Part of our series on mason jars wholesale — read the overview first if the format is new to you.

A glass jar is freezer safe when three things are true together: the body is straight-sided so ice can grow upward, the fill leaves enough headspace for water's ~9% expansion, and the glass is annealed well enough to ride out the temperature swing. For food brands holding product at -18°C (0°F), the jars we specify pass a 42°C thermal shock differential (ASTM C149 equivalent) and are filled only after the product has cooled below 30°C. Miss any one of the three and the usual result is a base that separates in a clean ring overnight.

Which products and processes suit glass in the freezer

Frozen jams and preserves, bone broth and stock, baby food and purees, and meal-prep lines moving through a cold chain are the natural users: products that sit at -18°C for weeks to months in a static freezer.

Three situations are a poor match for soda-lime jars, and it is better to know before sampling:

  • Deep freeze below -25°C with watery liquids in narrow-neck bottles, because the neck traps the expanding ice.
  • Blast-freezing filled jars. Stress in the wall rises with the cooling rate, and industrial blast rates go beyond what a soda-lime jar tolerates.
  • Lines that cannot hold a fill level. Headspace is physics, so a filler that wanders will produce breakage however good the glass is.

Jar geometry and the straight-sided range

Ice forms from the walls inward and from the top down. In a straight-sided jar the liquid core that freezes last pushes up into the empty space under the lid. Add a shoulder — a classic mason profile, a wine bottle, anything with a neck — and that shoulder freezes solid first, sealing the core underneath it. The pressure then leaves through the shoulder or the base.

That is why every freezer model below is straight-sided, or tapered with a draft that lets the frozen block slide out. A wide-mouth mason with very little shoulder taper can still work for low-expansion products if headspace is strict; a regular-mouth shouldered mason cannot. Our mason jar size and specification guide shows which profiles fall on which side, and the wide-mouth mason range covers the minimal-shoulder options. A straight wall has a practical bonus too: broth pucks and portioned purees release as a clean cylinder.

ModelCapacityFinishProfileJar weightPcs per 40HQTypical freezer use
FZ-4 portion4oz (120ml)58-400 CTStraight-sided~130g~150,000Baby food portions, samples
FZ-8 straight8oz (240ml)70-450 CTStraight-sided~210g~92,000Jam portions, baby food
FZ-12 straight12oz (355ml)70-450 CTStraight-sided~265g~70,000Sauces, purees
FZ-16 straight16oz (473ml)86-400 wide CTStraight-sided~340g~50,000Broth, soup, family-size jam
FZ-16 tapered16oz (473ml)86-400 wide CTTapered, freezer-safe draft~330g~52,000Broth pucks; the taper gives the quickest release
FZ-24 straight24oz (710ml)86-400 wide CTStraight-sided~460g~36,000Bulk broth, meal prep
FZ-32 straight32oz (946ml)86-400 wide CTStraight-sided~560g~27,000Foodservice stock, bulk puree
FZ-64 bulk64oz (1.9L)110-400 CTStraight-sided~980g~14,000Foodservice stock concentrate

Headspace by product type

Expansion needs both room and a direction, and headspace supplies the two. How much depends on how much free water the product carries.

  • 1 inch (25mm) minimum for broths, soups, thin purees and jams above 60% moisture.
  • 3/4 inch (19mm) for thick, low-free-water fills such as nut butters, dense fruit butters and high-sugar preserves. Sugar lowers the freezing point, so a good share of the water never freezes solid at -18°C.
  • 1.5 inches for fermented or still-gassing products, although those are better kept out of frozen glass altogether.

The rule is only as good as the filler. A line that swings ±5mm around a 1-inch target will send out jars with 20mm of headspace, which is failure territory for broth. Level-sensing or vacuum-assist fillers hold ±2mm, and the headspace figure should be set with that tolerance in mind.

When headspace is short, the ice core loads the junction between base and wall, where stress concentrates, and the base comes away as a ring. It tends to happen during the overnight freeze-down and get found at morning QC. A ring-shaped base break therefore points to headspace or shoulder geometry rather than to the glass itself.

Temperature limits at freeze-down and thaw

Thermal shock breakage occurs when the temperature difference across the wall creates more stress than the glass can carry. A frozen product meets that risk twice.

Going into the freezer

Broth hot-filled at 85°C and sent directly to a -18°C freezer faces a differential of 100°C+, which no soda-lime jar survives reliably; brands that skip the cooling step report 5–15% base-ring failures. Stage filled jars at ambient or run them through a spiral cooler for 2–4 hours until they are under 30°C. That keeps the working differential inside 50°C. Jam cooled to 20°C before freezing is already within the limit.

The same sequence allows a dual-use SKU. Jam hot-filled at 88°C, into jars pre-warmed to the usual 40–45°C, can be sold ambient or frozen later, as long as it cools below 30°C between the two steps. For fill preparation, see our notes on sterilising mason jars before filling.

Empty jars deserve attention as well. Keep them above 5°C before filling, because a jar that has soaked in an unheated warehouse starts with less thermal margin.

Coming out of the freezer

Consumers break more jars than fillers do. A jar taken from -18°C and dropped into 60°C water sees an 80°C differential. Print the thaw method on the label — refrigerator thaw, or cool water first — and add a clear warning against microwaving from frozen. We can supply suggested handling wording with a freezer-jar project.

Annealed, tempered or borosilicate glass

Correctly annealed soda-lime glass copes with disciplined freezing, and nearly all commercial frozen products in glass use it. The freezer jars we specify are annealed to pass the 42°C differential, and annealing is checked by polariscope to stress grade ≤3 at each sampling round.

Tempered glass tolerates a wider swing. Its trade-off is the way it fails: it granulates completely instead of cracking, which avoids large shards but loses the whole jar of product. For palletised cold-chain delivery to retail distribution centres, standard annealed glass is sufficient; tempering becomes a question mainly where consumer mishandling is the worry.

Borosilicate is the third route, with an expansion coefficient of 3.3 against 9.0 for soda-lime. It makes sense for freeze-to-oven products and is more than freeze-to-fridge needs; our page on borosilicate containers goes into the material. Even borosilicate fails when a hot jar goes directly into a blast freezer.

Freezer-rated glass jars in thermal shock testing and QC inspection
Freezer-rated jars under thermal shock testing and stress-grade inspection.

Lids and labels that survive frozen storage

Ice pushes up against the closure as well as out against the wall. A continuous-thread (CT) lid flexes a little under that load, as intended, while a lug cap can unseat once the pressure exceeds its thread engagement. For frozen retail the specification is a one-piece CT lid with a plastisol liner, which keeps its seal through normal freeze-thaw cycles. Other sealing styles are compared under airtight jar and lid options.

Seals usually fail because an under-torqued cap is lifted by expansion. On the 86-400 finish, apply 8–14 in-lbs and confirm with a retention torque check of ≥6 in-lbs after 24h.

Labels need their own specification. Standard acrylic adhesives turn brittle below -10°C and lift at the edges within weeks. Ask for a freezer-grade adhesive rated to -40°C on BOPP synthetic stock or varnished paper, since plain paper soaks up condensation during thaw and turns grey. Wet-glue and self-adhesive labels both work with the right adhesive. Give the printer the actual service temperature and request a 30-day freeze-thaw cycle test on the finished label construction.

Lot tests and your own validation run

Freezer jars get checks beyond the standard AQL 2.5 inspection:

  • Thermal shock at a 42°C differential, sampled per ASTM C149.
  • A filled freeze-cycle validation on request: retained samples are filled with water to your headspace, held at -20°C for 72 hours, thawed and inspected. This catches a geometry or annealing escape before your product meets it.
  • Vertical load of ≥180kg on 16–32oz jars, to cover frozen stacking.
  • Finish gauging so lids land inside the torque window.

Before committing to full production, freeze three cases of each SKU with your real product, headspace and freeze rate. Water is the worst case, and most food products freeze more gently. We send a validation protocol sheet with a freezer-jar project so the trial is run the same way each time.

Packing and shipping freezer jars

Export packing is divider cartons on ISPM-15 plywood pallets with double wrap. Breakage on a full container typically runs 0.5–1.5%, and rises to 1.5–3% when a smaller quantity travels as part-container cargo.

Freezer moulds carry thicker base sections, so these jars are heavier for their capacity than standard ones. A 40HQ of the 16oz model, at ~50,000 pieces, comes close to road weight limits on some US routes, and splitting the load across a pair of 20GP containers is sometimes the better booking.

The document set covers invoice, packing list, certificate of origin, an FDA food-contact statement, migration data and, on request, the lot freeze-validation report. Related formats sit in our bulk canning jar range and across the full glass jar collection.

Freezer jar questions from buyers

Our jars cracked and a competitor's similar jars did not. What should we check?

Start with process, not glass. Measure the real spread of fill heights instead of trusting the target, log the time and temperature between filling and the freezer, and compare the inside geometry of the two jars, since shoulder angles are hard to see at a glance. The crack pattern helps: a ring at the base indicates expansion, a vertical split indicates impact.

What happens if a consumer microwaves a frozen jar anyway?

Soda-lime glass can break on an extreme differential whatever the label says. The options are tempered glass for that SKU, prominent warning text, and a prompt replacement policy for the small share of jars that are mishandled.

Can filled, frozen jars be shipped on to co-packers or customers?

Yes, on two conditions. Frozen glass is brittle under impact, so each jar must be immobilised; dividers alone are not enough for parcel, and moulded pulp or foam is needed. Thaw timing also has to be controlled so the product does not linger at 4–40°C past its food-safety window. We review pack-out designs on request.

Are the same jars suitable for blast chilling and for static freezers?

The FZ jars are validated for static freezers at standard freeze rates. Blast chilling, meaning a -18°C core in under 4 hours, is not something standard jars are certified for. If the process must be blast, either move to borosilicate or spiral-cool the jars to 5°C first and blast from there.

What does a freezer jar cost?

Pricing is quoted per project and depends on the mould, decoration, quantity and freight terms. Freezer-grade annealing is the process standard on the designated moulds and is not an extra. To get a useful quote, send the product and its moisture content or water activity, the fill temperature and any cooling step, capacities and quantities with destination port, your current fill height (or ask us to recommend one), and whether distribution is retail cold chain, foodservice or parcel.