Diffusion on Demand to Control Precipitation Aging: Application to Al-Mg-Si Alloys

S. Pogatscher, H. Antrekowitsch, M. Werinos, F. Moszner, S. S. A. Gerstl, M. F. Francis, W. A. Curtin, J. F. Löffler, and P. J. Uggowitzer

Phys. Rev. Lett. 112, 225701 – Published 6 June 2014

Abstract

We demonstrate experimentally that a part-per-million addition of Sn solutes in Al-Mg-Si alloys can inhibit natural aging and enhance artificial aging. The mechanism controlling the aging is argued to be vacancy diffusion, with solutes trapping vacancies at low temperature and releasing them at elevated temperature, which is supported by a thermodynamic model and first-principles computations of Sn-vacancy binding. This “diffusion on demand” solves the long-standing problem of detrimental natural aging in Al-Mg-Si alloys, which is of great scientific and industrial importance. Moreover, the mechanism of controlled buffering and release of excess vacancies is generally applicable to modulate diffusion in other metallic systems.

Received 18 November 2013

DOI:https://doi.org/10.1103/PhysRevLett.112.225701

Authors & Affiliations

S. Pogatscher^1,2,*, H. Antrekowitsch², M. Werinos², F. Moszner¹, S. S. A. Gerstl³, M. F. Francis⁴, W. A. Curtin⁴, J. F. Löffler¹, and P. J. Uggowitzer¹

¹Laboratory of Metal Physics and Technology, Department of Materials, ETH Zurich, 8093 Zurich, Switzerland
²Institute of Nonferrous Metallurgy, Montanuniversitaet Leoben, 8700 Leoben, Austria
³Scientific Center for Optical and Electron Microscopy ScopeM, ETH Zurich, 8093 Zurich, Switzerland
⁴Institute of Mechanical Engineering, EPFL, 1015 Lausanne, Switzerland

^*stefan.pogatscher@mat.ethz.ch

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 112, Iss. 22 — 6 June 2014

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
Evolution of hardness during RT storage after quenching for the Al-Mg-Si alloy AA6061 with and without Sn addition. The increase in hardness is retarded with increasing the amount of Sn added. Sn additions above the solubility limit ( $\sim 100$ at.ppm) are marked with an asterisk.
Reuse & Permissions
Figure 2
Analysis of solute clustering during RT storage. Results are obtained by APT for Sn addition of $200^{*}$ at.ppm and 2 weeks of RT storage after quenching. A 3D atom map of Mg, Si, and Sn (magnified) is shown. No solute clusters are recognizable by eye. (For better visibility the reader is referred to the video included in the Supplemental Material [18].) An analysis of NN distances of Mg, Si, and Sn revealed no significant difference between the measured and a calculated randomized distribution of these elements.
Reuse & Permissions
Figure 3
Evolution of hardness during artificial aging at 443 K with and without Sn addition for various periods of RT storage. Data generated from laboratory scaled samples without Sn addition and RT storage $< 60 s$ are shown as a “best-case” reference, where the maximum kinetics and hardening potential is achieved. This best-case scenario cannot be realized in a commercial process, where the material undergoes a significantly longer RT storage (see the commercial case in the figure; a similar curve is obtained for RT storages of 2 weeks). The negative effect of RT storage can be eliminated by minute additions of Sn, which enhance kinetics and hardness (red arrows).
Reuse & Permissions
Figure 4
Artificially aged microstructure. Data obtained by APT for Sn addition of $200^{*}$ at.ppm and artificial aging of 12 h at 443 K after 2 weeks of RT storage. A 3D map of Al and Sn atoms is shown. MgSi precipitates are indicated by semitransparent isoconcentration surfaces of Mg and Si (4 at.%). Sn is found within the matrix and the MgSi precipitates and forms no individual precipitates (see magnified inset).
Reuse & Permissions
Figure 5
(a) DFT results for the $Sn − V$ configuration. The binding energy and the migration barriers relevant to assess Sn diffusion within the context of the five-frequency model are shown. (b) Comparison of the rate reduction $R$ factor calculated from Eq. (2) using two different values for $Δ E_{S V}$ (as calculated, and as to fit experiments) and experimental $R$ values deduced from Figs. 1 and 3, for Sn at concentrations of 40, 70, and/or $100$ at.ppm at RT and artificial aging temperature. The model captures the major trends shown experimentally: $R$ is significant at RT and at the Sn solubility limit, but is fairly small at the artificial aging temperature.
Reuse & Permissions

Physical Review Letters