RU2504049C2 - Superconducting josephson instrument and method of its manufacturing - Google Patents

Abstract

FIELD: instrument making.

SUBSTANCE: substance of the invention: DTN on the basis of a multi-layer thin-film heterostructure comprises two superconductors that form electrodes, and a layer with metal conductivity between them from a metal-alloyed semiconductor. The layer has a locally non-uniform structure and is made as capable of simultaneous formation of two independent current transport channels in its volume, one of which represents a combination of chains of admixture atoms of metal that connect both electrodes and form quasi-unidimensional channels with metal conductivity for transport of superconducting current, and the other one - consists of separately arranged admixture atoms of metal, which form localised conditions of admixture centres and provide for transport of normal tunnel current, besides, the specified quasi-unidimensional channels represent internal shunts for a tunnel current in the layer. The method includes serial application of the first and second layers of the superconductor onto the substrate, a layer of the metal-alloyed semiconductor between them, formed by spraying of the semiconductor and the metal.

EFFECT: elimination of direct flow of current via a layer with provision of resonance mechanisms of current transport, increased specific voltage and differential resistance of a DTN; improved reproducibility of parameters due to usage of thicker layers of an alloyed conductor.

17 cl, 8 dwg, 1 tbl

Description

The invention relates to the field of superconducting micro- and nanoelectronics and can be used in the manufacture of superconducting integrated circuits (LIS) for various purposes.

Known is a superconducting device with a Josephson junction (hereinafter referred to as SPD) based on a multilayer thin-film structure SNS (superconductor - normal metal - superconductor), consisting of a lower superconducting electrode S based on an Nb compound, a layer of tantalum nitride (Ta _x N) and an upper electrode S also based on Nb compound (US6734454 (B2), Van Duzer, et al. 05/11/2004).

The disadvantage of this device is that the weak bond is localized in the region of the normal (N) metal layer, and such an N metal must simultaneously satisfy two mutually contradictory requirements. On the one hand, to ensure high critical current densities at a technologically reasonable interlayer thickness, it must have a large effective coherence length ξN, i.e. to be low resistance. On the other hand, to prevent significant suppression of superconductivity in S electrodes, its transport properties must be significantly worse compared to similar parameters of superconductors (M.Yu. Kupriyanov, A.A. Golubov, ZhETF, T. 96, issue 4, 1420 , 1989), i.e. the material of the interlayer should be high resistance to the material of the electrodes. There is a significant suppression of superconductivity in the electrodes, as a result of which the characteristic transition voltage (V _c = I _c R _N , where I _c is the critical transition current and R _N is the normal resistance) remains low, and its high-frequency properties, due to the Josephson relation, remain unsatisfactory.

Known SPD based on a multilayer thin-film structure superconductor - normal metal - superconductor, including the formation of the lower superconducting electrode based on the Nb compound, a layer of the nonsuperconducting phase of niobium nitride (Nb _x N) and the upper electrode also based on the Nb compound (EP 1365456 (A2), YAMAMORI HIROTAKE et al., 11.26.2003). The technological advantage of this method is the use of only one source of spraying to form the entire Josephson structure. However, the main disadvantage of this method remains the metal layer and, accordingly, obtaining SPD SNS type with all the ensuing disadvantages.

Other analogs of SPD are also known using higher-resistance interlayers based on metal alloys, for example, PdAu, TiN alloys. As a result, in all of the above structures, SNS-type SPDs were obtained in which current was transported by direct current flowing through a layer with purely metallic conductivity. In such transitions, as a rule, the characteristic voltage did not exceed tens of microvolts, which corresponded to operating frequencies up to 20 GHz. For most practical applications, an SPD with a characteristic voltage of hundreds of microvolts and operating frequencies in the range of 20-200 GHz and higher is required. So, for example, for metrological applications, operating frequencies should be in the ranges of 70 GHz and 90 GHz.

Known SPD and the method of its manufacture on the basis of a multilayer thin-film structure of a superconductor - normal metal - superconductor. The SPD includes a first superconducting electrode of Nb, a layer of highly metal-doped amorphous silicon (α-Si), and an upper superconducting electrode of also Nb. The method includes the sequential deposition of the first and second layers of a superconductor on the substrate, a layer of a metal-doped semiconductor between them, formed by sputtering a semiconductor and a metal (see Gudkov A.L., Kupriyanov M.Yu., Likharev K.K. Properties of Josephson junctions with an interlayer from doped silicon. JETP. 1988. T. 94. Issue 7. P.319-332 - the closest analogue of the first and second inventions of the group).

The advantage of this device is that as an impurity a metal is used that forms deep impurity levels in the semiconductor, and the interlayer formation mode is chosen so that the amorphous semiconductor is saturated with a metal impurity to the state of complete degeneration with the formation of a material with metallic conductivity, after which alloying ceases .

As established (Kulikov VA, et a1., A mm-wave radiometer with planar Nb / α-Si / Nb josephson junction. IEEE Trans. On Magn. MAG-27, 1991, No. 2), the degree of doping of α-Si with Nb impurity is approximately 11%. The result is an SNS type device, and a fairly high resistance layer. These transitions of the end structure showed high current densities of up to 10 ⁵ A / cm ² and higher.

The disadvantage is the possibility of implementing SPD only submicron area of 0.01-1 μm ² . With a larger area, either large critical currents (7 s) are obtained, which leads to overheating of the junction, or small V _c . For a number of applications, including Volta programmable quantum standards, generator systems and other high-frequency devices, SPDs with an area of up to 100 μm ² and having V _c from 100 μV and above are required. In particular, metrological needs require large-area transitions with V _c ≥150 μV and I _c values from fractions to several milliamps. The requirement for the value of I _c is associated with the need to obtain stable Shapiro steps of large amplitude when exposed to an external operating frequency.

The group of inventions is aimed at eliminating the noted disadvantages of SPD.

Patented SPD based on a multilayer thin-film heterostructure contains two layers of a superconductor that form electrodes, and a layer with metal conductivity between them from a metal-doped semiconductor.

The difference is that the interlayer has a locally inhomogeneous structure and is capable of simultaneously forming two independent current transport channels in its volume, one of which is a set of chains of impurity metal atoms connecting both electrodes and forming quasi-one-dimensional channels with metallic conductivity for superconducting transport current. Another channel consists of separately located impurity metal atoms, forming localized states of impurity centers and providing transport of a normal tunneling current. The quasi-one-dimensional channels mentioned are internal shunts for the tunneling current in the interlayer.

SPD can be characterized by the fact that the conductivities of the current transport channels are preferably equal, and also by the fact that the channels consist of metal atoms forming deep impurity levels in the semiconductor, and the channels consist of atoms of refractory metals forming deep impurity levels in the semiconductor.

SPD can be characterized by the fact that the channels consist of clusters of metals containing at least two atoms, and, in addition, the fact that the semiconductor is an amorphous material.

Silicon can be used as a semiconductor, and a superconductor material or refractory metals selected from the group consisting of W, Ta, Mo can be used as a dopant. Channels can consist of metal atoms forming deep impurity levels in a semiconductor and metal clusters containing at least two atoms. The interlayer may contain at least one layer of metal or other conductive material parallel to the interface between the interlayer and the electrodes, the thickness of which is at least one atomic layer.

SPD can be characterized in that the metal or other conductive material within the interlayer has superconducting properties, and also in that the metal or other conductive material has ferromagnetic or antiferromagnetic properties.

A patented method for manufacturing a Josephson superconducting device involves sequentially depositing the first and second layers of a superconductor on a substrate, a layer of a metal-doped semiconductor between them, formed by sputtering a semiconductor and a metal.

The difference of the method lies in the fact that after the deposition of the first layer of the superconductor, the sputtering parameters of the semiconductor and metal are selected from the conditions for the formation of a conductive layer with a locally heterogeneous structure containing in its volume a combination of chains of randomly located impurity atoms connecting the first and second layers of the superconductor and not related chains of separately located impurity atoms or metal clusters forming localized impurity centers.

The concentration C of impurity atoms or metal clusters inside the layer is selected from the condition:

C ₂ >C> C ₁ , where C ₁ is the concentration of impurity atoms or clusters at which the average distance between them is much greater than the localization radius; C ₂ is the concentration of impurity atoms or clusters at which the average distance between them is less than the localization radius.

The method may be characterized in that the sputtering source of the semiconductor and metal has a semiconductor target with a given concentration of metallic impurity and / or a mosaic target of semiconductor and metal.

The method can be characterized by the fact that two independent sputtering sources of a semiconductor and a metal are used. Silicon can be used as a semiconductor, and the material of a superconductor or refractory metals selected from the group consisting of W, Ta, Mo can be used as a dopant, while the degree of doping is 6-11%.

The method can be characterized by the fact that the substrate temperature and spraying conditions are selected from the conditions for the formation of an amorphous semiconductor and the exclusion of compounds of crystalline phases of a semiconductor with an alloying metal impurity, as well as the fact that a locally inhomogeneous structure is a layered structure formed by applying metal layers without breaking the vacuum .

The technical result of the first invention of the group is an increase in the value of V _c for a given value of I _c and the value of R _{d of the} differential resistance of the SPD in the voltage region corresponding to the operating frequencies and the possibility of manufacturing a SPD of any area; the exclusion of direct current flow through the interlayer with the provision of resonant transport mechanisms for currents up to several milliamps; improved reproducibility of parameters through the use of thicker layers of doped semiconductor.

The technical result of the second invention of the group is the possibility of forming the specified parameters of doping the interlayer in the process without breaking the vacuum: increasing the characteristic voltage V _c > 150-300 μV, normal resistance RN> 0.5-1.0 Ohm and the operating frequency F _c > 75- 100 GHz and amplitudes of Shapiro steps, comparable with I _s .

The achievement of technical results is justified by the following explanations of the functioning mechanisms and experimental data.

It is possible to improve the characteristics of the PSD by simultaneously increasing two parameters - the value of V _c at a given value of I _c and the value of R _d transition in the voltage region corresponding to the operating frequencies. In this case, the current-voltage characteristic (CVC) of the transition should remain almost unambiguous, i.e. McCumber - Stuart parameter must satisfy the condition:

Based on the requirements of the problem being solved, the parameter β should be ≈1 or ≤1, and for Volta standards it can slightly exceed unity. Note that for the SIS type SDS, the parameter β >> 1, and for the SNS type, β << 1.

The desired result in the patented invention is achieved by choosing a layer with certain properties, as well as by choosing the mode and degree of doping of the semiconductor layer of an SDP in a heterostructure: superconductor - doped semiconductor - superconductor (Superconductor - Doped semiconductor-Superconductor), - in the implementation of SDS-type SDS , since the current transport in such a layer is carried out exclusively due to impurity centers. In the patented invention, in contrast to the prototype, elastic and inelastic mechanisms of resonant current transport through localized states formed by impurity centers are provided in the semiconductor layer.

The sputtering mode is carried out in such a way that the level of doping of the semiconductor with a metal ensures the flow of superconducting current along resonant percolation trajectories, which are quasi-one-dimensional channels with metallic conductivity formed by impurity centers. The normal current flow was carried out mainly due to the mechanism of inelastic tunneling through localized states. An additional condition for the implementation of SDS-type SPDs can be the approximate equality of the conductivity of both channels of current transport, and the zone width of quasi-one-dimensional channels should be significantly smaller than the superconductor gap (Г <Δ).

In the patented invention, the increased value of I _{c is} ensured by quasi-one-dimensional channels with metal conductivity and their density per unit area of the Josephson junction, and the value of R _d , as well as R _N, is determined by the mechanism of inelastic tunneling through localized states. The I – V characteristic is unambiguous and the absence of hysteresis (β≈1) is ensured by shunting of resonant tunneling by quasi-one-dimensional channels with metal conductivity, i.e. internal bypass surgery.

As a result of the choice of the mode and degree of doping of the semiconductor layer, a Josephson transition of the SDS type is obtained, which in its properties is fundamentally different from the transitions of the SNS and SIS type and occupies an intermediate position. This means that the degree of alloying is limited both from above and from below. Indeed, an increase in the degree of doping will lead to complete degeneration of the interlayer and the formation of an SNS-type transition, and a decrease in the degree of doping leads to a rarer arrangement of impurities in the interlayer at distances exceeding the localization radius. With a decrease in the degree of doping, quasi-one-dimensional channels with metallic conductivity are no longer formed, and the Josephson junction becomes a SIS type junction.

The indicated technical result in the manufacturing method of SDS type SDDs is achieved in that the doping degree is selected less than in the case of a semiconductor degeneration and in the indicated ratio of impurity concentration, so that with a decrease in the doping degree a gradual transition from the SNS type transition to the SIS type transition occurs. In this case, the shape of the CVC changes in the field of operating voltages and, accordingly, operating frequencies. The I – V characteristic becomes close to the I – V characteristic of the SINIS type transition, but the parameter β remains close to unity. As is known in tunnel junctions, the amplitude of the Shapiro steps exceeds the value of I _c and even crosses the stress axis. The proximity of the properties of SDS transitions to SIS and SINIS type transitions allows us to achieve maximum values of the amplitudes of Shapiro steps, which is confirmed experimentally.

It was experimentally established that the desired degree of doping of the layer of amorphous silicon with a metal (Nb or W), forming deep impurity levels, lies in the range of 6-11%. This is enough to smoothly regulate the conductivity of the semiconductor layer and to obtain reproducible SDS type SDS.

A method of manufacturing SPD involves the use of thin films of equipment and materials known in the technology. First, a superconducting heterostructure consisting of two layers of a superconductor and a layer between them of a doped semiconductor is applied to the prepared substrate by the method of ion-plasma sputtering or another method of forming thin films in a single vacuum cycle. The modes of heterostructure formation, such as the power of the sources, the temperature of the substrate, and the purity of the whole process, are chosen in such a way as to exclude mutual diffusion of materials through the interfaces of the layers and to ensure the purity of the layers. The formation of the interlayer is carried out by the deposition of two materials of a semiconductor and a metal at the same time by sputtering either a mosaic target from one source, or by simultaneously spraying two targets from two independent sources.

The invention is illustrated in the drawings, where:

figure 1, 2 shows a patented SPD in the context;

figure 3 - to an explanation of the principle of operation of the patented SPD (a) and the closest analogue (b);

figure 4 - comparison of the CVC patented SPD (a) and the closest analogue (b);

5-8 - experimental CVC (explanation in the text).

Figure 1 shows the design of a planar SPD in the context, in which the numbers indicate: 1 - substrate; 2 - a layer of a superconductor as a lower electrode; 3 - a layer of metal-doped semiconductor, performing the function of a layer between two superconducting electrodes; 4 - a layer of a superconductor as an upper electrode; 5 - insulation layer; 6 - a layer of superconducting wiring of the upper electrode.

Position 7 denotes the location of the superconducting contact between the layers of the superconductor 4 and 6.

Figure 2 shows a modification of the design of the SPD. The metal-doped semiconductor layer 3, which acts as an interlayer between two superconducting electrodes 2 and 4, may contain at least one layer 8 of metal or other conductive material parallel to the interface between the interlayer and the electrodes, the thickness of which is at least one atomic layer.

Figure 3 explains the principle of operation of patented SPD (a) and the closest analogue of SNS type (b), which differ in the concentration of metallic impurities in the semiconductor layer 3. Designated: 9 - superconducting electrodes; 10 - a layer of a semiconductor doped with metal; 11 - interface between the layer and the electrodes; 12 - single impurity metal atoms; 13 - impurity in the form of metal clusters consisting of several atoms.

Position 14 denotes a conditional trajectory along the current transport channel through a chain of impurities. The distance r _i between adjacent impurity centers is less than the electron localization radius and in a semiconductor for single metal atoms or less than the radius of the electronic environment (electron cloud) r _cl for a metal cluster.

Position 15 shows the conditional probable trajectories along the channels of elastic and inelastic resonant tunneling of electrons through localized impurity centers for which r _i > α, r _cl .

The model (Fig. 3, a) schematically shows the presence of two mechanisms of current transport simultaneously: a quasi-one-dimensional channel with metal conductivity (pos. 14) and a channel with resonant tunneling through localized states (pos. 15). The interlayer in this case has special transport properties for both superconducting current and normal current, and the Josephson junction is an SDS type junction. The model (Fig. 3, b) demonstrates a layer that is homogeneous and degenerated as a result of doping to metal conductivity, in which the concentration of the metal impurity is such that the average distance between the impurity centers is less than the localization radius (r _i <α, r _cl )

Figure 4 schematically shows the shape of the CVC: a) SDS type SDS and b) type SNS type (analog). Pos. 16, the I – V characteristics corresponding to the superconducting current are indicated; POS.17 - resistive sections of the CVC.

There are a number of fundamental differences. On the CVC (Fig. 4a), after section 16 there follows a section close to horizontal, which may also have hysteresis depending on the size of the dimensionless capacity β SPD. The behavior of resistive sections (pos. 17) of the current – voltage characteristic due to various mechanisms of current transport is very different and are described by different formulas. The shape of the resistive section of the I – V characteristic of the SDS SDS type is determined by the dominant processes of resonant tunneling through one or two localized states and is close to the I – V characteristic following from the Glazman-Matveev theory:

,

,

Here: S is the transverse area of the Josephson junction; g, E ₀ and α are the density of the resonance centers, the position of the energy level of the drug relative to the Fermi energy and the radius of the drug, respectively; λ _ep is a dimensionless constant characterizing the electron-phonon interaction; L is the distance between the electrodes.

The factors S / α ² and S / Lα in expressions (3), (4) determine the total number of statistically independent tunneling channels through one and two drugs, respectively. The difference is due to the fact that in the case of two drugs these centers should not be necessarily located along a straight line perpendicular to the planes of the boundaries of the structure. In this case, the factor (La) ^1/2 determines the effective distance by which the drugs can be shifted relative to each other in the volume of the interlayer without loss of efficiency of the resonant conduction channel. The factors (gα ³ E ₀ ) ² , (2gdα ² T) ² and (2gdα ² eV) ² determine the probability of the formation of the corresponding conduction channel. For a channel with one LAN, such a probability is determined by the effective volume of the interlayer space, which can be used by an electron to carry out the resonance tunneling process. For channels with two drugs, this probability also depends both on the volume of the interlayer that can be used to organize the hopping process and on that portion of energy (T or eV) that can be received or delivered by the electron during tunneling. The last of the factors in formulas (3), (4) determine the conductivity of the optimal tunneling channel.

The shape of the resistive section of the I – V characteristic of the SPD with the parameter β≥1 can be described by a simple formula:

with parameters σ _n and β _n , the values of which can be easily determined from experiment.

The traditional form of the resistive section of the I – V characteristic of the SNS type SPD (Fig. 4 b) is described by the well-known resistive model:

Above this section is the portion of the I – V characteristic corresponding to normal resistance (Km). Moreover, for SPD SNS type, it is characterized by the value of excess current I _ex .

In contrast, the portion of the current-voltage characteristic of the SDS SDS type (Fig. 4 a), corresponding to the normal resistance and located above the gap voltage (V _g ), is characterized by the current deficiency I _df . The magnitude of the current deficiency is proportional to Δ-G and occurs when G <Δ. For an SNS type transition, the excess current is proportional to Г, and the width of the conduction band Г exceeds 2Δ.

Figure 5 shows the experimental CVC of an SPD based on a Nb / α-Si / Nb superconductor heterostructure with an interlayer of amorphous silicon (α-Si) doped with W (transition parameters: β≈1, I _c = 0.18 mA, R _N = 0.2 Ohm, V _c = 0.036 mV): a) - autonomous CVC; b) - I-V characteristic of the transition on an enlarged scale in current; c) - I – V characteristics of the transition under the action of an external irradiation frequency. The transition area is 9 × 9 μm ² .

SPD was performed as follows. After the formation of the superconducting heterostructure by standard methods of photolithography and dry etching, the upper electrode was formed, which sets the transition size in plan and the topology of the lower electrode. Then an insulation layer is applied, in which the window to the upper electrode was opened by lithographic methods. Then, in a single vacuum cycle, the surface of the upper electrode was ionically cleaned and a reconnaissance layer of the superconductor was deposited. The lithography of this layer was used to form the wiring of the upper electrode. As a result, the formed SPD has a CVC that is fundamentally different in form from the CVC of junctions of both SNS and SIS types.

Used standard niobium technology for the formation of planar Josephson junctions. Amorphous silicon (α-Si) doped with W. was used as the interlayer material. The interlayer was formed by magnetron sputtering of the α-Si + W mosaic target. The entire Nb / α-Si (W) / Nb superconductor heterostructure was formed in a single vacuum cycle by the magnetron sputtering method. When forming the SPD pattern, standard lithography methods used in electronics were used. The modes of formation of the Josephson SDS type device for key technological operations are shown in the Table.

Figure 6 shows the experimental CVC based on a Nb / α-Si / Nb superconductor heterostructure with an α-Si layer doped with W (transition parameters: β> 1, I _c = 0.44 mA, R _N = 0.65 Ohm, V _c = 0.286 mV): a) stand-alone CVC; b) - I-V characteristic of the transition on an enlarged scale in current; c) - I – V characteristics of the transition under the action of an external irradiation frequency. The transition area is 9 × 9 μm ² .

Figures 7 and 8 show the experimental I – V characteristics of two SDS-type SPDs based on a Nb / α-Si / Nb superconductor heterostructure, but differing by the introduction of an additional metal layer (Nb) inside the α-Si layer doped with tungsten (a) - autonomous CVC; b) - I-V characteristic of the transition under the influence of an external irradiation frequency).

The metal layer is parallel to the interface between the layer and the electrodes. Both SPD are made in one technological cycle.

The graphs of FIGS. 7a, b are shown for the structure shown in FIG. 1, the graphs of FIG. 8a, b - FIG. 2. The thickness of layers 2, 4 of niobium is 200 nm and 100 nm, respectively; the thickness of layer 3 made of α-Si doped with tungsten is 8 nm. The thickness of layer 8 of niobium is 1-2 nm, while it is located in the middle of the thickness of layer 3, and the sum of the thicknesses of the doped α-Si layers in both cases is equal. The transition area is 9 × 9 μm ² .

It can be seen that the value of I _c SPD with layer 8 (Fig. 8) increased by 30 times compared with the control SPD (Fig. 7). This is proved by the fact that the total number of current transport channels, both quasi-one-dimensional channels with metallic conductivity and resonance tunneling channels, increased due to the presence of layer 8 of niobium.

Thus, the transport of the superconducting current through quasi-one-dimensional channels with metallic conductivity and the transport of the normal current due to resonant tunneling through localized states gives a new form of the current – voltage characteristic of the SDS type transition, improves the high-frequency properties of SPDs, which expands the range of high-frequency applications. The use of a fully amorphous doped semiconductor interlayer increases the reproducibility of parameters due to the uniformity of the interlayer thickness over the entire transition area. A layer of a semiconductor doped with metal clusters increases the conductivity of the resonant current transport channels due to the higher probability of the formation of such channels, since r _cl > r _i .

The use of the SDS type of a layered layer of a doped semiconductor with metal layers increases the value of 1 s by increasing the channels of current transport and improves the reproducibility of the parameters due to the possibility of using thicker layers of the doped semiconductor at a fixed critical current. For the SDP of the patented SDS type of large area, an increase in the characteristic voltage V _c > 150-300 μV, normal resistance R _N > 0.5-1.0 Ohm, operating frequency F _c > 75-100 GHz and amplitudes of Shapiro steps, comparable with value of I _c .

Table SPD FORMATION MODES FOR KEY TECHNOLOGICAL OPERATIONS No. Name of operations Technological Modes Layer characteristics one. Preparation of silicon substrate 1 Standard chemical washing - silicon wafer KDB - 12 ρ = 12 Ohm · cm 2. Application of a layer of 2 Nb Pressure ost. gas
P _rest . = (3-5) 10 ^-4 Pa
T _forged = 40-45 ° C
Magnetron Power = 800 W
t spraying = 90 s d = 0.2 μm
ρ _s = 0.9-1 Ohm / □
γ = R300 / R10 = 2.5 T _c = 9.2 K 3. Application of layer 3 α-Si (W) Cooling time after
deposition of Nb1 t _rest = 5 min
T _forged = 30-35 ° C
Pressure ost.gazov P _ost = (3-5) 10 ^-4 Pa
Magnetron Power = 200 W
t spraying = 20-30 s d (α-Si (W)) = 5-10 nm four. Application of a layer of 4 Nb Residual gas pressure
P _ost = 3-5 · 10 ^-4 Pa
T _forged = 35-40 ° C
t spraying = 40 s d = 0.08 μm
ρ _s = 0.9-1 Ohm / □
γ = R300 / R10 = 2
T _c = - 9.2 K 5. PCT (Nb2 / α-Si / Nb / α-Si /) Gas - CF ₄ + O ₂ (5: 1)
Pressure P = 0.2 Pa
Power P = 40 W
t etching = 20 min 6. PCT layer 2 Nb Gas - CF ₄ +0 ₂ (5: 1)
Pressure P = 0.2 Pa
Power Р = 40 W Duration of etching = 30-40 min 7. Application of insulation layer 5 (Al ₂ O ₃ ) Pressure ost.gazov P _ost = 3 × 10 ^-3 Pa
T _forged = 150-200 ° C
t spraying = 3 min d = 0.35 μm 8. Opening windows in the layer Al ₂ O ₃ Etchant: CrO ₃ -H ₃ PO ₄ -H ₂ O (100 g - 200 ml - 150 ml) T _mp. = 85 ° C
t etching = 4 min 9. IO layer 4 and applying a layer of 6 Nb Gas-Ar Pressure P = 0.67 Pa Power P _rf = 0.6 kW
t cleaning = 2 min. P _ost = 1.3 · 10 ^-4 Pa T _forg.d. = 100-150 ° С Duration. spraying = 8 min d = 0.25 μm
ρ _s = 0.9-1 Ohm / □
γ = R300 / R10 = 2
T _c = 9.2 K

Claims (17)

1. A Josephson superconducting device based on a multilayer thin-film heterostructure containing two layers of a superconductor forming electrodes, and a layer with metal conductivity between them of a metal alloyed semiconductor,
characterized in that
the interlayer has a locally inhomogeneous structure and is configured to simultaneously form two independent current transport channels in its volume, one of which is a set of chains of impurity metal atoms connecting both electrodes and forming quasi-one-dimensional channels with metal conductivity for transporting a superconducting current, and the other consists from separately located impurity metal atoms forming localized states of impurity centers and providing normal transport nnelnogo current, said quasi-one- dimensional channels are internal to the tunnel current shunts in the interlayer. 2. The device according to claim 1, characterized in that the conductivity of the current transport channels is preferably equal. 3. The device according to claim 1, characterized in that the channels are composed of metal atoms forming deep impurity levels in the semiconductor. 4. The device according to claim 1, characterized in that the channels are composed of atoms of refractory metals forming deep impurity levels in a semiconductor. 5. The device according to claim 1, characterized in that the channels consist of metal clusters containing at least two atoms. 6. The device according to claim 1, characterized in that the semiconductor is an amorphous material. 7. The device according to claim 1, characterized in that silicon is used as a semiconductor, and a superconductor material or refractory metals selected from the group consisting of W, Ta, Mo are used as a dopant. 8. The device according to claim 1, characterized in that the channels consist of metal atoms forming deep impurity levels in the semiconductor and metal clusters containing at least two atoms. 9. The device according to claim 1, characterized in that the layer contains at least one layer of metal or other conductive material parallel to the interface between the layer and the electrodes, the thickness of which is at least one atomic layer. 10. The device according to claim 9, characterized in that the metal or other conductive material has superconducting properties. 11. The device according to claim 9, characterized in that the metal or other conductive material has ferromagnetic or antiferromagnetic properties. 12. A method of manufacturing a Josephson superconducting device, comprising sequentially depositing on the substrate the first and second layers of a superconductor, a layer of metal-doped semiconductor between them, formed by sputtering the semiconductor and metal,
characterized in that
after applying the first layer of the superconductor, the sputtering parameters of the semiconductor and metal are selected from the conditions for the formation of a conducting layer with a locally heterogeneous structure, containing in its volume a combination of chains of randomly located impurity atoms and separately located impurity atoms or clusters not connected with the mentioned chains metals forming localized impurity centers,
moreover, the concentration C of impurity atoms or metal clusters inside the layer is selected from the condition:
C ₂ >C> C ₁ , where C ₁ is the concentration of impurity atoms or clusters at which the average distance between them is much greater than the localization radius; C ₂ is the concentration of impurity atoms or clusters at which the average distance between them is less than the localization radius. 13. The method according to p. 12, characterized in that the sputtering source of the semiconductor and metal has a semiconductor target with a given concentration of metallic impurities and / or a mosaic target of semiconductor and metal. 14. The method according to p. 12, characterized in that two independent sources of atomization of the semiconductor and metal are used. 15. The method according to p. 12, characterized in that silicon is used as a semiconductor, and a superconductor material or refractory metals selected from the group consisting of W, Ta, Mo are used as a dopant, and the degree of doping is 6-11 % 16. The method according to p. 12, characterized in that the temperature of the substrate and the spraying conditions are selected from the conditions for the formation of an amorphous semiconductor and the exclusion of the formation of compounds of the crystalline phases of the semiconductor with an alloying metal impurity. 17. The method according to p. 12, characterized in that the conductive layer in the form of a layered structure is formed without breaking the vacuum.

Images