Organic Chemistry · Part 2 of 9

Reactive Intermediates I — Carbocations & Carbanions

Reaction Mechanisms, Part 2 · 10 sections · about 20,393 words · CSIR-NET Chemical Sciences, GATE Chemistry & IIT-JAM

Almost every mechanism in this book passes through a species that cannot be put in a bottle. This part is about the two most important of them. Carbocations explain rearrangements, solvolysis rates and why Friedel–Crafts alkylation so often gives the wrong product; carbanions explain enolate chemistry and most of carbon–carbon bond formation. For each, the same four questions: what shape is it, what makes it stable, how do you make one, and what does it do next. Two layers on every section: a slow, hand-held beginner path and a research-grade advanced/reference path.

The 10 sections in Part 2

  • 1Geometry, hybridisation, and where the empty orbital points Free below
  • 2The stability ordering — and the reason behind every entry
  • 3Making a carbocation, and what it does next
  • 4Wagner–Meerwein 1,2-shifts — and predicting which group moves
  • 5Ring expansion and ring contraction
  • 6Bridged ions — the non-classical debate, and neighbouring-group participation
  • 7Geometry, hybridisation and the inversion barrier
  • 8What stabilises a carbanion — four mechanisms and one retracted explanation
  • 9Enolates and organometallics — carbanions you can actually use
  • 10Where this leads — a pointer to Part 3

Geometry, hybridisation, and where the empty orbital points

Free extract

Section B.1 of Part 2, reproduced in full from the book — figures and all. No sign-in, no paywall on this section.

Beginner layer

Three bonds, one empty orbital, one plane. Everything else in the chapter is a consequence of this picture.

Take a neutral carbon with four bonds and remove one bonding pair — not the atom, the pair, so the group leaves with both electrons. You are left with three bonds and six valence electrons. Three electron domains around a central atom means trigonal planar geometry, sp² hybridisation, and 120° bond angles. The 2p orbital that was not used in hybridisation is left over, and it is empty.

The carbocation: sp² carbon, trigonal planar, one empty p orbitalC+RRRempty 2pboth lobes emptyall R–C–R angles = 120°Valence-electron audit at C3 σ bonds × 2 e = 6 electronsempty 2p = 0 electronstotal = 6 — two short of an octetIsoelectronic with BH₃, isostructural with BF₃— the same 6-electron, trigonal planar,empty-p pattern at the central atom.The empty p is the LUMO. Nucleophiles attackinto it, from either face with equal ease.
Three σ bonds use all six valence electrons; the unhybridised 2p orbital is empty and perpendicular to the CR₃ plane. The bold wedge is the third substituent coming toward you — the three R groups and the carbon are coplanar. Because the two lobes of the empty orbital are equivalent, a nucleophile can be captured on either face, which is why a clean, free carbocation gives racemic product from an enantiopure substrate. Schematic; lobes show orientation, not computed shape.

Three consequences follow immediately, and they are worth stating separately because each one is examined on its own.

  1. The cation is flat. The three substituents and the cationic carbon lie in one plane. A carbocation centre that is prevented from becoming planar is very hard to form.
  2. A nucleophile can attack from either face. The two lobes of the empty p orbital are identical, so a stereocentre that becomes a free carbocation is destroyed: the product is racemic. In practice, ion pairing biases this, which is B.3.
  3. The empty p orbital is the LUMO, and it accepts from anything aligned with it. Neighbouring π systems, neighbouring lone pairs and neighbouring σ bonds can all donate into it — provided they are correctly oriented. Orientation, not just identity, is the recurring theme of this whole part.

The bridgehead test — proof that flatness matters

Theory says a carbocation must be planar. The cleanest experimental proof is a molecule that will not let it be. In a small bicyclic cage, the carbon at the bridgehead is held rigidly pyramidal by three bridges; it physically cannot flatten. If planarity is required, such a substrate should refuse to ionise. It does.

Three tertiary bromides. All three would give a tertiary cation. Only two of them do so at any useful rate: tert-butyl bromide is free to flatten, 1-bromoadamantane can flatten almost completely because its cage is large and the bridgehead carbon sits at the centre of a nearly ideal chair-chair framework, and 1-bromonorbornane cannot flatten at all.
Bridgehead bromideCation geometry availableRelative solvolysis rate
1-Bromoadamantaneessentially planar at the bridgehead1 (reference)
1-Bromobicyclo[2.2.2]octanemodestly pyramidalised~10−3
1-Bromobicyclo[2.2.1]heptane (norbornane)severely pyramidalised — cannot flatten~10−13 or smaller
Relative rates of bridgehead solvolysis, order-of-magnitude, after Schleyer. Ten orders of magnitude of rate difference between three tertiary bromides, produced by nothing but geometry.

1-Chloroadamantane solvolyses readily; 1-chlorobicyclo[2.2.1]heptane does not react at all under the same conditions. Both are tertiary chlorides. Explain. Easy

Both would give tertiary cations, so the usual 3° > 2° > 1° argument predicts both should be fast. It gives the wrong answer, so something else dominates.
The controlling factor is geometry at the cationic carbon. A carbocation needs to become trigonal planar. Ask, for each substrate, whether that is geometrically possible.
Adamantane is a fused three-dimensional cage in which every six-membered ring is a chair — four such rings in all, three of them independent. Its bridgehead carbon can flatten to something very close to trigonal planar with only modest strain, because the three bridges are long enough (two carbons each) to accommodate the change.
Bicyclo[2.2.1]heptane has bridges of two, two and one carbon. The one-carbon bridge locks the bridgehead angle far from 120°. Flattening would require an impossible distortion, so the ionisation transition state lies enormously high.
Answer: the reaction is not controlled by the substitution class of the cation but by whether the cationic centre can achieve planarity. It is a geometric prohibition, not an electronic one.
Advanced / reference layer

The spectroscopic and structural evidence that carbocations really look like this, and the quantitative cost of bending one.

The trigonal-planar picture is not an inference from products. Long-lived carbocations can be prepared in superacid media — typically SbF₅ in SO₂ClF or HF–SbF₅ (“magic acid”) at low temperature — where there is no base and no nucleophile strong enough to quench them, and they can then be looked at directly. This was Olah’s programme, and it is what the 1994 Nobel Prize in Chemistry recognised.

Two observables matter most:

  • ¹³C chemical shift of the cationic carbon. An sp² carbon carrying a full positive charge and an empty p orbital is enormously deshielded. The tert-butyl cation resonates near δ 335 and the isopropyl cation near δ 320 — compare a ketone carbonyl at ~200 and an ordinary alkene carbon at ~125. That shift is the fingerprint of a genuine, localised, tricoordinate cation.
  • X-ray structure. Crystalline salts of stabilised carbocations — the trityl cation, and eventually the 2-norbornyl cation itself — show the cationic carbon planar to within experimental error, with the sum of the three C–C–C angles at 360°.

The complementary number is the cost of pyramidalisation. Because the empty p is non-bonding, bending the cation out of plane converts it into an sp³-like orbital which is lower in energy on its own — but it also destroys the alignment of all the hyperconjugative donors and, more importantly, forces the three σ bonds into compressed angles. The result is a steep penalty. The bridgehead correlation quantifies it: Schleyer and co-workers showed that bridgehead solvolysis rates over a range of more than ten orders of magnitude correlate with molecular-mechanics strain-energy differences between the halide and the cation, with essentially no adjustable electronic parameter. The reactivity of these systems is their strain.

log(krel) ∝ −ΔΔEstrain(R–X → R+) / 2.303RT

The one place the planar picture is genuinely wrong

Not every carbocation is a tricoordinate carbenium ion. Two families sit outside the picture and both are examined:

  • Bridged (non-classical) ions, where a σ bond is shared between two carbons and the “cationic centre” is pentacoordinate. The 2-norbornyl cation is the famous case and gets its own section, B.6.
  • Acylium ions, R–C≡O+, where the carbon is sp-hybridised and linear, not trigonal, because the oxygen lone pair has already donated into the empty orbital to give a genuine triple bond. This is not a minor resonance contributor; it is the dominant structure, which is why acylium ions are so well behaved.
Med
Predict the geometry at the cationic carbon of CH₃CO+ and explain why Friedel–Crafts acylation does not suffer the rearrangements that plague Friedel–Crafts alkylation.
Show solution
The acylium ion is written CH₃–C+=O ↔ CH₃–C≡O+, and the second structure dominates. Oxygen donates a lone pair into the empty p orbital, giving a carbon–oxygen triple bond, an sp carbon, and a linear C–C≡O unit with the positive charge formally on oxygen. Every atom then has a complete octet. The consequence is that the acylium ion is not electron-deficient at carbon in the way an alkyl cation is: there is no empty orbital for a neighbouring C–H or C–C bond to migrate into, so it does not rearrange. That is the whole practical advantage of acylation-then-reduction over direct alkylation: you get the straight-chain product you asked for.
Med
Optically active (R)-3-chloro-3-methylhexane is solvolysed in aqueous ethanol. Predict the stereochemical outcome and state what an observed 60:40 product ratio would tell you.
Show solution
Ionisation gives a tertiary carbocation, which is planar and achiral, so nucleophilic capture from the two faces of the empty p orbital is equally likely and the ideal product is racemic. An observed 60:40 excess of inverted product tells you the reaction is not proceeding entirely through a free, symmetrically solvated cation: some product is formed at the contact ion-pair stage, where the departing chloride still shields the face it left from and the nucleophile is forced to the other face. Net inversion with partial racemisation is the normal experimental signature of SN1, and 100% racemisation is the exception rather than the rule.

Read the rest of Part 2

The remaining 9 sections of this part — The stability ordering — and the reason behind every entry, Making a carbocation, and what it does next, Wagner–Meerwein 1,2-shifts — and predicting which group moves, Ring expansion and… — and all nine parts of Reaction Mechanisms are part of ChemVidya Full Access, along with the other books, 55 Study Notes and 6,000+ practice questions.

See plans Open in the app

Continue through Reaction Mechanisms

Related Organic Study Notes