Dry Ice Alternatives for Frozen Parcel Shipping
The conversation usually happens late. A frozen range is ready, the boxes are specified, the freezer is full, and then somebody reads the carrier’s prohibited items list properly and finds dry ice on it. At that point the packaging is not the problem. The route to the customer is.
We see this pattern often enough at Hydropac that it is worth setting out properly, because the usual explanations are wrong in both directions. Dry ice is not banned. It is also not something most UK businesses can put into a parcel network, and the reason has very little to do with how cold it is.
Solid carbon dioxide is UN1845, a Class 9 dangerous good. It sublimes rather than melts, which means the coolant you pack is not the coolant that arrives. And it releases a heavier-than-air asphyxiant gas into vans, cold rooms and packing areas, which makes it a Control of Substances Hazardous to Health question before it is ever a packaging question.
Three different rule sets apply depending on how the parcel travels, and a fourth constraint sits on top of all of them. This piece works through each in turn, then through what actually replaces dry ice in a frozen pack-out, with the arithmetic that decides how much coolant you need.
Hydropac offers every customer a customized solution for chilled and conditioned shipping. For example, we help a customer with limited freezing capacity to deliver gel packs frozen and ready to use, and we can manufacture almost all shapes and sizes of cooling elements. As a customer, you come first: we are here to help you.
Three Sets of Rules, and the One That Usually Decides It
Dry ice is regulated very differently by road, by air and by sea. Most of the confusion we encounter comes from businesses applying air freight requirements to a domestic road movement, or assuming the light-touch road position carries over to an export parcel. It is worth separating them cleanly, because the answer changes the pack design, the paperwork and the training obligation.
On the road, dry ice sits almost entirely outside ADR
Great Britain gives effect to ADR through the Carriage of Dangerous Goods and Use of Transportable Pressure Equipment Regulations 2009, as amended. Northern Ireland has its own equivalent regulations.
ADR treats UN1845 unusually. The note to section 5.5.3 states that for UN1845 the conditions in that section apply to all kinds of carriage, whether as a coolant, as a conditioner or as a consignment in its own right, and that no other provisions of ADR apply. Dry ice by road is therefore governed by one short section rather than by the regulations as a whole.
In practice that means no orange plates, no ADR driver vocational training certificate and no requirement to appoint a dangerous goods safety adviser on the strength of the dry ice. It does not mean no training. General duties under health and safety and COSHH law still apply, and anyone loading, driving or unloading needs to understand the asphyxiation risk.
What 5.5.3 does require is that packages are marked CARBON DIOXIDE, SOLID, AS COOLANT, that the transport document carries the same wording alongside UN 1845 and states the total number of packages, and that the vehicle or container itself is marked where there is a real risk of asphyxiation inside it and it is not adequately ventilated. That last condition is a risk assessment rather than a label to be ticked off, and it is the one we see skipped most often.
In the air, everything becomes explicit
Air transport works the other way round. Under IATA, dry ice is handled through Packing Instruction 954 with a limit of 200 kg per package. The package must carry a Class 9 label, the UN number, the proper shipping name as either Carbon dioxide, solid or Dry ice, and the net weight of dry ice in kilograms. The packaging has to be designed to vent, because a sealed inner carton containing subliming dry ice is a pressure vessel.
The air waybill then has to reflect it, showing the UN number, the shipping name, the number of packages and the net weight of dry ice in kilograms. The one concession is that a full Shipper’s Declaration for Dangerous Goods is not required where dry ice is the only dangerous good in the consignment. If the dry ice is refrigerating other dangerous goods, that concession disappears and a declaration is required for those goods, with the dry ice shown as the refrigerant.
The carbon dioxide itself is a workplace duty
This is the part that sits outside the packaging specification and catches operations teams out. Carbon dioxide is classed as a substance hazardous to health under COSHH, and the workplace exposure limits published in EH40 are 5,000 ppm as an eight hour time weighted average and 15,000 ppm as a fifteen minute short term limit.
Because the gas is colourless, odourless and denser than air, it accumulates at low level in exactly the spaces cold-chain work happens in: vans, cold rooms, lifts, stairwells, pallet wells and enclosed packing areas. HSE records fatalities where carbon dioxide has entered a confined space and displaced oxygen.
British Compressed Gases Association guidance sets out what rising concentrations actually do. At around 1 to 1.5 per cent there is a slight increase in breathing rate. At 3 per cent breathing becomes deeper and faster, with reduced hearing and headache. At 4 to 5 per cent breathing rate and heart rate rise further, with a slight choking sensation. Between 5 and 10 per cent there is a pungent odour, laboured breathing and exhaustion. Above 10 per cent, unconsciousness can follow in under a minute, and then death.
Contact causes severe cold burns, which is why the same guidance is explicit that dry ice should never be handled with bare hands. It should be stored somewhere well ventilated and preferably not below ground, never in a gas tight container, and never placed in a working refrigerator or freezer. Vehicle heating and ventilation should be set to draw fresh air from outside rather than recirculate.
No statute requires fixed carbon dioxide monitoring to be fitted. Whether it is needed is decided by the COSHH risk assessment, and where that assessment identifies a real risk, monitoring can become a legal requirement rather than a precaution. For any operation handling meaningful daily volumes in enclosed spaces, that is the usual conclusion.
Then the carrier contract overrules all of it
Having worked through the regulations, most businesses discover they were never the binding constraint. Carriers set their own acceptance policies, and those policies are considerably stricter than the law requires.
Royal Mail is the clearest case: frozen water and dry ice are prohibited under its perishables rules, inland and international. Elsewhere the position varies by carrier and by destination rather than being a blanket ban, so it has to be checked service by service. Some volume carriers prohibit dry ice on international services while treating chilled and frozen food as excluded from compensation on domestic ones, which is a commercial exclusion rather than a refusal. The integrator express networks will carry it as declared dangerous goods, but normally only where the account has been approved and audited for dangerous goods first.
For a direct-to-consumer frozen brand the practical position is therefore simple, even though the regulatory position is not. Once the route to the customer runs through a mainstream parcel network, dry ice stops being a packaging decision and becomes a reason to rebuild the pack-out around a coolant the network will accept without conditions.
Rebuilding a Frozen Pack-Out Without It
Removing dry ice is not a coolant substitution. It is a change in how the system holds temperature, and it needs to be treated as a redesign rather than a swap. The good news is that the redesign is arithmetic rather than guesswork, and the arithmetic is more reliable than it was with dry ice in the box.
Why dry ice was never a predictable coolant in the first place
A conventional coolant absorbs heat by changing phase and then stays in the parcel. Dry ice absorbs heat by turning directly to gas and leaving. That difference is the reason hold times quoted for dry ice systems are so much softer than they appear.
FAA testing of small expanded polystyrene shippers, run under a simulated flight profile, measured sublimation at around 2.0 per cent per hour across a range of roughly 1.6 to 2.9 per cent per hour. In a parcel-sized box that is close to half the coolant mass gone in twenty four hours. The rate is highly dependent on the system: later testing of much larger containers measured well under one per cent per hour, and unwrapped blocks in open air lose dramatically faster.
The variables driving that spread are insulation type, wall thickness, box volume, fill ratio, headspace, ambient temperature, altitude and handling. None of them are stable across a real courier network, which is why a single published hold time for a dry ice pack-out is a difficult promise to keep. A coolant that melts rather than sublimes holds a defined mass through the journey and releases its latent heat at a known temperature, which is what makes a stated hold time something you can specify against.
The three levers that replace it
Where dry ice supplied brute force, a compliant frozen pack-out supplies three smaller adjustments working together.
The first is the phase change temperature. A frozen payload has to stay below freezing, so the coolant has to change phase below the product temperature rather than simply start out cold. A water based pack stabilises around 0 degrees C as it melts, which is exactly right for chilled distribution and not sufficient to hold a frozen product. Our HydroFreeze Technology is a phase change formulation developed for the frozen band for that reason, which is what allows Frozen Ice Packs to maintain frozen conditions inside a suitable insulated container rather than merely slow the thaw. The phase change point and the conditioning temperature it needs are both stated on the product pages, and both matter.
The second is coolant mass, which governs duration rather than temperature. Once the phase change point is correct for the payload, quantity follows from payload mass, route duration and the ambient profile the parcel will actually meet, which is why packs need to be sized against the lane rather than against what fits conveniently in the carton. Unlike dry ice, a phase change pack does not sublime away, so the mass loaded at despatch is still in the box on day two and you only buy the quantity the route requires.
The third is the insulation envelope, and it has to work harder than it did before. With the coolant no longer sitting eighty degrees below the payload, wall thickness, closure method and void fill stop being incidental. A well specified Frozen Shipping Box or a correctly matched Polystyrene Box becomes part of the thermal calculation rather than a container to put the calculation inside.
There is one more difference that matters more than any of the three, given where this article started. HydroFreeze is not classified as a dangerous good, so a frozen pack-out built around it needs no HAZMAT labelling and no specialist courier. That is the practical reason a frozen range can move onto a mainstream parcel network at all.
Over-specifying dry ice damages the product
When a pack-out is uncertain, the instinct is to add more coolant. With dry ice that instinct is actively harmful.
At roughly minus 78.5 degrees C, dry ice sits some 80 to 86 degrees below the 2 to 8 degrees C chilled band. Anything chilled rather than frozen that ends up in contact with it, or in the cold plume directly beneath it, is not being protected. It is being frozen, and for fresh produce, dairy, chocolate, certain beverages and most chilled ready meals that is product damage rather than product protection.
Mixed chilled and frozen parcels are where this bites hardest, because the same box has to hold two temperature bands apart. That is a pack-out design question rather than a coolant quantity question, and it is the sort of thing Cold Chain Optimisation is meant to resolve before it becomes a customer complaint.
The freezer is usually the real bottleneck
There is a practical consequence that rarely appears in guidance written by people who do not manufacture coolant. Moving from dry ice to a phase change coolant shifts the constraint from the supplier to your own freezing capacity, and the set point matters as much as the volume. A frozen phase change pack has to be conditioned in a freezer cold enough to take it fully through its phase change, and the time it needs in there depends on how cold that freezer actually runs. A pack that has been in a domestic-grade freezer overnight has not been conditioned, whatever it feels like in the hand.
This is a solvable problem, and it is one of the reasons we manufacture rather than simply distribute. Where a customer’s freezing capacity is the limiting factor, we can supply packs already frozen and ready to use, so despatch is not gated by blast freezer throughput. Where the constraint is space or handling instead, freezer tray formats and pack geometry can be adjusted to the racking that already exists. Both are conversations we would rather have before a range launches than after the first weekend of complaints.
How we would approach the conversion
Our Cold Chain Optimisation work follows the same sequence whether the trigger is a carrier refusal, a cost review or a compliance question. Assessment establishes what the current pack-out is actually doing and where the risk and cost sit. Analysis applies thermal testing, compliance review and operational mapping against the routes the parcels really take. Optimisation refines the pack-out, the coolant quantity and the format. Validation and implementation then documents the result and moves it into scalable manufacturing.
The reason that sequence matters for a dry ice conversion specifically is that the failure mode is quiet. A pack-out that is slightly under-specified does not fail on the first parcel. It fails on the one that misses a Thursday collection before a bank holiday, and by then the product is already with the customer. Testing against the worst realistic route rather than the average one is the whole point of the exercise, and it is why we would rather walk a customer through production and sealing at High Wycombe than send a specification sheet.
Where dry ice is still the right answer
For deep frozen pharmaceutical logistics at minus 70 degrees C and below, for dedicated haulage under an owned fleet, and for specialist air freight with declared dangerous goods handling already in place, dry ice remains the standard and most widely available option. It is not the only one at that temperature, but it is the one most of the supporting infrastructure is built around.
What has changed is the population of businesses using it. A great deal of UK frozen e-commerce inherited dry ice from an earlier stage of the supply chain, where the parcel went by dedicated vehicle and the operation had trained handlers. Once the same product moves to a consumer parcel network, the inheritance stops working, and the honest answer is not a workaround. It is a different pack-out, sized properly, with the coolant quantity proven against the route it will actually travel.
If that is the conversation you are currently having with a carrier, it is worth having with a manufacturer at the same time. Courier Safe Frozen Packaging is a category that exists because this problem is common, and Long Lasting Ice Packs are specified in hours precisely because the alternative cannot be.