Effect of the Growth Conditions on the Spatial Features of Re Nanowires Produced by Directional Solidification Srdjan Milenkovic, Achim Walter Hassel,* and Andre ´ Schneider Max-Planck-Institut fu ¨r Eisenforschung GmbH, Max-Planck-Strasse 1, D-40237 Du ¨sseldorf, Germany Received July 22, 2005; Revised Manuscript Received December 4, 2005 ABSTRACT The effects of the solidification parameters, such as growth rate and temperature gradient, on the distance and diameter of Re nanowires have been examined. Both the spacing and diameter increase with decreasing growth rate and temperature gradient, respectively. The ratio of fiber spacing to diameter is 9.1. In addition, it was demonstrated that the temperature gradient influences interface undercooling in the same way as the growth rate and may be used as an additional parameter to control fiber spacing and diameter. Introduction. One-dimensional nanostructures have attracted significant research interest due to their mechanical, electri- cal, magnetic, and optical properties. Besides, they provide models to study the relationship between electrical transport and optical and other properties with dimensionality and size confinement. Recently, we have developed a novel method for assembling nanowire and nanopore arrays via directional solidification and selective etching of eutectic alloys. 1,2 One of the crucial factors in the synthesis of nanowires is the microstructural control governed by the kinetics of the solidification process. The directional solidification of eutectic alloys under controlled growth conditions generally yields lamellar or fibrous structures, aligned parallel to the direction of heat extraction, i.e., the growth direction. Two important param- eters of eutectic structure, which are influenced by the imposed process parameters, are the phase diameter and the eutectic spacing. The classical theory of eutectic growth was proposed by Jackson and Hunt 3 in mid 1960s. To obtain an analytical solution for the diffusion problem at the solid/ liquid interface, they assumed a planar interface and equal undercooling of both solid phases growing in a coupled mode from a melt of eutectic composition. Jackson and Hunt 3 postulated a relationship between total interface undercooling, the solute concentration field, and the interface curvature. Since their analysis was derived from the Zenner 4 and Tiller 5 treatment of eutectic growth, it was necessary to introduce an arbitrary extremum condition because the problem would have been intrinsically indeterminate without it. Considering the minimal undercooling as the “extremum condition”, i.e., that eutectic growth occurs near or at minimal undercooling, Jackson and Hunt obtained a series of relations between fiber (or lamellae) spacing, solidification velocity and average undercooling. The average undercooling for fiber eutectic structures, being a sum of the constitutional and capillary undercooling could be expressed as where λ is the distance between fibers, ΔT is the undercool- ing of the solidification front, V is the solidification velocity, and K 1 and K 2 are summarized material parameters (see ref 3). The minimum undercooling criterion leads to which is well accepted and has been experimentally verified many times. 6 The formation of eutectic structures was theoretically reviewed in recent years and to a certain extent also verified or confirmed experimentally. Many investiga- tions for eutectic spacing have been performed in the lamellar eutectics, whereas there is little concern for the relationship between the fiber spacing and the controlling parameters in fibrous or rodlike eutectic alloy. Moreover, the relationship between the fiber diameter and the growth parameters has not been reported yet. Considering it of crucial importance for our investigations, to know the range of nanowires that can be produced, this aspect is analyzed as well. This work presents an experimental investigation on the pseudobinary NiAl-Re eutectic system, with the main objective to * Corresponding author: E-mail: hassel@elchem.de. ΔT ) K 1 λV + K 2 λ (1) λ 2 V ) K 2 K 1 (2a) ΔT  V ) 2  K 1 K 2 (2b) ΔTλ ) 2K 2 (2c) NANO LETTERS 2006 Vol. 6, No. 4 794-799 10.1021/nl0514238 CCC: $33.50 © 2006 American Chemical Society Published on Web 04/12/2006