27.12.2013 Views

COST 507 - Repositório Aberto da Universidade do Porto

COST 507 - Repositório Aberto da Universidade do Porto

COST 507 - Repositório Aberto da Universidade do Porto

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

2.1 Solid solubility measurements<br />

The <strong>COST</strong><strong>507</strong> effort on investigating the solid solubility of selected alloying elements<br />

in aluminium has been a collaborative project between Pechiney CRV and SINTEF<br />

Materials Technology. The systems targeted are Al-Mn, Al-Mn-Si, Al-Mn-Fe, and Al-<br />

Mn-Fe-Si. Pechiney CRV have been responsible for the casting and heat treatment of<br />

the alloys, as well as determining the Mn content by way of the Thermo Electric Power<br />

(TEP) technique [81Bou, 89Bor]. SINTEF has measured all the elements in solid<br />

solution by using the SIBUT dissolution technique [80Sim, 84Sim] combined with<br />

pertinent analysis techniques. In this fashion the measurement of the Mn content has<br />

been duplicated using two fun<strong>da</strong>mentally different techniques, which ensures reliable<br />

results. The results of these investigations have prompted a réévaluation of the<br />

reliability of the work of [43 Phi] in the solid state. Due to extremely slow reaction<br />

rates in the Mn containing systems, up to 6000 hrs homogenisation time was required<br />

to reach equilibrium. This is longer than can be explained by a small diffusivity of Mn<br />

in solid Al. It is therefore probable that nucleation is sluggish and that the driving<br />

forces for precipitation are small. This could in tum help to explain the discrepancy<br />

between the <strong>COST</strong><strong>507</strong> assessment and the commonly accepted diagrams in all systems<br />

containing Mn.<br />

The experimental tie line determinations have been conducted on the same samples<br />

used in the solvus measurements. By measuring the composition, type and amount of<br />

particles left on the filter after dissolution we are able to combine this with the<br />

corresponding solid solution composition to form an experimental tie line. Amongst<br />

the results of this work on the Al-Mn-Fe system, it was discovered that the Al 12 Mn<br />

phase is stable at 550°C in the pure Al-Mn system. However, an Fe content as low as<br />

20ppm was sufficient to stabilise the Al 6 (Mn,Fe) phase, and bring the sample into the 3<br />

phase field Al + Al^Mn + Al 6 (Mn,Fe). It is believed that as little as 100 ppm will be<br />

sufficient to completely surpress the formation of Al^Mn. These results are in<br />

contradiction to previous beliefs based on the work of Schaefer et al [86Sch] which<br />

suggests that this phase decomposes at temperatures above 512°C, but in agreement<br />

with unpublished work conducted at the CNRS Vitry [96Har].<br />

2.2 Determination of invariant points in Al-Mn-Si<br />

Due to problems in fitting all the invariant points on the liquidus surface<br />

simultaneously, we have experimentally redetermined the eutectic point [93Sim]. Our<br />

coordinate for the eutectic is lower in Mn than that of Phillips, which already belongs<br />

to the lower spectrum of values quoted in the literature. We <strong>do</strong>, however, feel<br />

confident that this measurement is correct since the optimiser in Thermo-Calc seems to<br />

draw this point to lower Mn values. Our values have been obtained by performing area<br />

analysis in the microprobe in regions of the sample occupied by the eutectic structure.<br />

In this way the Mn content should be overestimated if regions which have solidified<br />

preeutectically formed part of the analysing volume.<br />

The delta-AlMnSi phase has also been investigated from the point of view of<br />

determining the position of the invariant point comprising alpha, silicon and delta in<br />

addition to the melt. Our results contradict those reported by [76Mon], who claims the<br />

163

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!