RU2343591C1 - Josephson- transition super-conducting device - Google Patents

Abstract

FIELD: physics, measurements.

SUBSTANCE: invention relates to cryogenic devices and can be used in measuring instruments, radio communication and data processing hardware operated at low temperatures. Josephson-transition superconducting device substrate has a weak-bond area formed as a thin-film FNF-structure connected with superconductor electrodes. Aforesaid thin-film structure is made up of F layers of ferromagnetic material with magnetisation directions lying in the plane of structure, a layer N of normal metal being arranged between aforesaid ferromagnetic metal F layers for them to turn said magnetisation directions relative each other. Superconducting electrodes are connected to opposite lateral sides of aforesaid FNF-structure.

EFFECT: higher-efficiency control over Josephson-transition critical current by external magnetic field due to provision of several independent channels for aforesaid critical current to pass through.

8 cl, 6 dwg

Description

The invention relates to cryoelectronic devices and can be used in measuring equipment, radio engineering and information systems operating at low temperatures.

A large number of designs of superconducting devices based on the Josephson effect (hereinafter - SPD) that are promising for use in various low-current superconducting devices (fast single-quantum logic devices, transmitting and receiving devices, magnetometric devices) are described. The Josephson effect occurs in the so-called "Weak coupling" formed at the point of contact of two superconductors through a non-superconducting material with any type of conductivity. In this case, the main implementation problem is the selection of the physicochemical characteristics of the layer materials providing both high critical current I _c at a given distance L between superconductors and the ability to control the magnitude and sign of I _c .

Traditionally known SPDs are a multilayer thin-film structure formed on a dielectric substrate, including a superconductor, insulating, barrier, and functional layers. Depending on the purpose and design, the choice of substrate materials and the active media themselves is carried out.

So, an SPD is described, which is formed on a single-crystal dielectric substrate and has three layers: two layers of a YВа ₂ Cu ₃ О _7-x superconductor (YBCO), one of which is the lower one, placed directly on the substrate, separated by a barrier layer (US 6541789, Sato , et al. 04/01/2003). A weak bond is formed at the end of one of the superconductors.

Known SPD (JP 3190175, YUZURIHARA et al., 08/20/1991), which is a device with four current leads, in which the current, set through one of the pairs of current leads, transfers to the ferromagnetic state the existing film from the antiferromagnetic substance, which is not in the region Josephson contact. The magnetic moment arising in this case creates a magnetic field, which leads to the suppression of the critical current of the Josephson element located between two other current leads of the device and to the generation of a voltage pulse on it.

Also known SPD, designed to control the flow of electrons and having a multilayer structure "superconductor - normal metal - superconductor" and does not use dielectric barrier layers (US 6995390, Tsukui, 02/07/2006).

Known SPD, designed to control the critical current of five-layer double-barrier Josephson junctions, in which the material located inside the barriers contains a ferromagnetic film. Its purpose is to provide Zeeman splitting of the resonant electron levels in the intra-barrier region in order to control the critical current of the structure by controlling the position of the split levels relative to the Fermi energy of the electrodes by the voltage applied to the additional control contacts of the structure (US 6344659, Ivanov et al., 02/05/2002 - the closest analogue).

An analysis of the prior art shows that known devices with Josephson junctions, including the closest analogue, usually provide for setting the current across the thickness of the composite weak-bond region, that is, in the direction perpendicular to the plane of the multilayer thin-film structure. Such devices have significant drawbacks in terms of controlling the current parameters due to mutual screening of the fields by layers of the same structure (for example, one ferromagnetic layer by another), as well as by small depths of penetration of the superconducting state with respect to the same in the normal metal.

The objective of the invention is SPD, the design of which allows to eliminate these disadvantages, namely, to provide more efficient control of the critical current of Josephson junctions by means of an external magnetic field due to the organization of a number of independent channels of its flow.

The problem is solved in that a superconducting device with a Josephson junction includes a weak-bonding region formed on a substrate in the form of a multilayer thin-film structure connected with electrodes from a superconductor. Thin-film structure consists of layers of ferromagnetic material - normal metal - ferromagnetic material. Superconductor electrodes are attached to opposite side faces of the thin-film structure, and the directions of magnetization of the layers of the ferromagnetic material lie in the plane of the thin-film structure, and the layers themselves are capable of reversing the indicated directions of magnetization relative to each other.

The device can be characterized by the fact that the layers of the ferromagnetic material have different values of the coercive field, as well as the use of niobium or an alloy based on it, and the fact that the compound of rare-earth cuprates of the general formula ReBa ₂ Cu ₃ O _{7 is} used as the superconductor _x , where Re is a rare earth metal.

The device can be characterized, in addition, by the fact that Ni, Co, Fe or metal alloys based on them are used as a ferromagnetic material, and by the fact that an element from the group Cu, Au, Al, Pt is used as a normal metal.

The device can also be characterized in that the thicknesses of the layers of the ferromagnetic material and the normal metal are 10-100 nm.

The technical result of the invention is the ability to independently change the directions of the magnetization of the layers, control the magnitude, period of oscillation and the direction of the current through the SPD due to the organization of three independent channels of its flow. This is realized through the SPD structure with a new configuration of layers in the composite region: S- (FNF) -S, where S, N, F are the layers of a superconductor, a normal metal, and a ferromagnet, respectively. In this topology, the superconducting current is set in the direction parallel to the FN boundaries of the composite weak coupling region in the S- (FNF) -S structure.

The invention is illustrated in the drawings, where:

figure 1 presents the design of the patented device;

figure 2, 3 - dependence of the imaginary and real part of the wave vector q ₃ from the magnitude of the exchange energy of one of the layers of a ferromagnet, respectively;

figure 4 - dependence of the critical current on the distance between the superconducting electrodes;

figure 5 is the same as in figure 4, for a fixed distance value;

figure 6 - dependence of the normalized part of the critical current from the normalized value of the exchange energy.

The superconducting device (see FIG. 1) includes a substrate 1. On its surface, a multilayer thin film structure is formed consisting of a first layer 2 of ferromagnetic material, a layer 3 of normal metal and a second layer 4 of ferromagnetic material. Ferromagnetic films must be single-domain, their manufacturing technology is known. Layers 2, 4 of ferromagnetic material should have different values of coercive fields, which allows you to reverse the directions of magnetization in the layers relative to each other. This can be achieved, for example, by the manufacture of layers 2 and 4 with slightly different thicknesses (~ 30%) or film width, as well as the choice of substrate material 1 or normal metal in layer 3.

Opposite side faces of the structure are connected to the electrodes 5 from the superconductor and current leads 6. As a result, the superconducting current supplied through the current leads 6 to the electrodes 5 simultaneously flows through three independent channels of the FNF structure of length L formed by layers 2, 3, 4. Side edges 7 structures connected to the electrodes 5 of the superconductor.

As components of the FNF structure suitable for implementing the patented device, materials used in the technology of cryoelectronic materials and known to specialists can be used. For example, any standard substrates (for example, silicon, sapphire, etc.) can be used as substrate 1. As the ferromagnetic materials of layers 2, 4, pure ferromagnets Ni, Co, Fe or ferromagnetic alloys based on them: Pt _x Fe _1-x , Pt _x Ni _1-x , Pt _x Co _1-x , Pd _x Fe _1-x , Pd _x Ni _1-x , Pd _x Co _1-x , Cu _x Ni _1-x ; as layer 3 of a normal metal — Cu, Au, Al, Pt. As a material for superconducting electrodes 5 - niobium, niobium nitride, or MgB ₂ and compounds based on it, or high-temperature superconductors based on rare-earth cuprates of the general formula ReBa ₂ Cu ₃ O _7-x , where Re is a rare-earth metal, or other oxides ( see, for example, US 6011981, Alvarez et al., 04/01/2000), the technology of applying layers on substrates is known. Estimates show that typical thicknesses of the layers of ferromagnetic material and normal metal for the patented topology are 10-100 nm and are in the range technologically feasible for thin-film electronics.

Superconducting device operates as follows. When current is supplied through the current leads 6 to the electrodes 5 from the superconductor, the superconducting current simultaneously flows through three independent channels of the FNF structure of length L, formed by layers 2, 3, 4. In this case, the Cooper pairs injected through one of the side faces 7 are transferred to the opposite superconducting electrode 5, which ensures the flow of a given superconducting current through the structure.

The rationale for achieving the technical result, as well as the requirements for choosing the parameters of the substrate, the materials of the layers forming the structure, and the physical principles underlying the invention are explained by the numerical calculations, the results of which are shown in Fig.2-6.

Figure 2, 3 shows the dependence of the imaginary (figure 2) and real (figure 3) parts of the wave vector q ₃ on the value of the exchange energy H _{2 of} one of the layers 2, 4 of the ferromagnet with a constant value of the exchange energy of another ferromagnet H ₁ / πТ _C = 30. The dependence was calculated in the framework of the equations of the microscopic theory of superconductivity for several values of the parameter z = (ξ _N / ξ _F ) ² = 50, 150, 300, 600 and ξ _N / ξ _N = 4, ξ _N / ξ _F = 10, T = 0.5 T _C. Here: Т _С is the critical temperature of superconducting electrodes, ξ _N and ξ _F are the penetration lengths of the superconducting state from superconductors into normal and ferromagnetic materials, respectively.

The parameters ξ _N = ξ _N (γ _N ) ^1/2 and ξ _F = ξ _F (γ _F ) ^1/2 characterize the coupling coefficient between F and N layers. (γ _F = γ _B d _f / ξ _F ; γ _N = γ _B d _n / γξ _N ; γ = ρ _N ξ _N / ρ _F ξ _F ; γ _B = R _B A / ρ _F ξ _F , R _B and A is the resistance and the area of the FN boundaries, ρ _N and ρ _F are the resistivities F and N of the materials). It can be seen that in the case of the antiparallel orientation of the magnetizations of the F films, the value of Imq ₃ strictly vanishes at H ₂ = -H ₁ for all parameter values. The position of the second point on the H ₂ axis, in which Imq ₃ = 0, depends on the parameter z and can be either to the left or to the right of the value H ₂ = -H ₁ .

The obtained result indicates that by changing the direction of the magnetization vector of one of the ferromagnetic films to the opposite direction, one can switch from the regime with an oscillating dependence of the critical current I _C (L) to the regime in which there are no oscillations completely. For a given distance L between superconductors and a given temperature, such a transition can be accompanied by either a change in the sign of I _C (L), a several-fold increase in the value of I _C (L), or a simultaneous manifestation of these two effects. The presence of two values of H ₂ at which this effect takes place means that the switching is achieved in three ways: either as a result of complete magnetization reversal of one of the films (H ₂ = -H ₁ ), or partial magnetization reversal (H ₂ <-H ₁ ) , or additional magnetization in the opposite direction (H ₂ > -H ₁ ).

Figure 4 shows the dependences of the components of the critical current of the structure on the distance L between the electrodes 5, calculated numerically at Н ₁ / πТ _С = 30, z = (ξ _N / ξ _F ) ² = 300, ξ _N / ξ _N = 4, ξ _N / ξ _F = 10, Т = 0.5Т _С , γ _BF / γ _BN = 1, and Н ₂ / πТ _С = 30, -10, -30, -78.4 (γ _BF = R _BF A _F / ρ _F ξ _F ; γ _BF = R _BN A _N / ρ _N ξ _N ; R _BF -R _BN and A _F -A _N - resistance and area of SF and SN faces, respectively). Figure 5 shows the dependence of the normalized part of the critical current I _C1 calculated for H ₂ / πT _C = 30 for the same parameter values. It can be seen that in the case under consideration, the critical current components I _C1 and I _C2 = I _C1 turn out to be significantly less than the term I _C3 and decrease much faster with increasing L. It is also seen that in the practically interesting case L> ξ _{N, the} contribution to the critical current from I _C1 and I _{C2 is} negligible, so that with an exponential degree of accuracy I _C = I _C3 . The critical current decreases exponentially with increasing L and experiences oscillations associated with the transition of the structure from 0 to π states over a length of the order of (Imq ₃ ) ^-1 ≈ξ _N ”(Imq ₁ ) ^-1 , (Imq ₂ ) ^-1 ≈ξ _F. Oscillations disappear at Н ₂ = -Н ₁ and Н ₂ / πТ _С = - (γ _F ² H ₁ / πT _C ) ^-1 = -78.4.

This leads to important practical conclusions that in the given parameter range for L> ξ _N both the magnitude and sign of the critical current of the structure with an exponential degree of accuracy are determined by only one current component I _C3. This component is always positive in the region of small L <ξ _N and experiences damped oscillations with increasing L. The sign of I _C3 changes each time the dependence I _c3 (L) passes through zero, leading to the alternation of 0 (I _C3 > 0) and π (I _C3 <0) states. Both the characteristic attenuation scale (Req ₃ ) ^-1 and the oscillation period (Imq ₃ ) ^{-1 of the} critical current I _C = I _C3 (L) significantly (two to three orders of magnitude) exceed the similar parameters achieved in SFS, SFSF and SFNS Josephson structures in geometry with current perpendicular to the FN and SF boundaries (see, for example, VVRyazanov, VAOboznov, A.Yu. Rusanov, AVVeretennikov, AAGolubov, and JAarts, // Phys. Rev. Lett., v. 86, 2427 ( 2001); VA.Oboznov, VVBol'ginov, AKFeofanov, VVRyazanov, and A. Buzdin // Phys. Rev. Lett., V. 96, 197003 (2006)).

Figure 6 presents the dependence of the normalized part of the critical current module | I _C3 γ _BN eR _BN / 2πT | from the normalized value of the exchange energy Н ₂ / πТ _С for H ₁ / πT _C = 30, z = (ξ _N / ξ _F ) = 300, ξ _N / ξ _N = 4, ξ _N / ξ _F = 10, Т = 0.5 T _C , γ _BF / γ _BN = 1, calculated for L / ξ _N = 0.1, 2, 3, 4, with compression ratios on the graph of 10, 1, 0.2, 0.01, respectively. The parameter z = 300 is selected so that the critical current oscillation period at H ₂ = H _{1 is} minimal (see figure 2).

As follows from the graph, for L / ξ _N = 0.1, the structure is always in the 0-state. Therefore, when switching from H ₂ = H ₁ to H ₂ = -H ₁ , the sign of I _C3 does not change, but an almost triple increase in the critical current is observed. For H ₂ = H ₁ and L / ξ _N = 1, 2, the Josephson transition is in the π state (see Fig. 4). In this case, switching from H ₂ = H ₁ to H ₂ = —H ₁ leads to a transition from π to the 0 state. With this switching, I _c increases by about 7 times for L / ξ _N = 1 and 3 times for L / ξ _N = 2. Finally, in the case L / ξ _N = 4 for H ₂ = H _{1, the} system is in the 0-state and switching from H ₂ = H ₁ to H ₂ = -H ₁ leads to a three-fold increase in the critical current, which is not accompanied by a change in its sign.

It can be seen that the transition from 0 to the π state is possible if H ₂ lies in the range from 4πT _C to 15πT _C. When the sign of H ₂ changes from H ₁ to -H _{1, the} critical current increases approximately 6 times. These facts justify the possibility of controlling both the magnitude and sign of the critical current by reversing the direction of the magnetization of one of the ferromagnetic films of the structure. They also indicate that a change in the sign of H _{2 is} accompanied by an increase in the critical current of the structure.

The data presented show that during the transition from the ferromagnetic configuration (H ₂ = H ₁ ) to the antiferromagnetic geometry (H ₂ = -H ₁ ), the critical current I _{C of the} structure can increase significantly, especially near the transition between the “0” and “π” states. Far from the transition points, the gain can reach one order, which is due to a change in the characteristic decay length of the critical current.

Thus, in Josephson structures with the S-FNF-S topology, it is possible not only to effectively increase (compared to the SFS topology) the effective decay length of the critical current and its oscillation period to lengths of the scale ξ _N , but also control both the magnitude and the sign I _S

Claims (8)

1. A superconducting device with a Josephson junction, including a weakly bonded region formed on a substrate in the form of a multilayer thin-film structure connected with electrodes from a superconductor, characterized in that
thin-film structure consists of layers of ferromagnetic material - normal metal - ferromagnetic material,
electrodes from a superconductor are attached to the opposite side faces of the thin film structure, wherein
the directions of magnetization of the layers of the ferromagnetic material lie in the plane of the thin-film structure, and the layers themselves are made with the possibility of reversal of these directions of magnetization relative to each other. 2. The device according to claim 1, characterized in that the layers of the ferromagnetic material have different values of the coercive field. 3. The device according to claim 1, characterized in that niobium or an alloy based on it is used as a superconductor. 4. The device according to claim 1, characterized in that the compound of rare-earth cuprates of the general formula ReBa ₂ Cu ₃ O _7-x , where Re is a rare-earth metal, is used as a superconductor. 5. The device according to claim 1, characterized in that Ni, Co, Fe or metal alloys based on them are used as the ferromagnetic material. 6. The device according to claim 1, characterized in that an element from the group of Cu, Au, Al, Pt is used as a normal metal. 7. The device according to claim 5, characterized in that the thickness of the layer of ferromagnetic material is 10-100 nm. 8. The device according to claim 6, characterized in that the thickness of the normal metal layer is 10-100 nm.

Images