Papers by Author: Michel Gerland

Abstract: Aluminum alloys, widely used for constitutive parts of aircrafts are confronted to a wide range of temperature depending on altitude and climate, from room temperature on the ground down to some 223K at high altitude. The fatigue crack growth behavior of two new generation aluminum alloys, 2024A in T351 temper and 2022 in T351 and T851 tempers, have been investigated at both temperatures. It is shown that temperature and air humidity do not affect the crack growth resistance of the peak aged 2022 while these two parameters widely influence the crack growth in the under-aged alloys which exhibit in cold air a crystallographic retarded propagation similar to that in vacuum. The respective role of microstructure, temperature, atmosphere residual humidity and crack closure is discussed on the basis of a preexisting modeling framework for environmentally assisted fatigue.

Abstract: sothermal mechanical spectroscopy measurements were performed in an Al-51 at % Zn alloy at various temperatures below and above the eutectoid transition temperature: during a heating the α-β eutectoid mixture changes into α solid solution at 550 K. Damping experiments were performed in a very large frequency range (10^-5– 50 Hz) between room temperature and 673 K. Internal friction spectra performed between 200 K and 540 K, exhibit two thermally activated relaxation peaks (P1 and P2). P1 decreases and disappears with the increase of measurement temperature while P2 appears and increases. P2 totally disappears above the eutectoid transition temperature. Above 550 K, a new peak (P3) is evidenced at very low frequency. The relaxation parameters of P3 (limit relaxation time τ₀ = 9×10^-7 and activation energy H = 105 kJ/mole (1.1 eV)) allow to associate this peak with the motion of sub grain boundaries. P1 and P2 (τ₀ ≈ 10^-7 and H ≈ 70 kJ/mole (0.75 eV) for both peaks) are associated with a thermally induced atom diffusion across the α-β interface.

Abstract: Al-12 wt% Mg alloys have been studied by isothermal mechanical spectroscopy. The samples were quenched then annealed at various temperatures. Experiments were performed in a very large frequency range (10^-4 Hz – 50 Hz) between room and solidus temperatures. For each temperature of measurement, experiment started after complete microstructure stabilization of the sample and therefore the transient effects due for instance to β (( and β′ precipitation were not observed. Nevertheless, a new relaxation effect was obtained in the reversion temperature range. This effect is not thermally activated. It is maximal at about 0.1 Hz and increases with the temperature of measurement. It completely disappears after annealing at solid solution temperature and successive slow cooling and therefore is linked to the β precipitates. This effect is interpreted as a phase transformation at the precipitate surface induced by the applied stress.

Abstract: Internal friction peaks observed in single or polycrystals are clearly due to a dislocation relaxation mechanism. Because a sample observed by transmission electron microscopy (TEM) often exhibits in the same time various dislocation microstructures (isolated dislocations, dislocation walls, etc.) it is very difficult to connect the observed relaxation peak with a particular dislocation microstructure. Using isothermal mechanical spectroscopy (IMS), it is easier to compare, for instance, the evolution of a relaxation peak with measurement temperature to the microstructural evolution observed by in-situ TEM at the same temperatures. IMS was used to study a relaxation peak in a 5N aluminium single crystal firstly 1% cold worked and then annealed at various temperatures. TEM experiments performed in the same material at various temperatures equal to the temperatures used for the damping experiments made possible to link this internal friction peak with a relaxation effect occurring inside dislocation walls. In two other experiments in a 4N aluminium polycrystal and in a metal matrix composite with SiC whiskers, it is shown that the observed relaxation peaks are connected to the motion of dislocations inside polygonization boundaries in the first case and in dislocation pile-ups around each whisker in the second one. Theoretical models proposed to explain such relaxation peaks due to a dislocation motion inside a dislocation wall or network are discussed.