Critical currents in REBaCuO superconducting tapes in response to neutron irradiation

The typical MHL observed on all samples before neutron irradiation exhibited maximum close to the self-field and the MHL height continuously decreasing with increasing applied magnetic field, H, (figure 1(a)). Although we did not make self-field correction, for simplicity of units, we present all fields in terms of magnetic induction, B = μ₀H. The irreversibility induction, B_irr = μ₀H_irr, at which the MHL terminates [19], was determined from the semi- logarithmic plot of I_c(μ₀H) dependence, using the precision criterion I_c = 0.01 A. Below this value, the I_c(μ₀H) curve became noisy (figure 1(b), 77 K). At 77 K, all 22 investigated tapes exhibited μ₀H_irr between 8 and 9 T.

Zoom In Zoom Out Reset image size

Classical phenomenological models [20, 21] associate the prevailing vortex pinning interaction with position of the maximum on the F_n(h) plot. F_n is the normalized volume pinning force density, F_n = J_cμ₀H /F_max, and h = H/H_c2. In cuprates, giant relaxation causes that the irreversibility field is significantly lower than B_c2, in contrast to classical metallic superconductors [19]. Irrespective of it, the classical scheme is frequently used for high-T_c superconductors as well, only H being normalized to H_irr. This substantially deforms the original scheme. However, for a comparative study the procedure can be useful. An example of the F_n(h) dependence (h = H/H_irr), representative for most of the tapes studied here, is in figure 1(c). The irreversibility field fell into our field range, 0–9 T, only at T ≥ 77 K. For lower temperatures we normalized magnetic field by the μ₀H_irr values obtained by extrapolating to lower temperatures the μ₀H_irr(T) dependence from the high temperature range. We see that the F_n(h) curve for 77 K has maximum slightly below h = 0.2. This is a value predicted for dilute point-like defects in classical wires [22], with H normalized to H_c2. With decreasing temperature, the F_n(h) curve slightly narrows, but the peak position scales with temperature. We conclude that the operative pinning defect landscape becomes more uniform with decreasing temperature. Such behavior was observed on all tapes by SuNAM, SuperOx, and SCS 4050 of SuperPower. In contrast, the F_n(h) dependences of advanced pinning type samples of SuperPower behaved in a quite different manner. While at 77 K the F_n(h) had still a regular shape with a single peak at around h = 0.2, with temperature decreasing below 70 K, the peak broadened towards lower h values, developing an additional peak or a shoulder, indicating appearance of an additional pinning mechanism (figure 2). A similar behavior was characteristic for all AP type samples of SuperPower.

Zoom In Zoom Out Reset image size

First, the influence of neutron irradiation on the critical temperature, T_c, was studied. Eight representative tapes were submitted to three neutron irradiations, with (total) fluences 1.0 × 10²² m⁻², 2.1 × 10²² m⁻², and 8.03 × 10²² m⁻². We remind the reader that fast neutrons (E ≥ 0.1 MeV) represented about 1/3 of each total fluence, i.e. 2.9 × 10²² m⁻² in the last case. Assuming the irradiation makes an additive effect on the pinning landscape, we combined these irradiations. Altogether, we have samples with four different irradiation loads. The prerequisite to our assumption of additive effect of irradiation was that the pinning produced by the neutron irradiation did not change with time (i.e. an aging effect did not take place). To prove it, we repeated the MHL measurement on one of the irradiated samples 5 months after the first measurement, which was performed 6 months after the irradiation (the radioactivity decay time necessary for a secure manipulation with the sample). Figure 3 shows that the results were practically identical.

Zoom In Zoom Out Reset image size

The neutron irradiation effect on critical temperature, T_c, is presented in figure 4. We see T_c degradation proportional to the irradiation fluence for all tested samples. These results are qualitatively in accord with those from literature [23], the T_c degradation, is, however, smaller, in average only 1.9 K per fast neutrons fluence of 10²² m⁻².

Zoom In Zoom Out Reset image size

First, the effect of neutron irradiation on critical current is demonstrated for a rather low total neutron fluence of 2.1 × 10²² m⁻² (figure 5). The irradiation is sample specific, however, in any of the irradiated samples it did not result in a severe I_c degradation at any of the measurement temperatures. In all samples, I_c was slightly depressed at low magnetic fields. At high magnetic fields, I_c of all SuNAM samples significantly increased in all temperatures (as shown on sample A1, figure 5(a)). In SuperPower AP tapes doped by Zr (figures 5(b), (c)) the current enhancement at high magnetic fields was not so pronounced. It shows that in general the irradiation effect depends on a combined action of neutron fluence, temperature, magnetic field, the initial pinning composition, and relaxation rate discussed further in this paper.

Zoom In Zoom Out Reset image size

Figure 6 shows the further progress of the effect of neutron irradiation on critical currents, for the total fluence of 8.03 × 10²² m⁻² (2.9 × 10²² m⁻² of fast neutrons). The I_c depression at low fields further increased, in all measured samples and temperatures. At higher magnetic fields, the effect differed in dependence of temperature and the original sample structure. At 77 K, I_c decreased in all investigated samples in the whole magnetic field range, the least degradation was observed in the SuNAM samples. At low temperatures and high magnetic fields, I_c values responded to the irradiation in a different manner in different tapes; in all SuNAM tapes and SuperPower AP one, I_c significantly increased, in SuperPower 4050 and SuperOx tapes a moderate increase was observed, while in both SuperPower AP tapes doped by Zr the neutron irradiation caused only a marginal current increase. At intermediate temperatures (50 K) the neutron irradiation caused I_c enhancement in some samples (SuNAM and SCS 4050 AP), nearly no effect in SCS 4050 and SuperOx, and a slight I_c degradation in the samples with Zr. Figure 7(a) presents the evolution of the effect of neutron irradiation on I_c in the SuperPower AP sample, at 10 K. The image resembles that of Eisterer et al [24] for fast neutron fluences up to 2 × 10²² m⁻², obtained in their case by current transport experiment. The results in figure 7(a) look like a continuous I_c dependence on the neutron fluence, decreasing at low magnetic fields and growing at high fields, up to the fast neutrons fluence of 2.9 × 10²² m⁻². It seems to be in contradiction with the crossover in this dependence observed by the Vienna group [23, 24] at around the fast neutron fluence of 2–2.5 × 10²² m⁻². This discrepancy might be only coincidental, as there is a gap in our sequence of neutron fluences, just around the position of the crossover observed in Vienna. Figure 7(a) might combine results for fluences before the break with those after it, making an impression of a continuous decrease/increase. Anyway, I_c(μ₀H) varies not only with neutron irradiation, but quite strongly with temperature. Figure 7(b) demonstrates that for 50 K the I_c(μ₀H) for the highest total neutron fluence lies below that for the total fluence of 2.1 × 10²² m⁻², however still above the initial curve. Thus, the crossover exists in our measurements, too, being clearly seen only at high enough temperatures. It is even better seen on the data for 77 K. Comparing the respective figures 5(a), (b), (c) for the total fluence 2.1 × 10²² m⁻² with figures 6(a), (d), (e) for the total fluence of 8.03 × 10²² m⁻², we see that at 77 K I_c(μ₀H) for the higher neutron fluence is in all three samples significantly below that for the lower fluence. According to the figures, the irradiation effect on I_c is sample specific.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Irrespective of the evident densification of the pinning medium by new defects added by neutron irradiation, we see for the highest neutron fluence an I_c reduction at 77 K in the whole investigated field range. Evidently, thermally activated relaxation plays a significant role here. The higher the temperature is, the higher thermal activation is, in particular for small pinning defects.

Information concerning the defect type might come from the normalized pinning force density plotted as a function of reduced magnetic field. Figures 1(c) and 2(a)–(c) show that the F_n(h) dependence of intact tapes at 77 K exhibits a maximum at around h  =  0.2.

Comparing figures 2(b) and 8(a) for the SCS 4050 AP with 7.5% Zr, we see that after irradiation by the total neutron fluence of 8.03 × 10²² m⁻² the F_n(h) dependence for 77 K did not practically change, even with the significantly lower irreversibility field (figure 6(d)). However, for lower temperatures the F_n(h) curve narrowed, the double-peak shape transformed into the single-peak one, and the peak shifted to h = 0.2. Figures 8(b) and (c) demonstrate that this happened irrespective of the obvious decrease of irreversibility field. All these results show that the neutron irradiation made the operative pinning landscape more uniform, with dominating smaller defects (which enhance the high-field part of the I_c(μ₀H) curve and shift thus the F_n(h) maximum to higher magnetic fields).

Zoom In Zoom Out Reset image size

The thermally activated relaxation is governed by temperature. It gives enough energy to some of the pinned vortices to release the trap. The same thermal energy is obviously much more effective with weak pinning defects than strong ones. In any case, a change in the defect character, distribution, and concentration should be reflected in relaxation rate. We checked the effect of neutron irradiation on pinning landscape and the related change in relaxation rate by measuring the dynamic relaxation rate. Two MHL were measured with magnetic field sweep rates ${{\mu }}_{0}\displaystyle \frac{dH}{{\rm{d}}t}\,\,\,$of 0.72 and 0.36 T min⁻¹ and from them the normalized logarithmic dynamic relaxation rate [25]

$$\begin{eqnarray}&&Q=\displaystyle \frac{d\,\mathrm{ln}\left(m\right)}{d\,\mathrm{ln}\left(\displaystyle \frac{dH}{{\rm{d}}t}\right)}\cong \displaystyle \frac{{\rm{\Delta }}\,\mathrm{ln}\left(m\right)}{{\rm{\Delta }}\,\mathrm{ln}\left(\displaystyle \frac{{\rm{d}}H}{{\rm{d}}t}\right)}\end{eqnarray} \tag{ 2 }$$

was deduced. Here, m is the irreversible magnetic moment on the MHL measured with the magnetic field sweep rate $\displaystyle \frac{dH}{{\rm{d}}t}.$ In order to eliminate spurious reversible components (e.g. paramagnetic one or the thermodynamic reversible magnetic moment, to which m relaxes), we consider m = Δm/2, where Δm is the MHL height. For two different magnetic field sweep rates, we get different MHL heights. As the magnetic field sweep rates are constant over each experiment, the denominator in equation (2) is constant and Q is proportional to the difference of logarithms of the MHL heights. The Q(μ₀H) dependences for the SuperPower AP + 7.5%Zr tape, intact and irradiated by the total neutron fluence 8.03 × 10²² m⁻², are shown in figure 9 for three temperatures. The neutron irradiation resulted in Q enhancement in all three cases, with a rate increasing with temperature; while the Q enhancement at 10 K was only marginal, it was moderate at 50 and huge at 77 K (note that the figure is in the semi-logarithmic scale). As a result, the same growth of defect concentration due to irradiation led to different I_c response to irradiation at different temperatures; at 10 K only a slight I_c increase was observed at high magnetic fields, figure 6, (the change of relaxation was small and its role thus negligible), at intermediate temperatures the increase in defect concentration was masked by the stronger relaxation, and at 77 K thermal activation won, leading to a strong I_c degradation.

Zoom In Zoom Out Reset image size

The increase of I_c at low temperatures and high magnetic fields corresponds to the obvious increase of defects concentration, which has not yet reached a state of saturation or the material damage. It is also supported by the observed change in F_n(h) dependence. Putting these facts together with the increased relaxation rate, we come to the conclusion that pinning landscape densifies in favor of weaker (smaller) defects.

Figure 10 shows the critical current density, J_c, for four temperatures before (solid lines) and after neutron irradiation by the total fluence 8.03 × 10²² m⁻² (dotted lines) for SCS4050AP + 7.5%Zr tape, exhibiting the best electromagnetic properties of all tapes we measured. The superconducting layer thickness was c = 1.3 μm, according to our recent SEM experiment. This plot and Figure 6 show that the studied tapes are appropriate for use in high magnetic fields, especially at temperatures below 30 K. Considering the design neutron fluence for ITER, 10²² m⁻² [23], which is nearly three times lower than the maximum fluence of fast neutrons studied in the present work, it seems that neutron irradiation is not a severe obstacle for use of these tapes in ITER and DEMO HTS magnets, at least from the electromagnetic point of view. Taking into account the rapid progress in the field, even significantly better tapes can be expected soon.

Zoom In Zoom Out Reset image size