Two questions that are easy to mix up
By now you can build a complex on paper from the earlier rungs: a metal ion surrounded by ligands, its d orbitals split into the t2g and eg sets, its spin state and colour decided by the size of delta-o. Now we ask what such a complex actually does when you leave it in a beaker. Two very different questions hide behind the single word "stable", and almost every beginner's confusion in this corner of chemistry comes from running them together.
The first question is how far: if you mix the metal and the ligand, does the formation reaction run nearly to completion, or does it barely get going? That is a question about thermodynamic stability — about the position of equilibrium, the bottom of the energy valley. The second question is how fast: once the complex exists, how quickly does it trade one ligand for another from the solution? That is a question about kinetic lability — about the speed of the swap, the height of the energy hill in the way. The crucial, almost startling fact of this guide is that the answers are independent: knowing one tells you essentially nothing about the other.
Stability, measured: the formation constant
Thermodynamic stability has a clean number attached to it. When a ligand displaces water from an aqua ion, it does so one step at a time, and each step has its own equilibrium constant — a stepwise stability constant K1, K2, K3 and so on. Multiply all the steps together and you get the overall formation constant, usually written beta-n. A large beta means the formation reaction lies far to the right: at equilibrium, almost all the metal is locked up in the complex. A small beta means the complex barely forms. This number is pure thermodynamics — it reports the depth of the valley and says nothing about how fast you reach the bottom.
Concretely, the first step is [M(H2O)6] + L giving [M(H2O)5 L] + H2O with constant K1, the second swaps another water for K2, and so on; the overall reaction [M(H2O)6] + 6 L giving [M L6] + 6 H2O has beta-6 = K1 x K2 x K3 x K4 x K5 x K6. A large beta-6 marks a deep valley and a thermodynamically stable complex; a small one marks a shallow valley. But beta tells you only how far, never how fast.
Two honest footnotes. First, "stable" is always relative to something — a complex is stable with respect to a particular set of products, usually the free aqua ion plus loose ligand; the same complex could be unstable toward, say, decomposing into a metal oxide. Always ask: stable against what? Second, the steps usually get weaker as you go, so K1 > K2 > K3, simply because there is less room and fewer waters left to replace; the chelate ring you met earlier is the famous exception where a single multi-armed ligand makes beta enormous.
Lability, measured: the speed of the swap
Lability is the other axis entirely. Henry Taube drew the practical line: a complex is labile if it exchanges its ligands fast — within about a minute at room temperature and ordinary concentrations — and inert if the exchange takes much longer, hours or longer. The words describe rate, nothing else. Crucially, calling a complex inert is not a moral judgement on its bonds being strong; it only means the path to swapping a ligand runs over a tall activation barrier. The complex may be sitting in a shallow valley (thermodynamically unstable) and still be walled in by a high ridge (kinetically inert).
What, then, makes a complex inert? For the octahedral d-block ions the answer leans heavily on the d-electron arrangement you already understand. Substitution almost always means a ligand has to leave or arrive through the faces of the octahedron, and that motion is blocked when the t2g orbitals — the ones pointing into the gaps between ligands — are full, and made worse when the higher eg orbitals are empty. So the most stubbornly inert ions are precisely those with a filled-t2g, empty-eg pattern: low-spin d6 like Co3+ and Pt4+ (t2g^6 eg^0), and d3 like Cr3+ (t2g^3 eg^0). In rough crystal-field language, these configurations also carry a large crystal field stabilization energy that the transition state would have to give up, which raises the barrier.
At the other end, ions whose d-electron count leaves eg partly filled — high-spin d4 through d7, and d9 like Cu2+, and the spherical d0 and d10 cases — have those antibonding eg orbitals occupied or low CFSE, so distorting toward the transition state is cheap and they are usually labile. This is a model, not a law: it explains the broad pattern of which first-row ions are quick and which are sluggish, but it is built on the same point-charge crystal-field picture that is only an approximation, and real rates also depend on charge, size, and the incoming ligand. Use it to predict trends, not exact half-lives.
All four corners are real
Because the two axes are independent, every combination actually occurs in nature — and the two surprising corners are where the lesson lives. Stable but labile: [Ni(CN)4]2- has an enormous formation constant (beta around 10^30, a very deep valley), yet if you add carbon-13 labelled cyanide it scrambles into the complex almost instantly. The complex is thermodynamically rock-solid and kinetically frantic at the same time. Unstable but inert: the classic case is [Co(NH3)6]3+ in acid. The ammonia complex is thermodynamically unstable in acidic water — the equilibrium hugely favours [Co(H2O)6]3+ plus ammonium — so it ought to fall apart, yet it can sit in dilute acid for days because the low-spin d6 cobalt(III) centre is so inert that the barrier to releasing ammonia is enormous.
LABILE (fast swap) INERT (slow swap)
+-------------------------+------------------------+
STABLE | [Ni(CN)4]2- | [Co(CN)6]3- |
(deep | beta ~ 10^30 yet CN- | deep valley AND |
valley) | exchanges in seconds | walled in (d6 LS) |
+-------------------------+------------------------+
UNSTABLE | [Ni(H2O)6]2+ | [Co(NH3)6]3+ in acid |
(shallow | forms & breaks fast | WANTS to decompose, |
valley) | (ordinary aqua ion) | but barrier too high |
+-------------------------+------------------------+
Thermodynamics (rows) and kinetics (columns) vary independently.The other two corners are the ones that lull people into the wrong instinct. Stable and inert is [Co(CN)6]3-: deep valley and high walls together, the picture most people wrongly assume "stable" always means. Unstable and labile is plain [Ni(H2O)6]2+: a shallow valley with low walls, forming and breaking apart in a blur. Because these two reinforce the lazy idea that strong means slow, the off-diagonal cases above are the ones worth burning into memory.
Why the confusion costs you, and where this leads
Why insist so hard on the split? Because the most famous facts about complexes only make sense once you keep the axes apart. Alfred Werner could deduce the structures of cobalt(III) ammines — count the isomers, prove the octahedron — precisely because those complexes are kinetically inert: they hold their shape long enough to be crystallised, separated, and studied. A labile complex would have scrambled before he could look. Inertness was the silent gift that made classical coordination chemistry possible at all.
- When you read "stable", pause and decide which axis is meant: a deep valley (large beta, thermodynamics) or a tall barrier (slow exchange, kinetics). They are not the same word twice.
- If the question is about equilibrium amounts, reach for the formation constant beta; if it is about how fast a reaction happens, reach for labile-versus-inert and the activation barrier.
- To guess lability for an octahedral d-block ion, look at the d-electron count: filled-t2g, empty-eg patterns (low-spin d6, d3) tend to be inert; partly-filled-eg patterns tend to be labile.
With the two axes firmly separated, the rest of this rung becomes a study of just the kinetic axis: how, exactly, does a ligand leave or arrive? Does the old ligand depart first and leave a gap (a dissociative path), or does the new one push in before the old one goes (an associative path)? That mechanistic detail, the trans effect that steers which ligand in cisplatin is replaced, and the two great mechanisms of electron transfer are where we head next — all of them stories about the height of the ridge, never the depth of the valley.