Caprolactam supplied to high-tenacity PA6 filament polymerization units is constrained less by melt colour than by reactive impurities that shorten the amide chain before the draw zone. Fibre-grade monomer under GB/T 13254 is controlled for purity above 99.90% by gas chromatography, for water below 0.05% by mass, for volatile bases below 0.5 mmol/kg, and for UV absorbance at 290 nm below 0.05 absorbance units, because unsaturation and water shift the relative viscosity during continuous two-stage VK tube polymerization. The monomer is mixed with demineralized water and polymerized in a first stage at 240–270°C, then finished under vacuum at 1–5 kPa in the second stage to strip unreacted caprolactam to 0.2–0.6 wt% in the chip. The resulting chip shows a relative viscosity of 3.3–3.8 measured as a 1% solution in 96% sulfuric acid per ISO 307. Chip is crystallized in a desiccant dryer at 80–120°C with a dew point below -40°C, reducing moisture below 0.04% before melt extrusion. Spinning on single-screw extruders with 30:1 L/D and melt temperatures of 265–295°C uses spin pack filtration through 20–40 μm sintered metal media to remove gels that would otherwise initiate filament breaks under high draw ratios. Undrawn yarn is taken up at 400–900 m/min onto heated godets and drawn in 3–5 stages with roll temperatures from 150°C to 220°C at a total draw ratio of 4.0–5.5. Heat setting at 200–230°C under a controlled relaxed overfeed of 2–5% stabilizes crystallinity and reduces hot-air shrinkage to 4–7% measured at 190°C per ASTM D885. Final tenacity values of 800–950 mN/tex and elongation at break of 15–20% are obtained per ISO 2062, which meets the dip-processing requirements for tire cord. For tire cord, the greige cord is dipped in resorcinol-formaldehyde-latex adhesive at a dry pickup of 4–6% and cured in multi-zone ovens at 150–230°C; line tension during dip treatment is held between 0.5 cN/tex and 1.0 cN/tex to avoid modulus loss. End-use failure modes on production lines include fibre fusion in the spin pack when monomer moisture control drifts above 0.05%, and dip adhesion loss when heat-setting temperatures fall below 200°C.

What Limits Hydrolytic Stability in Underhood PA6 Components?

Injection-molding grades based on caprolactam-derived PA6 for underhood components are specified around the 2.8–3.2 relative viscosity window (ISO 307); glass-fibre-reinforced formulations typically contain 30 wt% chopped glass. Unfilled grades show melt flow rates of 15–30 g/10 min at 275°C under 5 kg load per ISO 1133-1:2022. Pellets are dried in closed-loop desiccant systems at 80°C to water content below 0.12% as verified by ISO 15512; exceeding 0.20% triggers barrel hydrolysis that reduces melt viscosity and produces silver streaks on moulded surfaces. Injection moulding uses screw compression ratios of 2.5:1 to 3.0:1, melt temperatures of 250–280°C, and mould temperatures of 80–90°C; multi-cavity tools for intake manifolds, rocker covers, and radiator end tanks commonly require clamp forces between 4,000 kN and 8,000 kN. The critical long-term failure criterion is not dry strength but hydrolytic stability in hot ethylene glycol/water coolant. Copper halide stabilizer packages are introduced at 30–150 ppm Cu with a halogen synergist; unstabilized or poorly dispersed GF30 PA6 exposed to 50 vol% glycol/water at 120°C shows tensile strength retention below 60% after 1000 h in published comparative studies. The mechanism is acid-catalyzed chain scission at the amide bond, accelerated by carboxylic acid end groups; the processing response is therefore controlled end-group balancing during polymerization of caprolactam, not merely external additive loading.

Representative dry-as-molded and conditioned values for GF30 PA6 specimens prepared to ISO 294-1 are summarized below. The difference between dry and conditioned properties is a supply-chain specification issue: parts validated on dry specimens can lose more than 30% of flexural modulus after moisture uptake in service.

Cast Nylon Polymerization Moisture Thresholds and Demolding Windows

Monomer casting of caprolactam into large PA6 stock shapes begins with flakes melted at 80–100°C in a jacketed vessel fitted with vacuum take-off. Absolute pressure is reduced to 0.5–1.5 kPa for 30–60 min; moisture above 100 ppm by Karl Fischer titration (ISO 15512) is sufficient to terminate anionic active centres before conversion reaches a load-bearing molecular weight. The dried melt is transferred through a 20 μm filter to remove insoluble particles that would otherwise nucleate local crystallization. Sodium caprolactamate catalyst is formed in situ from sodium hydride or sodium metal at 0.2–0.8 mol%, and an acyl-type activator such as N-acetylcaprolactam or blocked diisocyanate is added at 0.2–0.6 mol%. The mixed melt is cast into preheated moulds at 150–175°C. Part thickness controls thermal history more than any other variable. In sections below 20 mm, polymerization reaches near-complete conversion within 20–40 min, and demolding can proceed once the core cools below 140°C. In sections above 60 mm, the exotherm drives the centre to 190–210°C; without external water channels or mould rotation, cavitation and shrinkage voids form because crystallization and monomer evaporation compete with chain growth. Production-scale shops therefore apply staged cooling at 1–2°C/min from mould temperature to 80°C before demolding. Finished cast PA6 from caprolactam typically exhibits tensile strength of 75–85 MPa (ISO 527-2), elongation at break above 20%, and Shore D hardness of 75–85 (ISO 868). Saturation water absorption reaches 8–9% (ISO 62), which must be reflected in bearing clearance design; a cast gear blank machined dry will close running clearances after exposure to ambient humidity.

The processing conflict in thick sections is that the same low-moisture condition required for catalyst survival also increases melt viscosity and reduces heat transfer. Moulders compensate by preheating caprolactam to 90°C before catalyst addition, but residence time after activation must remain below 5–10 min or premature gelation occurs in transfer lines. Large rotating equipment parts such as gears, sheaves, and wear shoes are produced by this route because melt-casting permits section thicknesses not practical in injection moulding, but the maximum practical thickness is bounded by the adiabatic temperature rise. Published data for the exact void-free thickness limit in a particular mould configuration are limited; operators validate each new geometry with thermocouple arrays and cut-section inspection before serial production.

Film-grade polyamide 6 converted from caprolactam at relative viscosity 3.6–4.2 (ISO 307) enters a flat-die cast line at 245–265°C and is quenched on a polished chill roll maintained at 12–18°C. The quench target is to cool the melt below the glass transition before spherulitic crystallinity exceeds 5%; higher cast-sheet crystallinity creates thickness bands during transverse stretching and reduces gauge uniformity. Electrostatic pinning is used with quench air at a dew point below -20°C to prevent moisture pickup and die-lip condensation. Machine-direction orientation is applied at 3.0–3.5:1 on rolls heated to 70–90°C, followed by transverse orientation at 3.5–4.0:1 in a tenter frame at 110–130°C. Heat setting under 5–8% transverse relaxation at 200–210°C reduces free shrinkage in boiling water to below 3%. A 15 μm biaxially oriented PA6 film tested at 23°C, 50% RH shows oxygen transmission of 30–50 cm³/(m²·d·bar) per ASTM D3985, and water vapour transmission of 200–300 g/(m²·d) at 38°C, 90% RH per ASTM F1249. The high water vapour transmission relative to polyolefins is a defined limitation; PA6 film alone is not used for high-moisture food packaging unless laminated with polyethylene or polypropylene. Corona treatment is set to 42–46 mN/m wetting tension for lamination and printing. Food-contact assignments require compliance with Regulation (EU) No 10/2011; the specific migration limit for caprolactam is set at 15 mg/kg food simulant, and the resin falls under FDA 21 CFR 177.1500 for nylon resins in repeat-use and certain single-use articles.

The main process conflict on high-speed BOPA lines is moisture control. Since PA6 absorbs ambient moisture within minutes at relative humidity above 60%, edge trim and off-spec film re-feed require closed-loop handling with dry-air conveying at a dew point below -30°C. If recycled flake is fed at more than 10–15% without viscosity adjustment, die pressure varies and the tenter window narrows. Slitting and winding tensions below 0.3 N/mm² prevent blocking at the mill core, and roll hardness is checked with a durometer to maintain 85–95 Shore A.

When Caprolactam Feeds Carpet Staple Fibre Lines, Crimp and Spin Finish Chemistry Set the Processing Window

Staple fibre extrusion lines using caprolactam-derived PA6 chips with relative viscosity 2.5–2.9 (ISO 307) operate at melt temperatures of 255–280°C and spinneret hole diameters of 0.25–0.35 mm. Quench air at 18–22°C and 0.3–0.6 m/s cools the filaments, which are taken up at 600–1,500 m/min. Drawing is carried out in one or two stages to a total ratio of 3.0–4.0, followed by steam crimping at 8–15 crimps per 25 mm and dry heat setting at 190–210°C. Residual caprolactam extraction on the tow line uses hot water at 80–95°C; final oligomer content below 0.3 wt% is required to prevent yellowing during later latex backing or piece dyeing. Spin finish type affects fibre-to-fibre friction and tufting performance. A low-viscosity lubricant with antistatic properties is applied at 0.3–0.6% by mass; overdosing above 0.8% creates roll wrap and crimp loss. Carpet yarns of 1200–2600 dtex are heat-set in loops; linear shrinkage must remain below 4% at 160°C after 10 min to preserve tuft definition. In nonwoven covers, compatibility with polyolefin backing and calender bonding at 150–190°C determines the use of coarser denier.

Dye uptake is controlled by amine end group content in the caprolactam-derived polymer. Acid dyes used for nylon carpets require uniform amine ends typically in the 40–60 mmol/kg range to avoid barré after atmospheric dyeing. Deep-dye and light-dye variants are produced by shifting amine end group content during polymerization; batch-to-batch variation above 5 mmol/kg is visible as streakiness in cut-pile carpets under daylight. This is not a generic mixing issue; it is a polymer-chain-level specification that determines shade reproducibility and rework rates on dyehouse jigs.

PA6/Polyolefin Alloy Rheology at the Weld Line

Alloying caprolactam-derived PA6 with polypropylene using maleic anhydride-grafted PP as compatibilizer is carried out in a co-rotating twin-screw extruder with L/D 40:1 and a barrel profile from 230°C to 270°C. Screw speeds of 300–500 rpm are used to generate a phase morphology in which the polyolefin domain size remains below 1 µm; throughput above 500 kg/h without side-feeder capacity shifts the domain size upward and produces visible flow lines at the injected part surface. The compound is filtered through a 20–40 μm screen pack before pelletizing. Drying of finished alloy pellets at 80°C to moisture below 0.10% (ISO 15512) is mandatory before molding; un-dried alloy pellets cause melt instability at the nozzle and reduce weld-line strength. A representative comparison for a 70/30 wt% PA6/PP alloy is provided in the table. The notched impact at -30°C improves over neat PA6, while tensile strength and stiffness decrease. The dominant processing defect is not gross phase separation but weak weld lines. The weld-line tensile strength of PA6/PP alloys can be 40–50% lower than bulk strength because interfacial reaction at the flow front is incomplete; this is quantified on double-gated specimens per ISO 527-2.

The weld-line deficit is addressed partly by increasing mould temperature to 90–110°C and partly by reducing pellet moisture below 0.08%; both measures lower melt viscosity at the flow front and allow the dispersed PP phase to coalesce less. However, increasing mould temperature above 110°C extends cycle time beyond the window acceptable for high-volume connector and clips production. Published data for the exact weld-line improvement in a specific multi-cavity hot-runner tool are limited; molders validate each tool with short-shot studies and thermocouple-instrumented inserts before committing to a processing window.