WO1994002862A1

WO1994002862A1 - Superconductor thin film crossovers and method

Info

Publication number: WO1994002862A1
Application number: PCT/US1993/006687
Authority: WO
Inventors: Eric Jeffrey Smith; Michael Martin Eddy
Original assignee: Superconductor Technologies, Inc.
Priority date: 1992-07-20
Filing date: 1993-07-15
Publication date: 1994-02-03

Abstract

A method for providing a buried insulator (24, 25) in a high temperature superconductor (30) is disclosed. In the preferred method, a single layer high temperature superconductor (30) is subjected to ion implantation. In one embodiment, the method is used to provide a crossover in a single layer high temperature superconductor (30). This crossover can be used in a variety of devices and instrumentations such as, for example, antennas and pickup coils, detector electronic leads, multi-chip modules, SQUIDs, microwave components and integrated semiconductor-superconductor devices.

Description

DESCRIPTION

Superconductor Thin Film Crossovers and Method

Field of the Invention

This invention relates to useful devices fashioned from superconducting thin films. More particularly, it relates to superconducting cross-overs or effectively multi-layer superconducting devices.

Background

Superconductivity refers to that state of metals and alloys in which the electrical resistivity is zero when the specimen is cooled to a sufficiently low temperature. The temperature at which a specimen undergoes a transition from a state of normal electrical resistivity to a state of superconductivity is known as the critical temperature ("T_c") .

Until recently, attaining the T_c of known supercon- ducting materials required the use of liquid helium and expensive cooling equipment. However, in 1986 a super¬ conducting material having a T_c of 3OK was announced. See, e.g., Bednorz and Muller, Possible High Tc Superconduct¬ ivity in the Ba-La-Cu-0 System, Z.Phys. B-Condensed Matter 64, 189-193 (1986) . Since that announcement superconduct¬ ing materials having higher critical temperatures have been discovered. Currently, superconducting materials having critical temperatures in excess of the boiling point of liquid nitrogen, 77K at atmospheric pressure, have been disclosed.

Superconducting compounds consisting of combinations of alkaline earth metals and rare earth metals such as barium and yttrium in conjunction with copper (known as "YBCO superconductors") were found to exhibit supercon- ductivity at temperatures above 77K. See, e.g., Wu, et al. , Superconductivity at 93K in a New Mixed-Phase Y-Ba- Cu-0 Compound System at Ambient Pressure, Phys. Rev. Lett. 58, No. 9, 908-910 (1987) . In addition, high temperature superconducting compounds containing bismuth have been disclosed. See, e.g., Maeda, A New High-Tc Oxide Super¬ conductor Without a Rare Earth Element, J. App. Phys. 37, No. 2, L209-210 (1988) ; and Chu, et al. , Superconductivity up to 114K in the Bi-Al-Ca-Br-Cu-0 Compound System Without Rare Earth Elements, Phys. Rev. Lett. 60, No. 10, 941-943 (1988) . Furthermore, superconducting compounds containing thallium have been discovered to have critical tempera- tures ranging from 90K to 123K (the highest critical temperatures to date) . See, e.g., G. Koren, A. Gupta, and R.J. Baseman, Appl. Phys. Lett. 54, 1920 (1989) .

These high temperature superconductors ("HTSCs") have been prepared in a number of forms. The earliest forms were preparation of bulk materials, which were sufficient to determine the existence of the superconducting state and phases. More recently, thin films on various sub¬ strates have been prepared which have proved to be useful for making practical superconducting devices. For example, the applicant's assignee has success¬ fully produced thin film thallium superconductors which are epitaxial to the substrate. See, e.g., Olson, et al . , Preparation of Superconducting TICaBaCu Thin Films by Chemical Deposition, Appl. Phys. Lett. 55, No. 2, 189-190 (1989) , incorporated herein by reference. Techniques for fabricating and improving thin film thallium superconduct¬ ors are described in the following patent and copending applications: Olson, et. aJL. , U.S. Pat. No. 5,071,830, issued December 10, 1991; Controlled Thallous Oxide Evaporation for Thallium Superconductor Films and Reactor Design, SN: 516,078, filed April 27, 1990; In Situ Growth of Superconducting Films, SN: 598,134, filed October 16, 1990; Passivation Coating for Superconducting Thin Film Device, SN: 697,660, filed May 8, 1991; and Fabrication Process for Low Loss Metallizations on Superconducting Thin Film Devices, SN: 697,960, filed May 8, 1991, all incorporated herein by reference. There are three superconducting oxide systems which are being developed as high T_c thin films for operation above 77K. These are YBaCu oxide (YBCO) , BiSrCaCu oxide (BSCCO) and TlBaCaCu oxide (TBCCO) . Each has reached a different level of maturity in materials development and each has different advantages in areas of critical current density (J_c) , critical temperature (T_c) and weak-link coupled junction properties. The thallium system has the highest recorded T_c of 125K, and the liquid mediated growth techniques used for thin film preparation allow growth of relatively thick («lμm) films over large areas. Films on areas larger than 2" round are desirable for certain applications.

Several devices require both large areas and high quality, superconducting crossovers. For example, Superconducting Quantum Interference Device (SQUID) based sensor applications need continuous superconductor loops over very large areas to allow separation between detector and balancing loops (Fig. 1) . The large separation between detector and cancellation is essential for good sensitivity. The cancellation loop will also pick-up the signal of interest at small spacings.

SQUID loops are currently made in HTSCs using a tri- layer, HTSC-dielectric-HTSC structure grown in situ. See, e.g., W. Eidelloth, et a] ., Appl. Phys. Lett. 5_9, 3473

(1991) ; L.P. Lee, et a] ., Appl. Phys. Lett. 59., 3051

(1991) ; and M.J. Ferrari, et al . , Appl. Phys. Lett. 58,

1106 (1991) . Good isolation between HTSC layers has been produced although shorts are possible from protrusions within each layer. In addition, the quality of the super¬ conductor, despite being epitaxial over most of the cross¬ over, is very poor on the inclined sections (see Fig. 2) . However, the main drawback to using an in situ growth technique for building a superconducting crossover, is the difficulty in scaling to large areas. Currently, super¬ conducting YBa₂Cu₃0₇ thin films can be grown on 2" round substrates. On the other hand, separating the deposition and crystallization steps allows for much easier scaling of the process. The applicant's assignee has developed a two step, ex situ crystallization process which currently produces epitaxial Tl₂Ca₁Ba₂Cu₂0₈ films, primarily for use in microwave circuits. The process, known as the Epitaxial Low-loss Film (ELF) process is easily scalable to 4" or 5" wafers and is very cost effective. See, e.g., Controlled Thallous Oxide Evaporation for Thalliumm Superconductor Fiulms and Reactor Design, SN: 516,078, filed April 27, 1990, incorporated herein by reference.

At the applicant's assignee deposition of amorphous precursors can be performed over areas up to 5" round using laser ablation. The applicant's assignee has devel¬ oped the ELF process for manufacture of high performance thallium/barium/calcium/copper oxide (TBCCO) films on

LaA10₃ substrates. Targets for laser ablation are hot pressed from metal oxide powder mixtures. Single crystal

(100) LaAl0₃ wafers (2" diameter) are purchased and cleaned. Amorphous precursor films are deposited by laser ablation onto the substrates at room temperature. The precursors are converted to epitaxial superconducting films by annealing at high temperature, using a special reactor to control oxygen and thallium oxide partial pressures. Films are inspected and tested for composi- tion, defects, and electrical properties before being fabricated into devices. Dicing is completed after film growth.

Substrates compatible with HTSCs are not currently available greater than 3" round. Thus, the large area processing capability is used to deposit and process wafers in a batch mode. Three 2" round wafers are deposi¬ ted simultaneously and crystallized in a large furnace. The positioning of the substrates is shown in Fig. 8.

Fig. 9 shows the thickness and compositional uniform- ity over the 5" round area of deposition. Deviations in composition are within the precision of the measurement. Energy dispersive spectroscopy using an electron beam as the excitation source was used for these measurements.

To illustrate the uniformity in properties resulting from this kind of deposition, three 2" round TBCCO wafers which were deposited and processed as a batch were pat¬ terned using the standard characterization mask shown in Fig. 10. The mask has six resonators, small (200 μm) and long (2m) J_c bridges, patterning resolution and contact resistance structures. Figs. 11A-11C show the results taken from all wafers for one such batch experiment. The T_c data were taken from the sixteen 1cm² samples from the three wafers. The micro¬ wave performance shows the low power, unloaded Q data from eighteen 5.6 GHz test resonators which are designed to stress the film. All data is comparable to data collected from single wafer processing. Note the consistemcy of the data from wafer to wafer.

The critical current density ("J_c") was also measured on two of the short 4-probe structures. The design is 200μm long, with a width of 20μm. The J_c varied from 0.5 to 1.0 MA/cm². These results show the potential for large volume, high quality film production available at the applicant's assignee.

Thus, a single layer, large area superconductor growth method already exists. The present invention uses ion implanting techniques for developing a planar cross¬ over technology by defining channels within a single layer superconductor.

Ion implants are used routinely in the semiconductor industry. See, e.g., K.A. Picker, Ion Implantation in Silicon - Physics, Processing and Microelectronic Devices, Appl. Solid State Sci. 5 (1975) ; and J.F. Gibbons, ion Implantation in Semiconductors Part I Range Distribution Theory and Experiments, Proceedings of I.E.E.E. 56, 295 (1968) . Another related system more similar to HTSC systems is HgCdTe. Extensive research has been performed varying ions, energy and dose in HgCdTe. See, e.g., R.G. Wilson, J. Appl. Phys. 6_2, 5302 (1988) . This is very similar to the basic work required for HTSC films in this program. Fig. 12 shows the depth distributions for vari¬ ous ions implanted in HgCdTe at 200keV. This information is summarized in Fig. 13, where the peak depth covers the thickness of TBCCO films. Thus, if a similar distribution exists for TBCCO films, appropriate ions and energies should be available to define the structure necessary for a crossover. Many groups have explored the creation of buried dielectric layers as a means of forming silicon on insu¬ lator (SOI) for microelectronic applications. Most attention has been focused on the generation of Si0₂ as the insulating layer. See, e.g., P.L.F. Hennert, SPIE Proc. 530, 230 (1985) . This insulating layer is generated by ion implanting 0⁺ into the Si wafer. Figs. 14A and 14B show the SIMS depth profile of two samples generated by oxygen ion implanting. These profiles illustrate the effect of increasing the dose on the thickness of the implanted region. The substrate is maintained at the temperature of 500°C, so that when the volume concentra¬ tion of oxygen reaches 4xl0²² cm^"3 Si0₂, will be directly synthesized as a buried oxide.

As described above, ion implanting has been used extensively in the semiconductor industry to dope layers of semiconductor, to isolate semiconductor layers with an insulator, or to introduce a buried metal layer within a crystalline semiconductor. Ion implanting of supercon¬ ductors has been exploited by other groups but all of the experiments have been aimed at producing junctions or investigating the effects of ion induced disorder on the superconducting properties. See, e.g., S. Vadlamannati, et al., Appl. Phys. Lett. 5_7, 2265 (1990) ; and M.P. Siegal, et al. , Appl. Phys. Lett. 6_0, 2932 (1992) . Summary of the Invention

The devices of the present invention comprise high temperature superconducting (HTSC) crossovers. The present invention uses the isolation technique in super- conducting thin films. The present invention uses ion implants to define channels for a high quality supercon¬ ducting crossover. The crossovers are cost effective and compatible with wafers larger than two inches. The preferred embodiments comprise high temperature superconductor crossovers in a single layer material .

The present invention uses ion implanting techniques for developing a planar crossovers by defining channels within a single layer thin film superconductor to provide a high quality single layer thin film superconducting crossover. Optionally, physical channels such as ion milling may be partially used to define the crossover.

Accordingly, it is a principal object of this invention to provide a buried insulator in a high temperature superconductor. It is also an object to provide a high temperature superconductor crossover in a single layer material .

It is a further object to provide a method of manufacturing single layer HTSC crossovers.

It is an additional object to provide high temperature superconductor crossovers in a single layer thin film superconductor through the use of ion implanting techniques.

It is also an object to provide a superconducting crossover for enabling HTSC devices to be fabricated from large area films.

It is yet an additional object to provide a super¬ conducting crossover suitable for practical devices and instrumentation (e.g., antennas and pickup coils, detector electronic leads, multi-chip modules, SQUIDs, microwave components and integrated semiconductor-superconductor devices) . It is still a further object to provide a crossover for use in a high quality superconducting planar gradiometer for use with SQUIDS.

Brief Description of the Drawings Fig. 1 shows a schematic for a SQUID sensor.

Fig. 2 shows a cross-sectional view of a crossover grown by an in situ technique.

Fig. 3 shows a cross-sectional view of an ion implanted crossover. Fig. 4 shows a top view of the crossover of Fig. 3.

Figs. 5A-5G show the processing sequence for building the crossover of Fig. 3.

Fig. 6 shows the SIMS depth profile of a TBCCO film grown on NdGa0₃. Fig. 7A shows a test structure for electrical characterization of ion implanted crossovers.

Fig. 7B shows a cross-sectional view of the device of Fig. 7A taken along 7B-7B in Fig. 7A.

Fig. 8 shows a substrate holder for simultaneous deposition of three 2" round wafers covering a 5" round area.

Fig. 9 shows thickness and compositional uniformity across a 2" diameter of the large area deposition.

Fig. 10 shows a mask design used to characterize 2" round films.

Fig. 11A shows T_c and δT_c for the 1cm² from the three 2" round wafers.

Fig. 11B shows unloaded Q at low input power (-60dBm) and 77K for 5.6 GHz resonators from the 2" round wafers. Fig. 11C shows full width at half peak maximum for the (0010) reflection from the 1cm² from the three 2" round wafers .

Fig. 12 shows depth distributions for selected ions implanted in HgCdTe at 200keV. Fig. 13 shows values of profile peak depth (R ) versus ion atomic number (Z) for 200keV in HgCdTe. Figs. 14A and 14B show SIMS depth profiles for samples implanted with 200keV oxygen ions.

Detailed Description of the Invention

As described above, many devices require high temper- ature superconducting crossovers. For example, SQUID based sensors need continuous superconductor loops over large areas to allow separation between detector and balancing loops. As shown in Fig. 1, a SQUID sensor 10 comprises a SQUID 11 shielded by a magnetic shield 12 and a continuous superconductor loop which has a detector loop 13 and a cancellation loop 14 separated by a crossover 15. Traditional SQUID sensors 10 use crossovers 15 which have a tri-layer structure as shown in Fig. 2. This structure is fabricated on a substrate 16 and has a first layer of HTSC 17, a second layer of dielectric 18, and a third layer of HTSC 19. The quality of the superconductor is very poor on the incline sections 20. An additional problem with this structure is that it is grown in situ and, therefore, as described above, is difficult to scale to large areas.

The crossover device of the present invention is scalable to large areas . The present invention uses ion implants to define channels for a high quality supercon¬ ducting crossover in a single layer HTSC film. The use of ion implants to define a crossover structure represents a novel use of this technology. The structure of the present invention, a high quality crossover 21, is shown in Figs. 3 and 4. The crossover 21 comprises an underpass 22 and overpass 23 separated by deep and shallow ion implanted insulating, doped or amorphous regions 24 and 25, and suitable ion milled trenches 26.

The steps required to generate the desired crossover structure are shown in Figs. 5A-5G. As shown in Figs. 5A- 5E, the first step comprises isolating a bottom layer HTSC to form the underpass 22 of the crossover 15 described above (Figs. 3 and 4) . Figs. 5A-5E are oriented such that the underpass 22 is shown as being perpendicular to the plane of the drawing (i.e. coming out at the viewer) . As shown in Fig. 5A, photo resist 32 is added to a single layer HTSC thin film 30 fabricated on a substrate 31. Next, deep ion implants 24 are added to the film 30 (Fig. 5B) . Also shown in Fig. 5B are areas of ion damage 33 due to the addition of the deep ion implants 24. Then, as shown in Fig. 5C, the photo resist 32 is removed from the center section of the film 30 and a shallow ion implant 25 is added which isolates the underpass 22. Then the top HTSC is regrown or re-crystalized to re-establish the HTSC over the shallow ion implant 25 (Fig. 5E) . Thus, the underpass 22 is isolated.

As shown in Figs. 5F-5G, the second step comprises isolating a top layer HTSC to form the overpass 23 of the crossover 15 described above (Figs. 3 and 4) . Figs. 5F-5G are oriented such that the overpass 23 is perpendicular to the plane of the drawing (i.e. coming out at the viewer) and the underpass 22 extends along the plane of the draw- ing (i.e. from the right side of the drawing to the left or vice versa) . Fig. 5F shows that photoresist 32 is added to the film 30. Fig. 5G shows that trenches 26 are ion milled into the film 30 to a depth of just below the shallow ion insulator 25. Alternatively, the isolation may be made by ion implantation. Thus, the overpass 23 is isolated perpendicular to the underpass 22 (see Fig. 4) .

Incorporating the superconducting crossover technology of the present invention with large area, cost effective superconducting thin films enables, for example, the building of HTSC SQUID sensors. Large area films, up to 3" round, are currently available in small quantities. The key enabling technology for the production of large area films suitable, for example, for SQUID sensors is crossovers in a single layer material such as that of the present invention. The key materials processing step is the retention of a thin amorphous or insulating layer buried within the TBCCO film. A matrix of ion species, dose and energy can be used to determine damage and doping in the superconducting thin film. To further understand the role of dopant on the superconductor, the elements for this invention can be chosen to range from elements which are known to destroy the superconducting properties (Si, Al) to those which are already incorporated within the structure (O, Cu) . A discussion of the technical problems follows.

Ion Implanting of TBCCO films Light ions such as Li⁺ and B⁺ can be used in deep ion implants. Ions such as copper and silver have shallow projected ranges because of their relatively large mass, but can be useful if the TBCCO film is re-grown. Films can be exposed to different doses and different energies to define their projected ranges. For good isolation of different HTSC regions, a buried amorphous layer is desirable.

Different elements, energies and doses can be used. A number of possibilities are shown in Table 1. The energy range, elements and doses are well known in the art. The implants can be done in several stages to (i) find the optimum dose to create an amorphous layer in the superconductor, and (ii) find the combination of element and energy to place the doped/amorphous region in the correct position within the layer.

Table 1 Sample Ranges of Ion Species, Energy and Dose Element Ener

SIMS to evaluate projected ranges of the dopants

Despite the roughness of TBCCO films after suitable planarization, sputtered secondary ion mass spectroscopy (SIMS) can be used to investigate concentration gradients of ions within films made. Fig. 6 shows the depth profile of a film deposited on NdGa0₃. Gallium diffuses into the superconductor, degrading the electrical properties of the film. The Tc for this film was 86K. Films made on LaAlO-,> which have no detectable interdiffusion have T_cs greater than 100K.

By using sputtered SIMS on implanted films, the regions of maximum doping (the peak depth or projected ion range) can be detected. The roughness of the TBCCO films does degrade the quality of the data, but these results will be invaluable in determining the position of the implanted regions.

Healing of damaged regions

After ion implant, the area of film in the ion path is severely damaged; however, much of the crystallinity remains intact . An annealing process is required to recover the crystal quality of the original film. A temperature above 700°C will allow sufficient solid state diffusion to heal the defects. The film quality can be investigated primarily through electrical characterization as described below. However, some use of selected area electron channeling may prove useful if the HTSC is annealed to a sufficiently high quality. Above 700°C loss of thallium as Tl₂0 will result in film degradation. TBCCO Re-growth

For the prei red embodiment of the present invention, re-growing an extra TBCCO layer on the film surface can be advantageous. Although having the flexibility of re-growing TBCCO would be useful for thick layers or producing undoped superconductor, the advantage of the simple planar crossover technology of the present invention would be negated. Thus, the present invention provides a single layer procedure. To generate an extra TBCCO layer (recall Figs. 5D and 5E) , a shadow mask can be used to deposit the layer over the small area above the ion implanted region. The film then goes through an additional high temperature process¬ ing step to crystallize the extra layer of material. This technique is often used in the semiconductor industry, particular for there-growth of an epitaxial GaAs layer.

Electrical Characterization

Electrical characterization of the processed films can use a test structure 40 as shown in Figs. 7A and 7B. This structure 40 allows independent testing of the top and bottom layers in addition to characterizing their isolation. This is best achieved by ion milling "mesas" around the ion implanted regions 41, to a depth defined by the projected range of the ions. For example, trenches 26 can be ion milled through to the depth of the ion implant region 41. Gold contacts 42 can be attached to the top and bottom layers as shown in Fig. 7A. Both longitudinal

(i.e. underpass) and transverse (i.e. overpass) transport properties can be measured for the various implant and anneal conditions.

The size of the structures can be small so that an estimate of the critical current density can be derived. The properties of the different superconducting paths are characterized by paths between C-D and E-F shown in Fig. 7A. The insulator is checked through C-F and C-E shown in Fig. 7A. Testing

The element species, energy of the ion beam and ion dose can be varied to define the projected ranges and damage on TBCCO superconducting thin films. After ion implanting, the films can be characterized both electric¬ ally and by sputter SIMS to locate the ion implanted region.

SIMS depth profiles

To find the position of the doped layers, SIMS depth profiling can be used. This is useful only for those elements which do not already exist within the TBCCO structure.

Re-crvstallize top, damaged layer of the ion implanted film The ion implanted wafers can be diced and parts of the wafer re-crystallized at different temperatures.

Although the invention has been described with respect to specific preferred embodiments, many variations and modifications may become apparent to those skilled in the art. It is therefore the intention that the appended claims be interpreted as broadly as possible in view of the prior art to include all such variations and modifications.

Claims

1. A single layer high temperature superconductor having a buried insulator.

2. A single layer high temperature superconductor having a crossover.

3. The device of claim 2 wherein the crossover comprises a buried insulator.

4. A method of providing a buried insulator in a high temperature superconductor comprising the steps of isolating an underpass by incorporating deep ion implants and shallow ion implants into the superconductor, and isolating an overpass by forming trenches in the superconductor.

5. The method of claim 4 wherein the trenches are formed by ion milling.