20nm MgB2 superconducting film with TC above 34K and JC∼106Acm−2 at 30K
Qian Dai,Yinbo Wang,Xiaobai Ma,Qianqian Yang,Huai Zhang,Ruijuan Nie,Qingrong Feng,Furen Wang
DOI: https://doi.org/10.1016/j.physc.2011.05.001
2012-01-01
Abstract:Highlights ► We fabricated 20 nm-thick MgB 2 films by ex - situ annealing from Mg/B multilayer method. ► SiC (0 0 1), MgO (1 1 1) and sapphire (0 0 1) substrate was compared for 20 nm-thick MgB 2 films. ► 20 nm-thick MgB 2 films have its T c onset = 35.6 K and J C = 2 × 10 6 A cm −2 at 30 K. Abstract MgB 2 thin films were prepared by ex - situ annealing of Mg/B multilayers. Different precursor, substrate and annealing condition were investigated. After annealed in a flow of Mg vapor with reducing hydrogen ambient for 1–4 min, high quality 20 nm-thick c-oriented MgB 2 thin films with T c onset = 35.6 K and critical current density J C = 2 × 10 6 A cm −2 at 30 K were obtained. The mean roughness of the 20 nm-thick film was 2.68 nm. Keywords MgB 2 Thin films E-Beam evaporation 1 Introduction The discovery of superconductor of MgB 2 [1] has attracted great interest in its application in superconducting electronics. The high transition temperature T C of 39 K allows operation of MgB 2 devices and circuits above 20 K instead of 4.2 K for Nb-based superconducting devices. MgB 2 has two energy gaps [2] , and even the smaller one is larger than that of Nb [3] . It implies that MgB 2 devices can operate even faster than Nb-based superconducting devices. Compared to the high-temperature superconductors, its longer coherence length [4] makes it easier to obtain reproducible and uniform MgB 2 Josephson junction. The first step towards Josephson junction and device is to fabricate high quality MgB 2 thin film. Various methods have been explored for the synthesis of MgB 2 thin films including one step method such as hybrid physical–chemical vapor deposition (HPCVD) [5] , reactive evaporation (REC) [6] , molecular beam epitaxy (MBE) [7] , etc. and two step in situ [8,9] or ex situ [10,11] annealing of Mg–B or B precursor films. In this paper, we choose a two-step method: annealing Mg/B multilayer films in a flow of Mg vapor [12,13] . This method avoids the toxic gas of diborane, and provides a reproducible MgB 2 thin film. Another advantage is its convenience in achieving multilayer structure using different materials for MgB 2 Josephson junction. In our experiment, we obtained 20 nm MgB 2 films with critical temperature T c onset = 35.6 K and J C = 2 × 10 6 A cm −2 at 30 K. Our work may present a new way to fabricate ultrathin MgB 2 films, which may be a potential use in MgB 2 Superconducting single-photon detectors (SSPD) [14–17] . 2 Experimental details Mg/B multilayer films were deposited on sapphire (0 0 1), MgO (1 1 1) and 6H–SiC (0 0 1) substrates, using electron beam evaporation method [12,13] . Pure Mg (99.5%) and pure B (99.5%) was used as our evaporation sources. Samples were kept at room temperature during the deposition. Precursor thickness was monitored by quartz oscillator in situ . In the annealing process, pre-purified hydrogen was continuously introduced, and the totally base pressure was 5 kPa as a protective gas. Surrounding the sample, Mg slugs were settled to offer a high Mg vapor. The Mg/B films were heated to 710 °C and kept for 1–12 min. The crystallinity of the samples was characterized by X-ray diffraction (XRD). Film surface morphology was examined with atomic force microscope (AFM) and scanning electron microscope (SEM). Film thickness was measured on the cross section of film by SEM. Standard four probe method was used for the electrical measurement. The critical current density was measured on a 40 μm (width) × 250 μm (length) microbridge patterned by wet chemical etching. 3 Results and discussion In the very beginning, we annealed our 20–80 nm-thick sapphire-substrate Mg/B multilayer films in a Ta tube with high-purity Mg and Ar atmosphere [10] . Fig. 1 shows the annealing temperature dependence of transition temperature, T c onset with 80 nm, 60 nm, 40 nm and 20 nm-thick films. For 80 nm-thick, the most suitable annealing temperature appears to be 650–850 °C. Higher annealing temperature above 850 °C and lower temperature below 650 °C results in the lower T c onset value. We can also see that T c onset drops obviously as the film thickness decreases. The 20 nm film’s R–T curve is shown in Fig. 2 (the triangular symbols), which starts its conductivity transition at 25 K but does not show 0 Ω resistively as low as 10 K. It indicates that the initial growth layer contains significant amount of impurity phases, which may come from the inhomogeneous mixture of Mg and B. So we doubled the precursor layer number and halved the layer thickness. The circular symbols in Fig. 2 show its resistively-temperature curve. We obtained the 20 nm MgB 2 film with the critical temperature up to T C (0) = 12.3 K. We noticed some disadvantages existed in the annealing process described above. A little oxygen was introduced into the annealing chamber with argon gas, while Mg drops appeared degrading the surface. To overcome the deficiencies above, we continuously purged with pure H 2 gas as the background atmosphere. The reducing H 2 ambient at annealing temperature helps to avoid oxygen contamination. Extra Mg was taken away by H 2 flow in the gas phase. What’s more, we changed the substrate from sapphire (0 0 1) to SiC (0 0 1), and the lattice mismatch between MgB 2 and substrate reduces from 11% to 0.1%. R–T curve is shown in Fig. 2 (the square symbols) and the T c onset increases as high as 30.1 K. Different annealing time was attempted ( Fig. 3 ). A sufficiently high critical temperature was obtained with an annealing time as short as 1 min. It indicated a rather fast reaction between B and Mg. Further prolongation of the annealing time led to a decrease of T C and increase of Δ T C . After annealing for 9 min, little protective Mg slugs were residual. Lacking Mg vapor annealing process led to the decomposition of MgB 2 [18] . In the following experiment, we chose 1–4 min as our annealing time. Film thickness was examined by SEM on the film cross section. We can see a continuous layer on the substrate in Fig. 4 a, whose thickness is 18.2 nm. Fig. 4 b–d shows the SEM image of MgB 2 film surface on different substrates. We can see that the appearance of the Mg droplets was effectually eliminated. Films on MgO (1 1 1) and SiC (0 0 1) substrates are built up by grains less than 10 nm. However, 50 nm hexagonal-shaped grains appear in films on sapphire (0 0 1) substrates. AFM measurement was performed on SiC substrate film. The RMS roughness appears to be 2.68 nm. The different film components and epitaxial relationships were assessed by XRD ( Fig. 5 ). The three samples mainly show MgB 2 (0 0 2) peak. It clearly shows that our 20 nm-thick films contain preferentially c-oriented MgB 2 grains. In Fig. 5 b, we can see MgO (1 1 1), (2 2 2) peak in sapphire-substrate sample. These peaks may come from a MgO layer existing between sapphire substrate and MgB 2 . The MgO layer acting as a buffer layer can make up for the lattice mismatch between sapphire and MgB 2 [19] . The resistivity-temperature curve was measured by standard four point technique ( Fig. 6 ). It shows that residual resistivity ρ (40 K) of sapphire substrate film is 12.8 μΩ cm, much higher than that of MgO and SiC, which is about 4.5 μΩ cm. The T c onset of the film on MgO and SiC substrate is 33.0 K and 35.6 K respectively, a little higher compared to the same thickness MgB 2 thin film prepared by MBE [14] or co-evaporation [16] , but lower than HPCVD [20] . We attributed this suppressed T C to the incomplete crystallization which reflects in the XRD spectrum ( Fig. 5 ) too. Only week MgB 2 (0 0 2) peaks appear, suggesting a poor epitaxial growth. For the films on sapphire substrate, T c onset reduces to 17 K, and does not show zero resistance as low as 10 K. It is also confirmed in the XRD spectrum. In Fig. 5 b, MgB 2 (0 0 2) peak is much weaker than that of other substrates while MgO peaks are quite strong. The formation of MgO phase can be ascribed to the presence of residual oxygen in annealing chamber, which leads to the poor T C value. Also the R–T curves indicate that SiC (0 0 1) and MgO (1 1 1) substrates are more suitable for 20 nm MgB 2 films. Critical current density was measured on 40 μm wide bridge. In Fig. 7 , we can see that our films have quite high self-field critical current density, which is J C (10 K) = 4 × 10 6 A cm −2 . The critical current density slightly falls down as temperature goes up to 30 K. At 30 K, it’s still about J C (30 K) = 2 × 10 6 A cm −2 . It indicates a quite well connectivity between grains in our 20 nm-thick MgB 2 films. 4 Conclusion We described the procedure for obtaining high quality 20 nm-thick MgB 2 films using ex - situ annealing from Mg/B multilayer method. Mg vapor flow was used to keep MgB 2 thermodynamically stable, as well as H 2 to suppress the oxidation of Mg. With optimal annealing time 1–4 min, we found that SiC (0 0 1) and MgO (1 1 1) substrates were more suitable for growing 20 nm MgB 2 films than sapphire substrate. 20 nm-thick c-oriented MgB 2 films were obtained on SiC (0 0 1) and MgO (1 1 1) substrate with T c onset of 35.6 K and 33.0 K respectively. The critical current density for the 20 nm-thick film on SiC (0 0 1) substrate was 2 × 10 6 A cm −2 at 30 K. Our work shows a potential method to prepare ultrathin MgB 2 thin films for future application in SSPDs. Acknowledgments Project 10874005, 10874004 supported by National Natural Science Foundation of China. Project 2102023 supported by Beijing Natural Science Foundation. References [1] J. Nagamatsu N. Nakagawa T. Muranaka Y. Zenitani J. Akimitsu Nature 410 2001 63 [2] H. Schmidt J.F. Zasadzinski K.E. Gray D.G. Hinks Phys. Rev. Lett. 88 2002 127002 [3] A.W. Kleinsasser IEEE Trans. Appl. Supercond. 11 2001 1043 [4] D.K. Finnemore J.E. Ostenson S.L. Bud’ko G. Lapertot P.C. Canfield Phys. Rev. Lett. 86 2001 2420 [5] X.H. Zeng A.V. Pogrebnyakov M.H. Zhu J.E. Jones X.X. Xi S.Y. Xu E. Wertz L. Qi J.M. Redwing J. Lettieri V. Vaithyanathan D.G. Schlom Z.K. Liu O. Trithaveesak J. Schubert Appl. Phys. Lett. 82 2003 2097 [6] B.H. Moeckly K.E. Kihlstrom A.T. Findikoglu D.E. Oates IEEE Trans. Appl. Supercond. 15 2005 3308 [7] K. Ueda M. Naito Appl. Phys. Lett. 79 2001 2046 [8] J. Kim R.K. Singh N. Newman J.M. Rowell IEEE Trans. Appl. Supercond. 13 2003 3238 [9] P. Kus A. Plecenik L. Satrapinsky Y. Xu R. Sobolewski Appl. Phys. Lett. 16 2002 2199 [10] W.N. Kang H.J. Kim E.M. Choi C.U. Jung S.I. Lee Science 292 2001 1521 [11] M. Paranthaman C. Cantoni H.Y. Zhai H.M. Christen T. Aytug S. Sathyamurthy E.D. Specht J.R. Thompson D.H. Lowndes H.R. Kerchner D.K. Christen Appl. Phys. Lett. 78 2001 3669 [12] A.T. Matveev J. Albrecht M. Konuma B. Stuhlhofer U. Starke H.U. Habermeier Supercond. Sci. Technol. 18 2005 1313 [13] A.T. Matveev J. Albrecht M. Konuma G. Cristiani Y. Krockenberger U. Starke G. Schütz H.U. Habermeier Supercond. Sci. Technol. 19 2006 299 [14] H. Shibata T. Maruyama T. Akazaki H. Takesue T. Honjo Y. Tokura Physica C 468 2008 1992 [15] H. Shibata M. Asahi T. Maruyama T. Akazaki H. Takesue T. Honjo Y. Tokura IEEE Trans. Appl. Supercond. 19 2009 358 [16] H. Shimakage M. Tatsumi Z. Wang Supercond. Sci. Technol. 21 2008 095009 [17] E. Monticone C. Portesi S. Borini E. Taralli M. Rajteri IEEE Trans. Appl. Supercond. 17 2007 222 [18] Z.K. Liu D.G. Schlom Q. Li X.X. Xi Appl. Phys. Lett. 78 2001 3678 [19] S.D. Bu D.M. Kim J.H. Choi J. Giencke E.E. Hellstrom D.C. Larbalestier S. Patnaik L. Cooley C.B. Eom J. Lettieri D.G. Schlom W. Tian X.Q. Pan Appl. Phys. Lett. 80 2002 1851 [20] Y.Z. Wang C.G. Zhuang X. Sun X. Huang Q. Fu Z.M. Liao D.P. Yu Q.R. Feng Supercond. Sci. Technol. 22 2009 125015