In-situ critical current measurements of REBCO coated conductors during gamma irradiation

The high critical current densities, fields and temperatures of rare-earth-barium-copper-oxide (REBCO) superconductors make them an attractive material choice for the primary magnets of next-generation fusion reactors and pilot power plants [1–3]. Fusion reactors however, present a novel and challenging environment for superconducting magnet operation, due in part to the high neutron fluxes and consequent gamma radiation to which the magnets will be subjected. (n, γ) interactions within the material layers between the plasma and the magnets (first wall, blanket, neutron shielding etc.) and within the magnets themselves lead to the production of a broad-spectrum photon flux, including gamma rays with energies exceeding 10 MeV (figure 1). These gamma rays primarily interact with REBCO through photoelectric absorption (E $_\gamma$ $\lt$ 0.3 MeV) and incoherent scattering (0.3 MeV $\lt$ E $_\gamma$ $\lt$ 10 MeV) (figure 2). Both processes generate scattered electrons with energies up to the incident gamma ray energy. These scattered electrons impart their energy to the superconductor generating a transient 'hot spot' as they travel through and interact with the ion lattice destroying superconductivity locally [6, 7]. Energetic electrons can also collide with and displace atoms out of their lattice locations if the energy imparted to the atom exceeds its threshold displacement energy [8]. Stable defects in the lattice can reduce Cooper-pair density and affect the material superconducting properties [9].

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Reproducing the broad fusion gamma spectrum artificially is prohibitively difficult outside of a fusion reactor. For the work detailed here, we have used the Dalton Cumbrian Facility's (DCF) Co-60 gamma source [10]. The spectrum of gamma ray energies produced by Co-60 decay has peaks at 1.17 MeV and 1.33 MeV which primarily interact with REBCO through incoherent scattering (figure 2). All previously published gamma-irradiation measurements of macroscopic REBCO samples were performed ex-situ—that is, measurements of critical current were taken after samples were irradiated (at room temperature or cryogenic temperature). Such experiments are valuable to determine the engineering lifetime of superconductors as a function of radiation fluence, but offer no information as to how radiation affects superconductivity during current flow. They also have conflicting conclusions: some observe no change in critical current with fluence [11, 12]; others show an initial increase in critical current, followed by a decrease at larger fluences [12, 13]; others still only show a decrease in critical current with fluence [14].

This paper reports an in-situ experiment to investigate the effect of gamma rays on the critical current of REBCO coated conductor tapes. Current–voltage (I–V) traces were measured during gamma irradiation with samples submerged in liquid nitrogen at 77 K at self-field. The dose rate on the REBCO samples was ≈ 86 Gy min⁻¹, roughly equal to the maximum expected dose rate on the magnets of proposed fusion power pilot plants.

2.1. Sample preparation

The REBCO tapes tested were SuperPower SCS4050-AP (2011) coated conductors. Bridges of widths 0.5 and 0.25 mm were laser-cut into 100 mm long, 4 mm wide samples (figure 3). Using bridges reduces the tapes' effective critical current thereby reducing the current required from the power supply unit and the power deposited in the tape as a result of resistive heating during measurements. The laser cut channel had an ablation depth of 33.8 µm (approximately 10 µm into the Hastelloy substrate) and width of 30 µm. The copper stabiliser and silver diffusion barrier over-layers were not removed from the samples. This was done for ease of electrical connection, to assist with heat conduction away from the REBCO layer and to protect the REBCO layer from chemical damage.

Zoom In Zoom Out Reset image size

2.2. Sample mounting and measurement procedure

Monte Carlo N-Particle (MCNP) calculations of the DCF gamma chamber were performed to calculate the expected dose rate of the samples during testing. The dose rate calculations were validated against Radcal 10X6-0.18 high dose-rate ion chamber measurements [16], performed within the irradiation chamber prior to the experiment by DCF staff—accurate to ± 4%. The modelled samples were treated as a homogenised bulk weighted by percentage mass of the various tape constituent atomic species (inclusive of the tape Cu stabiliser, Ag protection layer, REBCO, buffer layers and Hastelloy substrate). The calculations yielded an average dose rate of 86 Gy min⁻¹ to the sample. The dose rate of the parts of the samples in the shadow of the clamps and bolts is approximately two times lower than the parts of the tape that were fully exposed. The flux across the bridge region was uniform. Figure 4 shows a 2D map of the flux across the irradiation chamber and the dose rate of the sample.

Zoom In Zoom Out Reset image size

The standard $E_{\mathrm{c}}$ = 100 µV m⁻¹ electric field criterion was used for critical current density. This corresponds to a voltage criterion of V $_{\mathrm{c}}$ = 4 µV in this study. Data were fitted between electric fields of E = 0.4 µV and E = 8 µV to the power law [17]

$$\begin{equation} V(I) = V_{\mathrm{c}} \times \left(\frac{I}{I_{\mathrm{c}}}\right)^n ~, \end{equation} \tag{ 1 }$$

where I is current, $I_{\mathrm{c}}$ is critical current. The critical currents of each sample for each measurement are summarised in figure 5. Individual I–V curves for the five measurements (once before irradiation, three in-situ tests and once after irradiation) are shown in figure 6 for samples A-F. Individual I–V curves for the four further measurements (once prior to irradiation and three in-situ tests) on samples A and B after an additional room temperature dose of 208 kGy are shown in figure 7.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

No effect was observed on the critical currents of any samples under Co-60 gamma irradiation at a dose rate of 86 Gy min⁻¹. Any gamma interactions with the REBCO tapes were sufficiently small as to be within the error of the measurement. MCNP calculations predict a total gamma-induced heat load on the tapes of 0.21 W so it is therefore not surprising that the effective cooling method of liquid nitrogen submersion eliminated any effect of heating (as was the intention). The variation in the critical currents of different samples is attributed to variation in critical current along and across a tape, or defects caused during sample preparation. The in-situ critical currents of samples A and B were also unchanged after receiving an additional room temperature 208 kGy dose. This suggests that a total absorbed dose to this level does not affect whether a gamma flux during current flow has an effect on superconductivity. Similarly, the critical currents of samples A and B after the additional 208 kGy dose were unchanged with respect to the initial before-irradiation, measurements.

The observed null effect of gamma fluence corroborates the negligible effect of gamma flux on 77 K critical current, $I_{{\mathrm{c}}}(77~K)$, at a dose of up to 27.4 MGy reported in recent work by Iio et al [11]. Older studies, however, observe conflicting effects. Cooksey et al [12] irradiated two YBCO samples using a Cs-137 source. In one sample they observed an initial increase to 1.2$~\times~I_{{\mathrm{c}}0}(77~\mathrm K)$ after a 6 kGy dose followed by a drop to 0.9$~\times~I_{{\mathrm{c}}0}(77~\mathrm K)$ after a 15 kGy dose. In a second sample they observed no change in $I_{\mathrm{c}}$ with dose. Aksenova et al [13] report an initial increase to 1.2$~\times~I_{{\mathrm{c}}0}(77~\mathrm K)$ after a 1 MGy dose followed by a decrease to 0.7$~\times~I_{{\mathrm{c}}0}(77~\mathrm K )$ after a dose of 7 MGy. However, the spectrum of the gamma rays to which the REBCO bulk sample was subjected to was not reported, making direct comparison difficult due to the different possible gamma ray interactions with matter depending on energy. Leyva et al [14] observe an initial rapid decrease in $I_{\mathrm{c}}(77~\mathrm K)$ of thick film, polycrystalline, REBCO samples to 0.8$~\times~I_{{\mathrm{c}}0}(77~\mathrm K)$ for Co-60 doses up to 100 kGy followed by a plateau in degradation up to a dose of 250 kGy and gradual degradation to 0.6$~\times~I_{{\mathrm{c}}0}(77~\mathrm K)$ by a dose of 400 kGy. The initial sharp drop is attributed to radiation induced damage at grain boundaries; the plateau is attributed to a combined effect of improvement in $T_{\mathrm{c}}$ with doses up to 300 kGy with this grain boundary damage; and the gradual degradation is attributed to an over-saturation of radiation induced point defects. However, in all of these studies [12–14], the YBCO layer was exposed to air during irradiation and would therefore have been subject to gamma-induced chemical reactions which may have contributed to the observed change in $I_{\mathrm{c}}$ [13]. In the aforementioned work by Iio et al [11] the samples were commercial REBCO tapes complete with their protective silver and copper layers (as in our study), which were additionally sealed in vacuum during irradiation and measurement to prevent these chemical reactions. The results from these works are summarised in table 1.

Table 1. Summary of the literature on the effects of gamma irradiation dose on high temperature superconductors on their critical current and critical temperature.

References	HTS	Irradiation temperature (K)	γ source	γ dose $\text{(MGy)}$	Observation
[11]	SCS4050-AP	293	Co-60	27.4	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 1.0
[12]	aYBa2Cu3O$_{7-x}$	293	Cs-137	$6.0~\times~10^{-3}$	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 1.2
				$1.5~\times~10^{-2}$	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 0.9
[12]	bYBa2Cu3O$_{7-x}$	293	Cs-137	$6.0~\times~10^{-3}$	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 1.0
				$1.5~\times~10^{-2}$	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 1.0
[13]	YBa2Cu3O$_{7-x}$	293	?	1.0	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 1.2
				3.0	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 0.8
				7.0	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 0.7
[14]	YBa2Cu3O$_{7-x}$	293	Co-60	0.1	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 0.8
				0.2	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 0.8
				0.3	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 0.7
				0.4	$I_{\mathrm{c}}/I_{{\mathrm{c}}0}$ = 0.6
[14]	YBa2Cu3O$_{7-x}$	293	Co-60	0.1	$\Delta T_{\mathrm{c}}$ = 1.5 K
				0.2	$\Delta T_{\mathrm{c}}$ = 2.0 K
				0.3	$\Delta T_{\mathrm{c}}$ = 0.0 K
				0.4	$\Delta T_{\mathrm{c}} = -1.0$ K
[18]	YBa2Cu3O$_{7-x}$	293	Co-60	1.3$~\times~10^{-2}$	$\Delta T_{\mathrm{c}}$ = 0.0 K
[19]	YBa2Cu3O$_{7-x}$	293	Co-60	1.0	$\Delta T_{\mathrm{c}}$ = 0.0 K
[20, 21]	YBa2Cu3O$_{7-x}$	293	Co-60	0.8	$\Delta T_{\mathrm{c}}$ = 0.0 K
[22]	YBa$_{2-y}$Sry Cu3O$_{7-x}$	293	Co-60	0.2	$\Delta T_{\mathrm{c}}$ = 0.0 K
				0.5	$\Delta T_{\mathrm{c}} = -7.0$ K
[23]	Y3Ba5Cu8O18	293	Co-60	$2.4~\times~10^{-3}$	$\Delta T_{\mathrm{c}} = -8.0$ K
				$1.2~\times~10^{-2}$	$\Delta T_{\mathrm{c}} = -14.5$ K
				$2.3~\times~10^{-2}$	$\Delta T_{\mathrm{c}} = -17.4$ K
				$4.5~\times~10^{-2}$	$\Delta T_{\mathrm{c}} = -47.1$ K
[24]	YBa2Cu3O$_{7-x}$	293	Cs-137	$2.7~\times~10^{-7}$	$\Delta T_{\mathrm{c}}$ = 2.2 K
[25]	Bi2Sr2CaCu2Ox	293	Cs-137	5.0$~\times~10^{-7}$	$\Delta T_{\mathrm{c}}$ = 9.0 K
			Co-60	0.6	$\Delta T_{\mathrm{c}}$ = 16.0 K
[26]	Bi-system	030	Proton	?	$\Delta T_{\mathrm{c}}$ = 5.3 K(1 h)
			irrad. Li		$\Delta T_{\mathrm{c}} = -4.0$ K(90 d)
[27]	EuBa2Cu3O$_{7-x}$	293	Co-60	$1.0~\times~10^{-2}$	$\Delta T_{\mathrm{c}} = -3.3$ K
				$2.0~\times~10^{-2}$	$\Delta T_{\mathrm{c}} = -4.7$ K
				$3.0~\times~10^{-2}$	$\Delta T_{\mathrm{c}} = -8.1$ K

^a 0.2 µm thickness, on MgO substrate. ^b 1.0 µm thickness, on LaAlO₃ substrate.

Other related work pertains to the effect of gamma irradiation on the critical temperature, $T_{\mathrm{c}}$, of high temperature superconductors which was not measured during this work. For completeness, the experiments are also summarised in table 1. The conclusions of different studies are inconsistent. Bohandy et al [18], Kutsukake et al [19], Albiss et al [20] and Özkan et al [21] observed no change in $T_{\mathrm{c}}$ after Co-60 gamma ray doses of 13 kGy, 1 MGy, 0.8 MGy and 0.8 MGy, respectively. Cooksey et al [12] similarly saw no change in $T_{\mathrm{c}}$ after a 15 kGy Cs-137 dose. Elkholy et al [22] saw no change in $T_{\mathrm{c}}$ in Sr doped YBa₂Cu₃O$_{7-x}$ up to doses of 200 kGy after which $T_{\mathrm{c}}$ steadily dropped with fluence, falling by 7 K by 500 kGy. Akduran [23] observed an unprecedented 47.1 K drop in the $T_{\mathrm{c}}$ of Y₃Ba₅Cu₈O₁₈ after a dose of 45 kGy and a drop of 8.1 K in $T_{\mathrm{c}}$ of EuBa₂Cu₃O$_{7-x}$ after 30 kGy. The significant decrease in $T_{\mathrm{c}}$ of the Y-based sample would render $I_{\mathrm{c}}(77~\mathrm K) = 0$, which is clearly in contrast with the unchanged $I_{\mathrm{c}}$ measured in this work. The environment of the irradiation chamber used by Akduran is not described, so the reduction in $T_{\mathrm{c}}$ could perhaps be due to gamma-catalysed chemical reactions as proposed in [13]. Leyva et al witnessed improvements in the $T_{\mathrm{c}}$ of YBa₂Cu₃O$_{7-x}$ of ≈ 2 K after 150 kGy Co-60 dose [14] and after a 270 mGy Cs-137 dose [24]. The cause suggested was gamma ray induced oxygen reordering and overall greater crystal uniformity (assumed to be non-optimal prior to irradiation), followed by degradation from 'overdoping'. The $T_{\mathrm{c}}$ of Bi-system has also been observed to increase by 16 K after a dose of 600 kGy [25]. Zhao et al [26] performed cryogenic irradiation of a Bi-system (the chemistry was undefined) at 30 K using gamma rays derived from a proton irradiated lithium target. An initial average increase in $T_{\mathrm{c}}$ of 5.3 K was measured, followed by an eventual average drop of 4.0 K after 90 days. The spectrum of the gamma rays was not disclosed however, making comparison with other work difficult. Different electronic (and indeed nuclear) interactions may take place depending on the incident gamma ray energy leading to, in principle, different microstructural damage. The conflicting nature of the literature suggests that more research is required before any firm conclusions can be drawn on the effect of gamma dose on high temperature superconductors. Future investigations should take pains to prevent chemical degradation of REBCO samples.

The self field, 77 K critical currents of SuperPower SCS4050-AP (2011) tapes were measured during 86 Gy min⁻¹ Co-60 gamma ray flux (approximately equal to the peak flux expected on the magnets of proposed fusion pilot power plants). No effect on the critical current was observed within error. This is a promising result for fusion magnets, suggesting that the current carrying capacity of REBCO magnets will not degrade as a result of incident gamma ray flux (of these energies) in-situ. The questions of the effects of (n, γ) interactions within the REBCO itself, higher energy gamma flux, and fusion-relevant in-situ neutron flux must still be answered. Additionally, no effect on critical current was observed for tapes that were irradiated to a total dose of ≈ 215 kGy within error. This finding corroborates more recent studies on the effects of gamma dose on commercial REBCO tapes, but is in conflict with older literature on the effects on REBCO lab-manufactured samples. This is perhaps due to protection offered by the tape copper and silver layers from gamma-catalysed chemical reactions not present in older work.

We acknowledge the support of The University of Manchester's Dalton Cumbrian Facility (DCF), a partner in the National Nuclear User Facility, the EPSRC UK National Ion Beam Centre and the Henry Royce Institute. We recognise R Edge, C Tyagi and K Warren for their assistance during the experiment. Thanks to H Campbell for organising sample laser cutting. Thanks also to T Todd, W Iliffe, S C Wimbush and S Speller for enlightening discussion and paper review.

The data that support the findings of this study are openly available at the following URL/DOI: https://doi.org/10.14468/b1ce-mg50 [28].