JPH06260656A - Quantum box assembly device - Google Patents

Abstract

(57) [Summary] [Object] To provide a quantum box assembly element capable of more reliably and easily detecting the distribution of electrons after information processing as output information. [Structure] A plurality of first to third quantum boxes each having a ground state energy E ₀₁ of the first to third quantum boxes on the electrode 1,
E ₀₂ and E ₀₃ are configured so that E ₀₃ > E ₀₂ > E _01, and electron tunneling is disabled between the first quantum box and the third quantum box, and the second quantum box In this mode, electrons can be tunneled between the second quantum boxes, and information processing is performed by changing the electron distribution between the second quantum boxes, and electrons or holes cannot be tunneled from the third quantum boxes, but the Coulomb force extends. A channel layer 9 is provided via a barrier layer 8 having a thickness, and the magnetic field dependence (magnetic fingerprint) of the conductivity between the electrodes at both ends of the channel layer 9 is measured to obtain an electron distribution in the third quantum box. The processing information is output by the processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a quantum box assembly element in which quantum boxes (dots) are formed in an integrated manner and information processing is performed by tunneling electrons between the quantum boxes.

[0002]

2. Description of the Related Art Transistor circuits have made dramatic progress due to their use as ICs, and have become an indispensable part of modern civilization. The basic structure of an IC is formed by connecting transistors having a non-linear function and the like with metal wiring having a linear characteristic. Here, linear means that Ohm's law holds. In this metal wiring part, Fermi degeneracy is established even in a thermal equilibrium system at room temperature, and dissipation of conduction electron energy is large. Therefore, with higher integration, heat generation and delay in wiring become a problem.

Although integration by microfabrication has been studied, it has become difficult to process aluminum vapor deposited with a wiring rule of submicron by ordinary lithography and etching processes. Further, if the wiring is extremely miniaturized, a problem of disconnection due to electro-migration may occur.

[0004]

In order to solve such a problem, the present applicant has proposed that the CPU (central processing unit) currently used in the above-mentioned Japanese Patent Application No. 3-150446.
We proposed a quantum box assembly element formed by combining multiple quantum boxes capable of processing information instead of. The structure of this quantum box assembly element will be described below.

When quantum dots (quantum boxes) each having a side of about 10 nm, which are composed of compound semiconductor heterojunctions, are arranged close to each other, electrons can move between them due to the tunnel effect. The movement of electrons can be controlled by the arrangement of dots. By utilizing this phenomenon, information processing can be performed by the distribution transition of electrons on the quantum dot aggregate, but the specific configuration is still under study. In order to realize the quantum box assembly element proposed by the above-mentioned Japanese Patent Application No. 3-150446, the following configuration is required.

1) Utilizing discrete electronic states in quantum dots. Therefore, the dot diameter needs to have an ultrafine structure of the order of 10 nm.

2) Quantum dots are arranged side by side so that electrons can move between the quantum dots by a tunnel effect. For this reason, the distance between dots must be on the order of about 10 nm. Therefore, a quantum dot in which electrons are confined by a semiconductor depletion layer cannot be used, and a quantum dot in which electrons are confined by a heterojunction of a semiconductor compound has to be used.

3) Electrons to be conducted utilize electron-hole pairs by photoexcitation, which are used as input information. Therefore, it is necessary to prevent the electrons from moving due to the tunnel effect before they are completely input to the entire system of the device by photoexcitation such as laser irradiation.

As a method for controlling this, the present applicant has previously proposed a method capable of performing input with high efficiency in Japanese Patent Application No. 3-265511. This utilizes a structure in which quantum dot arrays are arranged in two layers, one at the top and one at the end of inputting electrons into dots in the upper layer where the coupling between adjacent quantum dots is small and electrons cannot tunnel. Then, an external bias is applied to the quantum dot array in the lower layer to transfer the electrons in the upper layer. If the coupling between adjacent dots in the lower layer is strengthened, the input electron distribution changes due to conduction between the dots in the lower layer.

4) It is also necessary to prevent electron-hole recombination during input or electron conduction. Even when the above-mentioned two-layer array type is adopted, the electron-hole can be spatially separated by an external bias, but as a better method,
A quantum dot assembly device using a type II superlattice structure is proposed in, for example, Japanese Patent Application No. 4-360263 filed by the present applicant.

A type II superlattice is a superlattice in which the conduction band of one semiconductor and the valence band of the other overlap, and is, for example, InAs.
And a GaSb superlattice. This type II
By using the superlattice of, it becomes a quantum dot for electrons, but becomes an antidot for holes, and it is possible to allow the conduction of electrons by tunneling in the quantum box composed of this type II superlattice. A desired electron distribution can be changed by the arrangement of the quantum boxes, and so-called information processing can be performed.

5) Further, it is necessary to detect the distribution of the conducted electrons as output information after a lapse of a certain time. On the other hand, a method of detecting light emission due to pair annihilation of electron-hole pairs can be considered. In the present invention, the problems 1 to 4 described above are solved, and particularly the problem 5 described above, that is, the distribution of electrons after information processing can be more reliably detected as output information.

[0013]

In the present invention, as shown in FIG. 1 which is an enlarged schematic diagram of an example thereof, an electrode 1 and an upper part of the electrode 1 (in FIG. 1, the lower side is shown, and so on) The electron affinity φ ₁ provided on the
And a first semiconductor 11 having an energy gap E _g1
The first well layer 2 made of φ ₂ <φ ₁ and φ ₂ + E
The first plurality of structures surrounded by the first barrier layer 3 composed of the second semiconductor 12 having the electron affinity φ ₂ and the energy gap E _g2 satisfying both relations of _{g 2} <φ ₁ + E _g1 .
A quantum box to Q _1, provided corresponding to said plurality of first quantum box Q ₁ to the first upper barrier layer 2, in the sequence plane, the third semiconductor 13 having an electron affinity phi ₃ The second well layer 4 is φ ₄ <φ ₃ and φ ₃ −φ ₄ <φ ₁ −
_4th with electron affinity φ ₄ satisfying both relations of φ ₂
And a plurality of second quantum boxes Q ₂ each having a structure surrounded by the second barrier layer 5 made of the semiconductor 14.

Further, in the present invention, the second barrier
The plurality of second quantum boxes Q on the layer 5 ₂Corresponding to
The third well layer 6 formed of the first semiconductor 11
Barrier layer 7 made of a semiconductor of_FiveBut
Electron affinity φ of the second semiconductor 12₂Against φ_Five<Φ₂To
The fourth barrier layer 8 made of the fifth semiconductor 15 which satisfies the above
A plurality of third quantum boxes Q surrounded by a structure₃And the fourth above
And the channel layer 9 on the barrier layer 8 of
Source or drain electrode (S / D charge) provided at the end
Poles) 22 and 23.

The above-mentioned fourth barrier layer 8 is
Is the thickness that does not tunnel the holes,
Quantum box Q₁E of the ground state energy of₀₁, The second quantum
Box Q ₂E of the ground state energy of₀₂, The above-mentioned third quantum
Box Q₃E of the ground state energy of₀₃When you say E₀₃> E₀₂> E₀₁ The relational expression of is established.

According to the present invention, in the above structure, the electrode 1 has a shape having an opening 21 at a portion corresponding to the first quantum box Q ₁ .

Furthermore, the present invention has the above-mentioned structure,
E₀₂> E₀₁Apply a bias voltage to electrode 1 under the condition that
The first quantum selected corresponding to the information in the added state
Box Q ₁Information is input by irradiating light on the. And then E
₀₃> E₀₂Holds, and E₀₂> E₀₁Under the condition that
By applying a bias voltage to the electrode 1,
1 quantum box Q₁Second quantum boxes with electron distribution in
Q₂Transfer to and process information. Further E₀₃> E₀₂Is established
By applying a bias voltage to electrode 1 under the condition
Multiple second quantum boxes Q₂The electron distribution in
Quantum box Q of 3₃It is configured to be transferred to the quantum box described above.
Source and drain electrodes as well as application of an ias magnetic field
Bias voltage is applied between 22 and 23 for output
The configuration.

The present invention also provides the first aspect of the present invention in the above-mentioned configuration.
The semiconductor 11 and the third semiconductor 13 are made of the same material, and the widths of the first, second and third quantum boxes Q ₁ , Q ₂ and Q ₃ are respectively d ₁ , d ₂ and d ₃ . Then, it is configured as d ₁ > d ₂ > d ₃ .

Further, the present invention has the above-mentioned structure, wherein the first
And the third and fifth semiconductors 11, 13 and 15 with InA
s, the second and fourth semiconductors 12 and 14 are AlGaS
b, the fifth semiconductor 15 is made of GaSb.

[0020]

In the quantum box assembly element of the present invention, the height (φ ₁ −φ ₂ ) of the potential barrier by the first barrier layer 3 of the first quantum box Q ₁ is set to be relatively large, and a plurality of first quantum boxes Q ₁ are formed. by sufficiently weaken the strength of the bond between the quantum boxes Q _1, in between the first quantum box Q ₁ may be so is not permitted electron conduction by tunneling. The second height of the potential barrier of the second barrier layer 5 of a quantum box Q ₂ (φ ₃ -φ ₄₎ relatively small as well the second strength of the coupling between quantum box Q ₂ By making it strong, it is possible to allow the conduction of electrons by tunneling between the second quantum boxes Q ₂ .

And, in particular, on the second quantum box Q ₂ (see FIG. 1).
On the lower side), that is, a plurality of second barrier layers on the second barrier layer 5.
Of the first semiconductor 11 corresponding to the quantum box Q ₂ of
Of the well layer 6, the third barrier layer 7 and the electron affinity phi ₅ is a fifth semiconductor 15 which satisfies phi ₅ <phi ₂ to the electron affinity phi ₂ of the second semiconductor 12 made of the second semiconductor By providing a plurality of third quantum boxes Q ₃ configured to be surrounded by the fourth barrier layer 8, it is possible to prevent the conduction of electrons by tunneling between the third quantum boxes Q _3. At the same time, the Coulomb interaction of electrons distributed in each third quantum box can reach the channel layer 9 formed thereon via the fourth barrier layer 8.

As described above, for example, "U. Sivan, PM Solomon and H.
Shtrikman, Phys. Rev. Lett. 68 1196 (1992) ”(hereinafter referred to as Reference 1).

In the present invention, the fourth barrier layer 8
Since the source or drain electrodes (S / D electrodes) 22 and 23 are provided at both ends of the upper channel layer 9, a predetermined bias voltage is applied between the source / drain electrodes 22 and 23, and the channel layer 9 It is possible to reliably read out the electron distribution in the third quantum box Q ₃ to the outside by measuring the conductivity of the electron in a magnetic field.

Further, in the present invention, since the opening 21 is provided in the portion of the electrode 1 corresponding to the _first quantum box Q ₁ ,
Even if the material of the electrode 1 is opaque to the input light,
Input can be performed by irradiating the first quantum box Q ₁ with light through the opening 21 as shown by an arrow L.

Further, according to the present invention, the information is obtained by irradiating the _first quantum box Q ₁ selected corresponding to the information with light while the electrode 1 is biased under the condition that E ₀₂ > E ₀₁ is satisfied. Is input, and the electrode 1 is biased under the condition that E ₀₃ > E ₀₂ is satisfied and E ₀₂ > E ₀₁ is not satisfied, so that the electron distribution in the plurality of _first quantum boxes Q ₁ is changed to the plurality of second quantum boxes Q ₁ . It is transferred to quantum box Q _2, by forming a quantum box Q ₂ of the second as a predetermined pattern, each of the second quantum boxes Q
Information processing can be performed by changing the electron distribution by tunneling between the _two .

Then, the electron distribution in the plurality of _second quantum boxes Q ₂ is transferred to the plurality of third quantum boxes Q ₃ by biasing the electrode 1 under the condition that E ₀₃ > E ₀₂ is not satisfied,
By applying a bias magnetic field to the above-mentioned quantum box and applying a bias voltage between the source and drain electrodes 22 and 23, the conductivity of the channel layer 9 is measured in the magnetic field, whereby the distribution of electrons is determined. It is possible to read out, and thus the processing result can be output reliably.

Further, in the above structure, the first semiconductor 1
When the first and third semiconductors 13 are made of the same material, and the widths of the first, second and third quantum boxes Q ₁ , Q ₂ and Q ₃ are d ₁ , d ₂ and d ₃ , respectively. Then, by configuring as d ₁ > d ₂ > d ₃ , the above E ₀₃ > E ₀₂ > E ₀₁
The condition of can be easily realized.

Furthermore, in the above structure, the first and third semiconductors 11 and 13 are made of InAs and the second semiconductor 1 is made of InAs.
2 is AlGaSb, the fourth semiconductor 14 is GaSb, the fifth
By configuring the semiconductor 15 of (1) as GaSb, it is possible to have a configuration in which they are lattice-matched to each other, and it is possible to obtain a quantum box assembly device having a more stable structure.

[0029]

Embodiments of the present invention will now be described in detail with reference to the drawings. In this example, as shown in FIG. 1, on a substrate 20 made of GaSb or the like on which a channel layer 9 is provided,
The third quantum box Q ₃ , the second quantum box Q ₂ , the first quantum box Q ₁ and the electrode 1 having openings 21 having a pattern corresponding to the distribution of the _first quantum box Q ₁ are sequentially laminated and formed. The first and third semiconductors 11 and 13 are made of InAs, the second semiconductor is made of AlGaSb, the fourth semiconductor is made of GaSb, and the fifth semiconductor is made of AlSb. It was done.

FIG. 2 shows the band gap in the cross section indicated by broken lines α and γ in the direction along the surface of each layer of the first and third quantum boxes Q ₁ and Q ₃ in this case. In FIG. 2, the energy at the lower end of the conduction band of the first semiconductor 11 is represented by E
_CW1 , the energy at the top of the valence band is shown as E _VW1 , the energy at the bottom of the conduction band of the second semiconductor 12 is shown as E _CB1 , and the energy at the top of the valence band is shown as E _VB1 . in this case,
The first and third quantum boxes Q ₁ and Q ₃ composed of the first and second semiconductors 11 and 12 have electron affinities φ ₁ and φ ₂ and an energy gap of E _g1 and E _g2 are φ ₂ <Φ ₁ and φ ₂ + E _g2 <φ ₁ + E _g1 .

At this time, the barrier height h _{1 of} the first or third barrier layer made of AlGaSb, that is, the first semiconductor 1
1 electron affinity φ ₁ and second semiconductor 12 electron affinity φ
The difference phi ₁ -.phi ₂ and ₂ can be varied to 0.8~1.5eV by changing the mixed crystal ratio.

As described above, in each of the layers indicated by the broken lines α and γ, the tunnel height between the electron quantum boxes can be prevented by increasing the barrier height h ₁ . In the layer in which the first quantum box Q ₁ shown by the broken line α is formed (hereinafter referred to as the α layer), electrons are input as described in detail later, and the third quantum box Q ₃ shown by the broken line γ is formed. The layer (hereinafter referred to as the γ layer) to be formed is configured to output the electron distribution.

FIG. 3 shows the band gap in the cross section indicated by the broken line β in the direction along the arrangement of the _second quantum boxes Q ₂ provided corresponding to the _first quantum boxes Q ₁ . In FIG. 3, the energy at the lower end of the conduction band of the third semiconductor 13 is E _CW3 , the energy at the upper end of the valence band is E _VW3 , the energy at the lower end of the conduction band of the fourth semiconductor 14 is E _CB4 , and the valence band. The energy at the top of is shown as E _VB4 .

In the second quantum box Q ₂ , the third and fourth semiconductors 13 and 14 have barrier heights h ₂ ,
That is, the difference in electron affinity φ ₃ −φ _{4 is set} to be smaller than the barrier height h ₁ in the first and third quantum boxes Q ₁ and Q ₃ . That is, φ ₄ <φ ₃ and φ ₃ −φ ₄ <φ ₁
-Φ ₂ (h ₂ <h ₁ ). In this case, the barrier height of the second barrier layer 5 made of GaSb is about 0.8 eV, and the coupling between the quantum boxes is increased by designing the interval between the _second quantum boxes Q ₂ to be small. You can
Therefore, electrons can tunnel between the quantum boxes in the layer in which the second quantum boxes Q ₂ are formed, which is indicated by the broken line β (hereinafter referred to as β layer), and can be used as a layer for changing the distribution of electrons.

FIG. 4 shows the band gap in the vertical cross section indicated by the broken line δ that crosses each of the first to third quantum boxes Q _{1 to} Q ₃ . E _CW5 the bottom energy of the conduction band of the fifth semiconductor 15 in FIG. 4, the energy of the top of the valence band as E _VW5, showing the right electrode 1 side, the left direction as the substrate 20 side in FIG. 4 .

Ground energies of the first to third quantum boxes
E₀₁, E₀₂And E₀₃Is the state where no voltage is applied to electrode 1.
Then E₀₃> E₀₂> E₀₁ Is configured to be. That is, from α layer to β layer, γ layer
As the lower layer, the base energy increases
Done. In this example, as described above, each well layer
The materials 2, 4, and 6 are made of the same material, for example, InAs.
And the width d of the first to third quantum boxes₁~ D₃That is, first to
The thickness d of the third well layers 2, 4 and 6₁, D ₂And d
₃, D₁> D₂> D₃ As described above, the ground energy is formed as
The size of the ghee is selected.

By changing the ground energy of each well layer in this way, selective electron excitation in the quantum box,
Electrons can move between the layers.

Further, as shown in FIG. 4, the third well layer 6 is formed.
The (γ layer) has an electron affinity φ ₂ on the substrate 20 side.
In contrast, a material satisfying φ ₅ <φ _{2 is formed in contact} with the fourth barrier layer 8 made of the fifth semiconductor 15 such as AlSb in this case. The configuration is such that tunneling cannot be performed beyond.

That is, each of the layers 2 to 8 has the above-mentioned structure.
Electron affinity of each of the semiconductors 11 to 15, that is,
The electron affinity of the first to third well layers 2, 4 and 6 is φ.
₁(= Φ ₃), The electronic parents of the first and third barrier layers 3 and 6
Sum power φ₂, The electron affinity of the second barrier layer 5 is φ_Four, 4th
The electron affinity of the barrier layer 8 of_FiveThen, the magnitude relationship
Is φ_Five<Φ₂<Φ_Four<Φ₁Becomes

Then, as shown in FIG. 1, the fifth semiconductor 1
A channel layer 9 is provided on the fourth barrier layer 8 (lower side in FIG. 1) and the S / D electrodes 22 and 23 are formed on both ends of the channel layer 9.

As described above, in this case, the fourth barrier layer 8 in which the electrons in the well layer 6 of the third quantum box Q ₃ cannot tunnel is provided, and particularly the third well layer 6 is provided. The thickness of the fourth barrier layer 8 is selected so that the Coulomb interaction of electrons in 6 is transmitted to the channel layer 9 provided below the fourth barrier layer 8. As described above, the formation of the potential barrier that extends the Coulomb force but blocks the electron tunneling is described in the above-mentioned Document 1.

The principle of operation in such a configuration will be described below. Hereinafter, the energies E _VW1 , E _VW2 , and E at the upper end of the valence band of each well layer 2, 4 and 6 made of InAs are _shown .
_VW3 is H ₀ , the potential of the upper electrode 1 is V _g , the potential of the channel layer 9 is V _e , and the voltage applied to one S / D electrode 23 in this case is V _e .

Information is input by the above-mentioned Japanese Patent Application No. 4-3602.
It can be carried out by a method similar to the method described in the 63 application. First, the potential V of the upper electrode 1
_G is biased slightly negative while maintaining the condition of E ₀₁ <E ₀₂ . In this state, the aperture 21 corresponding to the first quantum box 1 in which electrons are to be generated is irradiated with monochromatic light satisfying hν _in = E ₀₁ −H ₀ .

At this time, the band gap in the vertical cross section indicated by the broken line δ is shown in FIG.
The electron (e) -hole (h) pair is excited only in the quantum box Q _{1 of} . The holes h are absorbed on the side of the electrode 1 to which the negative bias is applied, that is, on the right side in FIG. As a result, electrons are input into the first quantum box Q ₁ of the α layer. The barrier of this first quantum box Q ₁ , that is, the first
Since the barrier height of the barrier layer 3 is set to be relatively high, tunneling is not performed between the adjacent first quantum boxes Q ₁ , and the electron distribution can be fixed during input.

Then, E ₀₃ > E ₀₂ holds and E ₀₂
A bias voltage is applied to the upper electrode 1 so that the condition that> E ₀₁ is not satisfied is satisfied. At this time, the electron e transits to the _second quantum box Q ₂ of the β layer, as shown in the band gap in the vertical section indicated by the broken line δ, as shown in FIG. γ
The energy level E ₀₃ of the third quantum box Q ₃ represented by the layer is β
Since it is higher than the energy level E _{02 of the} layer, the third quantum box Q
_The electron e does not transition up to ₃ .

In this case, in particular, in the plane of the β layer, as shown in FIG. 3, the barrier height between the adjacent second quantum boxes Q ₂ , that is, the barrier of the second barrier layer 5 is set. Is selected low, the electron e can tunnel between the quantum boxes Q ₂ . Tunneling of electrons, since according to the arrangement of the quantum box Q _2, by forming a quantum box Q ₂ of the second with a pattern corresponding to the wiring pattern for example in a conventional central processing unit such as a change in the electron distribution With this, information processing can be performed.

Next, the output mode will be described. Upon output, the electron distribution is fixed in order to output changes in the electron distribution. Therefore, first, a bias voltage is applied to the electrode 1 under the condition that E ₀₃ > E ₀₂ is not established. At this time, as shown in FIG. 7 for the band gap in the vertical cross section indicated by the broken line δ, the electrons e confined in the _second quantum box Q ₂ of the β layer are transferred to the third quantum box Q ₃ indicated by the γ layer. Transition. In the plane of the γ layer, as described in FIG. 2 above, since the barrier between the quantum boxes is high, electrons cannot be tunneled.
The distribution of the electrons transferred to the quantum box Q ₃ of is fixed.

At this time, the channel layer 9 has a third
According to the distribution of the electrons e confined in the quantum box Q _{3 of} , the Coulomb potential that reflects the electron distribution is working. In this state, the magnetic field B is applied to the channel 9.
It is desirable to apply this magnetic field in the film thickness direction of the channel layer 9. Then, a bias voltage V _e is applied between the S / D electrodes 22 and 23 to conduct the holes h and change the magnetic field B to conduct the channel layer 9 as shown in the schematic illustration in FIG. The magnetic field dependence of the conductivity of the channel layer 9 is measured.

As is well known, when the conductivity of a system in which the potential is locally changed in this way, for example, a substance containing impurity ions or the like is measured in a magnetic field,
A spatial variation of the potential, for example, a characteristic magnetic field pattern corresponding to the spatial arrangement of impurity ions is obtained as a so-called “magnetic fingerprint” as shown in an example in FIG. 9 (for example, “ABFowler, A. Hartsein and RAWe.
bb, Phys. Rev. Lett. 48 , p.196 (1982) ”).

Therefore, from the magnetic field dependence of the conductivity measured as described above, the information of the electron distribution in the _third quantum box Q ₃ can be taken out as an output. In this case, the magnetic fingerprint shows a completely different pattern depending on the presence or absence of electrons in one quantum box of the third quantum boxes Q ₃ , and the pattern can be obtained in advance by theoretical calculation, for example. . Therefore, it is possible to reliably know the electron distribution after information processing collectively by measuring the magnetic fingerprint once.

Next, an example of a method of manufacturing such a quantum box assembly element of the present invention will be described. First, as shown in FIG. 10 as one step, a thickness d is formed on a substrate 20 made of GaSb or the like.
_4th barrier layer 8 of AlSb, In of thickness d ₃
Third well layer 6 made of As, AlGaSb having a thickness d ₂₃
A third barrier layer 7 made of InAs, a second well layer 4 made of InAs having a thickness d _{2, and} a second well layer made of AlGaSb having a thickness d ₁₂ .
Of the barrier layer 5, the first well layer 2 of InAs having a thickness of d ₁ and the first barrier layer 3 of AlGaSb having a thickness of d ₀ are sequentially formed by MBE (molecular beam epitaxy) and MOCVD (chemical metalorganic chemistry). Vapor phase growth method), MOMBE (molecular beam epitaxy method using an organic metal), and the like.

At this time, as described with reference to FIG. 4 above, in order to satisfy the condition E ₀₁ <E ₀₂ <E ₀₃ of the base energy of the InAs layers forming the well layers 2, 4, and 6,
Each layer is epitaxially grown so that the thicknesses d ₁ , d ₂ and d ₃ satisfy the relational expression d ₁ > d ₂ > d ₃ . For example, when configured as d ₁ = 20 nm d ₂ = 15 nm d ₃ = 13 nm, the electron energy in each well layer is H ₀ , where E is the energy at the bottom of the conduction band in the well layer of each quantum box. ₀₁ a _{-H 0 ≒ 0.1eV E 02 -H 0} ≒ 0.2eV E 03 -H 0 ≒ 0.3eV. In this example, the thickness of each of the other layers is 10
It was formed in the order of nm.

Then, after that, by lithography using an electron beam or STM (scanning tunneling electron microscope),
A mask 31 having dry etching resistance is formed as shown in FIG. For example, in the case of electron beam lithography, a negative resist or a so-called contamination resist that uses an organic substance diffused on the sample surface can be used. The contamination resist is an organic substance leaking from an exhausting means such as a diffusion pump, or an organic substance such as TMA (trimethylgallium), which is flowed on a sample and irradiated with an electron beam to solidify hydrogen carbide. It is used as a resist by adhering to the surface.

After that, the AlGaSb layer and InAs are formed.
For the layer, a gas such as (CH + He) or (SiCl + He) is used, and highly anisotropic dry etching such as RIE (reactive ion etching) or ECR-RIBE is performed.
By a method such as (electron cyclotron resonance-reactive ion beam etching), as shown in FIG.
Selective etching is performed as a columnar pattern corresponding to the quantum box pattern using the mask 31 as a mask until the AlSb layer on the substrate is exposed.

At this time, as shown in the process charts of FIGS.
Patterning using a positive resist
By doing so, a finer pattern can be more reliably formed in the semiconductor.
It can be done without causing damage to. FIG.
In A to C, parts corresponding to those in FIG.
The duplicate description is omitted. For example, as shown in FIG. 13A
Of the second semiconductor 12 made of AlGaAs
On top, SiO₂And Si ₃N_FourInsulator protection layer 32 such as
In terms of area, for example, photo-CVD which causes less damage to semiconductors,
Alternatively, a thickness of, for example, several tens of nanometers is formed by usual plasma CVD or the like.
Wear as m.

Then, an electron beam positive type resist such as PMMA (polymethylmethacrylate) is applied, and pattern exposure and development are performed to form a positive type resist 33 having a predetermined pattern as shown in FIG. 13B. . In this case, for example, a pattern having an opening corresponding to a quantum box pattern having an opening width of about 10 nm is formed. Then, a metal such as Au-Pd having a grain size as small as possible is deposited over the entire surface by vacuum vapor deposition or the like to have a thickness of, for example, several tens nm.

Then, the lower semiconductor layer is (CH ₄ + He) or (SiCl ₄ +) by using the metal layer 34 as a mask.
Each layer is patterned by dry etching having a strong anisotropy such as RIE using an etching gas such as He).

Thereafter, although not shown, the insulator protection layer 32 is removed by wet etching such as HF which causes less damage, and the metal layer 34 on the insulator protection layer 32 is also removed to obtain a structure similar to that shown in FIG. Can be patterned.

By using the positive resist in this manner, fine patterning of several tens of nm or less can be performed with higher accuracy, and the metal layer 34 serving as an etching mask can be formed through the insulator protective layer 32. Therefore, it is possible to avoid the fluctuation of the semiconductor characteristics due to the residual metal mask, and it is possible to further reduce the damage because the semiconductor is not directly irradiated with the electron beam.

Next, as shown in FIG. 14, AlGaSb
And the thickness d ₂₃ of the third barrier layer 6 on the _third well layer 6 made of InAs, which is obtained by patterning the third barrier layer 7 made of Al. epitaxially grown by MOCVD or the like as a thickness corresponding to ₃ + d _23), its upper surface, to be substantially flush with the third upper surface of the barrier layer 7 of columnar patterns. MOCVD, MBE
In epitaxial growth such as, it is possible to accurately control the thickness in the order of nm or less. In FIG. 14 and FIGS. 15 to 21 below, portions corresponding to those in FIG.

Subsequently, as shown in FIG. 15, a second barrier layer 5 made of GaSb or the like is patterned into a columnar In pattern.
The second well layer 4 made of As is epitaxially grown by MOCVD or the like with a thickness similar to the thickness d ₂ of the second well layer 4, that is, the upper surface of the second well layer 4 is controlled to be substantially flush with the upper surface of the second well layer 4.

Further, as shown in FIG. 16, AlGa
The first barrier layer 3 made of Sb or the like, the first barrier layer 3 and the first well layer 2 in which the thickness is patterned into the columnar shape described above.
And the sum of the thicknesses of the third barrier layers, d ₁₂ + d ₁ + d _0, are controlled so that the upper surface thereof is substantially flush with the upper surface of the uppermost columnar first barrier layer 3, and MOCVD or the like is performed. Epitaxially grow.

Next, as shown in FIG. 17 which is a process drawing seen from the upper surface, a resist 35 is formed by patterning so as to cover the region where the quantum box assembly element is to be formed. Then, using this resist 35 as a mask, patterning is performed using an etching gas such as (CH ₄ + He) or (SiCl ₄ + He) until the substrate 20 is exposed, as shown in FIG. The box collecting unit 27 is formed.

Then, as shown in FIG. 19, a resist 36 covering the upper portion and the side wall of the quantum box assembly 27 is formed by patterning, and a metal of Al, Au or the like is deposited on the resist 36 by vacuum deposition or the like. . And the resists 35 and 3
6 is removed by wet etching and the metal layer deposited thereon is removed by lift-off. As shown in FIG. 20, the S / D electrodes 22 are formed on both sides of the quantum box assembly 27, for example.
And 23 are formed. It is known that the base 21 made of GaSb is non-alloy and can make ohmic contact with holes.

Next, as shown in FIG. 21, a resist 37 having a pattern covering the periphery of the quantum box assembly 27 and the peripheral portion of the upper surface thereof is formed by patterning, and a metal layer made of Al, Au, or the like is entirely formed from above. By vacuum deposition etc.,
The resist 37 is removed by wet etching or the like, the mask layer 31 made of SiO ₂ or the like is removed by wet etching or the like, and the metal layer deposited thereon is lifted off. As shown in FIG. The electrode 1 having the openings 21 having a pattern corresponding to the quantum boxes 3 is formed. Al
It is known that the first barrier layer 3 made of GaSb is also non-alloy and can make ohmic contact with holes, and the quantum box assembly element according to the present invention can be obtained by the above manufacturing process.

Further, AlSb forming the fourth barrier layer 8
Is undoped and tends to have p-type characteristics.
By applying a predetermined gate voltage to the two-dimensional hole gas 10 in the channel layer 9 between the S / D electrodes 22 and 23.
As described above, the electron distribution in the third quantum box can be known by measuring the magnetic field dependence of the conductivity.

The present invention is not limited to the above-mentioned embodiments, but the semiconductor material, the electrode material and the like are changed, and the fourth barrier layer is made of AlGaSb, for example. It goes without saying that various modifications and changes based on the above are possible.

[0068]

As described above, according to the present invention, in the quantum box assembly element, holes among the electron-hole pairs generated in the plurality of first quantum boxes at the time of input by light irradiation are converted into holes. It is possible to avoid the fluctuation of the electron distribution at the time of input by making it possible to leave only electrons in the first quantum boxes by sucking them into the electrodes and to prevent the quantum mechanical tunneling between the quantum boxes. At the same time, desired information processing is performed here by enabling tunneling of electrons between the second quantum boxes capable of moving electrons by applying a predetermined bias voltage from the first quantum box. You can

In particular, in the present invention, electrons are not tunneled in the third quantum box capable of moving electrons by applying a predetermined bias voltage from the second quantum box, and this third quantum box is used. It is impossible to tunnel electrons in contact with the box, but a fourth barrier layer is provided so that the Coulomb force can be exerted, and a channel layer is provided through this fourth barrier layer, and the magnetic field dependence of conductivity in the channel layer, Since a so-called magnetic fingerprint is measured and an output is obtained, the distribution of electrons existing in the third quantum box is collectively processed without affecting the third quantum box, and a more accurate resolution is obtained. Output information can be obtained with.

[Brief description of drawings]

FIG. 1 is a schematic enlarged perspective view of an embodiment of the present invention.

FIG. 2 is a diagram showing a band gap in a cross section (α, γ) of an example of the present invention.

FIG. 3 is a diagram showing a band gap in a cross section (β) of an example of the present invention.

FIG. 4 is a diagram showing a band gap in a vertical cross section (δ) of an example of the present invention.

FIG. 5 is a diagram showing a bandgap for explaining an operation mode of the embodiment of the present invention.

FIG. 6 is a diagram showing a band gap used for explaining an operation mode of the embodiment of the present invention.

FIG. 7 is a diagram showing a band gap used for explaining an operation mode of the embodiment of the present invention.

FIG. 8 is an explanatory diagram of an operation mode according to the embodiment of the present invention.

FIG. 9 is a diagram showing an example of magnetic field dependence of conductivity corresponding to electron distribution in a quantum box.

FIG. 10 is a manufacturing process diagram of an example of the present invention.

FIG. 11 is a manufacturing process diagram of an example of the present invention.

FIG. 12 is a manufacturing process diagram of an example of the present invention.

FIG. 13 is a manufacturing process diagram of an example of the present invention.

FIG. 14 is a manufacturing process diagram of an example of the present invention.

FIG. 15 is a manufacturing process diagram of an example of the present invention.

FIG. 16 is a manufacturing process drawing of the example of the present invention.

FIG. 17 is a manufacturing process drawing of the example of the present invention.

FIG. 18 is a manufacturing process drawing of the example of the present invention.

FIG. 19 is a manufacturing process diagram of an example of the present invention.

FIG. 20 is a manufacturing process diagram of an example of the present invention.

FIG. 21 is a manufacturing process diagram of an example of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Electrode 2 1st well layer 3 1st barrier layer 4 2nd well layer 5 2nd barrier layer 6 3rd well layer 7 3rd barrier layer 8 4th barrier layer 9 Channel layer 10 2 dimensional Hole gas 11 First semiconductor 12 Second semiconductor 13 Third semiconductor 14 Fourth semiconductor 15 Fifth semiconductor 20 Base 21 Open 22 S / D electrode 23 S / D electrode

Claims (5)

[Claims] 1. An electrode and a first well layer made of a first semiconductor having an electron affinity φ ₁ and an energy gap E _g1 provided on the electrode are φ ₂ <φ ₁ and φ ₂ + E _g2 < a plurality of first quantum boxes having a structure surrounded by a first barrier layer made of a second semiconductor and having an electron affinity φ ₂ and an energy gap E _g2 satisfying both relations of φ ₁ + E _g1 ; A second well layer made of a third semiconductor having an electron affinity of φ ₃ is provided on the first barrier layer corresponding to the plurality of first quantum boxes, and the second well layer is φ ₄ <φ ₃
And a plurality of second quantum boxes having a structure surrounded by a second barrier layer made of a fourth semiconductor and having an electron affinity φ ₄ satisfying both relations of φ ₃ −φ ₄ <φ ₁ −φ ₂ . A third well layer made of the first semiconductor and provided on the second barrier layer corresponding to the plurality of second quantum boxes is a third barrier layer made of the second semiconductor. And the electron affinity φ ₅ is φ compared to the electron affinity φ ₂ of the second semiconductor.
A plurality of third quantum boxes having a structure surrounded by a fourth barrier layer made of a fifth semiconductor satisfying ₅ <φ ₂ , a channel layer on the fourth barrier layer, and both ends of the channel layer. The fourth barrier layer has a thickness that does not allow electrons or holes to tunnel, the ground state energy of the first quantum box is E ₀₁ , and the second quantum layer is The energy of the box ground state is E ₀₂ , the third above
The quantum box assembly element is characterized in that the relational expression of E ₀₃ > E ₀₂ > E ₀₁ is satisfied when the ground state energy of the quantum box is E ₀₃ . 2. The quantum box assembly element according to claim 1, wherein the electrode has an opening in a portion corresponding to the first quantum box. 3. The information is input by irradiating the first quantum box selected corresponding to information with light while applying a bias voltage to the electrode under the condition that E ₀₂ > E ₀₁ is satisfied. , E ₀₃ > E ₀₂ holds, and E ₀₂ > E ₀₁ does not hold, a bias voltage is applied to the electrodes so that the electron distribution in the plurality of first quantum boxes is changed to the plurality of second quantum boxes.
Information processing is performed by transferring the electron distribution in the plurality of second quantum boxes to the plurality of third quantum boxes by applying a bias voltage to the electrodes under the condition that E ₀₃ > E ₀₂ is not established. The transfer to a quantum box, a bias magnetic field is applied to the quantum box, and a bias voltage is applied between the source electrode and the drain electrode for output. The described quantum box assembly element. 4. The first and third semiconductors are made of the same material, and the widths of the first, second and third quantum boxes are d ₁ , respectively.
The quantum box assembly element according to claim 1, wherein d ₁ > d ₂ > d ₃ when d ₂ and d ₃ . 5. The first and third semiconductors are InAs,
2. The quantum box assembly element according to claim 1, wherein the second semiconductor is AlGaSb, the fourth semiconductor is GaSb, and the fifth semiconductor is AlSb.