JPH0745878A - Semiconductor combination superconductive element - Google Patents

Abstract

PURPOSE:To increase the quantization unit of conductance of a propagation channel which changes discontinuously due to quantization for the continuous change in the gate voltage of a quantization point contact element and to control the quantization unit of conductance. CONSTITUTION:A drain electrode 4 is constituted by a superconductor and the distance between the drain electrode 4 and a split type metal electrode 10 is reduced as compared with the mean free path of a two-dimensional electron gas 6. Then, an electrode 11 for applying magnetic field is formed on the drain electrode 4 via an insulation film 12.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor superconducting coupling element in which a two-dimensional electron gas formed in a semiconductor heterojunction portion is in contact with a superconductor.

[0002]

2. Description of the Related Art In the vicinity of a heterojunction interface of a plurality of compound semiconductor thin films stacked on a compound semiconductor substrate, a two-dimensional electron gas confined within a width at the junction surface is formed by space charges. Research on electronic devices using this two-dimensional electron gas as a carrier is progressing. As a typical semiconductor electronic device, HEMT (High
h Electron Mobility Trans
istor). In this HEMT, three electrodes of a source / drain and a gate are provided in a laminated thin film compound semiconductor, and a voltage is applied to the gate to change the two-dimensional electron gas concentration under the gate electrode to control the source / drain current. I am trying.

Here, by providing a gap in this gate electrode,
There are quantum point contact devices that confine and control a two-dimensional electron gas as a carrier in a smaller area.
FIG. 10 is a perspective view showing an example of a quantum point contact device having a structure in which the gate electrode is divided by providing a minute gap, that is, a split type gate electrode. In FIG. 10A, 1 is a semi-insulating GaAs substrate, 2 is a buffer layer, 3 and 4 are source and drain electrodes made of AuGeNi or the like, and 5 is undoped GaA.
s layer, 7 is a non-doped AlGaAs layer, 8 is an AlGaAs layer introduced with an n-type impurity, and 9 is a non-doped A
The lGaAs layer 10 is a split type metal gate electrode having a gap at a predetermined position.

In such a structure, the GaAs layer 5 and the Al
On the GaAs layer 5 side of the interface of the GaAs layer 7, n-type AlGa
The presence of the As layer 8 forms the two-dimensional electron gas 6.
Here, when a voltage is applied to the split type metal gate electrode 10, a depleted depletion region 11 is formed as shown in FIG. 10B, and the two-dimensional electron gas 6 in the region below the split type metal gate electrode 10 is formed. Will be completely depleted and will no longer function as a carrier. Therefore, the two-dimensional electron gas 6 in the non-depleted region under the gap of the split type metal gate electrode 10 becomes a conduction channel. At this time, the two-dimensional electron gas 6 on both sides of the split type metal gate electrode 10 is
The connection is made under the gap of the split type metal gate electrode 10.

Here, the split type metal gate electrode 10
Width W and length L of the two-dimensional electron gas 6 as a conduction channel left when a voltage is applied in the gap of
Is sufficiently shorter than the mean free path l _e of this carrier,
Further, when the width W is equal to or smaller than the Fermi wavelength λ _{F of the} electron, this conduction channel can be regarded as pseudo one-dimensional. This is called a quantum point contact, and in such a structure, it is shown in Reference 1 that the conductance G of the conduction channel changes discontinuously (Reference 1: BJ van Wees et al. Physical Review Letters, V
ol. 60, No. 9, p848, 1988). This is because, as shown in FIG. 11, the conductance G becomes discontinuous with the quantization unit "2" with respect to a certain change in the gate voltage.
e ² / h ”. Here, e is the charge of the electron, h is the Planck's constant, and the quantization unit is a universal size.

Since the width of the depletion region 11 is changed by the voltage applied to the split type metal gate electrode 10, the conduction channel width W is also changed by this, and the conductance G is also changed by this, but in a certain voltage region. It becomes discontinuously quantized. As described above, in this certain voltage region, the width W and the length L of the conduction channel are
It is sufficiently shorter than the mean free path l _e of this carrier, and the width W is about the same as or less than the Fermi wavelength λ _{F of the} electron.

[0007]

Since the conventional configuration is as described above, there is a problem that it is not possible to increase the quantization unit of the conductance of the conduction channel which is quantized and changed with respect to the gate voltage. It was Further, there is a problem in that it is impossible to control the quantization unit of the conductance, that is, it is impossible to control the amount of change in one step of the conductance that is quantized and changes stepwise.

The present invention has been made in order to solve the above problems, and the quantum point contact element is quantized and discontinuously changed in response to a continuous change in gate voltage. The purpose is to increase the quantization unit of the conductance of the channel. And, it aims to be able to control the quantization unit of the conductance.

[0009]

A semiconductor-coupled superconducting device according to the present invention is formed on a semiconductor substrate and serves as a carrier.
Semiconductor layer laminated body having a heterojunction surface for forming a three-dimensional electron gas, a gate electrode formed on the semiconductor layer laminated body via an insulating layer and having a minute gap at a predetermined position, and a semiconductor layer A first electrode formed to connect with a two-dimensional electron gas formed on the heterojunction surface of the stack, connected with the two-dimensional electron gas, and shorter than the mean free path of the gate electrode and the two-dimensional electron gas The second electrode is formed at intervals and is made of a superconductor. It is also characterized in that it has a third electrode for applying a magnetic field to the interface between the two-dimensional electron gas and the second electrode. On the other hand, the semiconductor device is characterized by having a fourth electrode formed on the back surface of the semiconductor substrate so as to face the gate electrode.

[0010]

At the interface between the two-dimensional electron gas and the second electrode, when one electron whose energy based on the Fermi level is within the superconducting gap enters the second electrode from the two-dimensional electron gas, At the second electrode, a Cooper pair composed of two electrons is generated and becomes a carrier. Another electron is required to generate this Cooper pair, but this electron becomes a hole and is generated on the semiconductor side where the two-dimensional electron gas exists. That is, Andreev reflection that reflects holes occurs on the semiconductor side. Due to these, if this Andreev reflection occurs 100%, the conductance becomes 2
Doubled.

In the two-dimensional electron gas with which the second electrode is in contact, a pair potential in which a Cooper pair can exist oozes out from the second electrode side, and the amount of this oozing out is the third electrode. It is changed by the magnetic field. On the other hand, when a negative voltage is applied to the gate electrode from the fourth electrode facing the gate electrode, the electron concentration of the two-dimensional electron gas decreases.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. Example 1. FIG. 1 is a perspective view and a plan view showing the structure of a semiconductor-coupled superconducting device (hereinafter referred to as a quantum point contact device) which is an embodiment of the present invention. In the figure, 1 is an InP substrate, 2 is undoped InAlAs
Layer, 3 is a source electrode (first electrode) which is a normal conductor made of AuGeNi, 4 is a drain electrode (second electrode) which is a superconductor, and 5 is non-doped InGaAs.
Layer, 6 is a two-dimensional electron gas formed on the surface of the InGaAs layer 5, 7 is a non-doped InAlAs layer, and 8 is n-type I
nAlAs layer, 9 is a non-doped InAlAs layer (insulating layer), 10 is a split type metal gate electrode having a gap at a predetermined position, 11 is an electrode for applying a magnetic field (third electrode), 12 is a drain electrode 4 and an electrode 11 is an insulating film for electrically separating the same.

In the above configuration, the two-dimensional electron gas 6
Is I at the interface between the InGaAs layer 5 and the InAlAs layer 7.
The n-type InAlAs layer 8 is formed on the nGaAs layer 5 (semiconductor layer stack). On the other hand, the source electrode 3 is not limited to the normal conductor but may be a superconductor. As the drain electrode 4, for example, Nb is an absolute temperature of 9
If used below K, it will be in a superconducting state, so this Nb can be used and the absolute temperature can be brought to 4.2 K with liquid helium or the like. In this quantum point contact device, an InAlAs layer 9 serving as an insulating layer and a split type metal electrode 10 constitute a MIS type gate, while a current is passed through the electrode 11 to generate a magnetic field.

The element structure of the first embodiment is different from the conventional quantum point contact element in that the drain electrode 4 is an electrode made of a superconductor and that an electrode 11 for applying a magnetic field is newly provided. Is. The distance between the drain electrode 4 and the split-type metal electrode 10 is smaller than the mean free path of the two-dimensional electron gas 6. The electron concentration N _s and the mobility μ of the two-dimensional electron gas 6 having the structure of Example 1 are N _s = 2.1 × 10 ¹² cm ⁻² and μ = 87200c at an absolute temperature of 4.2 K, respectively.
m was ² / V _s. The Fermi wavelength λ _F of the two-dimensional electron gas 6 is 17.3 nm, and the mean free path l _e is 2.1 μ.
m. Therefore, the length L of the conduction channel (two-dimensional electron gas 6) formed in the region under the gap of the split-type metal gate electrode 10 when the voltage is applied is 2.1 μm.
When the width W is sufficiently smaller and the width W is about 17.3 nm or less, the structure of Example 1 becomes a quantum point contact.

Here, at the interface where the two-dimensional electron gas 6 of the InGaAs layer 5 and the drain electrode 4 of the superconductor are coupled, the absolute value of energy based on the Fermi level of the secondary electron gas 6 is superconducting. When electrons with energy less than the body gap enter, Cooper pairs are generated on the superconductor side, and Andreev reflection occurs in which holes are reflected on the semiconductor side. In a semiconductor, one electron or one hole serves as a carrier for showing conductivity, but in a superconductor, an electron 2
The Cooper pair, which is a pair of pieces, becomes the carrier. FIG. 2 schematically shows this.

In normal operation, since almost no voltage is applied between the source electrode 3 and the drain electrode 4, I
The electrons of the secondary electron gas 6 which is the conduction channel of the nGaAs layer 5 have energy close to the Fermi level of the superconductor. Therefore, the electrons of the secondary electron gas 6 when the quantum point contact device of FIG. 1 is operated are incident on the superconductor, as shown in FIG. Then, in the drain electrode 4, which is a superconductor, a Cooper pair composed of two electrons is generated by the incidence of one electron, while in the InGaAs layer 5 in which the secondary electron gas 6 exists,
Generates individual holes. In this Andreev reflection, one electron is incident and one Cooper pair consisting of two electrons is generated, so two net charges are conducted,
The conductance G of the element including the interface is twice the conductance of the electrode made of a normal normal conductor.

Assuming that the Andreev reflection probability at the interface between the drain electrode 4 and the InGaAs layer 5 is A, the conductance G between the source electrode 3 and the drain electrode 4 is "4e2 / h" as shown in FIG. It is quantized in units of “× A” and takes discontinuous values. Therefore, according to the first embodiment, the conductance G between the source electrode 3 and the drain electrode 4 is 2A times the quantization unit in the conventional quantum point contact element.

By the way, when the superconductor and the semiconductor are in contact with each other, a region where a Cooper pair can exist is actually formed on the semiconductor side. FIG. 4 shows this state, in which the drain electrode 4 as a superconductor and the In as a semiconductor are used.
FIG. 4 is a state diagram showing a pair potential state showing a region where a Cooper pair can exist near the interface with the GaAs layer 5 (two-dimensional electron gas 6). Here, Z is the drain electrode 4 and I
It is a parameter indicating the scattering potential barrier at the interface with the nGaAs layer 5.

The pair potential of the superconductor leaks into the secondary electron gas 6 of the InGaAs layer 5 which is a semiconductor due to the proximity effect of the superconductor. When the penetration distance from the interface between the InGaAs layer 5 and the drain electrode 4 to the point where the bleeding of the potential is zero is X _N , when the incident energy is close to the Fermi level, the larger the X _{N, the} Andre The F reflection probability increases. The presence of pair potential intrusion in the semiconductor adjacent to the superconductor reduces the effect of the scattering potential barrier Z at these interfaces and increases the Andreev reflection probability when the incident energy of the electron is close to the Fermi level. (Reference 2: PC van Son et al., Physi
cal Review B, Vol. 37, No. 1
0, p. 5015, 1988).

FIG. 5 is a correlation diagram showing the relationship between Andreev reflection probability and incident electron energy. In the figure, 51 when the distance X _N oozes pair potential of 0, 52 distance X _N oozes pair potential 0.4
The correlation when μm is shown. As apparent from the figure, the distance X _N even 0.4μm even 0, Andreev reflection probability A is the electron energy incident superconductor 0.
The maximum is around 8 (E / Δ). However, when the incident electron energy, which is the energy of electrons when the device is actually operated, is near 0 (E / Δ), the distance X _N is 0.4
The Andreev reflection probability A is larger in μm. This is when Nb having a superconducting gap energy of 1.5 meV is used in the superconducting state as the drain electrode 4 of the superconductor.

The bleeding distance X _N is suppressed by applying a magnetic field to the interface between the drain electrode 4 and the two-dimensional electron gas 6. This is because when a magnetic field is applied to the interface between the semiconductor and the superconductor, the magnetic field destroys the pair potential that may exist in the Cooper pair exuding from the superconductor side to the semiconductor side. The relationship between the penetration length X _N , which is the exudation length of this pair potential, and the magnetic field to be applied is expressed by the following equation 1 as shown in Reference 3 (Reference 3: TY Hsiang et al. , Physical R
view B, Vol. 22, No. 1, p. 15
4, 1980).

[0022]

[Equation 1] Here, v _F is the Fermi velocity, D _N is the diffusion coefficient, H is the applied magnetic field, T is the absolute temperature, and k _B is the Boltzmann constant.

For example, when T = 0.5K, the penetration distance X _N becomes about 1/10 even when a weak magnetic field of about 4 mT which has almost no effect on the electrode of the superconductor is applied. Here, from the magnetic field applying electrode 11 shown in FIG.
When the distance to the interface between the GaAs layer 5 and the drain electrode 4 is 0.5 μm, a magnetic field H of about 4 mT can be obtained by passing a current of 10 mA through the electrode 11 for applying a magnetic field. That is, by supplying a current to the electrode 11 for applying a magnetic field, the pair potential penetration length to the secondary electron gas 6 side at the interface between the drain electrode 4 and the secondary electron gas 6 can be modulated.

Therefore, the Andreev reflection probability becomes variable, and as a result, the quantization unit of the conductance G can be continuously controlled by the magnetic field. As shown in FIG. 6, when the magnetic field H is large, the conductance G becomes 1
When the amount of change in step is small and the magnetic field H is small, the amount of change in conductance G for one step is large. That is,
The magnetic field H controls the quantization unit of the conductance G.

Here, if the relationship between the gate voltage and the conductance G shown in FIG. 6 is re-expressed as the relationship between the gate voltage and the source-drain voltage, it becomes as shown in FIG. FIG. 7 shows the relationship between the gate voltage and the source-drain voltage when the magnetic field H is zero. In the device of the present invention, the source-drain voltage changes stepwise depending on the gate voltage, as shown in FIG. 7, and changes sharply between steps. Here, the gate voltage is biased to the point P, and the input voltage V _{in is set} to the split type metal gate electrode 10.
(FIG. 1), the gate voltage becomes the value at point Q. Due to this change in the gate voltage, the source-drain voltage changes from 0.5 mV to 1.0 mV, and the output voltage V _out =
0.5 mV is obtained.

If the point P where the gate voltage is biased is below the thermal noise voltage V _N from the point Q, this operation will malfunction due to the thermal noise V _N. In order to prevent this malfunction, it is sufficient that the point P is separated from the point Q by at least the thermal noise voltage V _N = (4k _B TR _L B) ^1/2 . Here, B is a frequency band and _RL is a load resistance. Here, assuming that T = 0.5K, _RL = 1KΩ, and B = 10 GHz, V _N = 17 μV. Therefore, assuming that the signal / noise ratio is 5 or more, it is sufficient that the point P is 85 μV or more and distant from the point Q.

Therefore, in order to obtain the output voltage V _out = 0.5 mV without malfunction due to the thermal noise voltage V _N , it suffices if the input voltage is 85 μV or more, and at this time the gain (V _out /
VI _N) has gained practical value of about 6, and the as a transistor. Furthermore, in this embodiment, by applying a magnetic field, the amount of change in the conductance G in one step can be continuously changed, that is, the change in the source-drain voltage in one step can be continuously changed. It has a new function that can continuously change the gain from 0 to 6.

Example 2. The second embodiment of the present invention will be described below. FIG. 8 is a perspective view showing the structure of a semiconductor-coupled superconducting device of Example 2 of the present invention. In the figure, 11a is a back gate electrode (fourth electrode) formed on the back surface of the InP substrate 1, and the InAlAs layer 9 and the split type metal gate electrode 10 and the InP substrate 1 and the back gate electrode 11a are MIS type gates. Are configured, and the others are the same as in FIG. Also in such a structure, the length L of the conduction channel by the two-dimensional electron gas 6 formed in the region below the gap formed at the predetermined position of the split type metal gate electrode 10 is larger than the mean free path le. Is sufficiently short, and the width W is the Fermi wavelength λ of the electron.
Quantum point contact is achieved when the condition of being equal to or less than _F is satisfied.

By the way, under a zero magnetic field in which no magnetic field is applied, the penetration distance X, which is the leakage length of the pair potential from the adjacent superconductor to the semiconductor,
_N is shown by the following equation 2.

[0030]

[Equation 2] Where m ^* is the effective mass of the electron, h is Planck's constant, N
_S is the electron concentration of the two-dimensional electron gas, T is the absolute temperature, and k _B is the Boltzmann constant. Intrusion distance of potential X _N
Is proportional to the square root of the electron concentration N _S of the two-dimensional electron gas. Here, when a negative voltage is applied to the back gate electrode 11a, the electron concentration N _S decreases, so that the back gate voltage controls the penetration distance X _{N of the} potential.
The Andreev reflection probability A can be changed.

FIG. 9 is a correlation diagram showing the relationship between the voltage applied to the split type metal gate 10 and the conductance G, and the applied voltage V applied to the back gate electrode 11a.
_{When bg} changes, the quantization unit, which is the change amount of the conductance G in one step, changes. And this shows that the quantization unit of the conductance G can be continuously modulated. That is, similarly to the first embodiment, the quantum point contact element of this embodiment also functions as a gain variable amplification element.

Example 3. The intensity of the magnetic field can be measured by using the quantum point contact device having the structure of FIG. 1 described in the first embodiment. By previously knowing the relationship between the applied magnetic field and the quantization unit of conductance shown in FIG. 6, it is possible to measure the magnetic field applied to this quantum point element from the quantization unit of conductance G.

[0033]

As described above, according to the present invention, the conductance, that is, the conduction channel between the first and second electrodes which is quantized and discontinuously changes with respect to the voltage applied to the gate electrode. The effect is that the quantization unit of conductance can be increased. Further, there is an effect that the quantization unit of the conductance of the conduction channel can be continuously changed by applying the magnetic field from the third electrode or applying the negative voltage from the fourth electrode. As a result, the variable gain amplifier can be configured by one element.

[Brief description of drawings]

FIG. 1 is a perspective view showing a cross-sectional structure of a semiconductor-coupled superconducting device of the present invention.

FIG. 2 is an explanatory diagram showing a state of electrons in the vicinity of an interface between a superconductor and a semiconductor.

3 is a correlation diagram showing a correlation between a gate voltage and a conductance in the semiconductor coupled superconducting device of FIG.

FIG. 4 is an explanatory diagram showing the state of seeping out of the potential pair at the interface between the superconductor and the semiconductor.

FIG. 5 is a correlation diagram showing a relationship between Andreev reflection probability and incident electron energy.

6 is a correlation diagram showing a correlation between a gate voltage, conductance and an applied magnetic field in the semiconductor coupled superconducting device of FIG.

7 is a correlation diagram showing the relationship between the gate voltage and the source-drain voltage in the semiconductor coupled superconducting device of FIG.

FIG. 8 is a perspective view showing the structure of a semiconductor-coupled superconducting device of Example 2 of the invention.

9 is a correlation diagram showing a correlation between gate voltage and conductance in the semiconductor coupled superconducting device of FIG.

FIG. 10 is a perspective view showing an example of a quantum point contact device having a split type gate electrode.

11 is a correlation diagram showing a correlation between gate voltage and conductance in the quantum point contact device of FIG.

[Description of Reference Signs] 1 InP substrate 2 InAlAs layer 3 Source electrode 4 Drain electrode 5 InGaAs layer 6 Two-dimensional electron gas 7 InAlAs layer 8 n-type InAlAs layer 9 InAlAs layer 10 Split type metal gate electrode 11 Magnetic field application electrode 11a back Gate electrode 12 Insulating film

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location H01L 29/812 29/80

Claims (3)

[Claims] 1. A semiconductor layer laminated body formed on a semiconductor substrate and having a heterojunction surface for forming a two-dimensional electron gas serving as a carrier; and a semiconductor layer laminated body formed on the semiconductor layer laminated body with an insulating layer interposed therebetween. A gate electrode having a minute gap formed at a predetermined position, and a gate electrode formed on the heterojunction surface of the semiconductor layer stack 2
A first electrode formed so as to be connected to a two-dimensional electron gas; formed at a distance shorter than the mean free path between the gate electrode and the two-dimensional electron gas; And a second electrode composed of a semiconductor-coupled superconducting device. 2. The semiconductor-coupled superconducting device according to claim 1, further comprising a third electrode for applying a magnetic field to an interface between the two-dimensional electron gas and the second electrode. Superconducting element. 3. The semiconductor-coupled superconducting device according to claim 1, further comprising a fourth electrode formed on the back surface of the semiconductor substrate so as to face the gate electrode.