Why does the reaction \eqref{rxn:R1Q} take place, but the reaction \eqref{rxn:R2Q} does not?

$$ \begin{align} \ce{HF(aq) + KF(aq) &-> KHF2(aq)}\label{rxn:R1Q}\tag{R1} \\ \ce{HCl(aq) + KCl(aq) &-> KHCl2}\label{rxn:R2Q}\tag{R2} \end{align} $$

I found that $\ce{[HF2]-}$ forms due to hydrogen bonding among $\ce{HF}$ molecules. How exactly is the hydrogen bonding responsible for the formation of bifluoride ions?


2 Answers 2


The bichloride ion $\ce{HCl2^-}$ does exist, but it requires other environment than usual solutions in water. E.g. solution of $\ce{[N(CH3)4]Cl}$ salt and undissociated $\ce{HCl(solv)}$ in nitrobenzene solvent, as mentioned in the ACS link of the Oscar's answer. It is the scenario that makes the partial hydrogen charge in $\ce{H^{(+)}-Cl{(-)}}$ quite favourable for chloride anions.

For water solutions scenario, the key is the presence of undissociated $\ce{HF}$ and the ability of $\ce{F-}$ to make hydrogen bond with $\ce{HF}$. Neither is available for $\ce{HCl}$ and $\ce{Cl-}$. There are just very minor traces of undissociated $\ce{HCl}$ in water and $\ce{Cl-}$ would not have bound to it even if it had been there.

There are no $\ce{HCl}$ or $\ce{KCl}$ in water as molecules. There are hydrated ions $\ce{H+(aq)},$ $\ce{K+(aq)}$ and $\ce{Cl-(aq)}.$ Hydrated molecules $\ce{HF(aq)}$ are there in hydrofluoric acid or in acidified fluoride solutions, reacting:

$$\ce{F-(aq) + HF(aq) <=> HF2^-(aq)}\tag{R1}$$

Hydrofluoric acid is weak mainly because of forming the stable ionic pair, so ion $\ce{H3O+(aq)}$ is mostly not free:

$$\ce{HF(aq) + H2O(l) <=> H3\overset{+}{O}\bond{...}F-(aq) <=> H3O+(aq) + F-(aq)}\tag{R2}$$

The structure of $\ce{HF2-}$ is $\ce{[F\bond{...}H-F]-}.$ There is also $\ce{[H-F\bond{...}H]+}$ in concentrated hydrofluoric acid or liquid hydrogen fluoride:

$$\ce{3 HF <=> H2F+ + HF2^-}\tag{R3}$$

Liquid hydrogen fluoride has such a high boiling point due $\ce{HF}$ chain linked by hydrogen bonds:


  • $\begingroup$ I'm guessing that, more generally, true $\ce{MHX2}$ compounds with well-defined $\ce{HX_{2}^{-}}$ anions can exist for a number of halogens and metals, but only in non-aqueous conditions at low temperatures and/or high pressures. It's interesting to consider where the decomposition boundaries lie. $\endgroup$ Commented Mar 13, 2023 at 10:09
  • $\begingroup$ @NicolauSakerNeto Yes, I suppose that is true. I am aware that many of statements done in context of chemistry have limited scope of validity. At extreme enough environment, they are often may not valid any more. Typically for low T solid matrices or interstellar space. $\endgroup$
    – Poutnik
    Commented Mar 13, 2023 at 10:39
  • $\begingroup$ @andselisk when + is used with overset (I agree with it). should not be - at F in overset too? Like visually more consistent. $\endgroup$
    – Poutnik
    Commented Mar 13, 2023 at 10:42
  • 1
    $\begingroup$ @Poutnik Matter of taste, although the top right of F would be the preferred position, an overset is an acceptable one (doi.org/10.1351/pac200880020277, pp. 372–374). Besides, you have superscripts in the same equation right next to the drawn formula which sort of undermines the idea of consistent macro for the charge placement here. $\endgroup$
    – andselisk
    Commented Mar 13, 2023 at 11:40
  • $\begingroup$ The authors from the 1954 paper were able to isolate the solid salts as well as the nonaqueous solution species, but again this would be without water or any other good B-L base. $\endgroup$ Commented Mar 14, 2023 at 18:58

The bichloride ion, including isolated salts, has in fact been known since 1954 [1]; but it appears to require relatively bulky counterions whereas bifluoride has a broader range of stable salts.

To form a complex ion of the type $\ce{HX2^-}$ the extra bond strength derived from a three-center four-electron bond (in the complex ion) must exceed the more favorable electrostatic attraction of the counterion to a more compact monatomic halide ion or solvent dipole (which favors dissociation into $\ce{X^- + HX}$), a situation similar to that for homoatomic trihalide ions $\ce{X3^-}$. But unlike homoatomic trihalide ions, $\ce{HX2^-}$ has the central atom fixed at hydrogen which favors strong covalent bonding with lighter, more compact atoms. Thereby $\ce{HF_2^-}$ is more favored than its heavier halide ion counterparts; whereas (at least in water solution and isolated compounds) triiodide ion is favored over lighter homoatomic halogen counterparts.


  1. Harry F. Herbrandson, Richard T. Dickerson Jr., and Julius Weinstein (11954). "The Bichloride Ion". J. Am. Chem. Soc. 76(15), 4046. https://doi.org/10.1021/ja01644a066

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.