Curved Arrow Press Forum

An organic chemistry forum of Curved Arrow Press. (Oh, would you believe we accept donations.) Registration closed.

You are not logged in.

  • Index
  •  » OrgoForum
  •  » How to explain the acidity of phenol, inductive or resonance?

#1 2010-12-13 16:01:46


How to explain the acidity of phenol, inductive or resonance?

Phenol is acidic because it is resonance stabilized.

For many, this is the preferred explanation. Certainly, I have used that argument also. (If my professor expected that as the answer, that is the argument I would give.) However, as I have had to think about what makes compounds acidic, I became increasingly wary of relying on such an argument. Let me try to give some reasons why I prefer an inductive rather than a resonance argument.

Let's start with something simple. If acidity is determined only by resonance, then acetone and its enol should have the same acidity as they form the same resonance stabilized enolate. However, the enol is far more acidic than the ketone. I argue the inductive effect of the oxygen atom has a larger influence than the formation of the resonance stabilized enolate from the C-H. In this example, I far prefer using an inductive over a resonance effect to explain the acidity. (I used the pKa calculator found here to calculate the enol:

In the deprotonation of cyclohexenone, two product can form, a kinetic product and a thermodynamic product. The kinetic product is simply an inductive effect. The hydrogen alpha to the carbonyl group is more acidic and removed preferentially. Loss of the gamma hydrogens can form a more stable product, resonance stabilized. When the sigma bond electrons participate in the enolate, it gives a more stable enolate. Although this has a higher barrier to formation, it gives a more stable product. This is a resonance effect, but one that can only be revealed after deprotonation. The actual deprotonation process itself is an inductive effect. The gamma hydrogens are vinylogously linked to the carbonyl group, therefore their acidity is reduced by a lower electron withdrawing sp2-carbon and distance to the sp2-oxygen compared to the hydrogen alpha to the C=O. Together, the enone increases the acidity in an inductive manner. That is, they alter the electrons of the gamma C-H bond inductively, but the electrons of the C-H bond are not involved in any resonance until deprotonation actually occurs.

Now, let's consider an example in which I have used a resonance argument, p-nitrophenol. Because we can draw the resonance structure with the charge distributed on the oxygen atoms of a nitro group, it is inviting to use a resonance effect. It is easier to explain with resonance structures. What we don't ask is whether this is true or not. To compare, I have included p-trimethylammonium phenol as well. It has the same number of resonance structures as phenol, yet is far more acidic. I did not draw the structures for the meta substituted anions, but no similar resonance stabilization is possible. Arguably, this is entirely inductive. If we consider p-acetylphenol, a resonance argument can again be made, but no significant increase in acidity is present. A methyl group is an inductive donor. It reduces the acidity. The resonance structures show a transfer of electron density to different atoms and to which atom the charge is placed, but the effect is ultimately an inductive effect. The pKa of the different substituents are similar. It is not necessary to invoke resonance for one example and induction in another. It seems perfectly consistent to invoke an inductive effect for all. (You may need to think about that a little. A methyl group can be an electron donor but one in which the reactivity is increased on an ortho/para carbon more than ipso or meta. I am not disavowing that. I am simply arguing the donation is an inductive effect, not a resonance effect.)

My argument is simple, the acidity of substituted phenols are similar. The large effect is inductive. An ammonium salt has nearly the same effect as a nitro group. I have included a resonance structure for p-nitrophenol that I would argue is consistent with a resonance effect, but that is generally not what is argued in explaining the acidity of the parent phenol itself. (pKa values taken from here:

There are two aspects of this that I am assuming to be true. One, for a resonance effect to be true, the larger the number of electrons participating, the greater the resonance stabilization. This is the case for polyenes, e.g., carotene. This is reflected in the UV spectrum. If you have a carbocation, the greater the number of neighboring electrons that can stabilize it, the larger the effect. I am assuming a similar basis for carbanions. This would explain the thermodynamic stability of the anion of cyclohexenone. I am simply wary of including sigma electrons in that argument except as an inductive effect.

Secondly, I do NOT think that everyone understands or agrees with the notion that an sp2-carbon is electron withdrawing. I include myself in that paradox. Let me elaborate. If you have propene, the pi-electrons are nucleophilic, they can become donated in a reaction with HBr. The sigma electrons are pulled toward the sp2-carbon increasing the acidity, pKa 44. Inserting an atom can extend the electron withdrawing properties to the allylic atom. Thus, the CH3 group of propene is also more acidic, pKa 44. If an oxygen were to replace the carbon, the acidity of the enol ether is similarly increased, pKa 11.92.

In this example, carbon is a donor of pi-electrons and a withdrawer of sigma electrons.

These are my assumptions. A resonance effect explains why benzene is much more stable than predicted from the energy of a simple isolated double bond. Similarly, a conjugated diene is more stable than an isolated diene. In these instances, the pi-electrons are interacting in a manner that results in a more stable compound. In the cyclohexenone example, no resonance stabilization is present in the C-H sigma bond of the neutral compound. The kinetic deprotonation occurs on the CH2-group alpha to the C=O. However, a resonance stabilization can and does occur if a less favorable, that is less acidic proton is removed. Upon removal, the released electrons can participate in an extended resonance structure. This is similar to an energy stabilization of adding an extra set of pi-electrons in conjugation to a diene. That is, by the usual resonance rules, you can now write the three resonance structures for the thermodynamic enolate.

If you have read this far, you will note that I am skeptical of using sigma electrons in resonance structures. I argue if they did participate, then the stabilization we expect from resonance should be found in the chemical properties of compounds. If that were true, then one might expect that a compound like carotene should have a much lower pKa due to the extended conjugation. However, if the acidity is really a function of the electron withdrawing properties of an sp2-carbon and increased by conjugation to other sp2-carbons, then as the number of double bonds is extended, the inductive withdrawal should decrease if inductive. It should increase if resonance. The best I could do to estimate this effect is again with MarvinDraw. It does not give pKa values for enones, but it will for enols. Thus, it will calculate the pKa of conjugated enols up to three double bonds. The effect you will find is that as the conjugation is extended, the magnitude of the increase in acidity falls. (Even though you can use it, MarvinDraw is limited in its utility and accuracy. It is not difficult to find examples in which it clearly is wrong, e.g., compare acetaldehyde with acetyl chloride. Marvin draw think acetaldehyde is more acidic.)

I argue that acidity is generally an inductive effect. The increase in acidity from CH4<NH3<H2O<HF is inductive. The series below is also inductive with the an sp2-carbon playing a role as a neighboring atom and as an electron withdrawing atom itself. A C=O bond simply increases the effect. The acidity will vary considerably, yet the basic number of electrons in the resonance structure is the same. I would argue all of these are inductive effects. If these are inductive effects, substituted phenols show inductive effects, then phenol itself should be explained as an inductive effect (and not based upon the stability of its anion).


  • Index
  •  » OrgoForum
  •  » How to explain the acidity of phenol, inductive or resonance?

Board footer

Powered by PunBB 1.2.16
© Copyright 2002–2005 Rickard Andersson

//google analytics added