JPH01246869A - Schottky barrier semiconductor device - Google Patents

Abstract

PURPOSE:To protect the high-speed responding feature from degrading and to increase breakdown strength by a method wherein a Schottky barrier region is formed by arranging a first barrier electrode on the upper surface of a first semiconductor region and a second barrier electrode is so formed as to surround the first barrier electrode. CONSTITUTION:A first barrier electrode 29a is arranged on a first semiconductor region 23 for the formation of a first Schottky barrier. A second barrier electrode 29b is formed to surround the first barrier electrode 29a, positioned on the first semiconductor region 23 for the construction of a second Schottky barrier. The second barrier electrode 29b is electrically connected to the first barrier electrode 29a and is higher in sheet resistance than the first barrier electrode 29a. A second semiconductor region 24 is formed next to the first semiconductor region 23, is opposite to the first semiconductor region 23 in conductivity type, serves as a guard ring, and is provided with a surface electrically connected to the first barrier electrode 29a through the second barrier electrode 29b. In such a design, the high-speed responding feature is protected from degrading and the breakdown strength may be enhanced.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to a high voltage Schottky barrier semiconductor device.

A problem to be solved by the prior art and the invention: Schottky barrier diodes have been widely used in pan-frequency rectifier circuits, etc. by taking advantage of their good short-speed response, fast switching characteristics, and low loss. However, Schottky barrier diodes have a phenomenon in which the single peripheral withstand voltage (withstand voltage around the Schottky barrier) is significantly lower than the bulk withstand voltage (withstand voltage at the center of the Schottky barrier). We have the problem that t-.

Measures to solve this problem are introduced below. 8th
The figure shows a Schottky barrier diode with a guard ring. The semiconductor substrate 1 used is one in which an n-type semiconductor region 6 of higher resistance is formed on the upper surface of an n-type semiconductor region 2 of low resistance. An electrode 4 is formed on the soil surface of the n-type semiconductor region 6 to form a Schottky barrier at the interface with the n-type semiconductor region 3. The lower surface of the n-type semiconductor region 2 has an ohmic contact (C low resistance contact) with the n-type semiconductor region 2.
An electrode 5 is formed. Of the electrodes 4, the insulating film 6
The portion extending to the soil surface is called the field plate 7, and together with the guard ring, which will be explained later, contributes to improving the peripheral withstand voltage. That is, the field plate 7 has the function of expanding a depletion layer near the surface of the n-type semiconductor region 6 below it when a reverse voltage is applied. As a result, the electric field concentration around the Schottky barrier is alleviated, and the one-periphery breakdown voltage is improved. A low resistance p-type semiconductor region formed in an annular shape in contact with the periphery of the Schottky barrier is a region that acts as a guard ring 8.

In the Guard Link Honzo, the surrounding pressure resistance of the Schottky barrier is t-
The pn junction 9 takes on this role, and together with the action of the field plate 7, a high withstand voltage can be achieved even in a case that is not sufficiently strong. but.

When a large current flows in the 111 direction, the pn junction 9
Injection of minority carriers (holes) into the n-type thin plate 3 increases from . Therefore, even if the forward operation is switched to the reverse operation, the case where the minority carrier disappears will not be completely switched off. RO: A delay in switching response occurs, resulting in a reduction in high-speed response (high-frequency characteristics or high-speed switching characteristics).

A Schottky barrier diode having a structure as shown in FIG. 9 is disclosed in Japanese Patent Application Laid-Open No. 58-215079 as a guard ring structure that solves the above-mentioned reduction in switching response. That is, the electrode 4 forming the Schottky barrier
and the electrode 11 of the guard ring 8, and the technical fjR layer 10 having high sheet resistance formed on the insulating film 6.
electrically connected via. According to this structure,
The current in the 1111 direction mainly flows through the interface between the electrode 4 and the negative region 3, and the current flowing from the periphery of the pole 4 to the p shadow region is restrained by the insulating layer 10. For this reason.

Injection of minority carriers into the above n-shaded region 3 can be suppressed,
As a result, a Schottky barrier diode having a guard ring with almost no reduction in switching response can be provided.

However, it was not sufficient to achieve high voltage resistance. Generally, it is known that an electric field concentration point where the maximum electric field occurs exists in a semiconductor region with a large sheet resistance (under the boundary between two different coating layers).

In the Schottky barrier diode shown in FIG.

This electric field concentration point occurs in the n-type semiconductor region 3 below the boundary between the 'fR pole 4 and the insulating film 6.

Breakdown is likely to occur at this electric field concentration point.

As a result, the withstand voltage decreases.

Therefore, one aspect of the present invention is that high-speed responsiveness is less degraded.

It is an object of the present invention to provide a Schottky barrier semiconductor device having a guard ring capable of achieving a breakdown voltage of A.

[Means to solve the problem]

The present invention will be described with reference to the reference numerals in the drawings showing the embodiments.
, a first barrier electrode 29a, and a second barrier electrode 29
b, and a second semiconductor region 24, and the first barrier electrode 29a has a first barrier electrode 29a between the first semiconductor region 26 and the second semiconductor region 24.
The second barrier electrode 29b is arranged on the upper surface of the first raw conductor axis fjI123 to form a Schottky barrier, and the second barrier electrode 29b surrounds the first barrier electrode 29a in @ contact with the first barrier electrode 29a. The first semiconductor region 23 is arranged on the first semiconductor region 23 so as to form a second Schottky barrier between the first semiconductor region 23 and the second semiconductor region 23, and is in electrical contact with the first barrier electrode 29a. It has a sheet resistance larger than that of the first barrier electrode 29a, and the second
The semiconductor region 24 has a conductivity type opposite to that of the first semiconductor region 23 and acts as a guard ring.
A Schottky barrier conductor device, characterized in that it has a surface that is disposed adjacent to the semiconductor region 26 of and electrically connects the second barrier to the first barrier electrode 29aK via a pole 29b. It is related to.

[For production]

The second semiconductor region 24 in the above invention functions as a guard ring and contributes to high breakdown voltage. Pole 2 for the second barrier
Since the barrier electrode 9b has a sheet resistance larger than that of the first barrier electrode 1y29a, it is possible to suppress a decrease in high-speed response caused by providing the second semiconductor @ region 24 as a guard ring region.

[First Embodiment] A Schottky barrier diode and a method for manufacturing the same according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 4.

When manufacturing the Schottky barrier diode shown in FIG. 1, a semiconductor substrate 21 made of GaAs (gallium arsenide) is prepared for the casing, as shown in FIG. 300μm.

N-shaped candy area 22 with impurity concentration of 1 to 3 × 10”cm
±, thickness 10-20μm, impurity concentration 1-2X 1
015 cm-3 of n-shaded region 26 is epitaxially grown.

Next, as shown in Figure 2 CB), the n-type G a A s
Selectively in the n-shaded region 23 consisting of Zn (zinc)
is diffused to form a guard ring region 24 consisting of a p shadow region. The p+ type region depth is about 2 μm1 surface impurity Hf!
/IJ concentration is approximately 5×IQ”cryl.

Next, as shown in FIG. 2(C), a thin layer 25 of Ti (titanium) is formed on the entire soil surface of the shadow area 23 and the guard ring area 24 by vacuum evaporation, and then Continuously vacuum evaporate 26 layers of AI (aluminum) I on the entire soil surface. The thickness of the Ti thin layer 25 is 50~20 OA.
(0,005 to 0.02 am) and e-thin.

The thickness of the 41 layer 26 is about 2 μm, and the thickness of the sTi thin layer 25 is about 100 μm.
It is more than twice as floating. Furthermore, Au is added to the lower surface of the n-shaded area 22
An Au-Tac contact electrode 27 made of (gold)-Ge (germanium) alloy is formed by vacuum evaporation, and then,
Heat treatment is performed at 680° C. for 10 seconds.

Next, as shown in FIG.
The I layer 26a is left. Furthermore, the 'I'i# layer 25 is removed from the peripheral region of the element by photoetching, and the AIj
A Ti thin N 25 a located below ti 26 a and a Ti thin N 25 a surrounding it adjacently remain. Ti
Since the thin layers 25a, 25b are extremely thin films even though Tf itself is a conductor, they are resistance layers with a sheet resistance of 20 to 400 Ω/hole.

It has higher resistance than the AI layer 26a.

Next, heat treatment is performed in air at 300°C for 5 to 30 minutes. As a result, as shown in Figure 2 (■).

The undoped Ti thin layer 25b of the AI layer 26a is oxidized to become a thin layer 28 of titanium oxide.

The Ti thin layer 25a under the AI layer 26a is not oxidized because it is masked by the AI layer 26a.

Next, by photo-etching r) AI layer 26 a f
) The entire periphery is removed, and the unoxidized Ti thin layer 25c under the AI layer 26a is exposed in a kiln, as shown in FIG. 2 [F]. As shown in FIG. 3, the Ti thin layer 25c is the AI layer 26a.
A! so as to surround the outer periphery of! A thin titanium oxide layer 28 is disposed adjacent to the thin Ti layer 25c so as to surround the thin Ti layer 25c. Since both AI and Ti are metals that form a Schottky barrier with GaAs, At J So 26
The Schottky barrier forming gold m[pole'' consisting of the Ti thin layer 25d disposed under the AI bird 26b and the first barrier gold electrode 29a is referred to as the first barrier 'I! electrode or the first barrier gold electrode 29a. Schottky barrier forming gold i consisting of a thin layer of Ti thin N 25 c adjacently surrounding the barrier electrode @L29a.
The ii electrode will be referred to as a second barrier metal electrode 29b. This is because the Ti thin layer 25d is an extremely thin film.

In the first barrier gold J@electrode 29a, AI/! 12
It is not necessarily clear how the Ti thin layer 6b and the Ti thin layer 24d are respectively involved in forming the Schottky barrier. In addition to the role of the Ti4 layer 25d in forming a Schottky barrier, the Ti thin layer 25d also contributes to the adhesion between the A1 layer 26b and the n-shade region 23. Further, the Ti thin layer 25c forming the second barrier metal electrode 29b contributes to the electrical connection between the first barrier metal type b29a and the titanium oxide thin layer 28. Here, the sheet resistance of the first barrier metal electrode 29a, which serves as the main flow path for the 111''RL flow, is preferably 10/port or less, and in this embodiment, it is about 0.05 Ω/port. The titanium oxide thin layer 28, which is provided in a ring shape so as to surround the second barrier metal electrode 29b shown in FIGS.
It is a medium-insulating, high-resistance layer with a thickness greater than that of the thin layer 25b and approximately 70 MΩ/hole. That is, the titanium oxide thin layer 28 is a perfect insulator, and the cell Ti02 (2rjlR + fin) T&Zt! < , 1'1ChJ-
') 'bIt is considered that Tiex is a so-called oxygen-poor titanium oxide Tiex (where X is a value smaller than 2). In addition, the guard ring region 24 made of a p-type thin plate according to the present invention includes a Ti thin layer 25b, that is, a second barrier gold Ws*b 29b and a titanium oxide thin layer 2.
It is arranged in the n-type region 23 at the lower part of the boundary portion of 8. The surface of the guard ring area 24 is made of Ti thin 11m 25
c and the lower surface of both the thin titanium oxide layer 28. As shown in FIG. 3, the guard ring region 24 is formed in an annular shape along the peripheral edge of the second barrier gold Si electrode 29b.

Subsequently, the upper surfaces of the titanium oxide thin layer 28 and the second barrier metal electrode 29b are covered with an insulating film 30 to complete a semiconductor chip having a Schottky barrier, that is, a power Schottky barrier diode chip. By the way, is insulation 1-60 bad? CV D (Chemica I Vapor
It consists of a silicon oxide film formed by a deposition method. The insulating layer 60 is formed by plasma CVD
Although it is possible to use a silicon nitride film formed by a plasma or photo-CVD method or a polyimide resin film formed by a coating method, a silicon oxide film formed by a plasma (VD or photo-UVD method) is preferable. Although not shown in the figure, there are, for example, a Ti layer and an Au layer on the upper surface of the AI layer 26a.
Layers are provided sequentially.

This is normally used as an electrode for connection to a lead member. At this time, the outer peripheral sides of the Ti layer and the Au layer are covered with an insulating layer 30.
By disposing the insulating layer 30 on the upper surface of the insulating layer 30, the Ti layer and the Au layer can protect the insulating layer 30, and also add a field plate.

Examples of dimensions of each part in FIG. 6 are as follows. 1st
Barrier gold M t & 29 a f') @a is approximately 910
μm, the width of the second barrier metal electrode 29b is approximately 20 μm
, the width C of the titanium oxide thin layer 28 is approximately 140 μm.

Further, the width d of the guard rig region 240 is approximately 30 μm.

In this Schottky barrier diode.

A first Schottky barrier and a second Schottky barrier are provided between the first barrier metal [pole 29a and the n-type candy region 23 and between the second barrier metal pAim pole 29b and the n-type head 26, respectively. Not only does a thin titanium oxide layer 28
A sixth Schottky barrier also occurs between the N-type region 26 and the N-type region 26 . The occurrence of a Schottky barrier between the titanium oxide thin layer 28 and the n-type diode 26 was confirmed by the rectification characteristics, capacitance characteristics, saturation current characteristics, etc. of the Schottky barrier diode. For example, when the area of the titanium oxide thin layer 28 is increased from zero, the saturation current 18 increases approximately in proportion to the sum of the areas of the titanium oxide thin layer 28 and the first and second barrier metal electrodes 29a and 29b. do. It has been confirmed that this proportional relationship can be obtained at various temperatures of Schottky barrier diodes. The fact that the saturation current 18 changes approximately proportionally to the sum of the areas of the titanium oxide thin layer 28 and the first and second barrier gold mt electrodes 29a and 29b means that the first and second barrier This means that a reverse current flows through the titanium oxide NN 128 at approximately the same current density as that of the metal electrodes 29a and 29b. This phenomenon is caused by the barrier hide φB whose thin titanium oxide/1 m28 is approximately the same as the barrier metal electrodes 29a and 29b.
This clearly shows that a Schottky barrier with t- is formed.

Second barrier gold N electrode 29b and titanium oxide thin layer 28
forms a Schottky barrier with the n-type thin plate 23, but is thought to have a relatively low resistance contact with the p-type overhead or guard ring region 24, which has a slightly non-curved shape. Therefore, it can be regarded as conductive when a reverse voltage group is added, and a Schottky barrier is not formed for the guard ring@region 24. Therefore, the first Schottky barrier based on the first barrier metal electrode 29a and the second Schottky barrier based on the first barrier metal electrode 29a and the second
The second Schottky barrier based on the pole 29b is continuous with the @ contact, but the sixth Schottky barrier based on the titanium oxide thin layer 28 and the second Schottky barrier are connected via the guard ring region 24.

Not directly consecutive. The 1L second Schottky barrier and the third Schottky barrier can also be considered to be continuous via the boundary between the guard ring region 24 and the n shadow region 26, that is, the pn junction 61.

The guard ring region 24 contributes to improving the peripheral breakdown voltage of the Schottky barrier as in the conventional example. here.

The second barrier metal 'fit! compared to the first barrier metal 1fA'tlh29a! The sheet fold of 529b is quite large. Therefore, in the Schottky barrier diode shown in FIG. 1, the forward current fjft flowing through the guard ring region 24 into the n-shade region 23 when a 111-direction voltage is applied is limited by the second barrier metal electrode 29b. Therefore, from the guard ring region 24 through the pn junction 31,
Minority carriers (C holes) injected into the shadow region 23 are drastically reduced, and the provision of the guard link region 124 causes almost no deterioration in quick response. That is, the second barrier gold striped electrode 29b of the Schottky barrier diode shown in FIG. 1 is an electrode that forms the entire Schottky barrier, and also limits the injection of minority carriers.

The titanium oxide thin layer 28 forms a Schottky barrier between it and the n-shade region 23, and greatly contributes to improving the breakdown voltage as a highly resistive field plate. That is, when a reverse voltage is applied, a minute current flows between the inner and outer edges of the thin titanium oxide layer 28, creating a potential gradient in the direction of the thin titanium oxide layer 28. The spread of the depletion layer formed by the Schottky barrier formed by the titanium oxide thin layer 28 is large on the guard ring region 24 side due to the voltage distribution gradient, and becomes smaller toward the outer peripheral edge of the titanium oxide thin layer 28. When a reverse voltage is applied, the first barrier gold lAm electrode 29a and the second barrier gold m electrode 29b and the n-type lnj2
J! 23, first and second Schottky barriers formed between the pn junction 31. The depletion layers extending from each of the Schottky barriers of n i3 formed between the titanium oxide thin layer 28 and the n-shade region 23 are integrated, and the depletion layer 32 schematically shown in FIG. It is formed continuously on the surface of the region 23. As a result, 4) IL field concentration is alleviated locally in the Schottky barrier when a reverse voltage is applied, and the withstand voltage of the Schottky barrier diode is significantly improved. In addition, the titanium oxide thin layer 28 increases the voltage resistance! Regarding II4 construction, one was invented by the inventors of the present application, etc.
It has been filed as Japanese Patent Application No. 62-307196 by applicant FFlthl.

The Schottky barrier diodes of this example were able to achieve withstand voltages of 150V or more with high yield, and some even showed a total withstand voltage of approximately 250V, which is considered to be approximately equal to the bulk withstand voltage.

When no measures are taken to increase the withstand voltage (titanium oxide thin layer 28,
In the case where the Ti thin layer 25c and the p''-shaped region 24 are not formed, a withstand voltage of only about 60 V can be obtained.The Schottky barrier diode of this embodiment was also subjected to cylindrical frequency switching with a switching frequency of 500 kHz. When used as a regulator diode, rectifying operation with extremely low noise generation was confirmed.

This embodiment has the following advantages.

(The region surrounded by the guard ring on the surface of II n-type gjiM23 has a first Schottky barrier and a second Schottky barrier based on the first barrier metal 'w1 electrode 29a and the second barrier gold ljA electrode 29b. Therefore, when a reverse voltage is applied, the depletion layer etched from the first Schottky barrier, the depletion layer etched from the second Schottky barrier, and the depletion layer from the pn junction 61 are continuous. A good depletion layer 31 that alleviates field concentration can be obtained.

Even in the conventional structure having a field plate made of an insulator, a depletion layer continuous with the depletion layer extending from the pn junction 9 can be obtained in the region surrounded by the guard ring. but.

Since a depletion layer is formed through the insulating layer 6, a good depletion layer for alleviating electric field concentration cannot be obtained, and a strong breakdown voltage Jc as in this embodiment cannot be expected.

When biased in the +21 1111 direction, the current that flows through the pn junction 31 to the n shadow region 23 flows through the second barrier metal electrode 2.
9b. Therefore, high voltage resistance can be achieved without impairing high-speed response.

(31) The thermal instability, which is one of the drawbacks of the conventional SEIZO having a field plate with an insulating layer, is due to the presence of an insulating layer under the Ti thin layer 25c and the titanium oxide thin layer 28. It is solved by not having it.

(41 Stress concentration/imaginary and electric field concentration points can be separated,
We were able to improve the pressure yield. That is.

This is because the first barrier metal electrode 29a is thicker than the coating layer formed on the outer peripheral side thereof.

A stress concentration point occurs at the lower part of the outer peripheral edge of the first barrier metal paulownia' electrode 29a. At this stress concentration point, since strain is applied to the n-shaded region 26, the critical electric field 1M degree (Ecrit) is reduced, and breakdown is likely to occur if electric field concentration occurs here. In this embodiment, the electric field concentration point where the maximum electric field occurs is located near the outer periphery of the guard ring region 24, so that the electric field concentration point can be separated from the stress concentration point. In addition, the frequency of products with a breakdown voltage below a specified value is reduced, and the yield of breakdown voltage is improved. In addition, in the conventional JRI&ZO structure having a field plate with an insulator interposed therebetween, the electric field concentration point is located below the boundary between the barrier metal electrode 4 and the insulator 6, and substantially coincides with the stress concentration point. For this reason, no improvement in voltage yield can be expected.

Since the Ti thin layer 25c and the titanium oxide thin layer 28 are obtained in the extended portion of the Ti thin layer 25d provided directly under f51 AI 7 & 26 b, the desired second barrier gold w
Four electrodes 29b and a thin titanium oxide layer 28 are easily obtained. Further, the electrical connection between the first barrier metal electrode 29a and the second barrier metal electrode 29b and the titanium oxide thin layer 28 can be easily and reliably established.

(6) Since the titanium oxide thin layer 28 is provided, the breakdown voltage can be further improved. That is, by providing the titanium oxide thin layer 28''ff as described above, the guard ring region 24
There is also a depletion J that alleviates electric field concentration in the n-shaded region 26 on the outer peripheral side of
~ can be formed well.

Therefore, the breakdown voltage could be significantly improved.

[Second Embodiment] A Schottky barrier diode according to a second embodiment of the present invention shown in FIG. rExplain. However, in FIG. 5 and FIGS. 6 and 7, which will be explained later, parts that are substantially the same as those in FIGS. 1 to 3 are denoted by the same reference numerals, and the explanation thereof will be omitted. The Schottky barrier diode shown in Figure 5 has a ring-shaped titanium oxide thin layer 28ak on the n-type @R230 surface.
Furthermore, a ring-shaped titanium oxide thin 7ili
Has i28b. The first barrier metal electrode 29a is made of AI
It consists of a layer 26b and two Ti thin layers 25e and 25f. The second barrier metal electrode 29b has two Ti thin layers 25g, 2
It consists of 5 hours. This maki-zukuri can be formed by repeating the steps shown in FIG. 2(C) to ω twice. The upper titanium oxide thin layer 28b is formed to have the same thickness and oxidation degree as the titanium oxide thin layer 28 of the Schottky barrier diode shown in FIG. The lower titanium oxide thin layer 28a has a thickness of N- compared to the upper titanium oxide thin layer 28b, but the degree of oxidation is stronger than that of the upper titanium oxide thin layer 28b. The sheet resistance is higher than that of the titanium oxide layer 28b. Therefore, the current passing through the upper titanium oxide thin layer 28b is higher than the current passing through the upper titanium oxide thin layer 28b.
The current flow through a also increases, and the potential gradient is mainly determined by the upper f4INVC. Since the titanium oxide thin layer 28a in the latter half has a barrier hide φB, the saturation current 1s can be reduced.

Therefore, a Schottky barrier diode with a low reverse current level can be provided.

Furthermore, by making the second barrier metal electrode 29b thick, the resistance is reduced, so that the current at the time of breakdown is dispersed to several locations on the outer peripheral end side of the second barrier metal electrode 29b and flows smoothly. Therefore, the reverse surge load resistance is improved. however,
The sheet resistance of the second barrier metal electrode 29b needs to be set to a resistance value that can suppress injection of minority carriers into the n-shaded region 26 when +1vi voltage is applied.

Note that, similarly to the first embodiment, the first and second barrier metal electrodes 29a, 29b and the guard ring region 24 consisting of the p shadow region are harmonized;
Ru.

[Sixth Embodiment] FIG. 6 shows a Schottky barrier diode according to a third embodiment of the present invention. The Schottky barrier diode in FIG. 6 is a second barrier metal 1! On the outer circumferential side of the pole 29b, a first
Titanium oxide thin layer 28C1 Ti thin layer 251 for equalizing potential
．． The second titanium oxide thin layers 28d are arranged adjacent to each other in a ring shape and are electrically connected to each other.

[For equipotential] [ii] The annular @* made of the thin layer i has conductivity, so it can be one equipotential distribution @ area. As a result, it is possible to correct the non-uniformity of the potential distribution in plan view on the surface of the n-shaded region 23, to form a uniform depletion layer all over, and to equalize the breakdown voltage. Also.

The outer cylinder end of the titanium oxide thin layer 28d is Au (gold) -G
A connection region 64 made of e (germanium) alloy is electrically connected to the n shadow region 23 with low resistance.

Therefore, since breakdown does not occur at the outer peripheral edge of the titanium oxide thin layer 28d, a high breakdown voltage Schottky barrier diode for small signals that generates less noise can be produced. Of course,
In the Schottky barrier diode shown in Fig. 6, the first
Also, the effect of improving the breakdown voltage can be obtained by the combination of the second barrier metal electrodes 29a, 29b and the guard ring region 24 consisting of the p# region.

[Fourth Embodiment] The Schottky barrier diode of the fourth embodiment shown in FIG.
A thick Ti thin layer 33 is formed by the i thin layer 25 . The thickness of the Ti thin layer 33 is 100 to 400 Å, which is extremely thin compared to the AI layer 26c.

For the Schottky barrier diode in No. 7, a thin Ti layer 63 is formed in advance on the soil surface of the n-shade region 23, as shown in FIG. 2(C).
It can be manufactured using the process shown in ~(g). In the Schottky barrier diode shown in Figure 7, the Al layer is 26c.

A portion consisting of the Ti layer 25j and a part of the Ti thin layer 63 is used as the first barrier layer I! On the pole 29a, on the Ti thin layer 25, and on the T
The remaining portion of the i-thin layer 63 is made of second barrier gold IpI.
It can be called a four-electrode 29b. As in the sixth embodiment, a Schottky barrier diode with reduced sheet resistance of the second barrier gold W4 electrode 29b and improved reverse surge resistance can be provided. Further, since the barrier 8ite φB of the peripheral Tit 47133 portion of the Schottky barrier is improved, the peripheral breakdown voltage of the Schottky barrier is improved, and a Schottky barrier diode with an even higher breakdown voltage can be provided. In this case too.

The second barrier metal electrode h 29 b toy sheet resistance is set to a value that can appropriately limit the forward current. Note that this embodiment also provides the effect of improving the breakdown voltage due to the combination of the first and second barrier gold PA electrodes 29a, 29b and the guard link range 24.

[Modified example]

This is not limited to the above-described embodiments.

For example, the following transformations are possible.

fil If a high withstand voltage is not particularly required, the thin titanium oxide layer 28 may be omitted.

(The effective range of the sheet resistance of the 21 titanium oxide thin layers 28.28b to 28d varies depending on the actual structure and size of the semiconductor chip, but it is preferably 10 to 5000 MΩ/2, preferably 10 to 1000 MΩ/28. ]OMΩ/You should choose it.

(31 Fig. 2 (The thickness of the Ti thin layer 25 in Q is controlled by thickness control.

Considering the oxidation temperature, oxidation time, etc., the oxidation temperature should be at least 2OA. There is no upper limit as long as the above-mentioned predetermined sheet resistance can be obtained, but when a TiN film is thermally oxidized to form a titanium oxide thin layer.

Considering the oxidation temperature and oxidation time, it should be 300A or less. If you perform strong oxidation such as plasma oxidation,
This upper limit can be further expanded.

+41 Oxidize Tt thin layer 25 to form titanium oxide thin layer 2
The oxidation temperature for obtaining No. 8 is desirably 500° C. or lower, and when using Au thread IL, the oxidation temperature is 380° C. or lower. Regarding the lower value of oxidation degree, it is set at 200° C. or higher when using the #IrjlR method, but when using plasma oxidation, it can be lowered to a low temperature below room temperature. Oxidation time is TiWI layer 25 layer thickness 5. Oxidation temperature.

It varies depending on the oxidizing atmosphere, but it is desirable to keep it within the range of 5 seconds to 2 hours! Yes.

(Those corresponding to 51+ tan oxide thin layers 428.28a to 28d are formed by vapor deposition or sputtering of titanium oxide, and the Ti thin layers 25c, 25h, 25g, and 25i are replaced with titanium nitride layers with relatively high conductivity. Alternatively, the titanium nitride layer may be formed by masking off the A1 layer and nitriding the titanium oxide layer.

(61 A titanium oxide thin layer is suitable as a thin layer that has high sheet resistance and forms a Schottky barrier, but Ta
(Tantalum) It is also possible to use thin oxides of thread materials. In addition, the 7゛i rumor layer 25 and the titanium oxide thin layer 28
may be added with In'P Sn or the like.

It is also applicable to Schottky barrier semiconductor devices using Ill-V group compounds such as lnP (indium phosphide) or silicon instead of 171 GaAs.

(8+ In the case of forming a Schottky barrier conductor device in the circuit, it is necessary to
A shadow area 22 is provided to connect the 1st pole 27 to the n-type head 23.
It is also possible to have a brainer structure provided on the surface side.

f91 n shadow area 23. The n-shadow area 22 can be replaced with a p-shaped head.

(101) The effective range of the sheet resistance of the second barrier gold@electrode 29b as a thin layer according to the present invention varies depending on the conditions of the semiconductor chip.

The effects of the present invention can be effectively obtained within the range of 1 Ω/mouth to 1 MΩ/mouth. However, if a high-level reverse diagonal gate is required, it is desirable to set the resistance to 5Ω/port to 500Ω/port.

[Ill The thickness of the second barrier metal type h29b as a thin layer according to the present invention, the thickness of the titanium oxide thin layer 28, and the combined thickness of the titanium oxide thin layers fVI28a and 28b are designed to have reasonable numerical values. 5,000 yen so as not to substantially cause stress concentration points.

az Second barrier metal 1! The width of the L pole 29b should be 2 μm or more in order to separate the stress concentration point and the electric field concentration point. However, even if the thickness is 50 μm or more, the effect will be saturated and the area of the chip will only increase. Therefore 2-5
It is best to set it in the range of 0 am.

u31 flint tlt compound thin 428.28a, 28b
.

By setting mt (the sum of 28c and 28d) to about 10 .mu.m or more, the effect of improving the breakdown voltage appears, and by setting it to 30 .mu.m or more, the effect becomes remarkable.

However, it is more desirable to design the thickness to be 100 .mu.m or more in order to improve the yield at which a predetermined breakdown voltage can be obtained. Even if the width is set to 500 lIm or thicker, a sufficient voltage-resistance improvement effect can be obtained.

Therefore, although there is no upper limit to the width, even if the width is increased to 500 .mu.m or more, it is not possible to expect a proportional increase in breakdown voltage.

Therefore, it is desirable to have one width in the range of 30 to 500 μm!
Yes.

〔Effect of the invention〕

According to the present invention, there is a Schottky barrier semiconductor device t' in which the high-speed response is less degraded despite the provision of the second semiconductor region that functions as a guard ring.
can be provided.

[Brief explanation of the drawing]

FIG. 1 is a sectional view showing a Schottky barrier diode according to a first embodiment of the present invention. Figures 2-5 are sectional views showing the Schottky barrier diode of Figure 1 in the order of the leg end process. The third prisoner is a plan view showing the state of m2 diagram ■. FIG. 4 is a partially enlarged sectional view of a Schottky barrier diode schematically showing a nitrogen-depleted layer. FIG. 5 is a top view showing the Schottky barrier diode of the second embodiment. FIG. 6 is a sectional view showing a Schottky barrier diode of a third embodiment. FIG. 7 is a sectional view showing a Schottky barrier diode of a fourth embodiment. FIG. 8 is a sectional view showing a conventional Schottky barrier diode. The ninth figure is a cross-sectional view showing another conventional Schottky barrier diode. 22...n shadow area, 23...n shadow area, 24...
Guard ring region, 25c, 25d-Ti thin layer, 26
a... A1 layer, 27... Ohmic electrode, 28...
-Titanium oxide thin layer, 29a...first barrier metal 1
! Pole, 29b... second barrier metal electrode.

Claims (1)

[Claims] [1] A first semiconductor region (23), a first barrier electrode (29a), a second barrier electrode (29b), and a second
a semiconductor region (24), and the first barrier electrode (29a) is connected to the first semiconductor region (23) so as to form a first shot barrier between the first barrier electrode (29a) and the first semiconductor region (23). 23), the second barrier electrode (29b) adjacently surrounds the first barrier electrode (29a) and has a second barrier electrode between the first semiconductor region opening and the second barrier electrode (29b). disposed on the first semiconductor region (23) to form a shot barrier and electrically connected to the first barrier electrode (29a);
The second semiconductor region (24) has a higher sheet resistance than the first barrier electrode (29a), and has a conductivity type opposite to that of the first semiconductor region opening so as to function as a guard ring. is arranged adjacent to the first semiconductor region opening, and the second semiconductor region is arranged adjacent to the first semiconductor region opening;
A shot barrier semiconductor device characterized by having a surface electrically connected to the first barrier electrode (29a) via a barrier electrode (29b).