Cross-Couplings & C–C / C–X Bond Formation
The reactions that join two carbon fragments under a metal. A low-valent palladium(0) (or copper) centre acts as a molecular matchmaker: it inserts into a carbon–halide bond (oxidative addition), picks up the second carbon from a main-group organometallic (transmetalation), and expels the new C–C bond (reductive elimination), regenerating Pd(0) to do it again. Change only the organometallic partner and you change the name of the reaction: boron → Suzuki, zinc → Negishi, tin → Stille, magnesium → Kumada, silicon → Hiyama, Cu-acetylide → Sonogashira. Swap transmetalation for alkene insertion + syn-β-hydride elimination and the same metal runs the Heck, Wacker–Tsuji and Tsuji–Trost reactions; swap the carbon nucleophile for an amine and you get Buchwald–Hartwig C–N coupling. Two layers on every reaction: a hand-held beginner path (plain “what it does”, RDKit scheme, catalytic-cycle mechanism, graded worked examples, trap boxes) and a research-grade advanced/reference path (detailed cycle with the specific transmetalation step, ligand/base effects, chemoselectivity, scope & FG-tolerance tables, asymmetric variants, selectivity comparisons, research-level problems). Pd(0)/Pd(II) couplings: Suzuki–Miyaura, Negishi, Stille, Kumada–Tamao–Corriu, Hiyama, Sonogashira, Heck, Buchwald–Hartwig, Wacker–Tsuji, Tsuji–Trost. Copper: Chan–Lam, Ullmann, Glaser/Eglinton/Hay. Classical alkali/Na: Wurtz, Wurtz–Fittig. Plus the Fukuyama thioester coupling. — closing with a “which cross-coupling do I use?” decision guide by organometallic partner and a catalytic-cycle master figure.
The 19 sections in Part 8
- 1The four elementary organometallic steps Free below
- 2The Suzuki–Miyaura coupling (organoboron)
- 3The Negishi coupling (organozinc)
- 4The Stille coupling (organotin)
- 5The Kumada–Tamao–Corriu coupling (Grignard)
- 6The Hiyama coupling (organosilicon)
- 7The Sonogashira coupling (terminal alkyne + halide)
- 8The Mizoroki–Heck reaction (alkene arylation)
- 9The Buchwald–Hartwig amination (C–N coupling)
- 10The Wacker–Tsuji oxidation (alkene → methyl ketone)
- 11The Tsuji–Trost allylic alkylation (π-allyl Pd)
- 12The Chan–Lam coupling (Cu; boronic acid + amine/phenol)
- 13The Wurtz & Wurtz–Fittig couplings (Na; the classical alkane/arene couplings)
- 14The Ullmann coupling & condensation (Cu; biaryls and C–heteroatom bonds)
- 15The Glaser / Eglinton / Hay alkyne couplings (Cu; oxidative diyne formation)
- 16The Fukuyama coupling (thioester + organozinc → ketone)
- 17Decision table by organometallic partner (the transmetalation couplings)
- 18Decision table by target bond / product
- 19Selectivity & functional-group comparison
The four elementary organometallic steps
Free extractSection Y.0 of Part 8, reproduced in full from the book — figures and all. No sign-in, no paywall on this section.
Four moves do everything: oxidative addition, transmetalation (or migratory insertion), β-hydride elimination, and reductive elimination. Learn these once and every named coupling in this Part becomes a variation on a theme.
In one line: a cross-coupling is built from a small vocabulary of elementary steps at the metal. Master these and you can reconstruct the mechanism of any coupling from its name.
Step 1 — Oxidative addition (OA)
Pd(0) (electron-rich, coordinatively unsaturated, typically 14-electron L₂Pd(0)) inserts into the C–X bond of the electrophile. Two new bonds form (Pd–C and Pd–X), the electrophile's C–X bond breaks, and palladium is oxidised from 0 to +2. This is usually the rate-determining step for unreactive electrophiles (aryl chlorides).
Step 2 — Transmetalation
The second carbon group R′ is transferred from a main-group organometallic (R′–[M]) onto palladium. For the classical partners the halide leaves on that metal as X–[M] (Bu₃SnX, ZnX₂, MgX₂, R₃SiX). Boron is the exception: under aqueous base the halide is taken off palladium by the base and the boron departs as boric acid, B(OH)₃ — not as an X–[B] species (see the mechanism box below). This is the step that defines which named coupling you are running: boron → Suzuki, zinc → Negishi, tin → Stille, magnesium → Kumada, silicon → Hiyama, copper acetylide → Sonogashira. It very often needs an activator: a base (Suzuki), fluoride (Hiyama), or simply a more electropositive metal (Kumada).
Step 2′ — Migratory insertion & β-hydride elimination (the Heck manifold)
When there is no organometallic to transmetalate, an alkene coordinates and inserts into the Pd–C bond. This migratory insertion is strictly syn (the Pd and the carbon add to the same face). The resulting σ-alkyl–Pd then does a syn-β-hydride elimination, releasing the new alkene and leaving H–Pd–X, which a base converts back to Pd(0).
Step 3 — Reductive elimination (RE)
The two organic groups now on Pd(II) — which must be cis — couple to form the new C–C bond, and palladium falls from +2 back to 0, ready for another cycle. Reductive elimination is fastest for two sp² carbons and is accelerated by bulky ligands that push the two groups together.
Mechanism — A generic Pd cross-coupling (Suzuki as the template)
- Pre-catalysis. A Pd(II) pre-catalyst (e.g. Pd(OAc)₂) is reduced in situ to the active Pd(0)Lₙ; or a Pd(0) source (Pd(PPh₃)₄, Pd₂(dba)₃) is used directly.
- Oxidative addition of Ar–X gives Ar–Pd(II)–X.
- Transmetalation: the base acts at palladium — hydroxide replaces the halide (which leaves as a salt with the base) to give Ar–Pd(II)–OH, and this Pd–OH transmetalates with the neutral boronic acid through a Pd–O–B bridge. Ar′ migrates to Pd, giving Ar–Pd(II)–Ar′, and the boron is expelled as boric acid, B(OH)₃ (present as the borate B(OH)₄⁻ under the aqueous base) — not as an X–[B] species. The alternative boronate pathway (base first forms Ar′B(OH)₃⁻, which attacks Ar–Pd–X) is the minor, much slower route — Ar–Pd–OH plus the neutral boronic acid outruns it by orders of magnitude. Note that too much hydroxide is also bad, because it locks the boronic acid up as the unreactive boronate; see Y.1.
- Reductive elimination couples Ar–Ar′ and regenerates Pd(0)Lₙ.
⚠ Common mistakes & exam traps
- Oxidative addition needs Pd(0), reductive elimination gives Pd(0). Do not draw Pd(0) coupling the two carbons directly — the carbons couple from Pd(II).
- Reductive elimination is from cis. A trans-diorganopalladium must isomerise to cis first; bidentate ligands that enforce cis geometry (dppf) help.
- β-Hydride elimination is the enemy of alkyl couplings. Any σ-alkyl–Pd with a β-H tends to eliminate rather than couple — this is why C(sp³) partners were historically hard and why bulky, electron-rich ligands are needed. It needs three things: (i) a β-hydrogen, (ii) an M–C–C–H unit that can become syn-coplanar, and (iii) a vacant coordination site cis to the alkyl — blocking (iii) with a chelating ligand or a coordinatively saturated metal is one of the main ways it is suppressed.
- Electrophile reactivity: I >> OTf ≳ Br >> Cl. The huge gap is between the iodide and everything else: with Pd⁰(PPh₃)₄ in DMF at 20 °C the measured rate constants are 17 (PhI), 1.7 × 10⁻³ (PhOTf) and 1.0 × 10⁻³ (PhBr) — i.e. PhI adds about 10⁴× faster than PhOTf, while PhOTf is only slightly (≈2×) faster than PhBr, not orders of magnitude. Do not quote a large OTf/Br rate ratio; the practical difference between triflate and bromide lies mainly in the cationic Pd intermediate and in the later steps, not in the oxidative addition. Aryl chlorides are cheap but need special ligands (SPhos, XPhos, P(t-Bu)₃, NHCs); fluorides do not react.
- The Pd(0)/Pd(II) cross-coupling manifold runs under an inert atmosphere. Pd(0) and most organometallics are oxygen-sensitive; adventitious O₂ oxidises Pd(0) and causes homocoupling. The deliberate exceptions are the oxidative couplings, which need O₂: Chan–Lam and Glaser/Eglinton/Hay (Cu), and the Wacker–Tsuji oxidation, which is a palladium reaction on the Pd(II)/Pd(0) couple run under Cu/O₂.
Oxidation-state and electron bookkeeping, the rate order and mechanism of oxidative addition, the ligand toolbox, why C(sp³) is hard, and how each named coupling differs only in the transmetalation step.
Oxidative addition: rate order and mechanism
For aryl electrophiles, oxidative-addition rate tracks the C–X bond strength and the LUMO energy: Ar–I >> Ar–OTf ≳ Ar–Br >> Ar–Cl >> Ar–F. Put numbers on it: for oxidative addition to Pd⁰(PPh₃)₄ in DMF at 20 °C the rate constants are 17 (PhI), 1.7 × 10⁻³ (PhOTf) and 1.0 × 10⁻³ (PhBr) (Jutand & Mosleh, Organometallics 1995, 14, 1810; tabulated in Jutand, Eur. J. Inorg. Chem. 2003, 2017). So the iodide is roughly four orders of magnitude faster than the triflate, whereas the triflate is only about twice as fast as the bromide — the OTf/Br gap is small, and the exact ordering can even invert with ligand and solvent. What really distinguishes the triflate is that it adds through a cationic (σ-aryl)Pd(II) intermediate, because triflate is a non-coordinating counterion; the resulting chemoselectivity is coupling-dependent (triflate is displaced preferentially in Kumada, Negishi and Stille, but bromide usually wins in Suzuki–Miyaura — Espino, Kurbangalieva & Brown, Chem. Commun. 2007, 1742). Electron-poor aryls add faster (they stabilise the electron-rich, formally Ar−-like TS). Mechanistically, aryl/vinyl halides usually add by a concerted three-centred pathway giving cis-Ar–Pd–X (which rapidly isomerises to trans); activated or benzylic/allylic systems can go by SN2 or radical pathways. Bulky, electron-rich ligands (P(t-Bu)₃, SPhos, XPhos, NHCs) give monoligated L–Pd(0), the most reactive species for the tough aryl chlorides.
| Electrophile | Relative OA rate | Typical catalyst/ligand need | Notes |
|---|---|---|---|
| Ar–I | fastest (≈10⁴× Ar–OTf with Pd(PPh₃)₄/DMF) | Pd(PPh₃)₄ suffices | expensive; can over-react |
| Ar–OTf | comparable to Ar–Br (≈2×, ligand-dependent) | PPh₃, dppf | made from phenols; cationic Pd intermediate |
| Ar–Br | comparable to Ar–OTf | PPh₃, dppf | the workhorse electrophile |
| Ar–Cl | slow | SPhos/XPhos, P(t-Bu)₃, NHC | cheap; needs modern ligands |
| Ar–F / Ar–OMe | inert (normally) | — | not used as electrophiles (classically) |
| R(sp³)–X with β-H | variable | bulky L; avoid β-H elim. | historically hard |
The ligand toolbox
| Ligand class | Examples | What it does | Best for |
|---|---|---|---|
| Triarylphosphine | PPh₃ | moderate donor, cheap | Ar–I/Br, Stille, classic Suzuki/Heck |
| Bidentate phosphine | dppf, dppe, BINAP | enforces cis; chelation | RE-limited couplings; asymmetric (BINAP) |
| Bulky electron-rich | SPhos, XPhos, P(t-Bu)₃ | monoligated Pd, fast OA | Ar–Cl, hindered substrates |
| N-Heterocyclic carbene | IPr, SIPr | very strong σ-donor, robust | Ar–Cl, high-temperature work |
| Buchwald dialkylbiaryl | SPhos, RuPhos, BrettPhos | tuned OA + RE | Buchwald–Hartwig, aryl chlorides |
Why the name changes but the cycle does not
Every transmetalation coupling shares steps 1 and 3; they differ only in step 2 — the identity of R′–[M] and the activator it needs. The table below is the organising map for the whole Part.
| Named coupling | Nucleophile R′–[M] | Activator | Signature strength |
|---|---|---|---|
| Suzuki–Miyaura | organoboron (R–B(OH)₂, BPin) | base (OH⁻, CO₃²⁻) | cheap, non-toxic, water-tolerant, huge scope |
| Negishi | organozinc (R–ZnX) | (none; Zn is reactive) | fast, chemoselective, C(sp³) capable |
| Stille | organotin (R–SnBu₃) | (none; sometimes Cu/F⁻) | very FG-tolerant; tin is toxic |
| Kumada–Tamao–Corriu | Grignard (R–MgX) | (none) | cheapest; low FG tolerance |
| Hiyama | organosilane (R–SiR₃) | fluoride or base | cheap, non-toxic; needs activation |
| Sonogashira | terminal alkyne (via Cu-acetylide) | amine base (+ Cu) | the way to make aryl alkynes |
Show solution
Show solution
Show solution
Further reading.
- The 2010 Nobel Prize in Chemistry was awarded to Richard F. Heck, Ei-ichi Negishi and Akira Suzuki ‘for palladium-catalyzed cross couplings in organic synthesis.’ Standard, widely reported fact.
Read the rest of Part 8
The remaining 18 sections of this part — The Suzuki–Miyaura coupling (organoboron), The Negishi coupling (organozinc), The Stille coupling (organotin), The Kumada–Tamao–Corriu coupling (Grignard) — and all nine parts of Named Reactions are part of ChemVidya Full Access, along with the other books, 55 Study Notes and 6,000+ practice questions.
See plans Read it in the app