Why and how are ketones reduced?

[olee]

found this on this page
http://www.chemguide.co.uk/mechanisms/nucadd/reduce.html#top

can someone explain why it is simplified? and how does it really happen?

 

[DDTea]

This is the gist of what happens. However, there are a few ways in which it is simplified.

For one, all of those individual reactions are equilibrium processes and some are very reversible, especially the formation of the tetrahedral intermediate that results from the nucleophilic attack of hydride on the carbonyl group.

Also, the hydride ion is not floating around in solution. Hydride comes from some hydride reducing agent such as lithium aluminum hydride (LAH), sodium borohydride, sodium cyanoborohydride, etc. In the case of lithium aluminum hydride, the lithium or aluminum ion will stabilize the charge density on oxygen atom in the tetrahedral intermediate. Also, in the case of the LAH reduction, the resulting lithium alkoxide salt needs to be hydrolyzed to liberate the alcohol.

Furthermore, there is no mention of what solvents are used in these reactions. For example, the LAH reduction is typically done in ether or THF, while sodium borohydride and sodium cyanoborohydride reductions are done in aqueous solutions.

Finally, a note about water chemistry: in acidic solution, there is a high concentration of hydronium ion. There are never free protons floating around in solution, although in most introductory courses, it's perfectly acceptable to simply write "H+(aq)" as opposed to "H3O+(aq)."

Really, it could be a lot more complicated than this depending on how deep and quantitative you'd like to go! Organic Chemistry might seem like a very "fuzzy" or "squishy" branch of chemistry, but it actually depends on some cold, hard, physical laws.

 

[alxm]

Sounds good. Also, is there really a tetrahedral anion intermediate? I'm not sure that's actually the case in all circumstances.
You could have some kind of [BH₄-CR₁R₂O]^- intermediate. Although this probably depends a lot on the solvent.

(edit: Checked it out, seems that for sodium borohydride in a nonpolar solvent you wouldn't have an intermediate; the hydrogen is transferred to the carbon in concert with the oxygen binding to the sodium.
But I should point out that the 'simplified' picture, and mechanisms given in org chem textbooks are still useful even when not 100% correct, because knowing the stability of these hypothetical intermediates still tells you something about the reaction barrier of the real thing. Analogously, resonance structures don't really exist in-themselves, but are a useful tool in analyzing the stability.)

 

[DDTea]

But I should point out that the 'simplified' picture, and mechanisms given in org chem textbooks are still useful even when not 100% correct, because knowing the stability of these hypothetical intermediates still tells you something about the reaction barrier of the real thing. Analogously, resonance structures don't really exist in-themselves, but are a useful tool in analyzing the stability.)

One of the tricks in every branch of science is knowing which approximations are acceptable for a particular situation. So, while any description relying on valence bond theory (VBT) is not going to be the entire "truth" of the matter, it is certainly *a lot* easier to think in terms of VBT as opposed to Molecular Orbital (MO) Theory. Seriously, have you done MO calculations on even something small like CN-? It can be done, but it's not fun to do by hand!

If you ever study physical organic chemisry, you'll learn a lot of cool things along these lines--basically, everything you're taught in a sophomore O. Chem class is pretty much a limiting example or not *quite* accurate, although it suffices for rationalizing why reactions occur. It's disconcerting to "unlearn" what you've been taught, but also exciting to get that deeper understanding.

 

[alxm]

So, while any description relying on valence bond theory (VBT) is not going to be the entire "truth" of the matter, it is certainly *a lot* easier to think in terms of VBT as opposed to Molecular Orbital (MO) Theory.

Actually I think in MO theory terms almost exclusively. VB theory isn't so simple IMO, once you get into three-electron bonds, etc.

Seriously, have you done MO calculations on even something small like CN-?

Seriously? Well, after doing quantum chemistry full-time for the past 7 years, I've done a few. The largest would be on systems of >200 atoms, if you count Kohn-Sham MOs. Otherwise, it'd depends on the level of theory.

It can be done, but it's not fun to do by hand!

By hand? I'm not sure what kind of calculation you're be talking about then. Do you mean using Hückel theory? (In which case, sure, I've done it)

If you ever study physical organic chemisry

What made you think I haven't? (My first postgrad degree was in phys chem) Did you intend to address the original poster?
(And welcome to PF by the way)

 

[DDTea]

Haha, yes, the comment about Physical O. Chem was directed at the original poster. Actually, almost all of the post was. I was just quoting from your post, alxm, to keep some continuity in the discussion. :P

Cheers on the welcome!