*3.4. Hot Ductility Weldability Test Results*

Figure 7 describes the area reduction percentage determined in on-heating and oncooling hot ductility tests for the five casting heats. Continuous lines correspond to the grade 3 polynomial fitting of experimentally determined on-heating test results, whereas dashed lines depict on-cooling behaviour. Looking at on-heating curves, the ductility at temperatures between 950 and 1000 ◦C was higher than 60% for the five moulds, showing comparatively higher values in the case of mould E and O (low Si content). Between 1000 and 1050 ◦C, ductility started to drop, and at 1150 ◦C it was already below 2%, which meant that the capability to deform without breaking had been completely lost. Again, at intermediate 1100 ◦C, moulds E and O (low Si content) showed comparatively better performance in terms of ductility, which was the reason why fitting curves were slightly displaced towards higher temperatures. This means that the onset of ductility drop in these two moulds was delayed to some extent.

**Figure 7.** On-heating and on-cooling curves in terms of area reduction percentage and testing temperature for moulds E, O, and P (**a**), and N and NP (**b**).

Differences between casting heats were much more evident in on-cooling curves. Mould O (low Si content) presented remarkable ductility recovery behaviour, quickly reaching an area reduction value of 64% after testing at 1100 ◦C. Note that the thermal sequence of on-cooling hot ductility tests involved fast heating to peak temperature of 1195 ◦C and cooling down to the corresponding test temperature. Peak temperature was selected after defining NST in mould E samples that reached 1263.5 ◦C ± 5.8 ◦C. It was decided to limit peak temperature in on-cooling test to 1195 ◦C in order to ensure repeatable and stable ductility recovery behaviour.

After reaching ductility values or original parent material, mould O showed a ductility drop at lower testing temperatures down to 27% at 950 ◦C. This drop was not so remarkable either in mould E (from 69% to 46%) or high Si P (from 46% to 40%). Thus, it is quite clear that the ductility recovery rate is strongly related to the Si content of the alloys.

Moulds N and NP with slower cooling rates in the casting process depicted a completely different ductility recovery performance during on-cooling tests. In these two heats, restored ductility values did not surpass 14% and 24%, respectively, and the slope of the curves was drastically reduced. Calculated DRT and brittle temperature range (BTR) values are included in Table 6. BTR is the difference between peak temperature employed in

on-cooling tests and determined DRT, at which 5% of area reduction is recovered. BTR is a parameter widely used to conclude on the hot cracking susceptibility of superalloys [2,9,26].


**Table 6.** Weldability parameters obtained from hot ductility tests.
