Hydrogen is generally considered a non-metal, and a non-metal and a metal often makes a salt. However, lithium hydride seems like a special case. The wikipedia article does not clearly state if the compound is a salt or not, only describing it as "salt-like".

Structurally, it makes sense to classify it as a salt, due to the network structure, but are there any clear definition to settle where the compound belongs?


  • 1
    $\begingroup$ Many metals form hydrides; what's so special about lithium? $\endgroup$ Feb 18, 2016 at 9:40
  • $\begingroup$ @IvanNeretin Can hydrides be considered salts? $\endgroup$ Feb 18, 2016 at 9:42
  • $\begingroup$ Salt doesn't mean ionic compound. And there's no purely ionic or covalent compound. $\endgroup$
    – Mithoron
    Feb 18, 2016 at 18:29
  • $\begingroup$ Ah! A commonly used classification may be the best way to handle this then $\endgroup$ Feb 18, 2016 at 21:21

2 Answers 2


The Wikipedia article Salt (chemistry) starts off with "In chemistry, a salt is an ionic compound that results from the neutralization reaction of an acid and a base." This seems like a pretty good Chemistry 101 definition. So, in essence, this means having an anion and a cation in the salt, which further implies ionic bonding. This seems like a pretty reasonable place to start.

I'd think that the definition could be tweaked to say "a salt is a solid ionic compound..." That would indicate that one "molecule" isn't a salt. So one molecule of NaCl in the gas phase is a molecule, not a salt. I also have trouble thinking of molten NaCl as a salt. So I'd describe it as a "molten salt."

Kind of like way I think German works. I'd expect that German would have some compound word for "molten salt" which better denotes that there is one new idea. So in English if a "salt" is a solid that has a 3D crystal lattice how can a salt be "molten"?

In general, there are a lot more organic salts than inorganic, so dissolving in water isn't a good restriction.

These last paragraphs point out the sorts of problem in trying to make such a definition absolutely unambiguous. You get a one sentence definition, then two pages of qualifications which render the definition useless.

I can't think of a solvent in which you could dissolve lithium hydride and I don't see how you could get lithium hydride as a neutralization reaction, so I wouldn't think of it as a salt. I don't know a better word than "salt-like" to describe such a crystal structure of anions and cations.

  • $\begingroup$ According to your new definition, wouldn't sodium oxide also be a salt? $\endgroup$ Jun 26, 2017 at 14:45
  • $\begingroup$ Wikipedia has edited the article to meet with IUPAC's latest definition. Now, "In chemistry, a salt is a chemical compound consisting of an ionic assembly of cations and anions." $\endgroup$ Sep 19, 2021 at 0:52

Hydrides of metallic elements are ionic compounds. In the case of $\ce{LiH}$ there may be some covalent characters due to the small size of the $\ce{Li+}$ ion. This hydride on electrolysis in molten form gives $\ce{H2}$ on the anode, which is an indication of the presence of hydride ions and, further, this hydride reaction with water or protic acid evolves $\ce{H2}$ gas.

  • 4
    $\begingroup$ This answer is a bit vague. For instance, metal carbonyl hydrides often have particularly non-hydridic "hydride" ligands. When mixed with water, they may form $H_2$, but not before acting as a Lewis acid to form a more electron-rich metal complex (by displacing the carbonyl ligands with water/hydroxide ligands). $\endgroup$ Jan 15, 2017 at 18:59
  • $\begingroup$ Can you give a reference for the hydrogen release? The WP article identifies these compounds as acids, and in fact hydrogen is considered as having oxidation state +1 instead of -1 in such compounds. $\endgroup$ Sep 5, 2021 at 9:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.