Materials Science Forum Vols. 519-521

Abstract: In this work, the bake-hardening (BH) response of an Al-3.0Mg-1.0Cu (in mass%) alloy has been improved by the small addition of Ag as a good example of our proposed Nanocluster Assist Processing (NCAP) technique. From the detailed observation through high resolution transmission electron microscopy (HRTEM), it is found that the origin of the increased hardness in the Ag-added alloy is attributed to the densely and uniformly formed Z phase at the expense of Guinier-Preston- Bagaryatsky (GPB) zones and the S’ phase. It is new findings that the Z phase is formed even in the ternary alloy although the chemical composition lies in the (α+S+T) phase field. Based on the threedimensional atom probe (3DAP) results, furthermore, it is suggested that nanoclusters of Mg, Ag and/or Cu provide effective nucleation sites for the Z phase, whereas nanoclusters of Mg and Cu do less. Such unique characteristics of Ag are clearly seen in the newly constructed interaction energy map (IE map).

Abstract: In this work, the crystal structure of the rod-shaped precipitate in aged Al -1.0 mass% Mg2Ge alloy at 523 K has been investigated by high resolution transmission electron microscopy (HRTEM), electron diffraction technique and energy dispersive X-ray spectroscopy (EDS). The rod-shaped precipitate in its alloy has showed the similar arrangement of bright dots in its HRTEM images and selected area diffraction pattern (SADP) to those of the b’-phase in Al-Mg2Si alloy. But a lattice constant of this precipitate in Al-Mg2Ge alloy was slightly larger than the b’-phase in Al-Mg2Si alloy. In addition, the new metastable phase has been found out in Al-Mg-Ge alloy.

Abstract: The effects of composition and temperature on the ageing response and the microstructural development during ageing treatment of a series of dilute 6xxx series alloys have been investigated. The alloys contained between 0.22 and 0.79 wt% Si and 0.20 and 0.51 wt% Mg. Some of the alloys were copper-free, 0.001-0.002 wt% Cu, while others contained additions of 0.1 wt% Cu. Some of the alloys were ‘balanced’ while others contained excess Si (ExSi). The effects of solution treatment temperature and artificial ageing (T6) on the precipitation process were investigated using various techniques, including differential scanning calorimetry (DSC), transmission electron microscopy (TEM) and analytical transmission electron microscopy (ATEM). MT DATA has been used to predict the phase relationships as a function of temperature and the MT DATA predictions have been compared with the phases observed by DSC and ATEM. The morphology and crystal structures of the precipitates formed were monitored by TEM. The results showed a correlation between the composition and the ageing response of the alloys.

Abstract: The precipitation behavior of Mg and Si during storage at RT in Al-Mg-Si alloys pre-aged at 90°C was studied using a tensile test and differential scanning calorimetry (DSC) measurement. Specimens were solutionized at 530°C, water-quenched and then pre-aged for 2, 6 and 12 hours at 90°C during which small precipitates were formed. In the pre-aged alloy, the strengthening rate at RT has two stages. In the initial stage, the yield strength increases slowly with the aging time and in the final stage, it increases rapidly. In the initial stage, the strength in the pre-aged alloy is smaller than that in the non pre-aged alloy, while in the final stage, the strength in the pre-aged alloy is larger than that in the non pre-aged alloy. Furthermore, the period of the initial stage is dependent on the pre-aging period at 90°C. The DSC curves of alloys in the initial stage do not show the presence of clusters, while those in the final stage do. It seems that in the initial stage Mg and Si atoms accumulate around the small precipitates that have been formed in pre-aging at 90°C while in the final stage, the clusters of these atoms are formed.

Abstract: Specimens of three Al-Mg-Si alloys, 6060, 6005 and 6082, were solution heat treated, stored at different temperatures for different time, and artificially aged. Properties were measured before and after artificial ageing. The natural ageing response of the alloys is dependent on the storage temperature. Decreasing storage temperature leads to a delayed onset of natural ageing, but also to a higher strength after prolonged ageing, particularly for lean alloys such as 6060. The temperature and time of intermediate storage between solution heat treatment and artificial ageing has a significant effect on the strength of the artificially aged material. For the 6005 and 6082 alloys the processes that take place during natural ageing lead to a reduced strength after artificial ageing.

Abstract: The formation of nano-scale clusters (nanoclusters) prior to the precipitation of the strengthening β" phase significantly influences the two-step aging behavior of Al-Mg-Si alloys. In this work, two types of nanoclusters are found to be formed at different temperatures. The characterization of these two nanoclusters has been performed from the viewpoints of composition and thermal stability using a three-dimensional atom probe (3DAP) and differential scanning calorimetry (DSC). Mg-Si co-clusters formed at room temperature (RT), Cluster(1), play a deleterious role in the subsequent formation of the β" phase because of the high thermal stability even at the bake-hardening (BH) temperature of 443K. In contrast, the nanoclusters formed by pre-aging at 373K, Cluster(2), are effective in the formation of the refined β", suggesting that Cluster(2) transforms more easily into the β" phase than Cluster(1). The quantitative estimation of the chemical compositions of the two nanoclusters suggests that the Mg/Si ratio is one of the key factors in addition to the internal structures consisting of Si, Mg and probably vacancies. The detailed two-step aging mechanism in Al-Mg-Si alloys is proposed based on the characteristics of the two types of nanoclusters.

Abstract: In Al-Cu-Mg with compositions in the α+S phase field, precipitation hardening is a twostage process. Experimental evidence shows that the main precipitation sequence in alloys with Cu contents in excess of 1wt% is involves Cu-Mg co-clusters, GPBII/S'' and S. The first stage of the age hardening is due to the formation of Cu-Mg co-clusters, and the hardening can be modelled well by a modulus hardening mechanism. The appearance of the orthorhombic GPBII/S'' does not influence the hardness. The second stage of the hardening is due to the precipitation of S phase, which strengthens the alloy predominantly through the Orowan looping mechanism. These findings are incorporated into a multi-phase, multi mechanism model for yield strength of Al-Cu-Mg based alloys. The model is applied to a range of alloys with Cu:Mg ratios between 0.1 and 1 and to heat treatments ranging from room temperature ageing and artificial isothermal ageing to rapid heating to the solution treatment temperature. The predictive capabilities of this model are reviewed and its constitutive components are compared and contrasted with a range of other methods, such as the Kampmann-Wagner and JMAK models for precipitation as well as the LSW model for coarsening.

Abstract: The nonequilibrium δnon-phase was originally investigated in the work[1] when studying the ageing processes of the 1424 alloy ( Al-Li-Mg-system). As shown in the work[2] this phase may also be formed during the cooling from the temperature of SSHT. The δnon-phase precipitates are also present in 1420 alloy of the same system[3]. The basic structure investigations were carried out on the 1424 alloy sheets aged in accordance with the regime: 125°C, 32 hrs. The investigations were performed by TEM using the JEOL JEM 200CX microscope. The diffraction patterns from the δnon-phase on the crystallographic axes of zones close to <110>, <111>, <112> were obtained. It can be seen that the regular reflex networks appeared after the deviation from axes of zones by 3-4˚. It was established that the lattice parameters of δnon-phase and its orientation relation with the matrix can be approximately described by the following way: aδnon≈ aα/2[112] b δnon ≈a α/2[110] cδnon≈ a α [111] where aα- the period of the FCC- lattice of solid solution. The model of the crystallographic structure of the δnon-phase precipitates is proposed. On the basis of this model the mechanism of the δnon-phase formation is discussed. The late ageing stages are analyzed and it was shown that the δnon-phase particles are the nuclei for S1(Al2LiMg)- phase.

Abstract: The structure of Al-Li-Mg system alloy 1420, containing a small quantity of Sc, Zr, Ti was investigated in cast, homogenized, hot-pressed, quenched and aged conditions, using the methods of optical metallography, transmission electron microscopy and X-ray examination. An existence of areas, having fine grains (20-30 nm)- "Ultrafine Grain areas"(UFGA) was observed in all the investigated conditions. UFGA are located on the boundaries, sub boundaries and S1(Al2MgLi) phase particles. UFGA can also form near the particles of crystallization origin. These areas have a complex phase composition. Inside the UFGA the particles of S1(Al2MgLi) phase and also δnon-phase, investigated in [1] are always present whereas δ'(Al3Li) precipitations are absent. These areas are formed during crystallization and hot deformation. Their composition changes during the treatment. The nature of these changes is considered.

Abstract: One of the challenges for the Aluminium industry is to reduce the costs and lead times of the development of novel alloys. This can be achieved by applying increasingly sophisticated models to predict the microstructures and properties of novel chemistries and processing routes. At Corus RD&T, several physically based microstructure models and one process model have been developed and integrated into a Through Process Model (TPM). The TPM presented here is constructed from microstructural sub-models that predict precipitation, work-hardening, recovery and recrystallisation. Furthermore, there is a finite difference based process model that predicts the local process variables like strain, strain rate and temperature. The final sub-model translates the predicted microstructures into product properties. In this paper the integrated model has been applied to the production chain of brazing sheet (AA3103) covering all steps from homogenisation to the braze cycle as applied by the manufacturers of for instance heat exchangers. The model predictions have been verified by comparing them to the results of full-scale a plant trial. Microstructure and mechanical properties were experimentally characterized and predicted at various production steps. Due to the limited space, here, only the results of the through process modelling on microchemistry are presented. Nevertheless it can be concluded that the (fully predictive) results of the models compare well with those found experimentally which opens up the option to use such models for alloy development.