The selection of polyamide 6 (PA6) for retort pouch laminates requires a clear distinction between extrusion-grade resins, film-grade resins with controlled monomer stripping, and halogen-free additive packages that avoid chlorinated paraffins, polyvinylidene chloride, brominated flame retardants, and chlorinated external lubricants. Residual caprolactam in PA6 is regulated under EU Regulation 10/2011 with a specific migration limit of 15 mg/kg in food simulants, while FDA 21 CFR 177.1500 covers nylon resins with extractives limits that are differentiated by food type and use temperature. The halogen-free requirement does not, by itself, produce low migration; rather, it removes certain halogenated barrier and additive chemistries and places greater dependence on polyamide morphology, layer order, retort temperature, and sealant thickness. Retort sterilisation is normally conducted at 121 °C for 30 min or at 135 °C for a shorter hold interval, and these conditions drive migration through accelerated diffusion, steam plasticisation, and hydrolytic chain scission of the PA6 phase. Published data for specific coextruded halogen-free PA6 configurations under high-temperature retort are more limited than monolayer migration studies, so a valid compliance assessment must be obtained on the complete laminate rather than on isolated resin pellets or monolayer film.

What Limits Caprolactam Diffusion During Retort Steam Exposure?

Residual caprolactam concentration in the PA6 film is the first quantitative boundary because caprolactam has a molecular weight of 113.16 g/mol, high water solubility, and a diffusion coefficient that rises strongly in the amorphous phase when the layer is plasticised by steam. Commercial film-grade PA6 is normally produced with monomer stripping after polymerisation, and residual caprolactam is measured by extraction and HPLC-UV; film grades intended for retort applications are often specified below 0.15 wt%, although general-purpose film grades may be supplied below 0.3 wt%. Each finished lot requires verification because monomer content can rise at the film die when melt temperature exceeds 260 °C or when vacuum venting is inadequate. The retort cycle increases the diffusion coefficient of caprolactam in PA6 by several orders of magnitude relative to ambient storage, because water absorbed at 121 °C lowers the glass transition temperature of the amorphous phase and increases segmental mobility. Water also attacks the amide linkage, producing chain ends and regenerating caprolactam, especially at the film surface and at interfacial boundaries with tie resins. Cyclic dimer and cyclic trimer are additional PA6 migrants that are detected by LC-MS/MS after food-simulant exposure, and their migration is controlled less by total monomer level than by crystal morphology and orientation, because crystalline domains are effectively impermeable to oligomers. Biaxially oriented PA6 film with a density near 1.13 g/cm³ typically provides lower caprolactam migration than cast film of the same resin, because orientation and heat-setting increase crystallinity and reduce the mobile amorphous fraction. On a production line, monomer control is achieved with a twin-screw extruder having an L/D ratio of 40:1 to 52:1, vacuum devolatilisation between −0.08 MPa and −0.095 MPa, and melt temperature maintained between 240 °C and 260 °C. The processing window is narrow because lower melt temperature reduces monomer stripping while higher melt temperature regenerates caprolactam by thermal depolymerisation.

In coextruded retort pouch laminates, PA6 is rarely placed as the direct food-contact surface because cast polypropylene provides reliable heat sealing, puncture resistance, and a lower extractable profile in aqueous and fatty simulants. A common halogen-free structure is reverse-printed biaxially oriented PET at 12 µm, laminated to 15 µm biaxially oriented PA6, and then laminated or extrusion-laminated through a maleic anhydride-grafted polypropylene tie resin to a 70 µm cast polypropylene sealant. The PA6 layer is therefore not in direct contact with food, but its low-molecular-weight constituents can still migrate through the tie resin and polypropylene sealant when the retort time–temperature integral exceeds the lag capacity of the sealant. Tie-resin selection is critical because maleic anhydride grafted polypropylene with a high graft level improves adhesion but can introduce polar reaction products and low-molecular-weight fragments; halogen-free tie resins with an anhydride graft level of 0.5 % to 1.2 % and a melt flow rate of 2 g/10 min to 8 g/10 min at 230 °C under 2.16 kg are commonly used, but the exact grade must be validated after retort because absorbed water attacks the PA6/tie interface during cooling. Delamination at that interface is a failure mode that couples mechanical loss with migration because it creates channels and increases the effective diffusion area. Interlayer bond strength before retort may exceed 3 N/15 mm, while post-retort values can drop below 1 N/15 mm if steam condensate is trapped at the interface. Laminating adhesives must be retortable and aliphatic; aromatic polyurethane adhesives can generate primary aromatic amines under high-humidity retort conditions, and the combined migration of primary aromatic amines is limited to 0.01 mg/kg in food simulants under EU Regulation 10/2011. In a halogen-free structure, the absence of polyvinylidene chloride means that oxygen barrier must be supplied by EVOH or PA6; PA6 remains useful as a secondary oxygen barrier and flex-crack resistance layer even when EVOH is present, but its oxygen permeability increases sharply after steam absorption.

Moisture Uptake Turns the PA6 Amorphous Phase into a Diffusion Pathway

The oxygen transmission rate of biaxially oriented PA6 film is highly dependent on relative humidity, and this humidity dependence determines how the laminate behaves before, during, and after retort. At 23 °C and 0 % RH, a 15 µm PA6 film may exhibit an oxygen transmission rate in the range of 15 cm³/m²·day·bar to 25 cm³/m²·day·bar according to ASTM D3985, but at 85 % RH the value can rise above 100 cm³/m²·day·bar because absorbed water increases segmental mobility and oxygen diffusion. Moisture uptake of PA6 at 23 °C and 50 % RH is approximately 2.5 % to 3.0 % by mass according to ISO 62, and saturated steam at 121 °C forces the layer to a much higher moisture content within minutes. This absorbed water lowers the glass transition temperature of the polyamide from a dry value near 60 °C to below 20 °C; at retort temperature the amorphous phase is therefore rubbery, and the diffusion coefficients of oxygen, caprolactam, and cyclic oligomers are elevated. Hydrolysis competes with plasticisation because water cleaves the amide bond, producing chain ends and regenerating caprolactam monomer at the layer surface. The extent of hydrolysis during a standard 121 °C/30 min retort cycle is usually small for high-molecular-weight film grades with a relative viscosity above 3.6 in sulfuric acid solution, but repeated retorting or extended high-temperature cycles at 135 °C can lower molecular weight enough to reduce tensile elongation at break below 50 % and increase the extractable fraction. Migration testing of the laminate rather than the PA6 monolayer is necessary because the sealant and tie layers partition the migrant stream; extraction of the full pouch with food simulant A or B after retort and analysis by HPLC-UV or LC-MS/MS provides the compliance data needed for EU Regulation 10/2011. Oxygen barrier after retort is not a direct migration parameter, but oxygen ingress creates oxidative degradation products in the sealant and adhesive that can alter sensory quality and increase the extractable burden. In halogen-free systems, the removal of polyvinylidene chloride means that PA6 and EVOH must be combined in a validated layer sequence; a structure of PET 12 µm / PA6 15 µm / EVOH 12 µm / CPP 70 µm is used where very low oxygen ingress is required, but interfacial adhesion between EVOH and PA6 under saturated steam requires retortable tie resins and precise moisture management.

When Retort Temperature Exceeds 121 °C, Hydrolytic Degradation Accelerates

Under saturated steam at 135 °C, the migration risk profile differs from a standard 121 °C cycle because the rate of PA6 hydrolysis increases sharply with temperature and because the partial pressure of steam raises the water concentration at the PA6 layer surface. Published kinetic data for polyamide hydrolysis describe an Arrhenius activation energy in the range of 70 kJ/mol to 90 kJ/mol; when the retort temperature rises from 121 °C to 135 °C, this activation energy predicts a relative increase in hydrolytic chain scission of roughly one order of magnitude, although the exact value depends on moisture concentration, film crystallinity, and the presence of acid or base residues. The degradation products include caprolactam and linear and cyclic oligomers with molecular weights up to several thousand grams per mole; the oligomers have lower diffusion coefficients than caprolactam but can still migrate into food simulants and are often detected by LC-MS/MS after solvent extraction. In a multilayer structure, hydrolysis of PA6 under 135 °C retort is aggravated if the CPP sealant layer is too thin to buffer steam contact or if the pouch experiences rapid pressure drops during cooling because steam condenses inside the laminate and creates localised water pockets. The residence time at 135 °C is typically limited to 10 min or less, but the threshold for property loss is sensitive: PA6 film with a relative viscosity below 3.2 after retort can exhibit a loss of tensile strength above 25 % and an increase in specific migration of caprolactam above the EU Regulation 10/2011 limit when the sealant layer is also compromised. Retort processors measure come-up time, hold time, and cooling rate with calibrated sensors and must validate that the coldest pouch in the load achieves the target lethality without over-processing the PA6 layer. The halogen-free constraint means that acid scavengers and thermal stabilisers added to PA6 must not contain chlorinated species; phosphite-based processing stabilisers are halogen-free but their hydrolysis products can include phosphorous acids, which may accelerate amide hydrolysis if the layer is not adequately stabilised. Batch-to-batch variation in PA6 film from different casting and orientation lines is a production-scale concern: films with the same product code can differ in extractable content by 20 % to 30 % after 135 °C retort because of subtle differences in heat-setting temperature, line speed, and monomer stripping efficiency.

A production-scale coextrusion line processing halogen-free PA6 barrier or structural layers must address moisture content before the extruder throat because residual water above 0.06 % by mass will hydrolyse the polymer during plastication and increase the caprolactam concentration in the finished film. Desiccant dryers with a dew point below −40 °C and a drying temperature of 80 °C for 4 h to 6 h are used for film-grade PA6, while longer drying at higher temperatures risks yellowing and premature degradation; the dried resin is conveyed in dry air to a twin-screw extruder with an L/D ratio of 40:1 or higher, and the vacuum vent is operated at a negative pressure between −0.08 MPa and −0.095 MPa to strip residual monomer and moisture. The extruder barrel temperature profile begins near 240 °C at the feed zone and is held below 260 °C at the die to avoid thermal depolymerisation; die pressure and melt temperature are monitored continuously because a temperature excursion above 265 °C for more than 2 min can increase residual caprolactam by 0.05 wt% to 0.10 wt% in a single lot. In cast PA6 film extrusion for extrusion lamination, the quenching drum temperature is set between 20 °C and 40 °C to control crystal growth and limit migration; with biaxially oriented PA6 film, sequential stretching at draw ratios from 2.8 to 3.4 in machine direction and 2.8 to 3.6 in transverse direction is used to orient the amorphous phase and increase crystallinity. Orientation and heat-setting at temperatures between 180 °C and 220 °C reduce the mobile amorphous fraction and lower the effective diffusion coefficient for caprolactam, but excessive heat-setting can embrittle the film and reduce post-retort flex-crack resistance. Nucleating agents such as talc or treated silica at loadings below 1.0 wt% can refine crystalline morphology and reduce oligomer extraction, but these additives must be selected from halogen-free and food-contact-listed grades to avoid introducing their own migration problems. Silica with a BET surface area above 200 m²/g can absorb moisture and caprolactam, but it can also agglomerate and create gel counts if the twin-screw screw configuration does not include adequate distributive mixing elements.

Halogen-Free Additive Packages and Their Migration Boundaries

For retort-grade PA6, the designation halogen-free refers not to the polymer backbone but to the absence of chlorinated paraffins, brominated flame retardants, polyvinylidene chloride, and chlorinated solvents in the formulation and converting process. Low-migration additive packages for retort-grade PA6 therefore rely on high-molecular-weight phenolic antioxidants, phosphite stabilisers, acid scavengers, and non-migratory slip and antiblock agents. Erucamide slip is effective at loadings from 0.05 wt% to 0.15 wt% but its molecular weight of 335 g/mol and low melting point allow it to migrate to the film surface and into fatty simulants; for low-migration structures, erucamide is replaced or supplemented by high-molecular-weight polysiloxane-based slip additives or by modified polypropylene waxes that have lower diffusion rates. The use of silicone-based slip in a retort pouch must be validated for heat-seal compatibility because surface silicone can lower the heat-seal strength of the CPP sealant if transferred to the seal surface. Antioxidant selection follows the same logic: pentaerythritol tetrakis(3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionate) and tris(2,4-di-tert-butylphenyl) phosphite are halogen-free and listed for food-contact use in the EU and FDA, but their migration into fatty simulants must be evaluated by GC-MS or HPLC because their molecular weights and partition coefficients differ. The antioxidant loading in PA6 film for retort is often in the range of 0.1 wt% to 0.3 wt%, and residual phosphite oxidation products can be detected after retort if the additive is hydrolysed by steam; hydrolysis of the phosphite produces non-volatile organophosphorus species that may increase overall migration. Halogen-free acid scavengers such as hydrotalcite or calcium stearate are used at low loadings, typically below 0.1 wt%, to neutralise catalyst residues and reduce amide hydrolysis, but these additives can themselves contribute to inorganic extractables if not fully encapsulated in the polymer. The selection of an additive package therefore involves a multi-factor migration boundary: increasing antioxidant loading reduces oxidative degradation but increases the pool of potential migrants, while reducing additive loading improves migration compliance but may sacrifice thermal stability during repeated retorting. Analytical verification of an additive package requires a contamination screen by headspace GC-MS, liquid injection GC-MS, and LC-QTOF after the laminate is exposed to the retort cycle and the food simulants specified in EU Regulation 10/2011.

Because the cast polypropylene sealant layer is the food-contact layer in most retort pouches, its own migration profile and thickness determine the degree to which PA6-derived substances reach the food. Polypropylene homopolymer or random copolymer films for retort applications are typically formulated with low extractables, and food-contact compliance is established under FDA 21 CFR 177.1520 and EU Regulation 10/2011. A 70 µm CPP layer provides a substantial diffusive lag time for caprolactam and larger oligomers, but the lag time is not infinite; at 121 °C, the diffusion coefficient of low-molecular-weight migrants in polypropylene is high enough that a detectable amount of caprolactam can cross the sealant within the first hour. The layer-to-layer partition coefficient between PA6 and PP also matters: caprolactam is more soluble in the polar polyamide phase, which retards its transfer into polypropylene, while non-polar slip additives partition preferentially into PP. If the CPP layer is replaced by a thinner 50 µm sealant to reduce cost, migration of PA6 constituents into food simulants can increase by more than the thickness ratio would suggest because the diffusive lag time scales with the square of thickness. Retort pouches that use a multilayer film with PA6 as the direct food-contact surface are uncommon and require explicit migration testing because caprolactam solubility in aqueous simulants is high and the specific migration limit of 15 mg/kg can be approached if the PA6 layer is thick or the retort cycle is aggressive. The converting operation can also introduce migrants from printing inks and solvents; in a halogen-free structure, ketone- or acetate-based lamination solvents must be purged to low residual levels, and reverse-printed retort inks are formulated with high-molecular-weight binders to resist extraction. The absence of a chlorinated barrier means that any residual solvent in the printed outer PET layer can more readily partition through the laminate under steam pressure, so heat-seal strength and migrant screening should be performed on the finished pouch according to ASTM F88/F88M-21 and the extraction protocols of EN 1186-1:2002.

Does the CPP Sealant Layer Function as a Barrier to PA6 Migrants?

Whether a 70 µm cast polypropylene sealant layer functions as a barrier to PA6 migrants under retort depends on the migrant identity, the retort time–temperature profile, and the partition coefficient between PA6 and polypropylene. For caprolactam with a molecular weight of 113.16 g/mol and high water solubility, polypropylene is a poor barrier at 121 °C because the polymer is above its glass transition temperature and the amorphous phase is highly mobile; migration tests on PA6/PP laminates show that caprolactam is detectable in aqueous food simulants after retort even when the PA6 layer is not the direct food-contact layer. For cyclic dimer and larger oligomers, the CPP layer provides a greater lag time because molecular weight reduces the diffusion coefficient in polypropylene, but the high retort temperature still accelerates diffusion by several orders of magnitude compared with ambient storage. The functional barrier concept under EU Regulation 10/2011 applies only when the barrier layer reduces migration below the detection limit; a single CPP sealant is rarely accepted as a functional barrier at retort temperatures for low-molecular-weight migrants unless the PA6 has very low residual monomer and the retort cycle is minimised. For high-risk PA6 migrants, manufacturers often use a combination of low-residual PA6 film, a thicker CPP layer, and a barrier interlayer of EVOH or aluminium-containing film to achieve compliance. Measurement of actual migration requires the full laminate, not separate layers, because layer interactions during retort, including plasticisation, delamination, and adhesive splitting, change the effective diffusion path. Analytical results from monolayer PA6 tests cannot be divided by the number of layers to estimate a multilayer result because partition coefficients and lag times are non-linear. When a pouch is retorted repeatedly or is held under pressure for longer than the target cycle, the CPP sealant can undergo dimensional relaxation and allow steam channels to form at the interface, creating preferential pathways for PA6-derived migrants. Production-scale retort authorities usually require migration testing at worst-case retort load, with the pouch in contact with the food simulant under pressure and with the seal area included in the extraction; results are then reported per EU Regulation 10/2011 Article 12 or per FDA 21 CFR 177.1395 for high-temperature laminates. Published data for halogen-free PA6/CPP laminates under 135 °C retort remain limited, so regulatory acceptance typically relies on case-by-case migration studies rather than generic industry assumptions.