It is well known that terminal α-hydroxy ketones give Tollens’ test, via tautomerisation. Also, α-hydroxy ketones flanked by aromatic groups give Tollens’ test, such as benzoin.

But do all α-hydroxy ketones give Tollens’ test? Would for instance dimethyl glyoxal be able to give Tollens’ test? If so, what is the mechanism? The answer to this question was not satisfactory.

  • $\begingroup$ Have you gone through this question: chemistry.stackexchange.com/questions/73313/… $\endgroup$ Aug 27, 2021 at 8:06
  • $\begingroup$ And chemistry.stackexchange.com/questions/94620/… $\endgroup$ Aug 27, 2021 at 8:43
  • $\begingroup$ @NilayGhosh The first link contains an incorrect answer, and I've already referenced the second link in my question. $\endgroup$
    – Aspirant
    Aug 27, 2021 at 11:00
  • 2
    $\begingroup$ @user55119 Most questions have dealt with a specific case, like your answer. Nothing has dealt with the general case of alpha hydroxy ketones. There exist many contradictory answers; some sources say that all alpha hydroxy ketones give this test, while others say only specific cases do. This results in a great deal of confusion. $\endgroup$
    – Aspirant
    Sep 1, 2021 at 5:17
  • 1
    $\begingroup$ journals.sagepub.com/doi/pdf/10.3184/174751911X13206824040536 speculates about the mechanism and gives examples of aldehydes that give a negative Tollens' test. $\endgroup$
    – Karsten
    Sep 2, 2021 at 1:22

1 Answer 1


according to wikipedia :

Tollens' reagent (chemical formula Ag(NH3)2OH) is a chemical reagent used to distinguish between aldehydes and ketone functional groups along with some alpha-hydroxy ketones which can tautomerize into aldehydes

you could handle this problem considering the first step in tollen's test which is a redox reaction :

enter image description here

For tollen's test to give positive, your substrate must have a relatively high oxidation potential, that's why aldehydes give positive while ketones don't:

  1. aldehydes can be oxidized easier than ketones as hydrogen of -aldehydes is more electron-donating than ketones' alkyl group this causes the electrons occuping the MO of carbonyl oxygen to be higher in energy and thus more unstable and relatively easier to get lost (i,e better HOMO-LOMO interaction)
  2. alkyl shift (in ketones) is much harder than hydrogen shift (in aldehydes)
  3. alkyl groups in ketones stabilize electrons in carbonyl group making it harder to break that bond

Keeping that in mind, let's analyze alpha hydroxy ketones:

  1. hydroxyl group would decrease the dipole effect of carbonyl oxygen and stabilize carbonyl electrons (i,e, decrease its oxidation capacity)

  2. If alpha hydroxy ketone had to be deprotonated first before being oxidized, this step produces conjugated structure which can stabilize carbonyl group, this decreases the tendency to being oxidized. Evidence to this is that aromatic aldehydes don't react with Fehling's reagent (which is a weaker oxidizing agent than Tollen's) due to conjugation and increased stability of carbonyl group. If Tollen's reagent already can't reduce ketones how would it reduce stabilized ketones? enter image description here

  3. the only way alpha-hydroxy ketones can react with Tollen's reagent is as wikipedia states through tautomerization. this necessitates that hydroxyl group of the ketone be a terminal one (i,e primary hydroxyl group), there are some exceptions such as benzoin enter image description here


Benzoin is oxidized using copper(II) acetate in catalytic amount (1), so this means that it would be easier for silver ion to oxidize it. Another important aspect is that Fehling's reagent (Copper-based oxidizing agent) doesn't oxidize aromatic aldehydes besides the fact that benzoin can't tautomerize into an aldehyde. My explanation is that oxygen of carbonyl groups are cross-conjugated and not fully conjugated as the main conjugated path involves the two phenyl rings and the two carbons P-orbitals. So, I think it will be easily deprotonated and oxidized enter image description here


(1) Marvin Weiss and Mildred Appel Journal of the American Chemical Society 1948 70 (11),3666-3667 DOI: 10.1021/ja01191a036

  • 6
    $\begingroup$ It's an interesting idea, but I'm not sure whether this is really useful. Hard science > speculation, even if the speculation is logically sound. Consider using appropriate capitalisation, too, please. $\endgroup$ Sep 1, 2021 at 11:25
  • $\begingroup$ @orthocresol, I edited the answer $\endgroup$
    – mohamed
    Sep 2, 2021 at 11:44
  • $\begingroup$ I appreciate the answer, but I am looking for something substantiated by evidence. For instance, a source stating that dimethyl glyoxal cannot show Tollen's test. Or that it can, with an accompanying mechanism. $\endgroup$
    – Aspirant
    Sep 3, 2021 at 10:11

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.