IQF vs BQF: what the freezing method changes for a frozen fruit buyer

In the IQF vs BQF choice, neither acronym is defined by the Codex quick-frozen standards (CXS 320-2015, CXS 52-1981, CXS 69-1981, CXS 75-1981, CXS 103-1981) or by CXC 8-1976. The regulated term is quick frozen, and CXS 75-1981 puts completion at a core of −18 °C after stabilisation. IQF commits the seller to pieces that do not stick together. Block freezing removes the free-flow test, the cheapest cold-chain evidence available.

What do IQF and BQF actually mean, and what does neither term guarantee?

IQF, individually quick frozen, means pieces that are frozen separately and stay separate in the bag. BQF, block quick frozen, means fruit frozen as a fused mass. For example, one European platform lists a 20 kg block against 10 kg cartons of free-flowing dice. Both are commercial words. Neither tells you how fast the fruit was frozen or what its core temperature was when it left the tunnel.

The cheaper of two offers on your desk is usually the block. If you treat the choice as a packaging preference, you will sign a specification that controls the wrong thing. The format changes what you can measure on arrival and what evidence of a temperature failure survives the voyage.

Which standards define quick freezing for tropical fruit?

None defines it for tropical fruit specifically. Codex has quick-frozen standards for vegetables (CXS 320-2015) and for four temperate fruits: strawberries CXS 52-1981, raspberries CXS 69-1981, peaches CXS 75-1981 and blueberries CXS 103-1981. For banana, mango, pineapple, jackfruit, durian, dragon fruit, passion fruit, papaya, longan, lychee and pomelo there is no Codex standard, no USDA grade standard and no EU marketing standard, because EU marketing standards cover fresh fruit only.

None of those Codex quick-frozen standards, and not the Codex code CXC 8-1976, defines IQF or BQF. What these general instruments do regulate is freezing rate and end-point, and they write both the same way for every format.

The practical reading: for quick-frozen tropical fruit, every tolerance is a contract threshold the two parties set themselves. A clause claiming conformity to an international standard is empty, because there is no standard to conform to. The food safety certification a processing plant holds governs hygiene. It does not set your free-flow or drip-loss limit.

What does the Vietnam–China durian protocol require of the freezing step?

It requires freezing at or below −35 °C for a minimum of one hour, until the product core reaches −18 °C or colder, then storage and transport held at or below −18 °C continuously in accordance with CAC/RCP 8-1976. The protocol came into force on 19 August 2024 for five years with automatic five-year renewal. It is the one binding instrument found that writes the freezing step as a number.

It also permits three physical formats under that single clause: whole fruit in shell, segments with seed removed, and de-shelled pulp. One freezing requirement, three formats, and no mention of IQF or block. A lot is quick frozen or not on the evidence of temperature and time, not on whether the pieces are separate. Raw material must come from registered growing areas, packaging must be new and previously unused, and the label must carry the registered establishment code. That code is issued to the facility, not the exporter, and since 1 June 2026 it has been issued under GACC Order 280, which replaced Order 248.

Codex CXS 75-1981, which governs peaches rather than durian, defines completion by end-point alone: the core at −18 °C after stabilisation. The Vietnam–China durian protocol adds an exposure on top of that end-point: at least one hour at −35 °C. The unresolved divergence sits between that dwell minimum and the tunnel times equipment suppliers publish, 20–40 minutes at −35 to −40 °C for a straight-belt IQF tunnel. A fast line can therefore reach the −18 °C end-point in 25 minutes and still fall short of a literal reading of the protocol’s dwell minimum. Whether Chinese inspection enforces the dwell or only the end-point is not established in any document reviewed for this article. If a supplier quotes tunnel minutes for durian bound for China, ask which of the two requirements the figure answers.

What does freezing rate physically change in the fruit?

Rate controls ice-crystal size, and crystal size controls how much cell structure survives. Slow freezing is the recorded root cause of three separate defects: large ice crystals, high drip loss and mushy texture after thawing. Ice crystals cannot be measured at a factory or on receipt, so drip loss is used as the indirect measure, and it is the cheapest one available.

The one published comparison of the two formats found for this article comes from Nutrada, a European trading platform: drip loss of 3–8% for IQF against 10–20% for block, with freezing at −30 to −40 °C in 3–30 minutes. It carries no thaw regime, and thawing at 4 °C for 18 hours and at 20 °C for one hour do not give the same answer on the same product. Treat it as one platform’s claim, not a standard, and measure against the full protocol: 200 g weighed still frozen, a 2 mm sieve, thaw at 4 °C for 18 hours, reweigh the solids, and express the difference over the starting weight.

The four freezing routes in commercial use in Vietnam differ most in time to complete.

These are vendor parameters, not acceptance criteria. Use them to frame questions to a plant.

Two points change how a buyer reads the table. First, two freezer types are ruled out for ripe banana. A fluidised bed fails because a soft, sticky, high-sugar piece will not fluidise, and an impingement freezer fails because its high-velocity air lifts and flips thin soft slices off the belt. Belt-type tunnels, straight or spiral, are still IQF methods and remain usable, and block remains an option, so for this product the choice is partly what the fruit tolerates rather than a cost decision. Second, dewatering by vibration or air knife immediately before the tunnel is the step that decides the drip-loss figure, and incomplete dewatering is also why pieces fuse in store.

How do you test free flow, and what does a clump tell you?

Pour 1 kg of product still frozen onto a tray, separate it with light hand pressure, and weigh the portion still adhering. That is the whole free-flow test. It is enforceable only once the contract defines the unit of adhesion, and the published documents do not agree on that unit.

The one published sentence stating what IQF commits a supplier to comes from CBI buyer guidance:

for IQF, the pieces must not stick to one another

Read plainly, that is a commitment about separation and nothing else. The limits written against it vary:

A contract that borrows any of these limits should attach the 1 kg tray test described above as the method. They are different documents doing different jobs. Copy a number without the definition beside it and you have imported the figure and left the argument behind.

A clump is a diagnostic as well as a defect. Clumping has three root causes at three different stages: product still wet entering the tunnel; a tunnel overloaded or with airflow too weak; or partial thawing followed by refreezing in store or in transit. The third cause is the one a buyer in Rotterdam or Osaka cannot otherwise audit. It usually arrives with a companion signature: ice sheeting inside the liner, pieces stuck together, loose white snow in the carton. Weigh the bag, let the ice melt, weigh again; net weight excludes glaze, the rule CXS 320-2015 sets for vegetables, which a fruit contract should adopt in writing.

What does a buyer lose by choosing block frozen fruit?

The free-flow test, and with it the retrospective evidence of a thaw-and-refreeze event. In a 20 kg block the pieces are fused by design, so fusing caused by a temperature break in week three of the voyage leaves nothing for a tray test to find. The buyer has not saved the cost of a test. They have lost the test.

What replaces it must be specified deliberately. Where a lot is bought as block rather than free-flow, VinFruits Global writes the three-position data logger and the bore-hole core measurement into the specification, because the free-flow test that would otherwise have reported on the voyage no longer exists.

The replacement evidence, clause by clause:

• Core temperature, destructive, per CXC 8-1976. Bore at least 2.5 cm with a pre-chilled tool; probe accurate to ±0.5 °C, display resolution 0.1 °C. A non-destructive reading between cartons carries a 2.8 °C tolerance; use it to screen, and confirm destructively before rejecting.
• The Critical line. A core warmer than −12 °C, the CXC 8-1976 ceiling for the warmest package, is a Critical defect: reject the lot without price negotiation.
• Data loggers in every container at three positions: behind the door, mid-load and at the end wall. A logged breach is an independent ground for rejection, separate from any sensory finding.
• Drip loss still works on a block, but no published document says how to take 200 g from a frozen block without the sampling becoming breakage. Specify the tool and the sample state, and have both parties sign it.
• Metal detection. BRCGS Food Safety Issue 9 requires metal-detector sensitivity validated on the product and specifies no number. Frozen mass reduces real detector sensitivity, so ask for the validation report on this product in this format, and ask whether X-ray is used on block.

The premium for IQF is therefore partly the instrumentation you can stop paying for: buy block and you pay for loggers and destructive core measurement instead.

Does sampling and acceptance change between IQF and block?

No. Codex CXS 233-1969 at AQL 6.5 Level I puts pack units above 4.5 kg in one band, and both 10 kg cartons and 20 kg blocks fall inside it. CXS 320-2015 adds that Level II applies on dispute.

Keep legal thresholds, such as core temperature and Salmonella, apart from commercial ones, such as clumping, colour and size. A legal threshold cannot be traded for a discount. A workable split: Critical at zero and reject; Major at AQL 2.5; Minor at AQL 6.5. Write the failure path in advance. Major over AQL goes, in order, to 100% re-sorting at the seller’s cost, downgrade against a pre-published price scale, then return. Sieve while frozen and count after thawing, because sieving thawed product breaks pieces and biases the result against the seller. Every percentage needs three attachments: by weight or by count, on what sample size, in what physical state.

Neither format has a kill step. Freezing arrests micro-organisms; it does not destroy them.

Is block cheaper once you compare it properly?

Possibly, but price per kilo of finished goods is the wrong basis. The valid comparison is value returned per kilo of raw fruit entering the plant. Block is where broken pieces and fines legitimately go, and GG-2615 shows a whole-and-broken grade can be a saleable product with its own specification. A yield quoted without its definition, biological or factory-level, is two different numbers.

The price differential between IQF and block for the same fruit could not be found in public sources; it sits behind paid trade-data services such as Tridge, Mintec and Expana, and this article does not estimate it. The honest route is to ask one plant to quote the same fruit in both formats to the same specification. If there is no price difference, the faster technology is a cost to the plant, not an advantage to you.

What should a specification ask instead of the technology name?

Ask for results. Three questions do most of the work: what drip loss do you actually measure, show the core-temperature chart for the most recent lot, and what is the price difference between your fast-frozen line and your standard line. Then write the answers into a signed, versioned annex to the contract.

The clauses to carry:

1. Quick frozen per CXC 8-1976 and CXS 75-1981: rapid passage through −1 to −5 °C, core −18 °C after stabilisation.
2. Drip loss: 200 g weighed still frozen, 2 mm sieve, thaw at 4 °C for 18 hours, reweigh solids, (m1−m2)/m1 × 100. Name the regime; 20 °C for 1 hour gives a different number.
3. Free flow (IQF only): 1 kg tray test, clumps defined as ≤3 pieces and agglomerates as ≥5.
4. Signed defect photo set as an annex. Test it first: three assessors, the same five samples, agreement within 20% relative.
5. Sealed, countersigned retention sample stored under label conditions; remaining shelf life on arrival at least 75%, which works whichever shelf-life declaration is printed; inspection not more than 60 days before delivery, as the USDA Commodity Specification for Frozen Fruit of June 2017 sets.
6. Dedicated cold store, no seafood, onion or garlic; odour checked on opening the carton, before thawing.

Common questions

Is there a standard that defines IQF or BQF?

No. None of the Codex quick-frozen standards (CXS 320-2015, 52, 69, 75 and 103-1981) or the Codex code CXC 8-1976 defines either term, and no Codex, USDA grade or EU marketing standard exists for quick-frozen tropical fruit such as banana, mango, pineapple, jackfruit or durian. The regulated term is quick frozen. Codex CXS 75-1981 defines completion as a core of −18 °C after thermal stabilisation, and CXC 8-1976 requires rapid passage through −1 to −5 °C. Both apply equally to free-flowing pieces and to blocks.

What does IQF commit a supplier to that frozen does not?

Only separation of pieces. CBI buyer guidance states that for IQF the pieces must not stick to one another, and that is the whole commitment. It becomes enforceable when the contract defines the unit of adhesion, such as clumps of up to three pieces and agglomerates of five or more as in the GG-2616 mango specification, and names the test: 1 kg poured still frozen onto a tray and separated by light hand pressure.

How is drip loss measured, and what does the figure tell you?

Weigh 200 g still frozen, place it on a 2 mm sieve, thaw at 4 °C for 18 hours, weigh the remaining solids, and express the difference as a percentage of the starting weight. Drip loss is the cheapest indirect measure of freezing speed, because ice-crystal size cannot be checked at a factory or on receipt. A figure quoted without its thaw regime is not comparable with any other.

Can block-frozen fruit be sold as quick frozen?

Yes, if it meets the rate and end-point requirements. Codex ties quick freezing to how fast the product crosses −1 to −5 °C and to a −18 °C core, not to whether pieces are separate. The Vietnam–China frozen durian protocol in force since 19 August 2024 applies one freezing clause to whole fruit, segments and pulp alike. A block is quick frozen or not on the same evidence as IQF.

If block is cheaper, what does a buyer give up?

The free-flow test, and with it the retrospective evidence of a thaw-and-refreeze event in store or in transit. Pieces fused by design cannot show new fusing. A block buyer should instead specify destructive core measurement per CXC 8-1976, a −12 °C Critical rejection line, and data loggers at three positions in each container. The size of the price gap itself could not be found in public sources; it sits behind paid trade-data services.