Influence of Precursor Powder Fabrication Methods on the Superconducting Properties of Bi-2223 Tapes

Li-Jun Cui(崔利军)$^{1,2}$, Ping-Xiang Zhang(张平祥)$^{1,2,3\hyperlink{s}{**}}$, Guo Yan(闫果)$^{2}$, Yong Feng(冯勇)$^{2}$, \\ Xiang-Hong Liu(刘向宏)$^{2}$, Jian-Feng Li(李建峰)$^{2}$, Xi-Feng Pan(潘熙锋)$^{2}$, Sheng-Nan Zhang(张胜楠)$^{3}$, Xiao-Bo Ma(马小波)$^{3}$, Jin-Shan Li(李金山)$^{1}$

doi:10.1088/0256-307X/36/2/027401

⇦Chinese Physics Letters, 2019, Vol. 36, No. 2, Article code 027401⇨ Influence of Precursor Powder Fabrication Methods on the Superconducting Properties of Bi-2223 Tapes Li-Jun Cui (崔利军)1,2, Ping-Xiang Zhang (张平祥)1,2,3**, Guo Yan (闫果)2, Yong Feng (冯勇)2, Xiang-Hong Liu (刘向宏)2, Jian-Feng Li (李建峰)2, Xi-Feng Pan (潘熙锋)2, Sheng-Nan Zhang (张胜楠)3, Xiao-Bo Ma (马小波)3, Jin-Shan Li (李金山)1 Affiliations 1State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi'an 710072 2National Engineering Laboratory for Superconducting Materials, Western Superconducting Technologies Co. Ltd., Xi'an 710018 3Superconducting Materials Center, Northwest Institute for Nonferrous Metal Research, Xi'an 710016 Received 14 August 2018, online 22 January 2019 ^**Corresponding author. Email: cljwst@163.com Citation Text: Cui L J, Zhang P X, Yan G, Feng Y and Liu X H et al 2019 Chin. Phys. Lett. 36 027401 Abstract Bi-2223 precursor powders are prepared by both oxalate co-precipitation (CP) and spray pyrolysis (SP) methods. The influence of fabrication methods on the superconducting properties of Bi-2223 tapes are systematically studied. Compared to the CP method, SP powder exhibits spherical particle before calcination and smaller particle size after calcinations with more uniform chemical composition, which leads to a lower reaction temperature during calcination process for Bi-2223 tapes. Meanwhile, the non-superconducting phases in SP powder are more uniformly distributed with smaller particle sizes. These features result in finer homogeneity of critical current in large-length of Bi-2223 tape, higher density of filaments and better texture after heat treatment. Therefore, the SP method could be considered as a better route to prepare precursor powder for large-length Bi-2223 tape fabrication. DOI:10.1088/0256-307X/36/2/027401 PACS:74.25.F-, 74.25.Sv, 74.62.Bf © 2019 Chinese Physics Society Article Text As one of the most important practical high temperature superconductors, the current capacity of Bi$_{2-x}$Pb$_{x}$Sr$_{2}$Ca$_{2}$Cu$_{3}$O$_{y}$ (Bi-2223) tape mainly depends on the precursor powder, mechanical deformation process, and heat treatment process. Among these factors, the properties of precursor powder are the origin of the most crucial parameter to affect the critical current density ($J_{\rm c}$) of Bi-2223 tape. It is necessary to thoroughly understand the influence of precursor powder on $J_{\rm c}$ and to develop an ideal method to prepare precursor powder. There are extensive studies on effects of phase assemblage of precursor powder on the formation of Bi-2223 phase and $J_{\rm c}$ of Bi-2223 tape. Many results show that an ideal phase assemblage should be Bi$_{2}$Si$_{2}$CaCu$_{2}$O$_{y}$ (Bi-2212) as major phase balanced with (Ca, Sr)$_{x}$Cu$_{y}$O (AEC) and Ca$_{2}$PbO$_{4}$, or Ca$_{2}$CuO$_{3}$ and CuO, or Ca$_{2}$PbO$_{4}$ and CuO.[1-3] Ca$_{2}$PbO$_{4} $ content in precursor powder is very important because it provides the liquid phase to accelerate Bi-2223 formation and heals the cracks caused by the intermediate deformation.[4,5] It has been shown that an ideal phase assemblage of precursor powder is not limited to only one set. The phase assemblage is only one parameter which can affect final current capacity of Bi-2223 tape. The homogeneity of phase distribution, particle size and chemical composition are also essential factors for the fabrication of high quality precursor powder for large-length Bi-2223 tape fabrication. As a conventional method, solid state reaction was adopted to prepare the precursor powder in the early stage. However, to obtain powders with high homogeneity and small particle size, many milling steps were required, and long sintering time was necessary because of low reaction activity.[6] The spray drying method was also developed for preparation of Bi-2223 precursor powder. However, the particle size obtained with this method was too large. Thus powder need be milled several times during the process, which would bring more complexity to the fabrication technique and cause the intrusion of impurity.[7,8] To date, oxalate co-precipitation (CP) and spray pyrolysis (SP) are two commonly adopted methods to prepare Bi-2223 precursor powders. Therefore, many works have been carried out to obtain optimal process parameters. For the CP method, 'one-powder process' and 'two-powder process' have been developed.[9,10] The one-powder process can prepare precursor powder with higher homogeneity and smaller particle sizes, while the two-powder process can exactly control the phase assemblage of final powder. The SP method could be adopted to prepare Bi-2223 precursor powder with particle sizes smaller than 2 µm, and high reaction activity, both of which can greatly reduce the calcination time and improve the fabrication efficiency.[11,12] Hsueh et al. found that there were much faster development of Bi-2223 phase and less Bi-2201 phase forming in spray pyrolysis powder than in others.[13] It is easy to control particle size by adjusting solution concentration and reaction temperature.[11] Li et al. found that the optimized annealing temperature of the tapes appears decreasing as the particle size decreases.[14] Jiang et al. obtained that the critical current of the tape with the smallest particle size is most sensitive to the annealing temperature, and leads to fastest formation of the 2223 phase.[15] However, so far, there is still no systematical study on the influence of SP fabricated Bi-2223 precursor powder on the large-length Bi-2223 tape fabrication, which can meet the real requirements of commercialized application. In the present work, we systematically investigate the influence of the CP and SP methods on the homogeneity and superconducting properties of large-length Bi-2223 tape. The reasons that could cause these differences are also studied.

Fig. 1. SEM images of precursor powders of (a) original CP powder, (b) original SP powder, and (c) final CP powder mixed by Bi-2212 and Ca$_{1.1}$Cu$_{1.04}$O$_{y}$ powders (d) calcined SP powder.

Precursor powders, corresponding to the nominal composition of Bi$_{1.87}$Pb$_{0.34}$Sr$_{1.9}$Ca$_{2.1}$Cu$_{3.04}$O$_{y}$, were prepared by oxalate co-precipitation and spray pyrolysis methods, respectively. For the CP method, the precursor powder is prepared by two-powder process as described by Li et al.[9] First, Bi-2212 powder was prepared by the CP process, and calcined at 800$^{\circ}\!$C for 15 h. After an intermediate milling, the powder was calcined at 840$^{\circ}\!$C in Ar-8%O$_{2}$. Second, Ca$_{1.1}$Cu$_{1.04}$O$_{y}$ powder, including Ca$_{2}$CuO$_{3}$ and CuO, was prepared with a similar route, and then calcined at 900$^{\circ}\!$C for 20 h. Finally, the CP powder was obtained by mixing Bi-2212 and Ca$_{1.1}$Cu$_{1.04}$O$_{y}$ powders by a long time milling process to achieve homogeneity. The SP powder was prepared using a mixed nitrate solution with the total cation concentration of 0.5 M. After the SP process at 800$^{\circ}\!$C, the collected powder was sintered at 810$^{\circ}\!$C for 6 h without any milling process. The 37-filament Bi-2223 tapes were fabricated by the established powder in tube (PIT) technique and heat treated based on two heat treatments and one intermediate rolling technology with CP and SP precursor powders, respectively. The phase compositions were determined by x-ray diffraction (XRD) using Cu K$_{\alpha}$ radiation. The melting behavior of precursor powders was analyzed based on TGA and DSC measurement. Elemental analyses were made using inductively coupled plasma atomic emission spectrometry (ICP/AES). The microstructures of powders and tapes were analyzed by scanning electron microscopy (SEM). The critical currents of tapes were measured by the standard four-point probe technique in nitrogen liquid. Figure 1 shows the SEM images of precursor powders prepared by different methods. The original CP powder has no regular shape and agglomeration can be observed by many small particles, as shown in Fig. 1(a). However, the original SP powder is spherical and uniformly distributed with the average particle size of $\sim$2 µm. After the calcination process, micaceous-like particles are obtained on both powders, which is the typical shape of Bi-2212 grains. Because of the long calcination time for CP powder, the average particle size increases to $\sim$5 µm, as shown in Fig. 1(c). It is known that the powder prepared by spray pyrolysis has higher reaction activity, thus reaction can be finished within a short time, which results in smaller average particle size; as shown in Fig. 1(d). Due to spherical shape of SP powder, the reaction process can be restrained separately within each particle, which may be beneficial to the homogeneity of phase compositions.

Fig. 2. X-ray diffraction patterns of calcined precursor powders prepared by oxalate co-precipitation and spray pyrolysis methods, respectively.

Phase assemblages of two different precursor powders are determined by x-ray diffraction patterns, as shown in Fig. 2. The SP powder is composed of Bi-2212, Ca$_{2}$PbO$_{4}$, AEC and small content of Bi-2201 phases, and CP powder is composed of Bi-2212, Ca$_{2}$CuO$_{3}$, Ca$_{2}$PbO$_{4}$, CuO and also small content of Bi-2201 phases. The difference of phase assemblage can be attributed to different preparation and calcination techniques. There are notable diffraction peaks of Ca$_{2}$CuO$_{3}$ and CuO phase in CP powder, which are not detectable in SP powder. It should also be noted that more Ca$_{2}$PbO$_{4}$ phase appears in SP powder. As described above, both phase assemblages of CP and SP powders should be reasonable. Figure 3 shows the TGA and DSC curves of final calcined precursor powders prepared by oxalate co-precipitation and spray pyrolysis methods at a heating rate of 5$^{\circ}\!$C/min in air. As shown in Fig. 3(a), two precursor powders have similar tendency of weight loss with increasing temperature, and the onset temperature of rapid dropping of SP powder is lower than that of CP powder. It should be noted that step-like dropping of weight appears on the curve of CP powder below 800$^{\circ}\!$C other than SP powder, which suggests a more complex reaction in CP powder. The carbon contents of SP and CP powders were detected, which are about 90 ppm and 240 ppm, respectively. Jeremie et al. found that the milling of precursor powder can increase carbon content rapidly.[16] Thus, several milling steps of CP powder should be responsible to the high carbon content. During the heating process, the decrease of carbon content causes the step-like dropping and larger weight loss below 800$^{\circ}\!$C; as shown in Fig. 3(a).

Fig. 3. TGA and DSC curves of precursor powders prepared by oxalate co-precipitation and spray pyrolysis methods at a heating rate of 5$^{\circ}\!$C/min in air.

Figure 3(b) shows the DSC curves of two precursor powders. It is obvious that SP powder has a lower reaction temperature than CP powder, which is attributed to the smaller particle size. As reported by Mark et al.,[17] the different phase assemblages in two precursor powders may have some effects on the reaction temperature, which can also enlarge the difference of reaction temperature, but it should not be the main reason in our study. To analyze the difference of non-superconducting phases, the SEM images of longitudinal cross sections of 37-filament green tapes are presented in Fig. 4. In both the images, the gray and black particles all represent non-superconducting phases, namely CuO and AEC phase, respectively. In the tape fabricated with SP powder, the non-superconducting phases have smaller diameters (less than 1 µm) and uniform distribution. However, in the tape fabricated with CP powder, the average diameter of non-superconducting phases is larger than 3 µm, and the homogeneity is not good. The inhomogeneity and large-size non-superconducting phases may affect the superconducting property of final Bi-2223 tape.

Fig. 4. SEM images of longitudinal cross sections of 37-filaments green tapes fabricated with (a) SP powder and (b) CP powder, respectively.

Fig. 5. Particle size distribution of calcined precursor powders prepared by oxalate co-precipitation and spray pyrolysis methods.

The uniformity of particle size distribution has an important effect on non-superconducting phase and critical current in final Bi-2223 tape. Figure 5 shows particle sizes of calcined precursor powders prepared by oxalate co-precipitation and spray pyrolysis methods. It can be seen that SP powder has smaller average particle size than CP powder. The particle size of SP powder shows normal distribution. However, there are extra two peaks located at about 0.3 µm and 25 µm, which may be the result of milling process. Compared to CP powder, SP powder has better uniformity of particle size distribution. The deviation of elements to nominal composition has an important influence on the homogeneity of Bi-2223 tapes. Therefore, a good preparation method should not only be simple and easy operating but should also minimize the deviation of chemical composition with nominal values. Figure 6 shows the deviation of five elements of precursor powder, namely Bi, Pb, Sr, Ca and Cu, prepared by two methods after normalization to Bi. For SP powder, all elements have small deviation, which are well controlled within $\pm$0.1%, and only Cu exhibits larger deviation compared to other elements, which is still smaller than 0.1%. However, for CP powder, the deviation increases to $\pm$0.35%, and deviations of three elements are more than $\pm$0.2%. It is obvious that it is easier to control the chemical composition in SP powder than in CP powder, which should be beneficial to the repeatability of precursor powder during industrial fabrication.

Fig. 6. The deviation of five elements to nominal composition by normalization to Bi for SP powder and CP powder.

Fig. 7. Critical currents of Bi-2223 tapes fabricated by CP and SP powders at different positions in larger-length direction at 77 K.

Fig. 8. SEM images of heat treated Bi-2223 tapes fabricated by (a) SP powder and (b) CP powder.

The 37-filament Bi-2223 tapes were fabricated with CP powder and SP powder, respectively. Two tapes were heat treated at the same process. Figure 7 shows the critical current values of two tapes at different positions in larger-length direction. Because of fine control of phase assemblage of CP powder, the CP tape has the highest critical current, which is about 118 A. However, it has a worse homogeneity in a larger-length direction. A larger variation ranging from 85 A to 118 A can be observed. For SP tape, there is better homogeneity of critical current, which varies only in a smaller range from 102 A to 110 A. By combining the above results, it seems that the homogeneity of critical current is affected by many factors, such as small particle sizes, small and uniformly distributed non-superconducting phases, small element deviation from nominal composition and simple reaction process. The spray pyrolysis fabricated precursor powder exhibits many advantages for the preparation of large-length Bi-2223 tape. Figure 8 shows the typical SEM images of heat treated Bi-2223 tapes fabricated with CP and SP powders, respectively. The light particles are (Bi, Pb)$_{3}$Sr$_{2}$Ca$_{2}$Cu$_{1}$O$_{y}$ (Pb-3221), and the black particles are AEC phase. Compared to SP tape, the diameters of AEC phases are larger, and the content of Pb-3221 phases is smaller in CP tape. The results of Fig. 4 show that the original size of non-superconducting phases in CP green tape is larger than that in SP green tape, which may be the reason for the different secondary phase particle sizes in final tapes. It is interesting that higher density and better texture filaments are obtained in SP tape compared with that of CP tape, which may be another reason for the homogeneity of critical current. In conclusion, we have prepared precursor powders by oxalate co-precipitation and spray pyrolysis methods, and 37-filament tapes are fabricated with different powders to study the effect of precursor powder on the superconducting properties of Bi-2223 tapes. We have demonstrated that the SP method is ideal to prepare precursor powder for large-length Bi-2223 tape. The main findings are as follows: (1) because of high reaction activity, lower reaction temperature and simple reaction process are required for calcinations of SP powder; (2) compared to CP powder, SP powder has smaller particle sizes, uniform distribution and smaller diameter of non-superconducting phases and small elements deviation to nominal composition; and (3) the advantages of SP powder are beneficial to high homogeneity of critical current, which is a very important issue for large-length Bi-2223 tape. References Preparation of Ag - Bi-2223 tape by controlling the phase evolution prior to sinteringOptimization of the preparation parameters of monofilamentary Bi(2223) tapes and the effect of the rolling pressure on j _cSynthesis of highly pure Bi-2223 ceramics using defined precursorsCorrelation between reaction kinetics of the Bi(Pb)-2223 phase and critical current density in Bi(Pb)-2223/Ag tapesEffects of lead content and particle size of precursor powders on formation rate, grain growth and critical current density of BSCCO 2223 tapesMechanical grinding of precursor powder and its effect on the microstructure and critical current density of Ag/Bi-2223 tapesThe influence of process variables of precursor powders on the microstructure evolution and transport current properties of Bi-2223/Ag tapesOptimization of spray drying process for Bi-2223 precursor powders fabricationThe effect of processing parameters on the phase assemblage and critical current density in Bi(Pb)-2223 tapesPreparation and Characterization of Large-Quantity Bi-2223 Precursor Powder by Oxalate CoprecipitationSynthesis and Characterization of Fine and Homogeneous BSCCO-2223 Precursor Powder by Spray Pyrolysis Process for PIT ProcessThe fabrication of fine and homogenous Bi-2223 precursor powder by a spray pyrolysis processA comparison of the properties of Bi-2223 precursor powders synthesized by various methodsCritical current density enhancement in Ag-sheathed Bi-2223 superconducting tapesEffects of precursor powder particle size on critical current density and microstructure of Bi-2223/Ag tapesEffect of Precursor Phase Composition on 2223 Phase Formation in Ag-Sheathed Tapes

[1]	Wang W G, Bain P A, Horvat J, Zeimetz B, Guo Y C, Liu H K and Dou S X 1996 Supercond. Sci. Technol. 9 881
[2]	Grasso G, Jeremie A and Flükigre R 1995 Supercond. Sci. Technol. 8 827
[3]	Sotelo A, Majewski P, Park H S and Aldinger F 1996 Physicas C 272 115
[4]	Jiang J and Abell J S 1998 Supercond. Sci. Technol. 11 705
[5]	Yoo J, Ko J, Su X D, Kim H and Chung H 2001 IEEE Trans. Appl. Supercond. 11 3549
[6]	Kim W J, Kwon S C, Lee H J, Lee H G, Hong G W and Kuk I H 1998 Physica C 294 147
[7]	Yoo J, Ko J, Kang S, Kim H, Jiang C H and Chung H 2002 Physica C 372 1005
[8]	Bai L F, Zhang S N, Li C S, Hao Q B, Liu G Q and Zhang P X 2016 J. Mater. Sci.: Mater. Electron. 27 8862
[9]	Li C S, Zhang P X, Yu Z M, Zheng H L, Xiong X M, Liu Y S, Wang Q Y, Liu F S, Wu Y F and Zhou L 2003 Physica C 386 127
[10]	Cui L J, Zhang P X, Li J S, Yan G, Wang D Y, Pan X F, Liu G Q, Hao Q B, Liu X H and Feng Y 2016 IEEE Trans. Appl. Supercond. 26 1
[11]	Yoo J, Ko J W, Kim Y K and Chung K C 2005 IEEE Trans. Appl. Supercond. 15 2474
[12]	Yoo J, Kim S W, Ko J W and Kim Y K 2004 Supercond. Sci. Technol. 17 S538
[13]	Hsueh Y W, Chang S C, Liu R S, Woodall L and Gerards M 2001 Mater. Res. Bull. 36 1653
[14]	Li Q, Broderson K, Hjuler H A and Freltoft T 1993 Physica C 217 360
[15]	Jiang J and Abell J S 1998 Physica C 296 13
[16]	Jeremie A and Flukiger R 1994 IEEE Trans. Appl. Supercond. 30 1883
[17]	Mark O, André W, Severine A, Zimmer M and Bock J 2005 IEEE Trans. Appl. Supercond. 15 2499