JPH11330254A - Semiconductor device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a switching element suitable for a FPGA switch, which is small and MENS type and to reduce the wiring connection resistance of a logic variable LSI. SOLUTION: A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to a semiconductor element, a plurality of switching elements for arbitrarily connecting two of signal wirings, and a switching driving part for selectively driving the switch element with external signal, are provided. Here, each switching element comprises first and second connection wirings 11 and 12 connected, respectively to differing two of the signal wirings, a mobile wiring 30 allocated next to the connection wirings 11 and 12, and a control wiring 20 allocated next to the mobile wiring 30, and the control wiring 20 is applied with a voltage for moving the mobile wiring 30 under coulomb force, so that connection is established between the mobile wiring 30 and the first and second connection wirings 11 and 12.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor device, and more particularly to a semiconductor device in which wiring is switched by Coulomb force to switch wiring connection of an integrated circuit element.

[0002]

BACKGROUND OF THE INVENTION Today, LSI by scaling of the transistor due to fine processing technology, etc., will be up to several million transistors are mounted on a single chip of a few cm ² square, large-scale computers, personal computers, home appliances, automobile It has been used everywhere, such as mobile phones. Digital LSI such as microprocessor, microcontroller, DSP etc.
I, memory such as DRAM, SRAM, flash, etc., A
From large-scale LSIs such as analog LSIs such as D and DA converters, to small-scale LSIs for customers
C (Application Specific IC).

[0003] Among ASICs, standard cells for customizing most mask layouts are used for customers who require a small chip area and a high-speed operation because medium-volume production is sufficient. In addition, a gate array that customizes the mask layout only for metal wiring, via contacts, etc. is used for customers who have a short manufacturing period and can produce small quantities at the expense of slightly increasing the chip size and operating speed somewhat. Have been. However, even a gate array is expensive because it is not in the order of thousands to tens of thousands, and takes several months to deliver. Therefore, it is not suitable for the prototype stage of the system or for the purpose of small-volume production.

Therefore, recently, after LSI production,
A device that can define logical connection information inside I, that is, a device that can be customized (PLD: Programmable Logic Devices)
), Products such as Field PLA and Field Programmable Gate Array (FPGA) are used. In these, the user describes the desired logic circuit in software,
Information such as logical connection information, a logic circuit, and a sequential circuit is written into the FPGA using an FPGA writer or the like and used. FPGA
And the like, the logical connection information and the like are realized by electrically or physically connecting or disconnecting signal wirings inside the FPGA.

FIG. 32 shows an example of a conventional FPGA. Cell library blocks (CLB1, CLB2, CLB3) having various logic circuits are arranged in a matrix in a chip, and signal wirings are arranged in the horizontal and vertical directions between these cell library blocks. ing. Switch elements as shown in FIG. 33 are arranged at the intersections of these wirings and between the wirings. In this example, a transistor is used for the switch element, and this transistor is
It is determined whether N or OFF, that is, information on the gate voltage of the transistor is stored in a memory that controls this.

[0006] In a conventional FPGA, the following three methods are employed as switch elements.

(1) Amorphous Si is sandwiched between metal wirings to be insulated as initials, a current flows between desired metal wirings, and this insulating film is broken and short-circuited (anti-fuse method). Short-circuited with poly-Si or the like and then cut with a laser or the like (fuse method).

(2) As shown in FIGS. 32 and 33, a normal MOS transistor is used as a switching element, and the potential state of the gate is stored in a memory. As the memory, there are a memory using a volatile DRAM, an SRAM, or the like, a memory using a nonvolatile EEPROM, and a memory using a ferroelectric memory (FRAM) has been proposed. For volatile ones, the logical connection and wiring connection information are stored in the FPGA every time they are used.
Reconfigurable Logic, which has the disadvantage of having to be taken inside, but the information can be changed many times
Can be On the other hand, in the case of the non-volatile type, the power supply is O
There is an advantage that the logical connection and the wiring connection information can be maintained even if the FF / ON is performed.

(3) A method in which the switch element itself is composed of an EEPROM or the like, and the ON / OFF information of the transistor is stored.

However, in the case of using a transistor as a switch element, the transistor is connected via all the wirings, and the ON resistance of each transistor is 3 V / 1 mA = 3 kΩ even if an ON current of 1 mA can flow.
And the wiring delay of the LSI becomes very large.

Even if the anti-fuse or fuse method is used, the resistance is as large as about 500Ω. If there are four anti-fuses in the wiring path, the resistance becomes 2 kΩ by itself, and the delay becomes 2 ns when the load capacitance is 1 pF. Very large. Further, in the fuse method, it takes too much time to write logical information because a laser is applied one by one.
As described above, in the conventionally proposed FPGA system, there is a problem that the resistance of the switch element is large and the wiring delay is very large, and the operation speed is 10 times (operating frequency is 1) as compared with the conventional gate array and the like. / 10) There was a problem about slow.

In the conventional FPGA, not only the above-mentioned signal wiring and switch elements but also arbitrary switches arranged in a matrix and ON / OF of the switches are provided.
A control circuit for selecting the F information and writing the information is required.
For this reason, first, the switching element itself is large,
Second, because the memory area for storing the ON / OFF information of the switch elements is large, and third, because the ratio of the control circuit and the control circuit wiring is large, the degree of integration of the FPGA is 1/1000 to 1/100 that of the general-purpose MPU. 100, and there is a problem that the chip size (chip cost) is larger than that of the gate array by one digit or more. In addition, there is a problem that the operation speed of the FPGA is further reduced due to the low integration. FIG. 34 collectively shows the types and problems of various FPGAs.

On the other hand, by bending, moving, or rotating the wiring with a force such as Coulomb force (electrostatic force) between metal wirings in a micro size, a micro motor,
A micro-electro-mechanical system using micro-machining technology that realizes micro-gear wheels, micro-grippers, etc.
tem: MEMS) has been proposed and prototyped. However, these MEMS have problems of fatigue and wear due to actual movement of an object although they are micro, and have problems in practical use. For this reason, a microsensor which does not need to be worn or a micromirror device having a small degree of wear as a display device has only been put to practical use.

Conventionally, as shown in FIG. 35, there has been proposed a device in which one end of a wiring in a beam portion is fixed and one end is turned on / off like a pull transistor by a Coulomb force. This switch is different from an electrical switch in that when it is OFF, the metal wiring
Since the F characteristic is better than that of the semiconductor transistor and the metal wirings are in direct contact with each other when the transistor is ON, the ON resistance is 50
It has a characteristic that is extremely small, about an order of mΩ, which is an order of magnitude smaller than that of a semiconductor transistor.

However, such a switch is called LS
To use in place of transistors constituting logic circuits such as NAND, NOR, and inverter in I, 1) fatigue and wear are imposed, and the number of switching is restricted. 2) Wiring does not bend unless Coulomb force is large due to stress. For this reason, it is necessary to make the wiring length of the switch beam extremely large with respect to the wiring thickness of the beam. As a result, there is a problem that the size of the switch is much larger than the size of the semiconductor transistor. In addition, due to the above-mentioned problems of fatigue and wear, and the problem of the switch size, it has been impossible to replace a semiconductor transistor constituting a conventional LSI logic circuit.

Another problem is that the conventional MEMS
, The dielectric constant is ε, the voltage applied between the wirings is V,
If the distance between wires is d, the Coulomb force F
However, F = (1/2) ε × V ² / d ² , and there is a problem that the object is difficult to move unless a voltage higher than the power supply voltage Vdd is applied as it is.

[0017]

As described above, the conventional FP
In a PLD such as a GA, compared to an ASIC such as a standard cell or a gate array, one LSI
While a user can directly implement a desired custom LSI in I units, there are disadvantages such as a large ON resistance of a switch element and a large control circuit, and the operation speed is 1/10 times and the integration degree is 1/1000 to 1 / 100 times, and the cost per chip also becomes 10 times.

Further, the conventional MEMS has a problem of fatigue and wear, and has a problem in practical use. When a MEMS type switch is used as a transistor constituting a logic gate, not only a problem of fatigue and wear but also a problem is caused. The switch element itself becomes large, and the operating power supply becomes larger than the current 5 V and 3.3 V, so that there is also a problem in reliability.

The present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide an FPG capable of rewriting logical connection information and wiring connection information after manufacturing a chip.
In a logic variable LSI such as A, a small MEMS type F
It is an object of the present invention to provide a semiconductor device capable of realizing a high-speed logic variable LSI with greatly reduced wiring connection resistance by realizing a switching element suitable for a PGA switch and using the same for connection / disconnection of wiring.

It is another object of the present invention to provide a semiconductor device capable of applying a high voltage, eliminating or reducing a control circuit section, and enabling high integration and low cost.

[0021]

(Structure) In order to solve the above-mentioned problem, the present invention employs the following structure.

(1) A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of wirings respectively connecting any two of these signal wirings A semiconductor device comprising a switch element and switch driving means for selectively driving the switch element by an external signal, wherein each of the switch elements is connected to two different ones of the plurality of signal lines. First and second connection wires respectively connected to the first and second connection wires;
Movable wiring arranged adjacent to the connecting wiring of
And a control wiring disposed adjacent to the movable wiring, wherein a predetermined voltage is applied to the control wiring in a switch element selected by the switch driving means, whereby the movable wiring is formed. Are moved by Coulomb force to connect between the movable wiring and the first connection wiring, and between the movable wiring and the second connection wiring, respectively.

(2) A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of wirings respectively connecting any two of these signal wirings A semiconductor device comprising a switch element and switch driving means for selectively driving the switch element by an external signal, wherein each of the switch elements is connected to two different ones of the plurality of signal lines. First and second connection wires respectively connected to the first and second connection wires;
Movable wiring arranged adjacent to the connecting wiring of
And a control wiring disposed adjacent to the movable wiring, wherein a predetermined voltage is applied to the control wiring in the switch element selected by the switch driving means, whereby the movable wiring is All of the control wiring, the movable wiring and the first connection wiring, and the movable wiring and the second connection wiring are caused by the attractive force of Coulomb force to act on the movable wiring. And the first connection wiring, and the movable wiring and the second connection wiring are connected, and the movable wiring and the control wiring are not connected. Features.

(3) A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of signal wirings respectively connecting any two of these signal wirings A semiconductor device comprising a switch element and switch driving means for selectively driving the switch element by an external signal, wherein each of the switch elements is connected to two different ones of the plurality of signal lines. First and second connection wires respectively connected to the first and second connection wires;
Movable wiring arranged adjacent to the connecting wiring of
A first and a second arranged adjacent to the movable wiring
A control element, and by applying a predetermined voltage to both the first and second control lines in the switch element selected by the switch driving means, the movable line and the first control line are controlled. Coulombs between the movable wiring, the movable wiring and the second control wiring, the movable wiring and the first connecting wiring, and the movable wiring and the second connecting wiring. The movable wiring is connected to the first connection wiring and the movable wiring and the second connection wiring are connected by the attraction of force, and the movable wiring is connected to the first control wiring. And between the movable wiring and the movable wiring and the second control wiring are not connected.

(4) A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of wirings respectively connecting any two of these signal wirings. A semiconductor device comprising a switch element and switch driving means for selectively driving the switch element by an external signal, wherein each of the switch elements is connected to two different ones of the plurality of signal lines. First and second connection wires respectively connected to the first and second connection wires;
Movable wiring arranged adjacent to the connecting wiring of
The first and second control wirings are arranged along the moving direction of the movable wiring with the movable wiring interposed therebetween, and the first and second control elements are selected by the switch driving means. A predetermined voltage is applied to either one of the control wirings to control the Coulomb force between the movable wiring and the first control wiring or between the movable wiring and the second control wiring. By the attraction of
The movable wiring and the first connection wiring, and the movable wiring and the second connection wiring are connected, and the movable wiring and the first control wiring are connected to each other. The present invention is characterized in that it is configured such that the possible wiring and the second control wiring are not connected.

(5) A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of wirings respectively connecting any two of these signal wirings A semiconductor device comprising a switch element and switch driving means for selectively driving the switch element by an external signal, wherein each of the switch elements is connected to two different ones of the plurality of signal lines. First and second connection wires respectively connected to the first and second connection wires;
Movable wiring arranged adjacent to the connecting wiring of
First and second control wirings which are disposed to face once in the moving direction of the movable wiring, and third and fourth control wirings which are disposed to face the other end in the moving direction of the movable wiring. A predetermined voltage is applied to either one of the first and second control wirings or one of the third and fourth control wirings in the switch element selected by the switch driving means. By applying an attractive force due to Coulomb force between the movable wiring and the first and second control wirings, or between the movable wiring and the third and fourth control wirings, The movable wiring is connected to the first connection wiring, and the movable wiring is connected to the second connection wiring, and the movable wiring is connected to the first and second wirings.
And the movable wiring and the third and fourth control wirings are not connected to each other.

(6) A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of signal wirings respectively connecting any two of these signal wirings A semiconductor device comprising a switch element and switch driving means for selectively driving the switch element by an external signal, wherein each of the switch elements is connected to two different ones of the plurality of signal lines. First and second connection wires respectively connected to the first and second connection wires;
A bendable wiring arranged adjacent to the connection wiring of
A control element disposed adjacent to the bendable wiring, and applying a predetermined voltage to the control wiring in a switch element selected by the switch driving means, thereby forming the bendable wiring and The bendable wiring is formed by applying attractive force due to Coulomb force to all of the control wiring, the bendable wiring and the first connection wiring, and between the bendable wiring and the second connection wiring. And the first connection wiring, and the bendable wiring and the second connection wiring are connected, and the bendable wiring and the control wiring are not connected. Features.

(7) A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of wirings respectively connecting any two of these signal wirings A semiconductor device comprising a switch element and switch driving means for selectively driving the switch element by an external signal, wherein each of the switch elements is connected to two different ones of the plurality of signal lines. First and second connection wires respectively connected to the first and second connection wires;
Rotatable wiring arranged adjacent to the connecting wiring of
And a control wiring disposed adjacent to the rotatable wiring, and applying a predetermined voltage to the control wiring in the switch element selected by the switch driving means, thereby forming the rotatable wiring and The at least one of the control wirings, the rotatable wiring and the first connection wiring, and the rotatable wiring and the second connection wiring are subjected to attractive force by Coulomb force to thereby generate the rotatable wiring. And the first connection wiring, and the rotatable wiring and the second connection wiring are connected, and the rotatable wiring and the control wiring are not connected. Features.

Here, preferred embodiments of the present invention include the following.

(A) In the above (1) to (7), 1024 or more switch elements are integrated.

(B) In the above (1) to (7), the connection wiring, the movable wiring, and the control wiring are Al, C
u, Ni or other metal wiring.

(C) In the above (1) to (7), a groove is formed in the insulating film so that the movable wiring moves only in a certain direction, and a part of the movable wiring is formed in the groove of the insulating film. Being embedded.

(D) In the above (1), (2) and (3), the wiring which can be moved only when a positive voltage equal to or higher than the power supply voltage is applied to the control wiring.

(E) In (3) above, it is possible to move only when a positive voltage equal to or higher than the power supply voltage is applied to the first control wiring and a negative voltage equal to or higher than the absolute value of the power supply voltage is applied to the second control wiring. The wiring moves.

(F) In the above (4), the total capacitance between the first and second connection wirings and the movable wiring is the capacitance between the first and second control wirings and the movable wiring. Be greater than the sum of

(G) In (5), the total capacitance between the first and second connection wirings and the movable wiring is the capacitance between the first and second control wirings and the movable wiring. And the third
And the sum of the capacitance between the fourth control wiring and the movable wiring.

(H) In the above (2), (4), (6), and (7), a plurality of first selection lines arranged in parallel in one direction of the semiconductor substrate;
A plurality of second selection lines arranged in parallel in a vertical direction with respect to the first selection line, and an arbitrary first and second selection line selected from the plurality of first and second selection lines And a logic gate disposed at the intersection of the first and second selection lines and taking a logical product with at least the first and second selection lines as inputs. The output is connected to a switch element arranged at the intersection of the first and second selection lines.

(I) In the above (3) and (5), a plurality of first control wirings are arranged in parallel in one direction of the semiconductor substrate.
Are arranged in parallel in the vertical direction with respect to the first control wiring, and the switch elements are arranged at the intersections of the first and second control wirings. And a plurality of openings for directly connecting the first and second control wirings to the outside of the semiconductor device.

(Operation) As described above, the conventional PLD,
In a logic variable LSI such as an FPGA, an FPLA, and a reconfigurable LSI, there is a problem that high-speed operation is difficult due to a large resistance of a transistor, an anti-fuse, and a fuse portion of a wiring connection portion. On the other hand, according to the present invention (claim 1), the first connection wiring and the second connection wiring at a desired position are connected by a Coulomb force via a movable metal wiring, whereby contact resistance is reduced. Can be reduced to a very small value (on the order of several Ω or less).
I can be realized. Further, in the application of the conventional MEMS, there have been problems of fatigue and wear. However, when used in a logic variable LSI, only one switching may be performed, and fatigue does not become a problem.

Further, the conventional MEMS switch aims at replacing the transistor of the logic circuit, and fixes one end of a curved wiring so as to be suitable for switching many times.
This is a method of turning on / off the switch by using the bending of the wiring. Since the wiring needs to be bent with a small Coulomb force, it is necessary to take a long wiring to be bent, and the switch becomes large.

On the other hand, according to the present invention (claim 2),
Although it is used only once, by applying a voltage to the control wiring, the movable wiring and all the wirings constituting the switch, that is, the movable wiring and the control wiring, the movable wiring Attraction by Coulomb force acts on all of the first connection wiring and between the movable wiring and the second connection wiring, eliminating a portion where repulsion acts, and reducing the voltage applied to the control wiring to a small value. be able to. In other words, the wiring can be easily moved, and it is possible to operate the wiring that can be moved properly even if the distance between the wirings varies, as well as the downsizing of the element and the reduction of the contact resistance. become.

Further, even if the movable wiring moves, an extra distance is provided between the movable wiring and the control wiring so that the movable wiring and the control wiring do not come into contact with each other. By this effect, it is possible to prevent DC current from flowing through the control wiring before and after the movement, to reduce power consumption when writing wiring connection information, and to simultaneously write many switches. And a drop in the writing potential due to the flow of the DC current can be prevented. Further, by forming a groove into which a part of the movable wiring is fitted in the insulating film, the movable wiring can be stably moved only in a certain direction.

Further, in the present invention (claim 3), in addition to the above-mentioned configuration (claim 2), one control wiring is divided into two first and second control wirings. Thereby, only when a voltage is applied to both the first and second control wires,
The movable wiring can be moved. This is X
When the switch element of the present invention is arranged at the intersection of the control wiring arranged in the axial direction and the control wiring arranged in the Y-axis direction, an arbitrary switch element is selected. In addition, the necessity of inputting the result of the logical product of the signals of the X-axis control wiring and the Y-axis control wiring at the intersections of these wirings as the signal of the switch element control wiring is eliminated. The Y-axis control wiring can be directly used as the signal of the first and second control wirings.

In this case, first, a logic circuit for generating a logical product can be dispensed with, and there is an effect of increasing the degree of integration.
Second, a switch element using Coulomb force requires a high-voltage operation, so if a logic circuit can be eliminated, a large-sized high-voltage transistor can be eliminated. In addition, this is because when the difference between the power supply voltage and the voltage applied to the control wiring is large and a transistor with a high withstand voltage cannot be formed,
It is even more effective.

According to the present invention (claim 4), the first and second control wirings are provided on opposite sides of the movable wiring, and the first control wiring or the second control wiring is provided. By applying a voltage to the control wiring, it is possible to move between the applied first or second control wiring and the movable wiring by applying an attractive force due to a larger Coulomb force than between the other nodes. Wiring can be easily moved. Therefore, similarly to (claim 2), it is possible not only to reduce the size of the element and to lower the contact resistance, but also to operate the movable wiring properly even if the distance between the wirings varies. Become.

This is because the total capacitance between the first and second connection wirings and the movable wiring is larger than the total capacitance between the first and second control wirings and the movable wiring. Compared to the capacitance between the first or second control wiring to which the voltage is applied and the movable wiring, the first or second control wiring to which no voltage is applied, the first connection wiring, and the second connection wiring. The total capacitance between the connection wiring and the movable wiring increases. Therefore, when a voltage is applied to the first or second control wiring, a large potential difference is generated between the first or second control wiring to which the voltage is applied and the movable wiring. A small potential difference is generated between each of the first or second control wiring, the first connection wiring, and the second connection wiring that are not performed and the movable wiring. Accordingly, the attractive force between the first or second control wiring to which the voltage is applied and the movable wiring is increased, and the movement of the movable wiring is facilitated.

Further, the first and second connection wirings are movable in the direction in which the first and second control wirings and the movable wiring face each other, ie, in the direction in which the movable wiring moves. By setting the direction facing to the vertical direction,
With respect to the movement of the movable wiring, the attractive force between the first connection wiring and the movable wiring and the attractive force between the second connection wiring and the movable wiring are not affected.

Further, even if the movable wiring moves, an extra distance is provided between the movable wiring and the first and second control wirings so that the movable wiring and the first and second control wirings are separated from each other. 2 can be prevented from coming into contact with each other, and this effect allows the first control wiring to be connected to the first control wiring both before and after the movement.
It is possible to prevent the C current from flowing, reduce power consumption when writing wiring connection information, perform simultaneous writing of a large number of switches, and prevent a drop in the writing potential due to the flow of the DC current. Furthermore, by forming a groove in the movable wiring and the insulating film, it is possible to move stably only in a certain direction.

Further, the present invention (Claim 5) has a configuration in which the above (Claim 3) and (Claim 4) are combined, so that in addition to the effect of (Claim 3), (Claim 4) It has the effect of adding the effects of

Further, according to the present invention (claim 6), the first connection wiring and the second connection wiring at the desired position are connected by a Coulomb force via a bendable wiring, thereby providing contact. The resistance can be reduced to a very small value (about several Ω or less)
The same effect as the above (claim 2) can be obtained.

According to the present invention (claim 7), the first connection wiring and the second connection wiring at a desired position are connected by a Coulomb force via a rotatable wiring, whereby contact is achieved. The resistance can be reduced to a very small value (about several Ω or less)
The same effect as the above (claim 2) can be obtained.

The above (h) is the present invention (claim 2,
4, 6, 7) shows a matrix configuration of the logic variable LSI of the present invention applicable to the present invention, and by this configuration, any switch element of the present invention arranged in a matrix can be selected.

The above (i) is the present invention (claim 3,
5 shows a matrix configuration of a logic variable LSI of the present invention applicable to 5). With this configuration, any switch device of the present invention arranged in a matrix can be directly passed from an external pad without passing a logic circuit device. You can select and apply a voltage. This makes it possible to configure a logic variable LSI even when the potential required for moving the first movable wiring is high and the transistor breakdown voltage inside the LSI is low. Further, an extra control circuit can be completely eliminated.

[0054]

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

(First Embodiment) FIG. 1 shows a first embodiment of the present invention.
1 is for describing the semiconductor device according to the embodiment, and shows a switch element applicable to a logic variable LSI.

Although not shown in the figure, nM
After forming semiconductor elements such as OS and pMOS, the elements are connected to form a metal wiring layer forming a logic circuit. The switch element is formed of any one of the metal wiring layers, and is structurally constituted by two signal wirings (connection wirings) 11 and 12 and one control wiring 20. It is composed of one movable wiring 30. As will be described later, a gap is formed between the movable wiring 30 and the insulating film covering the same, and a groove is formed in the insulating film so that the movable wiring 30 can move in a certain direction. , A structure in which the lower part of the movable wiring 30 is fitted in an inverted convex shape thereon.

After the LSI is manufactured, the movable wiring 30
It is completely separated from other nodes. A plurality of switch elements of the present embodiment are provided on an LSI chip, an arbitrary switch element is selected from the plurality of provided switch elements, and a voltage is applied to the control wiring 20 of the selected switch element. As a result, an attractive force due to Coulomb force acts between the control wiring 20 and the movable wiring 30, and the movable wiring 30 moves up and down in the drawing, and the left connecting wiring 1 is moved.
1 and the movable wiring 30 and the right connecting wiring 12 and the movable wiring 30 are both in contact with each other, and as a result, the left connecting wiring 11 and the right connecting wiring 12 are conducted.

Between a number of connection wirings on an LSI chip,
By arranging this switch element and applying a voltage to the control wiring 20 of an arbitrary switch element, the wiring connection inside the LSI can be changed after manufacturing the LSI, and a desired logic circuit can be formed by this switching. Becomes possible. As the metal wiring material, Al, Cu, Ni, or the like can be used, and a material similar to that of a normal LSI wiring can be used.

The features of the present embodiment are as follows. Conventional PLD, FPGA, FPLA, Reco
In a logic variable LSI such as an nfigurable LSI, the resistance of a transistor, an anti-fuse, and a fuse part of a wiring connection part is large, and it is difficult to operate at high speed. On the other hand, according to the present embodiment, the metal wiring signal at the desired position and the metal wiring signal are connected by the Coulomb force via the movable metal wiring, so that the contact resistance is reduced to a very small value (several Ω or less). ), And a high-speed logic variable LSI can be realized. Further, in the application of the conventional MEMS, there have been problems of fatigue and wear. However, when used in a logic variable LSI, one switching may be performed, and one switching from OFF to ON as in the present embodiment. , This fatigue problem goes away.

Also, with the aim of replacing the transistor of the conventional logic circuit, one end of a curved wiring is fixed and a switch is turned on / off by utilizing the bending of the wiring so as to be suitable for a large number of switching operations. 1) Compared to the MEMS switch, 1) since the bending is not used, the size of the switch element can be reduced. 2) As shown in FIG. 1, a voltage is applied to the control wiring 20, and the connection wirings 11 and 12 are also arranged in the direction (lower side in the figure) in which the movable wiring 30 moves. When a voltage is applied to the wiring 20, all wirings, that is, the movable wiring 30 and the control wiring 20
, Movable wiring 30 and two connection wirings 11 and 12
There is no part where the repulsive force works because the attractive force by the Coulomb force works in everything in between.

As a result, the voltage applied to the control wiring 20 can be reduced, the wiring can be easily moved, or the surface on which Coulomb force acts can be reduced, and the size of the switch element can be reduced. These not only reduce the resistance at the time of contact, but also ensure that the movable wiring operates even if the distance between the wirings varies.

For example, the control wiring 20 is set to 0 V, the two connection wirings 11 and 12 are set to 0 V, and the movable wiring 30 is set to 0 V.
When the voltage V is applied to the control wiring 20 from the state shown in FIG. 2, the capacitance between the movable wiring 30 and the control wiring 20 becomes Cmc,
The distance is Tmc, the movable wiring 30 and the connecting wiring 11,
12 is Cms and the distance is Tms, the potential difference Vm between the movable wiring 30 and the connecting wirings 11 and 12
s is Vms = Cmc / (2Cms + Cmc) V, and the potential difference Vmc between the movable wiring 30 and the control wiring 20 is Vmc = 2Cms / (2Cms + Cmc).

The force acting between the movable wiring 30 and the connecting wirings 11 and 12 is an attractive force of Fms = (1/2) Vms ² / Tms ² , and similarly, the movable wiring 30 and the control wiring The force acting between the wirings 20 is the attractive force of Fmc = (1/2) Vmc ² / Tmc ² , and the total force is Ftotal = 2 × (1/2) Vms ² / Tms ² + (1/2) Vm
The attractive force is c ² / Tmc ² , and the movable wiring 30 moves downward in the figure regardless of the conditions of the capacity and the distance, that is, regardless of the variation or the like. Therefore, it is possible to move efficiently as much as there is no repulsive force that refuse to move, and it is possible to reduce the applied voltage.

Further, as shown in FIG.
Even if 0 moves, the movable wiring 30 and the control wiring 20
By providing an extra distance therebetween, the movable wiring 30 and the control wiring 20 can be prevented from coming into contact with each other. With this effect, the DC current flows through the control wiring 20 both before and after the movement. Can be prevented, the power consumption at the time of writing the wiring connection information can be reduced, the simultaneous writing of a large number of switches can be performed, and the drop of the writing potential due to the flow of the DC current can be prevented.

(Second Embodiment) FIG. 2 shows a second embodiment of the present invention.
FIG. 7 is a diagram illustrating a switch element applicable to a logic variable LSI, for describing the first embodiment. The same parts as those in FIG. 1 are denoted by the same reference numerals, and detailed description thereof will be omitted.

The present embodiment is different from the first embodiment shown in FIG. 1 in that the movable wiring 30 is not floating and is connected to a potential supply terminal 40 for supplying a potential of 0 V or -V. Is a point. Here, the movable wiring 30 and the potential supply terminal 40 are connected by a thin wiring that can be cut.

In the present embodiment, unlike FIG. 1, for example, 0 V is applied to the potential supply terminal 40 and the control line 20 is set to 0.
When the voltage V is applied to the control wiring 20 from the state of 0 V of the two signal wirings 11 and 12 and 0 V of the movable wiring 30, the capacitance between the movable wiring 30 and the control wiring 20 becomes Cmc. If the distance is Tmc, the capacitance between the movable wiring 30 and the connection wirings 11 and 12 is Cms, and the distance is Tms, the potential difference Vms between the movable wiring 30 and the connection wirings 11 and 12 is 0 V, The potential difference between the movable wiring 30 and the control wiring 20 is Vmc = V.

The force acting between the movable wiring 30 and the connecting wirings 11 and 12 is Fms = 0, and the movable wiring 3
Force acting between 0 and controller wires 20 becomes the attraction ^{Fmc = (1/2) V 2 /} Tmc 2, total force becomes attractive force ^{Ftotal = (1/2) V 2 /} Tmc 2, Coulomb force Since the voltage increases in proportion to the square of the voltage, the total attractive force in FIG. 2 is larger than that in FIG. 1 under normal conditions.

The potential supply terminal 40 and the movable wiring 30 are disconnected by the movement of the movable wiring 30 by this attractive force, and the signal wirings are connected to each other via the movable wiring. However, a cutting force is required to increase the attractive force, and depending on how the cutting portion can be weakened, FIGS.
Are easy to move.

When a potential of -V is applied to the potential supply terminal, the force acting between the movable wiring and the signal wiring is Fms = (1/2) V ² / Tms ² between the movable wiring and the control signal wiring. work force becomes the attraction of ^{Fmc = (1/2) V 2 /} Tmc 2 , the sum of forces Ftotal = 2 × (1/2) V 2 / Tms 2 + (1/2) V 2 / T
Attraction of mc ² and maximum.

(Third Embodiment) FIG. 3 shows a third embodiment of the present invention.
FIG. 7 is a diagram illustrating a switch element applicable to a logic variable LSI, for describing the first embodiment. The same parts as those in FIG. 1 are denoted by the same reference numerals, and detailed description thereof will be omitted.

This embodiment is different from the first embodiment shown in FIG. 1 in that the control wiring is a control wiring 21 for the X-axis.
And the control wiring 22 for the Y-axis. As a result, the X-axis and Y-axis control wires 21,
The movable wiring 30 can be moved only when a voltage is applied to both of them. This is because, when a plurality of control signal lines arranged in the X-axis direction and a plurality of control signal lines arranged in the Y-axis direction are used and the switch element of the present invention is arranged at the intersection thereof, the embodiment of FIG. There are the following effects as compared with the embodiment.

In the embodiment of FIG. 1 for selecting an arbitrary switch element, the control signal line of the X axis and the Y
Although it is necessary to input the result of the logical product of the signals of the control signal lines of the axis as the signal of the control wiring of the switch element, according to the embodiment of FIG. 3, the control signal line of the X axis and the Y axis Are directly used as the signal of the control wiring 21 for the X-axis and the signal of the control wiring 22 for the Y-axis,
A logic circuit for generating a logical product can be dispensed with, and there is an effect of increasing the degree of integration. Second, since the switch element of the present invention that uses Coulomb force requires high-voltage operation, if a logic circuit can be eliminated, a large-sized high-voltage transistor can be eliminated. This is even more effective when the voltage between the power supply voltage and the voltage applied to the control signal wiring is so large that a withstand voltage transistor cannot be formed.

(Modification of Third Embodiment) FIG.
Although the structure is the same in terms of equivalent circuit, the following describes modified examples in which the operating condition is changed by changing the contact area and distance between the wirings and the operating method is changed. In the third embodiment,
The movable wiring 30 can be moved only when a voltage is applied to both the X-axis and Y-axis control wirings 21 and 22. This means that the capacitance between the movable wiring 30 and the control wirings 21 and 22 is Cmc, the distance is Tmc, and the capacitance between the movable wiring 30 and the connection wirings 11 and 12 is Cmc.
When ms and the distance are Tms, the condition is satisfied only under the condition of Cms> Cmc.

In short, when Cms is large, when a voltage V is applied to the control wiring 21 or 22 for the X-axis or the Y-axis, a potential close to V is applied to the movable wiring 30 and the voltage. It extends between the control wires 21 or 22. Therefore, when a voltage is applied to both of the two control wires 21 and 22, the total attractive force increases. Movable wiring 30
When the attractive force of the threshold value at which the voltage moves is between the attractive force when a voltage is applied to one control wiring and the attractive force when a voltage is applied to two control wires, the operation as in the third embodiment is performed. Will be possible. That is, to realize the operation of the third embodiment,
It is necessary to design the switch so that Cms> Cmc.

On the contrary, under the condition of Cms <Cmc, as shown in FIG. 4, when a voltage is applied to only one control wiring, the movable wiring 30 moves. To put it simply, when Cmc is large, if a voltage is applied to the two control wires 21 and 22, conversely,
The potential of the movable wiring 30 becomes too high, and the Coulomb force between the movable wiring 30 and the control wirings 21 and 22 becomes smaller than when a voltage is applied to one control wiring. That's why.

As shown in FIG. 4, one of the X-axis and Y-axis control wirings 21 and 22 has a potential of + V and the other has a potential of -V.
V potential is applied, and a large number of control wirings 21 for the X axis are applied.
And selecting any one of the switch elements of the present invention disposed at the intersection of the matrix of a large number of control wires 22 for the Y-axis,
The movable wiring 30 can be moved with a strong Coulomb force. This is because the potential of the movable wiring 30 is offset to 0 V by the application of V and -V, and a large wasteful Coulomb force works in proportion to twice V ² .

For example, this can be easily realized by controlling the X-axis control wiring 21 at 0 V when not selected and V when selected, and controlling the Y-axis control wiring 22 at 0 V when not selected and -V when selected. it can. In this method, the X-axis control signal line and the Y-axis control signal line can be directly input to the X-axis control wiring 21 and the Y-axis control wiring 22 to generate a logical product. Since a logic circuit is not required, a negative potential can be easily used.

(Fourth Embodiment) FIG. 5 shows a fourth embodiment of the present invention.
FIG. 7 is a diagram illustrating a switch element applicable to a logic variable LSI, for describing the first embodiment. 2 and 3 are denoted by the same reference numerals, and detailed description thereof will be omitted.

This embodiment is a combination of the embodiments shown in FIGS. 2 and 3, and the control wiring is divided into an X-axis control wiring 21 and a Y-axis control wiring 22, which are movable. The wiring 20 is not floating, and is connected to a potential supply terminal 40 that supplies a potential of 0 V or −V. Here, the movable wiring 30 and the potential supply terminal 40 are connected by a thin wiring that can be cut. As the effect, the effects of both the embodiment of FIGS. 2 and 3 can be obtained.

FIG. 6 shows an example of operating conditions of the present embodiment.
As in the second embodiment, by setting the potential supply terminal to 0V, the voltage V applied to the control wirings 21 and 22 is set.
Represents the control wirings 21 and 22 and the movable wiring 3
Since the voltage is applied between zero, a large Coulomb force can be obtained, and the movable wiring 30 can be moved only when a voltage is applied to the two control wirings 21 and 22 without depending on the capacitance relation and the shape relation. . Further, as shown in FIG. 6, the movable wiring 30 can be moved even if + V and + V and + V and −V are applied to the two control wirings 21 and 22 while changing the voltage. Similarly, since the voltage is also shifted at -V and -V voltages, as a result, there is an effect that the degree of freedom on the control side of the switch element is increased.

(Fifth Embodiment) FIG. 7 shows a fifth embodiment of the present invention.
FIG. 7 is a diagram illustrating a switch element applicable to a logic variable LSI, for describing the first embodiment. In the drawings, 111 and 112 indicate connection wirings, 121 and 122 indicate control wirings, and 130 indicates a movable wiring. In the present embodiment, the control wiring 121 is disposed on one end of the movable wiring 130 in the moving direction of the wiring 130, and the control wiring 122 is disposed on the other end.

The main effects are the same as those of the embodiment shown in FIGS. The difference is that in the embodiments of FIGS. 1 and 2, the number of times of switching is only one, but in the present embodiment, switching can be performed many times. In this case, since the fatigue and wear occur, the number of times of switching is limited, but there is no problem in applications such as FPGA.

Here, switching by moving an object using Coulomb force cannot be realized with a simple configuration. FIG. 36 shows a conventional case where a conductor is arranged between an electrode 0 and an electrode 1. A voltage of 0 V was applied to the electrode 0 and V [V] was applied to the electrode 1, and the Coulomb force between each electrode and the conductor was calculated. At this time, the surface area between the electrode 0 and the conductor is S0, the distance is d0, and the surface area between the electrode 1 and the conductor is S1,
When the distance is d1, the capacitance between the electrode 0 and the conductor is C0 =
εS0 / d0, the capacitance between the electrode 1 and the conductor is C1 = εS1 /
Since it is d1, the potential of the conductor is V01 = (d0S1) / (d1S0 + d0S1).

Therefore, the attractive force acting between the electrode 0 and the conductor is: F0 = ε / 2 × (S0 × S1 ² ) × V ² / (d1S0 +
d0S1) ^2, and the attractive force acting between the electrodes 1 and the conductors, F1 = ε / 2 × ( S1 × S0 2) × V 2 / (d1S0 +
d0S1) ² . When S1 = S0 = S, the two equations are equal, and F = ε / 2 × S × V ² / (d1 + d0) ² .

In the case where the distance between the electrode 0 and the electrode 1 is constant, if this distance is d, d = d0 + d1, and as a result, F = 2 × ε × S × V ² / d ² = constant. This result indicates that the Coulomb force with the electrodes on both sides of the conductor is the same and constant, no matter where the conductor is, and that the conductor does not move when voltage is applied.

Returning to FIG. 7 of the present embodiment, in order to solve the immovable problem shown in FIG. 36 and move the movable wiring 130, two controls are provided above and below the movable wiring 130. Wirings 121 and 122 are provided, connection wirings 111 and 112 are provided on the left and right, and a movable wiring 13 is provided.
0 and the movable wiring 130 and the connecting wirings 111 and 112 as compared with the capacitance Cmc between the control wirings 121 and 122.
A configuration is adopted in which the adjacent area between the movable wiring 130 and the connecting wirings 111 and 112 is increased so that the capacitance Cms between them has a large value.

This is, for example, the case where two connection wirings 111,
A potential of 0 [V] is supplied to the
When 0 [V] is supplied to 1 and V [V] is supplied to the lower control wiring 122, the potential of the movable wiring 130 becomes 0V.
The value of Cmc / (2Cmc + 2Cms) × V, which is smaller than (1/2) V [V], because the capacitance adjacent to is large,
The potential Vmc between the movable wiring 130 and the lower control wiring 122 is movable and the upper control wiring 121 is movable.
Potential between: movable wiring 130 and left and right connection wiring 1
It takes a value larger than the potential between 11 and 112, and as a result, the total Coulomb force becomes a downward attractive force, and the movable wiring 130 moves downward.

Similarly, for example, a potential of 0 [V] is supplied to the two connection wires 111 and 112, 0 [V] to the lower control wire 122, and to the upper control wire 121. V
When a potential of [V] is supplied, the total Coulomb force becomes an upward attractive force, and the movable wiring 130 moves upward. In addition, according to the principle of FIG.
The total amount of the Coulomb force in the horizontal direction occupying most of the space between the connection wirings 111 and 112 is canceled out to zero, and the movable wiring 130 is easily moved.

(Sixth Embodiment) FIG. 8 shows a sixth embodiment of the present invention.
FIG. 7 is a diagram illustrating a switch element applicable to a logic variable LSI, for describing the first embodiment. The same parts as those in FIG. 7 are denoted by the same reference numerals, and detailed description thereof will be omitted.

This embodiment is different from the fifth embodiment shown in FIG. 7 in that the control lines 121 and 122 are
b, 121b and 122a, 122b. The principle of movement and the main effects are shown in FIG.
This is the same as the embodiment.

(Seventh Embodiment) FIG. 9 shows a ninth embodiment of the present invention.
FIG. 1 is a cross-sectional view of a switch element applicable to a logic variable LSI for explaining the embodiment of FIG. This cross-sectional view shows the shapes of the movable wiring and the insulating film, and can be applied to all the inventions of FIGS. 1 to 8 and also to the other latter inventions.

In FIG. 9A, a concave rail-shaped groove 251 is formed in the insulating film 250, and a movable wiring 230 is fitted thereon with a downwardly convex mold. With this rail configuration, the movable wiring 230 can be stably moved only in the front and back directions on the paper. The space between the insulating film 250 and the movable wiring 230 is filled with air, vacuum (low pressure), gas, or liquid 260. After completion of the LSI chip, these are filled and packed. .

In FIGS. 9B and 9C, the movable wiring 230 completely fits into the groove 251 of the insulating film 250 like a keyhole, and restricts the movement in the left, right, up, and down directions. Therefore, in this case, in addition to the effect of FIG.

(Eighth Embodiment) FIG. 10 is a diagram for explaining an eighth embodiment of the present invention.
FIG. 6 is a diagram showing a part of a switch element applicable to the present invention. This is a cross-sectional view of a plane cut along the X-axis-Y-axis plane including both the insulating film 250 in which the groove 251 is formed and the protruding portion below the movable wiring 230. It can be applied to all inventions, and also to other later inventions.

When the movable wiring 230 is in the upper or lower position, the lower insulating film 250 and the upper movable wiring 2
When the distance between the movable wirings 30 is long and the movable wiring 230 is located near the center, the distance between the lower insulating film 250 and the upper movable wiring 230 is reduced. Accordingly, the movable wiring 230 is stable when it is on the upper or lower side, is hard to move up and down when the Coulomb force does not work, and is movable only when the Coulomb force works on or lower. 230 can be moved. Thus, the influence of external gravitational acceleration or the like such as the falling of the semiconductor device can be eliminated, and the stability of the device can be improved.

(Ninth Embodiment) FIG. 11 and FIG.
FIG. 19 is a diagram for explaining a ninth embodiment of the present invention and showing a switch element applicable to a logic variable LSI.
(A) is a sectional view, and (b) is a plan view.

Although FIGS. 1 to 10 show examples of the planar switch element, this example shows an example in which the switch element is configured in a three-dimensional configuration. In FIG. 11, the control wiring 320 is formed of the upper metal 2, the two connection wirings 311 and 312 are formed of the lower metal 1, and the movable wiring 330 is formed of the metal 1 and the metal 2, The contact is formed.

In FIG. 12, the control wiring 320 is formed of the upper metal 2 and has two connection wirings 311, 31.
Reference numeral 2 denotes one formed of the lower metal 1 and the other formed of the upper metal 2, and the movable wiring 330 is formed by the metal 1 and the metal 2 and their contacts.

By configuring the switch element of the present invention by using a plurality of wiring layers in this way, in addition to the effects of the above-described embodiments, the element becomes smaller and the degree of freedom of wiring increases. For example, in FIG. Is suitable when the metal 1 is mainly used as the connection wiring and the metal 2 is mainly used as the control wiring. In FIG. 12, the X-axis direction signal line (connection wiring) made of the metal 1 is used. This is suitable for the case where a signal line (connection wiring) in the Y-axis direction formed of the metal 2 and the metal 2 is connected or disconnected by the switch element of the present invention disposed at the intersection. An example in which this example is turned upside down or an example in which three or more layers are formed can be easily realized.

(Tenth Embodiment) FIGS. 13, 14, and 15 are views for explaining a tenth embodiment of the present invention, and are diagrams showing switch elements applicable to a logic variable LSI. The present embodiment is different from the above-described embodiments in that switching is not performed by moving a movable wiring, but is performed by rotating a rotatable wiring based on the principle of leverage. In the figure, 411 and 412 are connection wirings, 420 is a control wiring, 430 is a rotatable wiring, 440 is a voltage supply terminal, and 460 is a shaft groove.

FIG. 13 shows a method in which the rotatable wiring 430 is floating, FIG. 14 shows a method in which the rotatable wiring 430 is connected to the voltage supply terminal 440 by a thin wiring, and a method in which the wiring is cut by Coulomb force, and FIG. In this example, a thin wiring 430 is connected to the voltage supply terminal 440 by a thin wiring, and the thin wiring is bent and used by Coulomb force. In each case, one end of the rotatable wiring 430 is fitted into the groove 460 of the insulating film so that the wiring 430 can only perform a rotating operation.

In these examples, the connection wirings 411 and 412 are farther than the control wiring 420 based on the rotation axis. However, the relationship may be reversed. However, FIGS.
As shown in (1), it is more effective to cut or bend a thinner wiring than the control wiring 420 based on the rotation axis.

(Eleventh Embodiment) FIGS. 16 and 17
Is for explaining an eleventh embodiment of the present invention, and shows a conventional switch element applicable to a logic variable LSI. In the figure, 511 and 512 are connection wirings, and 521 and 5
Reference numerals 22 and 523 denote control wirings, and reference numeral 530 denotes a bendable wiring.

The element structure includes a bendable wiring 530.
This is a method in which both sides are fixed and the bendable wiring length is made longer than the distance between the two fixed points as an initial, thereby stabilizing either up or down. This switch configuration is the same as that of a conventional MEMS switch, and it is difficult to use it in place of a transistor constituting a conventional logic circuit because of problems of fatigue and wear. When the present invention is applied to a logic variable LSI that can change the logic connection information and the wiring connection later, the number of times of switching may be finite, and there is a sufficient effect. However, problems such as an increase in element size remain.

FIG. 16 shows an example of a one-time switch in which control wirings 521 and 522 are arranged on one side of a bendable wiring 530. FIG. 17 shows an example of a switch in which a control wiring 523 is arranged on the other side of the bendable wiring 530 in addition to the configuration of FIG. In addition, a wire that bends or bends is used as one end of a signal wire.
MEMS connected to other signal wiring by bending, bending, etc.
Although a switch is conceivable, in order to realize a bend, a bend, or the like, a thin and long wiring is required, so that there is a problem that a resistance between signal wirings increases, which is not practical.

(Twelfth Embodiment) FIG. 18 to FIG.
This is for explaining the twelfth embodiment of the present invention, and shows a circuit configuration example including a switch element applicable to a logic variable LSI.

In the example of FIG. 18, a plurality of X-signal lines (connection wiring) and a plurality of X-selection lines (control wiring) are provided in the X-axis direction, and Y-signal lines (connection wiring) are provided in the Y-axis direction. ), A plurality of Y-selection lines (control wirings) are arranged, and switch elements of the type having two control wirings shown in FIGS. Show a part.

With this configuration, any X-signal line and Y-
The intersections of the signal lines can be connected and disconnected. As a result, first, a logic circuit for generating a logical product can be dispensed with, and there is an effect of increasing the degree of integration. Second, the switch element of the present invention that uses Coulomb force requires a high-voltage operation. If a logic circuit can be eliminated, a large-sized high-voltage transistor can be eliminated. This is even more effective when a transistor having a large withstand voltage cannot be formed because the difference between the power supply voltage and the voltage applied to the control wiring is large.

In the example of FIG. 19, a plurality of X-signal lines (connection lines) and a plurality of X-selection lines (control lines) are provided in the X-axis direction, and Y-signal lines (connection lines) are provided in the Y-axis direction. Wiring) and a plurality of Y-selection lines (control wirings) are arranged, and switch elements of a type having one control wiring as shown in FIGS. Show a part. With this configuration, by connecting the switch element of the present invention to the intersection via a logic circuit that implements a logical product, the intersection of any X-signal line and Y-signal line can be connected or disconnected.

In the example shown in FIG. 20, a plurality of X-signal lines (connection lines) and a plurality of X-selection lines (control lines) are provided in the X-axis direction, and Y-signal lines (connection lines) are provided in the Y-axis direction. Wiring) and a plurality of Y-selection lines (control wirings) are arranged, and switch elements of a type having one control wiring as shown in FIGS. Is shown.

With this configuration, a logic circuit for realizing a logical product is provided at the intersection, and a desired X-selection line and Y-selection line are input to the logic circuit from the X-selection line group and the Y-selection line group. By doing so, for example, in the case of N X-selection lines, there is a case of 2 ^N powers, and if they are used separately, the number of selection line groups can be significantly reduced to N / 2 ^N times as compared with FIG. it can.

In the example of FIG. 21, a plurality of X-signal lines (connection wires) are provided in the X-axis direction, an X-selection line (control wire) is provided in the Y-axis direction, and a Y-selection line is provided in the X-axis direction. (Control wiring)
A part of a case where a plurality of switch elements of the type having two control wirings shown in FIGS. 3 and 5 are arranged in a matrix at these intersections is shown. With this configuration, an arbitrary X-signal line and an X-signal line can be connected and disconnected.

In the example of FIG. 22, a plurality of Y-signal lines (connection wires) are provided in the Y-axis direction, and an X-selection line (control wire) is provided in the Y-axis direction, and a Y-selection wire is provided in the X-axis direction. Wire (control wiring)
A plurality of switches are arranged, and a part of the case where switches of the type having two control signal wirings shown in FIGS. With this configuration, an arbitrary Y-signal line can be connected and disconnected.

By combining the above-described configurations of FIGS. 18, 21 and 22, any X, Y-signal line can be extended or shortened, or the X and Y axes can be changed in the middle, and the wiring can be freely changed. Switching becomes possible.

In the example of FIG. 23, the X-axis or the Y-axis
(Y) -Signal line (connection wiring), X (Y)-in the Y-axis direction
A plurality of selection lines (control wiring) and a plurality of Y (X) -selection lines (control wiring) are provided in the X-axis direction, and one of the control wirings shown in FIGS. A part of a case where switches of a certain system are arranged in a matrix is shown. With this configuration, if the switch is connected to the switch element of the present invention via a logic circuit that implements a logical product at the intersection, the length of the interconnection can be changed by connecting or disconnecting the intersection of any X (Y) signal wiring. be able to.

FIG. 24 is similar to FIG. 23 except that an X-selection line group and a Y-selection line group are provided.
This configuration, 2 X- and select line group Y- and desired X- selected line from the selected line group, by entering Y- selection line to the logic circuit, is for example N number of X- selection lines ^N Since there is a power case, the number of selection line groups can be greatly reduced to N / 2 ^N times as compared with FIG. 23 when used separately.

(Thirteenth Embodiment) FIG. 25 is a view for explaining a thirteenth embodiment of the present invention.
FIG. 2 is a diagram illustrating a block or chip configuration in which a plurality of switch elements applicable to an SI are provided.

In a chip (or a block), a plurality of logic circuit blocks and switch elements as shown in FIGS. 1 to 17 are arranged around the logic circuit blocks. , Y-axis selection lines are selected, and their intersections are configured as shown in FIGS. 18 to 24 to switch ON / OFF the switch elements. A register is provided inside the X-line controller and the Y-line controller. If ON / OFF is sequentially written to each matrix-like switch element by an external signal,
Even after the LSI is manufactured, the user can obtain a desired logic LSI.

When the voltage applied to the switch using the Coulomb force is high, the X-line controller and the Y-line controller circuit may be formed using a transistor with a high withstand voltage. Even in this case, these circuits at the end of the chip or block need not be so large in overall size, even if the degree of integration is low.

(Fourteenth Embodiment) FIG. 26 is a view for explaining a fourteenth embodiment of the present invention.
FIG. 2 is a diagram illustrating a block or chip configuration in which a plurality of switch elements applicable to an SI are provided.

In a chip (or block), a plurality of X-selection lines and a plurality of Y-selection lines, and a switch element having two control wirings as shown in FIGS. 18, FIG. 21, FIG.
In the circuit configuration of FIG. 22, the X, Y-selection lines and the switch elements are connected. Pads for connection to the outside of the chip are arranged in a matrix on the entire area where a plurality of X-selection lines and a plurality of Y-selection lines are arranged. Any one of a plurality of X-select lines and a plurality of Y-select lines
They are connected in a one-to-one correspondence via contacts indicated by black circles in the figure. As a result, if there are N rows on the X axis and N rows on the Y axis, the number of pads is N × N, and each X, Y-selection line can be used N × N / 2 when divided into two. As a result, (N × N / 2) ² switch elements can be controlled.

According to the configuration of the present embodiment, an arbitrary switch element arranged in a matrix can be directly selected and applied with a voltage without passing a logic circuit element from an external pad.
As a result, a logic variable LSI can be configured even when the potential required to move the movable wiring is high and the transistor breakdown voltage inside the LSI is low. Further, an extra control circuit can be completely eliminated. As a result, in addition to the above-described high-speed operation of the switch element according to the present invention due to the low resistance, a highly integrated, large-capacity, low-cost logic variable LSI can be realized.

(Fifteenth Embodiment) FIG. 27 is a view for explaining a fifteenth embodiment of the present invention, and shows a specific example of a chip in the system shown in FIG.

As shown in this example, when two-dimensional pads are arranged at a pitch of 200 μm in a logic variable LSI of 1 cm square, the number of pads is 50 × 50 = 2500, and the X-select line is 1250 lines and 1250 Y-selection lines can be used, and the number of switch elements disposed at these intersections is 1250 × 1250 = 1,562,500.
In other words, it can be a huge amount of about 1.5 mega transistors,
Compared to the current 100,000-element FPGA, a logic-variable LSI with sufficiently high degree of freedom can be realized, as well as high-speed operation, high integration, large capacity, and low cost.

(Sixteenth Embodiment) FIG. 28 is a view for explaining a sixteenth embodiment of the present invention.
An example of a writing device used for SI will be described. This is shown in FIG.
This can be applied to FIG. 27 and the like.

As shown in FIG. 28A, the logic variable LS
A table 601 on which the I602 is mounted, a head unit 604 having two-dimensionally arranged pins on the side of the probe 603 (FIG. 28B) corresponding to the two-dimensionally arranged pads, and a head unit 604
A movable stage 605 for raising and lowering the position and aligning it, and a function for converting the logic designed by the user by software to turn on / off which switch element, and writing the logic wiring information by which method And a computer 606 that determines whether Here, needless to say, a function of verifying after writing can be provided.

In LSI manufacturing, a technique for positioning a pad having a pitch of 200 μm has been established and is realistic. By making the pitch finer, the number of switch elements that can be handled increases by a square. In this embodiment, even if the product group is different, only the probe 603 needs to be replaced.

(Seventeenth Embodiment) FIG. 29 is a view for explaining a seventeenth embodiment of the present invention.
It is an enlarged view of the pad part in SI. This is shown in FIG.
6 shows an enlargement of a pad portion 6. (A) is a plan view,
(B) is a sectional view.

An insulating film 7 provided on a vertically extending selection line 701
02, a contact 703 is opened, a pad is formed thereon with a pad wiring layer 704, a protective film 705 is applied thereon, and the protective film inside the pad is removed to complete the process. With this configuration, a pad wider than the selection line 701 can be easily realized. In addition, it can be configured without a selection line from a pad and a switching element and a logic circuit from a selection line. However, except at the time of writing, a high resistance such as negligible current somewhere is set so that this node becomes 0 V except for writing. Through, GND
It is desirable to connect to

(Eighteenth Embodiment) FIG. 30 is for explaining an eighteenth embodiment of the present invention.
It is sectional drawing which shows the element structure in SI. 80 in the figure
1 is a Si substrate, 802 is an element isolation insulating film, 803 is a source / drain region, 804 is a gate electrode, and 806 is a first electrode.
A layer metal 808 indicates a second layer metal. Also,
Reference numerals 811, 812, and 830 denote third-layer metals,
811 and 812 are connection wirings, and 830 is a movable wiring.

In this example, after the active elements such as nMOS and pMOS are formed, some wirings are invariably connected by lower wirings, and the switching elements of the present invention are formed by upper wirings to change wiring connection information. I can do it. Although omitted in the description of the drawings in this description, a logic variable LSI having a high degree of freedom can be realized by not only arranging the switch element of the present invention between cell blocks but also using it as a connection switching element in the cell block.

(Nineteenth Embodiment) FIGS. 31A and 31B are views for explaining a nineteenth embodiment of the present invention, and show a switch using Coulomb force. FIG.
(B) shows an OFF state.

In this example, a movable wiring is arranged in a region serving as a channel between a source and a drain, and gates (Gate, / Gate) are provided on both sides of the movable wiring in the moving direction.
, And can be used in place of the transistors constituting the logic circuit. Although there are problems of fatigue and wear, as described with reference to FIG. 7, in order to operate as a transistor element, bias nodes are arranged on both left and right sides of a movable wiring to apply a load capacitance to the gate and / Gate. If the potential of the movable wiring does not rise so much even when the voltage is applied,
A MEMS transistor that can be repeatedly switched can be realized.

The feature of this embodiment is that the ON / OFF current difference is larger than that of the conventional transistor and the movable wiring is on when the movable wiring is on the upper side, and the source / drain portion and the movable wiring are too close to each other. In the present embodiment, even if the movable object is on the upper side, even if it is small enough to be impossible to realize with a transistor, it conducts by the ballistic or Coulomb rocket effect, and can be completely insulated when it is lowered. further,
There is no problem of S factor and threshold drop unlike a semiconductor transistor, and operation is possible even at 0.1 V or less.

The present invention is not limited to the above embodiments, but can be implemented in various modifications without departing from the scope of the invention.

[0137]

As described above in detail, according to the present invention, in a logic variable LSI capable of rewriting the logic connection information after manufacturing the LSI using Coulomb force, 1) the resistance of the wiring connection portion is made larger than in the prior art. It is possible to realize a high-speed logic variable LSI that can be reduced by orders of magnitude. 2) A MEMS-type switch element that is less wasteful than the conventional one can be configured,
Low voltage operation, downsizing, low wiring connection resistance, variation tolerance, and stable operation are possible. Further, a control circuit unit other than the switch can be eliminated or greatly reduced, and high integration and low cost can be achieved while ensuring high reliability.

[Brief description of the drawings]

FIG. 1 is a plan view of a switch element for explaining a semiconductor device according to a first embodiment and applicable to a logic variable LSI;

FIG. 2 is a plan view of a switch element that can be applied to a logic variable LSI according to a second embodiment.

FIG. 3 is a plan view of a switch element applicable to a logic variable LSI according to a third embodiment.

FIG. 4 is a view for explaining a modification of the third embodiment;

FIG. 5 is a plan view of a switch element applicable to a logic variable LSI according to a fourth embodiment.

FIG. 6 is a view for explaining a modification of the fourth embodiment;

FIG. 7 is a plan view of a switch element applicable to a logic variable LSI according to a fifth embodiment.

FIG. 8 is a plan view of a switch element applicable to a logic variable LSI according to a sixth embodiment.

FIG. 9 is a sectional view of a switch element applicable to a logic variable LSI according to a seventh embodiment;

FIG. 10 is a sectional view of a switch element applicable to a logic variable LSI according to an eighth embodiment;

FIGS. 11A and 11B are a plan view and a cross-sectional view illustrating a switch element applicable to a logic variable LSI according to a ninth embodiment; FIGS.

FIGS. 12A and 12B are a plan view and a cross-sectional view illustrating another example of a switch element applicable to the logic variable LSI according to the ninth embodiment;

FIG. 13 is a sectional view showing a switch element applicable to a logic variable LSI according to a tenth embodiment;

FIG. 14 is a sectional view showing another example of a switch element applicable to the logic variable LSI according to the tenth embodiment;

FIG. 15 is a sectional view showing still another example of a switch element applicable to the logic variable LSI according to the tenth embodiment;

FIG. 16 is a sectional view showing an example of a conventional switch element applicable to the logic variable LSI according to the eleventh embodiment.

FIG. 17 is a sectional view showing another example of a conventional switch element applicable to the logic variable LSI according to the eleventh embodiment;

FIG. 18 is a diagram showing a circuit configuration example including a switch element applicable to a logic variable LSI according to a twelfth embodiment.

FIG. 19 is a diagram showing a circuit configuration example including a switch element applicable to a logic variable LSI according to a twelfth embodiment.

FIG. 20 is a diagram showing a circuit configuration example including a switch element applicable to a logic variable LSI according to a twelfth embodiment.

FIG. 21 is a diagram showing a circuit configuration example including a switch element applicable to a logic variable LSI according to a twelfth embodiment.

FIG. 22 is a diagram showing a circuit configuration example including a switch element applicable to a logic variable LSI according to a twelfth embodiment.

FIG. 23 is a diagram showing a circuit configuration example including a switch element applicable to a logic variable LSI according to a twelfth embodiment.

FIG. 24 is a diagram showing a circuit configuration example including a switch element applicable to the logic variable LSI according to the twelfth embodiment.

FIG. 25 is a diagram showing a block or chip configuration in which a plurality of switch elements applicable to the logic variable LSI according to the thirteenth embodiment are provided.

FIG. 26 is a diagram showing a block or chip configuration in which a plurality of switch elements applicable to the logic variable LSI according to the fourteenth embodiment are provided.

FIG. 27 is a diagram showing a block or chip configuration in which a plurality of switch elements applicable to the logic variable LSI according to the fifteenth embodiment are provided.

FIG. 28 is a view showing an example of a writing device used for a logic variable LSI according to the sixteenth embodiment.

FIG. 29 is an enlarged view of a pad portion in a logic variable LSI according to a seventeenth embodiment.

FIG. 30 is a sectional view showing an element structure in a logic variable LSI according to an eighteenth embodiment.

FIG. 31 is for describing a nineteenth embodiment.
The figure which shows the switch element using Coulomb force.

FIG. 32 is a block diagram showing an example of a conventional FPGA.

FIG. 33 is a block diagram showing an example of a conventional FPGA.

FIG. 34 is a view collectively showing a conventional FPGA system and problems.

FIG. 35 is a view showing an example of a conventional MEMS switch.

FIG. 36 is a diagram showing an example of a conventional force acting on a conductor sandwiched between electrodes.

[Explanation of symbols]

11, 12, 111, 112, 311, 312, 411, 412, 51
1, 512 ... wiring for connection 20, 21, 22, 121, 122, 320, 420, 521, 52
2 ... control wiring 30, 130, 230, 330 ... movable wiring 40, 440 ... potential supply terminal 250 ... insulating film 251 ... groove 430 ... rotatable wiring 460 ... shaft groove 530 ... bendable wiring 601 ... table 602 LSI 603 Probe 605 Head 605 CPU 701 Select line 702 Insulating film 703 Contact 704 Pad wiring 705 Protective film 801 Si substrate 802 Element isolation insulating film 803 Source / source Drain region 804 gate electrode 806 first-layer metal 808 second-layer metal 811 812 third-layer metal (connection wiring) 830 third-layer metal (movable wiring) Q positive charge -Q Negative charge + V, -V: Voltage application node Cms, Cmc: Capacitance Metal 1, Metal 2: Metal wiring SW: Switch

Claims (7)

[Claims] 1. A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of switches respectively connecting any two of these signal wirings. And a switch driving means for selectively driving these switch elements by an external signal, wherein each of the switch elements is provided on two different ones of the plurality of signal wirings. The connected first and second connection wirings, a movable wiring arranged adjacent to the first and second connection wirings, and a control wiring arranged adjacent to the movable wirings A predetermined voltage is applied to the control wiring in the switch element selected by the switch driving means, so that the movable wiring is moved by Coulomb force. Wherein a between moveable wiring and the first connecting wire, and between the movable wire and the second connecting wire is intended to be connected respectively. 2. A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of switches respectively connecting any two of these signal wirings. And a switch driving means for selectively driving these switch elements by an external signal, wherein each of the switch elements is provided on two different ones of the plurality of signal wirings. The connected first and second connection wirings, a movable wiring arranged adjacent to the first and second connection wirings, and a control wiring arranged adjacent to the movable wirings A predetermined voltage is applied to the control wiring in the switch element selected by the switch driving means, so that the movable wiring and the control wiring can be moved between the movable wiring and the control wiring. And the first connection wiring, and between the movable wiring and the second connection wiring, the attractive force of Coulomb force is applied to all of the movable wiring and the first connection wiring, and between the movable wiring and the first connection wiring. A semiconductor device, wherein the movable wiring and the second connection wiring are connected to each other, and the movable wiring and the control wiring are not connected to each other. 3. A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of switches respectively connecting any two of these signal wirings. And a switch driving means for selectively driving these switch elements by an external signal, wherein each of the switch elements is provided on two different ones of the plurality of signal wirings. The connected first and second connection wirings, a movable wiring disposed adjacent to the first and second connection wirings, and a first wiring disposed adjacent to the movable wirings And a second control wiring, and by applying a predetermined voltage to both the first and second control wirings in the switch element selected by the switch driving means, the movable wiring and the second 1 Between the control wire, between said movable wire and the second control wire,
All of the movable wiring and the first connection wiring, and the movable wiring and the second connection wiring are all attracted by the Coulomb force, so that the movable wiring is connected to the first connection wiring. Wiring is connected between the movable wiring and the second connection wiring, and between the movable wiring and the first control wiring, and between the movable wiring and the second control wiring. Wherein the semiconductor device is configured not to be connected. 4. A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of switches respectively connecting any two of these signal wirings. And a switch driving means for selectively driving these switch elements by an external signal, wherein each of the switch elements is provided on two different ones of the plurality of signal wirings. The connected first and second connection wirings, a movable wiring arranged adjacent to the first and second connection wirings, and the movable wiring along the moving direction of the movable wiring. A first voltage and a second voltage, and a predetermined voltage is applied to one of the first and second control lines in the switch element selected by the switch driving means. Application Thereby, an attractive force due to Coulomb force acts between the movable wiring and the first control wiring, or between the movable wiring and the second control wiring, and the movable wiring and the first control wiring. A connection is made between the connection wirings, and between the movable wiring and the second connection wiring, and between the movable wiring and the first control wiring, and between the movable wiring and the second control wiring. A semiconductor device characterized in that the wiring is not connected. 5. A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of switches respectively connecting any two of these signal wirings. And a switch driving means for selectively driving these switch elements by an external signal, wherein each of the switch elements is provided on two different ones of the plurality of signal wirings. The connected first and second connection wirings, a movable wiring arranged adjacent to the first and second connection wirings, and an opposing arrangement once on a side in the moving direction of the movable wiring. The first and second control wirings, and the third and fourth control wirings disposed opposite to each other in the moving direction of the movable wiring, and are selected by the switch driving means. Switch element By applying a predetermined voltage to either the first and second control wirings or to both the third and fourth control wirings, the movable wiring and the first and second control wirings are applied. Between the movable wiring and the first connecting wiring, and between the movable wiring and the first connecting wiring by applying attractive force by Coulomb force between the movable wirings or between the movable wiring and the third and fourth control wirings. Between the first wiring and the second control wiring, and between the movable wiring and the first and second control wirings.
And a semiconductor device, wherein the movable wiring is not connected to the third and fourth control wirings. 6. A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of switches respectively connecting any two of these signal wirings. And a switch driving means for selectively driving these switch elements by an external signal, wherein each of the switch elements is provided on two different ones of the plurality of signal wirings. The connected first and second connection wires, a bendable wire arranged adjacent to the first and second connection wires, and a control wire arranged adjacent to the bendable wires A predetermined voltage is applied to the control wiring in the switch element selected by the switch driving means, so that the bendable wiring and the controllable wiring are connected between the bendable wiring and the control wiring. And the first connection wiring, and between the bendable wiring and the second connection wiring, all of which are attracted by the Coulomb force, so that the bendable wiring and the first connection wiring, and A semiconductor device, wherein a bendable wiring and a second connection wiring are connected to each other, and the bendable wiring and the control wiring are not connected to each other. 7. A plurality of semiconductor elements integrated on a semiconductor substrate, a plurality of signal wirings connected to these semiconductor elements, and a plurality of switches respectively connecting any two of these signal wirings. And a switch driving means for selectively driving these switch elements by an external signal, wherein each of the switch elements is provided on two different ones of the plurality of signal wirings. The connected first and second connection wirings, a rotatable wiring disposed adjacent to the first and second connection wirings, and a control wiring disposed adjacent to the rotatable wirings. A predetermined voltage is applied to the control wiring in the switch element selected by the switch driving means, so that the rotatable wiring and the control wiring are connected between the rotatable wiring and the control wiring. And the first connection wiring, and between the rotatable wiring and the second connection wiring, the attractive force of Coulomb force is exerted on the rotatable wiring and the first connection wiring, and between the rotatable wiring and the first connection wiring, and A semiconductor device, wherein a rotatable wiring and a second connection wiring are connected to each other, and the rotatable wiring and the control wiring are not connected to each other.