If you can’t split it into components with any known method and if you can reproduce its properties with different samples there is a decent chance it’s an element. That process is not without issues and has produced some “elements” that were later shown to be a mixture of more than one element (e.g. “columbium”, later split into niobium and tantalum).
Many elements also follow predictable patterns, so if you have something that fits in a gap it’s more likely to be an element instead of a compound.
Latest Answers