Crystal Dissolution: How Solids Give Way to a Solvent

Dissolution is a race between detachment and diffusion

A crystal dissolving in a solvent is governed by two sequential steps, and whichever is slower sets the overall rate. First, an ion or molecule at the crystal surface must detach — break free of the lattice bonds holding it in place, a process controlled by surface chemistry and temperature. Second, once free, it must diffuse away from the surface through a thin boundary layer of solvent into the bulk solution, a process controlled by the diffusion coefficient and how vigorously the solution is stirred. The classic Noyes-Whitney equation captures the diffusion-limited case:

dm/dt = (D · A / L) · (C_s - C)
D = diffusion coefficient,  A = surface area,  L = boundary-layer thickness
C_s = saturation (solubility) concentration,  C = bulk concentration

Why corners and edges dissolve fastest

Not every surface ion is equally exposed. An ion sitting on a flat face is bonded to many lattice neighbours below and beside it; an ion at an edge has fewer neighbours; an ion at a corner or a kink in a step has fewer still. Fewer neighbouring bonds means a lower activation energy to break free, so corners and edges detach first and fastest — which is exactly why dissolving crystals round off into smoother, more rounded shapes over time rather than shrinking as a scaled-down copy of their original faceted form. This is the mirror image of crystal growth, where kink sites are also where new ions attach most readily.

The saturated boundary layer

As ions leave the crystal surface faster than they can diffuse away, a thin layer of nearly saturated solution builds up right against the crystal — the concentration gradient driving diffusion is steepest at the surface and flattens out toward the bulk. This boundary layer, typically tens of micrometres thick in still solution, is why stirring dramatically speeds up dissolution: agitation physically thins the layer L, steepening the gradient (C_s − C)/L without changing the solubility itself. A sugar cube dissolves far faster stirred than sitting still, even though the water's total capacity to hold dissolved sugar (C_s) has not changed at all.

Temperature, solubility and the exception of retrograde solubility

Temperature enters the Noyes-Whitney picture twice: it raises the diffusion coefficient D (faster random thermal motion moves ions away quicker) and, for most crystalline solids, it raises the saturation concentration C_s itself, widening the driving gradient. Both effects usually point the same way — hotter solvent dissolves solids faster — which is why the standard advice for dissolving salt or sugar is warm, stirred water. A handful of compounds, like calcium sulfate and some cellulose ethers, show retrograde solubility: C_s actually decreases with temperature, so heating slows or even reverses their dissolution, a detail that matters directly in scale formation inside boilers and pipes.

When detachment, not diffusion, is the bottleneck

The Noyes-Whitney picture assumes diffusion is the slow step, which holds for many salts in water. But for crystals with strong lattice bonding or a passivating surface layer (some silicates, many minerals under near-neutral pH), the surface reaction of breaking a lattice bond is slower than diffusion could ever be, and stirring barely helps — the rate becomes essentially independent of boundary-layer thickness and instead follows an Arrhenius-type law in temperature, dm/dt ∝ exp(−E_a/RT), governed by the activation energy E_a of detachment rather than by D. Distinguishing the two regimes experimentally is usually done by testing whether stirring rate changes the dissolution rate: if it does, diffusion is limiting; if it barely matters, surface reaction is.

Why this matters beyond the lab bench

The same physics governs pharmaceutical tablet dissolution (where the Noyes-Whitney equation is used directly to design how fast a drug releases into the gut), mineral weathering and karst cave formation (limestone dissolving in slightly acidic groundwater over millennia), ocean acidification's effect on carbonate shells, and the descaling of kettles and pipes. In every case the same two levers control the rate: increase the concentration gradient (undersaturate the solvent, or increase surface area by crushing) or thin the boundary layer (stir, flow, or otherwise move fluid past the surface faster).

Frequently asked questions

Try it live

Everything above runs in your browser — open Crystal Dissolution and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.