Is mercury(I) chloride ionic or covalent? If I look here, I see arguments either way. The article doesn't mention anything about a cation or an anion.

Some might think there's two mercury ions each 1+ but it seems to me that's not the case, as the article mentions that the chemical is also known as "mercurous chloride". And I see there is such a thing as a mercurious ion, which I understand to be a polyatomic ion with charge of 2+, so that would suggest it's an ionic compound.

This Wikipedia article also mentions that it's "Also known as the mineral calomel" and it says calomel is a salt "mercury salts (such as calomel)". So that indicates to me that it's an ionic compound.

But the article on mercury(I) chloride also says "Hg2Cl2 is a linear molecule". Well, that suggests it's a covalent compound.

I've also heard that it has "a relatively low sublimation point which also implies molecular".

The Wikipedia page also mentions "the Hg–Cl bond length in the linear Hg2Cl2 unit is 243 pm"

I suppose that might suggest covalent, because as I understand it, "bond length" is normally mentioned for covalent bonds rather than ionic bonds.. Though with ionic bonds the concept of bond length could apply though i've heard it's usually called "interionic distance".

Looking at the pictures, it has this one

enter image description here

And I understand that that style of diagram is usually used for ionic compounds..

It shows a unit cell (which seems to show maybe linear molecules)

enter image description here

And as I understand it, a unit cell could apply not just to an ionic crystal, but to a molecular crystal or a network covalent crystal

So I'm really confused as to what the structure is. I can't tell whether it's ionic or whether it's covalent molecules?

  • $\begingroup$ related chemistry.stackexchange.com/questions/37141/… $\endgroup$
    – Mithoron
    Oct 12, 2023 at 20:46
  • 3
    $\begingroup$ There is no clear-cut division between the two. Let that sink in. $\endgroup$ Oct 13, 2023 at 15:05
  • $\begingroup$ Hg2Cl2 crystals are definitely molecular, and HgCl2 too. They still dissociate in water to some degree. $\endgroup$
    – permeakra
    Oct 13, 2023 at 22:30
  • $\begingroup$ @permeakra yes I think it is ionic and molecular. Ionic because it's composed of ions $Cl^-$ --- $Hg_2^{2+}$ --- $Cl^-$ And molecular because those $Cl^-$ --- $Hg_2^{2+}$ --- $Cl^-$ entities are separate/discrete, molecules. No covalent or ionic bond between one of those entities and another - only within. And I suppose if attempted to dissolve in water then it'd separate into $Cl^-$, $Cl^-$, and $Hg_2^{2+}$. The $Hg_2^{2+}$ ion remaining intact as it's bonded covalently. $\endgroup$
    – barlop
    Oct 14, 2023 at 2:04

1 Answer 1


A ionic crystal is one with structure dominated by ionic interactions. The charge separation in $\ce{Hg_2Cl_2}$ is small, with Mullicken charges about 0.2e at most [1]. By comparison, alkali halides have Mullicken charges at about 0.8e [2] It is further evidenced by direct $\ce{Cl-Cl}$ contacts that would be impossible if there was significant charge separation in the molecules.

[1] https://doi.org/10.1016/j.matchemphys.2021.125379

[2] https://doi.org/10.1039/C9RA05190B

Background. Mercury is a weird metal that has unusually high for metal first and second ionization energy at 10.4 and 18.7 eV respectively. This is comparable with heavier halogens with first ionization energies in 10-15 eV range. At the same time, mercury cations are large at about 110-120 nm, which is comparable to calcium with ~110 nm and ionization energies at 6.1 and 11.2.

This means, that mercury strongly favors covalent bonding despite its relatively low formal electronegativity. This is evidenced by various alkylmercury compounds that are stable in contact with water.

Of course, there is a catch in that first and second bonds formed by mercury atom are different. When first bond is formed, it uses mercury's s-orbital that is very low in energy since it has considerable electron density near highly charged nucleus. When mercury forms two bonds, it by necessity utilizes at least one p-orbital, which is higher in energy. Same actually applies to zinc and cadmium but to much, much lesser degree. This is evidenced by formation and relative stability of $\ce{M2}$ moieties and annoying persistence of $\ce{CH3Hg+}$ cation in environment.

So, returning to your question. $\ce{Hg2Cl2}$ and $\ce{HgCl2}$ are mostly covalent compounds with molecular crystal structures. There is a degree of charge separation in those molecules, but it is fairly low. Both only partially dissociate in water. This simple picture is complicated by fairly diffuse outer shells of halogens and mercury, meaning there are strong inter-molecular dispersion interactions between individual molecules in the solids. But they are clearly not ionic.

  • $\begingroup$ Well, I commented that it's ionic and molecular. You agree that it's molecular. So the question is then if it's ionic. Ionic means composed of ions, it doesn't mean that the thing is an ion. e.g. NaCl is ionic but not an ion. (And indeed NaCl isn't molecular). You write "individual molecules in the solids. But they are clearly not ionic." <-- Each molecule consists of three ions $Cl^-$---$Hg_2^{2+}$---$Cl^-$ so it is ionic by definition(or at least is in that regard!)! $\endgroup$
    – barlop
    Oct 14, 2023 at 18:45
  • 1
    $\begingroup$ @barlop No, the molecules are covalent, that is they ARE NOT composed of ions. I edited the answer to give hard charge numbers, those were realatively easy to find. $\endgroup$
    – permeakra
    Oct 14, 2023 at 20:23
  • $\begingroup$ Thanks. What do you make of the fact that it has a reported ksp en.wikipedia.org/wiki/Mercury(I)_chloride ksp of 1.43×10^-18 . and the idea that a reported ksp would suggest that it breaks into ions when in water? $\endgroup$
    – barlop
    Mar 8 at 2:27
  • $\begingroup$ just calculate the ionic product using solubility data on the same page. $\endgroup$
    – permeakra
    Mar 10 at 21:18
  • $\begingroup$ I know that a ksp can be calculated from solubility data, and somebody could do that whether or not there are ions. The calculation does treat things as if there r ions. I wondered if maybe if a reliable source states something as having a ksp then theyd have checked that it's ionic, but I guess not? eg i've heard that atomic absorbtion spectroscopy can be used to measure ksp(and that'd really only apply where there are ions as it'd detect them). I suppose in this case it's (clearly) just calculated from solubility data not via AAS. So shouldn't be taken as an indication of there being ions? $\endgroup$
    – barlop
    Mar 12 at 3:53

Your Answer

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

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