JPH0136265B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to photosensitive thyristors, particularly photosensitive silicon planar thyristors.

A photosensitive thyristor is a thyristor formed so that it can be caused to transition from a forward blocking state to a forward conduction state by irradiation with an optical signal instead of applying a gate signal to a normal reverse blocking three-terminal thyristor having a four-phase PNPN structure, for example. It is.

These photosensitive thyristors can be divided into large current type and medium and small current type depending on the current they can handle. For large currents, a so-called bevel structure in which the blocking junction is exposed on the side of a bevel-shaped substrate is used, whereas for medium and small currents, a so-called planar structure is often used. There is.

This is because the planar structure can reduce leakage current, making it relatively easy to increase photosensitivity, and for medium and small current applications, the withstand voltage does not need to be so high.

However, as the application of photosensitive thyristors has recently progressed, there has been a strong desire to put into practical use photosensitive thyristors with higher breakdown voltage and higher photosensitivity.

The first is a schematic cross-sectional view showing the structure of such a conventional photosensitive silicon planar thyristor (hereinafter referred to as a photothyristor). (The oxide film, electrodes, etc. are omitted.)
Type emitter region ( _P1 region) 3, N-type base region ( _N1 region) 4, and P-type base region ( _P2 region)
5 and an N-type emitter region ( _N2 region) 6
Four layers, P ₁ , N ₁ , P ₂ , and N ₂ , are formed adjacent to each other in sequence.

FIG. 2 shows an example of a photosensitive test circuit for such a photothyristor. The photothyristor 7 is forward biased by a bias power supply 8 via an ammeter 9, and a photodiode 10 directs light onto the top surface of the photothyristor 7. It is becoming irradiated.

In such operating conditions, the junction J ₁ of P ₁ and N ₁ and the junction J ₃ of P ₂ and N ₂ are forward biased, and the junction J ₂ of N ₁ and P ₂ is reverse biased, so that the bias voltage is Most of the voltage is applied to junction J ₂ and a space charge layer is formed near J ₂ .

When light is irradiated in this state, the light energy generates hole-electron pairs, which are amplified near the junction V ₂ and flow into the P ₂ layer or N ₁ layer, causing a photocurrent.
_IR will be streamed. This photocurrent increases the multiplication factor of the minority carrier and turns it on, similar to the gate current in a normal thyristor.

As is clear from the above explanation, in order to increase the photosensitivity of this photothyristor, it is necessary to increase the generation efficiency of hole-electron pairs generated by light irradiation and to efficiently use the generated hole-electron pairs. It is necessary to reach junction J ₂ and increase the multiplication factor of the minority carrier. Therefore, for this purpose, P ₂ area 5
The depth of the junction J ₂ x _jp (see Figure 1) must be made as shallow as possible.

On the other hand, the peak-off voltage (hereinafter referred to as breakdown voltage), which is the forward breakdown voltage, largely depends on the shape of the junction J ₂ (more precisely, the shape of the space charge region formed). That is, the smaller the radius of curvature r (see FIG. 1) at both ends of the bottom of the junction _J2 , the stronger the electric field concentration and the lower the withstand voltage. Therefore, as mentioned above, the shallower the depth x _jp of the junction J ₂ is made to increase the photoresponsiveness, the lower the withstand voltage at which the photoresponsiveness increases, and in some cases, the product specifications may not be met.

Next, in order to make the electrons, which are minority carriers generated in the P ₂ region 5 by light, efficiently reach the junction J ₂ , the minority carriers in the P ₂ region 5 must be The carrier's lifetime must be extended.
For this purpose, it is necessary to reduce the surface impurity concentration C _S of the P ₂ region 5.

On the other hand, the rate of increase in off-voltage (dv/dt), which is one of the important characteristics that determines the stability of the turn-on characteristics of a thyristor, increases as the resistance of this _P2 region 5 increases, that is, the surface impurity concentration C _S decreases. Because the voltage drop due to the deflection current is large, dv/
dt becomes smaller.

As explained in detail above, when trying to increase the photosensitivity of conventional photothyristors, the withstand voltage and dv/dt decrease. It is extremely difficult to produce photothyristors, and we have overcome this difficulty to create products with high light sensitivity, high voltage resistance, and high
There is a strong desire to realize a photothyristor with dv/dt characteristics.

An object of the present invention is to overcome these difficulties and provide a photosensitive silicon planar thyristor with high light sensitivity, high breakdown voltage, and high dv/dt characteristics.

The photothyristor of the present invention includes a silicon substrate of the first conductivity type, an emitter region of the second conductivity type, a base region of the first conductivity type, a base region of the second conductivity type, and an emitter region of the first conductivity type. In the photosensitive silicon planar thyristor in which four layers are formed adjacent to each other in sequence, the surface impurity concentration of the base region of the second conductivity type is (2×10 ¹⁷ to 2
×10 ¹⁹ )/cm ³ and its junction depth is (43 to 60) μm
This is because it is.

Hereinafter, the present invention will be explained in detail using the drawings.

In order to solve the difficulties of the prior art described in detail above, the photothyristor of the present invention has the same structure as shown in FIG. _{P 1} , N 1 , P consists of an N-type base region (N ₁ region) 4, a P-type base region (P ₂ region) 5, and an N-type emitter region (N _{2 region} ) 6. As a result of experiments on a photothyristor in which four layers of ₂ and N ₂ are successively formed next to each other, the depth x _jp of the junction J ₂ in the P ₂ region 5 and the surface impurity concentration C _S were changed to different colors. Contrary to the reasoning based on the prior art described above, in order to increase the photosensitivity, the depth of the junction J ₂ must be made much deeper than before, and the surface impurity concentration must be further increased.
This is because we found that the optimal range _{of CS} also exists in a much larger range than before. As is clear from the explanation of the prior art described above, this also increases the withstand voltage and dv/dt, making it possible to obtain a photothyristor with high light sensitivity, high withstand voltage, and high dv/dt characteristics, which is the object of the present invention. .

Next, these experimental results will be explained in detail.

The test circuit used in the experiment was the circuit shown in FIG. 2 described above. As the LED 10 used here, a GaAs infrared light emitting diode, which is commonly used as a photothyristor coupler that combines an LED and a photothyristor, and whose emission wavelength peaks around 0.94 μm was used. Then, a bias voltage of 12 V is applied by the bias power supply 8, near-infrared light with an irradiance of 10 mw/cm ² is applied from the GaAs light emitting diode to the surface of the base region N1, and a photocurrent I _R flows through the photothyristor 7 by the ammeter 9. was measured.

(Experiment 1) Relationship between junction depth x _jp and photocurrent I _R of junction J ₂ .

When the surface impurity concentration in the P ₂ region is kept constant at 1×10 ¹⁸ (/cm ³ ) and only the junction depth x _jp is changed in color,
Figure 3 shows the _IR measurement results. The horizontal axis in the same figure is x _jp
(μm), and the vertical axis is I _R (μA).

As is clear from this figure, I _R is x _jp as
This can be shown by a curve with a maximum value near 50 μm, indicating that there is a region of high photosensitivity deeper than the conventionally used 40 μm or less. This means that the wavelength of the LED10 used
This can be easily understood considering that the average penetration depth of 0.94 μm infrared light into silicon is approximately 50 μm.

(Experiment 2) Relationship between surface impurity concentration C ₂ in N ₂ region and photocurrent I _R.

FIG. 4 shows the measurement results of I _R when the surface impurity concentration C _S of the P ₂ region is changed in color while the depth x _jp of the junction J ₂ is kept constant at 40, 50, and 60 μm, respectively. The horizontal axis of the figure is the surface impurity concentration C _S (/cm ³ ), and the vertical axis is the photocurrent I _R
(μA), and the parameter is x _jp (μm).

As is clear from this figure, I _R is 1× C _S
It can be represented by a curve with a maximum value around 10 ¹⁸ /cm ³ . Furthermore, it can be seen that almost no dependence on x _jp is observed, and each curve shifts in accordance with the relationship between I _R and x _jp shown in FIG.

This result also differs from the above-mentioned prior art inference that the smaller the surface impurity concentration C _S is, the higher the photosensitivity (corresponding to I _R ) should be, and it is clear that the optimum value exists where C _S is large. I have to.

This result can be better understood if considered as follows. Electrons as minority carriers generated in the P ₂ region 5 by light irradiation are transported toward the junction J ₂ by diffusion through the P ₂ region, and are also transported by a drift electric field induced by the concentration gradient of impurities in the P ₂ region. and is accelerated. And this drift electric field is C _S
The larger the number, the stronger it becomes. On the other hand, as explained above, some of the electrons disappear before reaching the junction _J2 due to recombination with holes. In other words, the rate M at which electrons generated by light irradiation reach the junction J ₂ is (1 -
M ₁ ) and the fraction M ₂ carried by the drift electric field {M∝(1−M ₁ )M ₂ }.
By the way, in _CS large, M ₁ large, M ₂ large (1-M ₁ ) small,
For C _S small, M ₁ is small, M ₂ is small, and (1 - M ₁ ) large, so for a specific C _S (C _SO , M is maximum. Thus, if x _jp is held constant and C _S is changed, C _SO The curve that maximizes I _R will be obtained.

Next, we will examine in what range x _jp and C _S should be set to obtain the highly light-sensitive photothyristor that is the object of the present invention.

First, since the photosensitivity of a photothyristor according to the prior art is 10 μA as an I _R value, a range that satisfies this is determined. In Figure 4, I _R =
If you draw a line of 10μA, x _jp = 60μm curve and C _S is 2×
Intersect at two points, 10 ¹⁷ /cm ³ and 2×10 ¹⁹ /cm ³ , and C _S is 1
It can be seen that I _R is 22μA at ×10 ¹⁸ /cm ³ . Next, in FIG. 3, when the value of x _jp corresponding to x _jp =60 μm where I _R becomes 22 μA is found, x _jp =43 μm is obtained. In other words, when I _R = 10 μA is the minimum standard, x _jp is (43 to 60) μm and C _S is (2 × 10 ¹⁷ to 2
It is sufficient to form the P ₂ region 5 having a diameter of 1.times.10 ¹⁹ )/cm ³ .

Furthermore, we will consider a case where the I _R value is set to 20 μA, twice that of the conventional photothyristor, as a highly light-sensitive photothyristor.
First, in Figure 4, C _S is 2×10 ¹⁷ /cm ³ and 2
Draw a curve similar to the curve of x _jp = 50 μm so that I _R = 20 μA at two points of ×10 ¹⁹ /cm ³ , and C _S
= 1×10 ¹⁸ ) _/ cm ³ as the value of I _R =
Get 30.5μA. Next, from Figure 3, this I _R = 30.5μA
Obtain 45 μm and 55 μm as the values of x _jp corresponding to .
In other words, if I _R = 20μA is the minimum standard,
x _jp as (45-55) μm and C _S as (2×
10 ¹⁷ /cm ³ to 2×10 ¹⁹ /cm ³ )/cm ³
All that is required is to form the _P2 region 5.

FIG. 5 illustrates the results obtained from the above study. Note that the representation of the figures is the same as before.

From FIG. 5, the _{photothyristor} of the present invention has a surface impurity concentration C _S (2×10 ¹⁷ to 2×
10 ¹⁹ )/cm ³ and depth of junction J ₂ x _jp (43~60) μm
It can be seen that it is easy to obtain a photosensitivity nearly four times larger than the conventional value of 10 μA as _IR .

Additionally, if you only need something with high light sensitivity,
C _S (2 × 10 ¹⁷ ~ 2 × 10 ¹⁹ ) / cm ³ and x _jp (45 ~
55) By using μm, it is possible to easily obtain a photothyristor with even higher photosensitivity, with an I _R value of 20 μA or more, which is twice that of the conventional one.

As is already clear from the above explanation, the photothyristor of the present invention has two other problems with the _conventional photothyristor, which are the breakdown voltage and dv/dt. Since _S is made higher than before, dv/dt is larger than before. Also, since the depth of the joint J ₂ is increased, the radius of curvature r at both ends of the bottom of J ₂ becomes larger, and therefore the withstand voltage becomes higher. (Both of these have been confirmed experimentally.) In the explanation so far, we have taken up the case where the silicon substrate 1 is N type, but this
It goes without saying that the present invention is similarly applicable to molds.

As explained above in detail, the photosensitive silicon planar thyristor of the present invention has a surface impurity concentration of (2×10 ¹⁷ to 2
×10 ¹⁹ )/cm ³ and its joining depth (43 to 60)
The ranges are determined so that a greater photosensitivity than the conventional one can be obtained based on the sensitivity curve, which is larger than the conventional one and has the maximum value for each photosensitivity. is on average about (2~
3) Has twice the high light sensitivity and has high voltage resistance and
Higher light sensitivity, higher voltage resistance, and higher dv/dt
This makes it possible to provide a photosensitive silicon planar thyristor with dt characteristics, which is highly effective.

[Brief explanation of drawings]

Fig. 1 is a schematic cross-sectional view showing the structure of a conventional photothyristor, Fig. 2 is a photothyristor test circuit diagram, Fig. 3 is a diagram of the junction _J2 depth x _jp vs. photocurrent I _R characteristic curve, Figures 4 and 5 use x _jp as a parameter.
FIG. 3 is a characteristic curve diagram of surface impurity concentration C _S versus photocurrent I _R in the P ₂ region. 1... N-type silicon substrate, 2... Breakthrough layer,
3...P type emitter region (P ₁ region), 4...N
Type base region ( _N1 region), 5...P type base region ( _P2 region), 6...N type emitter region ( _N2
area), 7... Photothyristor, 8... Bias power supply, 9... Ammeter, 10... Photodiode (LED), J ₁ , J ₂ , J ₃ ... Junction, x _jp ... Depth of junction J ₂ S, C _S ... surface impurity concentration of region P ₂ .

Claims (1)

[Claims] 1 an emitter region of a second conductivity type having a penetration layer surrounding a base region of the first conductivity type; the base region of the first conductivity type; a base region of the second conductivity type; and an emitter region of the first conductivity type. In the photosensitive silicon planar thyristor, the four layers of the second conductive type are formed adjacent to each other in sequence, and the flat surface of the base region of the second conductive type is irradiated with light from a GaAs infrared light emitting diode as a photocoupler light source. The surface impurity concentration of the base region of the mold is set to (2×10 ¹⁷ to 2×
10 ¹⁹ )/cm ³ and its junction depth is (43 to 60) μm
A photosensitive silicon planar thyristor characterized by being formed to have optimum sensitivity to near-infrared light.