Search

Sawh et al. recently reported experimental results that showed, in the case of a partially magnetized bulk superconductor, magnetized using zero-field-cooling, that flux creep resulted in a reduced field measured at the center of the top surface of the bulk. The authors reported that this may suggest magnetic flux vortices, in this case, move against the Lorentz force, contravening commonly-accepted theory. In this paper, we report the results of numerical simulations explaining the observed measurements, and show that the vortices do indeed move with the Lorentz force, but geometric effects from the finite geometry of the bulk and the form of the resulting induced supercurrent flowing within the bulk play a key role in this observed phenomenon. As a result, the relaxation of the magnetic flux can result in a measured magnetic field above the bulk superconductor that could be perceived as magnetic flux moving against the Lorentz force, when applying a simple Bean model (infinite slab) analysis to the problem.

The contribution offered by this paper is to point out the advantage gained by the use of ultra thin trapped field magnets (TFMs) in electromechanical devices. They closely match the torque and force generation capability of their thicker counterparts due to the small reluctance paths they present to the stator side of the magnetic circuit. These conclusions are based on recent experimental progress in the production of these thin TFMs which show they can support critical current densities in the range of 47 kA/cm 2 at 77 K. The results prove that the default electromechanical design objective of maximizing the TFM air gap magnetic field density should be reconsidered. Because of the reduction in the reluctance of the stator field path, substantial torques result even when the volume of superconducting material is vanishingly small. A second advantage is that the stator windings can more easily activate the rotor TFMs.

Experiments are reported on pulsed activation of zero-field-cooled trapped field magnets (TFMs) containing improved broken-columnar pinning centers (PCs). The YBCO TFMs have J C ~ 50 kA/cm 2 and maximum trapped field, BT,max ~ 2.2 T at 77 K. Several results are in disagreement with the well-established Bean model. Essentially full activation is obtained with an applied field at the surface of B A ≈ 1.0 X BT,max. The Bean model predicts B A ≥ 2 X BT,max. Low points in activation are observed at radial sample coordinates r ≈ 0.5 R, a result precluded by the model. Activation shows a discontinuous giant field leap, in disagreement with the smooth increase of B T,max vs. B A predicted by the model. Results are compared to TFMs containing Y211 PCs, which are well described by the Bean model. E.g., for Y211 PCs, field at inner r is the last to rise. For columnar PCs it is the first to rise, an additional conflict with the model. Also, the Bean model predictions are independent of J C , B T , and PCs, whereas major differences exist experimentally. We speculate that with high J C and B T , the very large, centrally-directed Lorentz force causes a discontinuous shift of the fluxoid mesh toward r = 0.

Trapped field superconductors hold promise as a potential game changer in a number of electromechanical applications. Predicting the trapped field is integral to their integration in these applications, and by necessity the critical state model is normally used in that regard. This paper offers three contributions towards that end. First, it offers a method to predict the trapped field based on a novel way to break down a mutual coupling matrix one cell at a time. Second, it discusses another predictive scheme based on equivalent column currents in the cell while taking into account the side of the YBCO crystal where seeding occurred. There is an experimental procedure that allows one to predict the equivalent thickness of the YBCO puck. Third, it offers some guidelines for when the simpler critical state Bean model can be employed with some confidence over the more complete Kim model. All the analyses are compared to experimental results on bulk YBCO pre-cooled to 77 °K, 20 mm in diameter with an axial length of 8 mm.

The field and current induced in a trapped field magnet during a transient current excitation is complicated by the fact that the conductivity of the material changes wildly depending on both the magnitude of the induced current and the local B field. A mutual inductance approach is presented for solving this coupled problem; it involves discretizing the bulk superconductor into cells and precomputing the inductance coupling matrix of all cells with one another and with the excitation coils. The benefits of this approach are that the numerical simulation is rapid and flexible. The latter is important because the conductivity of the individual cells is dependent on the current density, the magnetic field density magnitude, and the temperature. The proposed approach allows the modeling of a very sharp J-E relationship which has proved problematic using conventional finite element approaches. In addition, a phenomenological model relating the critical current density to the field is introduced for these tests in place of the Kim model. The field simulation is tested with a challenging dwell time experiment in which the trapped field above a bulk superconductor is predicted for various excitation current hold times.

Currents are set up in high temperature superconducting (HTS) material when it is cooled in a magnetic field (called field-cooling: FC), or exposed to a magnetic field after cooling (called zero-field-cooling: ZFC). The HTS material can be treated to make these currents persistent, thereby forming “trapped field magnets” (TFMs). TFMs have been improved over the past two decades by the efforts of a large number of worldwide research groups. However, applications using TFMs have lagged, in part due to the problem of high fields needed for activation. We describe herein experiments to characterize the behavior of TFM activation using repeated applications of permanent ferromagnets (called “pumping”). Significant partial activation is obtained using a non-uniform pumping field which is higher in the center of the HTS than at the periphery. ZFC, followed by pumping with such a field, results in trapping the full applied field, instead of half the applied field, as is trapped by zero-field-cooling followed by application of a uniform field. We also find that for FC partial activation, and subsequent additional activation by pumping, the resulting fields are additive. Additionally, we conclude that one contributing, positive component of activation by fluxoid pumping is the phenomenon of creep.

High temperature superconducting trapped field magnets (TFM) offer great potential as an alternative to second generation YBCO wire, both in cost and performance. The difficulty of activating this material is the primary problem hindering their incorporation into electromechanical devices. An alternate method called flux pumping is discussed in which the magnetic field trapped within a precooled TFM is increased, i.e., pumped up incrementally. A boundary element formulation is employed to illustrate what happens to the current density within the TFM during this pumping procedure. This is followed by a rigorous analytical formulation to simulate flux pumping within a 2-cm diameter TFM; the simulation is compared to an experiment using a NdFeB magnet as the pumping source.

The properties of discontinuous aligned pinning centers (PCs) created by high-energy heavy-ions are compared for bulk melt-textured and coated conductor HTS. Properties of PCs, which increase J c (pinning potential and entanglement), and negative properties which decrease J c (e.g., decreased T c and percolation paths) are evaluated. Mechanisms are proposed to explain the very large increases in J c resulting from multiple-in-line-damage (MILD) compared to continuous columnar pinning centers (CCPC). In particular, a mechanism which results in fluxoid entanglement, even for parallel (unsplayed) PCs, is discussed. The same mechanism is found to also account for restoration of much of the pinning potential expected to be lost due to the gaps in MILD PCs. It also accounts for the fact that at high fluence, J c increases as fluence is increased, instead of decreasing as expected. The very low self-field in coated conductor permits separation of the negative and positive effects of PCs. It is found that parameters developed to quantify the negative effects in bulk melt-textured YBCO, by 63 GeV U 238 ions, successfully describe damage to 2.1 μm thick coated conductor by 1 GeV Ru 44 ions. Coated conductor at 77 K and self-field is generally known to have J c about 100 times that of melt-textured YBCO. However, at 77 K and applied field of 1 T, when both forms of HTS are processed with comparable numbers of near-optimum MILD PCs, the difference in J c is reduced to a factor of 1.3–2. Whereas J c for melt-textured YBCO increased sharply, by a factor of up to 16.8 for high-fluence MILD PCs, J c in coated conductor increased by a smaller factor of 2.5–3.0. Nevertheless, 2.1 μm thick coated conductor, with near-optimum MILD PCs, exhibits J c = 543 kA/cm 2 at 77 K and applied field of 1.0 T, and I c = 114 A/cm-width of conductor. This is the highest value we find in the literature. The phenomenology developed indicates that for optimum MILD PCs in coated conductor, J c ∼ 700 ± 70 kA/cm 2 should be achievable at 77 K, 1.0 T.

Trapped field magnets (TFMs) are characterized by the maximum field they can retain when they are activated in air (~vacuum). In some applications however, TFMs are partially embedded in a ferromagnetic medium. We consider, experimentally and theoretically, some simple geometries involving contact between TFMs and ferromagnetic media. We find: an increase in maximum TFM field exceeding a factor of 2.53; a change in field profile from the well-known conical TFM field to a uniform field in the ferromagnetic material; and fields produced by present day TFMs capable of saturating Hyperco-50 alloy. The observed flux creep rate of the TFM field was found to be the same in air and in contact with ferromagnetic material.

Large grain YBCO superconductors, containing the same atomic percentage of Ce, and that of Pt, were produced via the melt-texturing process. Samples doped with nanometer cerium oxide exhibits an approximate 30% increase in trapped magnetic flux density, relative to submicrometer platinum doped YBCO. A 30% increase in trapped magnetic flux density represents ~ 80% increase in critical current density. Microstructure studies indicate that just like Pt, Ce refines Y211 particles to increase trapped magnetic flux density and critical current density. Cerium is much cheaper than platinum and is therefore, a desirable alternative to Pt in the large-scale production of bulk YBCO superconductors. Another interesting result of this experiment is that the trapped magnetic flux density of Ce-doped samples is unchanging over a large range of melting temperatures used for melt-texturing. In addition to refined Y211, Ce-rich deposits are also formed. The Ce-rich particles are insufficient in number to act as effective pinning centers.

High-temperature superconducting trapped field magnets (TFMs) offer great potential as an alternative to second-generation YBCO wire, both in cost and performance. Attention is given to the calculation of current distribution within YBCO disks at partial and full activation and comparing this to experimental values. The best results are obtained by treating the current as a sequence of nested current rings. The fields are computed by integrating the elliptic integrals representing the fields from these rings and using variable metric optimization to choose the ring radii to best match the activation field over the unactivated material. A technique for treating the subregions of the TFM as voltage fed coils appears most expeditious for computing forces.

Flux pinning properties and irreversibility fields Birr(T) of melt-textured YBCO with discontinuous or multiple-in-line-damage (MILD) columnar defects produced by irradiation with high-energy U238 ions were studied at a constant matching field of B = 10?T and for several energy losses between Se = 1.67 and 2.4?keV???1. With increasing Se and increasing length of the MILD pins, the critical current density jc(H,T) strongly increases and Birr(T) for fields along the c axis progressively shifts upwards reaching 9?T at 77?K. For Se = 2.4?keV???1, a pronounced kink is observed in Birr(T) at 8?T which is a strong indication of Bose-glass behaviour. The jc(H) dependence of this sample shows a peak at a low applied field Bp. This peak effect is explained by the entanglement of vortices. It is argued that for MILD pins single vortices interact simultaneously with many short columnar defects in neighbouring ion trails resulting in an automatic splay in the vortex orientation. The observed decrease of Bp with increasing temperature is estimated taking the increasing thermal fluctuations into account.

We discuss an experiment on large-grain YBCO, the intent of which is to compare the effectiveness of continuous columnar pinning centres (CCPC) to that of discontinuous aligned pinning centres, in their ability to increase Jc. We find that discontinuous pinning is far superior. The motivation and design of the experiment are reviewed. High-energy heavy ions were utilized to produce pinning damage of controlled morphology. The resulting increase in Jc is presented as a function of the variable Se, which is closely related to the discontinuity and diameter of the damage. The dependence of Jc on magnetic field, temperature, and Fi (the number of ion tracks per cm2) is presented and discussed. The dependence of Hirr on Se and Fi, and the temperature and fluence dependence of a Jc fishtail effect, are presented. The experimental results indicate that multiple-in-line-damage (MILD), despite its discontinuities, is far more effective in increasing Jc than CCPCs are.

In type II superconductor, pinning centers are essential to obtain high current density, Jc. Conventional wisdom holds that highest Jc is obtained by using continuous columnar pinning centers, which in turn maximize pinning potential. We find this conclusion to be in severe error. Using heavy high-energy ions to insert pinning centers into the YBCO high temperature superconductor, we find peak Jc where pinning centers have 66% discontinuity. Furthermore, this provides record Jc, and peak irreversibility field, Hirr. The values of Jc achieved using discontinuous pinning centers is 5 to 60 times that achieved using continuous columnar pinning centers. We review a phenomenological theory, in its early stages, which matches this experimental result. It indicates that Jc is highest not where pinning potential is largest, but where the combined effects of percolation paths, critical temperature, and pinning potential, are optimized.

A conundrum is observed where one type of small, non-superconducting particles fails to provide pinning, in textured YBCO superconductor, while other non-superconducting particles of comparable size and number density, pin very well. Pinning centers in high temperature superconductors (HTS) have been shown to greatly enhance critical current density (JC)- An experiment was performed in which large grain YBCO doped with uranium (U), but no platinum (Pt), was produced. Profuse deposits of a U-Y-Ba-O compound were formed during texturing. These U-rich deposits have an almost spherical morphology with an average diameter of ~300 nm and are uniformly distributed inside the YBCO. Conventional wisdom dictates that the U-Y-Ba-O sub-micron deposits should act as pinning centers and increase JC .However, in tests of JC via trapped magnetic flux density at 77 K, no increase was measured. This result is especially surprising because chemically similar and morphologically indistinguishable deposits of U-Pt-Y-Ba-O, found in (U + Pt)-doped YBCO, increased JC by a factor exceeding 2. The deposits compared had the same size and number density. Neither deposits significantly affected critical temperature (TC). One difference between these two compounds is that the non-pinning U-Y-Ba-O deposits have a single perovskite structure, while the U-Pt-Y-Ba-O pinning centers are a double perovskite. Data are presented on the remarkably similar morphology and strikingly different ability to trap magnetic flux density of the U-Y-Ba-O and U-Pt-Y-Ba-O deposits.

We report on the effect of randomly-distributed amorphous defects, ~ at the length scale of the superconducting coherence length, on Tc in YBCO. These damages were created by ions irradiation with a wide range of energy and fluence. Typically, in previous experiments, Tc was found to vary with ion fluence, species, energy, and ionization charge, and also on the chemical composition and density of the HTS. Neither a phenomenological nor a theoretical relationship between Tc and these ion properties is however known of. We now show that the decrease in Tc is a universal function of the fractional defect volume embedded in the HTS. We also re-analysed several experiments by others, and find that the same universality applies. This finding provides, for the first time, a means to calculate the effect on Tc of disorder as created by randomly-distributed amorphous nanosize defects.

This paper introduces a new electromechanical motor that incorporates free-spinning magnetized cylinders in the air gap. The motor has some of the features of a magnetic planetary gear in that it amplifies torque but it does so to a significantly higher degree than a conventional planetary gear with passive ferromagnetic elements in the air gap (79% higher). The cylinder motor offers the additional advantage that it eliminates cogging torque. The resulting machine constitutes a new class electromechanical device with the introduction of a freely rotating source of magnetomotive force (MMF) in the air gap that both couples and boosts the torque interaction of the rotor and the stator.

An experimental study of defects caused by ion irradiation has proven that very discontinuous multiple-in-line-damage results in much higher J c than continuous columnar pinning centers. This is contrary to conventional wisdom. We develop a first approximation to a theory relating J c to the morphology of pinning centers. Qualitative agreement with experiment is achieved. Conclusions indicate the relative importance of pinning potential, percolation, and critical temperature for various morphologies. Results suggest the leading importance of pinning, compared to texture, weak links and oxygenation in achievement of the goal of high J c especially in large grain HTS.

Scores of work have suggested that the pinning potential energy, Up, should be maximized to achieve high critical current density, J c . This has led several researchers to believe that the highest J c could only be achieved by means of continuous columnar defects, CCPCs, because they provide the largest Up. We now present the surprising experimental result that, in clear contrast with the conventional belief, J c for discontinuous pinning is much higher than for continuous. For example, in melt-textured YBCO we obtained a record-high J c ∼ 275 kA/cm 2 , at B = 1 T and T = 77 K, using discontinuous multiple-in-line-damage, MILD. Our data contrary to conventional pinning theories indicate that J c does not depend primarily on U p , but on the simultaneous optimization of three factors: the loss in current percolation, the reduction in T c , and Up. In conclusion the maximum J c in YBCO, at 10 K < T < 88 K and up to 8 T, is obtained for MILD pinning centers with a fractional length along the sample thickness of only 0.2-0.3, whereas CCPCs have fractional length of 1.

We review a recent experiment in which pinning centres, composed of discontinuous multiple-in-line damage (MILD), were compared to continuous columnar pinning centres (CCPCs). The methods used, results of the experiment, and the initial steps taken toward a phenomenological theory explaining the unusual results, are discussed. Experimental results show that MILD pinning centres, despite their reduced pinning potential, are far superior to CCPCs. Using MILD pinning, a new record Jc has been achieved in large grain YBCO. The value of Hirr (MILD) is comparable to, or larger, than Hirr (CCPCs). Pinning effects alone, with no modification of texture, weak links or oxygenation of the YBCO, increase Jc by as much as a factor of 17. Splay is expected to further increase this factor. The experimental dependence of Jc on the energy lost by the ion per unit length is approximately reproduced by a phenomenological theory, including only the effects of diminished percolation, critical temperature, and pinning potential.

Columnar pinning centres provide the largest pinning potential, Upin ,b utnot the greatest Jc or pinnable field, Bpin .C haracteristics of ion-generated columnar defects which limit Jc and Bpin are discussed, including reduction of the percolation path, and the need for a larger number of columns of damage, for pinning, than are usually estimated. It is concluded that columnar pinning centres limit Bpin to less than 4 T, and also severely reduce Jc .T hegoal of maximizing Upin ,v ia columnar centres, appears to have obscured a more rewarding approach and resulted in neglect of a large regime of ion interactions. Evidence is reviewed that multiple in-line damage (MILD), described herein, can provide orders of magnitude higher Jc and Bpin ,d espite providing lower Upin .T he MILDpinning centre morphology is discussed, and it is estimated that for present-day large grain high Tc superconductors, a Jc value of ∼10 6 Ac m −2 is obtainable at 77 K, even when crystal plane alignment and weak links are not improved. In addition, the pinned field is increased by over an order of magnitude. An experiment is proposed to confirm these calculations, directly compare MILD pinning to continuous columnar pinning, and determine the optimum MILD structure. Applications of MILD pinning are discussed.

An experiment was performed to test the effect of different uranium compounds on the properties of chemical pinning centres. UO2, UO3, and UO4·2H2O wereadmixed to Y 123+Pt, and textured. Tests of Jc via measurements of trapped field (Btrap) indicate a clear dependence of Btrap on the U compound admixed to create the pinning centres. In all three cases there is a monotonic increase in Btrap as the mass (MU) of U is increased. However, the magnitude of the increase in Btrap depends on the admixed U compound. The highest increase in Btrap is measured in samples doped with UO4·2H2O, and the lowest is obtained in samples doped with UO2. Microstructure studies indicate that the composition of the U-rich pinning deposits is the same in all three cases, i.e. all are the previously identified (U0.6Pt0.4)Y Ba2O6 compound. The primary difference among the three types of samples is that the size of the U–Pt–Y–Ba–O pinning deposits depends on the admixed U compound. While all are in the nanometre domain, the diameter of these deposits was markedly larger in UO2 doped samples than in UO3 doped samples, and smallest in UO4·2H2O doped samples. Because some form of poisoning limits the amount of U that can be added to create pinning centres, to wt%, smaller deposits result in a greater number of pinning centres. We conclude that UO4·2H2O is more effective than either UO3 or UO2 in the formation of U–Pt–Y–Ba–O pinning centres because of diminished pinning centre size, and consequent increase in pinning centre density.

In the U/n process enriched uranium is admixed to HTS powders, textured and then irradiated with thermal neutrons. Neutrons must penetrate several centimeters for such processing to be successful in textured materials. Bulk Sm123 cannot be U/n processed because thermal neutrons penetrate

When ion damage is used to create pinning centers, full columnar pinning centers provide the largest pinning potential, Upin, but not the greatest Jc or pinned field, Bpin. Some of the characteristics of columnar defects which limit Jc and Bpin are discussed, including reduction of percolation path, and the need for a larger number of columns of damage, for pinning, than are usually estimated. It is concluded that columnar pinning centers are limited to Bpin

Very small deposits are formed when two foreign elements, A and B, are added to textured (RE)BCO. These deposits increase the Jc of samples and hence the maximum trapped field. Deposit sizes are generally in the range 200–400 nm. Their chemical composition is (Ax,By)(RE)Ba2O6, where x+y=1 and x=y±0.1. Their structure is double perovskite. The deposits produce two types of pinning, one by refinement (e.g., of the Y211 phase) and the other due to the double perovskite deposits themselves. In those cases tested, the refinement increases Jc by factors exceeding 1.33, and the deposits increase Jc by factors exceeding 2, for a total increase by a factor exceeding 2.66. Element A can be U, W or Mo. Element B can be Pt or Zr. We speculate on possible additional elements in classes A and B.

Y123+30 mol% Y211 powders were doped with tungsten and platinum, and textured. Microstructure studies show the presence of profuse spherical deposits, 200–300 nm in diameter. These deposits were identified as (W 0.5 Pt 0.5 )YBa 2 O 6 , a double perovskite. The size of the W-rich deposits is independent of the W doping level. There is no substitution of W into the Y123 matrix. For Pt doping >0.5 wt.%, trapped field is observed to increase monotonically up to 40% for W doping of up to 0.48 wt.%. We conservatively estimate that this corresponds to a 60% increase in J c at constant field. Thus (W 0.5 Pt 0.5 )YBa 2 O 6 double perovskites deposits act as pinning centers. Similar studies of molybdenum doping indicate deposits 200–300 nm, of (Mo 0.5 Pt 0.5 )YBa 2 O 6 , also a double perovskite. The (W 0.5 Pt 0.5 )YBa 2 O 6 and (Mo 0.5 Pt 0.5 )YBa 2 O 6 deposits are remarkably similar to the (U 0.6 Pt 0.4 )YBa 2 O 6 deposits found earlier in U-doped Y123. Therefore, W and Mo are suitable non-radioactive substitutes for U.

The U/n process has been used to introduce quasi-columnar pinning centers into Nd123, textured in 1% O 2 , 99% Ar. The process involves adding 235 U to precursor powders of the superconductor, texturing, and irradiating with thermal neutrons. The two resulting ions, from each nuclear fission, cause aligned discontinuous electronic damage of length ∼5.4 μm, which acts as a pinning center. Best results for J c require U deposits spaced by s ⩽5.4 μm. This has been achieved by deposits containing two foreign elements, U and Zr, and three native elements. The deposits of UZrNdBaO and Nd422 refined by the Zr, act as additional non-aligned pinning centers. The U/n process increases J c by a factor R J ⩾10, to values of J c ∼274 kA/cm 2 at T =77 K, B =666 G.

The extended defects produced by the fission of neutron-irradiated U-235 inclusions in Ag/Bi2223 superconducting tapes act as very effective pinning centres. Substantial improvements in the magnetic field dependence of the critical current densities, J(c), and in the irreversibility line of treated tapes have previously been observed. The magnetoresistance of Ag/Bi2223 tapes treated with the doping and irradiation technique has been measured, and activation energies calculated from the Arrhenius plots. The findings corroborate the previous J(c), results, confirming the strong pinning action of the extended columnar defects, particularly in fields aligned along the c-axis of the tapes.

Melt textured YBCO was irradiated by 6.88 GeV Kr ions to determine the threshold for increase of J c by formation of ionization pinning centers. Trapped field was measured at front and back surfaces of 1 mm thick samples as a function of fluence and sample thickness. Trapped fields are lower at the front surface and higher at the back, consistent with ionization pinning centers in the last ∼200 μm of ion track. The discontinuous, aligned ionization damage has a threshold of ∼0.8 keV/A to increase J c from ∼20 to ∼90 kA/cm 2 at a fluence of 1.93×10 12 Kr ions/cm 2 .

Uranium doped BiSCCO 2223 tapes were irradiated by thermal neutrons. The resulting fission-induced defects improve flux pinning and shift the irreversibility line to higher fields. Significant enhancements of the transport critical current density as well as a reduction of the J/sub c/-anisotropy are found for the irradiated samples. Furthermore, inter- and intragranular critical current densities were determined from the remanent magnetic moments by SQUID magnetometry.

We report the effect of permanent magnet (PM) collars on the flux-creep rate of magnetized bulk HTS. The creep rates of single-grain, cylindrical samples are measured with attached collars activated to various fields, B PM, in the range 0??? B PM??? B PM,max, where B PM,max is the fully saturated field of the PM. As B PM varies, the creep rate of the HTS is found to maintain its well-known form?a constant fractional loss ?, of original residual field, per decade of time. However, the magnitude of ? decreases as B PM increases. The decrease in ? is found to be linearly dependent on increasing B PM. The collar field for which flux-creep extrapolates to zero is found to be comparable to the maximum trappable field of the HTS bulk, B T,max. The properties of the dependence of ? on the HTS peak field, B T,max, the PM field, B PM, and the creep rate ? 0 with B PM?=?0 permit the reduced creep rate in these experiments to be predicted by a universal equation.

BiSCCO-2223 tapes were produced by the powder-in-tube method; however, prior to processing, small amounts of uranium oxide were added to the precursor powders. Samples containing various amounts of 235 U are investigated and compared with undoped samples. Irradiation by thermal neutrons causes fission of uranium into two fission products, which create fission tracks in the sample, whereas irradiation by fast neutrons produces spherical cascade defects. Both types of defects are randomly distributed in the samples. The superconducting properties are determined by dc transport measurements in fields up to 6 T at temperatures between 40 K and 77 K. Very significant enhancements of the critical current densities and the irreversibility lines as well as a strong reduction of the J c anistropy are found.

Insertion of uranium into HTS precursor powders prior to texturing, and irradiation with thermal neutrons after texturing is called U/n processing. Damage caused by uranium fission fragments has been extensively studied for pinning center behavior in Y123, and Ag-BiSCCO tape. In both HTS systems results for J c , anisotropy, etc. are excellent. We consider the chemical and physical conditions under which U/n processing is successful, including spacing, S, and size of deposits of U compounds, and neutron absorption by the HTS. If δ=length of the quasi-columnar fission fragment defects, we find S

Levitation and stability forces have been studied using high-temperature superconductor (HTS) trapped field magnets (TFMs) to replace the permanent magnets (PM) usually used. TFMs of YBa2Cu3Oy with maximum trapped field, BTFM, up to 1.6 T were used and compared to a PM, with BPM = 0.34 T at its surface. The TFM or PM was opposed by a HTS cooled under various field cooling and zero field cooling conditions. It is found that the levitation force with the TFM can be increased by a factor of 10 or more. For example, for the TFM opposed by a saturated HTS with a maximum trapped field of 1.6 T, both 2 cm in diameter, the levitation force and pressure are FL = 393 N; pL = (125 N cm-2). The increase is limited only by BTFM,max of the TFM used. Restoring forces do not increase quite as rapidly as levitation forces when PMs are replaced with TFMs. However, we find that both levitation and stability forces can be significantly increased using multi-sample systems.

Results on the enhancement of critical current densities in melt-textured bulk (Y-123) superconductors, which were doped with and exposed to thermal neutron irradiation to create fission tracks, are presented. These tracks, which all originate from U-containing clusters of a stable UPtYBaO compound, form isotropic `starlike' defects with diameters of approximately 10 nm and lengths of . However, if we consider projections onto the symmetry plane of the flux line lattice and allow for a certain tilt modulus , their pinning-effective size is mostly (90%) between 10 and 20 nm, i.e. only twice as large as the characteristic defect structure (amorphous collision cascades with diameters of nm) introduced into undoped superconductors by fast neutron irradiation. We report on studies of and the irreversibility lines in superconductors containing various amounts (0.3-0.7 wt%) of uranium and various amounts of the fissionable isotope . In the best case (0.3 wt% U, thermal neutrons), we find (77 K) to be around (0 T) and (5 T). Control experiments on samples containing (nominally) only and exposure to thermal or to thermal plus fast neutrons confirm the above flux pinning considerations. As reported before for undoped melt-textured Y-123, the irreversibility lines remain practically unchanged due to the presence of strongly pinning Y-211 precipitates.

When uranium (U) is added to Y123 powders, after texturing the deposits containing U are found to be nm. These deposits act as pinning centres, and increase by 90% per 0.8 wt% U. Neutron irradiation further increases . Total increases of by factors of e.g. (at 77 K, 0.25 Tesla), and (at 50 K, 10 Tesla) result. Magnitudes of are (at 77 K, 0.25 T) and (at 50 K, 0.25 T). The U/n process is not yet optimized. Chemistry, microstructure, radiation effects and cost are discussed.

A study has been made of compounds formed when up to 1% uranium oxide is added to YBa 2 Cu 3 O 7− δ (Y123), prior to texturing. Initially, the study was to create pinning centers by addition of UO 4 , followed by neutron fission of some of the 235 U component. It was found that, prior to fission, the chemical reaction of U provides sufficient pinning centers to approximately double J c . The U forms in deposits about 300 nm in diameter. The number of such deposits is proportional to the mass of U. TEM and XRD studies identify the compound as (U 0.6 Pt 0.4 )YBa 2 O 6 , a double perovskite. Neither T c nor creep vary with %U. The uranium compound has also been produced outside of Y123. The behavior of uranium in high temperature superconductor (HTS) is thus of interest not only for introducing pinning centers by neutron fission of U but also as an effective chemical method to increase J c .

Quasi-permanent magnets made of melt-textured YBa2Cu3O7 - delta (MT-YBCO) superconductor can now trap multi-tesla fields, B sub t. The interaction of the trapped field and the critical current causes an outward pressure, proportional to B sub t (2), which can crack the magnet. The authors have done an experiment to observe such cracking in a mini-magnet fabricated from four MT-YBCO discs activated at 49 K using an applied field of 14 T. They have compared the results to existing theories which describe magnetic pressure in a trapped-field magnet (TFM) previously activated. They find that a modification is needed to describe magnetic pressure during the process of activation. They present the experimental results and the expanded theory, based on the simple Bean model. Theory and experiment show good agreement. The authors find that cracking is more likely during activation, and conclude that 10 T is achievable in TFM`s composed of present materials. Cracking is most probable at the center of a TFM, with the cracks running radially outward.

An experimental axial gap generator, with a 6 inch diameter rotor and quasi-permanent HTS magnets, was tested. A variable speed motor was used to rotate the rotor above a stator composed of copper wire wound in an 8 pole 3 phase configuration. The rotor and stator were both run in liquid nitrogen. A 3 phase variable resistor bank was the generator load. The HTS magnets were YBCO, each 2 cm in diameter and 1 cm thick, melt textured using a SmBCO seed, containing 60% excess Y, and about 1%Pt. They were not irradiated, and could trap maximum fields of 3500-4000 G. First, the HTS magnets were field cooled in a field of approximately 2200 G, provided by the stator while carrying 25 Amperes. The HTS magnets trapped about 1800 G. In this configuration, the generator was run for about one hour at speeds up to 2000 RPM, and developed a maximum power output of 33 Watts. Next the HTS magnets were warmed, and then zero field cooled. In this case, activation was accomplished by a /spl sim/15 ms pulse on the stator, providing 5000 G of pulsed field. The HTS magnets trapped about 2600 G. The generator was then run for about 1 hour at speeds up to 2265 RPM, and developed a maximum power output of just under 100 Watts. >

Irradiation may significantly increase I/sub c/ in HTS. A systematic pattern occurs: R=I/sub c/(afterirr.)/I/sub c/(beforeirr.) increases at low defect density, d. It reaches a peak, and then it falls below 1 at high d. The pinning center mechanism, which causes R to increase, has been extensively studied. The falloff in R has not. It has been considered a secondary effect. Here, we will show that the fall-off plays an important role in determining the maximum I/sub c/ enhancement achievable. A phenomenological model to describe the R-vs.-d curve, over the entire d range, is proposed. The idea is that R is the product of two competing effects. (i) Irradiation damage acts as pinning centers, hence increases critical current density, J/sub c/. (ii) Damage reduces the flow-area. Hence, it decreases the net critical current. Data on U/n processed Bi-2223 tapes are fitted to this model. The fitting indicates: (1) the reduction of the flow-area accounts for the majority of the R falloff; and (2) It is sufficient to describe J/sub c/ enhancement as linear with d, and it depending on field and temperature only through the ratio b=B/B/sub irr/, where B/sub irr/ is the irreversible field before irradiation.

An accurate noncontact method of characterizing Ag-BiSCCO tape is reported, using the trapped field, B/sub t/, of a field-cooled tape. The dimensions of the tape make it possible to calculate in closed form a very simple expression, relating trapped field to transport current. Short lengths of tape, 4 cm, are used to experimentally confirm the model. The model applies to any length of tape for which length is greater than 5 times the tape width. A schematic of a proposed test apparatus is presented which should permit on-line testing for magnitude of the current, and for cracks, while the tape is in motion during production.

For type-II superconductors, magnetic field can be trapped due to persistent internal supercurrent. Quasi-persistent magnetic fields near 2 T at 60 K (and 1.4 T at 77 K) have been measured in minimagnets made of proton-irradiated melt-textured Y-Ba-Cu-O (MT-Y123) samples. Using the trapping effect, high-field permanent magnets with dipole, quadrupole, or more complicated configurations can be made of existing MT-Y123 material, thus bypassing the need for high-temperature superconductor (HTS) wires. A phenomenological current model has been developed to account for the trapped field intensity and profile in HTS samples. This model is also a guide to select directions of materials development to further improve field trapping properties. General properties such as magnetic field intensities, spatial distributions, stabilities, and temperature dependence of trapped field are discussed. >

Both high levitation pressures (up to 22 N cm-2) and high restoring pressures (up to 11 N cm-2) are obtained for a superconducting trapped field magnet of 1.5 T levitating above the centre of a ring of zero-field cooled high-temperature superconductors. The ratio of levitation force to restoring force can be varied between 2.9 and 0.3 by changing the quality of the superconductors. This significantly improves the stability of levitation compared to commonly used single sample configurations.

Critical currents J c ( T , B ) for virgin and four 235 U doped Bi2223/Ag tapes irradiated with thermal neutrons (U/n treated) were measured in the temperature range T ⩾50 K and magnetic field B ⩽1 T. Below 97 K vortex pinning in U/n treated tapes increases with matching field B φ , as evidenced by a weaker B dependence of J c on increasing B φ . Hence, the field B max ( F p max =J c B max ) and the irreversibility field B irr (F p → 0) of U/n treated tapes are enhanced compared to those for virgin tape. Since the rate of the enhancement of B max and B irr with B φ depends on T , one can tune the maximum enhancement at desired T .

A production run of 60 melt-textured YBCO trapped field magnets (TFMs), 2 cm in diameter, is reported. The TFMs have pinning centers of Y211 and (Pt0.4U0.6)YBa3O6 deposits and also damage tracks of ions from uranium fission. The resulting average trapped field at the center of the seed-side surface is 〈BT(r = 0)〉 = 2.04 T at 77 K. The rms spread in BT is 5.9%. TFM samples rejected due to problems with the SmBCO seeds totaled ~10%, and an additional 2% were lost due to cracking under magnetic pressure. Methods to significantly reduce these losses are discussed. The ratio of BT on the TFM bottom to that on the seed side is ~67%. Evidence of a method to raise this ratio to 1.0 is presented. The magnitude and time dependence of the radioactivity resulting from the uranium fission are reported.

Experimental studies are reported for pulsed activation of trapped field magnets (TFMs), using a sequence of N equal pulses. These experiments are done on low field TFMs for which temperature increase is not yet a dominant limitation. For low and moderate amplitudes of activation pulses heights, BA, the trapped field at any distance r from the TFM centre increases as BT(r,N) ∝ klogN, where k is not a function of N. Thus the increment in trapped field is ∝1/N. Since logN increases without limit as N → ∞, while BT is limited by finite JC, limits are expected on the logN increase. The limiting behaviour is observed at higher values of BT(r,N), and is very well fitted by the multiplicative factor [1 − BT(r,N − 1)/B*(r)]. The resulting phenomenological equation fits the data at essentially all values of N, r, and BA. In this equation, B*(r) plays the role of the limiting value of BT(r) as N → ∞. A study of B*(r) leads to its identification as the trapped field reached in a zero-field-cool activation using a constant field of the same magnitude as the pulsed field.

We report experimental results of a study of pulsed field activation of YBCO trapped field magnets (TFMs), at 77 K, in which the applied field covers an area smaller than the TFM. Data on a subset of experiments, in which pulse amplitude is varied, show the effects on trapped field distribution and creep. At low pulse height, trapped field is observed at the sample periphery, even at applied field well below BC1. At increased fields, the measured J is not limited to JC or zero, as in the Bean model, but varies continuously from 0 to JC, possibly as a result of averaging over the spatially varying applied field. For partial activation, creep is a function of r. As the critical state is approached, the spread in creep rates collapses to a single value. Full activation is obtained by a single pulse of applied field approximately 3.3 times maximum trapped field on the TFM surface.

The magnetoresistance R(T, B) of virgin and four U-235 doped Bi2223/Ag tapes irradiated with thermal neutrons (U/n treated) has been measured in the temperature range T greater than or equal to 50 K and magnetic field B less than or equal to 1 T. For B greater than or equal to 20 mT the pinning potentials U-0 of U/n treated tapes were enhanced compared to that of virgin tape. The relative enhancement U-0(B, U/n)/U-0(8, virg.) increased with matching field B-phi and persisted to higher B at higher B-phi. For U/n treated samples, U-0 showed universal variation with B/B-phi : U-0(B less than or similar to B-phi) proportional to (B-phi/B)(proportional to) with alpha approximate to 0.5 and faster variation at higher B. The resistive irreversibility fields B-irr(T) of all U/n treated samples for T less than or similar to 100 K were also enhanced: B-irr (T, U/n)/B-irr (T, virg.) increased with B-phi and the enhancement persisted to lower T for higher B-phi. The maximum enhancements of B-irr scale well with the corresponding enhancements of U-0(B).

We comment on the paper by N Hari Babu et al (2002 Supercond. Sci. Technol. 15 104–10) and point out misinterpretations of the chemical composition of U-bearing deposits observed in Y123. The observed small deposits are those of new compounds which do not contain Cu, rather than refined Y211 plus U, as stated by the authors. We further note that extensive literature, not quoted, is in disagreement by nearly an order of magnitude concerning the values of Pt and U doping at which the optimum value of Jc is obtained. Other related information, presently in the literature, which may be helpful to those working with this high temperature superconducting chemical system, is presented.