch according to claim 12, wherein the switch is turned off and the first switch and the second switch are turned on. The field effect transistor is a normally-off type,
The control circuit includes:
A first diode in which the gate electrode and the cathode terminal are connected;
A second diode connected between the first ohmic electrode and the anode terminal of the first diode via a first switch so that a cathode terminal is on the first ohmic electrode side; ,
A third diode connected between the second ohmic electrode and the anode terminal of the first diode via a second switch so that a cathode terminal is on the second ohmic electrode side; ,
A third switch in which an anode terminal of the first diode and a first terminal are connected;
A first power source connected between a second terminal of the third switch and the first ohmic electrode with a fourth switch interposed therebetween and outputting a voltage higher than a threshold voltage of the field effect transistor; When,
A second switch is connected between the second terminal of the third switch and the second ohmic electrode with a fifth switch interposed therebetween, and outputs a voltage higher than a threshold voltage of the field effect transistor. Power supply,
In the shut-off state, the third switch is turned off, the first switch and the second switch are turned on,
In the conductive state, when the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, the third switch and the fourth switch are turned on and the first switch When the second switch and the fifth switch are turned off, and the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode, the third switch and the fifth switch 2. The bidirectional switch according to claim 1, wherein a switch is turned on and the first switch, the second switch, and the fourth switch are turned off. The control circuit includes a first drive circuit that drives the fourth switch and the fifth switch;
A second drive circuit for driving the first switch, the second switch, and the third switch;
The first driving circuit determines whether the fourth switch and the fifth switch are on or off based on a result of comparing the potential of the first ohmic electrode and the potential of the second ohmic electrode. Switch
The second drive circuit turns on the first switch and turns off the second switch and the third switch in the cut-off state, and turns off the first switch in the conduction state. The bidirectional switch according to claim 14, wherein the switch is turned off and the second switch and the third switch are turned on. The field effect transistor is
Having a semiconductor layer formed on a substrate;
The first ohmic electrode and the second ohmic electrode are formed on the semiconductor layer at a distance from each other,
The gate electrode is formed on the semiconductor layer such that a distance between the first ohmic electrode and the gate electrode is equal to a distance between the second ohmic electrode and the gate electrode. The bidirectional switch according to claim 1, wherein the bidirectional switch is formed. The field effect transistor has an insulating film formed between the first ohmic electrode and the gate electrode,
The bidirectional switch according to claim 1, wherein a breakdown voltage of the insulating film is higher than a breakdown voltage between the first ohmic electrode and the gate electrode.   The bidirectional switch according to claim 17, wherein the insulating film includes a first insulating film and a second insulating film that are sequentially stacked.   The bidirectional switch according to claim 18, wherein the first insulating film is made of silicon oxide containing phosphorus, polyimide, or benzocyclobutene.   The bidirectional switch according to claim 18 or 19, wherein the second insulating film is made of silicon nitride or silicon oxide. The substrate is conductive;
The field effect transistor is
A back electrode formed on the back surface of the substrate;
21. The wiring according to claim 16, further comprising a wiring penetrating the semiconductor layer and electrically connecting the first ohmic electrode and the back electrode through the conductive substrate. The bidirectional switch according to item 1. The field effect transistor is
A second ohmic electrode pad connected to the second ohmic electrode and a gate electrode pad connected to the gate electrode;
The bidirectional switch according to claim 21, wherein at least one of the second ohmic electrode pad and the gate electrode pad is formed on a region of the semiconductor layer having a high resistance. The field effect transistor has a p-type semiconductor layer formed between the gate electrode and the semiconductor layer,
The semiconductor layer includes a first layer and a second layer sequentially formed from the lower side,
The bidirectional switch according to any one of claims 16 to 22, wherein the second layer is made of an n-type semiconductor.   The bidirectional switch according to any one of claims 16 to 23, wherein the semiconductor layer is made of a nitride semiconductor or silicon carbide. A driving method for driving a field effect transistor as a bidirectional switch having a conduction state in which current flows in at least one direction and a cutoff state in which current does not flow between a first ohmic electrode and a second ohmic electrode,
Comparing the potential of the second ohmic electrode with the potential of the first ohmic electrode;
When the potential of the second ohmic electrode is higher than the potential of the first ohmic electrode, a bias voltage is applied to the gate electrode of the field effect transistor based on the potential of the first ohmic electrode, Applying a bias voltage to the gate electrode with respect to the potential of the second ohmic electrode when the potential of the second ohmic electrode is lower than the potential of the first ohmic electrode;
When the conductive state is set, the bias voltage is set to a voltage higher than the threshold voltage of the field effect transistor, and when the cutoff state is set, the bias voltage is set to a voltage lower than the threshold voltage. A method for driving a bidirectional switch characterized by the above.

Images