JOVANA
Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
All guides

The 18-Electron Rule

Why so many organometallic complexes settle on exactly eighteen valence electrons — the reasoning behind the number, and the honest, important exceptions that turn it from a law into a guide.

An octet, grown up

Back in your first chemistry lessons you met the octet rule: a main-group atom feels stable when it has eight electrons around it, the count that fills one s and three p orbitals and matches the nearest noble gas. The 18-electron rule is the same instinct, grown up and moved to the transition-metal side of the periodic table. A transition metal has more rooms in its house — it adds five d orbitals to that s and those three p orbitals — so its version of a full shell is not eight but eighteen. The rule says that many stable organometallic complexes arrange to surround the metal with exactly eighteen valence electrons, filling all nine of those orbitals and reaching the closed-shell calm of the next noble gas.

You already know how to do the headcount from the previous guide: it is the valence electron count, the total you get by adding the electrons the metal itself brings to the electrons every ligand donates. What changes here is the destination. Earlier we just tallied the number; now we ask what the number means. Eighteen is the target the count so often hits, the way a thermostat keeps drifting back to a set point, and learning to expect it lets you guess whether a complex will be content, hungry for another ligand, or eager to throw one off.

Why nine orbitals, why eighteen

The arithmetic is simple, but the reasoning underneath it is worth seeing clearly. A transition metal's valence shell offers one s orbital, three p orbitals, and five d orbitals — nine orbitals in all. An orbital can hold two electrons, so nine filled orbitals hold eighteen. When all nine are either filled with the metal's own electrons or used to accept a lone pair from a ligand, there is no low-lying empty orbital left to grab another bond and no high-lying filled orbital begging to dump electrons somewhere. That balance is what we call coordinative saturation, and it is why eighteen, not sixteen or twenty, is the comfortable number.

There is a deeper, molecular-orbital version of the same story, and it explains why the rule has teeth in some places and not others. When good ligands surround a metal, their orbitals combine with the metal's nine to form bonding, nonbonding, and antibonding combinations. In a happy octahedral complex you typically get nine bonding-or-nonbonding orbitals lying low — six metal-ligand bonds plus a set of three roughly nonbonding d orbitals (the t2g, which point into the gaps between ligands) — and the antibonding combinations sit well above. Filling the nine low orbitals and leaving the antibonding ones empty puts exactly eighteen electrons in safe places. The rule bites hardest when that gap to the antibonding orbitals is large.

9 valence orbitals: one s + three p + five d
  filled at 2 e- each  ->  9 x 2 = 18 electrons

Cr(CO)6 :  Cr (group 6) = 6 e-
           6 x CO @ 2  = 12 e-
           charge       =  0
           ------------------
           total        = 18  -> saturated, stable
Nine orbitals hold eighteen electrons; chromium hexacarbonyl lands on the number exactly.

Where the rule is almost a law

The rule earns its keep in one well-defined neighbourhood: low-valent metals from the middle and late transition series, wrapped in strong pi-acceptor ligands such as carbon monoxide. Here it predicts formulas with almost uncanny reliability. Why this corner? Because pi-acceptor ligands do more than donate a lone pair — through pi back-donation they also drain electron density out of the metal's filled d orbitals into their own empty antibonding orbitals. That back-donation pushes the antibonding metal-ligand orbitals up high and stabilizes the filled ones, carving out a big energy gap. With a wide gap, dropping below eighteen wastes good bonding and rising above eighteen would force electrons into steeply antibonding orbitals, so eighteen becomes the sharp minimum.

Watch it work. The simple binary metal carbonyls read out their formulas straight from the count. Chromium is group 6, so it needs six pairs from CO to reach eighteen: Cr(CO)6. Iron is group 8 and needs five CO: Fe(CO)5. Nickel is group 10 and needs only four: Ni(CO)4. The odd-electron metals cannot reach eighteen as monomers — manganese, group 7, would land on seventeen — so they pair up, forming a metal-metal bond that lets each metal borrow one electron from its partner. That is exactly why manganese carbonyl exists as the dimer Mn2(CO)10, with each manganese counting to a tidy eighteen.

The same logic explains the most famous molecule in the field. Ferrocene, Fe(C5H5)2, sandwiches an iron atom between two flat five-carbon rings. Iron brings eight electrons (group 8) and each cyclopentadienyl ring, sitting face-on through all five carbons, donates five — eight plus five plus five is eighteen. The count predicts a remarkably stable, air-tolerant orange solid, and that is precisely what ferrocene is. The 18-electron rule does not just rationalize after the fact; here it would have told you the molecule should be happy before you ever made it.

The honest exceptions

Now the part too many textbooks rush past. The 18-electron rule has whole families of exceptions that are not mistakes or unstable oddities — they are robust, common, and predictable in their own right. The most important is the square-planar 16-electron complex. When a metal has a d8 configuration — rhodium(I), iridium(I), palladium(II), platinum(II), nickel(II) in the right setting — its four ligands often sit at the corners of a square in one plane rather than wrapping all the way around. In a square-planar field one d-derived orbital, the one pointing up and out of the plane, gets shoved high in energy alongside the empty p orbital perpendicular to the plane, so the molecule contentedly fills eight of the nine orbitals and stops at sixteen.

Far from being a flaw, that empty seat is the whole point. A 16-electron complex is coordinatively unsaturated: it has a vacant site ready to bind a substrate or to swallow a small molecule in an oxidative addition, jumping to eighteen and then dropping back to sixteen as a catalytic cycle turns over. This is not a curiosity at the edge of the subject; it is the beating heart of homogeneous catalysis. So when you carefully count a late-metal d8 complex and get sixteen, do not reach for the eraser — you have almost certainly found a square-planar species behaving exactly as it should.

The exceptions run the other way too. Early transition metals — titanium, vanadium, the front of each row — routinely fall short of eighteen. They simply do not have enough d electrons of their own, and they are large enough that they cannot always pack enough ligands around themselves to make up the difference; their bonding orbitals are also less stabilized, so the penalty for an empty orbital is mild. Complexes like Cp2TiCl2 sit comfortably at sixteen or even fewer electrons without any urge to grab more. Meanwhile, ligands that are mainly sigma donors with little pi-accepting ability, and very bulky ligands that physically block extra coordination, both let a metal rest below eighteen. The wide-gap, pi-acceptor situation is special, not universal.

Using it as a guide, not a law

The skill is not to memorize eighteen as a commandment but to read what a count is telling you. A complex right at eighteen is probably coordinatively saturated and unreactive toward adding more — to make it do something you must first knock a ligand off. A complex at sixteen, especially a late-metal d8, is likely poised to react: it has room and is hungry. A count that comes out wildly off, like eleven or twenty, is more often a sign you have miscounted — slipped a charge, mixed the two counting conventions, or misjudged a ligand's hapticity — than a real molecule that disobeys all chemistry.

  1. Pick one counting convention — neutral or ionic — and commit; never mix them in a single count.
  2. Add the metal's contribution (its group number in the neutral convention; its d-electron count in the ionic one) to the electrons donated by every ligand.
  3. Adjust for the overall charge of the complex, then read the total.
  4. Interpret, don't judge: 18 means saturated; 16 with a late-metal d8 means a primed catalyst; an early metal below 18 is normal — ask the geometry and the periodic-table position before you trust the magic number.