Colligative properties: it's the count that matters, not the identity
Dissolve any non-volatile solute in a solvent and two things happen to the phase-change temperatures at once: the freezing point drops and the boiling point rises. Both are colligative properties — they depend on the number of dissolved particles per unit of solvent, not on what those particles chemically are. A mole of sugar and a mole of salt (before accounting for dissociation) depress the freezing point of water by comparable amounts per mole of particles, even though they are chemically nothing alike.
The two equations, and why they share a shape
Both effects reduce to the same underlying cause: adding solute lowers the chemical potential (loosely, the escaping tendency) of the solvent in the liquid phase, without directly affecting the solid or vapor phase. That shifts the equilibrium temperature between liquid and solid down, and between liquid and vapor up, by amounts proportional to the solute's molality:
DeltaT_f = i * Kf * m // freezing point depression DeltaT_b = i * Kb * m // boiling point elevation // m: molality of solute (mol solute / kg solvent) // Kf, Kb: cryoscopic and ebullioscopic constants, specific to the solvent // (for water: Kf = 1.86 degC*kg/mol, Kb = 0.512 degC*kg/mol) // i: van't Hoff factor — the number of particles each formula unit yields
The van't Hoff factor: dissociation changes the count
A non-dissociating solute like sugar (sucrose) has i approx 1 — one dissolved molecule, one particle. An ionic compound that dissociates changes the particle count directly: sodium chloride splits into Na+ and Cl-, so i approx 2; calcium chloride splits into one Ca2+ and two Cl-, so i approx 3. This is exactly why calcium chloride is a more effective road de-icer per mole than plain table salt — it is not a stronger interaction with water, it is simply more dissolved particles per formula unit doing the same colligative job. Real solutions show i slightly below the ideal integer value at higher concentrations because ion pairing keeps some fraction of the dissolved ions from acting as fully independent particles.
Freeze concentration: a moving target
Cooling a solution is not as simple as watching a single freezing point arrive and the whole thing solidify. As pure solvent crystallises out of solution (ice forming from salt water, for instance), the remaining liquid becomes more concentrated in solute, because the solute is largely excluded from the growing ice crystal lattice. A more concentrated remaining liquid has, by the same DeltaTf = i*Kf*m relationship, an even lower freezing point — so the freezing point of the shrinking liquid keeps dropping as freezing proceeds, rather than staying fixed. This freeze concentration effect continues until either all the liquid is gone or (for a solute with a eutectic system) the remaining liquid finally reaches the eutectic composition and solidifies completely at a single fixed temperature.
Freeze concentration is put to deliberate use in zone refining and cryoconcentration (used to concentrate fruit juice and to make eiswein/ice wine, where partially freezing the must and pressing off the still-liquid, sugar-concentrated fraction increases sweetness) and it is also the reason road salt works progressively worse as temperatures drop far below freezing — eventually the brine that forms is itself concentrated enough to be near its own eutectic point and stops melting any more ice effectively.
Why boiling point elevation gets less everyday attention
Boiling point elevation follows the identical mathematical form but is far less dramatic in ordinary life, because Kb for water (0.512 degC per molal) is roughly a third of Kf (1.86 degC per molal) — the same molality of solute raises the boiling point noticeably less than it lowers the freezing point. Adding salt to pasta water, a common kitchen claim, raises the boiling point by a small fraction of a degree at typical cooking concentrations, nowhere near enough to meaningfully speed up cooking; the practical reason to salt pasta water is flavour, not thermodynamics.
Frequently asked questions
Does it matter what solute is dissolved, or just how much?
For an ideal solution, colligative properties like freezing point depression depend only on the number of dissolved particles per unit of solvent, not on the solute's chemical identity — sugar and salt depress the freezing point comparably per mole of particles, once you account for the fact that salt dissociates into two ions and sugar does not.
Why is calcium chloride a more effective de-icer than table salt?
Calcium chloride dissociates into three ions per formula unit (one Ca2+ and two Cl-) versus sodium chloride's two (Na+ and Cl-), giving it a higher van't Hoff factor i. Since freezing point depression scales with i times molality, calcium chloride depresses the freezing point more per mole dissolved, not because it interacts with water any more strongly.
Why does salt water's freezing point keep dropping as it freezes, instead of freezing all at once at one temperature?
As ice crystallises out, it excludes most of the solute, so the remaining liquid becomes progressively more concentrated. A more concentrated solution has a lower freezing point by the same DeltaTf = i*Kf*m relationship, so the freezing point of the shrinking liquid keeps falling as freezing proceeds — this is the freeze concentration effect.
Try it live
Everything above runs in your browser — open Freezing Point Depression and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Freezing Point Depression simulation