Addition Reactions of Alkynes

Lindlar hydrogenation remains an industrial workhorse in retinoid (vitamin A family) and carotenoid manufacturing, exactly the application for which it was invented, because these molecules are built from long polyene chains whose biological and optical properties depend critically on maintaining specific cis or trans double bonds installed at precise points in a multistep synthesis. In total synthesis of natural products — macrolides, pheromones, polyketides — chemists routinely build a molecule's carbon skeleton using alkyne couplings (Sonogashira, alkylation of terminal alkynide anions related to Alkyne Structure and Terminal Alkyne Acidity) and then install the required alkene geometry at the very last step using Lindlar or Na/NH₃ reduction, because doing so avoids ever having to carry a geometrically fragile cis or trans alkene through many earlier synthetic steps where it might isomerize or react unwantedly. Pheromone synthesis is a particularly clean showcase: insect sex pheromones are frequently simple, straight-chain alkenes whose biological activity depends absolutely on E/Z geometry (a single wrong double-bond geometry can render a synthetic pheromone completely inactive, or even repellent, to the target insect), and the standard laboratory and industrial route is alkylation of a lithium acetylide followed by a stereochemistry-selecting reduction — Lindlar for the cis pheromone, Na/NH₃ or Li/NH₃ for the trans pheromone. Electrophilic halogenation of alkynes, while less clean for mono-addition (as the kinetic analysis above shows), is still used deliberately in cases where the tetrahalide or a specific vinyl halide is the actual target, and anti addition stereochemistry through the bridged halonium ion is a standard mechanistic teaching case connecting directly to The Electrophilic Addition Paradigm established for simple alkenes. The vinyl cation chemistry of HX addition, meanwhile, sets up the mechanistically related but practically far more important hydration of alkynes, in which Markovnikov addition of water (catalyzed by Hg²⁺ or, in modern green chemistry, by gold or other soft Lewis acid catalysts) generates an enol that tautomerizes instantly to a ketone — a transformation that depends on exactly the same vinyl-cation-like (or metal-stabilized cation-like) regiochemical logic developed here, and which is the natural next topic once mono- and di-addition control is understood.

Ask the tutor: Why does the vinyl radical intermediate in Na/NH₃ reduction isomerize so much faster than it gets reduced by the second electron, and what would happen stereochemically if that isomerization were somehow slowed down?

Start from orbital structure. Each acetylenic carbon is sp-hybridized: two sp orbitals point at 180° (one to the substituent, one to the other alkyne carbon via a σ bond), and two mutually perpendicular p orbitals form two π bonds, one in the plane of the page and one perpendicular to it. This cylindrical electron density is what makes the alkyne carbons accessible to electrophiles from any lateral direction — but it also means each π bond is a weaker "unit" than an isolated alkene π bond in terms of stabilizing a developing positive charge on an adjacent carbon, because sp carbons hold their electrons closer to the nucleus and are more electronegative than sp² or sp³ carbons. This is the single fact that governs alkyne electrophilic reactivity: alkynes are less nucleophilic per π bond than alkenes, so simple electrophilic additions to alkynes are often slower than the analogous addition to the alkene product — a fact you must know to explain why mono-addition is sometimes hard to stop and other times easy. The second governing fact is about intermediates. Protonation of an alkyne (Markovnikov HX addition) generates a vinyl cation, R–C⁺=CR′₂ or R₂C=C⁺R, in which the empty orbital is an sp-hybridized orbital in the plane of the remaining π bond, orthogonal to it. Vinyl cations are considerably less stable than analogous sp³ carbocations — poor hyperconjugative overlap, and the σ-framework rigidity around a linear cation center — so uncatalyzed HX addition to simple alkynes is sluggish and often needs strong acid catalysis or metal-ion catalysis (Hg²⁺, discussed in hydration chemistry) to proceed at a reasonable rate. Bridged halonium-ion formation (X₂ addition) sidesteps a fully localized vinyl cation and is mechanistically closer to alkene halogenation, giving clean anti-stereochemistry. The third governing fact is that reductive additions are not carbocation chemistry at all. Heterogeneous hydrogenation over Lindlar's catalyst is a surface phenomenon: the alkyne adsorbs flat on a Pd surface poisoned with lead acetate and quinoline (which block excess active sites so only weakly-bound alkyne adsorbs and the resulting alkene desorbs before over-reduction), and both H atoms are delivered from the same face of the metal, giving syn addition and thus the cis-alkene. Dissolving-metal reduction (Na or Li in NH₃(l)) is single-electron-transfer chemistry: an electron from a solvated-electron pool adds to the alkyne π orbital to give a vinyl radical anion, which is protonated by ammonia, generating a vinyl radical that isomerizes to its lower-energy (usually trans, less strained) geometry before a second electron and second proton complete the reduction to the trans*-alkene, anti to the first.

The four addition pathways can be organized by two orthogonal questions: (1) is the mechanism ionic (electrophilic) or radical/electron-transfer (reductive), and (2) is the stereochemical outcome syn or anti? Each answer follows mechanistically from the geometry of the rate-determining intermediate. For anti-addition of halogens through a bridged halonium ion, the net stereochemical relationship in the product dihalide (from an internal alkyne, one new stereocenter pattern per addition) is fixed by backside attack:

where: the first equivalent of $X_2$ gives an anti addition across the triple bond, producing a trans (E)-dihaloalkene if it is trapped once; a second equivalent of $X_2$ repeats the process on that alkene to give the tetrahalide. For the reduction branch, the governing relationship is a bond-count/electron-count bookkeeping statement plus a stereochemical rule:

where: stoichiometrically both Lindlar hydrogenation and Na/NH₃ reduction consume one mole of H₂-equivalent (two H atoms, or two electrons plus two protons) per triple bond reduced to a double bond, but which alkene diastereomer results is entirely a mechanistic (kinetic pathway) question, not a thermodynamic one — indeed Na/NH₃ delivers the thermodynamically less stable (E, more common "trans is more stable" logic is even reinforced here) alkene not because it is more stable overall in every case, but because the mechanism specifically funnels through the lower-energy vinyl radical geometry, which for most substituents happens to be the (E)-like arrangement. The Markovnikov regiochemical rule for HX addition to a terminal alkyne is best written as a relative-rate/stability statement:

where: protonation occurs preferentially at the terminal (less substituted) carbon because that places the resulting positive charge on the internal carbon, which bears an alkyl substituent capable of (weak, but real) hyperconjugative and inductive stabilization of the vinyl cation — exactly the logic of Markovnikov addition to alkenes, transposed to a less stable cation type.

Example 1 (foundational) — HBr addition to propyne, mono-addition regiochemistry. Propyne, CH₃–C≡CH, is treated with exactly one equivalent of HBr at low temperature, no peroxides present (ionic mechanism, not radical anti-Markovnikov). Step 1. Protonation must occur at one of the two alkyne carbons. Protonating the terminal CH carbon gives cation CH₃–C⁺=CH₂ analog — more carefully, protonating C1 (the terminal CH) places the developing positive charge on C2, giving the cation CH₃–C⁺=CH₂, a secondary vinyl cation stabilized by the adjacent methyl group's hyperconjugation and weak induction. Protonating C2 instead would place positive charge on the terminal carbon, CH₂=C⁺–H analog, i.e. a primary vinyl cation, drastically less stable (no adjacent alkyl substituent at all to help, and vinyl cations are already poor at stabilizing charge). Step 2. By Curtin–Hammett-type kinetic reasoning (the more stable transition state leading to the more stable cation is lower in energy, since the two protonation transition states resemble their respective cations reasonably well — Hammond postulate for an endothermic, cation-forming step), the pathway through the secondary vinyl cation dominates overwhelmingly. Step 3. Bromide traps the secondary vinyl cation CH₃–C⁺=CH₂ (written properly as CH₃–C⁺=CH₂, i.e. the cation is at C2 with the double bond now formally between C1 and C2's remaining π system — concretely the species is $\mathrm{CH_3-C^+=CH_2}$ is a resonance-poor way to write it; the honest Lewis structure is $\mathrm{H_2C=C^+(CH_3)}$, a vinyl cation) at the empty sp orbital, giving 2-bromopropene, $\mathrm{CH_2=CBr(CH_3)}$. Step 4 — check. This is Markovnikov: Br ends up on the more substituted alkyne carbon (C2, which bore the methyl), exactly analogous to Markovnikov addition to propene giving 2-bromopropane. The regiochemical rule transposes correctly from alkenes to alkynes.

Example 2 (applied) — Synthesizing a specific alkene stereochemistry: (E)-4-octene from 4-octyne via dissolving-metal reduction, with quantitative stoichiometry. A synthetic chemist has 6.40 g of 4-octyne (CH₃CH₂CH₂–C≡C–CH₂CH₂CH₃, $M = 110.20\ \mathrm{g/mol}$) and needs (E)-4-octene in high stereochemical purity, avoiding any over-reduction to octane. Step 1 — moles of substrate.

Step 2 — reagent choice and mechanism justification. Sodium metal (freshly cut, small pieces to control reaction rate and avoid violent H₂ evolution from trace water) is dissolved in liquid ammonia at $-33^\circ\mathrm{C}$ (its boiling point, used as the reaction solvent/temperature), generating the deep-blue solvated electron. Exactly 2 equivalents of electrons per triple bond are needed (Step 4–6 of Mechanism 2), so:

requiring at least $0.11616\ \mathrm{mol}$ of Na metal (in practice a modest excess, e.g. 2.2–2.5 equiv, is used to ensure complete reaction and to outcompete any adventitious proton sources that would quench solvated electrons unproductively), i.e.:

Step 3 — why no over-reduction to octane occurs. Once (E)-4-octene forms, it is a simple, unstrained alkene with no unusually low-lying π orbital or radical-stabilizing feature; solvated electrons do add to ordinary alkene π orbitals far more slowly than to the more easily reduced alkyne π system (alkynes have a lower-lying, more accessible LUMO along the electron-transfer coordinate relevant here, and the resulting alkene radical anion is much higher in energy / harder to generate under these conditions), so the reaction is kinetically self-limiting at the alkene stage — no catalyst poisoning trick is even needed here, unlike the Lindlar case. Step 4 — stereochemical outcome and limiting-case check. The product is essentially pure (E)-4-octene (typically >98% E by GC), by the radical-isomerization argument of Mechanism 2. As a limiting-case check: if the alkyne were <strong>symmetric with identical, very small substituents</strong> (imagine, hypothetically, replacing both propyl chains with hydrogen — i.e., degenerately, acetylene itself), there would be no E/Z distinction possible at all (both carbons bear only H), and indeed Na/NH₃ reduction of acetylene simply gives ethylene with no stereochemical question to ask — the mechanism's stereochemical selectivity* is only meaningful, and only tested, when the two alkyne substituents are distinguishable, confirming that the E-selectivity mechanism is a genuine steric-equilibration effect (vanishing when there is no steric difference to select between).

Example 3 (exam/competition level) — Mono- vs di-bromination selectivity and a quantitative kinetic argument. An alkyne, 1-phenylpropyne (Ph–C≡C–CH₃), is treated with Br₂ in CCl₄. Given that the relative rate of Br₂ addition to the alkyne is $k_{\text{alkyne}} = 1.0 \times 10^{2}\ \mathrm{M^{-1}s^{-1}}$ and to the resulting alkenyl dibromide-forming alkene intermediate (the (E)- or (Z)-1,2-dibromo-1-phenylpropene, itself capable of reacting with a second Br₂) is $k_{\text{alkene}} = 4.0\times 10^{4}\ \mathrm{M^{-1}s^{-1}}$, determine what fraction of alkyne survives as clean mono-adduct if exactly 1.05 equivalents of Br₂ are added all at once to a solution initially $0.20\ \mathrm{M}$ in alkyne, and explain the practical strategy actually used to avoid this problem. Step 1 — recognize the core kinetic difficulty. Because $k_{\text{alkene}}/k_{\text{alkyne}} = 4.0\times10^{4}/1.0\times10^{2} = 400$, the vinyl dibromide product is 400 times more reactive toward Br₂ than the starting alkyne is. This seems paradoxical (alkynes are supposed to be less reactive per π bond than alkenes, consistent with a bridged bromonium-like intermediate being harder to form when the alkyne's π electrons are more tightly held/lower in energy than an alkene's) — and indeed this ordering ($k_{\text{alkene}} > k_{\text{alkyne}}$) is exactly what is generally observed experimentally for uncatalyzed halogenation, which is why clean mono-halogenation of alkynes with X₂ is notoriously difficult by simple stoichiometric control alone. Step 2 — set up the competition. With Br₂ added all at once rather than added dropwise/slowly, both the alkyne and any vinyl dibromide already formed compete for the same, transiently high, local Br₂ concentration. Because the rate constant ratio strongly favors consumption of whatever alkene has already formed, once even a small amount of vinyl dibromide is generated, it is consumed by a second Br₂ equivalent far faster than fresh alkyne is consumed by its first equivalent — this drives the reaction toward a statistical mixture heavily weighted toward fully unreacted alkyne plus fully tetra-brominated product, with very little clean mono-adduct isolable, exactly as the large $k_{\text{alkene}}/k_{\text{alkyne}}$ ratio predicts. Step 3 — quantitative estimate. Treating this as parallel pseudo-first-order-in-Br₂ competition at the moment both species are present at comparable concentration, the instantaneous fraction of Br₂ that reacts with the alkyne rather than with any alkene already formed is approximately:

Even if $[\text{alkene}]$ is only $1\%$ of $[\text{alkyne}]$ at some instant (i.e. $[\text{alkene}]/[\text{alkyne}] = 0.01$), the ratio becomes:

meaning Br₂ is consumed by the small pool of already-formed alkene four times faster, per unit concentration, than by the much larger pool of remaining alkyne — confirming that mono-adduct cannot be cleanly isolated this way once any alkene has formed locally. Step 4 — the actual practical fix. In practice, chemists avoid this altogether by using N-bromosuccinimide (NBS) or controlled, very slow, dropwise addition of exactly 1.00 equiv Br₂ at low temperature with vigorous stirring to keep the instantaneous local Br₂ concentration low everywhere (so that no local excess ever accumulates to drive over-bromination), or — far more commonly in synthesis — they simply do not try to mono-brominate an alkyne at all, instead using the far more controllable Lindlar hydrogenation or a halogenation on the pre-formed cis-alkene obtained by a different, cleaner route. The kinetic argument above is precisely why alkyne dihalogenation, unlike alkyne hydrogenation, is rarely used industrially for mono-addition — the catalyst-poisoning trick available to H₂/Pd has no clean ionic analogue for Br₂.

Herbert Henry Mervyn Lindlar (1909–1998) was a Swiss industrial chemist working at Hoffmann-La Roche when, in 1952, he developed the poisoned palladium catalyst that now bears his name, originally in the context of vitamin A and carotenoid synthesis, where a specific cis double bond geometry had to be installed reliably and reproducibly on an industrial scale. The elegance of Lindlar's solution was not a new reaction but a controlled catalyst: by precipitating palladium onto calcium carbonate and then partially poisoning the metal surface with lead acetate (and often a trace of quinoline as an additional moderator), he tuned the number and reactivity of accessible Pd sites so finely that alkyne adsorption and syn-hydrogenation could proceed while alkene re-adsorption and further reduction to the alkane was suppressed almost completely. This is chemistry solved at the level of surface engineering rather than at the level of a new bond-forming step — a reminder that mechanistic control in organic synthesis is sometimes achieved not by inventing a new curved arrow but by managing which surface sites are available for the old ones to occur on.