1

Autopsy of an ITER TF conductor without Tcs degradation reveals filament fractures and strand

movement

Carlos Sanabria 1, Student Member, IEEE, Peter J. Lee 1, Senior Member, IEEE, Arnaud Devred 2, Senior Member, IEEE, and David C. Larbalestier 1, Fellow, IEEE

1 Applied Superconductivity Center, NHMFL, Florida State University, Tallahassee, FL 32310, USA 2 ITER Organization, Saint-Paul-lès-Durance, Cadarache, France

Abstract— Prototype ITER Toroidal Field (TF) and Central

Solenoid (CS) Cable in Conduit Conductors (CICCs) samples have been tested in the SULTAN facility under conditions that simulate ITER plasma operations, and they have shown a wide variety of degradation profiles. We have performed metallographic analyses on fully tested SULTAN samples to understand the physical changes that occur inside the cable during the electromagnetic cycling and warm-up cool-down (WUCD) cycles applied during the SULTAN tests. From the CS conductor autopsies, we learned that if transverse strand movement is blocked, there are no filament fractures and the current sharing temperature, Tcs, does not exhibit degradation vs. cycling in SULTAN. But among all the ITER TF conductors, there was only one prototype that did not exhibit Tcs degradation in SULTAN, and the metallographic autopsy of this sample revealed that significant transverse strand migration had occurred as well as fracture of some Nb3Sn filaments. The behavior of ITER CICCs cannot be fully understood unless the remarkable properties of this cable can be explained. In this paper we compare this cable with other TF as well as CS CICCs in an attempt to provide such an explanation.

I. INTRODUCTION

ROTOTYPE Cable-in-Conduit Conductors (CICCs) designed for the ITER fusion reactor are tested and

qualified in the SULTAN facility at CRPP in Switzerland [1]. They undergo a series of electromagnetic and warm-up-cooldown cycles to simulate the life time of a real coil. These tests have until recently shown degradation in performance during testing [2], but new prototypes have an increase in performance using a short twist pitch design [3]. The Applied Superconductivity Center (ASC) at the National High Magnetic Field Laboratory (NHMFL) has been tasked by the ITER Organization (IO) and US-ITER to disassemble and perform a detailed metallographic post-mortem analysis of the prototypes to quantitatively evaluate the condition of the superconducting cable after testing. An initial autopsy performed in 2012 revealed evidence of filament fracture and transverse strand movement that could be the reason for Tcs degradation [4]. In this new set of results we look at SULTAN samples with different configurations and degradation profiles to understand the effects of the Lorentz force inside the

Submitted in January 22, 2014. This work was supported by the ITER

Organization under contract ITER/CT/11/4300000511, by the State of Florida and by the US Department of Energy under award DE-FG02-06ER54881.

conductors and correlate that to the Tcs evolution with electromagnetic cycling seen in SULTAN

The degradation profiles of the ITER conductor prototypes to which metallographic autopsies were performed are shown in Fig. 1.

Regardless of the number of cycles and the CICC design, the conductors in Fig. 1 can be divided in two categories: conductors with an increase of Tcs and conductors with a decrease of Tcs during cycling. In this paper we will use metallographic techniques and image analysis algorithms to describe the physical state of the conductors after the SULTAN qualification.

II. PROCEDURE

Strand migration measurements were performed on very large images of full transverse obtained by assembling mosaics of multiple scanning laser confocal images.[4] These large images hold enough resolution to allow us to directly measure the displacement that the strands underwent during cycling due to Lorentz force by tracking the coordinates of the strands and comparing their average position to the center of the conductor. A total of six transverse cross sectional images like those in Fig. 2 were taken for each conductor in Fig. 1.

We examine the integrity of the strands by careful extraction of individual strands from the CICCs and then polishing them in longitudinal cross section. Field Emission Scanning Electron Microscope (FESEM) images like that shown in Fig. 3 allow us to count the number of fractured

P

Fig. 1. Summary plot of ΔTcs versus number of cycles, where ΔTcs is the difference between the Tcs at a given cycle number, N, and the Tcs

at N = 1. (Data courtesy of Arnaud Devred ITER IO).

IEEE/CSC & ESAS SUPERCONDUCTIVITY NEWS FORUM (global edition), October 2014 (Preview 1). ASC 2014 presentation and paper 2LOr2B-05; 2nd Prize in Best Student Paper Contest, Large Scale.

2

filaments and record evidence of local cold work in the copper sheath and the bronze matrix of the Nb3Sn strands. A representative population of 80 strand sections was extracted from each conductor and examined looking for evidence of filament fracture.

III. RESULTS

In the SULTAN facility there is a region of maximum magnetic field (10.8 T) that is homogeneous over a length of 410 mm that we call the High Field Zone (HFZ). We compare CICCs from the HFZ with those from a Low Field Zone (LFZ) with negligible magnetic field 1.4 m away from the center of the SULTAN magnet. None of the conductors studied showed filament fracture or transverse strand migration in the LFZ while the HFZ results are as follows:

A. Transverse Cross Section Analysis.

The average transverse movement of the strands is shown for each conductor in Fig. 4. It was seen that the cables assembled using the short twist pitch design [3] (CSIO2-2 and CSJA3-2) had the least strand migration.

B. Filament Fracture.

All conductors with the exception of the CSIO2-2 and CSJA3-2 developed filament fractures in a small population of strands. The filament fractures seen correspond to that of a bending scenario as seen in simulations of strand bending [5].

IV. DISCUSSION

Combining our crack density measurements with the average transverse migration of the strand in the CICCs in Fig. 4, it is not surprising that the short twist pitch conductors (CSIO2-2 and CSJA3-2) do not exhibit Tcs degradation, however, we measure significant strand migration in the TFRF4 conductor as well as filament fracture (Fig. 5) but the CICC shows no Tcs degradation.

V. CONCLUSION

Although there is a good correlation between our measurements of strand migration and filament fracture, there is a strong exception in the case of the TFRF4 cable that merits further investigation. It is well known that this RF strand has a barrier and filament geometries that are different from other ITER manufacturers [6] but electromechanical measurements of the individual strands [5] have yet to reveal a difference that can explain these results. Solving this mystery may lead to better strand and cable designs.

ACKNOWLEDGMENTS

Special thanks to Nicolai Martovetsky (ITER-US), Matthew Jewell (ITER, now at UW Eau-Claire) and Ian Pong (ITER, now at BNL) for fruitful discussions. The CICCs were provided by courtesy of Pierluigi Bruzzone (Plasma Physics Research Center) with agreement from the Japan Atomic Energy Agency (JAEA), ITER-US and ITER-Russia.

Fig. 2. Transverse cross sectional images of (a) the TFRF4 sample and

(b) the CSIO1-2 sample (jacket not shown) .The images shown are thumbnails of much larger images composed of more than 400 frames

taken using a digital stage Laser Scanning Confocal Microscope.

Fig. 3. (a) Longitudinal cross section of a strand extracted from the CSJA2-2 sample showing (b) filament fractures and (c) cold work

produced at the contact point of a neighboring strand.

Fig. 1. Average-strand-center distance from the conductor-center of

all the conductors shown here.

Fig. 5. FESEM images of strands from the CSJA3-1 sample (a and b) showing a crack free microstructure with a closer look at the

microstructure (b) and from the TFRF4 sample showing filament fractures in blue (c) and a closer look at the cracks (d).

IEEE/CSC & ESAS SUPERCONDUCTIVITY NEWS FORUM (global edition), October 2014 (Preview 1). ASC 2014 presentation and paper 2LOr2B-05; 2nd Prize in Best Student Paper Contest, Large Scale.

3

REFERENCES [1] P. Bruzzone, B. Stepanov, R. Wesche, E. Salpietro, A. Vostner, K.

Okuno, T. Isono, Y. Takahashi, H. C. Kim, K. Kim, A. K. Shikov, and V. E. Sytnikov, “Results of a New Generation of ITER TF Conductor Samples in SULTAN,” IEEE Transactions on Applied Superconductivity, vol. 18, no. 2, pp. 459–462, 2008.

[2] P. Bruzzone, B. Stepanov, R. Wesche, M. Bagnasco, F. Cau, R. Herzog, M. Calvi, M. Vogel, M. Jenni, M. Holenstein, and H. Rajainmaki, “Status Report of the SULTAN Test Facility,” IEEE Transactions on Applied Superconductivity, vol. 20, no. 3, pp. 455–457, 2010.

[3] A. Devred, D. Bessette, P. Bruzzone, K. Hamada, T. Isono, N. Martovetsky, N. Mitchell, Y. Nunoya, K. Okuno, I. Pong, W. Reiersen, C. M. Rey, B. Stepanov, Y. Takahashi, and A. Vostner, “Status of Conductor Qualification for the ITER Central Solenoid,” IEEE Transactions on Applied Superconductivity, vol. 23, no. 3, pp. 6001208–6001208, 2013.

[4] C. Sanabria, P. J. Lee, W. Starch, I. Pong, A. Vostner, M. C. Jewell, A. Devred, and D. C. Larbalestier, “Evidence that filament fracture occurs in an ITER toroidal field conductor after cyclic Lorentz force loading in SULTAN,” Supercond. Sci. Technol., vol. 25, no. 7, p. 075007, Jul. 2012.

[5] A. Nijhuis, R. P. P. van Meerdervoort, H. J. G. Krooshoop, W. a. J. Wessel, C. Zhou, G. Rolando, C. Sanabria, P. J. Lee, D. C. Larbalestier, A. Devred, A. Vostner, N. Mitchell, Y. Takahashi, Y. Nabara, T. Boutboul, V. Tronza, S.-H. Park, and W. Yu, “The effect of axial and transverse loading on the transport properties of ITER Nb3Sn strands,” Supercond. Sci. Technol., vol. 26, no. 8, p. 084004, Aug. 2013.

[6] V. I. Tronza, S. A. Lelekhov, B. Stepanov, P. Bruzzone, D. S. Kaverin, K. A. Shutov, and V. S. Vysotsky, “Test Results of RF ITER TF Conductors in the SULTAN Test Facility,” IEEE Transactions on Applied Superconductivity, vol. 24, no. 3, pp. 1–5, 2014.

IEEE/CSC & ESAS SUPERCONDUCTIVITY NEWS FORUM (global edition), October 2014 (Preview 1). ASC 2014 presentation and paper 2LOr2B-05; 2nd Prize in Best Student Paper Contest, Large Scale.