JPH02281757A - Semiconductor device - Google Patents

Abstract

PURPOSE:To prevent disconnection at the stepped section of a metallic electrode while keeping the controllability of a coating area to a semiconductor region of a metallic electrode by forming an upper section in a tapered shape and a lower section in an arcuate shape in the sectional shape of the opening section of an insulating film. CONSTITUTION:An upper section is formed in a tapered shape 17 and a lower section in an arcuate shape 18 in the sectional shape of an opening section 19, through which a metallic electrode 4 is applied in a semiconductor region 21, in an insulating film 12. Accordingly, a contact angle at the contact-section edge (b) of the metallic electrode 4 and the semiconductor region 21, the angle of the insulating film 12, is increased, the controllability of a coating area with the semiconductor region of the metallic electrode 4 is improved, and the upper section of the opening section 19 is formed in the tapered shape 17, thus augmenting the angle of the upper edge (a) of the opening section of the insulating film 12, then preventing disconnection at the stepped section of the metallic electrode 4.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a semiconductor device, and particularly to controllability of the deposition area of a metal electrode deposited on a semiconductor region through an opening in an insulating film and prevention of disconnection. It is something to improve.

[Summary of the invention]

The present invention provides a semiconductor device in which a metal electrode is adhered to a semiconductor region through an opening in an insulating film, in which the cross-sectional shape of the opening in the insulating film is tapered at the top and arcuate at the bottom. This makes it possible to prevent breakage of the metal electrode while maintaining controllability of the adhesion area of the metal electrode to the semiconductor region.

[Conventional technology]

Semiconductor diodes for high frequencies, e.g. GaA for mixers
In order to improve the frequency characteristics of the S Schottky diode, it is necessary to reduce the junction area 9, for example, to a diameter of 10 μm or less. In this case, stricter dimensional control is required to maintain device uniformity.

Typically, the formation of Schottky contacts for these diodes is
1 μm on the main surface of the semiconductor region (1) as shown in Figure 3.
A thick insulating film such as 5i02 film (2) is deposited, and this Si0g film (2) is selectively etched through a photoresist layer mask to form an opening (3). This is done by depositing a Schottky electrode or Schottky metal (4) on the semiconductor region (1). A pad metal (not shown) is deposited on the Schottky metal (4). And in this case, normal HF
The opening (3
), the upper edge a of the opening (3) becomes close to a right angle, and the Schottky metal (4) (including the band metal) is broken at this upper edge a, resulting in a problem of deterioration of reliability. Conventionally, in order to avoid this metal step break, a SiO□ film (2
) has a tapered cross-sectional shape (6) in the opening (3);
The angle of the upper edge a of the opening (3) is increased to prevent metal steps including the Schottky metal (4) from breaking.

[Problem to be solved by the invention]

However, in the configuration shown in FIG. 4, the edge of the contact area between the Schottky metal (4) and the semiconductor region (1) is already
The angle θ of the Oz film (2) becomes smaller. Since the positional dimension of this edge depends on Sin and the etched sheet β of the film (2) by β/tanθ, for example, when θ=20°, it changes by 2.75×β, which is the Schottky junction. This worsened the controllability of the width W and therefore the Schottky junction area.

In addition, with the reduction of the Schottky junction area, S i 0
2. Frequency characteristics deteriorated due to a relative increase in parasitic components due to the capacitance of the tapered portion of the film (2).

Although the above description deals with the case of Schottky electrodes, other electrode wirings that make ohmic contact with live conductor regions also have similar problems such as breakage and controllability of the adhesion area.

In view of the above-mentioned points, the present invention provides a semiconductor device that can prevent the metal electrode from breaking while maintaining controllability of the area to which the metal electrode is adhered to the semiconductor region.

[Means to solve the problem]

The present invention provides a semiconductor device in which a metal electrode (4) is adhered to a semiconductor region (21) through an opening (19) of an insulating film (12), in which the cross-sectional shape of the opening (19) is tapered at the top. (17), and the lower part is formed into an arc shape (18).

[Effect]

Since the cross-sectional shape of the opening (19) of the insulating film (12) formed on the semiconductor region (21) is arcuate (17) at the lower part in contact with the semiconductor region (21), the metal electrode (4)
The contact angle at the edge of the contact portion between the metal electrode (4) and the semiconductor region (21), that is, the angle θ of the insulating film (12) increases, and the metal electrode (4)
The controllability of the adhesion area with the semiconductor region is improved. Also, since the upper part of the opening (19) is tapered (17), y! ! , the angle of the upper edge a of the opening of the Li film (12) becomes large, and step breakage of the metal electrode (4) is prevented.

〔Example〕

The present invention will be explained in detail below.

As shown in FIG. 2A, the insulating film (12) formed on the base (11) has a three-layer structure, and each layer (13) (14)
(15) Etching rate β [β1. β2. The relationship between β3 and β3 is β3>β1)β2. Each layer (13) (14) (
15) The relationship between the film thickness d (L, dz, d3) is d, > d
z) Consider a configuration chosen so that dz. specifically,
Table I shows examples of the insulating films (13), (14), and (15) when a GaAs semiconductor is used as the base (11).

This three-layer structure consists of an insulating film (12) and a photoresist layer (1).
When selectively etching is performed using 6) as a mask, the etching progresses as shown in FIG. 2B and below.

That is, after etching the third layer film (I5) and the second layer film (14) and opening a window (see FIGS. 2B to 2E).
When the first layer film (13) is further etched using an HF-based etching solution, the cross-sectional shape of the first layer film (13) at the initial stage of etching is similar to that of the second layer film (13), as shown in FIG. 2F. 1
4) is used as a mask, the etching is isotropic, and the etching becomes an arc shape centered on the position B0 of the opening edge of the second layer film (14) at the start of etching.

However, as the etching progresses, the first layer III (
13) and the third layer film (14) are β, >β. Therefore, the relationship between the etching rate of the first layer film (13) and the second
The contact point A (hereinafter referred to as point A) of the layer film (14) and the contact point C (hereinafter referred to as point C) between the third layer film (15) and the second layer film (14) located further back than that. Interval I! , 1 gradually becomes larger. On the other hand, since the distance 12 between point 81 and point 0 is less than a certain value determined by etching rates β2 and β, point B1 will be inside point A at a certain point, and from then on, the first layer film (13) The second mask acts as a mask for
The point A formed by the N film (14) retreats at the etching rate β3 of the third layer film (15), and the upper cross section of the first layer film (13) takes on a tapered shape (17). (See Figure 2H). However, in this case, the isotropically etched part in the first half of the etching process has a circular arc shape (18 points) centered at 88 points.
Etching continues. Therefore, after the opening (19) reaching the base (11) of the insulating film (12) is formed, the cross-sectional shape of the opening is tapered (17) at the upper part and tapered at the lower part, as shown in FIG. 2I. has an arcuate cross-sectional shape (18). As a result, the opening size W can be controlled to the extent of the etching rate β1 of the first layer film (13), similar to non-taper etching.

Note that the time tx from which the first layer film (13) starts etching to tapered etching is tx=β3d2/β!
It is calculated as (β, −β1). In the specific example listed in Table 1, Lx = 4 minutes.

Another method is to make the insulating film (12) have a two-layer structure without using the third layer film (15), and in this case, the second layer film (15) is not used.
14) and photoresist j! ! This method utilizes the fact that the adhesion between the photoresist layer (16) and the photoresist layer (16) is reduced to some extent, and the etching solution gradually penetrates between the photoresist layer (16) and the second layer film (14) during etching. In this case, if the etching rate β3 of the third layer film (15) is replaced with the penetration rate of the etching solution, a partially tapered shape of the first layer film (13) can be obtained by the same effect, and the final An opening (17) similar to that in Fig. 2I has a tapered upper part (17) and an arcuate lower part (18) in cross-section.
19) is obtained.

The present invention utilizes the method described above, and then
An embodiment applied to an As Schottky diode will be described together with its manufacturing method.

As shown in FIG. 1A, first, second and third layers of films (13), (14) are formed on the main surface of the GaAs substrate (21).
An insulating film (12) consisting of (15) and (15) is deposited.

For each layer (13), (14) and (15), the same ones as shown in Table 1 are used, for example.

Next, as shown in FIG. 1B, a photoresist layer (16) having a predetermined opening (22) is deposited on the insulating film (12), and then a photoresist layer (16) is formed as shown in FIG. 1C. (
After selectively etching the third layer film (15) with an HF-based etching solution using 16) as a mask, the second layer film (14)
is selectively etched by dry etching using CF and 02 gas.

Next, the first layer film (13) is selectively etched using an HF-based etching solution. At this time, 50% overetching is performed. In this etching, the upper part of the insulating film (12) is tapered (17) due to the same effect as explained in FIG.
), and an opening (19) having a cross-sectional shape with an arcuate lower part (18) is formed (see first diagram).

Next, the photoresist layer (16) is peeled off, and the third layer film (15) is etched away using an HF-based etching solution. A Schottky electrode, that is, a Schottky metal (4) is formed, and a Schottky junction is formed between it and the GaAs1 plate (21). Table 1 shows each layer (13), (14) and (15)
When setting as follows, the height h of the non-tapered part of the opening (19) is 2000, and the width X of the tapered part is 2.
．． It becomes 4 μm. Schottky metal (4)
) Form a pad metal on top. In this way, the desired GaAs Schottky diode (23) is obtained.

In addition, in a method that does not use the third layer film (15), the portion related to the third layer film (15) may be omitted in the process shown in FIG.

According to the above-mentioned GaAs Schottky diode (23), since the cross-sectional shape of the opening (19) of the insulating film (12) is tapered (17) at the upper part, the opening of the insulating film (12) The angle of the upper edge a becomes larger, which prevents metal breakage including the Schottky metal (4), and the lower part is arcuate (18), so that the Schottky metal (4) and GaAs Substrate board (
The angle θ of the insulating film (12) at the edge of the contact portion with the GaAs substrate (21) becomes larger, and the opening dimension, that is, the opening width W, facing the surface of the GaAs substrate (21) is better controlled, and as a result, a Schottky junction is formed. It is possible to improve the controllability of the area. Also, the opening (
Since the lower part of 9) is arcuate (18), even if the Schottky junction area is reduced, the parasitic capacitance can be suppressed compared to the conventional case shown in FIG. Therefore, the frequency characteristics of the GaAs Schottky diode can be further improved.

Although the above method is applied to the formation of a Schottky metal in a Schottky diode, it can also be applied to the formation of an ohmic metal on a semiconductor region through an opening in an insulating film in other semiconductor devices.

(Effects of the Invention) According to the present invention, in a semiconductor device in which a metal electrode is adhered to a semiconductor region through an opening in an insulating film, the cross section of the opening is formed so that the upper part is tapered and the lower part is circular. By forming this, it is possible to prevent the metal electrode from breaking while maintaining controllability of the adhesion area of the metal electrode to the semiconductor region.

[Brief Description of the Drawings] Figures 1A to 1E are manufacturing process diagrams when the present invention is applied to a GaAs Schottky diode, and Figures 2A to 2 illustrate the method for forming an opening in an insulating film according to the present invention. 3 and 4 are cross-sectional views of essential parts of a conventional GaAs Schottky diode, respectively. (11) is the substrate, (12) is the three-layer insulating film, (13)
). (14), (15) are the respective layers, (16) is the photoresist layer, (17) is tapered, (18) is circular arc,
(21) is a GaAs substrate. 1-11hanpen A Jinjun A 2-・-5i0February

Claims (1)

[Claims] In a semiconductor device in which a metal electrode is deposited on a semiconductor region through an opening in an insulating film, the opening has a cross-sectional shape with a tapered upper part and an arcuate lower part. Semiconductor equipment.