Refrozen Before It Thawed: What Partial Re-Freeze Cycles Do to Gel

A reusable pack comes back to the depot at 11:00 with a hard core and a slushy shell. It goes straight into the freezer, because the rule is that packs go back in the freezer. Nobody records that it never fully thawed. Over a season, packs on this route see a hundred partial cycles instead of the forty full ones the lab tested, and they fail earlier than the cycle count predicts — usually by going soft in the middle of a shipment rather than at the depot.

Partial cycles are the common case, not the exception

Full freeze-thaw testing is done cleanly: −18 °C for 24 hours, 25 °C for 24 hours, repeat. Real return loops look nothing like that. A pack used on a 12-hour chilled delivery may come back at −4 °C in the core and 6 °C at the edges. It is then refrozen, and the next outbound trip starts from a pack with an uneven crystal structure. The lab cycle counts one event; the route counts two half-events.

A more honest metric is thermal work: the total energy absorbed and released by the pack over its life. A pack on a short chilled route may absorb only 40 percent of its capacity per trip, so its 50-cycle lab rating corresponds to well over 100 trips — but only if the gel chemistry tolerates many shallow cycles. Some do; some degrade faster under shallow cycling than under deep cycling, because the ice never fully recrystallises and the polymer network is worked repeatedly in a partially frozen state.

What degrades first

Three things change, in a fairly consistent order.

Free water appears. The first sign is a thin film of liquid inside the pouch after the pack has warmed. Syneresis, the slow expulsion of water from the polymer network, accelerates each time the gel is frozen, because ice crystals grow and rupture the crosslinked structure. A new pack releases under 1 percent of its water after 24 hours at 25 °C; a pack at 30 cycles may release 4 to 6 percent; by 60 cycles, 10 percent is common. Once free water is present, the pack freezes into a solid block with a liquid layer at the bottom, and the block no longer conforms to carton walls.

Shape goes. A gel that used to stand at 60 mm thick slumps to 40 mm under its own weight, which reduces contact area and adds an air gap exactly where the cold path was designed.

Film fatigues. Corner welds take the most stress, because a frozen block flexes at the corners every time it is handled. A pack that passes burst testing when new can fail at cycle 25 from a pinhole at a corner weld that was never stressed in the original test.

Measuring it without a laboratory. Track three numbers on ten sample packs from a route, checked at cycles 0, 15, 30 and 45.

• Mass loss: weigh dry, thaw, drain for 60 seconds, weigh again. Anything above 5 percent means the network is failing.
• Thickness under load: place a 2 kg plate on a thawed pack for 10 minutes and measure. A drop of more than 15 percent from new means the pack will slump in a stacked box.
• Freeze time: how long from 20 °C to a hard core at −18 °C. A degraded pack with less free water and more bound water can freeze slower, which quietly reduces throughput.

None of these need instrumentation beyond a scale, a ruler and a timer, and they catch degradation before a customer does.

A topping-up routine that extends life

Packs that lose water can often be recovered rather than scrapped, if the film is intact and the loss is under 8 percent. Weigh the drained pack, calculate the water deficit, and add that volume of demineralised water through a resealable port or by opening and re-welding the fill seam where the design allows. Let it stand 24 hours at 25 °C to re-swell, then re-freeze. This restores mass and thickness but not the original crosslink density, so treated packs should be marked and routed to less demanding lanes rather than returned to the tightest temperature band.

Do not top up with tap water on a food-contact pack. Hardness and chloride load will pull performance down further and will void the documentation that supports the fill.

For programs where refreezing is frequent, the specification should state a minimum retained-water figure after 30 cycles, not just a fresh absorption ratio. Fills with higher crosslink density trade some initial capacity for that durability, which is usually the right trade on a returnable pack and the wrong one on a single-use chilled pack that will only ever freeze once. Comparative data is set out on the SAP powder specification pages.