I am reading some theoretical papers on the defects in semiconductors. Theorists can calculate the formation energy of a certain kind of defect. If the number is too big, this defect is very unlikely to be seen. So what does this “formation energy” mean? Is that the energy necessary to move atoms together to form this defect, which is also the energy necessary to keep the defect structure from dissociating? That's my understanding but I don't know if it is correct or not.