The critical current density and trapped field of large grain RE123 bulk superconductors have drastically increased in the last few years. High-quality bulk superconducting magnets can trap a very high magnetic field of 3T at 77K and above 15T at 20K. The recent progress in bulk superconducting materials has led to many applications such as superconducting flywheel systems, current leads, superconducting motors, magnetic separation systems, magnetron sputtering machines. In this paper, we review recent research and development of the processing, critical current density, trapped field and mechanical properties of melt-textured bulk superconducting materials.
Large-grain RE (rare earth elements)-Ba-Cu-O bulk superconductors have potential trapping magnetic fields greater than 5T at liquid-nitrogen temperature. However, cracks are easily formed in large bulk samples due to their poor mechanical properties. Thus, the maximum trapped magnetic field is restricted by the mechanical properties rather than superconducting properties. It is therefore important to seek techniques to enhance the mechanical properties. In this review, I summarize the present status of investigations regarding the mechanical properties for melt-processed RE-Ba-Cu-O bulk superconductors.
In this review we summarize the present status of resin impregnation techniques for bulk high-temperature superconductors. Epoxy resin can permeate into the interior of bulk RE (rare earth element)-Ba-Cu-O superconductors when they are immersed in molten resin. Mechanical properties of resin-impregnated bulk RE-Ba-Cu-O are dramatically enhanced in addition to the enhancement of corrosion resistance. The thermal expansion coefficient can be controlled by adding quartz fillers to epoxy. Wrapping the bulk superconductors with carbon fabric prior to resin impregnation is effective in further enhancing the mechanical properties. Resin-impregnated bulk RE-Ba-Cu-O materials have already been employed for several practical applications.
Superconducting flywheels store energy in a rotating wheel that is supported by superconducting magnetic bearings. The superconducting magnetic bearing (SMB) consists of a bulk superconductor stator and a permanent magnet rotor. The rotor is levitated by the pinning force of the bulk superconductor. The SMB has lower rotational loss than conventional mechanical bearings and is suitable for long-term energy storage such as the diurnal load leveling of electric power. Research and development of a 10kWh system and a 100kWh SMB module using radial-type SMBs are being conducted by ISTEC, Shikoku Research Institute Inc., Ishikawajima-Harima Heavy Industries Co., Imra Material R & D Co., Sumitomo Special Metals Co. and Koyo Seiko Co. as a NEDO project. As a long-term goal, the NEDO project aims to develop a 10MWh power storage system to level diurnal power load. The group of Chubu Electric Power Co. and Mitsubishi Heavy Industries fabricated a thrust-type SMB of 620mm in diameter. Argonne National Laboratory tested a 2.25kWh system using a hybrid bearing consisting of SMBs and permanent magnet bearings. A 10kWh system was studied by Boeing Co. as a SPI (Superconductivity Partnership Initiatives) project in the USA.
We have developed a new, continuous water-purification system that uses high-Tc bulk superconductors to rapidly remove contaminants from water supplies, The maximum magnetic field of the magnetized bulk superconductors is 3.2 T. This membrane-magnetic separator consists of (1) a preapplication treatment unit in which magnetic flocs made up of suspended solids and seeded with ferromagnetic particles are coagulated in a flocculation vessel, (2) a membrane separator that purifies water using a rotating net to gather the flocs and (3) a magnetic separator that recovers the magnetic flocs from the net. This system simultaneously removed 98% of the contaminants from sewage at our factory and recovered sludge at a concentration of 40, 000mg/L. Moreover, it removed 90% of the oil from oily water and recovered sludge at a concentration of 23, 000mg/L. Treatment time was only four to five minutes, and we found that the new system is capable of treatment and sludge recovery at higher speeds than conventional flocculant deposition methods.
Multi-seeded bulk superconductors have been studied in order to improve a problem in the conventional multi-seeding method: the existence of excluded non-supercondcting phases (liquid phases and segregated RE211 phases) at the grain boundaries in the bulk superconductor. First, a crystal growth experiment using a precursor with holes was carried out instead of the experiment with two seed crystals. Exclusion and the disappearance of non-superconducting phases were observed around the holes. We believed that this phenomenon was caused when crystal growth changes to the same direction from opposite directions. Through the extension of this theory, a new multi-seeding method, the MUSLE (MUlti-Seeded seamLEss bulk) technique was derived. The basic concept of this technique is that the precursor is composed of two or more RE-Ba-Cu-O layers with different peritectic temperatures, thereby allowing contorol of the directionthe directions of the crystal growth from the seed crystals. RE-Ba-Cu-O superconductors were fabricated using the MUSLE technique with 4 or 9 seeds. The trapped field distribution obtained at 77K showed a single peak, indicating that the MUSLE technique is effective in eliminating the excluded phases formed at the grain boundaries in multi-seeded bulk superconductors.
We have succeeded in producing a strongly coupled superconducting Y-Ba-Cu-O joint by combining the control of crystal growth along the ‹110› direction and employment of a highly dense Er-Ba-Cu-O sintered solder. No segregation of the secondary phase and residual liquid phase was observed at the joint, which led to the formation of a strongly connected superconducting joint. Microstructural observation showed that many triangular crystals grew from the mother block in the (110)/(110)-joint, while the growth front was parallel to the (100) surfaces in the (100)/(100)-joint. Such a discrepancy was ascribed to the difference in the superconducting properties of the joints.
We have studied the microstructures of subgrains in large-grain RE123 (RE=Y, Nd, Sm) bulk superconductors through microscopic observation and micro-area x-ray pole figure analyses. We have found that misorientation of the subgrains increases along with the crystal growth, and also that the misorientation has a positive relation to the size and concentration of the second-phase particles. An electron backscatter diffraction technique was successfully utilized for characterizing the subgrain boundaries. We have also performed magnetic and electric measurements for several samples having different subgrain structures. The results of AC transport measurements indicated that the subgrains act as weak links, especially in a high magnetic field region.
We investigated the superconducting properties of melt-textured Ho123 and Er123 bulk superconductors containing 5 to 40mol% Ho211 and Er211, respectively. The samples were fabricated with the cold-seeding method in air using a Nd123 bulk seed crystal. The superconducting properties of the Ho-Ba-Cu-O samples were strongly affected by the Ho211 content and oxygen annealing temperature. In the Jc-B curves of the Ho123 samples with low Ho211 contents, a secondary peak effect was clearly observed. The peak position shifted to lower fields as oxygen annealing temperature increased, presumably due to an increase in oxygen vacancies. A secondary peak effect was also found in the Er123 sample with 5mol% of Er211. We believe that the peak effect in these materials is ascribed to oxygen deficiency, since RE/Ba substitution is not present in either system.
The thermal properties (thermal conductivity κ(T), thermal diffusivity α(T), thermoelectric power S(T) and thermal expansion dL(T)/L) of melt-processed Gd-based bulk superconductors are measured in the ab-plane and along the c-direction and compared with those of the Y-based system. The ab-plane κab(T) values of the Gd-based system show a characteristic large enhancement below the superconducting transition temperature Tc, which is suppressed by the magnetic field up to 10T applied perpendicular to the ab-plane. The results on the thermal properties suggest that GdBaCuO bulk crystals make a clean system similar to the Y-based system and are in marked contrast to the Sm-based system, where Sm-Ba substitution occurs.
In order to investigate the mechanical properties of Gd123 single-grain bulk superconductors fabricated using a modified QMG method, tensile tests in the direction of the a/b-or c-axis of small specimens cut from bulk samples have been carried out at 293K. As for the mechanical properties perpendicular to the c-axis, there was no significant difference between those in the crystal growth direction and those perpendicular to it. While the average value of Young's modulus of the bulk sample with 33.0mol%-Gd211 particles, 118GPa, was higher than that of the bulk sample with 28.7mol%-Gd211, 111GPa, the average value of the tensile strength of the former, 36MPa, was lower than that of the latter, 40MPa. Tensile strength and Young's modulus in the c-direction, 10MPa and 37GPa, were quite low compared with those mentioned above. Poisson's ratio based on the transverse strain parallel to the c-axis, 0.15, was significantly smaller than that perpendicular to it, 0.30. In the specimens with higher length, however, the difference was decreased to some extent. With regard to the anisotropy of the Poisson's ratio, the effect of a microcrack opening in the c-axis direction was discussed coupled with the constraints at the interface of the specimen with the sample holder.
A trapped field magnet is one application for melt-processed bulk high-Tc superconductors (HTSs). Evaluation of the maximum stress during magnetization is important from the viewpoint of the destruction of bulk HTS. Stresses in a bulk HTS are numerically evaluated during the magnetization process by field cooling. Shielding current distributions are obtained through macroscopic numerical simulation using Maxwell equations and the critical-state model. In the study of maximum stress, 3D axisymmetirc numerical solutions with open and fixed boundary conditions are compared with 1D analytical solutions. Analytical solutions are different from numerical solutions. It is shown that the evaluation of experimental results on the destruction of a bulk HTS will need numerical analysis.
We have measured the magnetic field and levitation force of bulk Y-Ba-Cu-O superconductors while the external field is controlled with a Bi-Sr-Ca-Cu-O superconducting magnet. The height dependence of the levitation force could be explained well in terms of the field gradient in the sample and radial component of the magnetic field. The field gradient also induced the trapped field and decreased the levitation force. The experiment data was different from the theoretical data as the magnetization process was repeated for levitated bulk superconductor.
We present the dependence of lift and stability on the arrangement of multiple HTS bulks in an active-maglev system composed of HTS bulks and an electromagnet. We experimentally investigated the influence of asymmetric magnetic field distribution in a field-cooling process on the characteristics of levitation such as levitation force and stable-levitation region. In spite of unstable levitation, levitation force was increased with the distance between the axes of the bulk and electromagnet. It may be considered that unstable levitation is caused by asymmetric supercurrent distribution within the bulk. We investigated a suitable arrangement of multiple HTS bulks in an active-maglev system experimentally. Levitation force was measured in four types of multiple-bulk arrangements. It was revealed in the experiment that the levitation force was influenced by the distance between the bulks and that the maximum levitation force was obtained when the bulks were located near the inner boundary of the electromagnet windings. It is believed that the most suitable arrangement is closely related to magnetic field distribution; i.e. electromagnet dimensions (inner and outer radii, thickness).
We have succeeded in constructing the world's first superconducting permanent magnet system capable of generating a strong magnetic field of 3.15T using a pair of meltprocessed bulk superconductors as trapped field magnets. The magnetic field was measured in a 2mm open gap between the magnetic poles set face-to-face after magnetization by the pulsed-field magnetization (“IMRA”) method operated at 35K. The strong field generator is composed of Smbased 123 bulk superconductors, vacuum pumps, pulsed-field coils and GM-cycle refrigerators with compressors. The performance of this compact field generator exceeds those of conventional rare earth magnets or large-scale electromagnets by far. Various application candidates for the strong field generator, such as magnetic separation systems, motors/generators, and a magnetron sputtering apparatus, are proposed in this report.
Our objective was to apply superconducting bulk magnets to a magnetic separation technique. A pair of Sm123 bulk superconductors were magnetized by the “IMRA” method and arranged opposite each other with a gap of 20mm. The maximum flux density at the surface of the magnetic poles containing the bulk magnets and at the center of the gap of the poles was 2.26T and 1.76T, respectively. We set up a magnetic separator by placing a pipe between the magnetic poles of face-to-face superconducting bulk magnets and conducted experiments for two types of magnetic separation: high gradient magnetic separation (HGMS) with the use of a stainless steel filter, and open gradient magnetic separation (OGMS) without filters. The former uses separation filters composed of ferromagnetic wires stuffed into a glass pipe, and therefore a high gradient magnetic field is generated near the wires. The latter does not use filters. The separation ratio is investigated by the experiments using water mixed with fine powder of Fe3O4 and the performance of about 100% is achieved in HGMS.
Because REBaCuO bulk superconductors have large flux pinning forces even in liquid nitrogen, various applications have been investigated. This paper is concerned with a motor using bulk superconductors in the rotor. In order to analyze the characteristics of a bulk superconducting motor in the transient state and those of a bulk superconductor in traveling fields, an experimental apparatus for a hysteresis motor with bulk superconductors in the rotor was constructed. In a pull-out test, the maximum torque of the superconducting hysteresis motor is generated at a speed which is slightly slower than the synchronous speed. The torque decreases with the slip frequency. In a locked rotor test, torque increases with the slip because Joule losses generated in the bulk superconductors decrease the critical current density of the superconductors and consequently their shielding capability. This paper describes the experimental device, the results of pull-out an d locked rotor test and transient characteristics of the motor.
Superconducting flywheel energy storage systems (SFESS) using superconducting magnetic bearings (SMBs) have been studied to contribute to daily load leveling. It is known that the levitation force of SMBs used for SFESS varies with time due to flux creep. In providing noncontact support to the whole weight of the SFESS rotor using a SMB, it should be noted that a decline in the levitation force of the bearing would result in rotor over time. Accordingly, an important consideration in providing long-time noncontact support to the rotor of an SFESS by a SMB is to control and minimize rotor descent. Among promising techniques to control descent are a method using the temperature characteristics of superconductors (a supercooling method), a method using the hysteretic characteristics of superconductors (a pre-load method), and a combination of these two methods. A study was conducted on these rotor descent control methods to examine their respective effects. In this study, a series of tests was carried out on the control methods using a radial superconducting magnetic bearing of the 126mm diameter class, The test results were compared and analyzed to find major characteristics and effects of the three rotor descent control methods as applied to the radial SMB of an SFESS.
A prototype of 250 A-class Y-Ba-Cu-O bulk current leads usable in magnetic fields has been fabricated and studied for transport, mechanical and thermal properties. In this current lead, a melt-processed bulk superconductor 5mm wide and 0.8mm thick was reinforced by a support made of glass fiber-reinforced plastic (GFRP). The current leads could carry 300 A in liquid nitrogen for 10min without a rise in voltage, although the bulk superconductor was under thermal stress due to the difference in thermal expansion between the bulk superconductor and GFRP. In the bending test, the bulk superconductor was not broken although the copper electrode was plastically deformed. Thermal input of the current lead was estimated to be 0.12W under the condition that the high-end temperature was 77.3K and the low-end temperature was 4.2K. Furthermore, using the numerical method, it was possible to visualize distribution of the temperature and the heat flux inside the current lead.
We have tested an axial-type motor that was fabricated using a disk-shaped Bi-2223 bulk rotor. Two-pole and three-phase copper coils were used for concentrated armature windings. Lock and load tests were performed as a function of temperature at a frequency of 60Hz. In order to examine the test results, the spatial harmonics of the air-gap magnetic flux density were measured at room temperature and 77.3K, and then analyzed by Fast Fourier Transformation. Hysteresis curves of the Bi-2223 bulk rotor were also measured in a rotating magnetic field at 77.3K. It was shown that the air-gap magnetic flux density includes large components of 5th and 7th harmonics due to the salient and concentrated armature windings, and these harmonics may be one of the main reasons for asynchronous torque as well as large slip operation. On the other hand, these harmonics are suppressed when the armature current is increased due to the heating with AC loss. Furthermore, it can be deduced from a comparison between the stational torque and the area of hysteresis loops that torque generation varies from hysteresis-dominant to induction-dominant mechanisms at the point the gradient of torque vs. armature current curve changes. These results are reasonably consistent with the temperature dependence of motor characteristics.