# Fileset

[Takeda_2023_Supercond._Sci._Technol._36_035004.pdf](https://mdr.nims.go.jp/filesets/17dfd594-bd2b-451d-ad6c-8262988e9ba2/download)

## Creator

[Y Takeda](https://orcid.org/0000-0001-7217-9853), [G Nishijima](https://orcid.org/0000-0001-7493-0559), K Inoue, [Y Takano](https://orcid.org/0000-0002-1541-6928), [H Kitaguchi](https://orcid.org/0000-0002-5998-2649)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[The effect of intermediate layer densification on the critical current of a Bi-2223 superconducting joint](https://mdr.nims.go.jp/datasets/46ef4499-8087-49f6-bc26-7141379a5d7a)

## Fulltext

The effect of intermediate layer densification on the critical current of a Bi-2223 superconducting jointSuperconductor Science and TechnologyPAPER • OPEN ACCESSThe effect of intermediate layer densification onthe critical current of a Bi-2223 superconductingjointTo cite this article: Y Takeda et al 2023 Supercond. Sci. Technol. 36 035004 View the article online for updates and enhancements.You may also likeDevelopment of a persistent-mode NMRmagnet with superconducting jointsbetween high-temperaturesuperconductorsY Yanagisawa, R Piao, Y Suetomi et al.-Review of recent developments in ultra-high field (UHF) NMR magnets in the AsiaregionY Yanagisawa, M Hamada, K Hashi et al.-Combination of high hoop stress toleranceand a small screening current-inducedfield for an advanced Bi-2223 conductorcoil at 4.2 K in an external fieldY Yanagisawa, Y Xu, S Iguchi et al.-This content was downloaded from IP address 144.213.253.16 on 24/01/2023 at 01:13https://doi.org/10.1088/1361-6668/acaccd/article/10.1088/1361-6668/ac2120/article/10.1088/1361-6668/ac2120/article/10.1088/1361-6668/ac2120/article/10.1088/1361-6668/ac2120/article/10.1088/1361-6668/ac5644/article/10.1088/1361-6668/ac5644/article/10.1088/1361-6668/ac5644/article/10.1088/0953-2048/28/12/125005/article/10.1088/0953-2048/28/12/125005/article/10.1088/0953-2048/28/12/125005/article/10.1088/0953-2048/28/12/125005Superconductor Science and TechnologySupercond. Sci. Technol. 36 (2023) 035004 (10pp) https://doi.org/10.1088/1361-6668/acaccdThe effect of intermediate layerdensification on the critical current of aBi-2223 superconducting jointY Takeda∗, G Nishijima, K Inoue, Y Takano and H KitaguchiNational Institute for Materials Science, 3-13 Sakura, Tsukuba, Ibaraki 305-0003, JapanE-mail: TAKEDA.Yasuaki@nims.go.jpReceived 16 September 2022, revised 11 December 2022Accepted for publication 19 December 2022Published 23 January 2023AbstractThe effect of intermediate layer densification on the critical current (Ic) of Bi-2223superconducting joints was quantitatively studied. First, we evaluated the phase purity, density,and intergrain critical current density (Jc) of Bi-2223 thick film samples simulating theintermediate layer. The samples were uniaxially pressed to increase the film density. After twoheat treatments of the pressed film, an increase in Jc was achieved. Second, we fabricatedsuperconducting joints by synthesizing an intermediate layer between two Bi-2223 tapes.Applying a uniform uniaxial pressure on the joint resulted in the formation of a homogeneousstructure. This process enables the reproducible fabrication of superconducting joints with highn values. The Ic of the superconducting joint was increased by intermediate pressing (IP) andtwo heat treatments. However, pressing at high pressures can mechanically damage filaments inthe Bi-2223 tapes, leading to a decrease in Ic. Sample characterization showed that the optimumIP pressure range to produce high Ic was 1.5–2 × 108 Pa. We confirmed that pressing densifiedthe intermediate layer of the superconducting joints. Our experimental results and analysesreveal that densification of the intermediate layer increases the Ic of Bi-2223 superconductingjoints.Keywords: superconducting joint, Bi-2223, HTS, critical current, densification(Some figures may appear in colour only in the online journal)1. IntroductionThe commercially available Ag-sheathed multifilamentary(Bi,Pb)2Sr2Ca2Cu3Oy [Bi-2223] high-temperature supercon-ducting (HTS) tape, DI-BSCCO® [1, 2], exhibits high criticalcurrent (Ic) up to 200 A at 77 K in self-field and >500 A at4.2 K under∼20 T parallel to the tape surface [3, 4]. This tapehas been used for magnets operating at a high temperature of∗Author to whom any correspondence should be addressed.Original content from this workmay be used under the termsof the Creative Commons Attribution 4.0 licence. Any fur-ther distribution of this work must maintain attribution to the author(s) and thetitle of the work, journal citation and DOI.20 K [2, 5, 6] and generating a high field of >24 T at around4 K [7–9].Superconducting joints are necessary for magnets operat-ing in the persistent current mode [10, 11]. Persistent currentmode magnets are used for magnetic resonance imaging andnuclear magnetic resonance systems because of their tempor-ally stable magnetic field. In the last decade, significant pro-gress on developing superconducting joints between HTS con-ductors has been made [11–18].We developed a high-Ic superconducting joint between Bi-2223 tapes [13] by synthesizing a Bi-2223 intermediate layerbetween the exposed filaments of the two tapes. The synthesiscomprises a slurry process, uniaxial pressing at room temper-ature, and heat treatment. It was found that the larger the num-ber of exposed filaments of the Bi-2223 tapes, the higher the1361-6668/23/035004+10$33.00 Printed in the UK 1 © 2023 The Author(s). Published by IOP Publishing Ltdhttps://doi.org/10.1088/1361-6668/acaccdhttps://orcid.org/0000-0001-7217-9853https://orcid.org/0000-0001-7493-0559mailto:TAKEDA.Yasuaki@nims.go.jphttp://crossmark.crossref.org/dialog/?doi=10.1088/1361-6668/acaccd&domain=pdf&date_stamp=2023-1-23https://creativecommons.org/licenses/by/4.0/Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alIc of the superconducting joint. To expose a large number offilaments, the Bi-2223 tapes were mechanically polished at asmall angle of less than 0.4◦.In a recent study [19], Ic was improved by applying two-step sintering. The process involves an initial heat treat-ment, intermediate uniaxial pressing, and a second heat treat-ment. The superconducting joint achieved a practical Ic of290–306 A at 4.2 K and 1 T under a resistance criterion of10−9 Ω.Although the Ic of the Bi-2223 superconducting joint hasbeen improved, the key parameters affecting Ic have not yetbeen clarified. To identify these parameters, it is necessary tounderstand the relationships between Ic and the microstruc-ture, density, and phase purity of the superconducting joint.The uniaxial pressing pressure on the Bi-2223 intermediatelayer significantly influences the microstructure and density,as demonstrated in bulk samples [20–22] and thick films [23].During the joining process we applied intermediate pressing(IP) pressure of about 200 MPa [19], which likely densifiedthe intermediate layer. However, the effect of densification hasnot yet been quantitatively evaluated.In this study, we clarify the effect of intermediate layerdensification achieved by IP on the Ic of the Bi-2223 super-conducting joint. Firstly, we evaluated the phase purity, dens-ity, and intergrain critical current density (Jc) of thick filmsamples. The relationship between density and intergrain Jcwas established. Secondly, the fabrication process of thesuperconducting joint was modified to achieve a uniform uni-axial pressure. The relationships between the IP pressure,density of the intermediate layer, and Ic of the superconduct-ing joints were determined.2. Relationship between density and Jc of thick filmsamples2.1. ExperimentalThick film samples simulating the intermediate layer of a Bi-2223 superconducting joint were prepared by a slurry process[23]. Figure 1 shows a schematic of the sample and flowchart of the sample preparation process. Commercially avail-able Bi-2223 precursor powder produced by TEP Co., Ltdwas used. The nominal chemical composition of the powderwas Bi:Pb:Sr:Ca:Cu = 1.6:0.45:1.9:2.0:3.0. The slurry wasobtained by mixing the precursor powder with EtOH/1-BuOHsolvent.The slurry was dropped onto a Ag sheet and dried in air. TheAg sheet was 15 mm long, 5 mm wide, and 0.2 mm thick. Thedried samples were uniaxially pressed at 2.0 × 108 Pa (about200 MPa). Sample TF1 was prepared at this stage as shownin figure 1(b). The other pressed films were each covered witha 30 µm thick Ag foil. We performed the first heat treatment(HT1) for these films. At this stage, sample TF2 was ready.Wecarried out the second heat treatment (HT2) for the remainingsamples after IP to prepare sample TF3–5. IP pressures (PIP)for samples TF3–5 are shown in table 1. HT conditions forboth HT1 and HT2 were at 815 ◦C for 36 h under a partialFigure 1. (a) Schematic of the thick film sample and (b) samplepreparation flow chart.oxygen pressure (PO2) of 3 kPa in a tube furnace. We con-trolled PO2 by flowing 3%O2/Ar gas in the furnace. The thick-ness of the oxide layer of the obtained samples was about0.1 mm. The Ag foil and sheet were removed as needed duringcharacterization.The density (ρ) of the samples was calculated using themass and dimensions of the rectangular oxide layer afterremoving the Ag foil and sheet. The theoretical density of6.3 g cm−3 [24] was used to calculate the relative density (D).θ/2θ surface x-ray diffraction measurements were per-formed using a Rigaku MiniFlex II. The measurements wereperformed on the surface of the oxide layer and the oxide/Aginterface. The molar fraction of the Bi-2223 phase (f Bi-2223)was estimated byfBi-2223 =IBi-2223IBi-2223 + IBi-2212(1)where IBi-2223 and IBi-2212 are the integrated intensities for the0014(Bi-2223) peak with 2θ of about 34◦ and 0012(Bi-2212[(Bi,Pb)2Sr2CaCu2Oy]) peak with 2θ of about 35◦ in the x-raydiffraction pattern.The intergrain Jc for the samples was evaluated at 77 K andzero external field by remanent magnetization measurements[23, 25] using a SQUID magnetometer (MPMS, QuantumDesign). We used the extended Bean model to calculate Jc.The magnetic field was applied normal to the surface of thesamples, i.e. parallel to the pressing direction.2Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alTable 1. Thick film sample specifications. Heat treatment conditions for both HT1 and HT2 were at a partial oxygen pressure (PO2) of 3 kPaand 815 ◦C for 36 h.Sample Heat treatment and IP PIP / 108 Pa f Bi-2223 / −(Surface) f Bi-2223 / −(Interface) ρ / g cm−3 D / %Intergrain Jc/ kA cm−2(77 K, zeroexternal field)TF1 — — 0.094 — 3.90 62 —TF2 HT1 — 0.769 0.853 3.62 57 0.25TF3 HT1 → IP → HT2 1.0 0.883 0.902 3.86 61 1.1TF4 HT1 → IP → HT2 1.5 0.932 0.886 4.02 64 1.4TF5 HT1 → IP → HT2 2.0 0.891 0.884 4.21 67 1.8Figure 2. Typical surface x-ray diffraction patterns of sample TF1–3. The Bi-2223 phase formation progressed significantly during HT1. Inthe oxide/Ag interface of TF2, Bi-2223 was the main phase. Bi-2223 was the main phase even on the oxide surface of TF3.2.2. ResultsFigure 2 shows typical surface x-ray diffraction patterns forthe samples. The main phase of TF1 was Bi-2212. There werealso small amounts of Bi-2223 and Ca2PbO4 phases. AlthoughCuO or (Ca,Sr)-Cu-O phases should also exist on the oxidesurface considering the nominal chemical composition, wecould not clearly find their peaks. This is probably because ofthe overlapping peaks, their scarce amount, or small crystallitesize.The Bi-2223 phase formation progressed significantly dur-ing HT1. The unreacted Bi-2212 phase remained on the oxidesurface of sample TF2, whereas Bi-2223 was the main phasein the oxide/Ag interface. It is well known that the presence ofAg aids in the formation and c-axis grain alignment of the Bi-2223 phase [26, 27]. The diffraction pattern in the interface ofTF2 indicates improved Bi-2223 phase formation and c-axisgrain alignment. Even on the surface of TF3, Bi-2223 was themain phase.Table 1 summarizes the f Bi-2223 values for the surface andinterface, ρ, D, and the intergrain Jc for sample TF1–5. ForTF3–5, f Bi-2223 values were not dependent on PIP and largerthan 0.88 on both the surface and interface. This indicates thatBi-2223 was the main phase for the entire sample after HT2.f Bi-2223 on the interface of TF2 was 0.853, which is larger thanthat on the oxide surface and comparable to that for TF3–5because of the presence of Ag.The ρ of TF1 was 3.90 g cm−3, which corresponds to aD of62%. This relatively low value was probably due to the insuf-ficient pressing pressure of 2.0 × 108 Pa. TF2 had a lower ρof 3.62 g cm−3. This is because the oxide layer swelled duringHT1 when the Bi-2223 phase was being formed [28, 29].In preliminary experiments, we obtained higher ρ of4.5 g cm−3 in a thick film by high-pressure pressing at1.0 × 109 Pa (about 1 GPa) before HT1. Due to the swelling,however, ρ of this thick film decreased to 3.7 g cm−3, which iscomparable to that of TF2. This indicates that it is difficult toimprove final ρ of thick films by high-pressure pressing beforeHT1.ρ increased after IP and HT2. IP contributed to the densi-fication, as demonstrated in a bulk sample [21]. In the presentwork, it was found that the higher the PIP, the higher the ρof the Bi-2223 thick film. Among all samples, the highest ρof 4.21 g cm−3 corresponding to a D of 67% was obtainedby TF5. This ρ value is comparable to that for bulk samplesprepared with a PIP of 1–2 × 108 Pa [21, 22]. However, thisρ value is still low compared with that of the filaments of theBi-2223 tape (D of about 100%) fabricated by rolling and over-pressure sintering processes [2].3Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alFigure 3. Relationship between ρ and intergrain Jc at 77 K and zeroexternal field for sample TF2–5. The gray dashed line was derivedfrom TF3–5 data points using the least-squares method. Jc increasedafter IP and HT2. In the thick film, an increase in intergrain Jc isachieved by densification.Figure 3 shows the relationship between ρ and intergrain Jcat 77 K and zero external field for TF2–5. The gray dashed linewas derived from TF3–5 data points using the least-squaresmethod. Jc increased after IP and HT2. This increase in Jc ismainly due to the high purification and densification. Thereappears to be a linear relationship between ρ and Jc in TF3–5.Therefore, it was verified that in the Bi-2223 thick film, anincrease in intergrain Jc is achieved by densification.3. Ic and microstructure evaluation forsuperconducting joint samples3.1. ExperimentalDI-BSCCO® Type H tapes 4.2 mm wide and 0.22 mm thick(without mechanical reinforcements) were used to fabricatethe Bi-2223 superconducting joints. Figure 4(a) shows thetransverse cross-sectional view and height profile of one sideof the Bi-2223 tape. This cross-section shows a bulge, i.e., thecenter is thicker than that of the edges. Height profile meas-urements in the lateral direction (x) using a laser microscope(Keyence VK-X1100/1000SP1976) showed that the height ofthe center (x = 2 mm) was about 20 µm larger than that of theedges.The uniaxial pressure on the joint will not be uniform usingsuch tapes, i.e., the high pressure is concentrated in the thickercentral part. This non-uniform pressure distribution is one ofthe major reasons why the densification demonstrated in thethick film could not be replicated in the intermediate layer ofa joint.To ensure that the applied uniaxial pressure is uniform, thetape was flattened by polishing both sides. Figure 4(b) showsthe transverse cross-sectional view and height profile of oneside of the flattened Bi-2223 tape. The thickness of the centerFigure 4. Transverse cross-sectional view and height profile of oneside in the lateral direction (x) for (a) as-purchased Bi-2223 tape and(b) flattened Bi-2223 tape. The gray dashed line in the heightprofiles is the reference level. The surface of the flattened tape isperfectly flat.of the flattened tape was 30–40 µm smaller than that of theas-purchased tape. Height profile measurements show that thesurface of the flattened tape is perfectly flat.In the present work, straight lap joint samples were fab-ricated using flattened Bi-2223 tapes 6 cm long. Figure 5(a)shows a schematic of the joint. One end of each flattened tapewas polished at 0.3◦ to expose most filaments [13, 19]. Thejoining procedure involving Bi-2223 intermediate layer syn-thesis was similar to that described in [13, 19] and was appliedusing the same slurry to prepare the thick film samples.We confirmed that the uniaxial pressure in pressing couldbe made uniform using the flattened tapes, as shown infigure 5(b). Joint samples using as-purchased tapes or flattenedtapes were prepared and pressed at 1.0–2.0 × 108 Pa withthe pressure measurement films (Fujifilm Prescale HS). Thesample using the as-purchased tapes showed a color gradi-ent at 1.0 × 108 Pa. This indicates that the high pressure isconcentrated in the thicker central part and the thinner edgesare not pressed. Even when pressing at 2.0× 108 Pa, althoughthe pressed area increased due to sample deformation, the thin-ner edges are still not pressed. In contrast, the joint using theflattened tapes showed no color gradient despite the pressingpressure. This indicates the uniform pressure distribution at4Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alFigure 5. (a) Schematic of the straight lap joint using the flattenedtapes. (b) Photograph of pressure measurement films (FujifilmPrescale HS). Joints using the as-purchased tapes or flattened tapeswere pressed at 1.0–2.0 × 108 Pa with the pressure measurementfilms. These films indicate that uniaxial pressure in pressing can bemade uniform by using the flattened tapes.1.0–2.0 × 108 Pa. The width of the region to which pressurewas applied almost completely corresponded to that of the tape(4.2 mm).The intermediate layer area was about 70 mm2. However,there was an area uncertainty of about 10%. This means thatthe uncertainty of the pressing pressure was also about 10%.The joint was uniaxially pressed at (2.0± 0.2)× 108 Pa beforeHT1. HT2 was performed for some samples after IP at a PIPof 1–2 × 108 Pa with ±10% uncertainty. HT conditions werethe same as those for the fabrication of the thick film samples,i.e., at PO2 of 3 kPa and 815 ◦C for 36 h.Transport measurements in the self-field were performedfor the samples in a liquid nitrogen bath (77 K). The conven-tional dc four-probe method was used. Ic was determined ata voltage of V = 0.2 µV. Assuming an empirical power lawmodel (V∝ In), the exponent n was calculated for a voltagerange of 0.2 µV ⩽ V ⩽ 0.7 µV.The microstructures of the samples were observed using afield emission scanning electron microscope, Hitachi SU-70.Secondary electron images were obtained from the polishedsurface of the transverse cross-section of the samples. Weestimated the filling factor (F) for the intermediate layer of thesamples by image analysis. More than ten images of the inter-mediate layer were obtained for each sample. The size of theimages was 25 µm× 18 µm. We estimated F values by imagethresholding, measuring F for each image, and calculating themean and standard deviation values.3.2. Results3.2.1. Ic and n values after HT1. Eight sample joints werefabricated. Table 2 summarizes the Ic, n and PIP for eachsample. Samples J1 and J2 were prepared using only HT1Table 2. Joint sample specifications. Ic and n values at 77 K inself-field were examined by transport measurements. The heattreatment conditions were at PO2 = 3 kPa and 815 ◦C for 36 h.SampleAfter HT1PIPa / 108 PaAfter HT2Ic / A n / – Ic / A n / –J1 37.3 14.1 — (only HT1)J2 31.6 15.2 — (only HT1)J3 33.8 18.1 1.0 40.3 17.6J4 42.6 13.1 1.5 73.1 11.5J5 39.3 15.7 1.5 68.7 15.2J6 24.7 18.9 2.0 73.3 16.3J7 33.8 16.5 2.0 45.1 8.35J8 22.8 16.5 2.0 5.1 8.63a PIP uncertainty was about 10%.Figure 6. V–I curves measured at 77 K in the self-field for sampleJ1–8 after HT1. All curves show a typical superconducting tonormal transition, indicating that the superconducting joints wereformed in all samples after HT1.(without IP). For J3–8, the Ic and n values after HT2 were alsoevaluated.Figure 6 shows the V–I curves at 77 K in the self-field forthe samples after HT1. All V–I curves show a typical super-conducting to normal transition, indicating that superconduct-ing joints were formed in all samples after HT1. As shown intable 2, the ranges of Ic and n values were 22.8–42.6 A and13.1–18.9, respectively. In the previously reported supercon-ducting joints, the n values were 6–10 [13, 19]. In this study,after HT1, the samples exhibited higher n values with highreproducibility.The higher n values are attributed to the homogenousstructure [10], as demonstrated in Nb-Ti wires and Bi-2223tapes [30, 31]. By using flattened Bi-2223 tapes, a uniformpressing pressure could be applied enabling the formation of ahomogeneous structure. This allows the reproducible fabrica-tion of superconducting joints possessing high n values.3.2.2. Increase in Ic after IP and HT2. In a previous study,we demonstrated Ic improvement in the superconducting joint5Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alFigure 7. Relationship between PIP and ∆Ic at 77 K in self-field for sample J3–8. The PIP error bars correspond to ±10% uncertainty. Thegray dashed line was derived from J3–6 data points by the least-squares method. There appears to be a linear relationship between PIP and∆Ic of these samples. At a PIP of 2.0 × 108 Pa, a large∆Ic was not reproduced.Figure 8. Secondary electron images of typical polished surfaces for the joining part of J1, J3, J4, J6, and J8. The position for theobservation and ∆Ic of each sample are also shown. Many cracks were observed in most J8 filaments. IP at a PIP of 2.0 × 108 Pa can causesignificant damage to the filaments in the Bi-2223 tape, resulting in the small∆Ic.by introducing IP and HT2 [19]. In this study, in order toquantify the increase in Ic, we used increments of Ic,∆Ic = Ic(after HT2)− Ic (after HT1). The relationship betweenPIP and∆Ic for J3–8 is shown in figure 7. The PIP error bars corres-pond to ±10% uncertainty. The gray dashed line was derivedfrom J3–6 data points by the least-squares method.The n values for J3–6 were >11 after HT2, as shown intable 2. There appears to be a linear relationship between PIPand∆Ic for these samples. This implies that a denser interme-diate layer was formed with a higher PIP, resulting in a larger∆Ic. J6 at a PIP of 2.0 × 108 Pa showed the largest ∆Ic of48.6 A.The∆Ic of J7 and J8 weremuch smaller than that of J6. ThePIP for J7 and J8 were similar to that for J6. A large ∆Ic wasnot reproduced at a PIP of 2.0 × 108 Pa. As shown in table 2,J7 and J8 had low n values (<9) after HT2. The cause of thesmall ∆Ic and low n values for J7 and J8 is discussed in thenext section.The ∆Ic of J3 at a PIP of 1.0 × 108 Pa was only 6.5 A.This may be because the densification of the intermediatelayer was insufficient to achieve a large ∆Ic. At a PIP of1.5 × 108 Pa, relatively large ∆Ic of 29.4 and 30.5 A wereobserved in J4 and J5, respectively. There is an optimum PIPof around 1.5–2 × 108 Pa to achieve a large ∆Ic with highreproducibility.3.2.3. Evaluation of microstructure and intermediate layerdensification. The microstructures of J1, J3, J4, J6 andJ8 were examined. An intermediate layer with a homogen-eous thickness of about 30 µm is formed in each sample,as reported in [19]. Figure 8 shows typical secondary elec-tron images of the polished surfaces of the samples. In eachimage, the upper and lower parts correspond to the inter-mediate layer and Bi-2223 tape, respectively. As reported in[13, 19], the grains were in good contact at the interfaces6Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alFigure 9. Typical secondary electron images of polished surfaces for the intermediate layers of J1, J3, J4, J6 and J8. The filling factor (F) ofthe intermediate layer was calculated using such images.between the intermediate layer and filaments. A large num-ber of voids were observed in the intermediate layer of eachsample.As shown in figure 8(e), many cracks were observed inmost filaments of J8. Few and thin cracks were observed inthe filaments of the other samples. There was a differencein the damage to the filaments between J6 and J8 pressedat a PIP of 2.0 × 108 Pa. This may be due to the uncer-tainty of PIP. The actual PIP applied to J8 was probably higherthan the nominal value, resulting in significant damage to thefilaments.Figure 9 shows typical secondary electron images of thepolished surface of the intermediate layer for J1, J3, J4, J6,and J8. General plate-like Bi-2223 grains were found in eachsample. The grain size was 5–10 µm, which was compar-able to that of the previously reported bulks heat-treated atPO2 = 3 kPa and 815 ◦C–825 ◦C [22].In J3, J4, J6, and J8, the main phase of the intermedi-ate layer was Bi-2223, which was not dependent on PIP.This result is consistent with that of the thick film samplesshown in section 2.2. We could not find coarse impurities suchas (Pb,Bi)3Sr2Ca2CuOy (Pb-3221), Bi-2212, or (Ca,Sr)-Cu-Ophases in the intermediate layer.There appeared to be no difference in the degree of c-axisgrain alignment among these samples. The c-axis grain mis-orientation angle for the intermediate layer of each samplewas estimated from the directions of the plate-like Bi-2223grains. Although there was broad distribution of the misorient-ation angle depending on the observation location, the meanmisorientation angle was about 15◦–30◦ calculated usingabout 30 grains in each sample. This suggests that a weaklygrain-oriented intermediate layer was formed in each sampleTable 3. PIP and calculated F of the intermediate layer with meanand standard deviation values.Sample PIPa / 108 Pa F / %J1 — (only HT1) 68.6 ± 3.0J3 1.0 71.2 ± 2.2J4 1.5 73.7 ± 1.9J6 2.0 77.4 ± 2.0J8 2.0 78.0 ± 1.9a PIP uncertainty was about 10%.because pressing pressure up to 2.0 × 108 Pa was too low topromote c-axis grain alignment.The calculated filling factors (F) with the mean and stand-ard deviation values for the intermediate layer of each sampleare summarized in table 3. The table quantitatively shows thatIP densifies the intermediate layer. The higher PIP, the denserthe intermediate layer.The F value of J8 is comparable to that of J6. This meansthat the densification by IP was achieved even in the interme-diate layer of J8, although the ∆Ic of J8 was much smallerthan that of J6. It has been reported that mechanical damageto the filaments deteriorates the Ic and n values of Bi-2223tape [32, 33]. The filaments in J8 were significantly damagedas shown in figure 8(e), resulting in the lowered ∆Ic and nvalues.Although the microstructure of J7 (PIP of 2.0 × 108 Pa)was not observed because J7 was degraded after the transportmeasurements, the corresponding F value is probably similarto that of J6 and J8. The small ∆Ic and low n value of J7 wasprobably caused by the damage to the filaments in the Bi-2223tapes.7Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alFigure 10. Relationship between F and ∆Ic for thesuperconducting joint samples J1, J3, J4 and J6. Gray dashed linewas derived from J3, J4, and J6 data points using the least-squaresmethod.∆Ic appears to be positively correlated with F, indicatingthat the densification of the intermediate layer produces a large∆Icof the Bi-2223 superconducting joint.4. Discussion4.1. Relationship between intermediate layer densificationand ∆Ic of the superconducting jointAs presented in section 2.2, IP increased the ρ of the thick filmsamples. The higher the PIP, the higher the ρ in sample TF3–5.An increase in intergrain Jc was achieved by the densification.Assuming that Jc of TF2 is a typical value for a thick film afterHT1, increments of Jc,∆Jc = Jc (after HT2)− Jc (after HT1)will also increase by the densification in TF3–5.In the superconducting joints, IP densified the intermediatelayer, as shown in table 3. The higher the PIP, the higher the Fin samples J3, J4, and J6. This densification of the intermediatelayer contributed to the large∆Ic of the superconducting joint.Figure 10 shows the relationship between F and ∆Ic forthe superconducting joint samples J1, J3, J4, and J6. J1, whichwas prepared with only one HT process (HT1), was plotted as∆Ic = 0. The gray dashed line was derived from J3, J4, andJ6 data points using the least-squares method. ∆Ic appears tobe positively correlated with F. It can be concluded that thedensification of the intermediate layer produces a large∆Ic inthe Bi-2223 superconducting joint.4.2. Achieving high Ic in the Bi-2223 superconducting jointWe clarified that the density of the intermediate layer is one ofthe key parameters affecting Ic of the Bi-2223 superconductingjoint. Figure 10 suggests that if the intermediate layer densityis increased to F > 90%, a∆Ic above 120 A can be achieved.Given that the Ic of the joints after HT1 ranged from 22.8 to42.6 A, an Ic above 140 A is possible after HT2 with a verydense intermediate layer. This Ic is comparable to that of thecommercially available Bi-2223 tape [2].High-pressure IP can cause mechanical damage to thefilaments in Bi-2223 tapes, and densification methods otherthan uniaxial pressing are needed to obtain dense intermediatelayers. Hot pressing, which has been demonstrated in Bi-2223tapes [34], may be effective. However, filament damage canoccur and needs to be suppressed. It may be possible to intro-duce an already very dense material prepared in advance as anintermediate layer. However, achieving good grain contact atthe joining interface may be difficult.The large∆Ic demonstrated in this study is attributed to theincrease in intergrain Jc of the intermediate layer due to densi-fication. Other methods for controlling the Jc of the intermedi-ate layer can be effective in producing a high Ic. The signific-ant parameters for controlling Jc include constituent phase [2],c-axis grain alignment [35], chemical composition [22, 23],and grain boundary structure [36]. Among them, improvingc-axis grain alignment is promising because the grain align-ment of the intermediate layer was poor, as shown in 3.2.3.5. ConclusionWe clarified the effect of Bi-2223 intermediate layer densi-fication achieved by IP on the Ic of Bi-2223 superconductingjoints. The following conclusions are drawn:(a) Bi-2223 thick films can be densified by IP. An increase inintergrain Jc can be achieved by densification.(b) Using flattened Bi-2223 tapes for making a joint enablesuniform uniaxial pressure to be applied. The uniformpressure probably promotes the formation of a homogen-eous structure. This allows the reproducible fabrication ofsuperconducting joints with high n values.(c) The Ic of Bi-2223 superconducting joints could beincreased by applying IP and HT2. However, the filamentsof the Bi-2223 tapes can be damaged by high-pressure IP,which then causes Ic deterioration in the joint. To achievea high Ic the optimum PIP is in the range of 1.5–2× 108 Pa.(d) The densification of the Bi-2223 intermediate layer by IPincreases the∆Ic of the superconducting joint. The densityof the intermediate layer is a key parameter affecting theIc of the Bi-2223 superconducting joint.(e) In order to further enhance the Ic of Bi-2223 superconduct-ing joints, densification methods other than uniaxial press-ing are needed. Controlling Jc of the intermediate layer isalso effective in producing a high Ic.Data availability statementThe data that support the findings of this study are availableupon reasonable request from the authors.s.AcknowledgmentsThis work was supported by JST Mirai-Program Grant Nos.JPMJMI17A2 and JSPS KAKENHI Grant No. JP22K14482,Japan.8Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et alORCID iDsY Takeda https://orcid.org/0000-0001-7217-9853G Nishijima https://orcid.org/0000-0001-7493-0559References[1] Ayai N et al 2008 DI-BSCCO wire with Ic over 200 A at 77 KJ. Phys.: Conf. Ser. 97 012112[2] Sato K, Kobayashi S and Nakashima T 2012 Present status andfuture perspective of bismuth-based high-temperaturesuperconducting wires realizing application systems Jpn. J.Appl. Phys. 51 010006[3] Miyoshi Y, Nishijima G, Kitaguchi H and Chaud X 2015 Highfield Ic characterizations of commercial HTS conductorsPhysica C 516 31–35[4] Bonura M, Barth C and Senatore C 2019 Electrical andthermo-physical properties of Ni-alloy reinforced Bi-2223conductors IEEE Trans. Appl. Supercond.29 6400205[5] Kusada S et al 2007 The project overview of the hts magnetfor superconducting maglev IEEE Trans. Appl. Supercond.17 2111–6[6] Terao Y et al 2013 Newly designed 3 T MRI magnet woundwith Bi-2223 tape conductors IEEE Trans. Appl.Supercond. 23 4400904[7] Hashi K et al 2015 Achievement of 1020 MHz NMR J. Magn.Reson. 256 30–33[8] Nishijima G et al 2016 Successful upgrading of 920-MHzNMR superconducting magnet to 1020 MHz usingBi-2223 innermost Coil IEEE Trans. Appl. Supercond.26 4303007[9] Awaji S, Watanabe K, Oguro H, Miyazaki H, Hanai S,Tosaka T and Ioka S 2017 First performance test of a 25 Tcryogen-free superconducting magnet Supercond. Sci.Technol. 30 065001[10] Brittles G D, Mousavi T, Grovenor C R M, Aksoy C andSpeller S C 2015 Persistent current joints betweentechnological superconductors Supercond. Sci. Technol.28 093001[11] Takeda Y, Maeda H, Ohki K and Yanagisawa Y 2022 Reviewof the temporal stability of the magnetic field for ultra-highfield superconducting magnets with a particular focus onsuperconducting joints between HTS conductorsSupercond. Sci. Technol. 35 043002[12] Park Y, Lee M, Ann H, Choi Y H and Lee H 2014 Asuperconducting joint for GdBa2Cu3O7−δ-coatedconductors NPG Asia Mater. 6 e98[13] Takeda Y, Motoki T, Kitaguchi H, Nakashima T, Kobayashi S,Kato T and Shimoyama J 2019 High Ic superconductingjoint between Bi2223 tapes Appl. Phys. Express12 023003[14] Ohki K et al 2017 Fabrication, microstructure and persistentcurrent measurement of an intermediate grownsuperconducting (iGS) joint between REBCO-coatedconductors Supercond. Sci. Technol. 30 115017[15] Chen P et al 2017 Development of a persistentsuperconducting joint between Bi-2212/Ag-alloymultifilamentary round wires Supercond. Sci. Technol.30 025020[16] Mukoyama S, Nakai A, Sakamoto H, Matsumoto S,Nishijima G, Hamada M, Saito K and Miyoshi Y 2018Superconducting joint of REBCO wires for MRI magnet J.Phys.: Conf. Ser. 1054 012038[17] Jin X, Suetomi Y, Piao R, Matsutake Y, Yagai T, Mochida H,Yanagisawa Y and Maeda H 2019 Superconducting jointbetween multi-filamentary Bi2Sr2Ca2Cu3O10+δ tapes basedon incongruent melting for NMR and MRI applicationsSupercond. Sci. Technol. 32 035011[18] Mousavi T, Santra S, Melhem Z, Speller S and Grovenor C2021 Superconducting joint structures for Bi-2212 wiresusing a powder-in-tube technique IEEE Trans. Appl.Supercond. 31 6400504[19] Takeda Y et al 2022 Critical current improvement andresistance evaluation of superconducting joint betweenBi2223 tapes Supercond. Sci. Technol. 35 02LT02[20] Asano T, Tanaka Y, Fukutomi M, Jikihara K, Machida J andMaeda H 1988 Preparation of highly orientedmicrostructure in the (Bi, Pb)-Sr-Ca-Cu-O Sintered OxideSuperconductor Jpn. J. Appl. Phys. 27 L1652–4[21] Ito A, Matsuda M, Iwai Y, Ishii M, Takata M, Yamashita T andKoinuma H 1989 Influence of intermediate pressing onsuperconducting characteristics in Bi-Pb-Sr-Ca-Cu-OSystem Jpn. J. Appl. Phys. 28 L380–1[22] Takeda Y, Shimoyama J, Motoki T, Kishio K, Nakashima T,Kagiyama T, Kobayashi S and Hayashi K 2017 Fabricationof Bi2223 bulks with high critical current propertiessintered in Ag tubes Physica C 534 9–12[23] Takeda Y, Shimoyama J, Motoki T, Nakamura S, Nakashima T,Kobayashi S and Kato T 2018 Development of high JcBi2223/Ag thick film materials prepared by heat treatmentunder low PO2 Supercond. Sci. Technol. 31 074002[24] Hu Q Y, Liu H K and Dou S X 1996 Effect of mechanicaldeformation on the mass density of Ag-clad(Bi,Pb)2Sr2Ca2Cu3O10 wire and tape Appl. Supercond.4 17–24[25] Müller K-H, Andrikidis C, Du J, Leslie K E and Foley C P1999 Connectivity and limitation of critical current inBi-Pb-Sr-Ca-Cu/Ag tapes Phys. Rev. B 60 659–66[26] Flükiger R, Grasso G, Grivel J C, Marti F, Dhallé M andHuang Y 1997 Phase formation and critical current densityin Bi,Pb(2223) tapes Supercond. Sci. Technol.10 A68–A92[27] Zhang L, Mironova M, Selvamanickam V and Salama K 2000Study of Ag/BSCCO interface in Ag-sheathedmultifilament Bi-2223 tapes Physica C 341–348 1471–2[28] Jiang J, Cai X Y, Polyanskii A A, Schwartzkopf L A,Larbalestier D C, Parrella R D, Li Q, Rupich M W andRiley J G N 2001 Through-process study of factorscontrolling the critical current density of Ag-sheathed(Bi,Pb)2Sr2Ca2Cu3Ox tapes Supercond. Sci. Technol.14 548–56[29] Kato T et al 2004 Development of high performance Agsheathed Bi2223 wire Physica C 412–414 1066–72[30] Ekin J W 1987 Irregularity in Nb-Ti filament area and electricfield versus current characteristics Cryogenics27 603–7[31] Rimikis A, Kimmich R and Schneider T 2000 Investigation ofn-values of Composite Superconductors IEEE Trans. Appl.Supercond. 10 1239–42[32] Ochiai S, Nagai T, Okuda H, Oh S S, Hojo M, Tanaka M,Sugano M and Osamura K 2003 Tensile damage and itsinfluence on the critical current of Bi2223/Agsuperconducting composite tape Supercond. Sci. Technol.16 988–94[33] Shin H S and Katagiri K 2003 Critical current degradationbehaviour in Bi-2223 superconducting tapes under bendingand torsion strains Supercond. Sci. Technol.16 1012–89https://orcid.org/0000-0001-7217-9853https://orcid.org/0000-0001-7217-9853https://orcid.org/0000-0001-7493-0559https://orcid.org/0000-0001-7493-0559https://doi.org/10.1088/1742-6596/97/1/012112https://doi.org/10.1088/1742-6596/97/1/012112https://doi.org/10.1143/JJAP.51.010006https://doi.org/10.1143/JJAP.51.010006https://doi.org/10.1016/j.physc.2015.06.004https://doi.org/10.1016/j.physc.2015.06.004https://doi.org/10.1109/TASC.2019.2892086https://doi.org/10.1109/TASC.2019.2892086https://doi.org/10.1109/TASC.2007.899691https://doi.org/10.1109/TASC.2007.899691https://doi.org/10.1109/TASC.2013.2239342https://doi.org/10.1109/TASC.2013.2239342https://doi.org/10.1016/j.jmr.2015.04.009https://doi.org/10.1016/j.jmr.2015.04.009https://doi.org/10.1109/TASC.2016.2524466https://doi.org/10.1109/TASC.2016.2524466https://doi.org/10.1088/1361-6668/aa6676https://doi.org/10.1088/1361-6668/aa6676https://doi.org/10.1088/0953-2048/28/9/093001https://doi.org/10.1088/0953-2048/28/9/093001https://doi.org/10.1088/1361-6668/ac5645https://doi.org/10.1088/1361-6668/ac5645https://doi.org/10.1038/am.2014.18https://doi.org/10.1038/am.2014.18https://doi.org/10.7567/1882-0786/aaf8b4https://doi.org/10.7567/1882-0786/aaf8b4https://doi.org/10.1088/1361-6668/aa8e65https://doi.org/10.1088/1361-6668/aa8e65https://doi.org/10.1088/1361-6668/30/2/025020https://doi.org/10.1088/1361-6668/30/2/025020https://doi.org/10.1088/1742-6596/1054/1/012038https://doi.org/10.1088/1742-6596/1054/1/012038https://doi.org/10.1088/1361-6668/aafc44https://doi.org/10.1088/1361-6668/aafc44https://doi.org/10.1109/TASC.2021.3064512https://doi.org/10.1109/TASC.2021.3064512https://doi.org/10.1088/1361-6668/ac45a3https://doi.org/10.1088/1361-6668/ac45a3https://doi.org/10.1143/JJAP.27.L1652https://doi.org/10.1143/JJAP.27.L1652https://doi.org/10.1143/JJAP.28.L380https://doi.org/10.1143/JJAP.28.L380https://doi.org/10.1016/j.physc.2016.12.005https://doi.org/10.1016/j.physc.2016.12.005https://doi.org/10.1088/1361-6668/aac11chttps://doi.org/10.1088/1361-6668/aac11chttps://doi.org/10.1016/0964-1807(96)00001-4https://doi.org/10.1016/0964-1807(96)00001-4https://doi.org/10.1103/PhysRevB.60.659https://doi.org/10.1103/PhysRevB.60.659https://doi.org/10.1088/0953-2048/10/7A/007https://doi.org/10.1088/0953-2048/10/7A/007https://doi.org/10.1016/S0921-4534(00)01083-2https://doi.org/10.1016/S0921-4534(00)01083-2https://doi.org/10.1088/0953-2048/14/8/307https://doi.org/10.1088/0953-2048/14/8/307https://doi.org/10.1016/j.physc.2004.05.006https://doi.org/10.1016/j.physc.2004.05.006https://doi.org/10.1016/0011-2275(87)90081-6https://doi.org/10.1016/0011-2275(87)90081-6https://doi.org/10.1109/77.828459https://doi.org/10.1109/77.828459https://doi.org/10.1088/0953-2048/16/9/305https://doi.org/10.1088/0953-2048/16/9/305https://doi.org/10.1088/0953-2048/16/9/309https://doi.org/10.1088/0953-2048/16/9/309Supercond. Sci. Technol. 36 (2023) 035004 Y Takeda et al[34] Fujii H, Garnier V, Giannini E and Flükiger R 2004 Effect ofhot uniaxial pressing on the microstructure and criticalcurrent density of (Bi,Pb)-2223 tapes Supercond. Sci.Technol. 17 263–8[35] Takeda Y, Iwami T, Saito Y, Motoki T and Shimoyama J 2021Fabrication of high Jc Bi2223 thick films through grainalignment technique using a permanent magnet PhysicaC.584 1353873[36] Kametani F, Oloye T A, Jiang J, Osabe G and Kobayashi S2019 Visualization of the grain structure in the filamentcross sections of uniaxially textured high Jc Bi-2223 tapesAppl. Phys. Express 12 09300210https://doi.org/10.1088/0953-2048/17/2/005https://doi.org/10.1088/0953-2048/17/2/005https://doi.org/10.1016/j.physc.2021.1353873https://doi.org/10.1016/j.physc.2021.1353873https://doi.org/10.7567/1882-0786/ab347ehttps://doi.org/10.7567/1882-0786/ab347e The effect of intermediate layer densification on the critical current of a Bi-2223 superconducting joint 1. Introduction 2. Relationship between density and Jcof thick film samples 2.1. Experimental 2.2. Results 3. Icand microstructure evaluation for superconducting joint samples 3.1. Experimental 3.2. Results 3.2.1. Ic and n values after HT1. 3.2.2. Increase in Ic after IP and HT2. 3.2.3. Evaluation of microstructure and intermediate layer densification. 4. Discussion 4.1. Relationship between intermediate layer densification and ΔIc of the superconducting joint 4.2. Achieving high Ic in the Bi-2223 superconducting joint 5. Conclusion References