JPH03272184A - Insulated-gate bipolar transistor - Google Patents

Abstract

PURPOSE:To obtain a semiconductor device improved in switching speed and lessened in loss by a method wherein crystal defect is partially formed in a second semiconductor layer of second conductivity type formed on the surface of a first semiconductor layer of first conductivity type. CONSTITUTION:A bipolar transistor of this design is different from a conventional one in following points: a metal absorber 60 of Al or the like is provided to the surface of a collector electrode formed on the other primary face; a mask 61 of stainless steel or the like provided with fine openings is provided thereon; and a light ion beam of He or the like is made to irradiate through the intermediary of the absorber 60 and the mask 61 for the formation of the bipolar transistor. At this point, the acceleration energy and the thickness of the absorber 60 are regulated so as to enable the range of the light ion beam 50 which passes through the fine openings provided to the mask 61 to lie inside a P<+> collector layer 1. By this setup, crystal defects are formed in a region correspondent to the pattern of fine openings provided to the mask 61, whereby a partial lifetime control can be carried out.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an insulated gate bipolar transistor (Insulated gate bipolar transistor).
5ulated Gate Bipolar Tran
The present invention relates to IGBTs (hereinafter referred to as IGBTs), and particularly to devices that aim to improve the trade-off relationship between turn-off time and on-resistance associated with lifetime control and reduce switching loss.

[Conventional technology]

Bipolar transistors generally have low output impedance, but also have a problem of low input impedance. On the other hand, field effect transistors (hereinafter referred to as MOSFETs) have a high input impedance, but also have a problem of high output impedance. In contrast, IGBTs are integrated to compensate for the shortcomings of these various transistors.
The aim is to achieve high human power impedance and low output impedance.

In other words, by creating a high a degree impurity diffusion layer of a conductivity type different from that of the substrate on the back side of the substrate on which the MOSFET is formed, a bipolar transistor and a MOSFET can be formed.
The bipolar transistor is controlled by the injected current by integrating the T and injecting the current generated when the MOSFET is turned on into the base region of the bipolar transistor (1) star.

Generally, an IGBT device has a structure in which a large number of IGBT elements (hereinafter referred to as IGBT cells) are connected in parallel. FIG. 7 is a sectional view showing the structure of a conventional n-channel type I GBT cell, and FIG. 8 is a circuit diagram showing its equivalent circuit.

In FIG. 7, (1) is a P collector layer, and an N epitaxial layer (2) is formed on one main surface thereof. In some areas of the surface of the N epitaxial layer (2),
A P-well region (3) is formed by selectively diffusing P-type impurities, and high-strength N-type impurities are selectively added to a part of the surface of the P-well head (3). An N+ emitter decoy region (4) is formed by diffusion into the N+ emitter decoy region (4). A gate insulating film (5) is formed on the surface of the P well region (3) sandwiched between the surface of the N epitaxial @ (2) and the surface of the N+ emitter covering region (4).
．． The edge film (51) is also formed on the surface m-h of the N epitaxial layer (2) so as to be integrated between adjacent IGBT cells.A gate made of, for example, polysilicon is formed on the gate insulation film (5). An electrode (6) is formed and an emitter electrode (7) of metal, e.g. aluminum, is formed to electrically connect both the P-well head (3) and the N+ emitter head (4) ing. Note that the gate electrode (6) and emitter electrode (7) have a multilayer structure via the #@edge film (8), so that they are electrically connected in common to all IGBT cells. A metal collector electrode (9) is integrally formed on the back surface of the P collector layer (1) for all IGBT cells.

The vicinity of the surface of the P well region (3) sandwiched between the N epitaxial layer (2) and the N+ emitter region (4) has an n-channel MO5 structure, and a positive electrode is connected to the gate electrode (6) through the gate terminal G. By applying a voltage, electrons are transferred to the N+ emitter region (4) through a channel formed near the surface of the P-well head (3) directly under the gate electrode (6).
) to the N epitaxial layer (2). Illustration ■.

represents the electron current flowing in this way. On the other hand, holes, which are minority carriers, are transferred from the +P collector layer (1) to N
It is injected into the epitaxial layer (2), a part of which recombines with the electrons and disappears, and the rest flows through the P-well head (3) as a hole current as shown. In this way, ■GBT is
Basically, it operates in a bipolar manner, and the N-epitaxial layer (2) increases the conductivity due to the effect of conductivity variation, achieving a lower on-voltage and larger current capacity than the conventional power MO8. There are advantages that can be achieved.

In FIG. 8, (10) is a parasitic NPN transistor consisting of an N epitaxial layer (2), a P well region (3) and an N+ emitter mount 4), and ul) is a P collector layer +1+ and an N epitaxial layer. PNP l-transistor consisting of (2 (and P-well head mounted (3), t1
2) is an NMOS transistor with a channel head mounted on the P well region (31 surface) under the gate electrode (6), RB is the diffusion resistance of the P well region (3), and RLC is the on-resistance of 1ull of the PNP transistor. .

Although the IGBT has the above-mentioned advantages, it has the disadvantage that the operating frequency cannot be increased because the hole current Ih decreases more slowly at turn-off than in a MOSFET or the like. This is a PNP transistor (]J)
When is in the on state, the N epitaxial layer @ (2) which is mounted on the base is filled with electrons and holes, and the MO
S) This is due to the fact that even if the transistor is turned off to block the injection of electrons into the N epitaxial layer (2), the genuine bill does not suddenly decrease due to its low mobility.

In order to shorten this turn-off time, two methods are known. One of them is to use modified metal atoms such as gold and platinum as so-called lifetime killers.
This is a means of introducing into the N-epitaxial layer (2) which is the base region of the P-transistor (11), and this lifetime killer becomes a recombination center for electrons and holes in the N-epitaxial layer (2). carrier will disappear within a short period of time.

The other method is to irradiate radiation such as electron beams, gamma rays, neutron beams, and various ion beams.Since these radiations introduce deep trap levels within the N-epitaxial layer (2), this Since the trap level serves as a recombination center for carriers, carriers can be annihilated within a short time at turn-off. These techniques are called lifetime control techniques and are applied to various elements such as thyristors and power diodes.

In general, lifetime control technology using radiation irradiation has yielded better results in terms of controllability and reproducibility than tunic metal diffusion. However, during radiation irradiation, electron beams,
In methods using gamma rays and neutron beams, the irradiation generates trap levels in the N epitaxial layer (2), and at the same time changes the film quality of the gate oxide film (5), resulting in changes in the value as well. There are problems that can cause changes and reduce its operational reliability. This problem can be solved by a method of irradiating various ion beams such as protons from the collector electrode (9) side. That is, as shown in Fig. 4, various light ion beams such as protons are irradiated from the side where the collector electrode (9) is formed, and the range position is set in the N epitaxial layer (2). (as shown by the broken line in Figure 7)
, by adjusting the acceleration energy, the gate insulating film (5) and other layers formed on the emitter side +31,
! Lifetime control can be performed without affecting 41 in any way.

Furthermore, as shown in Figure 9, crystal defects (mainly vacancies) due to irradiation with various ions such as protons occur intensively in the half-width W of the defect distribution peak, centered on the range, and other than that. It has characteristics that do not affect the location much. By utilizing this characteristic, it is possible to perform lifetime control with high controllability. For example, as shown in JP-A-64-19771, N-
By setting the range in the base region (corresponding to the N epitaxial layer (2) in FIG. 7), effective lifetime control can be performed. This is because carriers injected from the channel of the MOSFET are triggered by the carriers injected from the channel of the MOSFET, causing one major groove modulation.
Since it plays an important role, generating crystal defects in this part will increase the on-resistance, so
This is because it is desirable for the range of the ion beam to reach the N-base region closest to the P collector region, which is farthest from the T channel. Furthermore, in order to quickly capture holes that are continuously injected until the initial stage of off-operation, it is effective to intensively generate crystal defects in the N base region near the P+ collector region.

[Problem to be Solved by the Invention 1] However, all of the above-mentioned lifetime controls still generate crystal defects over the entire surface of the IGBT element. The resistance value of layer (2) inevitably increases, and the on-resistance RLC of the IGBT in FIG. 8 increases. In other words, there is a trade-off relationship between the on-resistance and turn-off time of the IGBT, and there is a problem in that the trade-off relationship is not optimal at present.

This invention was made to solve the above-mentioned problems, and it is a structure that optimizes the trade-off relationship between on-turn and turn-off times through lifetime control using irradiation with ionizing radiation such as ion beams. The purpose is to obtain an IGBT of

[Means to solve the problem]

The insulated gate bipolar transistor according to the present invention includes a second semiconductor layer of a first conductivity type, and a second semiconductor layer of a second conductivity type selectively formed on the surface of the first semiconductor region.
a semiconductor region, a second semiconductor layer, an insulating film formed on a surface of the first semiconductor region sandwiched by the second semiconductor region, and a control electrode formed on the insulating film. and a first main electrode formed across the first and second semiconductor layers, and a second main electrode formed on the back surface of the first semiconductor layer, 2 semiconductor “n
The structure is such that crystal defects are partially present near the junction with the first semiconductor region.

[Function] In the insulated gate bipolar transistor of the present invention, the bipolar transistor on the equinodal circuit, which is a component thereof, includes a first semiconductor region, a second semiconductor layer having no crystal defects, and a second semiconductor layer having no crystal defects. a first bipolar transistor consisting of a first semiconductor layer having a first semiconductor region, a second semiconductor layer having crystal defects, and a second bipolar transistor consisting of a first semiconductor layer having no crystal defects; It can be considered to be equally economical to configure it by connecting in parallel.

The first bipolar transistor has a crystal defect in the first semiconductor layer but does not have a crystal defect in the second semiconductor layer, so its advantage is that the on-resistance is low, and its disadvantage is that the turn-off time is long. On the other hand, since the second bipolar transistor has crystal defects in the second semiconductor layer, its disadvantage is that its on-resistance is high, and its advantage is that its turn-off time is short.

Since the first and second bipolar transistors are connected in parallel, the first bipolar transistor works dominantly in the on state, and the second bipolar transistor works dominantly in the turn-off state, resulting in low on-voltage and high-speed switching. Realize.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings.

Note that the description of parts that overlap with the description of the conventional technology will be omitted as appropriate.

FIG. 1 is a sectional view showing the structure of an IGBT according to an embodiment of the present invention. In the figure, (1) to (9) are the same as before.

The difference is that the collector electrode (9) on the other main surface side
A metal absorber (60) made of aluminum or the like, and a mask (61) made of stainless steel or the like with minute openings formed on the surface are provided on the surface, and a light ion beam (50) such as helium is irradiated through these. Here, the portion of the light ion beam (50) that has passed through the fine aperture of the mask (61) is charged with acceleration energy and absorber □ so that the range position is set within the N base layer (2+). The thickness of the mask (61) is adjusted so that the range position of the light ion beam 00) that has passed through the bright region excluding the fine aperture of the mask (61) is set within the P collector @ (1). The thickness of (61) is adjusted. As a result, partial lifetime control can be performed for loading on ships that almost reflects the pattern of fine apertures. Figure 2 is shown in Figure 1. It is a circuit diagram showing an isochoric circuit of things. In the figure, (10'l,
(121 is the same as the conventional one, (13) is a built-in PNP transistor formed in a bright region whose lifetime is not controlled, and tl, 11 is a built-in PNP transistor with a region whose lifetime is partially controlled.

Figure 3 shows an engineered GBT partially irradiated with a light ion beam ([10) and an IGBT fully irradiated with a light ion line width 0).
FIG. 3 is a diagram comparing trade-off characteristics with

In this case, divalent helium ions (He) are used as the light ion beam 150), and their acceleration energy is 20Me
It is V. Also, the absorber (60) has a thickness of 30 μm.
The mask (6I) has a thickness of 50 mm.
It is made of μm stainless steel. mask 161
) There are two types of fine openings formed in the mask 1 with radius γ
is 140/7m, with a pitch of 200μm between the windows.
The mask 2 has a radius γ of 160 μm.
The circular windows were formed with a pitch of 200 μm between the windows.

From this, it can be seen that the partial irradiation using Mask 1 and Mask 2 has an improved trade-off curve compared to the whole irradiation.

By the way, in Figure 4, tM and FEt-Von are approximately 3.3.
FIG. 3 is a diagram comparing the turn-off waveform characteristics of the IGBT material (2) subjected to full irradiation and partial irradiation. It can be seen that the Tilmi flow is reduced in the case where partial irradiation using mask 2 is performed compared to the case where the entire surface is irradiated.

Further, FIG. 5 is a diagram similarly comparing losses at turn-off. As the aperture of the mask (61) becomes finer, the turn-off loss decreases, and with partial irradiation using mask 2, it is possible to reduce the turn-off loss to 43910 with full irradiation even in high temperature conditions (TA = 125°C in the figure). It has become.

FIG. 6 is a sectional view showing the structure of an IGBT according to another embodiment of the invention. This device differs from the one shown in Figure 1 in that an absorber +6CI and a mask (61) are installed on the emitter electrode (7) on the − main surface side, and a light ion beam 00) is irradiated through it. It was formed. In this case as well, an absorber +6 (9 made of metal such as aluminum, mask +61) made of stainless steel or the like is used. The mask (61) is provided with a fine opening, and a light ion beam 1150 such as divalent helium ions (He 2 ) is irradiated through the mask (61). At this time, for the light ion beam that has passed through the area excluding the fine aperture according to the mask rule, the range position is set in the area near the junction with the +42175 layer (1) in the N base layer (2). In addition, the light ion acceleration energy conditions, absorber (60, and the lowliness of the mask punishment) are adjusted.
Light ion beam 115 passed through the fine aperture of the mask (61)
0), the thickness of the absorber battle is adjusted so that the range position is set within the 42175th layer (1). The IGBT circuit manufactured in this manner is shown in FIG. Therefore, in this case too, there is a trade-off relationship,
The improvement effect on loss etc. can be expected to be the same as that shown in FIG.

In the above embodiment, the case where helium was used as the light ion was shown, but the same effect as described above can be obtained even if other light ions such as protons are used.

〔Effect of the invention〕

As described above, according to the present invention, since crystal defects are partially formed in the second semiconductor layer of the second conductivity type, it is possible to improve the switching speed and obtain a semiconductor device with reduced loss. There is.

[Brief explanation of drawings]

Fig. 1 is a cross-sectional view showing an IGBT which is an embodiment of the present invention, Fig. 2 is a circuit diagram showing a circuit similar to that shown in Fig. I GB
Figure 4 is a diagram comparing the trade-off characteristics between T and an IGBT whose entire surface is irradiated. Figure 4 is a diagram which compares the waveform characteristics at turn-off of an IGBT which is partially and fully irradiated with light ion beams. Figure 5 This figure similarly compares the loss at turn-off, and FIG. 6 shows the IGB of another embodiment of the present invention.
7 is a sectional view showing the structure of a conventional IGBT, FIG. 8 is an equivalent path diagram of the one shown in FIG. 7, and FIG. 9 is a diagram showing the range of irradiated ions and FIG. 3 is a diagram showing the relationship between the distribution of crystal defects formed. In the figure, (1) is the P+ collector layer, (2) is the N base layer, (3) is the P well head, 141 is the N emitter region, (5) is the gate insulating film, (6) is the gate electrode, ( 7)
is the emitter electrode, (8) is the insulating film, (9) is the collector electrode, is the light ion beam, +61 is the absorber, 16++
is a mask. Note that the same reference numerals in each figure indicate the same or equivalent parts.

Claims (1)

[Claims] (1) a second semiconductor layer of a second conductivity type formed on the surface of the first semiconductor layer of the first conductivity type; and a second semiconductor layer selectively formed on the surface of the first semiconductor layer. a second semiconductor region of conductivity type 2; an insulating film formed on a surface of the first semiconductor region sandwiched between the second semiconductor layer and the second semiconductor region; and the insulating film. a control electrode formed above, a first main electrode formed across the first and second semiconductor regions, and a second main electrode formed on the back surface of the first semiconductor layer; An insulated gate bipolar transistor characterized in that the second semiconductor region has a crystal defect partially near a junction with the first semiconductor region.