Every winter, we spread salt on icy roads and the ice melts — even when the temperature is below 0 °C, ice's normal melting point. How can a little salt melt frozen water without any heat source? The answer is one of the most elegant in solution chemistry: it's freezing-point depression.
The equilibrium at 0 °C
At 0 °C, ice and liquid water coexist in equilibrium: at every instant, as many water molecules leave the ice for the liquid (melting) as the reverse (freezing). The two rates are exactly equal — which is what defines the melting point.
What happens if we dissolve salt in the liquid water? The Na⁺ and Cl⁻ ions take the place of some of the water molecules at the contact surface. As a result, fewer liquid water molecules are available to re-attach to the ice. The freezing rate drops, while the melting rate is unchanged (the ice itself is pure solid, salt-free). The balance tips toward melting: the ice melts. To restore equilibrium, we must cool further — hence a new, lower freezing point.
Seen another way, in the language of thermodynamics: dissolving a solute increases the liquid's disorder (entropy), which stabilizes it. Liquid water "prefers" to stay liquid down to a lower temperature before it sets.
A colligative property
The remarkable point: this effect depends not on the nature of the solute, only on the number of dissolved particles. This is called a colligative property (from Latin colligare, to bind together). The law reads:
ΔT_f = i · K_f · m
where m is the molality (moles of solute per kg of water), K_f the cryoscopic constant of water (1.86 °C·kg/mol), and i the van't Hoff factor — the number of particles released per dissolved unit.
This is where salt is clever: sodium chloride NaCl dissociates into two ions (Na⁺ + Cl⁻), so i ≈ 2. One mole of salt lowers the freezing point twice as much as one mole of sugar (which doesn't dissociate, i = 1). Calcium chloride CaCl₂ does even better: it releases three ions (Ca²⁺ + 2 Cl⁻), i ≈ 3.
Why CaCl₂ on the coldest roads
Salt has a limit. By saturating water with NaCl, we can only go down to about −21 °C (the eutectic point, reached at 23 % salt by mass). Below that, salt no longer works. That's why road crews switch to calcium chloride (effective down to ~−50 °C) or magnesium chloride in the harshest regions. Bonus: dissolving CaCl₂ is exothermic — it releases heat, further speeding up de-icing.
Note: salt needs a thin film of liquid water to start working (it has to dissolve somewhere). On perfectly dry, very cold ice, the process is slow to begin.
The same physics, elsewhere
Freezing-point depression is just one of the colligative properties. Its cousins: - boiling-point elevation: salt water boils above 100 °C (why we salt pasta water… though the effect is tiny); - osmotic pressure: it governs water exchange across cell membranes; - car antifreeze: ethylene glycol lowers the coolant's freezing point and raises its boiling point.
Why it matters
The story of salt on ice shows that a property can depend only on a count — how many particles, regardless of which. That's counterintuitive and deeply useful: it's how we measure molar masses (by cryoscopy), design antifreezes, and understand why cells swell or shrink in water. An everyday phenomenon that opens onto the whole thermodynamics of solutions.