High Temperature Crystalline Superconductors From Crystallized Glasses

The present invention relates generally to a method of producing a high critical temperature (T.sub.c) superconducting ceramic material. More particularly, the invention relates to a superconducting ceramic material composed substantially of a single phase prepared by heat treating a starting, glassy ceramic material within a preselected time and temperature transformation region.

High T.sub.c ceramic superconductor materials of the Bi-Sr-Ca-Cu-O system have recently been discovered, and substantial effort has been expended in an attempt to understand the fundamental characteristics of these ceramics. Preparation of the high T.sub.c ceramic materials typically has involved conventional ceramic processing techniques. In particular, work has been directed to the production of the highest T.sub.c phase of Bi-Sr-Ca-Cu-O wherein the T.sub.c is about 110.degree. K., compared to other crystalline phases having T.sub.c values of 20.degree. K. and 85.degree. K. These approaches usually involve heat treatment of sintered polycrystalline materials, but to date there has not been a suitable method devised for reliable Production of significant quantities of the 110.degree. K. phase. Current methods result in sizable quantities (typically more than about 10%) of the lower T.sub.c phases being produced, and these phases cannot be eliminated by conventional heat treatment. Such multi-phase structures have substantial commercial disadvantages compared to a single phase structure having a T.sub.c of 110.degree. K. Such polycrystalline starting materials include various inherent barriers to forming a single phase material, such as grain boundaries and contaminants which are difficult to overcome in order to form the desired single phase structure.

It is therefore an object of the invention to provide an improved ceramic superconductor material and method of manufacture.

It is another object of the invention to provide a novel ceramic superconductor material and method of manufacture of a single phase crystalline structure having a high T.sub.c.

It is a further object of the invention to provide an improved ceramic superconductor material prepared by heat treating an amorphous starting material under preselected time and temperature conditions.

It is an additional object of the invention to provide a novel ceramic superconducting material composed of Bi-Sr-Ca Cu-O having a Ca:Bi ratio in excess of or equal to about 1.0 and less than or equal to about 1.5.

A significant feature in accordance with the invention lies in the production of a high T.sub.c ceramic superconductor having substantially a single crystalline phase. The starting material has a predetermined composition, and an amorphous phase is prepared from the starting material. The amorphous phase undergoes a heat treatment within certain time and temperature transformation regions, which results in transformation from the amorphous phase to a crystalline material comprised substantially of a single phase. The amorphous phase can be prepared by conventional splat quenching methods, such as spin casting.

Further objects and advantages of the present invention, together with the organization and manner of operation thereof will become apparent from the following detailed description of the invention when taken in conjunction with the accompanying drawings.

FIG. 1 illustrates a method of preparation of high T.sub.c superconducting material;

FIG. 2A shows Magnetization Shielding Effect measurements on superconducting composition of the Bi-Sr-Ca-Cu-O system; FIG. 2B (1-3) shows resistivity measurements for selected annealing time periods on various compositions from the Bi-Sr-Ca-Cu-O system; and

FIG. 3 illustrates a time and temperature transformations diagram for one of the compositions of the superconducting ceramic system; and

FIG. 4A-C illustrates x-ray diffraction spectra for three compositions of the Bi-Sr-Ca-Cu-O system undergoing cyclic phase changes.

Referring now to FIG. 1, an improved high T.sub.c ceramic superconducting material can be prepared in accordance with the invention in the illustrated manner. The starting, raw material can be a chemical or mechanical mixture of the selected composition for any one of the new ceramic superconductors. For example, several types of compounds are the Bi-Sr-Ca-Cu-O system or the Y-Ba-Cu-O system. The preferred chemical composition range for the first system is Bi.sub.x Sr.sub.b Ca.sub.z Cu.sub.y O.sub.z with a ratio of a:x greater than or equal to 1.0 and less than or equal to 1.5, such as, Bi.sub.2 Sr.sub.2 Ca.sub.2 Cu.sub.4 O.sub.z, Bi.sub.2 Sr.sub.2 Ca.sub.3 Cu.sub.4 O.sub.z and Bi.sub.2 Sr.sub.2 Ca.sub.4 Cu.sub.5 O.sub.z. It has been found that in the preferred embodiment that one can readily obtain Bi.sub.2 Sr.sub.2 Ca.sub.2 Cu.sub.3 O.sub.z by starting with excess Ca (a:x.gtoreq.1) to compensate for loss of some Ca during processing. Typical raw materials are therefore appropriate quantities of Bi.sub.2 O.sub.3, SrCO.sub.3, CaCO.sub.3 and CuO. These materials are converted to powders by thoroughly mixing in a wet ball mill in a conventional manner. The premixed powders are calcined in air at about 800.degree. C. for 12 hours, and this prereacted material is melted in a platinum crucible. The molten mixture in the crucible is splat quenched onto a heat sink (such as a copper block) to form an amorphous, or glassy phase. Conventional methods of performing such splat quenching operations include spin casting.

Once the amorphous phase (or phases) has been formed, the amorphous material is subjected to a heat treatment for a preselected temperature and time, causing transformation from the amorphous state to the crystalline state. Various preferred compositions have undergone numerous temperature/time treatments in order to determine suitable transformation regions which will produce a substantially single phase crystalline material of the high T.sub.c superconductor.

Important advantages arise from genera high T.sub.c crystalline phase from an amorphous starting material. When prepared properly the amorphous starting material itself has little or no porosity. The resulting material crystallized from the amorphous material also exhibits a very low porosity and has a density approaching one hundred percent. In addition, nucleation and growth of crystalline grains from the amorphous state results in a much more stable structure, such as low energy grain boundaries. This is unlike sintered crystalline materials which exhibit high energy, mismatched grain boundaries and also such materials tend to have substantial intergranular porosity. The porosity and the high energy grain boundaries tend to diminish the electrical transport properties in such polycrystalline, sintered materials. The amorphous starting material used in the invention is also highly uniform compared to a starting mechanical mixture of polycrystalline metallic oxides to be combined to form the high T.sub.c ceramic materials. Consequently, if one begins with an amorphous starting material, the kinetics are much more favorable for the formation of a uniform high T.sub.c ceramic microstructure compared to beginning with mixtures of polycrystalline metal oxides (such as, Bi.sub.2 O.sub.3, SrCo.sub.3 , CaCo.sub.3 and CuO).

In the case of the Bi-Sr-Ca-Cu-O system, the desired ceramic crystalline phase exhibits a T.sub.c of about 110.degree. K. as determined by resistivity measurements and conventional Magnetization Shielding Effect measurements (See, FIG. 2). As can be noted from the data in FIG. 2A and B, there can be more than one crystalline phase produced from the amorphous material, depending on the composition, the time and the temperature of the heat treatment schedule. In FIG. 2A both the Bi.sub.2 Sr.sub.2 Ca.sub.2 Cu.sub.3 O.sub.x and Bi.sub.2 Sr.sub.2 Ca.sub.4 Cu.sub.5 O.sub.z, show evidence of at least two crystalline phases, having characteristic T.sub.c values of 110.degree. and 85.degree. K. For the illustrated heat treatment conditions only the Bi.sub.2 Sr.sub.2 Ca.sub.3 Cu.sub.4 O.sub.z has a single T.sub.c value of 110.degree. K. Due to the possibility of conduction occurring along a percolation path thru the material, conventional resistivity measurements (see, FIG. 2) cannot provide conclusive evidence that a material is only composed of one T.sub.c phase. Thus, in order to be certain the material is substantially a single phase, one should perform another measurement, such as, a conventional Magnetization Shielding Effect measurement which does account for the relative volumes of the different magnetic phases present in the superconducting material. As seen in FIG. 2 A, (upper plot) the Bi.sub.2 Sr.sub.2 Ca.sub.3 Cu.sub.4 O.sub.z, which has been heat treated for only seven days, does indeed exhibit the presence of a second phase having a T.sub.c of 85.degree. K. After further heat treatment for a total of ten days, the Magnetization Shielding Effect signal arising from the 85.degree. K. phase has virtually vanished, leaving substantially a single phase high T.sub.c (110.degree. K.) material (See FIG. 2A, lower plot).

For each of the preferred compositions of the Bi-Sr-Ca-Cu-O system, a matrix of heat treatment schedules were performed. Details of these results are set forth in the Examples described hereinafter. The treated specimens were each characterized by the Magnetization Shielding Effect measurements as well as resistivity measurements, and a time-temperature transformation diagram can be determined as shown in FIG. 3. The various regions resulting from this determination include amorphous regions, crystalline regions and mixtures of these two general phases. Similar such transformation diagrams can also be prepared for other systems, such as the Y-Ca-Cu-O system, in order to define other suitable transformation regions for forming high T.sub.c crystalline material from amorphous starting material.

In other forms of the invention, once the high T.sub.c ceramic, single-phase crystalline material has been formed, various commercial products can be produced as depicted generally in FIG. 1.

In another form of the invention a phase change cycle is observed during the crystallization annealing process. This cyclic phase change phenomenon has been followed by electron microscopy and also by x-ray diffraction results shown in FIGS. 4A, 4B and 4C for the samples having nominal composition 2223,2234, and 2245, respectively. The angle 2 theta and d-spacing values are indicated in Table I and II for the 85K and 110K phases, respectively.

The x-ray diffraction data for the "2223" samples annealed at 870.degree. C. for 1 day, 3 days, 5 days, 7 days, 10 days, and 16 days are shown in FIG. 4 (a-f) respectively. As seen in FIG. 4A, the annealing process does not have a strong effect in terms of phase changes. The crystallized product is mostly 85K phase, although certain texturing is observed. As can be seen from FIG. 4A (b-e) the (001) peaks are intensifying while the non-(001) peaks are disappearing. Electron microscopy results have indicated that the observed texturing is due to rapidly growing plate-like grains with diameter of 20 micrometers and thickness of 0.5 micrometers at 10 days anneal. These plate-like grains are then replaced by rod-like grains as the anneal is prolonged from 10 days to 16 days. As noted in FIG. 4A(f), the non-(001)) peaks are growing back again in a cyclic manner.

A larger amount of the 110K superconducting phase is formed in crystallized samples as the Ca and Cu contents of the samples are increased to 3 and 4, respectively. The x-ray plots for the 2234 samples are presented in FIG. 4B. As can be seen in the figure, the volume percent of the 110K phase is increasing with annealing time at 870.degree. C. The most noticeable peak of the 110K phase is the (0010) diffraction peak, shown in FIGS. 4B(b-e). However, as the anneal is prolonged from 10 days to 16 days at 870.degree. C., the amount of 110K phase is dropping as indicated in FIG. 4B(f). The texturing effect is also observed in the anneal of the 2234 samples. As seen from FIG. 4B(b-e), the intensities of all non-(001) peaks are decreasing. These peaks are growing back, accompanied with a decrease in the amount of 110K phase as the anneal time is extended from 10 days to 16 days. In this case, there appears to be a phase change cycling period of about 7 days.

The cyling period is shortened to about 3 days as Ca and Cu are further increased to 4 and 5 in the 2245 nominal composition. As shown in FIG. 4C(a and b), a large portion of the 110K phase has already formed in the sample annealed for 3 days at 870.degree. C. The 110K phase is then gradually vanishing as the anneal time increases from 3 to 7 days, as shown in FIG. 4A(c and d), and it should be noted that the disappearance of the 110K phase is accompanied with the development of the non-(001) peaks, and the texturing effect is found to be increasing as the amount of the 110 K phase increases. After 7 days anneal, the 110 K phase percentage increased again with the increasing texturing effect, as shown in FIG. 4C(e). The cycle is repeated at 16 days--the 110 K phase has disappeared and non-(001) peaks are large. When the 16 day-annealed sample is annealed for another 3 days at 870.degree. C. the 110K phase has again grown back.

The following Examples illustrate the preparation, processing and measurements on various high T.sub.c ceramic superconductors and are not meant to limit the scope of the invention which is set forth in the claims following.

Starting materials of Bi.sub.2 O.sub.3, SrCO.sub.3, CaCO.sub.3, and CuO are weighed with nominal compositions of Bi.sub.2 Sr.sub.2 Ca.sub.2 Cu.sub.3 O.sub.z (2223), Bi.sub.2 Sr.sub.2 Ca.sub.3 Cu.sub.4 O.sub.x (2234) and Bi.sub.2 Sr.sub.2 Ca.sub.4 Cu.sub.5 O.sub.x (2245) and thoroughly mixed by a wet ball mill in an agate container. The wet ball milling process used for the Bi-Sr-Ca-Cu-O system requires about forty weight percent of methanol. The milling time is about two hours. After ball milling, the mixed powders are dried in air and put into a platinum crucible and pre-reacted in a box furnace in air at 800.degree. C. for twelve hours. The heating rate is 26.degree. C./min and cooling rate is 7.degree. C./min. The pre-reacted materials are crushed into fairly loose powders and melted in a platinum crucible by a torch using natural gas mixed with oxygen. Two identical copper plates with polished surfaces are pre-heated to about 200.degree. C. prior to quenching. As the material is completely melted in the platinum crucible, it is quickly poured onto the surface of one of the copper plates. In approximately less than one second, the other copper plate is placed on the molten material to carry out a splat quenching operation. The copper plates have the dimension of 120 mm.times.80 mm.times.40 mm. Extremely dense (virtually 100%) glass plates of rapidly quenched materials are thus formed for the subsequent crystallization heat treatment. The glass samples with nominal compositions of 2223, 2234, and 2245 are annealed at 870.degree. C. in air for 1 day, 3 days, 5 days, 7 days, 10 days, and 16 days. The heating rate for the annealing is 26.degree. C./min and the cooling rate is 70.degree. C./min. The annealing experiment is carried out in a Lindberg box furnace. The techniques used for characterizing the annealed samples are x-ray diffraction, scanning electron microscopy, magnetization shielding measurements and resistivity vs. temperature measurements.