JPS582459B2 - reverse conducting thyristor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to reverse conducting thyristors, and more particularly to reverse conducting thyristors having relatively large current carrying capacity and improved switching characteristics.

When an on-current is passed through a semiconductor rectifier (diode) consisting of two layers of different conductivity types, and the forward current is forcibly interrupted by an external circuit (hereinafter referred to as commutation), a relatively short period of time is generated in the reverse direction. It is well known that pulse-width currents flow.

This current is due to the reverse flow of minority carriers accumulated in the semiconductor and is called a recovery current.

FIG. 1 is an explanatory diagram showing current and voltage waveforms of a diode during commutation.

During time T1, the forward current source flows, and during commutation (time T2
The current waveform in which the recovery current [present] flows during the period) is shown by a solid line.

On the other hand, the reverse voltage (C) rises quite steeply near the time tn when the recovery current (B) reaches its peak, and peaks at V
After reaching R(p), it generally falls again and reaches the steady value VR, which is shown by the dotted line.

Now, the rising slope dv/dt and the peak value VR (p) of the reverse voltage (C) are the decreasing rate di/dt of the forward current at the start of commutation and the recovery charge QR (the area indicated by diagonal lines in the figure). (equivalent to ) increases significantly, but another characteristic is that it has a similar strong dependence on the rate of decrease of the recovery current (di/dt)r.

Here, it is necessary to consider the relationship between the recovery current (E3) and accumulated carriers.

As a specific example, FIG. 2 shows temporal changes in the accumulated carrier concentration distribution in the n region (high resistivity layer), which determines the recovery current waveform, for a diode with a p+n junction structure.

Each curve p(x, tc), p(x, t1)..., p(x
, tn) is a minority carrier (
(holes) concentration.

p(x, tc) is the hole concentration at time tc when the current flowing through the diode changes from the forward direction to the reverse direction.

After time tc, holes are no longer injected from the p+ side, so holes near the pn junction are diffused and returned to the p+ side at the pn junction.

In this way, the minority carrier distribution changes over time with p(x
, t1), p(x, t2), p(x, t3),
At time tp, the distribution is approximately p(x, t4).

Furthermore, at time t5,...,tn, p(x, t5),...
, p(x, tn), and the recovered charge gradually decreases.

That is, since the recovery current is based on the diffusion phenomenon of minority carriers in the semiconductor, it can be seen that QR depends on the minority carrier lifetime.

Now, in the practical use of diodes, as mentioned above, steep reverse voltage pulses generated during commutation often cause undesirable phenomena.

For example, if the peak value VR(p) exceeds the withstand voltage of the diode, a decisive problem arises that the device will be permanently destroyed.

A more common and important example is that when a diode and a thyristor are used in antiparallel, the reverse voltage generated in the diode is forward to the thyristor, so dv
If /dt or VR(p) exceeds a certain limit value, Cylisk will fail to turn on, and in some cases, Cylisk will suffer permanent damage.

In particular, for reverse conduction silices, which have recently been put into practical use in applications such as inverters and choppers, the reverse voltage pulse generated during commutation on the diode side has a large effect on device performance.

In order to lower dv/dt and VR(p), it is effective to some extent to reduce Qr by keeping di/dt below a certain value or by shortening the lifetime of minority carriers in the semiconductor, but the former depends on the circuit conditions. However, there are limitations to countermeasures because the latter leads to deterioration in element performance such as a decrease in breakdown voltage and an increase in forward voltage drop.

In addition, there is a method in which the element is shunted with a series connection circuit (hereinafter referred to as CR absorber) of a capacitor and a resistor in order to absorb steep voltage pulses.

For example, FIG. 3 shows an example of a reverse conducting thyristor using a CR absorber.

However, since the introduction of the capacitor C increases the di/dt when the reverse conduction thyristor is turned on, C needs to be set below a certain level and R needs to be set above a certain level, and there is a limit to the effectiveness of the absorber.

Therefore, the main purpose of the present invention is to eliminate the above-mentioned drawbacks of the conventional method and reduce the steep voltage pulses dv/dt, VR generated during commutation without impairing other characteristics of the diode.
An object of the present invention is to provide a reverse conduction thyristor having an element structure capable of lowering (p).

Other objects and features of the invention will become more apparent from the following detailed description with reference to the drawings.

Briefly speaking, this invention provides a semiconductor device having a diode region made of a pn2 layer, with a portion to which an electrode is connected and a number of portions to which the electrode is not connected on the surface of the high concentration layer, and the latter is dotted within the former. By adopting a structure arranged so that the recovery current decreases [(d
The objective is to reduce dv/dt and VR(p) by reducing i/dt)r).

FIG. 4 is an embodiment of the present invention, and is an explanatory diagram of the element structure and operation of a reverse conduction circuit.

FIG. 4a is a structural sectional view of a reverse conducting thyristor wafer.

The conductivity type of this wafer 100 is n-type, and the specific resistance is 35 to 50.
Floating zone in Ω-cm
It is formed using a Si wafer as a substrate using a known diffusion technique.

The reverse conduction thyristor wafer 100 includes a p-type emitter layer 101, an n-type high impurity concentration base layer 102, an n-type base layer 103, a p-type base layer 104, and an n-type emitter layer 105.
The thyrisk part TH consists of four pnpn layers, an n-type base layer 102, an n-type base layer 103, and a p-type base layer 104.
It has a diode portion D made of a pn2 layer.

Here, the n-type high impurity concentration base layer 102 is formed using the epitaxial technique, the p-type emitter layer 101, the p-type base layer 104
, and n-type emitter layer 105 are each formed using a diffusion technique, and have a significantly higher impurity concentration than n-type base layer 103.

The impurity concentration, diffusion depth, etc. of the reverse conduction thyristor wafer formed in this manner are well known in the art.

For example, "Electronic Outlook Special Issue, Latest Thyristor Utilization Technology" (Seibundo Shinkosha, published June 15, 1948)
It is described on pages 28, 29, and 33 to 36 of .

This invention is a reverse conduction thyristor manufactured by using the above-mentioned known reverse conduction thyristor wafer and forming electrodes as described below. FIG. 3 is a cross-sectional view of the structure of a reverse conduction thyristor.

An anode electrode 112 made of molybdenum (Mo) is alloy-connected to a first surface 111 of the reverse conduction thyristor wafer 100 using aluminum (Al) foil as a solder, and a gate electrode 114 and a cathode electrode made of Al are connected to a second surface 113 of the reverse conduction thyristor wafer 100. 115 are formed by a vapor deposition method.

Furthermore, an anode electrode block 116 made of copper (Cu)
and the cathode electrode block 117 are the anode electrode 112
and the cathode electrode 115, respectively, and current is applied thereto.

For example, when using a Si wafer with a diameter of about 40 mm,
Approximately 400A on the Sirisk side and 200A on the diode side
A reverse conduction risk with an average current capacity of about

A notable feature based on this invention is that the cathode electrode 11
5 is formed to have a portion 118 that is not connected to a part of the second surface 113, that is, the surface of the p-type base layer 104.

A relatively large number of these portions 118 are provided in the diode portion D, and for ease of understanding, a plan view of the structure viewed from above is shown in FIG. 4C.

This electrode structure can be realized relatively easily by conventional selective deposition methods using mask techniques.

If the portion 118 is a small circle 119, the diameter d, the small circle 11
The number N of small circles 119 arranged at intervals of 9 is an important design element of the inventive structural element.

In order to see the effects of this invention in detail, reference is made to FIG. 4d, which shows an enlarged view of one component 130 of FIG. 4b.

Region E has electrodes 112 and 11 at opposing positions on both surfaces.
5 is a connected diode, and region F is the first surface 1
This is a diode in which an electrode 112 is connected only to the electrode 11.

When the diode part D is in the forward direction, that is, when a voltage is applied with a positive polarity to the terminal K and a negative polarity to the terminal A, the voltage changes from the p-type base layer (PB layer) 104 to the N-type base layer (NB layer) at the pn junction J2.
A large number of holes are injected into the electrode 103, and the holes travel toward the electrode 112 due to a diffusion phenomenon.

However, since the electrode 113 is not connected to the portion 118, there is a lateral resistance R of the pB layer 104, and the pn junction
Hole injection in the −2 portion does not occur much.

As mentioned above, there is a difference in the hole injection level between region E and region F, but the n-type base region 1-2-
Holes diffused from the n-type base region of region E are accumulated in 3-4 and contribute to the current propagation from terminal K to terminal A in no small measure.

The accumulation state of holes (majority carriers) is as shown in FIGS. 5a and 5b, respectively, when viewed in two directions, the direction of the symbol X and the direction of the arrow 2.

This is a distribution similar to p(x, to) shown in FIG.

Next, the forward current is cut off by an external circuit, and the cathode K
When an attempt is made to reverse the polarity of the voltage between the anode A and the anode A, the carriers accumulated in the NB layer 103 flow into the PB layer 104 through the pn junction J2 and become a recovery current, as described in FIG.

However, since the minority Gallium accumulated in the n-type base region 1-2-3-4 has a lateral resistance R, it becomes difficult to flow vertically to the pn junction J2 within the region F, and the region It has no choice but to flow to F with a lateral component.

That is, since the diffusion distance becomes longer, this phenomenon appears as an effect of lengthening the accumulation time.

In other words, the n-type base region 1-2-3-4 has the effect of adding a minority carrier storage location to the conventional diode (region E) and equivalently serves to increase the base thickness.

At this time, the speed of decrease W (di) of the recovery current (B) shown in FIG.
/dt)r becomes smaller and the recovery time T2 becomes longer.

Therefore, the reverse voltage (C) dv/dt and VR(p) can be kept to extremely low values.

It is preferable that the distance d of the portion 118 is set to be greater than or equal to the minority carrier diffusion length of the n-type base region.

In this embodiment, the diameter d of the portion 118 is 0.4 mm, and the distance between the portions 1 = 1 mm. Care must be taken as this will lead to an increase in voltage drop.

Preferably, the area occupied by portion 118 is between 5% and 25% of the total area of diode D.

In order to enhance the effect of the present invention and maintain other characteristics, it is preferable to suppress the diameter d to several times the diffusion length, and more preferably, to further increase the lateral resistance R, as shown in FIG. It is preferable to reduce the thickness of the p-type base layer 104 by removing the surface portion of the portion 118 by etching or the like, as shown in FIG.

Furthermore, if an electrically insulating layer 121 such as silicon oxide (SiO2) is provided on the surface of the portion 118 as shown in FIG. The advantage is that you don't need to be there.

In order to make the effects of the present invention more concrete, the results of a characteristic comparison test between the reverse conduction thyristor obtained in the example of the present invention and the reverse conduction thyristor obtained by the conventional method are shown in Figures 7a and b.
Shown below.

Specifically, this characteristic comparison test, as shown in Figure 3,
When the CR absorber brown wire is connected to the two reverse conduction thyristors to be tested and a pulse current of peak value current ID 1000 A and width 60 μs is passed through the diode section of each reverse conduction thyristor, the rise of the reverse voltage that occurs during commutation. Slope dv/dt and rate of decrease in reverse current during commutation di
/dt and the peak value VR(p
) and the rate of decrease di/dt of the forward current during commutation are shown in FIGS. 7a and 7b, respectively.

Other conditions at this time, namely, the steady value of the reverse voltage V
R is 400V, CR absorber capacitance C is 0.5μ
F and resistance value R were measured at 10Ω.

In FIGS. 7a and 7b, curve 1 is the measurement result of the conventional reverse conduction thyristor, and curve 2 is the measurement result of the reverse conduction thyristor according to the present invention.

In this way, as a result of comparing the conventional element and the element according to the present invention under the same conditions, the dv/dt (V/μS) and VR(p) (V) at the same di/dt (A/μs)
For example, when di/dt is 100A/μs, dv/dt is 450V/μs in a conventional reverse conducting thyristor.
S is 250V/μs for the reverse conduction thyristor of the present invention, VR(p) is 1120V for the conventional reverse conduction thyristor, and 700V for the reverse conduction thyristor of the present invention (see Figures 7a and b). From Figure 7a and b, indirectly the rate of decrease (di/dt)r of the recovery current decreases, dv/dt
It can be seen that VR(p) and VR(p) decreased.

Therefore, the reverse conducting thyristor according to the present invention has a higher dv/d
It is clear that both t and VR(p) are significantly improved over conventional reverse conduction thyristors, and the effects of the present invention can be clearly recognized.

In the embodiments of the invention described above, the portions 118 are formed in small circular shapes and are scattered in a pattern, but are not limited to this pattern.
It goes without saying that the effects of the present invention can be obtained even if a ring-like or finger-like shape is used.

Although the above embodiments of the present invention are applied to reverse conduction thyristors, the present invention is not limited to this, and can be applied to semiconductor elements having a rectifying function region made of a pn2 layer, such as diodes, to achieve sufficient effects.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing current and voltage waveforms of a diode during commutation. FIG. 2 is a diagram for explaining the temporal change in the accumulated carrier concentration distribution in the n region of a p+n junction structure diode. FIG. 3 is an explanatory diagram of a reverse conduction thyristor in which a CR absorber is connected. FIG. 4 is an explanatory diagram of the structure and operating mechanism of a reverse conduction thyristor according to an embodiment of the present invention. FIG. 5 is an explanatory diagram of the accumulated carrier concentration distribution in the base. FIG. 6 is an explanatory diagram showing a part of an element structure according to another embodiment of the present invention. FIG. 7 is an explanatory diagram comparing the characteristics of a reverse conduction thyristor according to an embodiment of the present invention and a reverse conduction thyristor having a conventional structure. In the figure, 100 is a reverse conduction thyristor wafer, 101
102 is a p-type emitter layer, 102 is an n-type high impurity concentration base layer, 103 is an n-type base layer, 104 is a p-type base layer, 10
5 is an n-type emitter layer, 118 is an electrode non-connection part, and 115 is a cathode electrode.

Claims (1)

[Scope of Claims] 1. A device comprising a thyristor region and a rectifier region including two layers of at least first and second conductivity types having different impurity concentrations, which are incorporated in antiparallel in a first semiconductor substrate. In the reverse conductivity thyristor, a first electrode is connected to a first surface of the first conductivity type semiconductor layer having a relatively low impurity concentration among the two layers, and a first electrode is connected to a first surface of the first conductivity type semiconductor layer having a relatively low impurity concentration among the two layers; A second surface of the second conductivity type semiconductor layer having an impurity concentration is provided with a first portion to which the second electrode is connected and a second portion to which the second electrode is not connected, and the second portion is connected to the second surface. The reverse conduction thyristor is arranged to be interspersed in the first portion so that lateral resistance exists in a portion of the second conductivity type semiconductor layer where the second portion is provided.