Crystal structure and stability of T₁ precipitates in aged Al–Li–Cu alloys

A crystal structure of the T ₁ phase (Al₂CuLi), commonly observed in aged Al–Cu–Li alloys containing ~3%Cu and 2%Li, is proposed. It is an ‘ideal’ crystal structure in the sense that it predicts X-ray peaks with intensities in fair agreement with those reported semiquantitatively by Hardy and Silcock, but for which additional structure refinement is not warranted. With the ideal crystal structure as a basis, the expected electron diffraction patterns for the zone axes [00l], [110], [111], [112], [013], and [114] have been predicted. These are demonstrated to be in excellent agreement with those observed experimentally. However, in addition to the structure factor for the unit cell, multiple diffraction, the T ₁ plate thickness and shape, the number of T ₁ variants contributing to a given reflection, and possibly faulting within the T ₁ particles all contribute to the intensities of the diffracted electron beams. The roles of plate thickness and multiple diffraction are particularly important since many of the T ₁ reflections observed are not centred on T ₁ reciprocal lattice points lying exactly in the zero order Laue zone of the matrix. The observed T ₁ reflections are therefore frequently a consequence of the intersections of streaks of diffracted intensity with the reflecting sphere. Since all the observed diffraction effects are attributable to T ₁ precipitates there is no need to invoke the existence of a precursor T ₁′ phase, as has been recently proposed. It is concluded that there is no evidence for the existence of a metastable plate-shaped precipitate preceding T ₁ particles in the aging sequence of the alloy.