JP5090971B2 - Superconducting quantum interference device - Google Patents

Description

  The present invention relates to a superconducting quantum interference device.

  A technology that uses a superconducting quantum interference device (SQUID) to accurately measure a very weak magnetic field that is weaker than the geomagnetism. It is used for observation.

  The superconducting quantum interference element has a structure having a Josephson junction in a ring of a superconductor, a superconducting quantum interference element having one Josephson junction is rf-SQUID, and a superconducting quantum interference element having two Josephson junctions Is called dc-SQUID. Usually, when simply saying SQUID, it is common to point out dc-SQUID widely used from the point of high sensitivity.

The dc-SQUID uses the DC Josephson effect that superconducting current flows according to the macroscopic phase difference of each superconductor when the superconductors are weakly connected. . That is, the magnitude of the superconducting current I that flows when the phase difference between the superconductors is θ is expressed as I = I _C sin θ. Here, I _C is the maximum superconducting current flowing through the Josephson junction and is called the critical current.

FIG. 4 is a schematic diagram showing a configuration of a dc-SQUID having two Josephson junctions (hereinafter simply referred to as SQUID for simplicity. In the case of showing rf-SQUID, it is strictly described as rf-SQUID). FIG. Two U-shaped superconductors 101 and 102 are weakly connected by Josephson junctions at points A and B, respectively. When a current is supplied from the current terminal 106, superconducting currents according to the respective phase differences θ _A and θ _B flow at points _A and _B. At this time, if a magnetic field is applied to the SQUID, the magnetic field is blocked in the superconductors 101 and 102 due to the Meissner effect, and the magnetic field cannot enter except for a very shallow region on the surface. A magnetic flux Φ can exist in the inner magnetic flux trapping region 105 surrounded by. Based on Maxwell's law, the round-trip integral of the vector potential when it makes a round in the superconductors 101 and 102 is equal to the magnetic flux penetrating this, and the superconducting phase is based on quantum mechanical requirements. Since it needs to be uniquely determined, θ _A −θ _B = 2πΦ / Φ ₀ . Here, Φ ₀ is called a flux quantum, and its magnitude is a very small value of Φ ₀ = h / 2e = 20.6gauss / μm ² . That is, it is possible to easily detect a change in magnetic flux of several gausses in a small region on the order of microns.

  The Josephson effect is manifested by weakly coupling two superconductors. For this reason, a narrow constriction or a step is formed between two superconductors, or a Josephson junction is formed by inserting a metal of several microns or less or an extremely thin insulating layer of several tens of angstroms. can do. In a sandwich device with an insulating layer in between, a Cooper pair (a pair of two electrons with opposite momentum) in the superconductor moves to the other electrode by the tunnel effect, so a tunnel junction or a superconductor (Superconductor ) -Insulator-Superconductor and is called SIS junction. In addition, when a metal (Normal metal) is sandwiched, the acronym of superconductor (superconductor) -metal (normal metal) -superconductor (superconductor) is taken and called an SNS junction. In particular, a junction in which superconductors are connected by point contact is called a point contact junction, and a junction in which fine conductors or steps are connected is called a weak link. In a broad sense, Josephson elements other than tunnel junctions are classified as weakly coupled elements.

In order for SQUID to operate with high sensitivity, it is necessary that the Josephson junctions A and B in FIG. 4 have the same characteristics (the critical current value I _C and the junction resistance value _RN are the same), and usually In order to extract a change in magnetic flux using a voltage change, it is better that the resistance value of the junction is large. In view of this, the SQUID using the tunnel junction is most suitable in consideration of the mechanical properties of the element, the stability over time, and the ease of the manufacturing process including the peripheral circuit. For this reason, at present, the SQUID usually refers to the SQUID of the tunnel junction.

However, the tunnel junction has the same element structure as the capacitor, and therefore has a larger junction capacitance C than the weakly coupled element. For this reason, when the superconducting current flowing through the element exceeds the critical current value, the state transitions from the zero voltage state to the voltage state. After that, unless the current is returned to zero once, the latching (latching) in which the voltage cannot return to the zero state is called Perform the action. Usually, the SQUID performs measurement using a change in magnetic field as a change in voltage, so this latching operation is not desirable. For this reason, in a SQUID using a tunnel junction, it is necessary to put a branch resistance R (shunt resistance) that bypasses the current in parallel with the Josephson junction. The current SQUID using niobium achieves a sensitivity of about 2 × 10 ⁻⁷ Φ ₀ / Hz ^1/2 (see Non-Patent Document 1).

  As described above, the SQUID has a very high sensitivity, but the magnetic field to be actually measured is a relatively large region of several millimeters or more. However, since the SQUID itself has a size of about several tens to several hundreds of micron squares, it is necessary to devise in order to measure a change in a magnetic field in a large region. For this reason, a form in which a secondary coil called a pickup coil that is inductively coupled with a SQUID superconductor loop has been used to detect a change in a magnetic field in a relatively large region. Depending on the shape of the pickup coil, there are magnetic field measuring instruments such as a magnetometer and a gradiometer. A method using such a secondary coil is useful for detecting a magnetic field change in a relatively large region, but conversely, a pickup coil is used to measure a minute magnetic field in a nano-sized local region. First, the sensitivity is higher when the SQUID and the object to be measured are directly inductively coupled.

As described above, the SQUID measures a change in magnetic field as a change in voltage. For this reason, the sensitivity of the SQUID is determined by the limit of the magnetic field that can be measured when there is a voltage fluctuation. The minimum energy that can be detected by the SQUID is expressed by the following equation using the inductance L and the tunnel junction capacitance C of the SQUID loop.

Here, k _B represents a Boltzmann constant, T represents an absolute temperature, and R represents a branch resistance. Β _C is called McCumber coefficient,

The sensitivity is maximized when. For this reason, the branch resistance R is adjusted at the tunnel junction.

It is trying to become. As apparent from the equation (1), the sensitivity is higher as the temperature is lower and the inductance L and the junction capacitance C are smaller. In the equation (1), the noise becomes zero at the absolute zero T = 0K, but actually, the reaction due to the noise current flowing through the SQUID loop must be considered, and the detection energy limit is the Planck constant as in the following equation (3). It becomes half of.

In order to reduce the inductance L and the junction capacitance C in order to increase the energy sensitivity, the loop size of the SQUID may be reduced and the area of the Josephson tunnel junction may be reduced. However, when the junction area is reduced, the coastal current value I _C is also reduced, so that the branch resistance R must be increased. Also, if the junction capacitance C is made too small, the charging energy E _C = e ² / 2C of the tunnel junction itself increases, and the particle characteristics of the Cooper pair carrying the superconducting current become significant, so that the phase of the SQUID Fluctuations and detection sensitivity decreases.

  For this reason, it is necessary to keep the area of the tunnel junction at least about 500 nm square. Therefore, in the case of a SQUID having a nanometer-sized loop, it is considered that a weakly coupled device is more advantageous (see Non-Patent Document 2).

The main conditions imposed on the SQUID capable of detecting a single spin include, as is clear from the equation (1), the operating temperature T of the SQUID as low as possible, the junction capacitance C of the Josephson junction, and the inductance L of the SQUID loop. Is to make it as small as possible. In a weakly coupled Josephson element that does not have a tunnel barrier, the junction capacitance C is very small. In order to reduce the inductance L, it is necessary to reduce the SQUID loop, and accordingly, the Josephson junction itself is also reduced.
D. Koelle, R. Kleiner, F. Ludwig, E. Dantsker and John Clarke, "High-transition-temperature superconducting quantum interference devices", Reviews of Modern Physics, 1999, Vol. 71, No. 3, p.631- 633 John Gallop, "SQUIDs: some limits to measurement", Superconductor Science and Technology, 2003, Vol. 16, p.1575-1582 John Gallop, PW Josephs-Franks, Julia Davies, Ling Hao and John Macfarlane, "Miniature dc SQUID devices for the detection of single atomic spin-flips", Physica C, 2002, Vol. 368, p.109-113 Sergei V. Sharov, Andrei D. Zaikin, "Influence of parity on the persistent currents of superconducting nanorings", Physical Review B71, 2005, p.014518.1-014518.7

  However, the current characteristics of the weakly coupled device are directly dependent on the device structure and its dimensions. Even with the current state-of-the-art nanotechnology, it is extremely difficult to produce the same weakly coupled device in the sub-nano size. For this reason, it has been experimentally shown that it is difficult to bring the SQUID characteristic itself to the quantum measurement region corresponding to the theoretical limit even in the conventional weakly coupled device.

  The present invention has been made in view of the above, and an object of the present invention is to provide an ultrasensitive superconducting quantum interference device having a magnetic field detection sensitivity enhanced to the limit.

  A superconducting quantum interference device according to the present invention includes a two-dimensional electron gas layer formed in a heterostructure, first and second superconducting electrode layers connected to the two-dimensional electron gas layer, and a two-dimensional electron gas layer. And a region from which the central portion of the substrate is removed.

  In the present invention, by removing the central portion of the two-dimensional electron gas layer of the superconductor-two-dimensional electron gas-superconductor junction, the magnetic flux is captured in a region where the central portion is removed with ultra-high sensitivity. A small superconducting quantum interference device can be provided. In addition, the superconducting quantum interference device loop is small and the influence of background magnetic field noise can be reduced. Furthermore, since the pickup coil is not used and the structure is simple, the combination with other functional elements is easy.

  The superconducting quantum interference device is characterized in that a gate electrode is provided on the heterostructure.

  The superconducting quantum interference device is characterized in that a gate electrode is provided on a side surface of the heterostructure.

  In the above superconducting quantum interference device, the heterostructure is formed on a substrate, and has a gate electrode on a surface opposite to the surface on which the heterostructure is formed.

  The superconducting quantum interference device is characterized in that electrical characteristics of two current paths in the two-dimensional electron gas layer are controlled by a voltage applied to the gate electrode.

  In the present invention, it is possible to control the carrier concentration and the mean free path of the two-dimensional electron gas layer by arranging the gate electrode at a position corresponding to the heterostructure formed with the two-dimensional electron gas layer. The electrical characteristics of the two current paths in the two-dimensional electron gas layer can be made the same. Therefore, an ultrasensitive superconducting quantum interference device can be provided.

  ADVANTAGE OF THE INVENTION According to this invention, the supersensitive superconducting quantum interference element which raised the magnetic field detection sensitivity to the limit can be provided.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view showing a configuration of a superconducting quantum interference device according to the present embodiment.

The superconducting quantum interference device shown in FIG. 1 has a superconductor-two-dimensional electron gas-superconductor in which superconductor electrodes 1 and 2 are connected to a two-dimensional electron gas 3 formed on a heterojunction surface of a semiconductor heterostructure. Body joint. The central portion of the two-dimensional electron gas 3 is removed to form a region 5 that captures the magnetic flux. By forming the region 5, the current path between the superconductor electrodes 1 and 2 is divided into two, and this junction itself also functions as a superconducting ring and operates as a superconducting quantum interference device. A gate electrode 4 is formed on top of the heterostructure corresponding to the two current paths. Unlike the SNS junction using a normal metal, the gate electrode 4 can control the current path, that is, the carrier concentration and the electron mean free path, by using the two-dimensional electron gas 2 instead of the metal. The gate electrode 4, the electrical characteristics as well as the same in the two current paths, it is possible to obtain an optimum critical current I _C and the resistance R _N.

Conditions required for an ultrasensitive superconducting quantum interference device are as follows: (1) The I _{C RN} products of two Josephson junctions of the superconducting quantum interference device are the same, and (2) no pickup coil is used. so that small in order to improve detection sensitivity as possible inductance L of the SQUID, i.e. to reduce the superconducting quantum interference device loop, (3) to take the operation margin I _C is as large as possible (4) There are four points of making the temperature of the element as low as possible (see Non-Patent Document 3). These conditions of the calculation of the parameters of the optimal superconducting quantum interference device for measurement of small magnetic fields, that the highest sensitivity is obtained in the critical current value I _C is few milliamperes, the resistance value R _N is several tens ohms situation I understand.

  FIG. 2 is a plan view showing the configuration of another superconducting quantum interference device according to the present embodiment. The superconducting quantum interference device shown in the figure is formed by connecting two two-dimensional electron gases 3A and 3B to the superconductor electrodes 1 and 2, respectively, so as to form a region 5 from which the two-dimensional electron gas is removed. is there.

  FIG. 3 is a plan view showing a configuration of still another superconducting quantum interference device according to the present embodiment. The superconducting quantum interference device shown in the figure has a gate electrode 4 formed on the side surface of a heterostructure. Thus, the resistance value in the Josephson junction can be adjusted even in the side gate structure in which the gate electrode 4 is disposed on the side surface. Further, a back gate structure in which a gate electrode is arranged on the back surface of a substrate (not shown) on which a superconductor-two-dimensional electron gas-superconductor junction is arranged can also be used.

Next, component materials constituting the superconducting quantum interference device will be described. Any superconducting material including an oxide superconductor can be used for the superconductor electrodes 1 and 2. However, good contact (ohmic contact) between the superconductor and the semiconductor heterostructure is desired, and materials with high critical temperatures are suitable. For this reason, niobium (Nb) having the highest critical temperature as a single element, or a metal superconductor such as aluminum (Al) that is easy to fabricate and has good wettability with a semiconductor heterostructure can be used. It is also possible to use a composite superconductor such as niobium nitride (NbN), magnesium diboride (MgB ₂ ), an oxide high-temperature superconductor, or the like.

As conditions for the two-dimensional electron gas 3 between the superconductor electrodes 1 and 2, it is desirable that good contact without a Schottky barrier is obtained with the superconductor, and that the carrier concentration of the two-dimensional electron gas is high. Any two-dimensional electron gas can be used. Therefore, indium arsenide (InAs) n-type surface inversion layer, indium arsenide (InAs) / indium arsenic (InAs) p-type InAs surface inversion layer, indium gallium arsenide (InGaAs) / indium aluminum arsenic (InAlAs), indium arsenide A semiconductor heterostructure having a two-dimensional electron gas layer such as (InAs) / aluminum anmotin (AlSb), indium arsenic (InAs) / aluminum gallium anmotin (AlGaSb), aluminum gallium nitride (AlGaN) / gallium nitride (GaN), etc. , Superlattice structure using silicon (Si) / silicon germanium (Si _x Ge _1-x ), graphene and carbon nanotubes that are monoatomic layers of graphite, and inversion layers of silicon (Si) Can It is.

  The material of the gate electrode 4 that controls the carrier concentration of the two-dimensional electron gas 3 includes aluminum (Al), gold (Au), gold compounds with good adhesion (AuGeNi, AuGe), haanium (Hf), titanium (Ti ) Or the like can be used. It is important to dispose an insulating layer under the gate electrode so that the gate electrode does not leak into the two-dimensional electron gas. In the case where a side gate is used, an insulating layer is provided so as to prevent leakage from the side surface, or a structure that does not contact the side surface is employed.

Next, the size of the superconducting quantum interference device will be described. The detection sensitivity of a single spin is proportional to the size r of the long side of the two-dimensional electron gas and the energy detection sensitivity ε of equation (1). When the resistance value R and the critical current value I _C are optimized to satisfy the condition (2), it can be calculated as 10 spins at r = 1 μm and 0.5 spins at r = 100 nm. Considering that the tolerance of about 5 times the maximum detection sensitivity is necessary against the influence of other noises, the optimum dimensions are r = 50 nm, junction resistance R = 20-30Ω, critical current value I _C = 2 μA or so is calculated as a desirable value. For this reason, the size of the two-dimensional electron gas layer is about 150 × 200 nm.

  Next, an application example of the present superconducting quantum interference device will be described.

  This superconducting quantum interference device is sensitive enough to detect a single spin change, and can be used as a read / detector for a spin current generated by a spin transistor or spin FET. In order to increase the detection sensitivity, it is necessary to arrange the object to be measured in the region 5 where the magnetic flux is detected. Moreover, it is clear from the simulation that the sensitivity is higher at the position shifted to one of the four corners than at the center. The object to be measured is not limited to semiconductors, but is used for various materials such as magnetic materials such as single-molecule magnets and nanomagnets, molecules or low-dimensional superconductors, organic molecules, and bio-related substances such as proteins and neurons. be able to.

  If the object to be measured is a nano-sized material, it is assumed that the intrinsic quantum mechanical state of the constituent atoms / molecules is expressed, and there is concern about the change in the quantum state of the object to be measured due to measurement. The In this superconducting quantum interference device, the measurement sensitivity can be freely changed by changing the gate voltage, so that measurement under conditions that do not destroy the quantum state of the object to be measured is also possible. In addition, the change in the carrier concentration with respect to the change in the gate voltage can respond quickly because the dimensions of the two-dimensional electron gas layer itself are small, and can be operated at several tens of gigahertz, realizing high-speed measurement. it can.

  Even when the measurement temperature region is changed, the highest sensitivity can be achieved by changing the gate voltage.

  The present superconducting quantum interference device can be used as a method of reading a qubit such as a superconducting magnetic flux bit device, a superconducting phase quantum device, or a spin bit device. When used as a readout of qubits, it can be measured by interacting with qubits only when necessary by varying the gate voltage. For this reason, the influence on the qubit can be minimized.

  By combining this superconducting quantum interference device with a scanning probe microscope (SPM) such as an atomic force microscope (AFM) or a scanning tunneling microscope (STM), it is possible to detect a minute magnetic flux while scanning an object. it can. By combining with a scanning probe microscope, it is possible to compensate for the disadvantage that the measurement area is limited to a very small area.

  When there is a changing magnetic flux current or spin current, the superconducting quantum interference device can be used as a spin amplifier that amplifies the change in the signal as a voltage. In this case, the dynamic range of the detection signal depends on the size of the superconducting energy gap and the bias point of the material constituting the superconducting quantum interference device.

  In an SNS ring in which a metal is sandwiched between a part of a superconductor ring, if the superconducting current flowing through the ring is several channels, a permanent current flows and the magnitude depends on the parity of the flowing electrons. (See Non-Patent Document 4).

  When a gate voltage is applied so that the superconducting current flowing through the superconducting quantum interference device flows using only one or several channels of the two-dimensional electron gas and the input current is zero, the two-dimensional electron gas region is permanently Current is expected to flow. In this case, as shown in Non-Patent Document 4, the permanent current largely depends on the parity (odd number, even number) of the number of electrons. That is, a permanent current flows when the number of electrons is an even number, but the permanent current is zero when the number is an odd number. This parity effect is not affected by the magnetic field. Therefore, it is possible to realize a highly sensitive ammeter that can detect the entry / exit of one electron by utilizing this parity effect.

  As described above, according to the present embodiment, by connecting the superconductor electrodes 1 and 2 to the two-dimensional electron gas 3 having the region 5 for capturing the magnetic flux, the superconductor-two-dimensional The electron gas-superconductor junction can be used as an ultrasensitive superconducting quantum interference device. Further, since the superconducting quantum interference device loop itself is small, the influence of the background magnetic field noise is reduced, and the relative magnetic field noise can be reduced.

According to this embodiment, by providing the gate electrode 4, it is possible to vary the critical current value I _C and the resistance R _N of the superconducting quantum interference device.

  According to the present embodiment, since a pickup coil having a large size is not used and the manufacturing process is relatively simple, the size is small, and it is easy to manufacture in combination with other quantum elements.

It is a perspective view which shows the structure of the superconducting quantum interference element in one embodiment. It is a top view which shows the structure of another superconducting quantum interference element in one embodiment. It is a top view which shows the structure of another superconducting quantum interference element in one Embodiment. It is a top view which shows the structure of the conventional superconducting quantum interference element.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,2 ... Superconductor electrode 3, 3A, 3B ... Two-dimensional electron gas 4 ... Gate electrode 5 ... Magnetic flux trapping region 101, 102 ... Superconductor 105 ... Magnetic flux trapping region 106 ... Current terminal

Claims (5)

A two-dimensional electron gas layer formed in the heterostructure;
First and second superconducting electrode layers connected to the two-dimensional electron gas layer;
A region from which a central portion of the two-dimensional electron gas layer is removed;
A superconducting quantum interference device comprising:   The superconducting quantum interference device according to claim 1, further comprising a gate electrode on the heterostructure.   The superconducting quantum interference device according to claim 1, further comprising a gate electrode on a side surface of the heterostructure.   2. The superconductor according to claim 1, wherein the heterostructure is formed on a substrate and has a gate electrode on a surface of the substrate opposite to a surface on which the heterostructure is formed. Quantum interference element.   5. The superconducting quantum interference device according to claim 2, wherein electrical characteristics of two current paths in the two-dimensional electron gas layer are controlled by a voltage applied to the gate electrode. 6.

Images