Wade-Mingos Rules: Counting Electrons in Boron Cage Clusters

The puzzle boranes posed — and why classical valence failed

By the 1950s the boron hydrides were a scandal for structural theory. Take B₅H₉ (pentaborane-9): five borons, nine hydrogens, and only 24 valence electrons — 12 electron pairs. A classical Lewis structure needs a two-center, two-electron (2c-2e) bond for every connecting line, but the observed square-pyramidal cage has far more B–B contacts than there are pairs to fill them. Boron is simply electron-deficient: with only three valence electrons and four valence orbitals per atom, it cannot supply enough pairs to draw localized bonds along every edge.

William Lipscomb's answer (Nobel Prize, 1976) was the three-center, two-electron (3c-2e) bond — most famously the B–H–B bridge in diborane, B₂H₆, where a single pair of electrons binds three atoms. Lipscomb's styx topological bookkeeping (counting s = B–H–B bridges, t = 3c-2e BBB bonds, y = 2c-2e B–B bonds, x = terminal BH₂ groups) rationalized many boranes, but it was combinatorial and did not predict which polyhedron a given formula would adopt. You still had to know the structure to assign the bonds.

The conceptual break came from treating the cage the way Hückel treats benzene. In benzene, six π electrons delocalize over a ring and you don't ask which carbon owns which electron — you count the filled bonding MOs of the whole ring. Kenneth Wade, at Durham, realized in 1971 that a borane deltahedron is the three-dimensional version of this idea: the skeletal electrons occupy a closed set of cage molecular orbitals, and the number of skeletal pairs — not their localization — fixes the geometry. This reframing turned an intractable topology problem into arithmetic.

The counting recipe: fragments, skeletal pairs, and the n+1 rule

The heart of Wade-Mingos counting is the separation of each vertex fragment's electrons into two jobs. A conical BH vertex uses one of boron's four orbitals and two of its electrons to form the exo (outward-pointing) terminal B–H bond. That leaves three orbitals and two electrons pointing inward toward the cage: one radial sp-hybrid pointing at the center, and two tangential p orbitals. Those inward-pointing electrons are the skeletal electrons; the three inward orbitals are exactly what a deltahedral vertex needs. So a neutral BH unit contributes 2 skeletal electrons.

To count a whole cluster, sum the skeletal electrons from every vertex fragment, add electrons from any charge, then divide by two to get the number of skeletal electron pairs (SEP). The standard fragment donations are:

• B–H (or any main-group E–H with 3 valence electrons): 2 skeletal e⁻
• C–H (4 valence electrons): 3 skeletal e⁻ — the extra electron is why carbon substitutes for BH⁻
• A bare heavy main-group atom E with v valence electrons (no exo H): v − 2 skeletal e⁻ (two electrons tied up in an exo lone pair) — e.g. Pb, Sn: 4−2 = 2; Bi: 5−2 = 3
• A transition-metal L₄M or (CO)₃M fragment: v + x − 12 skeletal e⁻, where v is the number of valence electrons of the metal (its group number) and x the electrons from cage-facing ligands — the −12 reflects the twelve electrons the metal buries in its non-skeletal (t₂g-like) and exo orbitals

Now the rule that ties count to shape. Wade's central result: a closo cluster built on an n-vertex deltahedron is stable when it has n + 1 skeletal electron pairs. The physical reason, made rigorous by Anthony Stone's Tensor Surface Harmonic theory (1980), is that an n-vertex deltahedron possesses exactly one strongly bonding radial MO (fully symmetric, the S orbital) plus n bonding tangential MOs — together n + 1 bonding cage orbitals. Fill all n + 1 and you have a closed shell. Remove a vertex without removing electrons and the same n + 1 pairs now blanket an (n−1)-vertex framework, giving n + 2 pairs per remaining vertex — a nido cage. Remove two vertices: arachno, n + 3 pairs. Each vertex you strip opens the deltahedron by one, and the electron count per occupied vertex climbs by one pair.

closo, nido, arachno: reading structure straight off the count

The predictive payoff is that you never need the structure in advance. Compute the SEP count, compare it to the number of occupied vertices n, and the difference names the family. Let p be the total skeletal pairs:

• p = n + 1 → closo (from clovo/klóbos, cage): a complete, closed deltahedron. Example: B₆H₆²⁻, n = 6.
• p = n + 2 → nido (Latin nidus, nest): the deltahedron with one vertex missing, an open face. Example: B₅H₉, n = 5.
• p = n + 3 → arachno (Greek arachnē, web): two adjacent vertices missing. Example: B₄H₁₀, n = 4.
• p = n + 4 → hypho (net): three vertices missing — rarer, e.g., some borane-Lewis base adducts.

The deep structural insight is the parent-deltahedron principle: closo, nido, and arachno clusters that share the same total skeletal pair count are all fragments of the same parent deltahedron. A cluster with 7 skeletal pairs is derived from the octahedron regardless of how many vertices are actually occupied — as closo-B₆H₆²⁻ (6 vertices), nido-B₅H₉ (5 vertices), or arachno-B₄H₁₀ (4 vertices). This is why the whole periodic zoo of boranes falls onto a single genealogical chart, first drawn by R. E. Williams (whose empirical structural correlations predated and inspired Wade's formalization).

Crucially, when you remove vertices you remove them from the highest-connectivity positions of the parent deltahedron, and the missing vertices leave the open face that the bridging and endo hydrogens then decorate. That is exactly why B₅H₉ is a square pyramid (octahedron minus one apex) with four B–H–B bridges around the open basal face, and why the count alone tells you where the extra hydrogens live.

Worked examples: from B₁₂H₁₂²⁻ to a metallacarborane

B₁₂H₁₂²⁻. Twelve BH vertices at 2 e⁻ each = 24 skeletal electrons; the 2− charge adds 2 more = 26 skeletal electrons = 13 pairs. With n = 12, we have p = n + 1, so this is closo: the closed 12-vertex deltahedron is the icosahedron (Iₕ). Every vertex is equivalent, every B carries one exo H, and the 13 filled bonding cage MOs make it aromatic in three dimensions — hence its extraordinary thermal and chemical stability (it survives boiling in concentrated acid).

C₂B₁₀H₁₂ (o-carborane). Ten BH (20 e⁻) plus two CH (2 × 3 = 6 e⁻) = 26 skeletal electrons = 13 pairs, neutral. Twelve vertices, p = n + 1 → closo icosahedron again. This is the exact isoelectronic logic Wade's rules make transparent: two CH units (3 e⁻ each) replace two BH⁻ units (also 3 e⁻ each), so o-carborane is the neutral analogue of B₁₂H₁₂²⁻ and adopts the identical icosahedral cage.

B₅H₉. Five BH (10 e⁻) plus four extra bridging/endo H atoms (4 × 1 = 4 e⁻) = 14 skeletal electrons = 7 pairs. With n = 5, p = n + 2 → nido. Seven pairs points to the octahedral parent; removing one vertex gives the square pyramid, with the four bridging H's decorating the open basal square. Prediction and reality match exactly.

[(η⁵-C₅H₅)Fe(C₂B₉H₁₁)]⁻ and the metallacarboranes. Here Wade's rules and the isolobal analogy fuse. The dicarbollide ligand C₂B₉H₁₁²⁻ is nido (icosahedron minus one vertex) and presents an open C₂B₃ pentagonal face whose frontier orbitals are essentially those of cyclopentadienyl (Cp⁻). M. Frederick Hawthorne recognized this in 1965: a transition-metal fragment slots into the missing vertex to complete the icosahedron. Counting the whole 12-vertex cage — 9 BH + 2 CH + a d-metal fragment supplying the right skeletal electrons — returns 13 pairs, closo, icosahedral. The metal literally caps the cage as the twelfth vertex.

Mingos's extensions: capping, fusion, and metal clusters

Wade counted single deltahedra; D. Michael P. Mingos generalized the framework to the messier real world of condensed and metal-rich clusters, and it is this generalization that earns the compound name Wade-Mingos rules (the whole edifice is also called PSEPT, polyhedral skeletal electron pair theory).

• Capping principle: adding a vertex that caps a triangular face of a deltahedron contributes no new skeletal bonding MO. Each capping vertex adds 12 electrons to the total valence count but leaves the skeletal pair requirement of the underlying deltahedron unchanged. A mono-capped n-vertex cluster keeps the n+1-pair count of its parent.
• Condensed-cluster (fusion) rule: for polyhedra sharing a vertex, edge, or face, the total electron count equals the sum of the counts for the individual polyhedra minus the count of the shared unit (a shared atom, edge, or triangular face). This single subtraction rule handles fused clusters like B₂₀H₁₆ that no single-deltahedron count can.
• Main-group / transition-metal unification: Mingos showed that the same electron-counting logic bridges electron-precise organometallic clusters and electron-deficient boranes through the magic numbers 4n+2 (main-group closo, ~4 electrons per vertex) and 14n+2 (transition-metal closo, e.g. 86 valence electrons for an octahedral M₆ cluster) — the 18-electron rule per metal vertex plus the shared skeletal pairs. The isolobal bridge (Roald Hoffmann, Nobel 1981) makes a BH vertex, a CH⁺ vertex, and a d⁸ Fe(CO)₃ fragment interchangeable — each a 2-skeletal-electron donor offering three frontier cage orbitals; a neutral CH vertex, a 3-electron donor, is isolobal instead with Co(CO)₃.

The reach is remarkable. The same arithmetic that shapes B₆H₆²⁻ also predicts the tetrahedral, trigonal-bipyramidal, and octahedral geometries of naked Zintl ions — Pb₅²⁻ (closo, trigonal bipyramid), Sn₉⁴⁻ and Ge₉⁴⁻ (nido/closo nine-vertex), Bi₉⁵⁺ — where a bare main-group atom with v valence electrons donates v − 2 skeletal electrons (Pb, Sn: 4−2 = 2; Bi: 5−2 = 3). For most of these the count reproduces the crystal structure, though Bi₉⁵⁺ is a well-known partial exception: the simple Wade count predicts a nido cage, yet the observed geometry is a tricapped trigonal prism.

Where the rules bend: limits, exceptions, and the deeper theory

Wade-Mingos counting is an astonishingly accurate heuristic, but it is a topological theory of orbital counting, not an energy calculation, and it has well-mapped failure modes. Students should know them.

• hypercloso / isocloso clusters. Certain metallaboranes adopt n-vertex closed deltahedra with only n skeletal pairs — one pair short of Wade's n+1. These 'hypercloso' or 'isocloso' species (studied extensively by John Kennedy and by Fehlner) genuinely violate the count; the resolution involves the metal's ability to accommodate the reduced electron count through its d orbitals, and the exact deltahedron is often distorted from the ideal.
• Very electron-rich or four-connect clusters. For low-connectivity or three-connected polyhedra (cubes, prisms), the simple deltahedral MO template breaks down; Mingos's separate rules for three-connected clusters (electron count 5n) apply instead. The n+1 rule is specifically a deltahedral (all-triangular-face) rule.
• Ambiguous parents. Some counts fit more than one plausible polyhedron, and second-order factors — heteroatom placement, steric strain, HOMO-LUMO gaps — decide. Wade's rules tell you the family and connectivity but not always which isomer (e.g., ortho vs meta vs para carborane) is most stable.

The rigorous justification of why n+1 works — and of when it should fail — is Anthony Stone's Tensor Surface Harmonic (TSH) theory (1980). Stone treated the cage surface like a sphere and expanded the vertex orbitals in spherical harmonics: the radial orbitals generate S, P, D... surface functions, and the tangential orbitals generate π-type Sπ, Pπ, Dπ sets. For a deltahedron the bonding combinations number exactly n + 1 (one radial S plus n tangential), which is Wade's rule derived rather than asserted. TSH also explains the exceptions: when the P-type radial set becomes bonding or antibonding out of the ideal order, you get hypercloso and other anomalies. So the practical rule and the deep theory close the loop — a rare and satisfying situation in structural chemistry, and the reason PSEPT remains the first tool any inorganic chemist reaches for when a new cluster formula lands on the bench.

The three principal deltahedral cluster classes under Wade-Mingos counting

Property	closo	nido	arachno
Skeletal electron pairs (SEP)	n + 1	n + 2	n + 3
Parent deltahedron	n-vertex closed	(n+1)-vertex, 1 vertex removed	(n+2)-vertex, 2 vertices removed
Vertices occupied	all n	n of n+1	n of n+2
Archetypal borane	B₆H₆²⁻ (octahedron)	B₅H₉ (square pyramid)	B₄H₁₀ (butterfly)
General neutral BₙHₘ formula	BₙHₙ²⁻ / CₐBₙ₋ₐHₙ	BₙHₙ₊₄	BₙHₙ₊₆
Shape descriptor	closed cage	nest / open face	web / two open vertices