JPH0342688Y2 - - Google Patents

Description

[Detailed explanation of the invention] This invention is a field-effect transistor (hereinafter referred to as
The present invention relates to a FET structure suitable for integrating analog circuits.

Conventionally, improvements in FET performance, that is, higher gain, lower noise, and higher speed, have been achieved mainly by shortening the gate electrode length, channel length, or increasing the channel width by increasing the width of the gate electrode. This has been achieved through expansion. Increasing the channel width makes it possible to increase the gain of the FET and reduce noise, but at the same time, increasing the channel width also increases the dimension of the FET in the direction parallel to the gate. Therefore, in order to create a FET with high gain and low noise, a long FET in the direction parallel to the gate is required.
It becomes necessary to make . However, when incorporating a FET into an integrated circuit, a long FET like this
When each element in a circuit is arranged in an integrated circuit, disadvantages arise, such as a reduction in the degree of freedom of arrangement and an increase in the overall dimensions of the integrated circuit. In order to avoid the above-mentioned inconveniences, the IBM Journal Research Development (IBM Journal Research Development)
As shown in "Development" (March 1970), p. 125, a method has been devised in which the gate electrode is arranged in an annular shape so that the FET can be used effectively. The degree of freedom of the parts is small, and the utilization efficiency of the device plane cannot be said to be high.

The purpose of this invention is to solve the above-mentioned drawbacks and to
The actual size of the FET is changed from a long one in one direction to one that spreads to the same extent in two directions, making it easier to use the FET.

In this invention, in order to make the width of the gate electrode of the FET as large as possible and to make the shape of the FET close to a square, the entire FET is connected to the gate electrode, source electrode,
A method is adopted in which the electrode is formed in a spiral shape on the plane where the drain electrode exists. By forming the FET in a spiral shape, in this invention the entire semiconductor surface can be covered.
Can be used as a FET. Furthermore, since it is formed in a spiral shape, an FET exists next to an FET. In other words, the FET that has gone around the spiral is
It exists alongside the FET. For this reason,
Another gate can be inserted between these adjacent identical FETs. By inserting this new gate, the FET gate width can be effectively doubled with the same area compared to the conventional electrode arrangement.

A concrete embodiment of the present invention will be described with reference to FIG. It can always be applied to horizontal FETs, but as an example, G ^as shot barrier ^type
Explain FET.

First, N-type is grown on ^a semi-insulating ^GaAs substrate by epitaxial growth or ion implantation.
Forms ^{a GaAs} semiconductor. Next, N-type impurities are further added to the source electrode portion shown by 2 in FIG. distribution according to the law, etc. this is,
This is to form an ohmic contact with the electrode metal. Next, the source electrode section indicated by 2 in FIG.
Metal is deposited on the drain electrode portion shown by , the source electrode pad portion 6, and the drain electrode pad portion 8, that is, the portion into which N-type impurities are doubly introduced by the above method, to form electrodes. Next, gate electrode part A indicated by 1, gate electrode part B indicated by 3, 5 and 7
A metal that forms an N-type semiconductor layer and a shot barrier is deposited on the gate electrode pad portion shown by .
The gate electrode part A indicated by 1 and the gate electrode part B indicated by 3 are connected at the innermost part of the spiral FET as shown in the figure, and this part is also sandwiched between the source electrode indicated by 2 and the drain electrode indicated by 4. , operates as a FET. In such a FET, all parts except each pad part operate as an FET. In addition, the parts of the spiral that are located at different circumferences also operate as FETs, and the lengths of the source and drain electrodes are longer than normal.
Even if it is the same as a FET, it operates as a FET with twice the length.

With the above configuration, in this invention,
Although the FET has an effective game width twice as long as the conventional one, the area it occupies does not increase, and its shape is closer to a square. This makes it easy to use as a FET, and in the case of FET manufacturing, FETs with long gate widths can be manufactured without creating elongated chips, which has advantages such as improved yield and easier handling after chip scribing. There are many. Also, when using such a FET in an integrated circuit, the FET
Since it does not have an elongated shape, wiring is easy. Since the FET has the same shape as other elements, the degree of freedom in arranging elements within the integrated circuit increases, such as allowing the FET to be placed in the same way as other elements.

Figure 1 shows the square case, but the FET
The shape may be circular as long as it is spiral, or it may be any polygon.

[Brief explanation of the drawing]

Figure 1 shows an embodiment of the present invention, in which the FET is
It shows the electrode part viewed from a direction perpendicular to the plane of the chip where the FET is formed. 1... Gate electrode part A, 2... Source electrode part,
3... Gate electrode part B, 4... Drain electrode part 5
...Gate electrode part A pad, 6... Source electrode part pad, 7... Gate electrode part B pad, 8... Drain electrode pad.

Claims (1)

[Scope of utility model registration request] A source electrode and a drain electrode are arranged in parallel in a spiral shape, a gate electrode is provided in every gap between the two electrodes, and a connection pad for each electrode is connected to each electrode at the outermost periphery of the spiral. Field effect transistor formed at the end.