Updated Data and Clarifications

[2025 IEEE SciVis Contest]

Dear all,

there has been a slight update of the contest data,
after we discovered a small mistake in the data conversion.
Please use the updated version here: 
https://zenodo.org/records/14182255

Further, we received some questions, that we would like to provide clarifications to:

"...The in-detail description of the data is not up to date anymore, e.g.the description has information on phosphorus-containing phases, while the data has information on zinc-containing phases instead (not shure yet if something else is different). Can you update this please?"

You are right, P is not included in the uploaded dataset. I apologize for the confusion!

"Two columns have only NAN values (Vf_AL3X and Vf_MG2ZN3). Is that really from the simulation, such that those phases cannot be simulated in general, or is there an error?"

This should come from the calculation. The mixed compositions contain only very small amounts of elements that form these phases. So in most cases, it is probable that the phase does not form and a “nan” value is recorded. There might also be some cases where a calculation error occurred because of inconvenient compositions for which the calculation parameters did not fit (see answer to next question). It is also worthy to note, that the volume fractions, Vf, that are calculated in this dataset, result from non-equilibrium Scheil calculations, which means they simulate a rapid solidification. If we were to calculate the volume fraction for these phases in equilibrium, i.e., with very slow solidification rates, these phases will most certainly form.

"The description talks about not finding an equilibrium, what is meant by that?"

The calculations conducted are thermodynamic calculations in equilibrium. Basically it is about finding minimal Gibb’s energies. Sometimes, e.g., calculation parameters such as start temperatures, etc, are not optimal or when the alloy system contains several element concentrations that are close to 0, it can happen, that the minimal Gibb’s energy and the equilibrium state cannot be found, thus no value will be provided for the output variable that we want to calculate.

"Am I right to assume, that sometimes new phases can be made, which are not added to the dataset? Sometimes the sum of all cases do not add up to 100%".

Please download the new version of the dataset. Now all mixing ratios as well as the mixed compositions have a sum of 100%.

Hope this helps!
Best Tim