Commercial ice cream is frozen twice and the second time is the point

One job at home, two jobs in a factory

A domestic ice cream machine chills and churns until the mixture is as firm as it is going to get, and then it goes into a container in the freezer to finish. The industrial process looks superficially similar and is organised quite differently. A continuous freezer takes the mix down to a soft, aerated state and no further, the product is filled and packed while it is still soft, and only then does a separate machine finish the freezing.

The division matters because the two stages are optimising different things. The first is about beating in air and creating a large number of very small ice crystals under intense mechanical agitation. The second is about freezing the remaining water as quickly as possible so that it forms new small crystals rather than adding to the ones already there. Doing both in one slow step, which is what a home freezer amounts to, gets neither right.

The continuous freezer draws it soft on purpose

In a continuous freezer the mix is pumped through a chilled cylinder while a rotating dasher with scraper blades removes the frozen layer from the wall as fast as it forms. Air is injected under pressure at a metered rate rather than being whipped in, which is why overrun can be specified precisely instead of estimated. The product leaves at a temperature around minus five or six degrees, soft enough to pump through a filling nozzle.

At that point only part of the water is frozen. Roughly half of it, depending on the formulation, is still liquid in a concentrated sugar solution surrounding the crystals, and it is that fraction the second stage has to deal with. Drawing colder would give more crystals immediately but the product could no longer be filled, ripple-fed or extruded, so the draw temperature is a compromise dictated by the packaging line as much as by the texture.

Hardening is a race against crystal growth

The filled containers go into a hardening tunnel — a blast freezer with air moving fast at a temperature far below anything domestic, commonly in the region of minus thirty or lower. Speed is the entire objective. Water that freezes quickly forms many small crystals, because there isn’t enough time at any one temperature for existing crystals to grow at the expense of their neighbours. Water that freezes slowly forms fewer and larger ones, and large crystals are what the tongue detects as coarseness.

The moving air is as important as the temperature. Heat has to leave through the container wall and then through the product itself, and ice cream is a poor conductor, which is why a small tub hardens well and a large one hardens unevenly. Fast-moving cold air removes heat from the surface far more efficiently than still air at the same temperature, which is the same reason a windy day feels colder than a calm one.

A home freezer cannot do the second stage

A domestic freezer holds around minus eighteen, has almost no air movement, and is usually holding other things that also need cooling. A container of soft ice cream put into it takes hours to reach that temperature all the way through, and every one of those hours is time in which crystals grow. The result is coarser than the same mixture would be if it had been hardened properly, and no amount of care with the base changes that.

What a domestic kitchen can do is reduce the distance the heat has to travel. A shallow, wide metal container hardens far faster than a deep plastic tub, because metal conducts and depth is the enemy. Pre-chilling the container helps. So does turning the freezer to its coldest setting a few hours in advance, and not putting warm items in alongside. These are marginal gains rather than a substitute, but they are real and they compound.

Scaling runs into the same wall in both directions

This is a specific instance of a general problem with scaling any frozen preparation. Heat removal depends on surface area and the distance to the centre, while the amount of heat to be removed depends on volume, and the two don’t grow together. A larger batch is not simply a longer wait; it is a slower freeze throughout, which changes the product rather than merely the schedule.

It also explains why small artisan producers who buy a batch freezer and a proper blast cabinet get closer to industrial smoothness than a home cook with a good recipe ever will, and why the same producers so often say that the freezer capacity was the purchase that changed their product. The recipe is available to everybody. The rate of heat removal is equipment, and it is not really negotiable.