The question is to find the product in the following reaction-

enter image description here

In the answer given, they have replaced Cl with I and inverted the configuration at that carbon(Finkelstein reaction).

But $\text{I}^-$ being a good base, wouldn't it first take away O's acidic H and then $\text{O}^-$ will show Neighboring Group Participation attacking the carbon with chlorine and finally resulting in retention of configuration?

  • 3
    $\begingroup$ I- is not a good enough base to deprotonate OH $\endgroup$
    – Waylander
    Jul 2, 2020 at 8:39
  • $\begingroup$ How do you say that? $\endgroup$ Jul 2, 2020 at 8:42
  • 2
    $\begingroup$ I- is really not a good base. The pKa of the OH proton is around 16. Iodide in EtOH does not give EtO- and HI $\endgroup$
    – Waylander
    Jul 2, 2020 at 8:59
  • $\begingroup$ Ok, now it makes sense $\endgroup$ Jul 2, 2020 at 9:13
  • 1
    $\begingroup$ Iodide is the very opposite of a good base. HI is the strongest hydrohalogenic acid. $\endgroup$
    – Jan
    Jul 3, 2020 at 6:18

1 Answer 1


You are thinking too far ahead for a simple reaction. First of all, $\ce{I-}$ is not a very strong base to abstract a protone from alcohol. $\ce{I-}$ is the conjugate base of very strong acid, $\ce{HI}$. Yet, $\ce{I-}$ is a very good nucleophile. Therefore, the condition given ($\ce{NaI}$ in anhydrous acetone; $\ce{NaI}$ dissolves in anhydrous acetone and therefore exists as $\ce{I-}$ in solution) is suitable only for nucleophilic substitution reaction and also favored $\mathrm{S_N2}$ mechanism. Since the byproduct form ($\ce{NaCl}$) is not soluble in acetone, the reaction favors only the forward reaction:

$$\ce{R-Cl + Na+ + I- -> R-I + NaCl_{(s)}}$$

Since it is a $\mathrm{S_N2}$ reaction, inversion is the result ($(1R,2S)$ to $(1R,2R)$ with closed to $100\% e.e.$).

However, note that there is a neighboring group participation (NGP) possible in 1,2-halohydrins in certain reactions (Ref.1). As OP suggested, if NGP is contributed here (without strong base I doubt it though), the product from NGP is going to be the retention product with $(1R,2S)$ configuration. Thus, total $\%e.e.$ of final product is $\lt 100\% e.e.$.


  1. T. A. Geissman, Richard I. Akawie, "Rearrangements of Halomagnesium Derivatives of Halohydrins," J. Am. Chem. Soc. 1951, 73(5), 1993–1998 (https://doi.org/10.1021/ja01149a029).
  • 2
    $\begingroup$ While OP's mistaking iodide for a strong base is mind-boggling, he has a point - why NGP wouldn't change the picture here? $\endgroup$
    – Mithoron
    Jul 2, 2020 at 17:11
  • $\begingroup$ @Mithoron: I agree with you. There is a possibility but it is very faint. But worth mentioning it. :-) Accordingly I change my answer $\endgroup$ Jul 2, 2020 at 19:47
  • $\begingroup$ @Mithoron What do you want it to do? Form an epoxide? $\endgroup$
    – Zhe
    Jul 3, 2020 at 2:31
  • $\begingroup$ @Zhe -OH can well enough participate in SN1, but it's effect is much more transient then that, more like first step on the road to epoxide and then step backwards. $\endgroup$
    – Mithoron
    Jul 3, 2020 at 14:52

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.