# How can benzaldehyde have a pKa of 14.9?

There are numerous websites out there which claim that the pKa of benzaldehyde – C6H5CHO – is 14.90. (Just do a Google search for benzaldehyde pka to see what I mean.) This doesn't make sense chemically, as there are no protons in benzaldehyde which would be so acidic.

However, it doesn't seem to have been made up. The CRC Handbook of Chemistry and Physics lists the same value of 14.90 for the pKa of benzaldehyde, at 25 °C in aqueous solution. But I can't find this information in the primary literature.

If this value refers to the acidity of the hydrate, does it include the equilibrium constant for hydrate formation? i.e. is this Ka the equilibrium constant for the reaction

$$\ce{PhCHO + 2H2O <=> H3O+ + PhCH(OH)O-}$$

or is it just

$$\ce{PhCH(OH)2 + H2O <=> H3O+ + PhCH(OH)O-}?$$

I would appreciate some kind of definitive evidence. Using chemical reasoning and intuition is great, but if this value is to be of any use to anybody, then we need to know exactly what it describes.

• Is it possible this is a value for the hydrate? – Dennis Cao Jan 7 at 19:57
• @DennisCao, indeed, that's my primary suspicion: it might be analogous to the "pKa" of CO2. – orthocresol Jan 7 at 19:59
• I think this must be for the deprotonation of the hydrate. – Waylander Jan 7 at 22:05
• Not that it's the definitive source, but the book "Dictionary of Food Compounds" (see Google Books preview) suggests it is indeed the hydrate. – SendersReagent Jan 7 at 22:07

Significant amount of geminal diol of benzaldehyde exists in an aqueous solution of benzaldehyde at 25 °C because $$\mathrm{p}K_{\text{hyd}} = 2$$ (Ref. 1)

The $$\mathrm{p}K_{\mathrm a}$$ of benzyl alcohol is listed as 15.40 (Wikipedia). Thus, one can reasonably assume that the given value of $$\mathrm{p}K_{\mathrm a}$$ 14.9 represents a composite equilibrium constant for the hydration of benzaldehyde and dissociation of the geminal diol thus formed.

In his paper, "Acidity constants of benzimidazolium ketone and pyridinium aldehyde hydrates" (Ref.2), Terence C. Owen states that:

It is known that the acidity constants of gem-diols typically are about 2.5 units lower than those of the corresponding monohydric alcohols.

When do some extensive literature survey, one can find few examples to backup that statement. For example, $$\mathrm{p}K_{\mathrm a}$$ of methanol is reported as 15.7 while that of formaldehyde hydrate is 13.3, between which $$\Delta\mathrm{p}K_{\mathrm a} = 2.5$$ (Ref.2). Interestingly, $$\mathrm{p}K_{\mathrm a}$$ of 1,1,1,3,3,3-hexafluoropropan-2-ol is reported as 9.22 while that of hexafluoroacetone hydrate is 6.45 where $$\Delta\mathrm{p}K_{\mathrm a} = 2.77$$ (Ref.3). However, $$\mathrm{p}K_{\mathrm a}$$ of 2,2,2-trifluoroethanol is reported as 12.37 (Ref.4) while that of 2,2,2-trifluoroethanal hydrate is 10.05 (Ref.3) where the difference is < 2.5 ($$\Delta\mathrm{p}K_{\mathrm a} = 2.33$$).

Thus, we can conclude that the $$\mathrm{p}K_{\mathrm a}$$ of benzaldehyde is derived from its hydrate (gem-diol).

Also see: Yoshiro Ogata and Atsushi Kawasaki, In The Chemistry of Carbonyl Group, Volume 2; Jacob Zabicky, Ed.; John Wiley & Sons Ltd.: New York, NY, 1970, Chapter 1: Equilibrium additions to carbonyl compounds, pp 1–69 (https://doi.org/10.1002/9780470771228.ch1).

• A bit of further research I did backs this up (several sources also mention the hydration equilibrium constant, e.g. nrcresearchpress.com/doi/pdf/10.1139/v79-084), and I think it is plausible that the gem-diol is more acidic than benzyl alcohol (inductive withdrawal). However, I'm still holding out for a more detailed confirmation of what exactly this 14.9 entails, specifically whether it includes the hydration equilibrium or not. – orthocresol Jan 7 at 23:09
• This is a good reference. Accordingly, $pK_a$ of hydrated Phthalaldehyde is suggested as 11.6. – Mathew Mahindaratne Jan 7 at 23:33
• Note that if you combine the 2 orders of magnitude from the hydration equilibrium and the 2.5 orders of magnitude from the gem-diol expected acidity constant, you would expect that the pKa is about 0.5 units smaller, which is actually spot on. – Zhe Jan 14 at 15:30

I'm a little reluctant to throw this out there, because I am not an expert in organic chemistry (and it's been ... a while since I was an undergrad researcher in an orgo lab), but I would guess that the extra electrons on the oxygen can help stabilize the phenyl group anion that forms when an H+ ion is removed. (I am assuming that the H+ to leave comes from the phenyl, not from the aldehyde).

Looking here, suggests that a typical aldehyde has a pka ~ 17, but this example uses just an ankane base, not an aromatic base. (Interesting to note that they also do not talk of the hydrogen right next to the oxygen being lost as H+.) So, the resonance along the chain from oxygen's non-bonded pairs, down the double bond, single bond, and onto the ring should help stabilize the negative charge that comes from losing the H+. Thus, benzaldehyde's pKa of ~15, compared to benzene's of ~43, makes sense as the extra bonds and electrons stabilize the anion.

Well, that's my hypothesis, anyway.

• Sorry, but this isn't plausible to me. As you note, the pKa of 17 is for the proton alpha to the C=O, not the one bonded to it. There is no resonance stabilisation of the conjugate base in benzaldehyde; the lone pair would be in a sigma orbital. Lastly, even an electron-withdrawing group can't reduce the pKa by almost 30 units. (For an example where there actually is resonance stabilisation, you can compare phenol, pKa ~10, with picric acid, pKa ~0.) – orthocresol Jan 7 at 20:01