Page 45 - Read Online

Page 8 of 25 Park et al. J Mater Inf 2023;3:5 https://dx.doi.org/10.20517/jmi.2022.37

Figure 5A and B present the DSC response during the heating of samples I and V (note that the results are
corrected by an arbitrary shift of the voltage). Detailed information on the interpretation of DSC signals to
[55]
determine phase transformations can be found in the reference . The first deviation from the baseline is
[56]
observed between 231.5-235.6 °C, corresponding to the melting of pure Sn (231.91 °C) . According to the
phase diagram in Figure 1, this phase transformation is predicted only for sample V from Table 2, indicating
that due to non-equilibrium cooling during the preparation of the samples, the microstructure did not reach
the equilibrium state at ambient temperature. Above 400 °C, only the second heating cycle of the remelted
sample is plotted in Figure 5. Both alloys show identical phase transformations. Again, this transition
sequence is calculated under equilibrium conditions only for a composition of X = 0.727. It can be
Sn
-1
therefore concluded that even by slow cooling in the calorimeter (-30 °C min ), the strong unmixing
tendency of Fe and Sn could not be avoided. The first onset at 517-520 °C may be assigned to FeSn + FeSn
2
↔ FeSn + Liquid and/or FeSn + Liquid ↔ FeSn + Liquid in the phase diagram [Figure 1]. However, this
2
assumption is based only on the calculated phase equilibria. Between 778 °C and 804 °C, the Fe Sn + Liquid
2
3
and/or Fe Sn + Fe Sn two-phase regions exist. Above 907.5 °C, the liquid is in equilibrium with the bcc
2
3
3
5
solid solution, representing the first phase boundary stable for all investigated compositions in the Fe-Sn
phase diagram [Figure 1]. As expected, the onset temperatures in Figure 5 do not systematically depend on
the used sample mass or the applied HR.
The monotectic temperature (bcc + Liquid ↔ Liquid + Liquid ) in Figure 5 is characterized by a peak in the
2
1
DSC signal. It is well known that peak temperatures are generally shifted to higher temperatures with
increased HR and larger sample mass [48,49] . This fact was also observed in the present results. The peak
temperature depends on the liquid and bcc fractions below the monotectic temperature, which are
transformed into two liquid phases when the monotectic line is reached. According to the lever rule in
Figure 1, less bcc is stable in the bcc + liquid two-phase region with increasing X . Due to the lower amount
Sn
of melting solid phase, the peak intensity was reduced and the peak temperature was detected at lower
values of about 1,130 °C; see for comparison 1,150 °C at X = 0.365. Therefore, the actual aim of the pre-
Sn
measurements of identifying the liquidus temperature (Liquid + Liquid ↔ Liquid) could not be achieved.
2
1
-1
Even increasing the HR to 20 °C min and using twice the mass of the samples (400 mg) did not result in a
detectable signal change by crossing the liquidus line above 1,200 °C.
Based on the pre-test, a reliable DSC analysis was possible for melting the intermetallic compound Fe Sn
3
5
(bcc + Fe Sn ↔ bcc + Liquid and/or Fe Sn + Liquid ↔ bcc + liquid). Both phase equilibria are stable at all
5
3
5
3
chemical compositions investigated in the phase diagram. The formation of non-equilibrium
microstructure components can therefore be neglected. The sample mass of 200 mg was selected, and a
heating rate of 10 °C min was defined. The DSC analysis of samples I to V is summarized in Figure 6A.
-1
The bcc + Fe Sn ↔ bcc + Liquid and/or Fe Sn + Liquid ↔ bcc + Liquid phase transformation is obtained at
5
5
3
3
907 ± 1 °C. More significant deviations are observed in the peak of the monotectic temperature, which can
be explained by the different bcc fractions melting during heating, influencing the intensity of the DSC
signal. The present work proposes a temperature of 1,139 ± 5 °C for the bcc + Liquid ↔ Liquid + Liquid
2
1
phase transformation. Figure 6B shows the enlarged section of the miscibility gap. Unfortunately, the
binodal line could not be obtained for any sample composition.
Results of the electromagnetic levitation test
Figure 7 shows the time-temperature profile change for sample I (X = 0.365). The sample was held at
Sn
1,760 °C for some time and subsequently was cooled by blowing He(g). The temperature of the sample
decreased to 1,385 °C. During the cooling, there was no visible change in the slope to the time-temperature
profile, as was expected. Below 1,385 °C, the sample could not be levitated anymore due to the loss of the
levitation force. According to some of the previous reports [30,33,44] , the binodal temperature was higher than