Physica C 415 (2004) 57–61 www.elsevier.com/locate/physc
Eﬀect of seed layers on the preparation of SrTiO3 buﬀer layers on Ni tapes via sol–gel method X.B. Zhu a
, L. Chen a, S.M. Liu a, W.H. Song a, Y.P. Sun a, K. Shi b, Z.Y. Sun b, S. Chen b, Z. Han b
Key Laboratory of Materials Physics, Institute of Solid State Physics, Chinese Academy of Sciences, Hefei 230031, PR China b Applied Superconductivity Research Center of Tsinghua University, Beijing 10084, PR China Received 8 January 2004; received in revised form 20 May 2004; accepted 25 July 2004 Available online 3 September 2004
Abstract SrTiO3 (STO) buﬀer layers with diﬀerent STO seed layers on Ni (2 0 0) substrates are fabricated using the spinning coating technique. It is found that the seed layers can remarkably aﬀect the orientation of the subsequent STO precursor layer and the relationship between the orientation of the STO layer and the seed layer is discussed. Ó 2004 Elsevier B.V. All rights reserved. PACS: 68.55. a Keywords: Buﬀer layer; SrTiO3; Sol–gel; Texture; YBCO
1. Introduction YBa2Cu3O7 d (YBCO) coated conductors have a great potential for applications due to their high critical current density and strong ﬂux pinning at liquid nitrogen temperature . Currently, two competing methods are being explored for fabrication of YBCO coated conductors; namely, the ion beam assisted deposition (IBAD) and the rolling
Corresponding author. Tel.: +86 551 5591436; fax: +86 55 5591434. E-mail address: [email protected]
assisted textured substrates (RABiTS) processes . The RABiTS process is a non-vacuum method, with lower requirement to the equipments, being easy to be scaled up. Additionally, it has been reported that using an all-chemical method, YBCO coated conductors have critical current densities exceeding 106 A/cm2 at 77 K and zero applied magnetic ﬁeld . To fabricate YBCO coated conductors, it needs a suitable buﬀer layer which plays an important role in the sandwich structure for its ability to prevent diﬀusion between the metal substrate and the YBCO ﬁlm as well as to transfer texture of the substrate. Until now, it has been reported that
0921-4534/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.physc.2004.07.018
X.B. Zhu et al. / Physica C 415 (2004) 57–61
various buﬀer layers can be grown on textured Ni substrates via solution-based methods [4–11]. Among these various buﬀer layers, SrTiO3 (STO) has the character of perovskite structure and is stable. In addition, the lattice parameter of 0.3905 nm of STO provides a very close lattice match with YBCO (2.3% mismatch) and Ni substrate (11.1% mismatch). Therefore, STO is considered to be one of the most suitable buﬀer materials for the preparation of YBCO coated conductors. Many research works have been published focusing on the sol–gel processing approaches for STO buﬀer layers [12,13]. To obtain high quality STO buﬀer layers, Siegal et al. [3,14] pointed out that a very thin template serving as seed layer was ﬁrst grown to provide orientation followed by the growth of subsequent precursor STO layers to provide thickness. However, what solution concentration for seed layers is the most suitable to fabricate STO buﬀer layers has not been investigated until now. In this work, we have grown STO buﬀer layers on textured Ni substrates using the spin-coating technique from a solution-based precursor. The motivation of this study was to investigate the effects of solution concentration of the seed layers on the orientations of the subsequent precursor STO buﬀer layers by X-ray diﬀraction (XRD), scanning electron microscopy (SEM) and atomic force microscopy (AFM) methods.
2. Experimental The detailed solution preparation for STO could be found elsewhere . Brieﬂy, the STO solutions were produced by reacting titanium isopropoxide with acetylacetone before combining it with a solution of Sr acetate dissolved in triﬂuoroacetic acid (TFA) and diluted by acetylacetone to the desired concentration. In our work, the seed layer solution concentration was 0.04, 0.065, 0.08 M and the solution concentration for the subsequent STO precursor layer was 0.25 M. A spinning rate of 4000 rpm and time of 60 s were used in the process of deposition both for the seed layers and the subsequent precursor layers on Ni (2 0 0) tapes, followed by a hot-plate treat-
ment at 300 °C. These dried ﬁlms were heated quickly to 900 °C and kept at this temperature for 2 h under 4% H2/N2 atmosphere. The treatment for each layer, seed layers as well as subsequent precursor STO layers, was repeated up to the numbers of layers. Typically, a seed layer and two subsequent precursor layers (about 100 nm for each STO layer determined by AFM) were deposited for our samples. For the sake of description, we deﬁne here the seed layer samples fabricated using solution concentration of 0.04, 0.065, 0.08 M and no seed layer as samples As, Bs, Cs and Ds, respectively. Correspondingly, the STO samples deposited on these four seed layers are deﬁned as samples A, B, C, and D, respectively. The microstructure was identiﬁed by an Inlens detector in a thermal ﬁeld emission gun scanning electron microscopy (FEG-SEM) LEO1530 equipped with a HKL electron backscattering diffraction pattern (EBSD) detector and a Park Scientiﬁc Instruments designed Autoprobe CP type atomic force micrograph (AFM); A Philips XÕpert PRO X-ray diﬀractometer (XRD) with CuKa radiation was used to carry out the h–2h scan, the out-of-plane x scan and the in-plane u scan investigations.
3. Results and discussion The results (not shown here) of the XRD h–2h scan, the out-of-plane x scan (full width in half maximum, FWHM 7.2°) and the in-plane u scan (FHWM 8.1°) of our Ni (2 0 0) tapes indicates that the Ni (2 0 0) tapes have a strong cubic texture. A series of STO precursor buﬀer layers are deposited on diﬀerent seed layers buﬀered Ni (2 0 0) tapes. The h–2h scans of STO/seed layer/ Ni as a function of seed layer solution concentrations are shown in Fig. 1(a)–(d). The peaks appearing at 32.1° and 46.2° belong to the (1 1 0) and (2 0 0) planes of STO buﬀer layers and the peaks appearing at 51.7° and 44.3° belong to the cubic Ni substrates. It can be seen that the strongest absolute intensity of STO (2 0 0) peak is attributed to the sample B. However, the STO (2 0 0) peaks of other samples are relatively weak. The results indicate that the crystalline qualities of STO
X.B. Zhu et al. / Physica C 415 (2004) 57–61
Fig. 1. The h–2h scans of the samples A (a), B (b), C (c), and D (d).
buﬀer layers can be changed when the seed layers are diﬀerent. The above diﬀerence of crystallization qualities of STO buﬀer layers can be uniquely
attributed to the eﬀect of seed layer solution concentrations because there are no other diﬀerences than the seed layer solution concentrations when fabricating these STO buﬀer layers. The SEM results (not shown here) show that sample A has a network structure, but that in sample B, the STO buﬀer layer consists of ﬁlaments, and that for sample C and D, the STO buﬀer layers are relatively dense compared with samples A and B. The diﬀerences between pictures for these samples can thus be only attributed to the eﬀects of the seed layer solution concentrations. In order to trace the diﬀerences of XRD and SEM patterns of the above samples, the observations of AFM and SEM are carried out to discern the morphology of diﬀerent seed layers. The SEM patterns are shown in Fig. 2(a)–(d). It can be seen that the grains of sample As and Bs are isolated, with grain sizes about 30–50 nm, and the distance
Fig. 2. The SEM patterns of the samples As, Bs, Cs and Ds, respectively.
X.B. Zhu et al. / Physica C 415 (2004) 57–61
between two grains in sample As is larger than that in sample Bs, i.e., at the same area there are more grains in sample Bs than in sample As. In sample Cs, the grains are also isolated but sparser than in sample As and Bs. Additionally, the grain sizes of sample Cs are not uniform, with grain sizes varying from 40 to 200 nm; for sample Ds ﬁnally, there exist no isolated grains but connected crystalline grains. The diﬀerence can also be observed in AFM (not shown here). It has been known that the subsequent STO precursor layers nucleate at the grains of seed layers. When the numbers of grains in the seed layer decrease, i.e., the nucleate centers decrease, the crystallization of the subsequent STO precursor layers will degrade. Additionally, as discussed by
Schwartz et al. , in the fabricating process of highly oriented ﬁlms the seed layer should be thin enough to suppress the grain growth along the thickness orientation. Combined with our experimental results, we suggest that although the grain growth along the thickness orientation of the seed layer fabricated with low solution concentration, i.e., 0.04 M in our experiments, is strongly suppressed, the nucleation centers for the subsequent STO layer are too sparse to induce all grains in the subsequent STO layer to nucleate at these oriented nucleation centers resulting in weak crystallization as well as weak orientation. At an intermediate solution concentration, i.e., 0.065 M in our experiments, the orientation of the seed layer is high due to the suppression of grain
Fig. 3. The SEM (a, b), the AFM (c) and the out-of-plane x scan pattern (d) of the sample B. From the SEM, it can be seen that the STO buﬀer layer consists of ﬁlaments with relatively dense structure. The root-mean-square (RMS) roughness of 10.1 nm obtained from the AFM indicates that the STO buﬀer layer is relatively smooth. The FWHM of 5.3° of the out-of-plane x scan reveals that the STO buﬀer layer has a strong out-of-plane orientation.
X.B. Zhu et al. / Physica C 415 (2004) 57–61
growth along the thickness orientation, additionally, the nucleation centers for the subsequent STO layer are suﬃcient resulting in the strongest crystallization quality and orientation of the STO buﬀer layer in our experiments. At the high solution concentration of seed layers, i.e., 0.08 M and no seed layer condition, the decreasing of the nucleation centers for the subsequent STO layers as well as the suppression of grain growth along the thickness orientation in seed layers led to the weak crystallization qualities and orientations for the subsequent STO layers. Based on the analysis as well as the experimental results, we suggest that there exists an optimal solution concentration for the seed layers to obtain high quality STO buﬀer layers, i.e., 0.065 M in our experiments. As a result of the most excellent oriented STO buﬀer layers (sample B), SEM at diﬀerent magniﬁcations, AFM and the out-of-plane x scan are carried out and the results are depicted in Fig. 3. It can be observed that oriented, crack-free, relatively dense and smooth STO buﬀer layers can be obtained in our experiments.
4. Conclusion STO buﬀer layers were fabricated on diﬀerent seed layers on textured Ni tapes using the spincoating technique. It was shown that the seed layer solution concentration remarkably inﬂuenced the orientations of the subsequent STO precursor layers. Either at a low or high seed layer solution concentration, the crystallization of subsequent STO precursor layers was insuﬃcient which could be attributed to the small number of nucleation centers. Only at an intermediate seed layer solution concentration (0.065 M in our experiments), the STO buﬀer layers were highly (2 0 0)-oriented. AFM and SEM revealed that the STO buﬀer layers were relatively dense and smooth. Further works are being carried out to fabricate epitaxial,
smoother, denser and thicker STO buﬀer layers for the preparation YBCO of coated conductors.
Acknowledgment This work was supported by the National Superconductivity 863 project under nos. 2002AA306211, 2002AA306281 and the Fundamental Bureau, Chinese Academy of Sciences. The authors thank J. Meng at Tsinghua University for the measurement of SEM.
References  D. Larbalestier, A. Gurevich, M. Feldmann, A. Polyanskii, Nature 414 (2001) 368.  T. Araki, I. Hirabayashi, Supercond. Sci. Technol. 16 (2003) R71.  M.P. Siegal, P.G. Clem, J.T. Dawley, R.J. Ong, M.A. Rodriguez, D.L. Overmyer, Appl. Phys. Lett. 80 (2002) 2710.  E. Celik, Y.S. Hascicek, Mater. Sci. Eng. B 106 (2004) 1.  Y. Akin, Z. Aslanoglu, E. Celik, L. Arda, W. Sigmund, Y.S. Hascicek, IEEE Trans. Appl. Supercond. 13 (2003) 2673.  J.S. Morrell, Z.B. Xue, E.D. Specht, A. Goyal, P.M. Martin, D.F. Lee, R. Feenstra, D.T. Verebelyi, D.K. Christen, T.G. Chirayil, M. Paranthaman, C.E. Vallet, D.B. Beach, J. Mater. Res. 15 (2000) 621.  M.S. Bhuiyan, M. Paranthaman, S. Sathyamurthy, T. Aytug, S. Kang, D.F. Lee, A. Goyal, E.A. Payzant, K. Salama, Supercond. Sci. Technol. 16 (2003) 1305.  E. Celik, J. Schwartz, E. Avci, H.J.S. Muntau, Y.S. Hascicek, IEEE Trans. Appl. Supercond. 9 (1999) 2264.  E. Celik, Y.S. Hascicek, Mater. Sci. Eng. B 96 (2002) 300.  M.I.E. Kawni, H. Okuyucu, Z. Aslanoglu, Y. Arkin, Y.S. Hascicek, J. Supercond.: Incorp. Novel Mag. 16 (2003) 533.  E. Celik, I.H. Mutlu, Y.S. Hascicek, IEEE Trans. Appl. Supercond. 10 (2000) 1162.  Y.X. Zhou, S. Bhuiyan, S. Scruggs, H. Fang, M. Mironova, K. Salama, Supercond. Sci. Technol. 16 (2003) 901.  S. Sathyamurthy, K. Salama, Supercond. Sci. Technol. 13 (2000) L1.  J.T. Dawley, P.G. Clem, Appl. Phys. Lett. 81 (2002) 3028.  R.W. Schwartz, Chem. Lett. 9 (1997) 2325.