JPH0214574A - Bi-directional thyristor - Google Patents

Abstract

PURPOSE:To attain high sensitivity and to improve a temperature characteristic of a gate trigger current by a method wherein the third semiconductor region had an invasion in a state of cutting deep into the fifth semiconductor layer, the invasion part is exposed to the surface of a semiconductor base body and a gate electrode is connected to the third semiconductor region at a recessed part of the invasion part. CONSTITUTION:An n4 region (the fifth semiconductor region) has a cut-into part and here an invasion part 11 of a p2 region (the third semiconductor region) is arranged. The invasion part 11 is the part masked when the n4 region is formed by impurity diffusion vertically dividing the n4 region. A narrow region 2 between an n2 region and the n4 region is utilized as a current path of a reactive current component IGTb of a gate current; further, the invasion part 11 is also utilized as its part so that an inlet of the invasion part 11 is arranged near the right end of the narrow region 2. A recess of the invasion part 11 consisting of the p2 region, that is, a tip part 11a is not coated with an insulating film 3, but the part excepting this is coated with the insulating film 3. A gate electrode G is arranged so as to cover almost the whole of the n4 region and the invasion part 11 while the invasion part 11 is connected only to the tip part 11a.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a bidirectional thyristor generally called a triac, and more specifically to a bidirectional thyristor that has achieved high levels of both high sensitivity and temperature characteristics. Regarding thyristors.

[Prior Art and Problems to be Solved by the Invention] Bidirectional thyristors (bidirectional self-progression three-terminal thyristors), or triacs, are widely used in electronic circuits such as motor control circuits as switches for controlling alternating current. Figure 8-1
The conventional bidirectional thyristor shown in Figure 1 is of the first conductivity type (n
a first semiconductor region n1 (hereinafter simply referred to as n1 region) of a type (type) and a second semiconductor region P1 of a second conductivity type (p type) that is in contact with one side (lower side) of this n1 region. (hereinafter simply referred to as the P1 region) and a third semiconductor region P2 of the second conductivity type (p type) that is in contact with the other side (upper side) of the n1 region (hereinafter simply referred to as the p2 region). A silicon semiconductor substrate 1 is provided. Note that the n1 region is made of an n-type silicon semiconductor substrate that is the starting material. Both the P1 region and the P2 region are formed by impurity diffusion, and the p2 region located on the upper surface side of the base 1 is surrounded by the n1 region. There is. Inside the P2 region, a fourth semiconductor region n (hereinafter referred to as
(simply referred to as the n2 region) and the fifth semiconductor region n (hereinafter referred to as
(simply referred to as an n4 region) is formed by impurity diffusion.

In the region 91, a sixth semiconductor region n3 (hereinafter simply referred to as an n3 region) of the first conductivity type (n type) is formed by impurity diffusion. A seventh semiconductor region P3 (hereinafter simply referred to as region 23) of the second conductivity type (p type) is formed in a side region of the semiconductor substrate 1 by impurity diffusion. This P3 region is formed so as to be continuous with the 21 region by diffusion of impurities from the top and bottom surfaces of the semiconductor wafer before element layering.As is clear from FIG. 9, the n-type region and the P2 region The planar shape is approximately rectangular, and the n2 area is approximately 17 out of 92 areas.
It is formed to have an area of about 3. The n region is about 178 facet seed regions out of 22 regions. Figure 8~
The bidirectional thyristor shown in FIG. 11 has a corner gate structure in which the n4 region is placed close to one corner of the 22 regions. Note that the patterns of the n2 region and the n4 region are determined so that the narrow region 2 of the p region is generated between the n region and the n4 region.

As is clear from FIG. 11, the n3 region is formed in a part of the 21 region, and is arranged so as to have a portion that overlaps with the n4g4 region when viewed in plan.

As is clear from FIGS. 8 and 10, the first main electrode T1 made of an aluminum electrode formed by vacuum evaporation is as follows.
It is arranged so as to be in contact with most of the P region and the n2 region. Tt (titanium)-N formed by vacuum evaporation method etc.
Tit! The second earth dragon consisting of f! T2 is arranged so as to be in contact with the p1 region and the n3 region. A gate electrode G made of an aluminum electrode formed by vacuum evaporation is arranged so as to be in contact with most of the n region and a part of the 22-head region. The first main electrode T1 and the gate electrode G are in contact with each region through the opening of the 5102-based insulation #3. Note that the first main electrode T is connected to the P region and n2
The gate electrode G is formed so as to partially short-circuit the n4 region and the p2 region.

The bidirectional thyristor has a first main electrode T1 and a second main electrode T.
2 and the voltage of the gate electrode G with respect to the first main electrode T1, that is, the gate voltage V. The switching operation can be performed regardless of the change in the positive or negative polarity of . In the first mode in which the potential of the second main electrode T2 with respect to the first main electrode T1 is positive and the potential of the gate electrode G with respect to the first main electrode T1 is positive, the p
The first region consists of a 1 region, an n region, a P region, and an n2 region.
The cyrisk part 4 of is conductive.

The second earth dragon fI for the first earth dragon iT1! positive at the potential of T2 and the gate electrode G relative to the first main electrode T1
In the second mode in which the potential of is negative, the first thyrisk portion 4 is conductive as in the first mode.

The second earth dragon' for the first main electrode T! ! f! In the third mode in which the potential of T2 is negative and the potential of the gate electrode G with respect to the first main electrode T1 is negative, the second thyristor section consisting of the n region, the P region, the n region, and the p2 region 5 is conductive.

In the fourth mode in which the potential of the second main electrode T2 with respect to the first main electrode T1 is negative and the potential of the gate electrode G with respect to the first main electrode T is positive, it is similar to the third mode. The second thyristor section becomes conductive.

As is clear from FIG. 8, the first and second thyristor sections 4 are equivalent to two thyristors connected in antiparallel. The part 6 on the right side of FIG. 8 consisting of the nfi\ area, the p area, the n area, the P1 area, and the n3 area is as follows:
This is a portion used to conduct the first and second thyristor portions 4.5.

By the way, one of the important issues in bidirectional thyristors is to increase the sensitivity, that is, how to make the first thyristor part 4 or the second thyristor part 5 conductive with a small gate trigger current.

However, in the conventional bidirectional circuit shown in FIGS. □ was not very small, making it difficult to reduce sensitivity. That is, the gate trigger current component T is the active current component I that contributes to the conduction of the first thyristor section 4 or the second thyristor section 5, that is, the trigger, and the reactive current component I that does not contribute to the trigger. Tb, and is expressed as the sum of the above two current components. Therefore, in order to achieve high sensitivity, it is necessary to reduce the reactive current component (2) that does not contribute to the trigger. For example, the first thyristor section 4 and the second thyristor section 5 The conduction is between n1 region and n4
The product of the resistance (base resistance) R of the p2 region sandwiched by the p2 region and the current flowing there is 22B. The region begins by injecting electrons.

Therefore, in the thyristor having the structure shown in FIGS. 8 to 11, n
The gate trigger current flowing through the p2 region in the lower example of the region is the effective current component, and the game GT
GTa Gate current flowing in the current path shown by arrow 7, which consists of 22 dark areas, between the connection part A between the upper electrode G and the p2 region, and the connection part B between the first main pole T and the 92 region. becomes the reactive current T flow component I that does not contribute to the above trigger, so y! of this yi path! ! LHI
Reactive current component I by increasing resistance Tb value R. It is desirable to reduce □, but in the bidirectional thyristor with the structure shown in FIGS. 8 to 11, the resistance value of the gate current path shown by arrow 7 is increased while electrically separating the n2 region and the 04 region. That was difficult.

In order to solve this problem, as shown in FIGS. 12 to 14, the p2 region is left in the form of an island within the n region, and the gate electrode G is not connected to the island-like portion 8. Direct thyristors have been proposed. In FIG. 13, the thickness direction is significantly enlarged, so the 14 region is shown thicker, but this thickness is significantly smaller than the lateral length of the n4 region, so the gate current indicated by arrow 9 The resistance value from one end A of the path to the other end iB is almost determined by the lateral resistance value of the p2 region below the n region. The impurity concentration of the p region below the n4 region is significantly lower than the impurity concentration of the surface of the 22 region. Concentration is 1015-1016 cm
- Designed to 3 orders of magnitude.

With the structure shown in FIGS. 12 and 13, only the path indicated by the arrow 9 including the island-shaped P2 region carries the gate trigger current I. □
becomes a current path. Therefore, all of the gate trigger current I becomes the effective current component I, and GT
GTa reactive current component is almost eliminated, therefore, Tb is gate-triggered current component than the structure of FIGS. 8 to 11.

It is possible to reduce □. Since the impurity concentration is low in the lower portion of the n4 region in the p2 region, the temperature dependence of the resistance in this portion (resistivity change due to temperature change) becomes large. This is clear from FIG. 15, which shows the relationship between the impurity concentration and resistivity of the semiconductor. As a result, the temperature dependence (rate of change in current value due to temperature change) of the gate trigger current I6□ increases, making it difficult to obtain good temperature characteristics of turn-on characteristics. In the case of FIGS. 8 to 11, GTb is equal to the amount of reactive current component.

Although it is disadvantageous in terms of high sensitivity, reactive current component ■. The temperature dependence of resistance is small near the surface of the p2 region, which becomes the current path of □, and the gate trigger current flow is smaller than that in FIGS. 12 to 14. The temperature dependence of 1 is small.Also, the second earth dragon [! The first and second T2 has a positive potential.
In this mode, a depletion layer extending from the pn junction between the 22 region and the 01 region spreads in the lower part of the n4 region of the 92 region. This depletion layer acts to narrow the gate current path shown by arrow 9 and increase the resistance value R1. In other words, the phenomenon in which the gate trigger current I6□ depends on the main electrode voltage V becomes stronger, which is not desirable. In addition, in a bidirectional thyristor, a positive gate voltage V is applied to the gate ring rjIG with respect to the first main electrode. In the first and fourth modes of applying , forward biasing the pn junction between the n2 region and p2 region triggers turning on the first thyristor section or the second thyristor section. Here, in the bidirectional thyristors shown in FIGS. 12 to 14, the resistance value RPB of the gate current path is
and gate electrician. The gate trigger current in the first mode is lower than that of the bidirectional thyristors shown in FIGS. 8 to 11 because the bias voltage corresponding to the product of Engineering.

, increases.

Therefore, the present invention solves the above problems and provides a gate trigger current engineer. It is an object of the present invention to provide a bidirectional SI risk that can achieve high sensitivity by reducing □ and also prevent deterioration of other turn-on characteristics accompanying this high sensitivity.

[Means for Solving the Problems] To achieve the above object, the present invention will be described with reference to the reference numerals in the drawings showing the embodiments. a second semiconductor region P1 of the second conductivity type adjacent to one side of the semiconductor region n1; and a third semiconductor region P2 of the second conductivity type adjacent to the other side of the first semiconductor region n1. The third semiconductor region p2 has a fourth semiconductor substrate of the first conductivity type surrounded by the third semiconductor region P2 and has a portion exposed on the surface of the semiconductor substrate. fifth semiconductor region n
, n4 are formed, and the second semiconductor region p has a portion exposed to the surface of the semiconductor substrate and is of a first conductivity type surrounded by the second semiconductor region p1. A sixth semiconductor region n3 is formed,
The sixth semiconductor region n3 has a portion overlapping the fifth semiconductor region n4 when viewed in plan, and the surfaces of the third and fourth semiconductor regions P and n4 are connected to the first main electrode T.
the third and fifth semiconductor regions p2,
The surface of n4 is in contact with the gate electrode G, the second and sixth semiconductor regions p, and the surface of n3 is in contact with the second main electrode T.
2, the third semiconductor region p2 has an intrusion portion 11 intruding into the fifth semiconductor region n4 in the form of a notch, and the intrusion portion 11
relates to a bidirectional thyristor, characterized in that the gate electrode G is exposed on the surface of the semiconductor substrate, and the gate electrode G is connected to the third semiconductor region P2 at a recessed portion 11a of the intrusion portion 11. It is.

[Function] The intrusion portion 11 of the third semiconductor region p2 into the fifth semiconductor region n4 in the above invention has a gate current ■. 1. Reactive current component work. □, acts as a passageway. Since this intrusion portion 11 is not a region for isolating a plurality of semiconductor regions from each other, it has a large degree of freedom in design. Therefore, the length and width of the intrusion portion 11 can be optimized to obtain the optimum resistance value. That is, reactive current component I. 1
．． By increasing the resistance value of the current path by the resistance of the intrusion part 11, the reactive current component ■. 1. As a result, the gate trigger current is reduced. It is possible to achieve a high level of sensitivity. In addition, reactive current component
Since the intrusion portion 11Tb, which acts as a path, includes a surface region with a low impurity concentration of the third semiconductor region p2, a reactive current component is generated. The temperature characteristics of 1b are good. Therefore, the gate trigger current engineer. The temperature characteristics of 1 are also relatively good, and the gate trigger current control of the voltage between the main electrodes is also relatively good.

The impact on □ will also be reduced.

[Embodiment] Next, a bidirectional thyristor (TRIAC) according to an embodiment of the present invention will be described with reference to FIGS. 1 to 5. However, in FIGS. 1 to 5, parts that are substantially the same as those in FIGS. 8 to 11 are designated by the same reference numerals, and their explanations will be omitted.

The bidirectional thyristors shown in FIGS. 1 to 5 are substantially the same as those shown in FIGS. 8 to 11, except that the pattern of the n4 region and the arrangement of the gate electrode G are changed. are configured identically.

As is clear from FIG. 1, the n4 region (fifth semiconductor region) of this embodiment has a notch, in which the intrusion portion 11 of the 22 region (third semiconductor region) is arranged. This intrusion portion 11 is a portion masked when forming the n4 region by impurity diffusion, and as is clear from FIG.
It is formed to divide four areas from top to bottom. n
The dark narrow region 2 between the 21 region and the n4 region is the reactive current component of the gate current ■. 1. In order to use the intrusion part 1 as a current path and also use the intrusion part 11 as a part of it, the intrusion part 1
The entrance 1 is located near the right end of the narrow region 2. That is, the intrusion portion 11 is disposed off to the right side of the n4 region. The length of the intrusion portion 11 is approximately 577 times the width W of the n region in the direction in which it extends. Further, the width of the intrusion portion 11, that is, the cut width W2 is approximately 177 times the width of the n4 region in the left-right direction in FIG. However, other parts of the intrusion portion 11 are covered with insulating rIA3 as shown in the third country. The gate electrode G is arranged so as to cover almost the entire n4 region and the intrusion portion 11, but in the intrusion portion 11, it contacts only the tip Lla. Therefore, 92 regions from the tip portion 11a of the penetration portion 11 that contacts the gate electrode G to the entrance of the penetration portion 11 can be used as a reactive current component path GTb having a relatively large resistance value. The bidirectional thyristor of this embodiment has the following effects.

(1) Since the reactive current component I can be limited, Tb is the gate trigger current factor. 1 is reduced, and high sensitivity can be achieved. That is, the gate electrode G has a penetrating portion 1 with respect to the p2 region.
Only the tip portion 11a of 1 is in contact with the other.

Therefore, the intrusion portion 11 becomes a part of the path of the reactive current component (GTb), and the resistance of this portion can limit the reactive current component (2). Required Tb Basically, reactive current component I. -mA of 16 passages is intrusion part 1
1, and the other 4B is the part where the contact part between the first main electrode T and the region 92 is closest to the n4 region, and the narrow region 2 and the intrusion between A and B are Since the intervening portion 11 is present, the resistance value becomes larger by the intrusion portion 11 than that shown in FIGS. 8 to 11, and the reactive current component I
decreases Tb and gate trigger current I. □ becomes smaller. Figures 1 to 5
With the structure shown in the figure, the value of the gate trigger current, that is, the value of the gate trigger current, can be improved compared to the bidirectional thyristors shown in sections 8 to 11.

(2) Gate trigger current■. 1 has good temperature characteristics. i.e. gate trigger current engineer. 1, the current path of the reactive current component I6□ is formed in a region with a high impurity concentration on the surface side of the region 92, where the change rate of resistance value due to temperature is relatively small. The temperature dependence of Tb is small, so the reactive current component I and the active current component
Game GTb with
The temperature dependence of GTa trigger current ■6□ is 8th~
It can be made almost as small as the bidirectional thyristor shown in Figure 11,
Gate trigger current engineering due to increased sensitivity. A large decrease in the temperature characteristics of □ can be prevented.

(3) The intrusion portion 11 is formed as a region in which a depletion layer is not formed due to the voltage between the main electrodes.

Therefore, the gate trigger current engineer. The dependence of □ on the main y7rWl voltage is small.

(4) Increase in gate trigger current I in the first mode can be prevented.

[Modifications] The present invention is not limited to the above-described embodiments, and for example, the following modifications are possible.

(1) As shown in FIG. 6, a bent portion may be provided in the narrow intrusion portion 11. By doing so, the length of the intrusion portion 11 increases, and as a result, the resistance of the current path increases. can.

(2) As shown in Figure 7, there is a notch 1 in the gate electrode G.
3 is provided, and the gate electrode G is formed into a planar pattern that almost corresponds to the n4 region.
The gate electrode G may be located only on the area a.

(3) A guard ring region and an EQR (equipotential ring region) consisting of a highly concentrated p-type head region may be formed on part of the surfaces of the P2 region and the p3 region.

(4) It is desirable that the gate type iG be connected near the intrusion part 11, but no matter where it is connected inside the intrusion part 11, it will be ineffective due to the resistance from the connection part of the intrusion part 1 to the entrance part. The current component I can be limited by Tb.

<5) It is also effective for bidirectional thyristors of the type in which the n4 regions (gate regions) are formed other than near the corners of the 22 regions (center gate bidirectional thyristors and side gate bidirectional thyristors).

(6) G by devising the shape of the T1 electrode side of the n2 region.
It is also effective to combine "Mode I." with an improved bidirectional thyristor.

[Effects of the Invention] As described above, according to the present invention, it is possible to provide a bidirectional thyristor that achieves high deterioration and also has excellent temperature characteristics of gate trigger current and voltage dependence between main Ti. .

[Brief explanation of the drawing]

FIG. 1 is a plan view showing the arrangement of each region on the surface of a semiconductor substrate of a bidirectional thyristor according to an embodiment of the present invention, and FIG. 2 shows a state in which an insulating film and an electrode are provided on the semiconductor substrate of FIG. 1. 3 is a sectional view of a portion corresponding to ■-■ bare in FIGS. 1, 2, and 5; FIG. 4 is a partial sectional view of IV-IV II in FIG. 2; 5 is a sectional view taken along the line V-V in FIG. 3, FIG. 6 is a plan view showing a modified example corresponding to FIG. 2, and FIG. 7 is a diagram showing the gate electrode and its vicinity in another modified example. FIG. 8 is an enlarged plan view of a conventional bidirectional thyristor, FIGS. 9 and 10,
A cross-sectional view of a portion corresponding to the line ■-■ in FIG. 11. FIG. 9 is a plan view showing the surface of the semiconductor substrate of the bidirectional thyristor in FIG. 8. The T%10 section is insulated on the semiconductor substrate in FIG. 11 is a sectional view taken along the line XI-XI in FIG. 8; FIG. 12 is a plan view showing another conventional bidirectional thyristor; '741311141! XII[-XI[! !
FIG. 14 is a sectional view taken along the line XrV-XrV in FIG. 13, and FIG. 15 is a diagram showing the relationship between impurity concentration, resistivity, and concentration of a semiconductor. 2... Narrow distribution area, 11... Intrusion part, 11a...
Tip, nl...first semiconductor region, n2...fourth
semiconductor region, n3...sixth semiconductor region, n4...
-Fifth semiconductor region, pl...second semiconductor region, P
2... Third semiconductor region, T1... First main electrode,
T2...second main electrode, G...gate electrode.

Claims (1)

[Claims] a first semiconductor region (n_1) of a first conductivity type;
a second semiconductor region (p_1) of the second conductivity type adjacent to the semiconductor region (n_1) of the first semiconductor region (n_1);
1) has a semiconductor substrate having a third semiconductor region (p_2) of a second conductivity type adjacent to the other side of the semiconductor substrate, and the third semiconductor region (p_2) is exposed on the surface of the semiconductor substrate. fourth and fifth semiconductor regions (p_2) of the first conductivity type surrounded by the third semiconductor region (p_2);
n_2) (n_4) is formed, and the second semiconductor region (p_1) has a portion exposed to the surface of the semiconductor substrate and is surrounded by the second semiconductor region (p_1). A sixth semiconductor region (n_3) of one conductivity type is formed, and the sixth semiconductor region (n_3) has a portion overlapping with the fifth semiconductor region (n_4) when viewed in plan. , the third and fourth semiconductor regions (p_2
)(n_2) is in contact with the first main electrode (T_1), and the surface of the third and fifth semiconductor regions (p_2)(n_
4) is in contact with the gate electrode (G), and the surface of the second
In a bidirectional thyristor in which the surfaces of the sixth semiconductor regions (p_1) (n_3) are in contact with the second main electrode (T_2), the third semiconductor region (p_2) is in contact with the fifth semiconductor region (n_4). ) into the notch-like part (
11), the intrusion part (11) is exposed on the surface of the semiconductor substrate, and the gate electrode (G) is connected to the third semiconductor at a deep part (11a) of the intrusion part (11). A bidirectional thyristor characterized in that it is connected to a region (p_2).