Metastable transformation states of decagonal Al-Co-Ni due to inhibited decomposition

Abstract

The transformation behavior of high-Co decagonal Al-Co-Ni is studied. Starting from stability considerations based on a number of annealing treatments we check theoretical predictions of the stabilization mechanism and transformation mechanism by phason strain analysis and tiling analysis. The quasicrystal transforms in the first stages into nanodomain structures and finally decomposes into a quasicrystal and a normal crystalline phase. Before decomposition, the material is locked in a disordered state exceeding the periodic order allowed for a one-dimensional quasicrystal. This is interpreted as the maximum periodic order achievable without diffusion. From considering different reasons for phason strain we conclude that disorder in tilings related to phase transformations can wrongly be associated with that suggested by the random tiling hypothesis.

Twenty years of structure research on quasicrystals. Part I. Pentagonal, octagonal, decagonal and dodecagonal quasicrystals

“…5D structure model based on the atomic surfaces modelling method; P10, wR ¼ 0.060 for 2767 reflections (Cervellino, Haibach, Steurer, 2002). According to Döblinger, Wittmann, Grushko (2001) the superstructure type II is not a d-phase but corresponds to an orientationally fivefold twinned nanodomain structure. According to several in situ HT X-ray diffraction studies (Baumgarte, Schreuer, Estermann, Steurer, 1997;Steurer, Cervellino, Lemster, Ortelli, Estermann, 2001), the HT-phase is a quasiperiodic phase without superstructure reflections.…”

“…The transformation behavior of a homogenized (1290 K, 43 h) and annealed (1140 K, 384 h, 1125 h, 2718 h) sample with composition Al 72.5 Co 15.5 Ni 12 was studied by SAED and HRTEM (200 kV) (Döblinger, Wittmann, Grushko, 2001). It was found that a homogenous d-phase (superstructure type S1) of this composition transforms via intermediate states (such as 'superstructure type II', which resulted to show first a fivefold twinned 1D QC and then a twinned h4,6i-approximant) to a mixture of the monoclinic X-phase and a d-phase (superstructure type I, unfinished transition probably due to not equilibrated chemical order) after 2718 h at 1140 K. This kind of transformation has been predicted by Steurer (1999Steurer ( , 2000 and Honal, Haibach, Steurer (1998).…”

Continuous transition between decagonal quasicrystal and approximant by formation and ordering of out-of-phase domains

“…These NDS's are not necessarily periodic in three dimensions, multiply twinned 1D QC's 8 or states whose order according to a 1D QC is only partially developed were also observed. 9 There is still a lack of experimental data about the transition of NDS's to larger domains of periodic approximants.…”

“…10,22,23 NDS's were observed in the system AlNi-Co by high-resolution x-ray diffraction 8,24,25 and highresolution transmission electron microscopy ͑HRTEM͒. 9 At compositions close to the binary Al-Co line the phase diagram seems to be very complicated. 26 Pentagonal quasicrystals are reported 27,28 to exist between Al 73 Ni 7 Co 20 and Al 72.5 Ni 10 Co 17.5 .…”

“…With respect to a 2 nm tiling, high-quality variants of a one-dimensional quasicrystal and superstructure S1 can be projected into this acceptance domain. 9,31 In the diffraction pattern linear phason strain causes peak shifts. 32 Thus E can be determined according to k ʈ ϭk id ϩE † k perp , where k id corresponds to the peak position of an ideal quasicrystal which is shifted by the transpose of E to the measured peak position k ʈ .…”