IMSBC Group A cargoes: cargoes that may liquefy

IMSBC Group A consists of cargoes which possess a hazard due to moisture that may result in liquefaction or dynamic separation if shipped at a moisture content in excess of their transportable moisture limit. That is the definition in section 1.7 of the International Maritime Solid Bulk Cargoes Code, in the form given by amendment 06-21 and carried into amendment 07-23, IMO resolution MSC.539(107), which has been mandatory since 1 January 2025.

The definition describes a hazard mechanism rather than a commodity list, which is the single most important thing to understand about the group. The same material can be Group A or Group C depending on its particle size distribution and its drainage behaviour, and a certificate can move a consignment from one schedule to the other. Bauxite and coal both demonstrate it, and both are treated below.

The three groups, and where Group A sits

The IMSBC Code sorts every solid bulk cargo into one of three groups by hazard. Group A cargoes may liquefy or undergo dynamic separation. Group B cargoes possess a chemical hazard, whether as dangerous goods in solid form in bulk or as materials hazardous only in bulk. Group C cargoes are neither.

A cargo can be in two groups at once. The Code writes this as group A and B, and it is a declared value on the cargo declaration form rather than a fourth group: the form’s group cell carries four checkboxes, group A and B, group A, group B and group C, and section 1.7 defines only the three. FLUORSPAR, the METAL SULPHIDE CONCENTRATES family and DIRECT REDUCED IRON (D) are all group A and B, and each carries both the moisture regime and a chemical one.

Amendment 06-21 also drew a distinction inside Group A itself, defining cargoes which may liquefy and cargoes which may undergo dynamic separation as two separate terms. Both sit within the group; they are not synonyms.

Liquefaction: the mechanism the Code states

Section 7.2.1 sets out three steps. The void volume between particles reduces as the cargo compacts under ship motion; that raises the water pressure in the remaining space; and the raised water pressure reduces the friction between particles and therefore the shear strength of the cargo mass.

Section 7.2.1 also carries the sentence that undermines the most common misconception about the hazard: Group A cargoes may liquefy during a voyage even when they are cohesive and trimmed level. A cargo that looks stable on completion of loading, and that has been trimmed exactly as required, is not thereby safe.

Section 7.2.2 gives the exclusion, and it is the principle the whole classification scheme rests on: liquefaction does not occur where the cargo consists of large particles or lumps and water passes through the spaces between them without any increase in water pressure. Drainage, not dryness, is what makes a coarse cargo safe.

Section 7.2.4 describes the consequence for the ship. The fluid flows to one side on a roll and does not completely return, so the ship may progressively reach a dangerous heel and capsize quite suddenly. It is a one-way transfer rather than an oscillation, which is why the list develops across a series of rolls and why the end arrives without much warning. The physics is treated at greater length at cargo liquefaction and cargo fluidisation and aeration .

The wet base, and why an average moisture content is not enough

Section 7.2.3 addresses moisture migration during the voyage. A shift may occur where the moisture content exceeds the transportable moisture limit, but the section then adds a warning that changes how a declaration should be read: some cargoes are susceptible to moisture migration and may develop a dangerous wet base even where the average moisture content is less than the TML. Although the cargo surface may appear dry, undetected liquefaction may take place.

That is why section 4.2.2.11 requires the shipper to declare the likelihood of formation of a wet base as a separate item of cargo information, and why an average figure comfortably under the limit is not by itself an answer to a hold that has been rolling for a week. The mechanism has its own treatment at dynamic separation of solid bulk cargoes .

The acceptance rule, and the exemption most descriptions omit

Section 7.3.1.1 is the operative provision: Group A cargoes shall only be accepted for loading when the actual moisture content of the cargo is less than its transportable moisture limit. It is drafted as a prohibition on acceptance, not as a discretion, and it is the source of the practical right to turn a wet cargo away.

Section 7.3.1.2 then disapplies a large part of the regime by ship type. Where the cargo is carried on a specially constructed or fitted ship under 7.3.2, or a specially constructed cargo ship for dry powdery cargoes under 7.3.3, sections 4.2.2.9, 4.2.2.10, 4.3.2 to 4.3.5, 4.5, 4.6 and the whole of section 8 need not apply. The entire testing and certification apparatus is switched off for those ships, because the ship’s structure rather than the cargo’s moisture content is what limits the consequence of a shift.

A specially constructed ship has permanent structural boundaries confining any shift to an acceptable limit. A specially fitted ship uses portable divisions, which shall not be constructed of wood, with the bounding structure strengthened and the plan and stability conditions approved by the Administration. The submission package under 7.3.2.3 is structural drawings including scaled longitudinal and transverse sections, stability calculations taking cargo shift into account, and any other assisting information. A 7.3.3 ship is designed to carry solely dry powdery cargoes and to handle them by closed pneumatic systems that prevent exposure to weather. The type is covered at specially constructed or fitted cargo ships .

Three further provisions round out section 7.3. No cargoes containing liquids may be stowed in the same space above or adjacent, other than packaged canned goods and the like; measures are required to prevent liquids entering the space; and masters are cautioned about using water to cool these cargoes at sea, because it may bring the cargo to a flow state, with spray to be used where water is necessary.

What the shipper must provide

Section 4.2.1 requires the shipper to provide the information sufficiently in advance of loading, and section 4.2.2 requires it to be confirmed in writing and by appropriate shipping documents prior to loading, listing eighteen items. Four matter here: the cargo group, given as A and B, A, B or C; the bulk density as required by SOLAS regulation XII/10; the certificate on moisture content and transportable moisture limit in the case of a Group A cargo; and the likelihood of formation of a wet base.

Section 4.3.2 sets out the two-document rule, and its asymmetry is easy to miss. The shipper provides the master with a signed certificate of the TML, and a signed certificate or declaration of the moisture content, each issued by an entity recognised by the competent authority of the port of loading. The TML certificate must contain or be accompanied by the result of the determining test; the moisture content declaration must contain or be accompanied by a statement that the figure is, to the best of the shipper’s knowledge and belief, the average moisture content at the time it is presented.

Section 4.3.3 is the third document, and it is the one most often absent from an account of the regime. The shipper must establish procedures for sampling, testing and controlling moisture content so as to ensure the content is less than the TML when on board; those procedures must be approved, and their implementation checked, by the competent authority of the port of loading; and the authority’s document stating the approval must be provided to the master or their representative.

Two extensions follow. Section 4.3.4 requires the 4.3.3 procedures to include protection of cargo on barges from precipitation and water ingress, which is the Indonesian and Philippine transhipment case written into the Code. Section 4.3.5 requires per-space certification where a Group A cargo goes into more than one cargo space, unless standard sampling shows the moisture content is uniform throughout the consignment. The declaration itself is covered at IMSBC section 4 and the shipper’s cargo declaration .

The two intervals

Source: IMSBC Code section 4.5, as given by MSC.539(107).

The TML interval carries a standing retest duty: where the composition or characteristics of the cargo are variable for any reason, the shipper must ensure a further test is conducted after it is reasonably assumed that such variation has taken place. A six-month certificate on a variable cargo is not by itself compliance.

The rain limb of 4.5.2 is anchored to completion of loading, not to commencement, and it is a duty rather than an invalidation. If the cargo has been exposed to significant rain or snow between the time of testing and the date of completion of loading, the shipper is responsible for ensuring the moisture content is still less than the TML and for providing evidence of that to the master as soon as practicable. The declaration does not become void; the shipper acquires an obligation to re-evidence it.

Sampling

Section 4.6.3 requires subsamples to be drawn approximately 50 cm below the surface. Section 4.6.4 sets the density at one 200 g subsample per 125 tonnes up to 15,000 tonnes, per 250 tonnes from 15,000 to 60,000 tonnes, and per 500 tonnes above 60,000 tonnes.

Section 4.4.6 requires moisture samples to be placed immediately in suitable airtight, non-absorbent containers with a minimum of free air space, because a sample that dries in transit understates the very quantity being measured. Section 4.4.3 requires the shipper to facilitate access to stockpiles for inspection, sampling and subsequent testing by the ship’s nominated representative, and section 4.4.8 limits stationary stockpile sampling for unprocessed mineral ores to cases where access to the full depth of the stockpile is available. Section 4.7 lists example sampling standards for information only, and notes of one of them that in situ sampling of ships and stockpiles is not permitted under that standard. The practice is treated at sampling of solid bulk cargoes .

The six test methods

Appendix 2 paragraph 1 opens by stating that six methods of testing for the transportable moisture limit are currently in general use.

Source: IMSBC Code appendix 2 paragraph 1, as given by MSC.539(107).

There is no general rule that the transportable moisture limit equals 90 percent of the flow moisture point. That relation belongs to the flow table and penetration routes only. The three Proctor/Fagerberg routes never determine a flow moisture point at all: they read a critical moisture content off a compaction curve at a stated degree of saturation. Presenting the 0.9 relation as the definition of the TML is wrong for iron ore fines, for coal, for bauxite and for every concentrate tested by Proctor/Fagerberg.

One exception sits inside the flow table method itself. For peat moss, bulk density is determined first, and above 90 kg per cubic metre on a dry weight basis the TML is 85 percent of the flow moisture point rather than 90 percent.

Both modified tests carry a free-draining outcome, and they are worded differently. For coal, where the moisture drains so freely that the compaction curve does not reach 70 percent saturation, paragraph 1.5.1 treats the result as indicating a cargo where water passes between the particles with no increase in pore water pressure, so the cargo is not liable to liquefy. For bauxite, paragraph 1.6.1.4 goes further and states that the cargo is not classified as group A. The coal wording addresses the hazard; the bauxite wording addresses the classification.

Appendix 2 also names its normative references, including the standard flow table apparatus and, for the coal method, a set of national and international sampling and sizing standards. The methods and their procedures are covered at IMSBC appendix 2 test procedures and IMSBC section 8 , and the individual routes at the penetration test .

Appendix 2 is a fallback, not a mandate

Section 8.3 states that the recommended methods for determining the transportable moisture limit are given in appendix 2. Recommended, not prescribed. Section 4.1.4 sets the hierarchy: the determination is made in accordance with test procedures approved by a competent authority in the country of origin where such procedures exist, and by the appendix 2 procedures only in their absence.

That matters when a certificate is challenged. A TML certificate is not invalid merely because it does not name an appendix 2 method, provided the procedure used carries competent authority approval. What can be challenged is whether the method suits the cargo, since several appendix 2 methods exclude specific materials by their own terms, and whether the authority that approved it is a competent authority within the meaning of the Code, whose definition ends with the requirement that it shall operate independently from the shipper.

The can test

Section 8.4 gives a shipboard complementary procedure. Half fill a cylindrical can or similar container of 0.5 to 1 litre capacity with the material; bring it down sharply onto a hard surface from a height of about 0.2 m; repeat 25 times at one or two second intervals; then examine the surface for free moisture or a fluid condition. If either appears, arrange for additional laboratory tests before accepting the material.

Section 8.4.2 is the whole point of it: a dry can test result does not prove the moisture content is below the transportable moisture limit. The test can indicate a problem and cannot certify its absence, which makes it a screening tool for the ship’s staff rather than an answer to a bad declaration. It is treated separately at the can test .

Bauxite: the split, and the certificate that moves a cargo between schedules

BAUXITE is Group C. BAUXITE FINES is Group A. The dividing line is particle size, and it runs both ways on a certificate.

The Group C BAUXITE schedule applies to cargoes containing 30 percent or less of particles under 1 mm, or 40 percent or less of particles under 2.5 mm, or both, or where the shipper provides the master with a certificate, in accordance with a test approved by the competent authority of the port of loading, stating that the moisture freely drains so that the degree of saturation is not liable to reach 70 percent.

The Group A BAUXITE FINES schedule applies to cargoes containing both more than 30 percent of particles under 1 mm and more than 40 percent under 2.5 mm, and carries the same free-draining escape in reverse: such a cargo may be carried under the Group C schedule on the same form of certificate.

So a bauxite cargo can move between the two schedules on a certificate rather than on its name, which is the single most practically important thing about the commodity and is missing from most accounts of it. The Group A schedule’s hazard cell is also the only place in the Code where the dynamic separation mechanism is stated in schedule prose, describing instability due to moisture resulting in dynamic separation and the formation of a liquid slurry of water and fine solids above the solid material, and stating that the cargo is not liable to undergo dynamic separation when shipped below its TML. Both schedules have their own treatment at the BAUXITE schedule and the BAUXITE FINES schedule .

The regulatory chain behind the split is short and dated. CCC.1/Circ.2 of 20 October 2015 followed consideration at CCC 2 of a Bahamas submission on the loss of the Bulk Jupiter, recording that the cargo had been declared as Group C and that the potential for bauxite to liquefy was not specifically addressed in the Code. CCC.1/Circ.2/Rev.1 of 20 September 2017 superseded it, recording that work by the industry Global Bauxite Working Group with competent authorities indicated that bauxite presents a moisture risk and that some cargoes should be treated as Group A. CCC 4 finalised the test procedure, the BAUXITE FINES schedule and the amendments to the Group C schedule, and they arrived with amendment 05-19, mandatory from 1 January 2021.

The principal Group A cargoes

Source: IMSBC Code appendix 1 schedules, MSC.539(107).

The group is wider than ore. The heavy mineral sands including ILMENITE SAND, ILMENITE CLAY, TITANOMAGNETITE SAND and OLIVINE SAND are Group A while plain SAND, RUTILE SAND and ZIRCONSAND are Group C. So are COAL SLURRY, COKE BREEZE, COPPER SLAG, FLY ASH WET, GROUND GRANULATED BLAST FURNACE SLAG, MAGNESITE FINES and a long tail of process residues. ALUMINIUM FLUORIDE and BARYTE FLOTATION CHEMICAL GRADE make a useful worked distinction: both appear in the appendix 3 list of cargoes that are non-cohesive when dry, and both are Group A when wet.

Individual schedules are covered at the IMSBC individual cargo schedule and its mandatory and informative fields , with worked examples at iron ore fines , nickel ore , mineral concentrates , iron ore concentrate , copper concentrate , lead concentrate , zinc concentrate and coal .

No total count of Group A schedules appears here. The appendix 4 index carries around a hundred rows marked Group A or Group A and B, but roughly half are secondary names cross-referring to a bulk cargo shipping name, the Mineral concentrates schedule alone absorbs dozens of index entries, and appendix 4 is informative rather than normative under section 1.4.2. Verify a cargo against its appendix 1 schedule and its bulk cargo shipping name , not against a count.

Coal

The COAL schedule head reads that coal shall be classified as group A and B unless classified as group B only by a test determined by the appropriate authority, or where it has not more than 10 percent by weight of particles less than 1 mm and not more than 50 percent by weight of particles less than 10 mm. Notwithstanding that, a blend of two or more coals is group A and B unless all the original coals in the blend are group B only.

The default is group A and B, and the escape is to group B only. Descriptions that state the reverse, or that give a 5 mm fines threshold, are wrong on both counts: there is no 5 mm figure anywhere in the coal provisions, and the 5 mm that circulates belongs to the top size limit of the standard Proctor/Fagerberg test, which may not be used for coal at all.

The schedule’s group cell reads “B (and A)” and its hazard cell records that coal may create flammable atmospheres, may heat spontaneously, may deplete the oxygen concentration and may corrode metal structures, in addition to liquefying above its TML. The chemical half of that is covered at self-heating and spontaneous combustion , oxygen depletion in cargo spaces and carbon monoxide in cargo spaces .

The Group A schedule boilerplate

Every Group A schedule carries the same five-limb weather precaution, conditional on the ship not complying with section 7.3.2: the moisture content shall be kept less than the TML during loading operations and the voyage; unless the schedule expressly provides otherwise, the cargo shall not be handled during precipitation; unless the schedule expressly provides otherwise, all non-working hatches of the spaces loaded or to be loaded shall be closed during handling; the cargo may be handled during precipitation under the conditions stated in the procedures required in 4.3.3; and a cargo space may be discharged during precipitation provided the total amount of cargo in that space is to be discharged in the port.

The fourth limb is the practical answer to whether a Group A cargo can be loaded in the rain: only under the competent-authority-approved procedures, and the approval document must already be in the master’s hands. A flat statement that loading stops in rain describes the default, not the rule. Precipitation practice and its interaction with hold condition is covered at cargo hold preparation standards , dust suppression in bulk cargo operations and cargo hold bilge well preparation .

The precautions cell requires bilge wells to be clean, dry and covered as appropriate to prevent ingress of the cargo, and the bilge system of a space to be loaded to be tested to ensure it works. The loading cell carries the high-density hook: where the stowage factor is equal to or less than 0.56 cubic metres per tonne, the tank top may be overstressed unless the cargo is evenly spread to equalise the weight distribution, which links the moisture regime to the separate structural regime at high-density solid bulk cargo , alternate hold loading and SOLAS chapter XII .

The master’s position

The Code confers no express right to refuse a Group A cargo. It gives the master two named rights. Section 5.1.3 gives the right to require that the cargo be trimmed level where there is any concern regarding stability. SOLAS regulation VI/7.5, reproduced verbatim inside the Code at section 1.6, gives the right to suspend loading or unloading where the ship’s limits in the agreed plan are exceeded or are likely to become so, together with an obligation to notify the appropriate authority of the port State with which the plan has been lodged.

The practical right to turn a wet cargo away comes from reading section 7.3.1.1 as a prohibition on acceptance rather than from any express grant. The distinction matters commercially, because the master’s position against a shipper, a charterer and a port authority is a contractual and P&I question layered on top of the Code, and it is treated at the master’s right to refuse cargo rather than here. The relevant SOLAS hook is chapter VI , and the loading plan itself sits under cargo distribution and trimming provisions .

Trimming is worth one further word, because it is often offered as a mitigation and is not one. Section 7.1.2 distinguishes two types of cargo shift, sliding failure and the consequence of liquefaction, and states that trimming in accordance with section 5 can prevent sliding failure. It says nothing of the kind about liquefaction, and section 7.2.1 says the opposite in terms.

The casualty record

Three losses are set out here because each has a named investigating body and a report that can be traced. Losses widely attributed to liquefaction in trade commentary without a published flag State report are not listed, because a casualty count assembled from commentary is not a casualty count.

Bulk Jupiter. The Bahamas Maritime Authority investigated the loss of the bulk carrier in the South China Sea on 2 January 2015 and issued its report on 18 August 2015. The ship, IMO 9339947, had loaded 46,400 tonnes of bauxite at Kuantan, Malaysia, and carried 19 crew, of whom 18 were lost, with one survivor. The investigation ran under the IMO Casualty Investigation Code mandated by SOLAS regulation XI-1/6.

The finding is not the one usually attributed to it. The report concluded that, having deduced that the probability of structural failure was low as a singular causal event, it was most probable that either liquefaction or a free surface effect induced an unrecoverable list. It did not find dynamic separation, and could not have: the term did not enter the Code until amendment 06-21, seven years later. The report also recorded that the shippers’ declaration forms were considered generic.

Emerald Star, IMO 9449261, a Hong Kong China flag bulk carrier lost on 13 October 2017 with 10 lives, investigation report on IMO GISIS dated 2 June 2021.

Nur Allya, IMO 9245237, an Indonesian flag vessel lost on 20 August 2019 with 27 lives, GISIS report dated 12 November 2021.

Stellar Daisy is the counter-example and belongs in the record for that reason. It was investigated as a structural failure and no liquefaction finding was made. Grouping it with the liquefaction losses inflates the count and obscures a distinct failure mode in converted very large ore carriers. Casualty investigation practice generally is covered at marine cargo damage investigation .

What the industry record shows

INTERCARGO publishes an annual Bulk Carrier Casualty Report on a rolling ten-year window. The 2026 edition, published 7 July 2026 and covering 2016 to 2025, records 17 bulk carriers over 10,000 dwt lost, totalling 1.63 million dwt, with 71 lives, and identifies cargo liquefaction as the leading cause of loss of life at 37 fatalities, over half the total. The 50,000 to 59,999 deadweight band accounts for 4 of the 17 losses but 52.1 percent of the lives lost, being the same 37.

The 2025 edition, covering 2015 to 2024, gives 20 losses, 89 lives, and 55 lives or 61.8 percent attributed to liquefaction from 3 ship losses.

One qualification governs how those figures should be cited. The report’s own footnote states that liquefaction is used to describe liquefaction and dynamic separation, so the number is a combined moisture-hazard total rather than a liquefaction total. It is still the most useful single figure in the field, because it says plainly what the group costs: a hazard that accounts for a minority of ship losses and a majority of the deaths.

Amendment history for Group A

Sources: the adopting IMO resolutions.

Amendment 07-23 replaces the complete text of the Code, which has a consequence for any statement about what it changed: the resolution carries the whole text rather than a list of amendments, so a claim that 07-23 added a particular schedule or a particular declaration field cannot be read off the resolution and has to be traced through the earlier amendment that introduced it. Amendment 08-25 has been voluntary since 1 January 2026 and is adopted and accepted but not yet in force, so it is not the applicable text for a cargo loaded today. The cycle itself is covered at the IMSBC Code amendment cycle , and the route for a cargo the Code does not list at cargoes not listed and tripartite agreements , with the alternative route at exemptions and equivalent measures .

Limitations

No count of Group A schedules is given, deliberately. The appendix 4 index is informative rather than normative and counts secondary names, and three independent parses of appendix 1 disagree with each other. A count would be a number with no source behind it.

A group letter is a property of the consignment, not of the commodity. Bauxite, coal, iron ore, manganese ore and several of the sands each appear in both a Group A and a Group C schedule, and the dividing line is particle size and drainage behaviour, modified in some cases by a certificate. Any statement that a named commodity is or is not Group A has to be checked against the specific schedule and the specific consignment.

Casualties attributed to liquefaction in trade commentary are not listed unless a flag State report can be named. Several losses commonly cited in this context could not be traced to a published investigation this pass, and are therefore absent rather than hedged. Where an investigation is open, no cause is stated here.

The INTERCARGO figures combine liquefaction and dynamic separation by the report’s own definition, so they should be cited as a moisture-hazard total. They also cover bulk carriers over 10,000 dwt only, so smaller vessels and non-bulk-carrier losses fall outside them.

This article describes the Code, not a specific national regime. Competent authorities in the major loading countries publish their own approved test procedures and their own conditions, which rank ahead of appendix 2 under section 4.1.4, and a port State may add requirements of its own. Read the loading country’s rules alongside the Code, and read the flag State’s position on any exemption.

Related Articles