Observation of Superconductivity in Granular Bi Nanowires Fabricated by Electrodeposition Mingliang Tian,* ,²,‡ Jinguo Wang, ²,§ Nitesh Kumar, ²,‡ Tianheng Han, ², | Yoji Kobayashi, ⊥ Ying Liu, ‡ Thomas E. Mallouk,* ,²,‡, ⊥ and Moses H. W. Chan* ,²,‡ Center for Nanoscale Science (CNS), Department of Physics, Department of Chemistry, and Materials Research Institute (MRI), PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802-6300, and Department of Physics, the Hong Kong UniVersity of Science and Technology, Clear Water Bay, Hong Kong, People’s Republic of China Received August 2, 2006; Revised Manuscript Received October 26, 2006 ABSTRACT Bulk rhombohedral Bi at ambient pressure is a well-known semimetal, and its transition to a superconductor has not been observed, at least down to 50 mK. We report that, unlike bulk rhombohedral Bi, granular Bi nanowires with well-defined rhombohedral grains of ∼10 nm diameter, fabricated by electrochemically depositing Bi into porous polycarbonate membranes at ambient pressure, are superconducting with two transition temperatures, T c , of 7.2 and 8.3 K. These T c values coincide with T c values of the high-pressure phases Bi-III and Bi-V, respectively. Analysis of our structural and transport data indicates that the superconductivity in granular Bi nanowires probably arises from grain boundary areas where there are structural reconstructions between the grains showing a preferred orientation within a small angular distribution. It is well-known that bulk rhombohedral Bi at ambient pressure is not a superconductor down to 50 mK. 1 Unlike other materials, rhombohedral Bi can be a metal, semimetal, and semiconductor. Its electronic properties depend sensi- tively on the sample size and geometry (bulk, 1,2 films, 3,4 cylinder or filament, 5 nanowires, 6-10 or tubes 11 ), quality (defects or impurity), and morphology (single-crystal, 1-6 polycrystal 4,8 or granular 9 ). Because of its unusually rich electronic properties, Bi has long been the subject of experimental and theoretical studies. 1-11 At high pressure, however, Bi is superconducting at low temperatures due to the formation of high-pressure metallic polymorphs, namely, monoclinic Bi-II at 2.55 GPa, 12,13 a complex tetragonal Bi-III at 2.7 GPa, 12,14-16 and a body- centered cubic (bcc) Bi-V at 7.7 GPa. 12,16,17 The transition temperatures of these polymorphic phases are respectively 3.9, 18,19 7.2, 18-20 and 8.3 K. 20,21 It was found that these high- pressure phases could be preserved in a metastable state by subjecting the bulk crystal to compression cycles at room temperature and releasing the pressure at helium tempera- tures. 20 When the samples are annealed at a temperature higher than 20 K, the metastable high-pressure phases disappear. In these compression cycles, internal microstresses are usually formed as a result of plastic deformation due to an increase in the number of dislocations and a greater degree of polycrystallinity. 20 At ambient pressure, Bi can be superconducting under three specific conditions: (1) Bi amorphous film with disorder on the atomic scale, fabricated by condensing Bi onto a liquid helium cooled substrate. The T c is around 6.0 K. 22 (2) Bi granular composite films, fabricated by codepositing Bi and matrix gas (A ) Kr, Xe, O 2 , and H 2 ) onto cold substrates at T < 40 K. These composite films consist of rhombohedral Bi grains sur- rounded by insulating solid matrix gas with a concentration ratio of Bi:A ) x:(1 - x) with x ∼ 0.34-0.89. 23,24 The transition temperature, which is generally below 5.5 K, is strongly grain size dependent and is influenced by the surrounding solid matrix. Without the solid gas matrix, a pure Bi granular film is not superconducting. The supercon- ducting property in these granular Bi composite films was initially interpreted by Weitzel and Micklitz 23 as a quantum size effect due to a strongly enhanced density of states at the Bi grain surface. However, Vossloh, Holdenried, and Micklitz 25 presented new data that indicated that the super- conductivity in Bi granular composite films is not related to the quantum size effect. They speculated that the original * Corresponding authors. E-mail address: mut1@psu.edu (Tian), tom@chem.psu.edu (Tom), and chan@phys.psu.edu (Chan). ² Center for Nanoscale Science (CNS), Pennsylvania State University. ‡ Department of Physics, Pennsylvania State University. § Materials Research Institute (MRI), Pennsylvania State University. | Department of Physics, the Hong Kong University of Science and Technology. ⊥ Department of Chemistry, Pennsylvania State University. NANO LETTERS 2006 Vol. 6, No. 12 2773-2780 10.1021/nl0618041 CCC: $33.50 © 2006 American Chemical Society Published on Web 11/10/2006