Lewis Acid-Base Theory

How Lewis acid-base theory works

Arrhenius defined an acid as anything releasing H⁺ in water and a base as anything releasing OH⁻. That worked for vinegar but failed outside water — ammonia is plainly a base in liquid ammonia. Brønsted-Lowry fixed half the problem by reframing acids as proton donors and bases as proton acceptors. Lewis went further. Looking at proton transfer microscopically, he saw that the base's lone pair always did the work of capturing H⁺. Strip that requirement away and you get the cleanest definition: an acid is anything an electron pair can attack, a base is anything that has an electron pair to attack with.

The territory broadens enormously. Every cation with an empty orbital — H⁺, Li⁺, Mg²⁺, Cu²⁺, Fe³⁺, Al³⁺ — is a Lewis acid. Every electron-deficient neutral molecule — BF₃, AlCl₃, SO₃, carbocations — is too. Every species with a lone pair is a Lewis base: water, ammonia, alcohols, ethers, halide ions, cyanide, the lone pair on sulfur in a thioether, even the π-electrons of a benzene ring. The acid-base reaction is no longer proton transfer but bond formation, drawn with a curved arrow from base lone pair to acid empty orbital.

Worked example: the BF₃ + NH₃ adduct

Boron trifluoride is the canonical neutral Lewis acid. In its trigonal-planar resting state, boron is sp²-hybridised with three B-F sigma bonds and one empty 2p orbital perpendicular to the molecular plane. Boron has only six valence electrons — two short of an octet — and that vacant 2p is hungry. Ammonia, by contrast, has a lone pair pointed along the sp³ axis, sitting on top of the trigonal pyramid of N-H bonds.

When they collide with the right orientation, the nitrogen lone pair drops into the empty boron 2p:

     F                   F
     |                   |
F — B          +     :NH3   →    F — B — NH3
     |                              |
     F                             F
   (sp², trigonal             (sp³, tetrahedral
    planar, empty p)           around B; formal -)
                                    |
                                  +NH3 (formal +)

   Net: F3B ← :NH3   (coordinate bond from N to B)

Boron rehybridises from sp² to sp³ — the fluorines bend back, the molecule becomes tetrahedral. Formal charges land at +1 on N and −1 on B. The H₃N-BF₃ adduct is a stable solid that sublimes near 125 °C and serves as a standard delivery form of BF₃ in synthesis.

The same arrow-pushing explains why AlCl₃ grabs a chloride to form AlCl₄⁻, why Cu²⁺ binds four ammonia molecules to form deep-blue [Cu(NH₃)₄]²⁺, and why a carbocation reacts with a nucleophile in the second step of SN1. All are Lewis acid-base bond formations.

Three definitions of acid-base — Arrhenius, Brønsted, Lewis

	Arrhenius (1884)	Brønsted-Lowry (1923)	Lewis (1923)
Acid	Releases H⁺ in water	Proton donor	Electron-pair acceptor
Base	Releases OH⁻ in water	Proton acceptor	Electron-pair donor
Solvent required	Water	Any (proton-transferring)	None — gas, liquid, solid
NH3 in liquid ammonia	Not classifiable	Acid or base depending on partner	Base
BF3 + NH3 reaction	Not an acid-base reaction	Not an acid-base reaction	Acid-base reaction
Metal cation hydration	Not classifiable	Indirect (via acidic hydrate)	Direct: metal acid + water base
Captures organic mechanisms	Almost never	Sometimes	Most of them

Each definition is a strict superset of the previous. Arrhenius is a special case of Brønsted (water is the proton-transferring solvent). Brønsted is a special case of Lewis (the proton's empty 1s orbital accepts the base's lone pair). The Lewis picture is the most powerful but also the most permissive, which is both its strength and the source of most of its pitfalls.

Hard and soft acids and bases

Lewis theory tells you whether an acid-base reaction can happen but not which pair will react preferentially when several options compete. Ralph Pearson's hard-soft framework (1963) supplies the missing rule: hard acids prefer hard bases, soft acids prefer soft bases.

• Hard acids are small, highly charged, and not easily polarised — H⁺, Li⁺, Na⁺, Mg²⁺, Al³⁺, Fe³⁺, Ti⁴⁺. Their orbitals are tight and the acid-base interaction is dominated by ionic, electrostatic attraction.
• Hard bases mirror this — F⁻, OH⁻, NH₃, NO₃⁻, CO₃²⁻, SO₄²⁻. Small, electronegative donor atoms, lone pair tightly held.
• Soft acids are large, low-charge, polarisable — Cu⁺, Ag⁺, Au⁺, Hg²⁺, Pd²⁺, Pt²⁺, I₂, BH₃. Their valence electrons are diffuse and the acid-base interaction is dominated by covalent overlap.
• Soft bases match — I⁻, S²⁻, RS⁻, R₃P, CN⁻, CO, alkenes. Big, polarisable, lone pair loosely held.

The rule predicts geology and biochemistry well. Hard oxophilic metals (Al, Ti, rare earths) concentrate as oxide ores. Soft metals (Cu, Ag, Au, Hg, Pb) concentrate as sulfide ores — soft S²⁻ pairs with soft cations. Mercury and lead toxicity tracks HSAB: both bind cysteine thiols, choking enzymes.

Real-world applications

• Friedel-Crafts catalysis. AlCl₃ coordinates to the chloride leaving group of an alkyl halide, generating a carbocation that attacks an aromatic ring. Industrial ethylbenzene (≈30 Mt/year, the styrene precursor) runs on this Lewis-acid cycle.
• Chelation therapy. EDTA is a hexadentate Lewis base that wraps around hard cations like Pb²⁺ and Ca²⁺. Clinical EDTA infusions strip lead from tissue.
• Hemoglobin. Iron(II) in a porphyrin ring is a Lewis acid; it binds O₂ as a Lewis base. CO is a tighter Lewis base than O₂, which is why CO poisoning displaces oxygen and is fatal above 1200 ppm.
• Zeolite cracking. Al-substituted silicate frameworks expose Lewis-acidic Al³⁺ sites that crack long-chain hydrocarbons in fluid catalytic crackers handling ~11 million barrels per day worldwide.

Variants

• Frustrated Lewis pairs (FLPs). Bulky acid and base that cannot combine because of steric repulsion. Discovered by Stephan in 2006, they cleave H₂ heterolytically at room temperature — historically the province of transition-metal catalysts.
• π-acceptor ligands. CO and CN⁻ donate a σ lone pair but also accept back-donation into empty π* orbitals. Standard Lewis arrows undercount the bonding.
• Solid Lewis acids. Acidic sites on alumina, silica-alumina, and zeolite surfaces — the working currency of heterogeneous petroleum catalysis.
• Usanovich (1939). Even broader: any reaction with cation-anion pairing or electron transfer counts. Rarely used — too general to constrain mechanism.

Common pitfalls

• Calling every reaction Lewis acid-base. If the bond formed isn't coordinate (radicals, pericyclics), Lewis is the wrong frame. Restrict the label to mechanisms where one partner contributes both electrons.
• Confusing Brønsted and Lewis basicity. Pyridine is a stronger Brønsted base than triethylamine in water but a weaker Lewis base toward soft metals — different probes test different things.
• Ignoring HSAB when comparing strengths. Cu⁺ binds RS⁻ far more tightly than RO⁻ despite oxygen being more electronegative. Pure electrostatic intuition fails for soft cations.
• Forgetting solvent effects. Gas-phase Lewis acids (BF₃, AlCl₃) react with water before doing anything else in aqueous solution. Mechanism arrows must honor the actual reactive species.
• Curly arrow backwards. Arrow goes from base's lone pair to acid's empty orbital, never the reverse — the most common arrow-pushing error in intro organic.