Rare-earth barium-copper-oxide (REBCO) superconductors are high-field superconductors fabricated in a tape geometry that can be utilized in magnet applications well in excess of 20 T. Due to the multilayer architecture of the tape, delamination is one cause of mechanical failure in REBCO tapes. During a mechanical slitting step in the manufacturing process, edge cracks can be introduced into the tape. These cracks are thought to be potential initiation sites for crack propagation in the tapes when subjected to stresses in the fabrication and operation of magnet systems. We sought to understand which layers were the mechanically weakest by locating the crack initiation layer and identifying the geometrical conditions of the slitter that promoted or suppressed crack formation. The described cracking was investigated by selectively etching and characterizing each layer with scanning electron microscopy, laser confocal microscopy, and digital image analysis. Our analysis showed that the average crack lengths in the REBCO, LaMnO 3 (LMO) and Al 2 O 3 layers were 34 μ m, 28 μ m, and 15 μ m, respectively. The total number of cracks measured in 30 mm of wire length was between 3000 and 5700 depending on the layer and their crack densities were 102 cracks mm −1 for REBCO, 108 cracks mm −1 for LMO, and 183 cracks mm −1 for Al 2 O 3 . These results indicated that there are separate crack initiation mechanisms for the REBCO and the LMO layers, as detailed in the paper. With a better understanding of the crack growth behavior exhibited by REBCO tapes, the fabrication process can be improved to provide a more mechanically stable and cost-effective superconductor.

Two sections of heat-treated (HT) and non-heat-treated (NHT) Cable-in-Conduit Conductor (CICC) of a design similar to the ITER tokomak have been imaged using very high energy X-ray tomography at the ESRF beamline ID19. The sample images were collected at four temperatures down to 77 K. These results showed a greater degree of movement, bundle distortion and touching strands in the NHT sample. The HT sample showed non-linear movements with temperature especially close to 77 K; increasing non-circularity of the superconducting fibre bundles towards the periphery of the CICC, and touching bundles throughout the CICC. The images have highlighted where future design might improve potential weakness, in particular at the outer perimeters of the conductor and the individual sub-cable, 'petal' wraps.

This project assesses the mechanical performance limits of two advanced superconducting technologies for HEP magnets, Bi-2212 round wire and REBCO tape. We specify the microstructural basis for critical current degradation in Bi-2212, and establish techniques for investigating the processing maturity of the powder-in-tube technology. For REBCO, we investigate a variety of defect mechanisms including delamination and damage from mechanical slitting of the wire.

Brittle fracture of Nb3Sn filaments is one mechanism by which the current-carrying capacity of composite Nb3Sn wires is degraded. However, there are relatively little data in the literature on the intrinsic material fracture properties of Nb3Sn filaments, because the complex composite structure (matrix, secondary phases, and defects such as voids) acts as an integrated mechanical unit. In this study, we extracted individual Nb3Sn filaments from a fusion-style Nb3Sn composite wire and conducted tensile testing to determine the fracture strength distribution of the isolated filaments. The distribution is modeled using a Weibull function. The relative fracture propensity of fully reacted filaments versus those with unreacted Nb cores is compared. The presence of a Nb core reduces, on average, the fracture strength of a filament by 38% and the strain to failure by 29%. Understanding the fracture probability of Nb3Sn as a function of both stress and volume will allow strand and conductor modeling efforts to more accurately represent the relative contributions of fracture and other effects (such as plasticity) to irreversible current density degradation, and may assist wire manufacturers in assessing the mechanical impact of wire design changes.

Cables made with Nb3Sn-based superconductor strands will provide the 13 T maximum peak magnetic field of the ITER central solenoid (CS) coils and they must survive up to 60 000 electromagnetic cycles. Accordingly, prototype designs of CS cable-in-conduit-conductors (CICC) were electromagnetically tested over multiple magnetic field cycles and warm-up-cool-down scenarios in the SULTAN facility at CRPP. We report here a post-mortem metallographic analysis of two CS CICC prototypes which exhibited some rate of irreversible performance degradation during cycling. The standard ITER CS CICC cable design uses a combination of superconducting and Cu strands, and because the Lorentz force on the strand is proportional to the transport current in the strand, removing the copper strands (while increasing the Cu:SC ratio of the superconducting strands) was proposed as one way of reducing the strand load. In this study we compare the two alternative CICCs, with and without Cu strands, keeping in mind that the degradation after the SULTAN test was lower for the CICC without Cu strands. The postmortem metallographic evaluation revealed that the overall strand transverse movement was 20% lower in the CICC without Cu strands and that the tensile filament fractures found were less, both indications of an overall reduction in high tensile strain regions. It was interesting to see that the Cu strands in the mixed cable design (with higher degradation) helped reduce the contact stresses on the high pressure side of the CICC, but in either case, the strain reduction mechanisms were not enough to suppress cyclic degradation. Advantages and disadvantages of each conductor design are discussed here aimed to understand the sources of the degradation.

Systematic studies of the intrinsic irreversible strain limit e(irr,0), microstructure, and microchemistry were made on several internal-tin Nb3Sn pre-production wires, fabricated for the domestic agencies of the USA and China participating in the International Thermonuclear Experimental Reactor. These wires were produced by Luvata, Oxford Superconducting Technology (OST), and Western Superconducting Technologies (WST), and were intended for the tokamak's toroidal-field coils. The results of this study show that, for a final heat-treatment at 650 degrees C to form the A15 phase, both epsilon(irr,0) and the de-pinning field B-c2* improved by increasing heat-treatment duration beyond 100 h for the Luvata wires. On the other hand, we saw no improvement in these two parameters as a function of heat-treatment duration in the OST wires. Furthermore, micro-chemical analysis of OST wires revealed that some Nb3Sn filaments have a Sn- and Ti-rich phase at the interface between Cu(Sn) matrix and Nb3Sn in the form of a shell around individual filaments. This phase is far less prominent in the Luvata and WST conductors, and could inhibit diffusion of Sn and Ti into Nb3Sn filaments during the reaction and may potentially be the reason for the lack of noticeable change in B-c2* with heat-treatment duration in the OST wires. The increase of epsilon(irr,0) and B-c2* with heat-treatment duration in the Luvata wires and the lack of increase in the OST wires may suggest a possible correlation between epsilon(irr,0) and the stoichiometry of the A15 composition. Investigation of the samples' microstructure revealed only a small number of cracked Nb3Sn filaments despite the significant and permanent degradation of their critical current I-c when subjected to longitudinal tensile strain epsilon beyond epsilon(irr,0). The scarcity of cracks indicate that I-c(epsilon) measurements are highly sensitive to crack formation in Nb3Sn filaments, especially at low electric-field criteria <= 0.1 mu V cm(-1), even when the sizes of the individual filaments are only few micrometers. All the strands contained substantial Kirkendall porosity, but we found that the quantity and distribution of the Kirkendall voids vary significantly with strand design. Luvata wires have the least porosity, followed by WST wires, and then by OST strands. However, even though the presence of cracks in the Nb3Sn filaments that are in close proximity to Kirkendall voids suggest a correlation between crack initiation and the proximity of the filaments to these voids, the porosity investigation established no definitive relationship between porosity and epsilon(irr,0) in the wires studied.

The International Thermonuclear Experimental Reactor (ITER) cable-in-conduit conductor used in the superconducting magnet system consists of a cable made of 300 to 1440 strands housed in a stainless steel tube (called as jacket or conduit). There are circular, square, as well as circle-in-square jackets, made of either a very low carbon AISI 316LN and AISI 316L grade stainless steels, or a high Mn austenitic stainless steel developed for ITER called JK2LB. Selected mechanical properties of the base material and weld joint were tested at room temperature and/or cryogenic temperatures (<; 7 K) at predefined mechanical deformation and heat treatment condition. The domestic agencies' reference laboratories and the ITER-IO appointed reference laboratories, CERN and Karlsruhe Institute of Technology performed mechanical tests such as tensile strength, fracture toughness, and fatigue crack growth rate. This paper will compare the test results (e.g., elongation to failure) from different laboratories, present the statistics, and identify any systematic differences.

In Tokamak fusion reactors, such as ITER, superconducting strands are subjected to repeated Lorentz force loading and unloading which may degrade performance over time. The Cu matrix which surrounds the brittle Nb3Sn filaments allows the possibility of some elastic-plastic deformation that can initiate filament cracking. We seek to understand if there are strand design variables that might ameliorate such degradation but before being able to do such experiments, we need to establish procedures that can unambiguously detect the cracking caused by loading, rather than by subsequent metallographic examination. Here we make a first report of our procedures after fatigue testing at 77K. Filament crack densities were quantified from large montages covering approximate to 20 mm length of strand. Three types of cracks were present. The most common were cracks transverse to the filament axis adjacent to voids. Cracks away from voids were of low density until close to the fracture strain. A third kind of crack which generally initiates at unreacted Nb cores of filaments extends radially and in the plane parallel to the wire axis. Example results on one bronze strand fatigued for 1000 loading cycles at axial strains from 0.4% to 1.14% are shown.

A benchmarking experiment was conducted to compare strain measurement facilities at the National Institute of Standards and Technology (NIST) and the University of Twente. The critical current of a bronze-route Nb3Sn wire, which was fabricated for the International Thermonuclear Experimental Reactor (ITER), was measured as a function of axial strain and magnetic field in liquid helium at both institutes. NIST used a Walters' spring strain device and University of Twente used a bending beam ("Pacman") apparatus. The ITER bronze-route wire investigated had a very high irreversible strain limit that allowed comparing data over a wide range of applied strain between -1% and +1%. Similarities of the data obtained by use of the two apparatuses were remarkable, despite the many differences in their design and techniques.

The ITER machine will require approximately 250 tons of NbTi strands and 500 tons of Nb3Sn strands. NbTi will be used in the Poloidal Field (PF) coils, Correction Coils (CC) and feeder busbars, whereas Nb3Sn will be used in the Central Solenoid (CS) and Toroidal Field (TF) coils. The large amount of superconducting strands needed requires worldwide procurement, involving suppliers from six of the seven ITER Domestic Agencies (DAs). To ensure reliable test results, it is necessary to benchmark the test facilities at each supplier and at each DA reference laboratory for physical and superconducting properties measurement, as well as sample preparation techniques. Following previous benchmarking efforts related to ITER procurement in the mid-1990's and to supplier and DA laboratory qualification performed on bronze route Nb3Sn strands in 2009, we report here the latest rounds on internal tin Nb3Sn and NbTi strands. Ten participants from five DAs (China, EU, South Korea, Russia, and the U.S.) together with CERN (the ITER Organisation's reference laboratory) took part in the benchmarking of internal tin Nb3Sn strands, and six participants from China and Russia, plus CERN, participated in the benchmarking of NbTi strands.

In the above titled paper (ibid., vol. 22, no. 3, p. 4802606, June 2012), there are errors in (2). The correct formula is presented here.

The ITER CICC will undergo a number of cool-down and warm-up cycles over the lifetime of the plant. The standard SULTAN based ITER conductor qualification test normally includes one thermal cycle in the test sequence. In many samples a performance degradation was observed following this thermal loading. In order to investigate the effect of multiple thermal cycles on the TF conductor short sample, additional repeated thermal cycles to liquid nitrogen temperature were carried out on the left leg of the CNTF3 sample and the JATF5 sample.Thermal cycles using SULTAN are very time consuming, about four days, with a corresponding cost of around 32 kEuro. Ten thermal cycles will give an estimation of the degradation upon repeated thermal loading, but would require a prohibitive amount of time in SULTAN, and therefore cause a significant delay in the testing of other time critical samples. As a large fraction of the change in thermal contraction occurs between room temperature and liquid nitrogen temperature, a purpose made facility and program was developed.This ad-hoc facility allowed faster, more cost effective thermal cycles that crucially did not interfere with SULTAN's ongoing test program. Cooling and heating was provided by means of forced flow nitrogen. The sample was contained within a vacuum to prevent the formation of moisture or ice. During warm-up, a heater distributed around the CNTF3A was also used.These cycles were performed both before and after electromagnetic loading. The results of these tests indicated that thermal loading before the first electromagnetic load cycle did not result in a worsening of the conductor performance. The tests following repeated thermal cycling after electromagnetic loading show a thermal cycle causes a performance degradation but a large number of consecutive thermal cycles do not appear to have a significant effect.

In the above titled paper (ibid., vol. 22, no. 3, p. 4802606, June 2012), there are errors in (2). The correct formula is presented here.

In a tokamak, such as ITER, superconducting strands suffer from bending and uniaxial strain due to Lorentz force loading/unloading and thermal cool down which may de- grade the performance over time due to specific cable-in-conduit conductor (CICC) design choice. Under repeated uniaxial loading the Cu(Sn) matrix which surrounds the brittle filaments allows the possibility of some elastic-plastic deformation that can initiate filament cracking. Here we present a metallographic study of filament cracking under increasing uniaxial loading cycles (0, 1000, 10,000 and 30,000 cycles) for one ITER Toroidal field (TF) bronze-process strand (tested at 0.4%, 0.6% and 1% strain) and one ITER TF internal tin strand (tested at 0.4%, 0.6% and 0.7% strain). Significant cracking of filaments was found at close to the respective fracture limits (strain at which strand breaks under uniaxial tensile loading) for both strands. After 0.6% strain, filament cracking in the bronze-process strand tends to increase with increasing number of loading cycles up to 10,000 and then remains almost constant after increasing the loading cycles from 10,000 to 30,000. The internal tin strand on the other hand showed an increase in filament cracking with increasing loading cycles to 10,000 up to 0.6% strain. For both types of strand and in all conditions the cracks were most likely to be found adjacent to voids.

The ITER Central Solenoid (CS) conductor is composed of 576 superconducting strands and 288 Cu strands assembled together into a multistage cable and protected by a circle-in-square jacket with the outer dimension of 49 mm × 49 mm. In R&D to prepare for the ITER CS conductor manufacturing, mechanical tests of jacket, welding tests and manufacturing of 181-m long dummy conductor have been performed. In this paper, the R&D activities are presented, showing that as a result of this R&D, the CS conductor manufacturing technologies have been preliminary defined to start the procurement of the CS conductor.

We analyzed the ITER TFEU5 cable-in-conduit conductor (CICC) after the full SULTAN conductor qualification test in order to explore whether Lorentz force induced strand movement inside the CICC produces any fracture of the brittle Nb3Sn filaments. Metallographic image analysis was used to quantify the change in void fraction of each sub-cable (petal); strands move in the direction of the Lorentz force, increasing the void space on the low force side of the CICC and producing a densification on the high force side. Adjacent strand counting shows that local increases in void space result in lower local strand-strand support. Extensive metallographic sampling unambiguously confirms that Nb3Sn filament fracture occurred in the TFEU5 CICC, but the filament fracture was highly localized to strand sections with high local curvature (likely produced during cabling, where strands are pivoted around each other). More than 95% of the straighter strand sections were free of filament cracks, while less than 60% of the bent strand sections were crack free. The high concentration of filament fractures on the tensile side of the strand-strand pivot points indicates that these pivot points are responsible for the vast majority of filament fracture. Much lower crack densities were observed in CICC sections extracted from a lower, gradient-field region of the SULTAN-tested cable. We conclude that localized filament fracture is induced by high Lorentz forces during SULTAN testing of this prototype toroidal field CICC and that the strand sections with the most damage are located at the petal corners of the high field zone.

NbTi strands to be used in four of the six ITER poloidal field (PF) coils, all the correction coils (CC) and all the superconducting feeder busbars are being produced in China. Short full-size qualification conductor (cabled and jacketed) samples have been developed at ASIPP and tested at CRPP. Single pinning mechanism parametrization for this Chinese strand (type S2) has been obtained using the Bottura scaling law. The determination of the scaling parameters using a Kramer-type regression method will be described. A comparison between the critical temperature at the operating current and field of a single strand as determined by the parametrization and the current sharing temperature (T-CS) of a few conductor samples tested at the SULTAN facility will be made. The validity and limitation of the estimation will be discussed. The estimated T-CS dependence on various (superconducting critical as well as geometric and volumetric) parameters will be assessed using the modelled critical surface. Errors propagated from critical current (I-c) measurements of the strands and parameter fitting, and other uncertainties, will be quantified.

The performance of the toroidal field (TF) magnet conductors for the ITER machine are qualified by a short full-size sample (4 m) current sharing temperature (T-cs) test in the SULTAN facility at CRPP in Villigen, Switzerland, using the operating current of 68 kA and the design peak field of 11.8 T. Several samples, including at least one from each of the six ITER Domestic Agencies participating in TF conductor fabrication (China, European Union, Japan, Russia, South Korea and the United States), have been qualified by the ITER Organization after achieving T-cs values of 6.0-6.9 K, after 700-1000 electromagnetic cycles. These T-cs values exceed the ITER specification and enabled the industrial production of these long-lead items for the ITER tokamak to begin in each Domestic Agency. Some of these samples did not pass the qualification test. In this paper, we summarize the performance of the qualified samples, analyze the effect of strand performance on conductor performance, and discuss the details of the test results.