Giant transport anomalies in (Fe,Ru)-Cu-Al quasicrystals can be explained by an internal structural model based on two large building blocks, one icosahedral and one layered. The model explains a wide variety of electronic and structural regularities.PACS numbers: 72.15.-v, 61.42.+h The remarkable similarity of the diffraction pattern of a three-dimensional Penrose tiling to those observed for icosahedral quasicrystals at first led to the assumption that the atomic structure corresponds to small-scale simple decorations of each of two types of tiles. x Careful comparison of experimental results with various atomicstructure models suggests, however, that in the absence of relaxation many distinct and context-dependent decorations are required for any reasonable model. In other words, the simple assumption of exact homogeneity on the scale of the quasilattice constant (typically ~5 A) may be shown to be inconsistent with diffraction data. 2 Transport data probe the atomic structure in a different way, but at first it was thought that such data revealed only the presence of the kind of short-range disorder found in amorphous or metallic glasses. The latter typically exhibit at low temperature resistivities 3 of order 50 jnflcm (binaries) or 100-300 /idem (ternaries). However, stable CuI^AU quasicrystals have resistivities 4 as high as 900 /idem, while in stable FeisC^oA^s quasicrystals the resistivity can be as high as 4500 //a cm, with the Hall carrier density ~10~2 of the free-electron value. 5 Purified stable z-Ru-Cu-Al exhibits 6 low-temperature resistivities as large as 30000 pClcm. This means that the electronic mean free path X calculated in the normal way 4 can be as small as 0.1 A, or 30 times smaller than the atomic spacing. Such a result in a metal is clearly nonsensical, and at present there is no material-specific theoretical model which explains this result. In this Letter we propose a novel materialspecific structural model which does explain these anomalies. Our model is phenomenological and heuristic and the large-scale structural variations which we propose have so far not been observed experimentally. It is important to realize that any theoretical model which explains these giant anomalies in a systematic way represents a great advance over prior understanding, and this is the first such model to do so.The electronic properties of quasicrystals composed of s-p metallic elements of similar size may in fact show strong similarities to the properties of metallic glasses and even related crystalline compounds. For example, in the metastable quasicrystal Mg32Alj7Zn32 the resistivity is low (p300K~~100 //(lem) and N(E F ) is close to the free-electron value, similar to amorphous phases of nearby composition, while the temperature dependence of the Hall coefficient resembles that of the crystal. 5 However, the behavior of highly ordered samples of the stable quasicrystals Al-Li-Cu and Al-Cu-Fe seems to depart markedly from the free-electron model in both cases. In Al-Li-Cu N(Ef) measured by specific ...
