JPH0452632B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a superconducting element having a semiconductor at a junction.
That is, it relates to a superconductor-semiconductor-superconductor coupled device.

[Summary of the invention]

The present invention provides a semiconductor-coupled superconducting element in which two superconducting electrodes are coupled by two-dimensional electron gas (2DEG) formed in an n-type inversion layer on the surface of a P-type semiconductor.

[Conventional technology]

Since the invention of the tunnel-type Josephson device, much research has been conducted on superconducting three-terminal devices, which correspond to transistors and FETs in semiconductors. In this regard, many attempts have been made to develop semiconductor-coupled superconducting devices because they have a low barrier height and can have a wide electrode spacing, and also have the possibility of three-terminal operation through electrical control of the semiconductor. However, nothing of practical use has been obtained.

FIG. 13 shows a cross-sectional structure of a conventional semiconductor coupling device.
A superconducting current is obtained in a structure in which single-crystal Si is used as a semiconductor, and superconducting electrodes 2 are formed close to each other as shown in FIG. 13 as highly concentrated P-type semiconductors by diffusion or ion implantation. Regarding this, RC
Ruby & T. Van Duzar: IEEE Trans.Elect.
Device, ED- 28 , 1394, ('81).

Incidentally, the characteristics of a semiconductor-coupled superconducting device are closely related to the superconducting diffusion length ξ _N in the semiconductor and the interface characteristics between the superconductor and the semiconductor. Superconducting proximity effect theory J. Seto & T. Van Duzer: Low Tempera−
ture Physics−LT− 13 , 328, New York,
According to Plenum, ('74), the maximum superconducting current Ic
is Ic∝Tj ² exP(−L/ξ _N )/ξ _N − (Formula 1). Here, Tj is the tunneling probability of electrons at the superconductor/semiconductor interface, and it can be seen from the above equation that in order to obtain a large Ic, Tj must be large and ξ _N must be long. Generally, a Schottky barrier is formed at the metal/semiconductor interface, and Tj increases as the barrier height decreases and the barrier width decreases.

FIG. 14 is an energy band diagram of a superconductor-P-type silicon-superconductor device, where E _F is the Fermi level, and E _C and _EV are the lower and upper end energy levels of the conduction band and valence band, respectively. In the case of P-type silicon, the barrier height E _b is 0.2 eV. The barrier width W depends on the carrier concentration n, and the larger n is, the thinner it becomes. Therefore, in an element using P-type silicon, it was necessary to increase n to 10 ²⁰ cm ^-3 as much as possible. On the other hand, ξ _N represents the mobility of the semiconductor as μ(cm ² /
VS), then ξ _N ∝μ ^1/2 n ^1/3 − (Equation 2).

In the case of the P-type Si mentioned above, μ60cm ² /
VS is small, and ξ _N is short at approximately 0.01μ. Like this P
In the case of a device using type silicon, ξ _N is short even though n is as large as 10 ²⁰ cm ⁻³ , and only a device with a device length L of around 0.1 μm could be realized. Carrier concentration
At 10 ²⁰ cm ^-3, it can no longer be called a semiconductor and is metallic, and even if an MIS (voltage-driven) or MES (current-injection driven) structure is realized, it will be sensitive to changes in the applied gate voltage and inflow current. However, it becomes extremely insensitive, and cannot take advantage of its characteristics as a semiconductor such as a transistor or FET element. Furthermore, if the superconducting electrode spacing is 0.1 μm, it is very difficult to form the third terminal on the semiconductor.

[Problem that the invention seeks to solve]

The present invention solves the problems with the conventional superconductor-semiconductor-superconductor coupling device described above, namely, the distance between the superconducting electrodes must be extremely short in order to obtain a superconducting current, and the carrier concentration of the semiconductor must be reduced. The aim is to solve the problem of having to increase the height extremely high and to realize a superconducting two-terminal or three-terminal device with excellent characteristics.

[Means for solving problems]

The present invention provides that a two-dimensional electron gas (referred to as 2DEG) is formed in the surface inversion layer of a P-type semiconductor, and that in two dimensions, for a superconducting diffusion length ξ _N ∝μ ^1/2 n ^1/3 , This was done by focusing on the analysis result that ξ _N ∝μ ^1/2 n ^1/2 as described later.

In the present invention, in the surface inversion layer of a P-type semiconductor,
A semiconductor coupled superconducting element in which two superconducting electrodes are coupled by 2DEG is provided.

In the structure of the present invention, an n-type inversion layer is formed on the surface of a P-type semiconductor, a 2DEG is formed therein, and the semiconductor layer in which the 2DEG is formed is in ohmic contact with the two superconducting electrodes. It is necessary that there be.

The present invention will be explained in more detail below along with its operation.

[Effect]

According to the proximity effect theory, the superconducting diffusion length ξ _N in a semiconductor is given by ξ _N = (〓D/2πk _B T) ^1/2 . Here, h is a blank constant, D is a diffusion coefficient, k _B is a Boltzmann coefficient, and T is a temperature.

When D is three-dimensional, D=1/3v _F l (v _F is Fermi velocity, l is mean free path), but in two-dimensional case, D=1/2v _F l, and l=μv _F m〓 /e (m〓 is the effective mass), k _F = (2πn _S ) ^1/2 , v _F =〓k _F /m〓, so ξ _N of the two-dimensional system is ξ _N = (〓 ³ μ/4πk _B Tem〓) ^1/2 (2πn _S ) ^1/2 ∝μ ^1
/2
It is found as n _S ^1/2 − (Equation 3). Compared to ξ _N ∝μ ^1/2 n ^1/3 in the three-dimensional case, for example, when both n and n _S increase by one order of magnitude, the effect is greater in the two-dimensional case, and the controllability improves accordingly. Then you can say. Also, in order to increase ξ _N , n and n _S must be increased, but
As mentioned above, in the case of three-dimensional (bulk) semiconductors,
If a large amount of dopant is doped in order to increase n, it becomes a scatterer and μ decreases, resulting in a decrease in ξ _N. However, 2
In the case of dimensions, this is not the case for the following reasons.

Generally, in order to increase the carrier concentration n of a bulk semiconductor, it is necessary to dope it with a dopant of the same degree as this, and as a result, the mobility decreases due to scattering by this dopant (as mentioned above). In the case of silicon, n = 10 ²⁰ cm ^-3 and μ
60cm ² /VS), but in the case of 2DEG in the surface inversion layer of a P-type semiconductor, even if the carrier concentration of the substrate is originally small, it naturally occurs in a narrow region of 2 to 10 nm.
Alternatively, carriers are gathered together by an electric field. For this reason, the areal density n _S of the carrier is as large as 10 ¹² to 10 ¹³ cm ^-2 (more than 10 ¹³ cm ^-2 when converted to n), and since the number of scatterers in the substrate is originally small, the mobility μ is also large. Furthermore, n _S
can be controlled by an externally applied electric field.

Next, in the present invention, it is required that the contact between the superconducting electrode and the semiconductor layer in which the 2DEG is formed is ohmic, and the meaning of this ohmic in the present invention will be explained.

An ohmic contact between a semiconductor and a metal means that the current-voltage characteristics of the contact conform to Ohm's law. Generally, the contact area between metal and semiconductor is provided with a shot key barrier or a first barrier as shown in Fig.
An oxide barrier as shown in Figure 5 and a barrier between the two are formed. At extremely low temperatures such as liquid helium temperature (4.2K at 1 atmosphere), the current flowing through this barrier is mainly due to the tunnel effect, and its current-voltage characteristics are ohmic in the sense mentioned above, and the contact resistance due to this occurs. This resistance is inversely proportional to the tunneling probability Tj, and Tj strongly depends on the height and width of the barrier, especially the width. Therefore, in order to lower the contact resistance and allow more current to flow, it is necessary to make the barrier width thinner.
Let's apply this to the element. The product of superconducting critical current (maximum superconducting current) Ic and normal conducting resistance R _N is:
When superconducting metals such as Nb and Pb are used for electrode 2, the maximum voltage is on the order of 2 mV, and for observability, it is thought that a minimum of about 10 _μA is required for Ic, so R _N is required to be 200 Ω or less. . If R _N
Considering only the above contact resistance, the contact area between the superconducting electrode 2 and the 2DEG of the inversion layer 5 is 5 nm ×
Assuming 100 μm (thickness of 2DEG×device width), the contact resistance needs to be 5×10 ⁻⁷ Ωcm ² (5×10 ⁻¹¹ Ωm ² ) or less.

The ohmic properties referred to in the present invention refer to such barrier properties having low contact resistance.

〔Example〕

(Example 1) FIG. 1 shows a cross-sectional structure of an example of the present invention. In the figure, 1 is a P-type InAs substrate, 2 is a superconducting electrode, 5 is an inversion layer, 6 is an insulating film such as SiO or SiO ₂ formed by sputtering or vapor deposition, and 7 is gold (Au), etc. This is the third electrode.

The surface of P-type InAs is n regardless of the carrier concentration.
The shape is reversed, and 2DEG is formed in this reversed layer. FIG. 2 shows a band diagram at the third electrode (gate) portion of the embodiment shown in FIG. As shown in the figure, the surface of P-InAs is inverted to n-type, and 2DEG is generated on the interface side of the inversion layer with the insulator.
The 2DEG carrier concentration n is controlled by the voltage Vg applied to the third electrode 7, thereby controlling the superconducting current flowing in the 2DEG between the two superconducting electrodes 2.

As mentioned above, the surface of P-InAs is inverted to n-type regardless of the carrier concentration, and there are
2DEG is formed, but considering the existence of leakage current due to tunneling from the n inversion layer to the P layer, the carrier concentration is preferably 2 to 3 × 10 ¹⁷ cm ^-3 or less, and in this case, at 4.2 K, n _S = 2.5 ×10 ¹² cm ^2-2 , μ=5000cm ² /
VS has been realized (E. Yamaguchi:
Extended Abstract of the 1984 International
Conference on Solid State Devices and
Materials, Kobe, ('84), 371), ξ _N also
It is long at 0.19μm.

Furthermore, in this embodiment, the superconducting electrode 2 and the semiconductor 1
As mentioned above, it is required that P be in contact with the ohmic (electron tunneling probability Tj at the semiconductor/superconductor interface is large).
As shown in FIG. 3, the Schottky barrier height of the n-type InAs layer of the inversion layer on the surface of the InAs-type InAs substrate 1 to the metal is always negative, forming an ohmic contact. Therefore, the electron tunneling probability Tj at the semiconductor/superconductor interface gradually increases compared to P-type silicon or the like.

In this way, in this example, since ξ _N is long and the tunneling probability Tj is large, a large Ic (maximum superconducting current) can be obtained from equation 2 shown earlier (holds true regardless of two-dimensional or three-dimensional). It is clear that

The relationship between electrode spacing L and ξ _N is given by IEEE
TRANSAC-TION ON ELECTRON
DEVICES, VOL.ED−28, NO.11,
It is shown in NOVEMBER 1981, PP 1394-1397 that L can range from several times to about 10 times _ξN .

In this example, for this reason, the electrode spacing L can be set as long as about 0.5 μm, which makes it possible to create not only a two-terminal structure but also a three-terminal structure.
Two-terminal operation is possible in which a superconducting current flows during 2DEG, and furthermore, like a MOSFET in a semiconductor, n _S can be changed as described above by applying a voltage Vg to the electrode 7, thereby changing the two-terminal characteristics. Control,
So-called three-terminal operation can be performed.

(Example 2) FIG. 4 shows an example in which a metal electrode (third electrode) 7 is directly formed on the n-type inversion layer 5 on the semiconductor surface of the P-type InAs substrate 1. A metal electrode 7 is formed in the opening. The superconducting electrode 2 is the same as in the first embodiment.

FIG. 5 shows an energy band diagram for the gate portion (third electrode 7). In this case, the two-terminal operation is the same as in the first embodiment, but the three-terminal operation is different. That is, in this case
MESFET-like or three-terminal operation using current injection.

(Example 3) Figure 6 shows an example of a two-terminal structure in which a ridge (protrusion) 8 is formed on the surface of a P-InAs substrate 1, and a Nb layer of a superconducting electrode 2 is formed on both sides of the ridge. be.

FIG. 9 shows its IV characteristics. In the figure, a is the IV characteristic when direct current is applied, the superconducting current flows in the region, and is the voltage state. b is the characteristic of irradiation with _10GHZ microwave. In addition, for ease of viewing, a, b
The origin of is shown shifted.

(Example 4) FIG. 7 shows a three-terminal element with an MIS structure that is the same as Example 1 except that a ridge 8 is formed.

FIG. 10 shows how the superconducting current changes when the gate voltage is changed in this example (Ic in the figure is the maximum superconducting current and R _N is the normal resistance).

(Example 5) The example shown in Fig. 8 is a three-terminal element with an MES structure.
The structure is the same as FIG. 4 of the second embodiment except that it has a ridge 8 structure.

FIG. 11 shows the change in Ic when the injection current Ij is changed in this example, and as shown, the superconducting current is controlled by Ij.

(Example 6) Figure 12 shows P-type InAs on the back side of P-type InAs substrate 1.
Metals that form ohmic contact with e.g.
Electrode 1 made of Au and Zn alloy (Au90% + Zn10%), etc.
This is an example in which 0 is formed. The electrode 10 corresponds to a junction gate, and the width of the depletion layer in the p-n junction (in this case, the n-layer is an inversion layer) is controlled by applying a voltage between the electrode 10 and the electrode 2 or 7. Sideways
Performs three-terminal operation to change _nS . Note that when applying a voltage between the electrodes 10 and 2, the electrode 7 and the insulating film 6 are not necessary.

Although the embodiments have been described above, the present invention is not limited thereto. For example, other semiconductors in which two-dimensional electron gas (2DEG) is generated in the surface inversion layer can be used, and the superconducting electrode material can also be Nb. , various superconducting materials other than Pb can be used.

〔Effect of the invention〕

As described above, the present invention provides n on the surface of a P-type semiconductor.
The present invention provides a semiconductor-coupled superconducting element in which two superconducting electrodes are connected by a 2DEG formed in a shape inversion layer, and is capable of two-terminal operation in which a superconducting current flows in the 2DEG. There is an advantage that the electrode spacing L can be made wider than that of conventional elements. Furthermore, a so-called three-terminal operation in which the two-terminal characteristics are controlled by the third electrode can be performed, and according to the present invention, a two-terminal or three-terminal element with excellent characteristics can be realized.

[Brief explanation of the drawing]

1 to 3 are a sectional view of a first embodiment of the present invention, a band diagram for a gate region, and a band diagram for a superconducting electrode region, and FIGS. 4 and 5 are a sectional view of a first embodiment of the present invention, respectively. 6 to 8 are sectional views of the third to fifth embodiments of the present invention, and FIGS. 9 to 11 are the cross-sectional views of the third to fifth embodiments of the present invention, respectively. -V characteristic diagram of the device of the sixth embodiment of the present invention, a diagram showing the change in superconducting current depending on the gate voltage, a diagram showing the dependence of Ic on the injection current, and Fig. 12 is a cross-sectional view of the sixth embodiment of the present invention. , FIG. 13 and FIG. 14 are a cross-sectional view and band diagram of a conventional element, and FIG. 15 is an energy band diagram showing an oxide barrier. 1... P-type InAs substrate, 2... superconducting electrode, 5
... (n-type) inversion layer, 6 ... insulating film, 7 ... third
Electrode, 8... Ritsuji.

Claims (1)

[Claims] 1. A P-type semiconductor substrate in which a two-dimensional electron gas in which a superconducting current flows is formed by diffusion of superconducting electrons due to the proximity effect in a surface inversion layer, and a surface of the P-type semiconductor substrate. A semiconductor-coupled superconducting device comprising an inversion layer and two superconducting electrodes in contact with an ohmic. 2. The semiconductor-coupled superconducting device according to claim 1, wherein the P-type semiconductor substrate is a P-type InAs substrate. 3. A P-type semiconductor substrate in which a two-dimensional electron gas in which a superconducting current flows is formed by diffusion of superconducting electrons due to the proximity effect in the surface inversion layer, and ohmic contact with the surface inversion layer of the P-type semiconductor substrate. two superconducting electrodes, and an MIS installed between the two superconducting electrodes.
A semiconductor-coupled superconducting element characterized by having a gate type, MES type, or junction type gate. 4. A P-type semiconductor substrate in which a two-dimensional electron gas in which a superconducting current flows is formed by diffusion of superconducting electrons due to the proximity effect in the surface inversion layer, and ohmic contact with the surface inversion layer of the P-type semiconductor substrate. What is claimed is: 1. A semiconductor-coupled superconducting element comprising: two superconducting electrodes, and an ohmic electrode formed on the P-type semiconductor substrate side.