![Show Menu](styles/mobile-menu.png)
![Page Background](./../common/page-substrates/page0098.jpg)
Figure 6
: Cooling ranges of intermetallic phases calculated by SiStruc
®
. The temperatures
delimiting the melting range are written on the side of each bar
It is not true that the longer the melting range, the more phase will be present. The
amount of some elements in the material will be the determining factor. Ca and Ti are
present in a lower extent compared to Fe and Al. Ca is limiting the quantity of Ca-
containing phases, and Ti does the same for the TiFeSi
2
phase.
4.2 Refining and segregation effects in the annealed sample
Changes in composition were expected in the annealed sample. Silicon was molten in
air. A white layer was noticed on the top of the cast, once the sample was extracted
from the furnace. This layer is generated by the oxidation of the material during the
melting step. Elements diffused to a slag phase from MG-Si. The slag will contain
mainly Si, Al and Ca oxides. All these elements have negative
ǻ
G° at 1450°C [20],
which is the highest temperature reached during the treatment. Ca also starts to
evaporate at temperatures slightly above 1450°C. This transformation could have
happened as well, thus further reducing the content of calcium.
Minor oxidation effects could have altered the quantity of Fe and Ti. However,
their
ǻ
G for the oxidation reaction is less negative than silicon. Once Si becomes
solid, elements are not able to diffuse rapidly to the slag, hence the lower distribution
coefficients for Ti and Fe compared to Al and Ca [21]. The segregation effect
becomes the most plausible hypothesis for Fe and Ti decrease and displacement in the
sample (Figure 7). The material for the chemical analysis was taken from the sides of
the cast and not from the center. The segregation is controlled by the heat exchange
coming from the lateral sides of the mold, as well as from vertical solidification. A
solidification front is developing from the bottom with a slower velocity compared to
the solidification velocity from the sides and from the top. This can be said since the
sample was standing on a warm surface. The temperature gradient developed from the
bottom was therefore small.
ϭϬϱϮ
ϵϴϲ
ϵϴϲ
ϵϭϯ
ϵϬϰ
ϴϵϱ
ϴϱϭ
90