Boron Clusters & Wade's Rules

An element that is short on electrons

The previous guide left boron in an awkward spot. With the configuration 2s2 2p1, a boron atom brings only three valence electrons to share, yet it has four valence orbitals (one 2s and three 2p) waiting to be used. That mismatch — more orbitals than electrons — is the definition of an [[electron-deficient-compounds|electron-deficient compound]], and it is boron's whole personality. A carbon atom, with four electrons for four orbitals, can pair off neatly into four ordinary two-centre two-electron bonds. Boron simply cannot. It is always one electron short of the party, and what it does about that shortage is one of the most elegant stories in the p-block.

You already met boron's first survival trick: when there are not enough electrons to glue atoms together one bond at a time, it spreads each bonding pair over *three* atoms at once. That is the [[three-center-two-electron-bond|three-centre two-electron bond]], the heart of diborane B2H6, where two electrons hold three nuclei together in a banana-shaped bridge. [[electron-deficient-bonding|Electron-deficient bonding]] of this kind is not a defect to be apologised for — it is a different, perfectly legitimate way to build molecules, one where electrons are shared communally rather than handed out in private pairs. The cage chemistry in this guide is simply that idea taken to its dazzling conclusion.

Cages built on a deltahedron

If you cannot afford one bond per pair of neighbours, the smart move is to huddle. Instead of a chain or a sheet, boron atoms gather into a compact ball where every atom touches several others, so the few electrons available can be smeared over the whole framework at once. The natural shapes for such huddling are deltahedra — closed polyhedra whose every face is a triangle (the name comes from the Greek capital delta). The octahedron (6 vertices, 8 triangular faces) and the icosahedron (12 vertices, 20 triangular faces) are the famous ones. Put a boron atom at each vertex, point one B–H bond radially outward from each, and you have the skeleton of a [[boranes|borane]] cluster. These are real, isolable substances: B6H6 2-, B10H10 2-, the strikingly stable B12H12 2- icosahedron.

Not every cage is a perfect closed ball, though, and the open ones are where the family gets its names. Picture the closed parent deltahedron, then knock atoms off its vertices like teeth from a comb. A complete, closed cage is called closo (from the Greek for 'cage'). Lop off one vertex and you get an open bowl, nido (Latin for 'nest'). Lop off two adjacent vertices and the bowl opens further into a arachno cage (Greek for 'spider's web', because it looks lacy and open). There is even an extra-opened hypho class. The marvel is that nido and arachno clusters are not built on their *own* compact polyhedra — they keep the skeleton of a *larger* parent deltahedron with vertices simply missing. A nido six-atom cluster is an octahedron with one corner empty, not a five-vertex shape of its own.

Wade's rules: counting the skeleton

Here is the question that organises everything: given a cluster's formula, how do we predict its shape without building a model in the lab? Kenneth Wade answered it around 1971 with a counting scheme of startling reach, now called [[wades-rules|Wade's rules]] (or the polyhedral skeletal electron pair theory, PSEPT). The key insight is to separate two jobs the electrons are doing. Each vertex atom uses some of its electrons and one orbital to hold onto its outward-pointing terminal hydrogen (or other external group); those are *not* part of the cage. What is left over — the electrons each vertex contributes *inward*, into the shared skeleton — are the only ones that decide the shape.

Counting the skeletal electrons is mechanical once you see the bookkeeping. A B–H vertex brings 3 (from boron) + 1 (from its hydrogen) = 4 electrons, spends 2 of them on the terminal B–H bond, and so donates 2 electrons to the skeleton. A C–H vertex, as in a carborane, has one more electron than B–H, so it donates 3. Add up every vertex's inward contribution, then add any electrons from the overall charge of the ion, and divide by two: that is the count of skeletal electron pairs, the number we call S. The shape then follows from comparing S to n, the number of vertex atoms actually present.

1. Count the vertex atoms actually present and call that number n (e.g. for B5H9, n = 5).
2. Add up the skeletal electrons each vertex donates inward (B–H gives 2, C–H gives 3), plus the extra hydrogens beyond one per vertex (each bridging or extra H gives 1), plus electrons from the ion's overall charge.
3. Divide that total by 2 to get S, the number of skeletal electron pairs.
4. Read the shape from S: S = n + 1 means closo (n-vertex deltahedron); S = n + 2 means nido (parent deltahedron of n + 1 vertices, one missing); S = n + 3 means arachno (parent of n + 2 vertices, two missing).

Worked example: B5H9 (pentaborane-9)

  vertices present:  n = 5
  5 (B-H) units      -> 5 x 2  = 10 skeletal e-
  4 extra bridging H -> 4 x 1  =  4 skeletal e-
  charge = 0         ->            0
                                 ----
  total skeletal electrons       14  -> S = 14/2 = 7 pairs

  S = 7 = n + 2  ->  NIDO
  parent = (n+1) = 6-vertex octahedron, ONE vertex removed
  => square-pyramidal B5 cage.  (Exactly what is observed.)

Carboranes and the isolobal leap

The rules get their real power from a swap. Replace one B–H vertex with a C–H vertex and almost nothing changes in the count — except that carbon brings one extra electron, so the skeletal-electron total goes up by one. Crucially, a neutral C–H unit (4 electrons, donating 3 to the skeleton) is bookkeeping-equivalent to a B–H unit that has gained one electron, i.e. to [B–H]-. So a [[carboranes|carborane]] like C2B10H12 — two carbons and ten borons — counts exactly like the closo B12H12 2- icosahedron: same 13 skeletal pairs, same 12-vertex closed cage. The famous ortho-, meta-, and para-carboranes are simply that icosahedron with the two carbons sitting in adjacent, separated, or opposite vertices. Carboranes are some of the most thermally and chemically robust molecular cages known, stable enough to survive boiling in air.

Now comes the genuinely surprising part, and the reason Wade's rules outgrew boron entirely. What matters to the cage is not which element sits at a vertex but how many electrons and orbitals that vertex offers to the skeleton. A vertex needs roughly three orbitals pointing inward and a certain electron contribution. A great many fragments fit that bill — including transition-metal fragments. A metal carbonyl unit such as Fe(CO)3 or Co(CP), once you subtract the electrons it uses on its own external ligands, turns out to donate the same kind of inward contribution as a B–H or C–H vertex. Fragments that offer the same number of frontier orbitals with the same number of electrons are called isolobal — a term coined by Roald Hoffmann. A B–H vertex and an Fe(CO)3 vertex are isolobal, even though one is a tiny main-group scrap and the other a hefty metal carbonyl.

Because of that equivalence, the very same counting predicts the shapes of [[metal-cluster-compounds|metal cluster compounds]] — the polyhedral carbonyl clusters like Os6(CO)18 or the mixed metal-and-boron metallaboranes. You just count skeletal pairs with each metal fragment standing in for a vertex (a transition-metal vertex contributes its valence electrons minus the 12 it tucks away in the three orbitals it does not aim at the cage), then read closo, nido, or arachno exactly as before. A rule discovered to tame boron's electron poverty turned out to be a unifying principle stretching across half the periodic table. That is the deep reward of this guide: the same idea explains both a tiny borane and a six-osmium metal cage.

What the rules are, and are not

Be honest about what kind of tool this is. Wade's rules are an electron-counting *correlation*, not a law of nature; they are a brilliantly reliable bookkeeping device, much like the oxidation states and the 18-electron rule you met earlier — useful, predictive, and riddled with edge cases. There are well-known exceptions: clusters that adopt shapes the count would not predict, hypho and even more open classes, clusters with interstitial atoms hidden inside the cage, and bare post-transition-metal Zintl clusters like Pb9 4- whose counting needs care. The right attitude is the same one you bring to any model in this subject: trust it as a first prediction, then let experiment have the final word.

Step back and notice what just happened. Boron's so-called weakness — too few electrons for ordinary bonds — is exactly what forced it into a richer kind of architecture, and that architecture gave chemistry a counting rule of extraordinary breadth. It even echoes a theme from the start of this rung: boron's [[diagonal-relationship|diagonal relationship]] with silicon and its borderline metalloid character all trace back to the same electron-starved, orbital-rich starting point. Keep this mindset as you move on to carbon and silicon next: in the p-block, an element's shortcomings are usually the doorway to its most beautiful chemistry.