JP5037804B2 - Electronic devices using vertically aligned carbon nanotubes - Google Patents

Description

  The present invention relates to an electronic device using vertically aligned carbon nanotubes, and relates to an electronic device using vertically aligned carbon nanotubes characterized by the alignment state and electrode structure of carbon nanotubes for constituting a mechanical switch or nonvolatile memory. It is about.

  Conventional transistors turn on / off between the source and drain by applying a voltage to the gate, but only change the resistance value between the source and drain by about 3 to 6 digits. Are not physically cut and separated, the effects of parasitic resistance and capacitance remain between the wires, making it difficult to completely eliminate interference between circuits.

  In particular, in an RF (high frequency) device, electrical interference between circuits is large, and physical disconnection between circuits is eagerly desired, and a mechanical switch that is very small and operates at high speed is desired.

  In addition, the conventional transistor has a problem that the leakage current continues to flow when OFF and the power consumption of the circuit increases, and this leakage current is accompanied by the improvement of the integration degree of the LSI, that is, the miniaturization of the transistor. It is increasing as performance increases, and is expected to continue increasing in the future, and there is an urgent need to improve it.

On the other hand, information devices such as mobile phones have a wide variety of functions and improved performance. Therefore, while power consumption increases, short battery life after charging has become a major problem. Yes.
Therefore, also from this point, a circuit that is not in use is desired to be a minute mechanical switch that cuts off immediately and cuts off leakage current.

  On the other hand, for memory, cost competition was intense on a global scale, including Korea, China, and Europe, and the development of a low-priced and large-capacity memory was desired. However, conventional memory has already reached its limits. There was a feeling of approaching.

  In addition, in the memory market, information is not lost even when the power is turned off from the "volatile memory" where all stored contents are lost when the power is turned off as in DRAM (Dynamic Random Access Memory). Demand is shifting to “memory”.

Under the above circumstances, since the Nantero Inc. has been proposed a non-volatile memory using carbon nanotubes, explain these nonvolatile memory with reference to FIGS. 19 to 23.

See FIG .
FIG. 19 is a schematic perspective view of a non-volatile memory of Nantero, in which a word line 103 and a bit line 105 made up of a single carbon naon tube are vertically arranged on a substrate 101 with an insulating film 102 interposed therebetween. It arrange | positions so that it may cross | separate a little apart (it constitutes a capacitor) (for example, refer patent document 2).
The word line 103 is placed on the support member 104 so as to intersect the bit line 105 via a space.

  When a voltage is applied between the specific word line 103 and the bit line 105 via the electrode 106 and the electrode 107, an electrostatic force acts between the two upper and lower carbon nanotubes (capacitors) at the intersection, and the two carbon nanotubes are 1-bit information is stored in contact.

On the other hand, the carbon nanotubes once attached adhere to each other by van der Waals force, and even when the voltage is turned off, the two are not separated, information is stored as it is and operates as a nonvolatile memory (see, for example, Patent Document 2).
However, this configuration represents only a basic concept, and a specific configuration shown in FIG. 20 has been proposed (see, for example, Patent Document 1).

Here, with reference to FIGS. 21 and 22, illustrating the wire Across the bar a memory of the manufacturing steps shown in FIG. 20.
21. First, an SiN film is formed on a silicon substrate 111 via an SiO ₂ film 112, and then an SiN support 113 is formed by patterning in a stripe shape. Then, an n-type silicon layer is deposited on the entire surface. After that, etching is performed so that the upper surface is lower than the upper surface of the SiN support 113 to form the striped electrode 114, and then thermal oxidation is performed to form a sacrificial oxide film 115 on the surface of the n-type silicon layer constituting the striped electrode 114. Form.

Next, after the catalyst fine particles 116 made of Fe, Mo, Co or the like are dispersed by a spin coating method or the like, the carbon nanotubes are grown using the catalyst fine particles 116 as growth nuclei using the CVD method to form the sheet-like carbon nanotubes 117. To do.
In this case, the single-walled carbon nanotube becomes a thin sheet having a thickness substantially equal to one single-walled carbon nanotube, which is tangled in a planar state.

Figure 22 Referring then by etching the sheet-shaped carbon nanotube 117 a resist pattern (not shown) as a mask, after the belt-like sheet electrode 118 which is perpendicular to the stripe electrodes 114, a sacrificial oxide film 115 is removed by etching.

Finally, by providing the electrode 119 at the end of the belt-like sheet electrode 118, the basic configuration of the wire crossbar memory is completed.
In this case, the cavity formed by removing the sacrificial oxide film 115 becomes a movable space for the strip-shaped sheet electrode 118 to move up and down.

Then, again, the operation of the memory of the Wye Across the bar a memory with reference to FIG. 20.
See FIG. 20 again. When a positive or negative voltage is applied to the belt-like sheet electrode 118 via the electrode 119 and the other voltage is applied to the strip-like electrode 114, the belt-like sheet made of carbon nanotubes is crossed. A capacitor is formed between the electrode 118 and the stripe electrode 114, and an electrostatic force acts between the two electrodes so that the belt-like sheet electrode 118 bends downward and contacts the stripe electrode 114.

  Since the strip-shaped sheet electrode 118 that has once contacted is attached to the stripe-shaped electrode 114 by van der Faels force, it does not return to the original state even when the applied voltage is cut off. It will be remembered.

  Carbon nanotubes are nano-sized nanomaterials in the diameter direction, but can be formed in the length direction in principle as long as possible. In this sense, carbon nanotubes are also huge single-crystal molecules. Therefore, it has an excellent property that it is extremely excellent in mechanical strength and can be bent flexibly and flexibly about 5 times as much as steel in terms of tensile tension.

  Moreover, unlike metal, it is thought that an oxide film that does not allow electricity to flow on the surface is not formed, and good contact can be maintained even at the contact point of the switch. By making use of the properties of such carbon nanotubes, it is possible to construct a very small micron-sized switch.

  In addition, the carbon nanotube exhibits ballistic conduction in which electrons do not collide with atoms of the carbon nanotube up to a distance of about 600 nm in the longitudinal direction, the resistance is very low, and even at a length of 600 nm or more, Cu It is considered that it exhibits a low resistance exceeding that.

  In general, when a large amount of current is passed through a conducting wire, an electromigration phenomenon occurs where the conducting wire atoms move and the conducting wire breaks along with the flow of electrons. A large amount of current can flow (that is, the critical current density is 1000 times that of Cu).

However, in this Wai Across the bar a memory, the method erasing information, ie, the open operation principle of adhesion state of carbon nanotubes is not disclosed, then, improved showing an open operation principle with reference to FIG. 23 A carbon nanotube switch will be described (see, for example, Patent Document 3).

Figure 23 reference
FIG. 23 is a conceptual cross-sectional view of an improved carbon nanotube switch, in which a carbon nanotube 123 is provided between a pair of insulating support members 121 and 122, and one conductor facing the carbon nanotube 123 through a space. 124 is arranged as an attractor electrode, and the other conductor 125 is provided as a release electrode on the carbon nanotube 123 via a space.

  As shown in the figure below, the carbon nanotube switch is closed by applying a voltage of one polarity to the carbon nanotube 123 and applying a voltage of the other polarity to one conductor 124 serving as an attractor electrode. By attaching the nanotube 123 to the attractor electrode, the switch is turned on or the information is written.

As shown in the above diagram, the principle of opening the carbon nanotube is as follows: a voltage of one polarity is applied to the carbon nanotube 123 and a voltage of the other polarity is applied to the other conductor 125 serving as a release electrode. Thus, by pulling the carbon nanotube 123 back to the release electrode, the connection with the attractor electrode is released, and the switch is turned off or the information is erased.
US Pat. No. 6,574,130 (B2) Japanese translation of PCT publication No. 2003-504857 US Pat. No. 6,784,028 (B2)

However, as a result of examining the switch according to the above proposal, there are some problems. Since securing realistic mass and reliability unless overcome these problems it is difficult to explain this situation with reference to FIGS. 24 and 25.

See FIGS . 24 and 25
In the case of the wire crossbar memory shown in FIG. 20 , when the belt-like sheet electrode is enlarged, it is as shown in the upper part of FIG. 24 , and inside, as shown in the lower part, the carbon nanotubes 120 are in contact with each other or cross each other. 25 , the carbon nanotubes 120 meandered as shown in the upper diagram of FIG. 25 , and the end portions of the carbon nanotubes 120 are discontinuous as shown in the lower diagram. There are problems that not only the physical properties are stable and the physical properties are not stable, but also very weak against repeated fatigue.

  For example, in the portion where the carbon nanotube 120 and the carbon nanotube 120 are simply overlapped (in contact with each other), the strength possessed by the single carbon nanotube 120 cannot be obtained, and mechanical repeated fatigue can be prevented. Is very weak and is thought to tear from this part.

  Further, the resistance in the diametrical direction in which the carbon nanotubes 120 and the side walls of the carbon nanotubes 120 are connected to each other is very high, about 1000 times that of the previous longitudinal direction, and the maximum density of current that can be passed is extremely low. When making such a switch or memory, it is very important to align the direction, orientation, alignment, etc. of the carbon nanotubes 120.

Figure 26 reference
Ideally, as shown in FIG. 26 , it is desirable that the carbon nanotubes are continuous without any discontinuous portions of the carbon nanotubes, and are straight without bending, and the orientations of the carbon nanotubes are aligned.
However, it is very difficult to obtain such an ideal carbon nanotube array by the above-described conventional method.

  Further, in the conventional technology, a cavity that allows deformation (movability) of the carbon nanotube 123 or the belt-like sheet electrode 118 is required between the belt-like sheet electrode 118 and the stripe-like electrode 114 or between the carbon nanotube 123 and the attractor electrode 124. However, in order to form such a cavity, it is necessary to remove the sacrificial film using a wet process.

Figure 27 reference
However, when this wet treatment is performed, as shown in the figure, the carbon nanotube 123 or the strip-shaped sheet electrode 118 adheres to the attractor electrode 124 or the stripe-shaped electrode 114 due to surface tension, and when it is further dried, it is more firmly fixed. a phenomenon that put away, that is, cause stearyl I Kkingu phenomenon, there is a problem that does not operate as a switch or the like.

  Further, for the same reason, a phenomenon that carbon nanotubes adhere to each other and become bundled, that is, a bundling phenomenon also occurs, and there is a problem that stable production is difficult and yield is difficult to improve.

  In the above-described prior art, some of the catalyst fine particles 116 formed on the surface actually do not have carbon nanotubes. Therefore, when the sacrificial oxide film 115 is removed, the sacrificial oxide film 115 is removed. The catalyst fine particles 116 on which the carbon nanotubes do not grow are drifted in the etching solution together with the removal of the sacrificial oxide film 115, and are adhered between the carbon nanotube sheet and the striped electrode 114, and the stripes are formed via the catalyst fine particles. There is a problem that the yield electrode is short-circuited between the strip-shaped electrode 114 and the strip-shaped sheet electrode 118 and the yield is lowered.

  Furthermore, in order to form a mechanism for erasing the memory, a series of steps for forming the release electrode is required. For example, when removing the sacrificial oxide film 115, it is necessary to remove by side etching. There is a problem that it takes time for the removal, or the removal is incomplete and the operation space of the carbon nanotube cannot be obtained.

  In addition, when trying to make the above-mentioned switch or memory three-dimensional with a multi-layer structure, an interlayer insulating film enters the movable part of the lower layer switch or memory and the free movement of the carbon nanotubes is hindered. There is a problem that it is very difficult.

Finally, a special package that secures a cavity that does not obstruct the moving parts is required, but such a package is as expensive as a chip, and it is difficult to reduce the price. There's a problem.
As described above, in order to realize a switch or memory formed of carbon nanotubes, it is necessary to overcome and solve the above problems.

  Therefore, an object of the present invention is to realize an element structure that can be manufactured through a process that is reliable in operation and high in productivity.

Here, means for solving the problems in the present invention will be described.
To solve the above Symbol object, the present invention is an electronic device using the vertically aligned carbon nanotubes, and carbon nanotubes oriented in the direction perpendicular to the main surface of the substrate, and provided adjacent to the carbon nanotube A laminated electrode comprising: insulating layer / first electrode / insulating layer / second electrode / insulating layer, and a surface of the first electrode and the second electrode facing the carbon nanotube is recessed from an end of the insulating layer characterized by comprising and.

  In this way, by making the device structure vertical instead of horizontal, and by providing the vertically aligned carbon nanotubes 2 that grow perpendicular to the substrate 1 between the electrodes, the orientation and shape of the carbon nanotubes 2 can be made uniform, Since the carbon nanotubes 2 can be formed continuously and without breaks, variations in electrical characteristics can be reduced, and sufficient strength against repeated fatigue can be obtained.

  Further, by utilizing the property that the carbon nanotube 2 grows straight without bending, it becomes possible to secure a movable space without using a wet process, and the problem of sticking and bundling can be avoided.

  In addition, since the first electrode 4 and the second electrode 7 serving as an attractor electrode or a release electrode can be formed at the same time, the attractor electrode and the release electrode can be formed at the same time in a single process, thereby simplifying the manufacturing process and reducing the number of processes. Therefore, the cost can be reduced.

In particular, the laminated electrode is composed of an insulating layer / first electrode / insulating layer / second electrode / insulating layer, and the opposed surfaces of the first electrode and the second electrode to the carbon nanotube are made to recede from the end of the insulating layer. since it is, it is possible to construct a switch utilizing a local bending of mosquitoes over carbon nanotubes.

  Also, the first electrode is an attracting electrode that adsorbs carbon nanotubes, ie, an attractor electrode, and is an attractor electrode, and the second electrode is an open electrode that separates carbon nanotubes from the first electrode and is not conducting, ie, a release electrode Alternatively, the first electrode and the second electrode may be a suction electrode that adsorbs carbon nanotubes to make it conductive, and various functions can be achieved by changing the functions of the first electrode and the second electrode. Different actions are possible.

  Further, the laminated electrode may be configured such that three or more electrodes are laminated via an insulating layer, and the surface of the electrode facing the carbon nanotube is set back from the end of the insulating layer. Thus, a vertical rotary switch can be configured.

  In addition, the above-mentioned laminated electrodes may be arranged in the circumference of three or more or three or more pairs, whereby a rotary switch can be configured, or a multi-bit can be realized, and the degree of integration can be increased. Improvements can be made and the bit unit price can be lowered.

  The carbon nanotubes constituting the electronic device described above may be a single carbon nanotube, a plurality of carbon nanotubes that do not constitute a carbon nanotube bundle, or a carbon nanotube bundle, and may be appropriately selected depending on the application.

  In addition, since the carbon nanotube has elasticity, the end surface of the carbon nanotube opposite to the substrate may be fixed or may not be fixed.

In addition, it is desirable that the space on the carbon nanotube is closed with a cap member, which eliminates the need for using a special package, and enables cost reduction.
Furthermore, the memory can be stacked in layers after the lid is closed, the degree of integration can be increased, and the bit unit price can be lowered.

  In addition, a nonvolatile memory having a configuration similar to that of a conventional semiconductor memory cell is configured by using a substrate as a semiconductor substrate and vertically aligning carbon nanotubes on a drain electrode that contacts a drain region formed on the semiconductor substrate. Is possible.

In the present invention,
(1) The orientation of the carbon nanotubes is aligned, and the movable part can be composed of one continuous carbon nanotube, so that the electrical characteristics are stable,
It can withstand repeated fatigue and can improve yield and reliability.
(2) Since the movable space can be formed without relying on wet processing, the yield can be improved.
(3) Since the sealed structure can be made while maintaining the movable space, it is possible to arrange multiple layers at a high density, and the degree of integration is improved and the price is low, such as omission of the package itself and the use of an existing inexpensive package. It becomes advantageous for the conversion.
(4) Multi-level logical recording becomes possible, and the degree of integration can be improved.

  In addition, since the electronic device of the present invention constitutes a nano-sized very fine mechanical switch, the electrical resistance when turned off can be effectively made infinite and the leakage current can be extremely reduced. Since it is possible to flow a large current by utilizing the low resistance of the carbon nanotubes when turned on, it is possible to replace with a power transistor.

In the present invention, the first electrode and the second electrode, which are attractor electrodes or release electrodes, are composed of insulating layer / first electrode / insulating layer / second electrode / insulating layer, and the carbon of the first electrode and the second electrode. The surface facing the nanotube is configured as a laminated electrode that is recessed from the end of the insulating layer, or three or more electrodes are stacked via the insulating layer, and the surface facing the carbon nanotube of the electrode There the insulating layer end and at least a pair of laminated electrodes are recessed constructed from the, by the carbon nanotube growth oriented in a direction perpendicular to the main surface of the substrate between the laminated electrodes facing each other, the carbon A switching operation or a memory operation is performed in a contact state between the nanotube and the attractor electrode.

The switch and the memory are essentially the same, and can be said to be a switch and a memory from a different perspective.
For example, the switch creates ON and OFF states, and can be handled as a memory if the switch is set to 1 and the switch is set to 0.

  Of course, in order to achieve product specifications, it is necessary to provide a matrix of word lines and bit lines, and to place these switches at the intersections, and at these intersections there are transfer gates (cell transistors) and diodes. Necessary, and of course, a peripheral circuit is also necessary, but when viewed only in the storage area of each cell, the switch may be regarded as a memory.

  Next, assuming that the memory part is composed of fine switches, the difference between “volatile memory” and “nonvolatile memory” will be explained. The difference between the two is the resilience of carbon nanotubes and the van der working between carbon nanotubes and electrodes. It is determined by the magnitude relationship of the Fars force.

  In other words, if the restoring force of the carbon nanotubes is superior to the van der Waals force, after turning off the voltage, the carbon nanotubes are separated from the electrode by their own restoring force and become a volatile memory, whereas the restoring force of the carbon nanotubes If the van der Waals force is inferior, the carbon nanotube is not separated from the electrode by the van der Waals force even after the voltage is turned off, and becomes a non-volatile memory.

  This van der Waals force depends on the material of the part in contact with the carbon nanotubes and the contact area, so it can be adjusted appropriately by changing the material and these areas, and the restoring force of the carbon nanotubes is also the length of the carbon nanotubes. (Length from fulcrum to contact part), carbon nanotube thickness (depending on catalyst type and catalyst size), number of carbon nanotube layers and chirality (graphite sheet winding), etc. Can be adjusted.

Here, with reference to FIG.1 and FIG.2, the carbon nanotube device of Example 1 of this invention is demonstrated.
Refer to FIG. 1. First, as shown in the upper figure, after a lower electrode 12 made of TiN is formed on a silicon substrate 11 by sputtering, an SiO ₂ film having a thickness of, for example, 300 nm using sputtering or CVD. 34, a conductor film made of Al with a thickness of 100 nm, for example, a SiO ₂ film 36 with a thickness of 300 nm, a conductor film with a thickness of, for example, 100 nm, and a thickness of 300 nm, for example. The SiO ₂ films 38 are sequentially deposited.

Next, dry etching is performed to form a movable space 39 having a width of 0.2 μm, and a laminated electrode 47 composed of SiO ₂ film 38 / attractor electrode 50 / SiO ₂ film 36 / attractor electrode 49 / SiO ₂ film 34. Then, a laminated electrode 48 composed of the SiO ₂ film 38 / release electrode 52 / SiO ₂ film 36 / release electrode 51 / SiO ₂ film 34 is formed.

  Next, by performing selective etching, the exposed end surfaces of the attractor electrodes 49 and 50 and the release electrodes 51 and 52 are side-etched to form the recesses 53 to 56.

  Next, a Co film is selectively deposited only on the lower electrode 12 exposed from the movable space 39 using a mask vapor deposition method to form the catalyst layer 13.

Next, the basic structure of the carbon nanotube device is completed by growing the carbon nanotube bundle 57 using the catalyst layer 13 as a growth starting point by the CVD method using C ₂ H ₂ gas.

  In this case, since the carbon nanotubes constituting the carbon nanotube bundle 57 grow perpendicular to the substrate, the carbon nanotube bundles 57 cannot grow into the recesses 53 to 56 formed on the end faces of the attractor electrodes 49 and 50 and the release electrodes 51 and 52. Since it does not adhere to the attractor electrodes 49 and 50 or the release electrodes 51 and 52 at the stage, the reliability of the switch operation is remarkably improved, and the carbon nanotubes 58 to 60 grow densely to form a carbon nanotube bundle. Since the concave portions 53 to 56 are the only movable spaces, the degree of freedom of operation is improved.

The lower diagram of FIG. 1 is an equivalent circuit diagram in which two switches are connected in parallel.

See Figure 2 Figure 2 is an explanatory view of the operation of the carbon nanotube device of Example 1 of the present invention, where a representative carbon nanotube 58 to 60 operation is likely as three understanding of the carbon nanotube bundles 57 It shows.
Here, the illustration of the catalyst layer is omitted.

  In this case, when a voltage is applied between the attractor electrode 50 and the carbon nanotubes 58 to 60, an electrostatic force acts between the attractor electrode 50 and the carbon nanotube 58 in the vicinity of the attractor electrode 50. It adheres to the attractor electrode 50.

  On the other hand, when canceling the voltage, once the voltage between the attractor electrode 50 and the carbon nanotubes 58 to 60 is made zero, and then a voltage is applied between the carbon nanotubes 58 to 60 and the release electrode 52, the release electrode 52 The first electric field exerts a force on the carbon nanotube 60 in the vicinity and also exerts a force on the carbon nanotube 58 in the vicinity of the carbon nanotube 58 in the vicinity of the attractor electrode 50, and the switch is released by pulling the carbon nanotube 58 away from the attractor electrode 50. Made.

Similarly, a switch is formed between the lower left and right electrodes, and performs the same operation as the upper switch.
In this case, when the width of the movable space 39 is wide and the number of carbon nanotubes constituting the carbon nanotube bundle 57 is large, the force of the electric field exerted from one electrode to the other electrode is weakened. Therefore, it is desirable that the width of the movable space 39 is narrow.

Next, the carbon nanotube device according to the second embodiment of the present invention will be described with reference to FIGS. 3 and 4. The carbon nanotube device according to the second embodiment of the present invention is substantially the same as the above-described embodiment just by setting the release electrode on the lower side and the attractor electrode on the upper side. Although it is the same as that of Example 1 , in this case, only the recessed part formed in the edge part of an attractor electrode and a release electrode becomes a substantial movable space.
The catalyst layer is not shown.
The upper part of FIG. 3 reference figures 3 is a conceptual configuration diagram of a carbon nanotube device of Example 2 of the present invention, also shown below is an equivalent circuit diagram.
In this case, the release electrodes 63 and 65 are on the lower side and the attractor electrodes 64 and 66 are on the upper side, and each laminated electrode 61 and 62 itself constitutes an independent switch or memory.
The detailed configuration of the memory will be described later separately.

As shown in an equivalent circuit diagram in the lower diagram of FIG. 3 , four switches connected in parallel can be configured with the lower electrode 12 as a common electrode.
If considered as a memory, four states of A, B, C, and D can be stored, so that multi-value storage of 2 bits (00, 01, 10, 11) is possible, so that the recording density is double that of the conventional case. Can be achieved.

See Figure 4. Figure 4 is an explanatory view of the operation of the carbon nanotube device of Example 2 of the present invention, where, in order to facilitate the understanding of the operation of one carbon nanotube 58 of the carbon nanotube bundles 57 Only the operation is shown.
Since the operation of the carbon nanotube bundle 57 is actually performed, only the concave portions 53 to 56 formed at the end portions of the attractor electrode and the release electrode become a substantial movable space.

  As shown in the figure, the bending of the carbon nanotube 58 formed by applying a voltage to the attractor electrode 64 is reset by the bending of the carbon nanotube 58 formed by applying a voltage to the release electrode 63, and conversely The bending of the carbon nanotube 58 formed by applying a voltage to the release electrode 63 is reset by the bending of the carbon nanotube 58 formed by applying a voltage to the attractor electrode 64.

Thus, in the carbon nanotube device of Example 2 of the present invention, writing and erasing can be performed between electrodes constituting the same laminated electrode.
This is because the allowance (bending) allowed for the carbon nanotube is only one bend, so that one bend is acceptable, but two bends are not acceptable. The length of the carbon nanotube, the depth of the recesses 53 to 56, and the height of the recesses 53 to 56, that is, the thickness of the release electrodes 63 and 65 and the attractor electrodes 64 and 66, the thickness of the carbon nanotube, and the like are adjusted. This is possible.

Such an operation cannot be realized if the carbon nanotubes 58 are attached and fixed to the ends of the SiO ₂ films 34, 36, 38, but the carbon nanotubes 58 are perpendicular to the vicinity of the SiO ₂ films 34, 36, 38. Since it grows in the direction, it does not adhere to the ends of the SiO ₂ films 34, 36, 38.

  In addition, as the actual movement of the carbon nanotube bundle 57, the entire carbon nanotube bundle 57 does not move, but only the carbon nanotube 58 located in the vicinity of each electrode moves. The electrodes 61 and 62 constitute a switch that moves independently.

Thus, in Example 2 of the present invention, even if carbon nanotubes stand in the opening, only the carbon nanotubes near the recess need to move, so the size of the opening can be freely set.

Further, since only the carbon nanotubes in the vicinity of the recesses are operated, it is not necessary to apply an electric field from the opposite laminated electrode side, so that low voltage operation is possible.
In addition, since two switches or memories that move independently by the two stacked electrodes 61 and 62 are configured, the degree of integration can be doubled.

Next, the carbon nanotube device of Example 3 of the present invention will be described with reference to FIG. 5 , but only a cap is provided on the carbon nanotube device of Example 2 described above, so only a conceptual configuration diagram is shown. .

See Figure 5. Figure 5 is a conceptual configuration diagram of a carbon nanotube device of Example 3 of the present invention, which cap layer 67 is provided so as to cover the top of the stacked electrodes 61 and 62 and the carbon nanotube bundles 57 Since the carbon nanotube bundles 57 are densely grown, the cap layer 67 cannot enter and grow deep into the gap between the carbon nanotubes one by one. Only the cap layer 67 is provided.
In this case, the cap layer 67 may be an insulator or a conductor. When the cap layer 67 is composed of a conductor, the cap layer 67 may have the same function as the lower electrode 12.

The basic operation of the carbon nanotube device of Example 3 is exactly the same as that of Example 2 described above, but in this Example 3 , each switch is housed in a small package having a movable space. It becomes a state.

  In this package, not only the minute spaces of the recesses 53 to 56 are maintained without being closed, but also the movement of the carbon nanotube bundle 57 in the longitudinal direction is not hindered. And the degree of integration can be improved three-dimensionally.

  Further, by providing the cap layer 67, a package is not required or a general-purpose package can be used, so that a significant cost reduction can be achieved.

Next, with reference to FIG. 6 , the carbon nanotube device according to the fourth embodiment of the present invention will be described. However, the substantial configuration is the same as the carbon nanotube device according to the first embodiment just by increasing the number of stacked layers. Only a conceptual configuration diagram and an equivalent circuit diagram are shown.

Left side of FIG. 6 reference 6 is a conceptual configuration diagram of a carbon nanotube device of Example 4 of the present invention, also, the right view is an equivalent circuit diagram.
As shown in the figure, eight layers of electrodes are provided here, and the laminated electrodes 68 and 69 on both the left and right sides form a pair to constitute one switch.

In addition, as shown in the equivalent circuit diagram on the right side of FIG. 6 , by providing the multilayer electrodes 71 and 72 via the SiO ₂ film 70, the memory capacity can be significantly increased according to the number of layers, In the case of eight layers, eight states A to H can be realized.

Next, with reference to FIG. 7 , the carbon nanotube device of Example 5 of the present invention will be described. However, the substantial configuration is the same as the carbon nanotube device of Example 2 just by increasing the number of stacked layers. Only a conceptual configuration diagram and an equivalent circuit diagram are shown.

Left side of FIG. 7 reference 7 is a conceptual configuration diagram of a carbon nanotube device of Example 5 of the present invention, also, the right view is an equivalent circuit diagram.
As shown in the figure, eight layers of electrodes are provided here, and the laminated electrodes 68 and 69 on both the left and right sides independently constitute two switches.

In addition, as shown in the equivalent circuit diagram on the right side of FIG. 7 , by providing the multilayer electrodes 71 and 72 via the SiO ₂ film 70, the memory capacity can be remarkably increased according to the number of stacked layers. Since the left and right laminated electrodes 68 and 69 independently constitute two switches, 16 states A to P can be realized in the case of eight layers.

Next, a carbon nanotube device according to Example 6 of the present invention will be described with reference to FIGS. 8 and 9 , but the structure is substantially the same as that of the carbon nanotube device according to Example 1 except for the operation. Therefore, only the configuration and operation are shown.

The upper part of FIG. 8 see Figure 8 is a conceptual configuration diagram of a carbon nanotube device of Example 6 of the present invention, also shown below is an equivalent circuit diagram.
As shown in the figure, a single carbon nanotube 58 constitutes a rotary switch that realizes four states A to D.

9. FIG. 9 is an explanatory view of the operation of the carbon nanotube device of Example 6 of the present invention. In the state in which the voltage of one polarity is applied to the lower electrode 12, the other of the electrodes 71 and 72 is on the other side. rotary switch by applying a polarity of the voltage, since the carbon nanotubes 58 performs the switching operation by attaching is attracted to one of the electrodes 71 and 72 to which a voltage is applied, can only connect to one electrode of the Configure.

Next, the carbon nanotube device of Example 7 of the present invention will be described with reference to FIG. 10. However, since the substantial structure is the same as that of the carbon nanotube device of Example 6 described above except that the number of stacked layers is different, Only a conceptual configuration diagram and an equivalent circuit diagram are shown.

The upper part of FIG. 10 reference figures 10 is a conceptual configuration diagram of a carbon nanotube device of Example 7 of the present invention, also shown below is an equivalent circuit diagram.
As shown in the figure, since the number of stacked electrodes 71 and 72 is eight, a single carbon nanotube 58 constitutes a rotary switch that realizes 16 states A to P.

Next, a carbon nanotube device of Example 8 of the present invention will be described with reference to FIG. 11. The structure of each stacked electrode is the same as that of any of the above examples, but the carbon nanotube is the center. Three or more laminated electrodes are arranged on the circumference.

Figure 11 Referring to FIG. 11 is a schematic plan view of a carbon nanotube device of Example 8 of the present invention, the above figure is a thin carbon nanotube bundles 74 provided at the center portion in the center of four stacked openings 73 The case where the electrode 75 is arrange | positioned on the periphery is shown, and the lower part has shown the case where the six laminated electrodes 75 are arrange | positioned.

In this case, if the voltage applied between each laminated electrode 75 and the carbon nanotube bundle 74 is lowered, the bending operation is performed without bending , and if the voltage is increased, the same operation as in the first or sixth embodiment is performed. The operation will be performed.

In the eighth embodiment of the present invention, since three or more laminated electrodes 75 are arranged around a single carbon nanotube bundle, the number of states that can be realized increases according to the number of laminated electrodes 75 arranged. can do.
When the number of laminated electrodes 75 to be arranged is an odd number, each laminated electrode 75 constitutes a single switch as in the second embodiment.

Next, the carbon nanotube device of Example 9 of the present invention will be described with reference to FIG. 12. The basic configuration is the same as that of Example 8 , but the carbon nanotube bundle fills the opening. It is a thing.

See FIG. 12
FIG. 12 is a conceptual plan view of the carbon nanotube device according to the ninth embodiment of the present invention, and the upper figure has four stacked electrodes 75 arranged on the circumference around the carbon nanotube bundle 76 filling the opening 73. The lower part shows a case where six laminated electrodes are arranged.

Also in this case, if the voltage applied between each laminated electrode 75 and the carbon nanotube bundle 76 is lowered , an operation of bending without bending is performed, and if the voltage is increased, the same operation as in the first or sixth embodiment is performed. Will be performed.

In the ninth embodiment of the present invention, since three or more laminated electrodes 75 are arranged around the carbon nanotube bundle 76, the number of states that can be realized increases according to the number of laminated electrodes 75 arranged. Can do.
When the number of laminated electrodes 75 to be arranged is an odd number, each laminated electrode 75 constitutes a single switch as in the second embodiment.

Next, a carbon nanotube device of Example 10 of the present invention will be described with reference to FIG. 13. Although the basic configuration is the same as that of Example 1 , the carbon nanotubes constituting the switch electrode and carbon Nanotube plugs are grown simultaneously.

See FIG. 13
First, a gate electrode 83 is provided on a silicon substrate 81 via a gate insulating film 82, a source region 84 and a drain region 85 are formed, and then a Co region is formed on the surfaces of the source region 84, the drain region 85, and the gate electrode 83. After the silicide electrode 86 is formed, an interlayer insulating film 87 is formed on the entire surface.

Then, Chi was to form a resist pattern for lift-off so as to cover the MOSFET region (not shown), the lower electrode 12, SiO ₂ film 34, conductive film 35 made of Al formed of TiN by using a spatter method, A SiO ₂ film 36, a conductor film 37 made of Al, and a SiO ₂ film 38 are sequentially deposited.

Next, after removing the resist pattern, dry etching is performed to form a movable space 39 and a pair of SiO ₂ film 38 / attractor electrode 42 / SiO ₂ film 36 / release electrode 43 / SiO ₂ film 34. The stacked electrodes 40 and 41 are formed.
In this dry etching process, a via hole 88 for the Co silicide electrode 86 of the MOSFET is formed.

Next, by performing side etching to selectively remove the exposed end portions of the attractor electrode 42 and the release electrode to form a recess, it is selectively formed on the lower electrode 12 and the Co silicide electrode 86 exposed in the movable space 39. The basic structure of the carbon nanotube device is completed by spraying catalyst fine particles (not shown) on the substrate and then simultaneously growing the carbon nanotube bundle 57 and the carbon nanotube bundle plug 89 by the CVD method using C ₂ H ₂ gas. To do.

  In this way, when integrating carbon nanotube devices in a semiconductor integrated circuit device, at least a part of plugs for MOSFETs, including transistors for driving carbon nanotube devices, are formed at the same time as a bundle of carbon nanotubes constituting the carbon nanotube device. can do.

Next, with reference to FIG. 14, the nonvolatile memory will be described using a carbon nanotube device of Example 11 of the present invention.
See FIG. 14
The upper diagram of FIG. 14 is a conceptual perspective view of a nonvolatile memory using the carbon nanotube device of Example 11 of the present invention, and the lower diagram is an equivalent circuit diagram thereof .

Since the operation of the nonvolatile memory of the eleventh embodiment can realize four states as in the first embodiment, a multi-valued logical nonvolatile memory can be configured depending on which electrode it is attached to.

Next, with reference to FIGS. 15 and 16 , a non-volatile memory using the carbon nanotube device of Example 12 of the present invention will be described.
See FIGS. 15 and 16
FIG. 15 is a conceptual perspective view of a nonvolatile memory using the carbon nanotube device of Example 12 of the present invention, and FIG. 16 is an equivalent circuit diagram thereof.

Since the operation of the nonvolatile memory according to the twelfth embodiment can realize the number of states corresponding to the number of stacked layers as in the fourth embodiment, the multi-valued logic non-volatile having six values or more depends on which electrode is attached. A memory can be configured.
In the figure, since the lowermost electrode is not selected, the value is 14 values.

Next, with reference to FIGS. 17 and 18 , a nonvolatile memory using the carbon nanotube device of Example 13 of the present invention will be described.
17 and 18 refer to
FIG. 17 is a conceptual perspective view of a nonvolatile memory using the carbon nanotube device of Example 13 of the present invention, and FIG. 18 is an equivalent circuit diagram thereof.

Since the operation of the non-volatile memory of this example 13 can realize the number of states corresponding to the number of stacked layers as in the case of the above-described example 6 , it is 6 values or more depending on which electrode is attached (in the case of the figure). 16-value) multi-valued logic nonvolatile memory can be configured.

The embodiments of the present invention have been described above, but the present invention is not limited to the configurations and conditions described in the embodiments, and various modifications can be made .

  In each of the above embodiments, a silicon substrate is used as the substrate. However, the substrate is not limited to a silicon substrate, and a quartz substrate, a glass substrate, a sapphire substrate, a stainless steel substrate, or the like may be used. Is.

  In each of the above embodiments, TiN is used as the lower electrode and Al is used as the attractor electrode and the release electrode. However, the present invention is not limited to TiN or Al, but Ti, Ni, TiN, W, Cu. , Ag, Au, At, WN, Fe, Mo, or other conductors may be used.

  In each of the above embodiments, Co is used as a catalyst. However, the catalyst is not limited to Co, and other catalysts such as Fe, Ni, Pt, Ti, Pa, TiCo may be used. is there.

  In each of the above embodiments, a catalyst layer is used when growing carbon nanotubes. However, the catalyst layer does not have to be spread in a plane, and is formed by spraying catalyst fine particles. In this case, the density of the carbon nanotubes can be controlled by the density of the catalyst fine particles.

In Example 10 above, the plug is formed of a carbon nanotube bundle. However, the plug is not necessarily formed of a carbon nanotube bundle, and is formed of carbon nanotubes having a sparse density that does not form a carbon nanotube bundle. The number of carbon nanotubes required as a carbon nanotube device may be determined.

Further, in the above-described Examples 11 to 13 , although described as a non-volatile memory, the restoring force due to the elasticity of the carbon nanotube and the adhesion force to each electrode are adjusted, and no voltage is applied to each electrode. In this case, a volatile memory such as a DRAM can be obtained if the carbon nanotubes are separated from each electrode as time passes.

Further, in Example 11 to Example 13 above, but using a conventional MOSFET as a transfer over the gate, it is intended may be used a carbon nanotube device of the present invention as a transfer gate, in that case, A memory device can be configured without using a semiconductor.

  Furthermore, in each of the above embodiments, carbon nanotubes are used. However, carbon nanotubes are not limited as long as they are conductive materials capable of growing in a rod shape with any nucleus as the base point. For example, aluminum, It is known that metal whisker such as iron or Si can be grown in a rod shape, and such a whisker or a rod-shaped semiconductor may be used.

In each of the above embodiments, acetylene (C ₂ H ₂ ) is used as a source gas when growing the carbon nanotubes, but is not limited to C ₂ H ₂ , and methane (CH ₄ ), Other gases such as ethylene (C ₂ H ₄ ), alcohols, CO (carbon monoxide) may be used.

Here , the detailed features of the present invention will be described again.
(Appendix 1) Carbon nanotubes oriented in a direction perpendicular to the main surface of the substrate , adjacent to the carbon nanotubes , and comprising an insulating layer / first electrode / insulating layer / second electrode / insulating layer and An electronic device using vertically aligned carbon nanotubes, comprising: a laminated electrode in which opposing surfaces of the first electrode and the second electrode facing the carbon nanotube are recessed from an end of the insulating layer .
(Supplementary Note 2 ) The first electrode is a suction electrode that adsorbs the carbon nanotubes to make the conductive state, and the second electrode is an open electrode that pulls the carbon nanotubes away from the first electrode and makes the conductive state non-conductive. An electronic device using the vertically aligned carbon nanotubes according to appendix 1, which is characterized in that it exists.
(Supplementary note 3 ) The electronic device using vertically aligned carbon nanotubes according to supplementary note 1 , wherein the first electrode and the second electrode are suction electrodes that adsorb the carbon nanotubes to make them conductive.
(Supplementary Note 4) and carbon nanotubes oriented in a direction perpendicular to the main surface of the substrate, the while being oppositely arranged with the carbon nanotubes, wherein the stacked and the electrode 3 or more layers of electrodes through the insulating layer An electronic device using vertically aligned carbon nanotubes, comprising: at least a pair of stacked electrodes whose opposite surfaces to the carbon nanotubes are recessed from the end portions of the insulating layer.
(Appendix 5 ) An electronic device using the vertically aligned carbon nanotubes according to any one of Appendix 1 to Appendix 4 , wherein three or more or three or more pairs of the stacked electrodes are arranged on the circumference.
(Additional remark 6 ) The said carbon nanotube consists of one carbon nanotube, The electronic device using the vertically aligned carbon nanotube of any one of Additional remark 1 thru | or Additional remark 5 characterized by the above-mentioned.
(Supplementary note 7 ) The electronic device using the vertically aligned carbon nanotube according to any one of Supplementary notes 1 to 5 , wherein the carbon nanotube includes a plurality of carbon nanotubes that do not constitute a carbon nanotube bundle.
(Supplementary note 8 ) The electronic device using the vertically aligned carbon nanotube according to any one of supplementary notes 1 to 5 , wherein the carbon nanotube is a carbon nanotube bundle.
(Supplementary Note 9 ) An electronic device using the vertically aligned carbon nanotube according to any one of Supplementary notes 1 to 8 , wherein an end surface of the carbon nanotube opposite to the substrate is not fixed and is in an open state.
(Appendix 10 ) An electronic device using vertically aligned carbon nanotubes according to any one of appendices 1 to 8 , wherein an end face of the carbon nanotube opposite to the substrate is fixed.
(Additional remark 11 ) The electronic device using the vertically aligned carbon nanotube of any one of Additional remark 1 thru | or Additional remark 10 characterized by the space on the said carbon nanotube being closed with the cap member.
(Supplementary note 12 ) Any one of Supplementary notes 1 to 11 , wherein the substrate is a semiconductor substrate, and the carbon nanotubes are vertically aligned on a drain electrode in contact with a drain region formed in the semiconductor substrate. 2. An electronic device using the vertically aligned carbon nanotube according to 1.

  As an application example of the present invention, a non-volatile memory is typical, but it is also applied to a volatile memory such as a DRAM by adjusting the restoring force and adhesive force by carbon nanotubes and elasticity, Furthermore, the present invention is also applied to a switching device that can realize a multi-value state, and a micro electrical switch that completely cuts off a leakage current, like a mechanical switch.

It is explanatory drawing of the carbon nanotube device of Example 1 of this invention . It is explanatory drawing of operation | movement of the carbon nanotube device of Example 1 of this invention. It is explanatory drawing of the carbon nanotube device of Example 2 of this invention. It is explanatory drawing of operation | movement of the carbon nanotube device of Example 2 of this invention. It is a notional block diagram of the carbon nanotube device of Example 3 of the present invention. It is explanatory drawing of the carbon nanotube device of Example 4 of this invention. It is explanatory drawing of the carbon nanotube device of Example 5 of this invention . It is explanatory drawing of the carbon nanotube device of Example 6 of this invention. It is explanatory drawing of operation | movement of the carbon nanotube device of Example 6 of this invention. It is explanatory drawing of the carbon nanotube device of Example 7 of this invention. It is a notional top view of the carbon nanotube device of Example 8 of the present invention. It is a notional top view of the carbon nanotube device of Example 9 of the present invention. It is explanatory drawing of the manufacturing process of the carbon nanotube device of Example 10 of this invention. It is explanatory drawing of the non-volatile memory using the carbon nanotube device of Example 11 of this invention. It is a conceptual perspective view of the non-volatile memory using the carbon nanotube device of Example 12 of this invention. It is an equivalent circuit schematic of the carbon nanotube device of Example 12 of the present invention. It is a conceptual perspective view of the non-volatile memory using the carbon nanotube device of Example 13 of this invention. It is an equivalent circuit schematic of the carbon nanotube device of Example 13 of the present invention. 1 is a schematic perspective view of a non-volatile memory of Nantero. FIG. It is a specific block diagram of the non-volatile memory of Nantero. FIG. 21 is an explanatory diagram of the manufacturing process up to the middle of the wire crossbar memory shown in FIG. 20. FIG. 21 is an explanatory diagram of the manufacturing process subsequent to FIG. 21 for the wire crossbar memory shown in FIG. 20. It is explanatory drawing of an improved carbon nanotube switch. It is an enlarged view of a strip-shaped sheet electrode. It is an enlarged view of a strip-shaped sheet electrode. It is an enlarged view which shows the ideal state of a strip | belt-shaped sheet electrode. It is explanatory drawing of a sticking phenomenon.

11 Silicon substrate 12 Lower electrode 13 Catalyst layer 34 SiO ₂ film 35 Conductor film 36 SiO ₂ film
38 SiO ₂ film 39 movable spatial
47 laminated electrode 48 laminated electrode 49 attractor electrode 50 attractor electrode 51 release electrode 52 release electrode 53 to 56 recess 57 carbon nanotube bundle 58 to 60 carbon nanotube 61 laminated electrode 62 laminated electrode 63 release electrode 64 attractor electrode 65 release electrode 66 attractor electrode 67 Cap layer 68 Laminated electrode 69 Laminated electrode 70 SiO ₂ film 71 Electrode 72 Electrode 73 Opening 74 Carbon nanotube bundle 75 Laminated electrode 76 Carbon nanotube bundle
81 Silicon substrate 82 Gate insulating film 83 Gate electrode 84 Source region 85 Drain region 86 Co silicide electrode 87 Interlayer insulating film 88 Via hole 89 Carbon nanotube bundle plug 90 Drain electrode 91 Source electrode 101 Substrate 102 Insulating film 103 Word line 104 Support member 105 Bit Line 106 Electrode 107 Electrode 111 Silicon substrate 112 SiO ₂ film 113 SiN support 114 Striped electrode 115 Sacrificial oxide film 116 Catalyst fine particle 117 Sheet-like carbon nanotube 118 Strip-like sheet electrode 119 Electrode 120 Carbon nanotube 121 Insulating support member 122 Insulating support Member 123 Carbon nanotube 124 Conductor 125 Conductor

Claims (3)

Carbon nanotubes oriented perpendicular to the main surface of the substrate ;
The insulating layer is provided adjacent to the carbon nanotube, and includes an insulating layer / first electrode / insulating layer / second electrode / insulating layer, and a surface of the first electrode and the second electrode facing the carbon nanotube is the insulating layer. electronic devices using the vertically aligned carbon nanotubes, comprising the a laminated electrode is retracted from the end <br/>. Carbon nanotubes oriented perpendicular to the main surface of the substrate ;
At least a pair of electrodes that are arranged to face each other via the carbon nanotubes, and that three or more electrodes are stacked via an insulating layer, and a surface of the electrodes facing the carbon nanotubes recedes from an end of the insulating layer. electronic devices using the vertically aligned carbon nanotubes, comprising the laminated electrode with <br/>. 3. The electronic device using vertically aligned carbon nanotubes according to claim 1, wherein three or more or three or more stacked electrodes are arranged on the circumference.

Images