Epitaxial growth of superconductors such as Nb.sub.3 Ge superconductors

Abstract

Nb.sub.3 Ge, Nb.sub.3 Al and Nb.sub.3 Si are prepared in A15 structures at the exact stoichiometric ratio with epitaxial growth on selected substrates. The phase diagrams of the materials are altered by this technique and only the A15 phase is stable at the stoichiometric ratios thus permitting attainment of higher and sharper transition temperatures.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to superconducting materials and to a method of preparing such materials by epitaxial growth.

2. Description of the Prior Art

Superconductivity, a thermodynamic state characterized by zero electrical resistivity and perfect diamagnetism, was discovered in 1911 by Kammerlingh Onnes at Leiden University when he found that the resistance of mercury vanished below 4.2 degrees K. Although it was subsequently found that numerous other elements, alloys and compounds are superconductors when cooled to sufficiently low temperatures, it was also discovered that, even if the temperature was below the critical or transition temperature, superconducting properties could be destroyed by a sufficiently strong magnetic field which might be either an external field or one associated with the superconducting current. Commercial interest in the first discovered superconducting materials soon waned because even moderate magnetic fields destroyed the superconducting properties and superconducting properties were thus limited to low magnetic fields and currents.

Upon the discovery of what is now referred to as hard, or Type II superconductivity, the entire field advanced and revived commercial interest. Type II superconductors are often characterized by, among other anomalous properties, high transition temperatures and high critical magnetic fields and although they may have any one of several crystal structures, they are best exemplified by the A15 or beta-wolfram structure. The discovery of Nb.sub.3 Sn by Matthias, Geballe, Geller and Corenzwit in 1954 and the further discovery by Kunzler and coworkers of its unexpectedly high critical magnetic field established benchmarks which are still typical of the commercially promising high T.sub.c superconductors and are easily sufficient to justify the extensive studies during recent years directed toward commercial exploitation of superconductors. Commercial uses suggested include very high magnetic field devices used as magnetic containers for high temperature plasmas in fusion power sources or as magnets for bubble chambers or high energy elementary particle accelerators. Although much effort was directed towards studying Nb.sub.3 Sn, investigation of other materials having the A15 structure also proceeded.

Possibly the most significant advance in the latter respect was the discovery by Matthias and Geballe and their coworkers, U.S. Pat. No. 3,506,940, that an alloy of Nb.sub.3 Ge and Nb.sub.3 Al had a transition temperature higher than any known at that time including Nb.sub.3 Sn. This tended to support the thesis that Nb.sub.3 Ge is probably, from the standpoint of the parameters of concern, the most valuable of the studied A15 materials. Based on this and other studies which indicated that the maximum transition temperature of Nb.sub.3 Ge would occur at the stoichiometric composition, it was generally accepted that if Nb.sub.3 Ge were available as a pure A15 material, it would supersede Nb.sub.3 Sn for many purposes. Efforts made to produce Nb.sub.3 Ge in the desired structure have resulted in advances and identification of directions for fruitful exploration.

For example, a metastable A15 phase in Nb.sub.x Ge.sub.y films has been prepared by electron beam codeposition at an oxygen partial pressure of 2.times.10.sup.-6 Torr on a substrate held at 775 degrees C. during deposition. This technique extends the A15 phase boundary to about 23 atomic percent Ge. The A15 Nb.sub.3 Ge structure obtained, competing with the stable Nb.sub.5 Ge.sub.3, is, however, extremely sensitive to nucleation conditions, contamination, deposition rate and the substrate temperature. Thermodynamic considerations that favor formation of phases other than the A15 at the stoichiometric composition are the primary reason for the lack of success in preparing Nb.sub.3 Ge in a pure A15 structure at the stoichiometric composition.

It was also generally accepted that other materials, such as Nb.sub.3 Al and Nb.sub.3 Si, would have their maximum transition temperatures, and have increased commercial value, if they could be prepared in a pure A15 phase at the stoichiometric composition. Thermodynamic considerations similar to those described for Nb.sub.3 Ge cause phases other than the A15 to also be stable at the stoichiometric composition and have prevented preparation of the desired materials.

SUMMARY OF THE INVENTION

It has been found that epitaxial growth of superconductors on suitable substrates alters the phase diagram of the superconductor and allows the formation of desired phases at the stoichiometric composition. The substrate and the desired phase should have the same crystal structure and approximately matching, normally within plus or minus 0.04 Angstroms, lattice constants. Superconducting bodies, such as magnets, can be formed from the compositions.

For example, Nb.sub.3 Ge, Nb.sub.3 Al and Nb.sub.3 Si can be prepared in a perfect A15 structure at the stoichiometric compositions by epitaxial growth of the compositions on A15 substrates having lattice constants matching that of the desired phase within plus or minus 0.01 Angstroms. In a preferred embodiment, Nb.sub.3 Ge is grown on an Nb.sub.3 Ir substrate.

BRIEF DESCRIPTION OF THE DRAWING

FIG. 1 is a sectional view of a magnetic configuration consisting of an annular cryostat containing a plurality of windings of a superconducting material such as Nb.sub.3 Ge;

FIG. 2 shows experimentally measured lattice constants of Nb.sub.x Ge epitaxially grown on polycrystalline Nb.sub.x Ir versus atomic percent of Ge;

FIG. 3 is a plot of the transition temperature and width as a function of atomic percent of Ge for Nb.sub.x Ge samples epitaxially grown on Nb.sub.x Ir;

FIG. 4 shows experimentally measured lattice constants of Nb.sub.x Ge epitaxially grown on polycrystalline Nb.sub.x Rh versus atomic percent of Ge; and

FIG. 5 is a plot of transition temperature as a function of atomic percent of Ge for Nb.sub.x Ge samples epitaxially grown on Nb.sub.x Rh.

DETAILED DESCRIPTION

Referring specifically to FIG. 1, an annular cryostat 1 with approximate dimensions 18 inches outer diameter, 6 inches inner diameter and 30 inches long, filled with coolant and containing 2,000 turns per centimeter length of windings 2 is depicted. Windings 2 are made from a superconducting material having a critical temperature and may be on a substrate as well as insulated by, e.g., a thin coating of copper which also serves to protect the superconducting winding if it should return to its normal state by acting as an electrical shunt. Terminal leads 5 and 6 are shown emerging from the coil. A pumping means, not shown, may be attached to the cryostat to permit a temperature variation resulting from a concomitant variation in boiling point of, for example, liquid helium as the pressure varies. Variations in magnetic configurations may be made in accordance with well-known and established practice depending upon the ultimate intended use.

It has been found that epitaxial growth of superconductors on selected substrates alters the phase boundaries so that desired phases can be grown at stoichiometric compositions and will be thermodynamically stable. The substrate and desired phase should have the same phase and approximately matching lattice constants, normally within plus or minus 0.04 Angstroms. As examples, the boundary of the stable A15 phase can be extended to at least the stoichiometric composition for Nb.sub.3 Ge, Nb.sub.3 Al and Nb.sub.3 Si, and mixtures thereof.

Growth on a substrate having the same crystal structure and a lattice constant matching that of the phase whose growth is desired should result in the Gibbs-free energy of the material grown being a minimum if the overgrowth lattice matches that of the substrate. With these growth conditions, nucleation and growth of the competing and unwanted phases are less energetically favorable than growth of the desired phase which matches the substrate parameters. When the free energy minima are sufficiently altered by such epitaxial growth, the A15 phase boundary extends to or beyond 25 atomic percent Ge, Al or Si and allows formation of stoichiometric Nb.sub.3 Ge, Nb.sub.3 Al or Nb.sub.3 Si in a pure A15 phase. An another example, the B1 phase boundary can be extended beyond 50 atomic percent C and stoichiometric NbC and VC, and mixtures thereof, can be formed in a pure B1 phase.

An exemplary embodiment, the growth of Nb.sub.3 Ge, will be described in some detail with the understanding that appropriate modifications, in accordance with the general considerations given, are made when growing other materials. This superconducting material may be epitaxially grown on a single crystal substrate having an A15 structure with a lattice constant matching, within plus or minus 0.01 Angstroms, that of the superconducting material. An alternative is the use of a polycrystalline thin film of a material having an A15 structure and a matching lattice constant, as previously defined, as a substrate. As the latter growth technique is perhaps of more general utility and less easily understood than the former, it will be described in detail.

The term epitaxy as used here refers to growth on polycrystalline, as well as single crystal, substrates.

The initial deposit is forced to be related epitaxially to the substrate to minimize the free energy and, consequently, growth conditions of the desired Nb.sub.3 Ge A15 phase are more favorable than those of the competing second phase, or phases, at the stoichiometric composition. If the presence of the matching substrate sufficiently reduces the free energy of the desired phase, growth of the unwanted second phase or phases will be completely suppressed. The polycrystalline substrates are typically grown on highly polished carriers, such as sapphire, prepared with conventional techniques. Other carrier materials such as carbon, fused quartz or silicon might also be used.

A particularly convenient and well-known method, electron beam coevaporation, of epitaxially growing the materials, both substrate and superconducting Nb.sub.3 Ge, will be described in some detail although other methods, such as sputtering, might also be used. The substrate and superconducting material were grown on a sapphire carrier. Two separately controllable electron guns were used. One activated a fixed Nb source, while the other activated a Ge, Si or Al, Ir or Rh source. The latter sources are placed in separate crucibles located on a rotating table to permit convenient selection of the desired crucible and material. The sapphire carriers were held at 875.+-.10 degrees C. during evaporation using a temperature controlled oven. The growth rate was between 12 and 20 Angstroms/sec. The growth rate chosen is not crucial but rates faster than 20 Angstroms/sec. make attainment of equilibrium on the surface difficult. At higher temperatures, adhesion becomes a problem. The substrate and superconducting material should be at least 0.5 .mu.m thick but there is no upper limit to the thickness.

After termination of growth, electron microprobe analysis was used to determine the composition of the film to an accuracy of approximately .+-.1 atomic percent. The lattice constants were determined from X-ray diffractometer traces. The diffraction patterns were recorded on photographic films using a Reed diffraction camera and the photographs were used in well known manner to determine any preferred orientation and the number of crystalline phases present. The superconducting transition temperature was simultaneously measured with standard inductive and resistive techniques to an accuracy of .+-.0.2 degrees K.

EXAMPLE 1

A polycrystalline Nb.sub.x Ir film grown on sapphire was used as a substrate for growing Nb.sub.3 Ge. x represents the variation in composition of the Nb-Ir binary system over the range of interest. Both the Nb.sub.3 Ge and the Nb.sub.x Ir layers were approximately 0.5 .mu.m thick.

Nb.sub.x Ir films on sapphire: The A15 structure in the Nb-Ir binary system extends over the range 21.5-28.5 atomic percent Ir and electron beam coevaporation in the manner previously described of Nb and Ir on a sapphire carrier held at 875 degrees C. resulted in polycrystalline Nb.sub.x Ir films with a pure A15 structure. The Nb.sub.x Ir lattice constant varies between 5.125 angstrom units and 5.169 angstrom units as the atomic percent of Ir varies within the range of interest between 28 percent and 20 percent and is thus well-suited for the epitaxial growth of Nb.sub.3 Ge which should have a lattice constant of approximately 5.135 angstrom units at the stoichiometric composition if extrapolation from thermodynamic equilibrium is valid.

Epitaxial Growth of Nb.sub.x Ge on Nb.sub.x Ir: Nb.sub.x Ge was grown, as described by electron beam coevaporation, on Nb.sub.x Ir substrates. Samples were grown with different Nb.sub.x Ir compositions of which some were designed to expand and some to contract the Nb.sub.x Ge lattice constant.

FIG. 2 shows the experimentally measured lattice constants for the Nb.sub.x Ge compositions grown on Nb.sub.x Ir. The different signs used for the experimental points represent, as also in FIGS. 3, 4 and 5, measurements on different samples. It is evident that the Nb.sub.x Ge lattice neither expanded nor contracted but, within experimental errors, grew with its natural lattice constant as obtained from a straight line extrapolation of the bulk equilibrium value. The Nb.sub.x Ge lattice constant is not sensitive, within the limits previously given, to the Nb.sub.x Ir lattice constant. It is also apparent that the boundary of the A15 phase has been extended from 19 atomic percent to approximately 26 atomic percent Ge. The presence of the Nb.sub.x Ir film having an A15 phase has caused the free energy of the Nb.sub.x Ge system to minimize by forming a pure A15 phase. The observed linear variation in lattice constant is typical of the equilibrium range of a given phase.

The effect of the epitaxial growth, and the presence of the A15 phase at the stoichiometric composition, on the transition temperature is shown in FIG. 3 which plots the T.sub.c onset temperature of Nb.sub.x Ge versus atomic percent Ge. The maximum transition temperature occurs around 25 atomic percent Ge where a minimum in transition width of about 0.9 degrees K. also occurs. Transition temperatures in excess of 20 degrees K. are easily obtained. Neither the lattice constant nor T.sub.c is particularly sensitive, within plus or minus 0.01 Angstroms, to the Nb.sub.x Ir lattice constant.

EXAMPLE 2

A polycrystalline Nb.sub.x Rh film grown on sapphire was used as a substrate for growing Nb.sub.3 Ge. x represents the variation in composition of the Nb-Rh binary system over the range of interest. Both the Nb.sub.x Ge and Nb.sub.x Rh layers were approximately 0.5 .mu.m thick.

Nb.sub.x Rh films on sapphire: The Nb.sub.x Rh system is more complicated than the Nb.sub.x Ir system. In the composition range of interest, extending from 24 to 32 atomic percent Rh, X-ray diffractometer traces indicate that the Nb.sub.x Rh films generally contain three phases including an A15 structure with a lattice constant that varies between 5.13 Angstrom units and 5.09 Angstrom units.

Epitaxial Growth of Nb.sub.x Ge on Nb.sub.x Rh: Samples were prepared by electron beam coevaporation, as previously described, with Nb.sub.x Rh deposited at 400 degrees C. on a sapphire substrate. The substrate temperature was raised to 875 degrees C. prior to the deposition of Nb.sub.x Ge. The lattice constants for these Nb.sub.x Ge compositions grown on Nb.sub.x Rh versus atomic percent Ge are shown in FIG. 4 and it is apparent that the boundary of the Nb.sub.x Ge A15 phase has been extended to approximately 25 atomic percent Ge. The values of the lattice constants are, however, approximately 0.0035 Angstrom units smaller than the extrapolated equilibrium values.

FIG. 5 plots the T.sub.c onset temperature as a function of atomic percent Ge for the Nb.sub.x Ge samples epitaxially grown as described on Nb.sub.x Rh substrates. The peak transition temperature appears to be in the vicinity of 24 atomic percent Ge. The transition temperatures obtained are comparable to those obtained on Nb.sub.x Ir substrates although the transition widths are about three times larger than for the compositions prepared on Nb.sub.x Ir. The good results obtained on a substrate containing a mixture of phases including the A15 illustrates that the substrate may contain several phases only one of which need match, with respect to type and lattice constant, that of the material grown.

The techniques described can also be used to prepare Nb.sub.3 Al and Nb.sub.3 Si in a pure A15 structure at their stoichiometric compositions. Suitable substrates for Nb.sub.3 Al and Nb.sub.3 Si are Nb.sub.3 Pt and Ti.sub.3 Au, respectively. Additionally, mixtures, such as Nb.sub.3 Ge.sub.x Al.sub.1-x, x between 0 and 1, of the described compositions may be epitaxially grown on suitable substrates.

Additional superconducting materials that may be grown with the method described include NbN and Nb.sub.x V.sub.1-x C, x between 0 and 1, which may be grown on a silicon carbide substrate.

Claims

1. A superconducting body comprising a substrate; and a composition having a phase and a stoichiometric composition, said composition having a lattice constant, characterized in that said composition is epitaxially grown on a substrate having a lattice constant matching that of said composition within plus or minus 0.04 Angstroms; said composition is selected from the group consisting of Nb.sub.3 Ge, Nb.sub.3 Al, Nb.sub.3 Si and mixtures thereof. NbN and Nb.sub.x V.sub.1-x C, x between 0 and 1; said substrate being selected from the group consisting of Nb.sub.x Ir, Nb.sub.x Rh, Nb.sub.3 Pt, Ti.sub.3 Au and SiC.

2. A body as recited in claim 1 in which said lattice constants match within plus or minus 0.01 Angstroms.

3. A body as recited in claim 2 in which said substrate is a niobium containing composition.

4. A body as recited in claim 3 in which said composition is Nb.sub.3 Ge and said substrate consists essentially of a composition selected from the group consisting of Nb.sub.x Ir and Nb.sub.x Rh.

5. A body as recited in claim 4 in which said substrate composition is Nb.sub.x Ir and in which said substrate has a lattice constant between 5.169 Angstrom units and 5.125 Angstrom units.

6. A body as recited in claim 4 in which said substrate composition is Nb.sub.x Rh and in which said substrate has a lattice constant between 5.13 Angstrom units and 5.09 Angstrom units.