Aluminium Chemistry and the +3 State

Modern primary aluminium production still runs the Hall–Héroult/Bayer combination essentially unchanged in chemical principle: bauxite ore is digested in hot NaOH (dissolving amphoteric Al₂O₃ as aluminate while leaving basic Fe₂O₃ and much of the silica as red mud), the aluminate liquor is cooled and seeded to precipitate Al(OH)₃, which is calcined to pure Al₂O�3 and then electrolytically reduced in cryolite melts at industrial scales exceeding 60 million tonnes of aluminium per year worldwide. The energy intensity of the reduction step (the ~2.3 V thermodynamic floor computed above, inflated to 4+ V operating voltage) is why aluminium recycling is so attractive: remelting scrap aluminium consumes roughly 5% of the energy of primary production, because it bypasses the Al³⁺-to-Al⁰ reduction step entirely. Anodizing exploits the same passivation chemistry deliberately: an aluminium part made the anode in dilute sulfuric or oxalic acid grows a thick, porous, controllable Al₂O�3 layer (up to tens of micrometers) that can be dyed and sealed, giving the colored, hard-wearing surfaces seen on architectural cladding, cookware, and consumer electronics casings. The corundum structure of α-Al₂O₃ — a hexagonally close-packed oxide lattice with Al³⁺ in two-thirds of the octahedral holes — is also the host lattice for ruby (Cr³⁺-doped, substituting for Al³⁺ owing to nearly identical ionic radius and matching +3 charge) and sapphire (Fe²⁺/Ti⁴⁺-doped), both of which are synthesized industrially (Verneuil flame-fusion process) for watch crystals, laser rods, and scratch-resistant optical windows — a direct materials-science consequence of Al³⁺'s small, hard, charge-matched octahedral site. In medicine, aluminium hydroxide gel is a common antacid, exploiting exactly the acid-dissolution branch of the amphoteric equilibrium (neutralizing excess stomach HCl by converting Al(OH)₃ to soluble Al³⁺ species, later re-precipitated and excreted), and aluminium salts (notably aluminium hydroxide and aluminium phosphate) are used as adjuvants in many vaccines, where the Al³⁺-based colloidal particles adsorb antigen and enhance immune response. In organic synthesis, the Lewis acidity of AlCl₃ — the same empty-3p-orbital hunger for electron density discussed throughout this article — makes it the classical catalyst for Friedel–Crafts alkylation and acylation, and related organoaluminium compounds (triethylaluminium, methylaluminoxane) are essential co-catalysts in Ziegler–Natta and metallocene polyolefin manufacturing, underpinning a large fraction of global polyethylene and polypropylene production. Water treatment plants use aluminium sulfate ("alum," KAl(SO₄)₂·12H₂O or plain Al₂(SO₄)₃) as a coagulant, precisely because adjusting pH into the amphoteric minimum-solubility window (as derived above) precipitates a flocculent Al(OH)₃ gel that sweeps out suspended particulates and organic matter. These threads — metallurgy, materials science, medicine, catalysis, and environmental engineering — all trace back to the same starting fact examined here: aluminium commits all three valence electrons, every time, and the resulting small, hard, triply charged cation has a chemical personality unlike almost anything else in the periodic table.

Ask the tutor: How does the bonding picture change between monomeric AlCl₃ (gas phase) and its dimer Al₂Cl₆, and why does BCl₃ never dimerize the same way?

Aluminium is atomic number 13, electron configuration [Ne]3s²3p¹. The valence shell has three electrons in orbitals of comparable energy (3s and 3p are not split by a large gap the way that, say, a filled d-subshell would create). Removing all three costs a lot of energy in absolute terms, but there is no energetic "cliff" between removing electron two and electron three the way there is, for example, between removing s-electrons and reaching into a filled inner d-shell in the post-transition metals. Compare this with thallium (period 6, [Xe]4f¹⁴5d¹⁰6s²6p¹): there, relativistic contraction and poor screening by the filled 4f and 5d shells stabilize the 6s² pair so strongly that Tl(I) is the dominant, often more stable, oxidation state, and Tl(III) compounds are oxidizing. Aluminium, at the top of the group, shows none of this — Al(I) and Al(II) species are known only as reactive, high-temperature gas-phase curiosities or in specialized low-temperature matrix-isolation and cluster chemistry, never as bulk, everyday compounds. The consequence is that virtually every aluminium compound you will encounter — the oxide, the halides, the hydroxide, organometallics, the hydrated ion — has aluminium in the +3 state with a formal electron count of zero valence electrons on the metal (all three donated or shared away). This gives Al³⁺ a very particular chemical personality: it is a small, "hard" Lewis acid (in Pearson's HSAB sense) that prefers hard, oxygen- and fluorine-donor ligands, forms strong ionic lattices with O²⁻ and F⁻, and is an extremely effective catalyst for electrophilic reactions precisely because its empty 3p orbital wants electron density from somewhere. Structurally, Al³⁺ almost always appears in six-coordinate octahedral or four-coordinate tetrahedral environments, obeying its ionic radius ratio with common anions. In γ-Al₂O₃ and in aqueous solution, Al³⁺ sits in an octahedral hole surrounded by six oxygen donors; in AlCl₄⁻ and in many AlR₃ derivatives, it is tetrahedral. The switch between coordination numbers is governed by radius-ratio rules and by whether the ligand can bridge to satisfy the metal's Lewis acidity (chloride bridges readily; water and hydroxide do not need to, because Al³⁺ is already saturated at CN 6 with those smaller, harder donors).

The dominance of Al(III) is a thermodynamic balance, not a statement that ionization is "easy." The relevant energetic bookkeeping for forming an ionic lattice MX₃ from the elements is the Born–Haber cycle:

where:

• $\Delta H_{\text{sub}}(M)$ is the enthalpy of atomizing/subliming the metal,
• $IE_i(M)$ are the successive ionization energies of the metal (here $i=1,2,3$),
• $\Delta H_{\text{diss}}(X_2)$ and $EA(X)$ are the dissociation and electron-affinity terms for the nonmetal,
• $\Delta H_{\text{lattice}}$ is the lattice enthalpy of the solid MX₃, always large and negative for a small, triply-charged cation.

Reading it: the huge, unfavorable sum of ionization energies (for Al, $IE_1+IE_2+IE_3 \approx 577+1817+2745=5139$ kJ/mol) is paid for almost entirely by the lattice enthalpy term, which scales with the product of ionic charges and inversely with interionic distance via the Kapustinskii/Born–Landé relationship. Because Al³⁺ is small (ionic radius 53.5 pm, six-coordinate) and triply charged, that lattice term is enormous and negative — easily large enough to make $\Delta H_f^\circ$ strongly exothermic overall. The second governing relationship is the amphoteric solubility behavior, captured qualitatively by the charge-density (polarizing power) parameter:

where:

• $z$ is the formal ionic charge,
• $r$ is the ionic radius (in the same units, e.g. pm or Å),
• $\phi$ (sometimes written as ionic potential) measures how strongly the cation polarizes a coordinated anion or water molecule.

Reading it: cations with very low $\phi$ (large, singly charged, e.g. K⁺, Cs⁺) form purely ionic, basic oxides/hydroxides. Cations with very high $\phi$ (small, highly charged, e.g. Si⁴⁺, P⁵⁺) form covalent, acidic oxides. Al³⁺, with $z=3$ and $r=53.5$ pm, falls squarely in the intermediate "amphoteric belt" of the Cartledge/Kapustinskii style ionic-potential map — this is why Al₂O₃ and Al(OH)₃ dissolve in both acids and strong bases, unlike the clearly basic Na₂O or clearly acidic SO₃.

Example 1 (foundational): Electron configuration and the "why" of +3 only. Write the electron configuration of Al and Al³⁺, and use it to explain why Al²⁺ compounds are not encountered in ordinary chemistry. Neutral Al: $1s^2 2s^2 2p^6 3s^2 3p^1$, i.e., [Ne]$3s^23p^1$. Removing the single 3p electron gives Al²⁺: [Ne]$3s^2$, a species with one electron still unpaired... no, actually with two electrons left in 3s and a formal charge of +2. The issue is not configuration validity but energetics: $IE_2 = 1817$ kJ/mol (removing the first 3s electron from Al⁺) is large but not anomalously so; however, once you remove that second electron, the resulting Al²⁺ is a good enough Lewis acid, and the third ionization energy is close enough in magnitude ($IE_3 = 2745$ kJ/mol, only about 1.5× larger), that lattice/solvation energetics for typical anions always overcompensate for stripping the third electron as well. There is no configurational reason (no big s/p gap, no filled inner shell) for the ionization to stop at Al²⁺. Compare with Tl: $IE_1(\mathrm{Tl}) = 589$, $IE_2 = 1971$, $IE_3 = 2878$ kJ/mol — individually similar in scale to Al's — yet Tl(I) persists because the relativistically contracted, poorly shielded 6s² pair in Tl⁺ is disproportionately stabilized in its own right (a genuine electronic effect, the inert pair), not simply an energetics coincidence. Aluminium has no such filled-shell stabilization mechanism at 3s², so nothing arrests the process at +2; the cascade runs to completion at +3.

Example 2 (applied): Mass and charge balance in the Bayer process. A bauxite ore sample contains 45.0% Al₂O�3, 20.0% Fe₂O₃, and the remainder inert silicates/moisture, by mass. If 1000 kg of this ore is digested in hot concentrated NaOH, calculate (a) the mass of Al(OH)₃ that can theoretically be recovered after precipitation, and (b) the moles of NaOH consumed in the dissolution step (ignoring silica reactions). Digestion dissolves only the amphoteric Al₂O₃, via

Fe₂O₃ is basic (iron(III) is far less amphoteric at these conditions) and remains as insoluble "red mud." Mass of Al₂O₃ = $0.450 \times 1000 = 450$ kg. Molar mass of Al₂O�3 = $2(27.0) + 3(16.0) = 102.0$ g/mol.

From the digestion equation, moles NaOH used $= 2 \times 4412 = 8824$ mol, i.e. about 353 kg NaOH (M = 40.0 g/mol). On precipitation (seeding and cooling reverses the equilibrium):

Each mole of Al₂O₃ gives 2 mol NaAl(OH)₄, hence 2 mol Al(OH)₃, so

Molar mass Al(OH)₃ = $27.0 + 3(16.0+1.0) = 78.0$ g/mol, giving

This Al(OH)₃ is then calcined ($2\,\mathrm{Al(OH)_3} \to \mathrm{Al_2O_3} + 3\mathrm{H_2O}$, releasing water) to regenerate pure Al₂O₃ for the electrolysis step, closing the loop between the amphoteric-separation chemistry and the metallurgical reduction chemistry. As a limiting-case check: if the ore contained no Fe₂O₃ at all (pure Al₂O₃/silica ore), the same calculation would apply essentially unchanged (Fe₂O₃ never entered the NaOH mass balance), confirming that the Fe₂O₃ percentage is a distractor for this particular sub-calculation — a common exam trap.

Example 3 (exam/competition level): Estimating the Hall–Héroult cell voltage and current efficiency. Molten cryolite–alumina electrolysis at ~960 °C effectively runs the net reaction

Given $\Delta G^\circ$ for this reaction at 960 °C is approximately $-1340$ kJ per mole of Al₂O₃ reacted (i.e., $-2680$ kJ for the reaction as written per 2 mol Al₂O₃), and that the reaction transfers 12 electrons per formula unit as written (4 Al³⁺ each requiring 3 electrons), estimate the theoretical decomposition voltage, then account for why real industrial cells run at ~4.0–4.5 V rather than this theoretical value. Using

where:

• $n = 12$ mol electrons per reaction as written (4 Al × 3 e⁻ each),
• $F = 96{,}485$ C/mol,
• $\Delta G^\circ = -2{,}680{,}000$ J,

So the thermodynamic minimum cell voltage is about 2.3 V. Yet industrial cells operate at 4.0–4.5 V, and the difference (roughly 1.7–2.2 V) is not wasted arbitrarily — it is consumed by three real physical processes: (i) overpotentials at the carbon anode (kinetic barrier to CO₂ evolution, typically 0.3–0.5 V), (ii) ohmic (IR) losses through the molten cryolite electrolyte and busbars, which dominate because the electrolyte, though ionically conductive, still has appreciable resistance over cell dimensions of meters, and (iii) back-reaction/current-inefficiency losses from partial reoxidation of Al by dissolved CO₂ or from carbon-anode side reactions producing some CO instead of CO₂ (which changes the effective $n$ and lowers Faradaic efficiency to typically 90–95%). As a limiting case: if the cell instead operated reversibly at zero current (open-circuit, infinitesimally slow), the measured voltage would approach the thermodynamic 2.3 V exactly — this is the textbook distinction between the reversible EMF (a thermodynamic state function) and the operating voltage (which must overpay for kinetics and resistance to deliver industrial current densities of order 0.7–1.0 A/cm²). This is also why the Hall–Héroult process is so energy-intensive (roughly 13–15 kWh per kg Al produced) — most of that energy above the thermodynamic minimum is dissipated as heat that in practice helps keep the cryolite bath molten.

Friedrich Wöhler improved on Ørsted's method in 1827, reducing AlCl₃ with potassium metal to obtain aluminium powder pure enough to characterize its density and basic reactivity, definitively establishing it as a distinct light metal. For decades aluminium remained a laboratory curiosity and a luxury item — Napoleon III reportedly reserved aluminium cutlery for his most honored dinner guests, ranking it above gold and silver, precisely because chemical reduction from the exceptionally stable oxide was so difficult and expensive. The breakthrough came in 1886, when Charles Martin Hall in the United States and Paul Héroult in France, working independently and almost simultaneously (both aged 22, born the same year, and both dying the same year in 1914), discovered that alumina dissolves in molten cryolite (Na₃AlF₆) at ~950 °C, allowing electrolytic reduction at a temperature far below alumina's own melting point of 2072 °C. This single insight — using an amphoteric-oxide-in-fluoride-flux electrolyte rather than trying to melt or chemically reduce the oxide directly — collapsed the price of aluminium roughly a thousandfold within a generation and remains the basis of essentially all primary aluminium production today.