Chocolate feels cool in the mouth because melting costs energy

A fat with an unusually sharp melting curve

Most natural fats are mixtures of many different triglycerides and therefore melt gradually across a wide span, softening long before they liquefy. Butter is like this, which is why it can be spreadable and solid at the same time. Cocoa butter is the exception. Its triglycerides are unusually uniform, dominated by a small number of closely related molecules, and the result is a fat that stays hard almost to the end and then melts across a range of only a few degrees.

That range finishes somewhere in the mid thirties Celsius, just under the temperature of the human mouth. So a bar is rigid at room temperature, unchanged in the hand for a while, and then fully liquid within seconds of being eaten. No common fat does this as cleanly, which is a large part of why chocolate is the confection it is and why substituting the fat is so noticeable.

The cooling is real, not a metaphor

Melting any solid requires energy, the latent heat of fusion, and that energy has to come from somewhere. In the mouth it comes from the tongue and the surrounding tissue, which lose heat to the melting fat. Cocoa butter has a comparatively high heat of fusion for a fat, and a mouthful of chocolate contains a lot of it, so the temperature drop is genuine and measurable rather than imagined.

Several other confectionery ingredients exploit the same effect more dramatically. Sugar alcohols such as xylitol and erythritol absorb a substantial amount of heat as they dissolve, which is why sugar-free mints can taste startlingly cold. In chocolate the effect is milder and mixes with everything else going on, but it contributes to the impression of freshness that a well-tempered bar gives and a badly tempered one doesn’t.

Temper decides how much of it you get

The stable crystal form melts higher and more sharply than the unstable ones, so a properly tempered bar holds its solidity until it is well into the mouth and then releases all at once. A poorly tempered bar contains a proportion of lower-melting crystals that have already softened at room temperature, so part of the melting has happened before you ate it and the thermal drama is muted.

This is why temper is perceptible even to someone who could not name what they are perceiving. Snap, gloss and cool clean melt are three symptoms of the same underlying crystal population, and they arrive together or not at all.

It has a further consequence for anything a chocolatier makes. A shell moulded from tempered chocolate contracts very slightly as those crystals form, which is what releases it from the mould, and an untempered shell simply sticks. The property that produces the sensation in the mouth is the same one that makes the manufacture possible, which is a tidy piece of physics.

Flavour release is timed by the melt

Aroma compounds in chocolate are largely fat-soluble, which means they are held in the cocoa butter and released to the nose only as it liquefies. A fat that melts sharply delivers its aroma in a rush; one that melts slowly dribbles it out over a longer period and against a background of residual solid. The first reads as intense and the second as flat, even where the underlying compounds are identical.

This is also why chocolate eaten straight from a refrigerator tastes of very little. The bar is far below its melting range, it takes a long time to get there, and by the time the aroma is released the first impression has already been formed. Twenty minutes on a bench before eating changes the experience more than most people expect, and it costs nothing.

The narrow range is also the difficulty

Everything convenient about cocoa butter follows from the same property that makes it awkward. A fat that goes from solid to liquid across three or four degrees is a fat with almost no working margin, which is why tempering demands the precision it does, why a warm room defeats a display, and why chocolate shipped in summer needs either refrigeration or a reformulation.

Confectioners in hot climates have dealt with this for a long time by blending in higher-melting fats or by accepting that some products simply belong to the cooler months. Neither solution is free. What you gain in stability you lose in exactly the clean, cool, complete melt that made the material desirable to begin with, and that trade has no clever way around it.