X-Ray Studies: Phase Transformations and Microstructure Changes

Authored by: Christian Scheuerlein , M. Di Michiel

Handbook of Superconductivity

Print publication date:  July  2022
Online publication date:  July  2022

Print ISBN: 9781439817360
eBook ISBN: 9781003139638
Adobe ISBN:

10.1201/9781003139638-4

 

Abstract

Conventional materials characterisation of superconducting wires or tapes implies the destructive preparation of the samples by cutting, grinding and polishing, for instance, for microscopic studies. A destructive sample preparation is also required for X-ray diffraction (XRD) experiments with laboratory diffractometers, which commonly use Cu Kα radiation (ECu-Kα∼8.03 keV) with a penetration depth of some tens of µm in metallic samples.

 Add to shortlist  Cite

X-Ray Studies: Phase Transformations and Microstructure Changes

G1.2.1  Introduction

Conventional materials characterisation of superconducting wires or tapes implies the destructive preparation of the samples by cutting, grinding and polishing, for instance, for microscopic studies. A destructive sample preparation is also required for X-ray diffraction (XRD) experiments with laboratory diffractometers, which commonly use Cu Kα radiation (ECu-Kα∼8.03 keV) with a penetration depth of some tens of µm in metallic samples.

In contrast, both neutrons and high-energy photons can penetrate mm-thick strongly absorbing superconductors, which enables non-destructive diffraction experiments, for instance, with Bi-2223 tapes [1, 2]. The X-ray transmission through typical Ta-alloyed Nb3Sn wires as a function of the X-ray energy is presented in Figure G1.2.1. X-ray energies above 50 keV enable non-destructive experiments in transmission geometry with Nb3Sn wires [3].

X-ray transmission through Ta-alloyed Nb

Figure G1.2.1   X-ray transmission through Ta-alloyed Nb3Sn wires of equal composition with different diameters.

Typical acquisition times of neutron diffraction pattern of individual superconducting wires are in the order of hours [4]. State-of-the-art high-energy synchrotron sources can provide very high monochromatic photon flux densities, and diffraction measurements can be performed within seconds when using fast read-out area detectors in Debye–Scherrer transmission geometry. This chapter focuses on high-energy synchrotron radiation in situ studies of entire processes, which can be much faster than conventional materials studies that require a series of samples to be prepared after different processing steps. The non-destructive in situ studies also avoid experimental uncertainties caused by sample quenching, sample inhomogeneity and preparation artefacts.

The relatively small scattering angles of high-energy X-rays, facilitate to add auxiliary equipment, for instance, a furnace for heat treatment (HT) studies. For the experiments presented here, two furnaces of the ESRF ID15 beamline have been used. For in situ studies in inert gas or air at ambient pressure, the ID15 diffraction and tomography furnace was used [Figure G1.2.2(a)]. After alignment with respect to the X-ray beam, the position of this furnace remains fixed during the experiment. The sample is mounted onto a ceramic stick that enters the furnace from a bottom hole. For sample alignment and rotation, the ceramic stick is mounted onto a goniometer and the rotation and translation stages. The thin Al foil windows at both sides of the furnace are nearly transparent for high-energy photons. The furnace temperature can be regulated using the temperature reading of a thermocouple that can be inserted into the furnace through the bottom hole. Large sample temperature uncertainties can be avoided when the thermocouple is spot welded onto the sample.

(a) ID15 furnace for

Figure G1.2.2   (a) ID15 furnace for in situ XRD and tomography at ambient pressure. (b) Furnace for in situ XRD at pressures up to 200 bar in measurement position and (c) capillary furnace withdrawn from the high-pressure cell.

For overpressure in situ studies, a set-up consisting of a capillary furnace around a high-pressure cell made of a single crystal sapphire tube has been used [Figure G1.2.2(b)]. This furnace was developed for combined in situ high-energy X-ray diffraction and mass spectrometry investigations during catalysed gas/solid or liquid/solid reactions [5]. The connection of the high-pressure cell to the pressure controller is done with a flexible stainless steel line that allows rotating the high-pressure cell up to 360° during the acquisition of diffraction patterns. For the study of superconducting wires, the high-pressure cell was modified such that a thermocouple can be spot welded onto the superconductor sample [6].

Other sample environments are possible too, for instance, a cryostat and a tensile rig can be added. This makes it possible to study the superconductor electromechanical behaviour and the damage development by XRD measurements at well-defined uniaxial tensile stress or strain at cryogenic temperatures, for instance, in liquid Helium [3, 7, 8] or in liquid Nitrogen [9]. Lattice distortions, superconducting properties and mechanical properties of high-temperature superconductors can be measured simultaneously [10].

X-ray absorption micro-tomography (µ-CT) can provide three-dimensional images of the superconductor bulk. A spatial resolution of µ-CT in the order of 1 µm is today routinely obtainable with both laboratory and synchrotron sources. Fast µ-CT [10, 11], where the acquisition of about 1000 radiographs needed for the reconstruction of one tomogram lasts not more than 1 minute, is required for in situ studies of microstructural changes and porosity formation during the processing of superconductors with temperature ramp rates in the order of 100°C/h.

Different synchrotron techniques, for instance, high-energy XRD and µ-CT, can be combined in one experiment. This combination has been pioneered at the ESRF ID15A beamline for an in situ study of the void growth mechanisms in Nb3Sn wires [12]. This was achieved using two X-ray beams, a high-intensity filtered white X-ray beam for µ-CT and a monochromatic X-ray beam (energy 88 keV, energy bandwidth 0.1 keV) for the XRD measurements [13]. The sample and the furnace needed to be aligned in both X-ray beams, and during the entire HT cycle, they were continuously moved from the white beam to the monochromatic beam for the alternating XRD and µ-CT measurements. After the installation of the new ID15 insertion device in 2008, the flux density of high-energy monochromatic photons has been further increased such that XRD and fast µ-CT can now be performed both with the same monochromatic photon beam.

The goal of this chapter is to illustrate the potential of high-energy synchrotron radiation experiments for in situ studies of the processing of superconductors. We present case studies describing the Nb3Sn wire diffusion HT, the transformation HT of Nb3Al precursor wires and the melt processing HT of Bi-2212 wires.

G1.2.2  Nb3Sn Diffusion HT

G1.2.2.1  Phase Transformations during the Diffusion HT of Nb3Sn Superconductors

The Nb3Sn phase in multifilament wires is produced during a diffusion HT, where the precursor elements Nb and Sn interdiffuse with the Cu matrix, forming various intermetallic phases and finally the superconducting Nb3Sn [14]. The intermediate phase transformations can degrade the microstructural and microchemical homogeneity of the fully reacted superconductor. This is most easily observed in the tubular strand types (powder-in-tube [PIT] [15] and tube type [16]), where typically 25% of the Nb3Sn volume consists of coarse grains that are not well connected and cannot conduct significant supercurrents.

High-energy synchrotron X-ray diffraction is an excellent tool to monitor the phase changes in superconducting wires in situ during the processing HT [17, 18]. As an example, Sn, Cu6Sn5, NbSn2, Nb6Sn5, Nb3Sn and the ternary phase (Nb0.75Cu0.25)Sn2 [19] can be identified in the diffraction pattern that have been acquired during Nb3Sn PIT wire HT (Figure G1.2.3). The respective temperature intervals where these phases are present are easily revealed in the sequence of diffractograms.

Summary of the diffraction patterns acquired

Figure G1.2.3   Summary of the diffraction patterns acquired in situ during the Nb3Sn PIT wire reaction HT. (© IOP Publishing. Reproduced with permission. All rights reserved.)

In situ XRD measurements during the reaction HT of restacked rod process (RRP) type [20] and tube type [21] Nb3Sn wires revealed a similar phase sequence. In particular, (Nb0.75Cu0.25)Sn2, NbSn2 and Nb6Sn5 are detected in the high Jc strands. In the RRP type wire, the amount of these phases is comparatively small, which presumably explains the relatively small volume fraction of Nb3Sn coarse grains in the fully processed RRP wire. During the processing HT of a low Sn content internal tin wire [22], a markedly different phase sequence is observed, and in particular, the phases (Nb0.75Cu0.25)Sn2, NbSn2 and Nb6Sn5 are not formed [7] (see G1.2.5).

G1.2.2.2  Nb3Sn Nucleation and Growth

Nb3Sn nucleation and growth in multifilament wires is accompanied by changes of the Sn content distribution, and the Nb3Sn grain size distribution. These microstructure and composition changes, which have a strong influence on Jc [23, 24], can be followed in situ by high-energy synchrotron XRD measurements.

By monitoring the Nb3Sn diffraction peak area evolution, the Nb3Sn formation kinetics in different wires can be compared. In an internal tin wire with low Sn content, the Nb3Sn phase growth follows a parabolic law [17], indicating that in this wire Nb3Sn growth is diffusion controlled. This is in contrast to the Nb3Sn phase growth in state-of-the-art high Sn content RRP- and PIT-type wires, where Nb3Sn growth is not purely diffusion controlled [17, 20].

Figure G1.2.4 compares the Nb3Sn growth in a RRP wire with 80 µm subelement size with that in PIT wires with 30 µm and 50 µm subelement size. All wires followed the same HT cycle with 100°C/h heating rate and three isothermal steps (4 h-700°C, 1 h-800°C and 1 h-900°C). Usually the processing peak temperature of Nb3Sn superconductors does not exceed 700°C. Here the 800°C and 900°C plateaus were added to explore the full reaction within a duration that allows to perform in situ synchrotron experiments. It is assumed that in the three wires the maximum possible amount of Nb3Sn was formed during the HT. Therefore, the maximum Nb3Sn peak areas measured during the HT cycles can be normalised, and the Nb3Sn growth kinetics in the wires with different elemental composition and architecture can be compared.

Evolution of the integrated intensity of prominent
                                    Nb

Figure G1.2.4   Evolution of the integrated intensity of prominent Nb3Sn peaks in the PIT Ø=0.8 mm, PIT Ø=1.25 mm and RRP Ø=0.8 mm wires during identical HT. (Courtesy J. Kadar.)

In the RRP wire with 80 µm subelement size, Nb3Sn is already detected at about 540°C, and about 80% of the maximum possible Nb3Sn volume is formed after the 4 h-700°C plateau. In the PIT wires, Nb3Sn is first detected during the 700°C plateau, and the Nb3Sn formation kinetics and the duration needed to transform Nb6Sn5 entirely into Nb3Sn are significantly influenced by the subelement size.

Keeping a small grain size for flux pinning and at the same time have maximum Nb3Sn volume and Sn content are conflicting needs of the HT. In ideally homogeneous wires, the Nb3Sn grain size and Sn content evolution can be monitored simultaneously with the Nb3Sn volume by XRD measurements. The Sn content can be determined from Nb3Sn lattice parameter measurements [25]. Assuming that the Nb3Sn grains nucleate and grow in a nearly stress-free state, the decrease of diffraction peak width after deconvolution of the instrument function is associated with the increase of grain size, and the use of the Scherrer formula allows for a rough calculation of the mean crystallite size [26]. In order to monitor crystallite sizes up to 200 nm, the diffraction experiment needs to be optimised for minimising instrumental peak broadening.

Since the RRP wire has a comparatively homogeneous Nb3Sn microstructure, it has been selected for the Nb3Sn nucleation and growth in situ study [27]. Figure G1.2.5 compares the changes of the Nb3Sn volume with the average crystallite size evolution (a) and the average Sn content (b) during 100°C/h HT with the three isothermal steps at 700°C, 800°C and 900°C. The average crystal size is a rough estimate, and only relative changes are meaningful.

Average Nb

Figure G1.2.5   Average Nb3Sn crystallite size and volume from (a) Nb3Sn (200) reflection and (b) Nb3Sn lattice parameter as a function of temperature. The relative lattice parameter variation induced solely by thermal expansion in the temperature interval 540°C–900°C is shown for comparison. (Reproduced with permission from Appl. Phys. Lett. 99, 122508. Copyright 2011, AIP Publishing LLC.)

At the onset of detectability (at 540°C), the mean Nb3Sn crystallite size estimated from the Nb3Sn (200) peak width is about 60 nm. During the 700°C plateau, the average crystallite size increases from 110 nm to about 180 nm. At the same time, the Nb3Sn volume increases by about 35%, and the Nb3Sn lattice parameter increases from 5.3140 Å to 5.3156 Å. This indicates that the average Sn content increases by more than 1%, which in turn corresponds to a strong critical field B c20 increase of 5 T [23]. At high magnetic field, such a strong B c2 increase outweighs the reduction of flux pinning force due to the simultaneous Nb3Sn grains growth. Further increase in temperature and HT duration only slightly increases the Nb3Sn volume but has a detrimental influence on the Nb3Sn crystallite size, which results in limited Jc when flux pinning has a dominating influence.

G1.2.2.3  Nb 3 Sn Texture Formation

Static texture analysis of bulk materials can be performed best by neutron diffraction measurements because of the relatively low neutron absorption. Synchrotron XRD in transmission geometry using an area detector can be very fast and is therefore better suited for monitoring texture formation in situ during processing HT [28]. Diffraction images of different Nb3Sn wires acquired with a Trixell Pixium 4700 two-dimensional flat-panel digital detector are compared in Figure G1.2.6. Preferential crystallite orientation is revealed by intensity fluctuations along the Nb and Nb3Sn diffraction rings [8].

Diffraction pattern of a bronze route, PIT and RRP
                                    Nb

Figure G1.2.6   Diffraction pattern of a bronze route, PIT and RRP Nb3Sn wire. The dashed arrows indicate the positions of intensity maxima in the Nb3Sn (200) Debye rings. (© IOP Publishing. Reproduced with permission. All rights reserved.)

The homogeneous intensity along the Nb3Sn rings of the BR wire shows that the BR process produces randomly oriented Nb3Sn crystallites. In contrast, intensity maxima are seen in the Nb3Sn (200) rings of the RRP- and PIT-type wires, but the intensity maxima are in different positions. Further texture analysis by electron backscatter diffraction (EBSD) revealed an Nb3Sn <110> texture in the PIT-type wire, whilst in the RRP type wire Nb3Sn grows with a <100> texture in the wire axis direction [29]. EBSD also confirmed a strong <110> Nb texture parallel to the wire axis, as it is commonly observed in cold-drawn body-centred cubic (BCC) Nb.

G1.2.2.4  Void Growth Mechanisms in Nb3Sn Superconductors

The presence of porosity in superconductors is often unavoidable, and the fabrication route can have a strong influence on the porosity volume and the distribution of voids that remains in the fully processed superconductor. Porosity generally reduces the useful superconductor volume in the composite, and in some cases, it may degrade the irreversible strain limit of brittle superconductors. If the porosity is distributed inside the superconducting phase, it can block the supercurrent.

The visualisation and quantitative description of the distribution of voids in the superconductor can help to better understand the porosity formation and redistribution mechanisms, and how porosity influences the superconducting properties. When the void shape and distribution are irregular, two-dimensional metallographic observations of void formation can be erratic and misleading. In contrast, X-ray µ-CT can provide non-destructively three-dimensional (3D) quantitative information about the porosity and particle size distribution.

MAt modern synchrotrons, tomograms can be acquired in less than 1 minute, which enables time-resolved in situ µ-CT studies of entire processes. Figure G1.2.7 shows a sequence of tomograms that was acquired in situ during the processing HT of a low Sn content internal tin Nb3Sn wire [22] with a ramp rate of 60°C/h, using the tomography furnace shown in Figure G1.2.2(a). In order to obtain a 3D view of the porosity inside the wire, the strand materials have been transparently depictured in the image reconstructions.

3D view of the porosity inside an internal tin Nb

Figure G1.2.7   3D view of the porosity inside an internal tin Nb3Sn wire acquired in situ by synchrotron µ-CT at different temperatures. (Reproduced with permission from Appl. Phys. Lett. 90, 132510. Copyright 2007, AIP Publishing LLC.)

In Figure G1.2.8, the phase evolution during the HT, based on diffraction peak area measurements, is compared with the porosity volume evolution, which is determined from the simultaneously acquired tomograms (Figure G1.2.7).

Evolution of prominent diffraction peak areas of all
                                 Sn-containing phases, apart from α-bronze, that exists in the IT
                                    Nb

Figure G1.2.8   Evolution of prominent diffraction peak areas of all Sn-containing phases, apart from α-bronze, that exists in the IT Nb3Sn strand during the reaction HT up to 540°C. Diffraction peak areas have been scaled such that the values correspond with the relative phase volume in the wire. The liquid Sn evolution is estimated from the amount of the detected phases. The total void volume is shown for comparison. (Reproduced with permission from Appl. Phys. Lett. 90, 132510. Copyright 2007, AIP Publishing LLC.)

The phase evolution during the HT of the low Sn content internal tin wire differs strongly from that of high Sn content PIT- and RRP-type wires (Figure G1.2.3). In particular, the Nb-containing phases NbSn2, (Nb0.75Cu0.25)Sn2 and Nb6Sn5 are not formed in the low Sn content wire. The analysis of the simultaneously acquired µ-CT and XRD results allows to distinguish between different void formation mechanisms.

The growth of the globular voids up to a temperature of about 200°C is driven by a gain in free energy through a reduction of the total void surface area when smaller voids present in the as-drawn wire agglomerate to larger globular voids. At 200°C, the maximum ratio of void volume to void surface area is obtained. At this temperature the total void volume corresponds to 2.5% of the pure Sn volume in the as-drawn wire. The correlation between void volume and Cu3Sn content, which is obvious in Figure G1.2.8, is due to the 4% higher density of Cu3Sn with respect to the Cu and Sn in their stoichiometric quantities.

G1.2.3  Transformation HT of Rapidly Quenched Nb3Al Precursor

The Nb3Al phase in superconducting wires is produced during a Rapid heating, quenching and transformation (RHQT) process [30]. During a HT at roughly 1900°C, a Nb(Al)SS solid solution is obtained, which can be retained during rapid quenching to ambient temperature. The rapid heating and quenching (RHQ) stages are followed by a transformation HT with a peak temperature of typically 800°C, during which fine-grained Nb3Al with high Al content is formed from the Nb(Al)SS solid solution.

The phase evolution during this transformation HT can be studied in situ by high-energy synchrotron X-ray diffraction [31]. The two-dimensional diffraction pattern acquired in transmission geometry with an area detector can be caked into sectors, in order to distinguish between reflections from the crystalline planes oriented both perpendicular and parallel to the wire drawing axis, which are in the following referred to as the axial and transverse directions, respectively.

The pattern presented in [31] have been caked into 128 sectors. The 222 filament Nb3Al precursor wire without Cu stabiliser that was studied has a partial interfilamentary Ta matrix. Since the strain-free Ta and Nb lattice parameters at RT differ by about 0.03% only, these phases could not be distinguished by their lattice spacing. Therefore, in the following, Nb diffraction peak refers to both overlapping peaks of Nb and Ta.

The axial and transverse Nb (110) diffraction peaks of the RHQ wire are presented in Figure G1.2.9. The axial Nb (110) peak is about eight times more intense than the transverse peak, which shows that the Nb (and/or Ta) texture, which is developed during the cold drawing of BCC metals, is partially retained during the RHQ process. The axial and transverse Nb peaks exhibit two maxima, which are characteristic for pure Nb and Ta (larger d-spacing) and Nb(Al)ss supersaturated solid solution (with roughly 1% smaller d-spacing).

Axial and transverse Nb(110) diffraction peak, consisting of two
                              components characteristic for pure Nb and for Nb(Al)SS.

Figure G1.2.9   Axial and transverse Nb(110) diffraction peak, consisting of two components characteristic for pure Nb and for Nb(Al)SS.

The evolution of the Nb (110), Nb3Al (200) and Nb3Al (211) diffraction peak shape and intensity during the RHQ Nb3Al precursor wire transformation HT with a ramp rate of 800°C/h and a final 800°C plateau lasting 30 minutes can be seen in Figure 5 of Reference [31]. The Nb (110) peak shape change is caused by the vanishing of the Nb(Al)SS peak component, upon formation of Nb3Al. When heating with a ramp rate of 800°C/h, Nb3Al (200) and Nb3Al (211) peaks are detected at about 780°C. When heating with a ramp rate of 160°C/h, the transformation from a Nb(Al)SS supersaturated solid solution into Nb3Al occurs at roughly 60°C lower temperature than during the 800°C/h HT [31].

G1.2.4  Bi-2212 Wire Melt Processing

G1.2.4.1  Phase Evolution during Bi-2212 Wire Melt Processing

In order to form well-connected and textured Bi-2212 filaments, the Bi-2212 precursor particles in the as-drawn Bi-2212 PIT wire need to be melted when the wire is at its final size and shape [32]. During the melt processing HT, an external oxygen supply through the oxygen-permeable Ag wire matrix is needed in order to re-form Bi-2212 out of the melt. The phase evolution during the melt processing HT can be studied in situ by high-energy synchrotron XRD measurements. Oxygen can be supplied conveniently in a flow of air at ambient pressure, using the X-ray transparent furnace shown in Figure G1.2.2(a).

The sequence of diffraction pattern acquired during the melt processing of a state-of-the-art Bi-2212 PIT wire in air (oxygen partial pressure pO2=0.21 bar) is presented in Figure G1.2.10 [33]. An initial Bi-2212 diffraction peak growth with increasing temperature is observed, which is attributed to crystallisation of Bi-2212 that was amorphised during the wire drawing process. The main impurity phase Bi-2201 is first detected when the temperature exceeds approximately 200°C, and a maximum amount of Bi-2201 is detected at about 500°C. Bi-2201 decomposes completely at 850°C, and reforms again upon cooling at approximately 850°C. The diffraction peaks that occur upon Bi-2212 melting around 880°C in a pO2=0.21 bar process gas have been tentatively identified as Cu-free phase Bi2(Sr4-yCay)O7.

Sequence of XRD patterns acquired during Bi-2212 wire HT in
                                 ambient air. The diffraction peaks which are labelled with arrows
                                 have been tentatively identified as the Cu-free phase
                                    Bi

Figure G1.2.10   Sequence of XRD patterns acquired during Bi-2212 wire HT in ambient air. The diffraction peaks which are labelled with arrows have been tentatively identified as the Cu-free phase Bi2(Sr4-yCay)O7). (© IOP Publishing. Reproduced with permission. All rights reserved.)

Overpressure (OP) processing at pressures of up to 100 bar is a key for achieving homogeneous high critical currents in long lengths of Bi-2212 wires [34]. OP processing also enables varying the oxygen partial pressure in a wide range, and it is of interest to verify how pO2 influences the phase sequence and the Bi-2212 precursor melting and recrystallisation behaviours.

In order to study the influence of pO2 on the Bi-2212 phase stability inside the Bi-2212/Ag wire by in situ high-energy synchrotron XRD measurements, the high-pressure cell and capillary furnace shown in Figure G1.2.2(b, c) have been used. This furnace allows to explore pO2 above ambient pressure, with total process gas pressures up to 200 bar. Another advantage of this furnace is that 5-cm-long wire samples with closed ends identical to the samples typically used for Bi-2212 critical current measurements can be studied.

The diffraction pattern acquired in situ during HTs at different pO2 shows that increasing pO2 reduces the Bi-2212 stability [6]. At pO2=1.5 bar, Bi-2212 decomposes partly prior to melting, and the precursor decomposition temperature is about 20°C lower than it is at pO2=1.05 bar. At pO2=5 bar, the Bi-2212 precursor particles in the state-of-the art Bi-2212 multifilament wire decomposes completely in the solid state.

G1.2.4.2  Void Formation and Redistribution during Bi-2212 Wire Melt Processing

Porosity and second phase particles formed during melt processing are considered to be the main current limiting defects in Bi-2212 wires. It is therefore of great interest to visualise and to quantify the porosity and second phase distribution during the different processing steps.

The potential of µ-CT to visualise these features inside a superconducting wire depends equally on the spatial and density resolution of the µ-CT experiment. The calculated linear absorption coefficients of 70 keV photons in the main wire constituents Ag and Bi-2212, and the main impurity phase Bi-2201, are µAg=40 cm−1, μBi−2212=15 cm−1 and µBi−2201= 18 cm−1, respectively. Because of the different X-ray attenuation in Ag, Bi-2212 and porosity, high-energy synchrotron µ-CT is particularly well suited to monitor Bi-2212 microstructure changes and the porosity formation and redistribution inside Bi-2212/Ag wires [33]. On the other hand, because of the relatively small difference of the X-ray attenuation in Bi-2212 and Bi-2201, these phases cannot be distinguished in the X-ray absorption tomograms.

The void redistribution during the melt processing of a 37 × 7 filament Bi-2212 wire at ambient pressure can be followed in the longitudinal µ-CT cross-sections shown in Figure G1.2.11, which have been acquired in situ at different temperatures. In order to show a more detailed view of the voids, the images have been cropped from the longitudinal cross-sections showing the entire wire cross-section. The black areas represent voids, the bright grey areas the strongly absorbing Ag matrix, and the dark grey areas are Bi-2212 with a small amount of Bi-2201.

Detailed view of Bi-2212 wire longitudinal tomographic
                                 cross-sections acquired

Figure G1.2.11   Detailed view of Bi-2212 wire longitudinal tomographic cross-sections acquired in situ at different temperatures during HT to T max=915°C in air. A time lapse movie showing the changes occurring over the whole heating and cooling cycle is available (https://edms.cern.ch/document/1153082/1). (© IOP Publishing. Reproduced with permission. All rights reserved.)

Filament microstructure changes can be first observed at about 850°C when the Bi-2201 impurity phase decomposes (as seen in the simultaneously acquired XRD pattern). At this temperature, the finely divided porosity, which is in the as-drawn wire uniformly distributed between the precursor particles, coalesces into lens-shaped defects. On Bi-2212 melting, the lens-shaped voids grow to bubbles of a filament diameter.

Upon cooling, nucleation of Bi-2212 is first observed in the tomogram acquired at 877°C at the filament periphery. The Bi-2212 formed upon cooling partly bridges the void space, but bubbles remain and cause an obstacle to the current flow in the Bi-2212 wires that are melt processed at ambient pressure [35].

The importance of complete Bi-2212 precursor melting is obvious when comparing the longitudinal µ-CT cross-section acquired at the end of a processing HT to T max=915°C, during which Bi-2212 was completely melted [Figure G.1.1.2.12(a)], and to T max=875°C, in which only a fraction of the Bi-2212 powder was melted [G.1.1.2.12(b)]. The tomograms show clearly that after the T max=875°C HT, the filaments remain interrupted by a regular array of lens-shaped voids, and that filament connectivity is only achieved after the porosity rearrangement that occurs during complete Bi-2212 melting and recrystallisation.

Tomographic cross-sections of the Bi-2212 wire acquired after

Figure G1.2.12   Tomographic cross-sections of the Bi-2212 wire acquired after in situ HT to (a) T max=875°C and (b) T max=915°C. 3D reconstructed images of selected filaments are shown in the insets. (© IOP Publishing. Reproduced with permission. All rights reserved.)

The overall porosity volume in Bi-2212 wires that are short enough to allow relief of internal pressure through the open ends does not strongly change, because the Bi-2212 processing does not involve phase transformations associated with important density variations, as it is, for instance, the case in Nb3Sn conductors [7]. In long Bi-2212 wires and in wires with closed ends, additional porosity is formed during processing at ambient pressure when internal gas pressure leads to creep of the Ag matrix [36]. OP pressing strongly reduces the porosity volume that is present in the as-drawn wire [34].

G1.2.5  Outlook

Today the time-resolved combined XRD and µ-CT experiments for in situ studies of superconductors that are described above can be routinely performed at advanced high-energy synchrotron beamlines.

Thanks to the fast data acquisition at modern synchrotron beamlines, XRD and u-CT results can be compared with those obtained with identical ramp rates by other in situ techniques like differential scanning calorimetry (DSC) to achieve an even deeper understanding of the endothermic and exothermic phase transformations [37].

The continuously improving brilliance of synchrotron sources and new efficient X-ray focusing optics makes it possible to use nanometer scale X-ray beams, enabling new non-destructive in situ experiments on length scales that so far were only accessible to destructive techniques [38].

Grain size and grain orientation have a dominant influence on the performance of most superconductors, and studies of the thermal growth of grains and the grain orientation evolution are examples where future superconductor research can profit from new synchrotron experiments with X-ray nanobeams. Such studies can be performed in two dimensions, averaging over the sample depth that is penetrated by the X-ray beam. When applying tomographic methods (e.g. XRD-tomography [39]) spatially resolved in situ studies of phase composition, crystallite size distribution and texture become possible.

Acknowledgments

All XRD and µ-CT experiments presented here have been performed at the ESRF ID15 beamline. We are grateful to Julian Kadar for the Nb3Sn diffraction peak analysis of Figure G1.2.4.

References

1
T.R. Thurston , P. Haldar , Y.L. Wang , M. Suenaga , N.M. Jisraw , U. Wildgruber , “In situ measurements of texture and phase development in (Bi, Pb)2Sr2Ca2Cu3O10–Ag tapes”, J. Mater. Res., 12(4), (1997)
2
D.K. Finnemore , “X-ray studies: phase transformation and texture”, in Handbook of Superconducting Materials 1st edition, edited David A. Cardwell , David S. Ginley , (2003)
3
C. Scheuerlein , M. Di Michiel , F. Buta , “Synchrotron radiation techniques for the characterisation of Nb3Sn superconductors”, IEEE Trans. Appl. Supercond., 19(3), (2009), 2653–2656
4
C. Scheuerlein , U. Stuhr , L. Thilly , “In-situ neutron diffraction under tensile loading of powder-in- tube Cu/Nb3Sn composite wires: effect of reaction heat treatment on texture, internal stress state and load transfer”, Appl. Phys. Lett., 91(4), (2007), 042503
5
J. Andrieux , C. Chabert , A. Mauro , H. Vitoux , B. Gorges , T. Buslaps , V. Honkimäki , “A high pressure and high temperature gas loading system for the study of conventional to real industrial sized samples in catalyzed gas/solid and liquid/solid reactions”, J. Appl. Cryst., 47, doi:10.1107/S1600576713030197, (2014)
6
C. Scheuerlein , J. Andrieux , M.O. Rikel , J. Kadar , C. Doerrer , M. Di Michiel , A. Ballarino , L. Bottura , J. Jiang , T. Kametani , E.E. Hellstrom , D.C. Larbalestier , “Influence of the oxygen partial pressure on the phase evolution during Bi-2212 wire melt processing”, IEEE Trans. Appl. Supercond., submitted
7
L. Muzzi et al, “Direct observation of Nb3Sn lattice deformation by high energy xray diffraction in internal tin wires subject to mechanical loads at 4.2 K”, Supercond. Sci. Technol., 25, (2012), 054006
8
C. Scheuerlein , M. Di Michiel , F. Buta , B. Seeber , C. Senatore , R. Flükiger , T. Siegrist , T. Besara , J. Kadar , B. Bordini , A. Ballarino , L. Bottura , “Stress distribution and lattice distortions in Nb3Sn/Cu multifilament wires under uniaxial tensile loading at 4.2 K”, Supercond. Sci. Technol., 27, (2014), 044021,
9
R. Bjoerstad , C. Scheuerlein , M. Rikel , A. Ballarino , L. Bottura , J. Jiang , M. Matras , M. Sugano , J. Hudspeth , M. Di Michiel , “Strain induced irreversible critical current degradation in highly dense Bi-2212 round wire”, Supercond. Sci. Technol., 28, (2015), 062002
10
M. Di Michiel , J. M. Merino , D. Fernandez-Carreiras , T. Buslaps , V. Honkimäki , P. Falus , T. Martins , O. Svensson , “Fast microtomography using high energy synchrotron radiation”, Rev. Sci. Instrum., 76, (2005), 043702
11
J. Y. Buffiere , E. Maire , J. Adrien , J. P. Masse , E. Boller , “In situ experiments with X ray tomography: an attractive tool for experimental mechanics”, Experimental Mechanics, 50, (2010), 289–305,
12
C. Scheuerlein , M. Di Michiel , A. Haibel , “On the formation of voids in Nb3Sn superconductors”, Appl. Phys. Lett., 90, (2007), 132510
13
“Void formation in Nb3Sn Superconductors”, http://www.esrf.eu/Apache_files/Highlights/HL2007.pdf, (2007), 27–28
14
M.T. Naus , P.J. Lee , D. C. Larbalestier , “The interdiffusion of Cu and Sn in internal Sn Nb3Sn superconductors”, IEEE Trans. Appl. Supercond., 10(1), (2000) 983–987
15
H. Veringa , E. M. Hornsveld , and P. Hoogendam , Adv. Cryo. Eng., vol. 30, (1984), 813–821
16
E. Gregory , X. Peng , M. Tomsic , M. D. Sumption , A. Ghosh , “Nb3Sn superconductors made by an economical tubular process”, IEEE Trans. Appl. Supercond., 19(3), (2009), 2602–2605
17
M. Di Michiel , C. Scheuerlein , “Phase transformations during the reaction heat treatment of powder-in-tube Nb3Sn superconductors”, Supercond. Sci. Technol., 20, (2007), L55–L58
18
A.B. Abrahamsen , J.C. Grivel , N.H. Andersen , M. Herrmann , W. Häßler , K. Saksl , “In-situ synchrotron x-ray study of MgB2 formation when doped by SiC”, J. Phys. Conf. Ser., 97 (2008), 012315
19
S. Martin , A. Walnsch , G. Nolze , A. Leineweber , F. Léaux , C. Scheuerlein , “The crystal structure of (Nb0.75Cu0.25)Sn2 in the Cu-Nb-Sn system”, Intermetallics 80, (2017), 16–21
20
C. Scheuerlein , M. Di Michiel , G. Arnau , F. Buta , “Phase transformations during the reaction heat treatment of Internal Tin Nb3Sn strands with high Sn content”, IEEE Trans. Appl. Supercond., 18(4), (2008), 1754–1760
21
C. Scheuerlein , M. Di Michiel , L. Thilly , F. Buta , X. Peng , E. Gregory , J.A. Parrell , I. Pong , B. Bordini , M. Cantoni , “Phase transformations during the reaction heat treatment of Nb3Sn superconductors”, J. Phys. Conf. Ser., 234, (2010), 022032
22
M. Durante , P. Bredy , A. Devred , R. Otmani , M. Reytier , T. Schild , F. Trillaud , “Development of a Nb3Sn multifilamentary wire for accelerator magnet applications“, Physica C, 354, (2001), 449–453
23
R. Flükiger , D. Uglietti , C. Senatore , F. Buta , “Microstructure, composition and critical current density of superconducting Nb3Sn wires”, Cryogenics, 48, (2008), 293–307
24
P.J. Lee , D.C. Larbalestier , “Microstructural factors important for the development of high critical current density Nb3Sn strand”, Cryogenics, 48, (2008), 283–292
25
H. Devantay , J. L. Jorda , M. Decroux , J. Müller , R. Flükiger , J. Mater. Sci., 16, (1981), 2145
26
J.I. Langford , D. Louër , P. Scardi , “Effect of a crystallite size distribution on X-ray diffraction line profiles and whole-powder-pattern fitting”, J. Appl. Cryst., 33, (2000), 964–974
27
L. Thilly , M. Di Michiel , C. Scheuerlein , B. Bordini , “Nb3Sn nucleation and growth in multifilament superconducting strands monitored by high resolution synchrotron diffraction during in- situ reaction”, Appl. Phys. Lett., 99, (2011), 122508
28
H.R. Wenk , S. Grigull , “Synchrotron texture analysis with area detectors”, J. Appl. Cryst., 36, (2003), 1040–1049
29
C. Scheuerlein , G. Arnau , P. Alknes , N. Jimenez , B. Bordini , A. Ballarino , M. Di Michiel , L. Thilly , T. Besara , T. Siegrist , “Texture in state-of-the-art Nb3Sn multifilamentary superconducting wires“, Supercond. Sci. Technol., 27 (2014) 025013
30
T. Takeuchi , “Nb3Al Conductors –Rapid Heating, Quenching and Transformation Process”, IEEE. Trans. Appl. Supercond., 10(1), (2000), 1016–1021
31
C. Scheuerlein , A. Ballarino , M. Di Michiel , X. Jin , T. Takeuchi , A. Kikuchi , K. Tsuchiya , K. Nakagawa , T. Nakamoto , “Transformation heat treatment of rapidly quenched Nb3Al precursor monitored in situ by high energy synchrotron diffraction“, IEEE Trans. Appl. Supercond., 23(3), 6000604, (2013)
32
K. Heine , J. Tenbrink M. Thoner , “High field critical current densities in Bi2Sr2CaCu2O8-/Ag wires”, Appl. Phys. Lett. 55, (1989), 2441–2443
33
C. Scheuerlein , M. Di Michiel , M. Scheel , J. Jiang , F. Kametani , A. Malagoli , E.E. Hellstrom , D.C. Larbalestier , “Void and phase evolution during the processing of Bi-2212 superconducting wires monitored by combined fast synchrotron micro-tomography and X-ray diffraction”, Supercond. Sci. Technol., 24 (2011), 115004
34
D.C. Larbalestier , J. Jiang , U.P. Trociewitz , F. Kametani , C. Scheuerlein , M. Dalban-Canassy , M. Matras , P. Chen , N.C. Craig , P.J. Lee , E.E. Hellstrom , “Isotropic round wire multifilament cuprate superconductor for generation of magnetic fields above 30 T”, Nature Mater., 13, (2014), 375–381
35
F. Kametani , T. Shen , J. Jiang , A. Malagoli , C. Scheuerlein , M. Di Michiel , Y. Huang , H. Miao , J.A. Parrell , E.E. Hellstrom , D.C. Larbalestier , Supercond. Sci. Technol., 24, (2011), 075009
36
A. Malagoli , P.J. Lee , A. Ghosh , C. Scheuerlein , M. Di Michiel , J. Jiang , U.P. Trociewitz , E.E. Hellstrom , D.C. Larbalestier , “Evidence of length-dependent wire expansion, filament dedensification and consequent degradation of critical current density in Ag-alloy sheathed Bi-2212 wires”, Supercond. Sci. Technol., 26, (2013), 055018
37
C. Scheuerlein , J. Andrieux , M. Michels , F. Lackner , C. Meyer , R. Chiriac , F. Toche , M. Hagner , D. Michiel , “Effect of the fabrication route on the phase and volume changes during the reaction heat treatment of Nb3Sn superconductors”, Supercond. Sci. Technol. 33, (2020), 034004
38
G.E. Ice , J.D. Budai , J.W.L. Pang , “The race to X-ray microbeam and nanobeam science”, Science, 334, (2011), 1234–1239
39
S.D.M. Jacques , M. Di Michiel , A.M. Beale , T. Sochi , M.G. O'Brien , L. Espinosa-Alonso , B.M. Weckhuysen , P. Barnes , “Dynamic X-ray diffraction computed tomography reveals real- time insight into catalyst active phase evolution”, Angewandte Chemie International Edition, 50(43), 2011, 10148–10152
Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.