Handling a POM Batch Deviation: Disposition Without Losing the Lot

Polyoxymethylene (POM) is a semi-crystalline engineering thermoplastic with a density of 1.41–1.42 g/cm³ and a melting temperature between 165 °C and 183 °C depending on copolymer or homopolymer structure. A batch deviation in POM typically appears as a shift in melt volume-flow rate (MVR) measured in accordance with ISO 1133-1:2022 at 190 °C and 2.16 kg, as an increase in volatile condensables after drying, or as an unexplained change in notched impact. When a lot is flagged, disposition is initiated by reproducing the out-of-specification result, comparing the retained reference sample, and isolating the failure mode before any downgrade is applied. Automatic rejection is avoided because a deviated POM lot can still be assigned to a less demanding application, blended into virgin resin, or reprocessed into controlled regrind if the degradation vector is understood. The disposition criterion is therefore the measured property profile against the end-use requirement, not the original grade specification alone.

What Does a Melt Flow Rate Shift Reveal About Molecular Weight Distribution?

A shift in MFR above the grade specification—for example from 9 g/10 min to 13 g/10 min—indicates a reduction in average molecular weight. Capillary rheometry in accordance with ISO 11443:2014 is used to compare apparent shear viscosity at 190 °C and 1,000 s⁻¹. For a POM copolymer with a nominal MFR of 9 g/10 min, apparent shear viscosity at 1,000 s⁻¹ is typically in the range 50–150 Pa·s; an increase in MFR to 13 g/10 min moves the viscosity toward the lower end of this range. The exact value is grade-dependent and is confirmed on a capillary rheometer before disposition. The viscosity reduction changes pressure-limited mold filling and cushion consistency, but does not by itself render the lot non-conforming. The lot is retested after moisture removal because water can depress viscosity through plasticization, creating a false MVR elevation. Melt density is calculated from the piston displacement and the known melt density at 190 °C, which for POM is approximately 1.19–1.20 g/cm³; this conversion is necessary when comparing MFR and MVR values.

MVR and MFR are not interchangeable without density correction. ISO 1133-1:2022 reports MVR in cm³/10 min; MFR in g/10 min is derived by multiplying MVR by the melt density at test temperature. A retained reference sample is always tested in the same melt indexer to remove instrument offset. Molecular weight changes are also confirmed by dilute solution viscosity, but this method is slower and is reserved for dispute resolution. A rise of 15–30% in MFR above the grade target is commonly managed by blending the deviated lot with a low-MFR POM at a ratio of 20:80 to 40:60. The blending ratio is determined by the measured MFR, not by a fixed schedule, because POM viscosity changes do not scale linearly across the entire shear rate range. The final blend is re-qualified against the same standard before release.

On a production floor, a deviating POM lot often presents first as screw recovery time drift or nozzle drool. Injection molding machines with 80–120 t clamp force have recorded screw recovery time changes of 10–25% when MFR shifts by more than 30%, because lower melt viscosity reduces pressure feedback and alters screw slip. The lot is sampled from three zones of the gaylord or silo; a retained reference pellet sample is run alongside the deviation under identical conditions. If the deviation is confined to one packaging lot and no contamination is observed, blending with virgin material at a ratio of 20:80 is evaluated before downgrade. Moisture content by ISO 15512:2019 is measured on the retained material; levels above 0.10 wt% require drying at 80 °C for 3–4 h in a desiccant dryer with a dew point below −30 °C. If the MVR remains elevated after drying, the root cause is not surface moisture but hydrolytic or thermal chain scission. The same dryer is not used for materials with hygroscopic polyesters to avoid cross-contamination.

Thermal Stability and Formaldehyde Evolution During Reprocessing

POM degradation is autocatalytic when formaldehyde accumulates. Reprocessing a thermally deviated lot on a 25 mm co-rotating twin-screw extruder with an L/D ratio of 40:1 and zone temperatures of 170–200 °C can trigger additional chain scission if the vent port vacuum is above −0.08 MPa. Differential scanning calorimetry in accordance with ISO 11357-1:2016 detects shifts in crystallization temperature; a reduction in Tc from 146 °C to 138 °C is often associated with molecular weight loss. Thermogravimetric analysis per ISO 11358-1:2014 is used to measure the onset decomposition temperature; stabilized POM typically shows onset degradation between 280 °C and 310 °C. A lower onset or a mass loss step between 200 °C and 250 °C indicates contamination or extensive chain scission.

Volatile by-products are detected as a pressure drop increase across a 40 µm melt filter on a gear-pump-assisted extruder. If the pressure differential exceeds 0.5 MPa during regrind processing, the vent capacity is insufficient to strip formaldehyde, and the material should not be reprocessed further. In many cases, re-extrusion with an acid scavenger masterbatch at 1–2 wt% reduces free formaldehyde and allows reuse; however, published data for this specific configuration is limited, and the exact dosage must be confirmed in a laboratory twin-screw run before a production batch is committed. Residence time should remain below 2 min, and the melt temperature at the die should not exceed 195 °C for copolymer POM.

Where a lot is borderline high in MFR, re-extrusion is performed at 15–25 kg/h with a vacuum of −0.09 MPa at the vent. After one pass, MVR is retested. If MVR returns to within ±10% of target, the lot is reclassified as controlled regrind and is not returned to the original specification. If degradation products remain volatile, further re-extrusion is discouraged because each additional pass reduces notched impact strength. Mechanical testing on the regrind is mandatory for any load-bearing application. Tensile specimens are molded in accordance with ISO 527-2:2012 and tested at 23 °C and 50% RH. A POM copolymer with a nominal tensile modulus of 2600 MPa can fall to 2400 MPa after one additional heat history; a drop greater than 10% in tensile elongation at break is used as a disposition boundary for high-load components. Notched Charpy impact per ISO 179-1:2010 is measured; typical values for unfilled POM range from 5.5 kJ/m² to 8.0 kJ/m². If the measured impact is below 4.0 kJ/m², the lot is not assigned to snap-fit assemblies or safety-critical clips. The regrind is then directed to non-load-bearing spacers, guides, or housings where the reduced impact is acceptable.

When Shrinkage and Crystallization Deviate Together

Lot deviation may also appear as a change in post-molding shrinkage. POM is highly crystalline; mold shrinkage measured per ASTM D955-18 typically ranges from 1.8% to 2.5% for standard specimens. A shift in crystallization rate caused by nucleating agent loss or thermal degradation can increase shrinkage anisotropy. Molded parts are measured after 24 h and 48 h under ISO 291:2018 standard atmosphere; a deviation greater than 0.2% compared with the reference lot is investigated. If the dimensional change is batch-wise and not accompanied by an MVR shift, the root cause is often mold temperature variation or packing pressure, not the resin itself. The lot remains assignable to applications with wide dimensional tolerance after the molding parameters are returned to the validated window: melt temperature 210–220 °C, mold temperature 80–100 °C, and packing pressure 50–70 MPa.

If the deviation includes both MVR increase and shrinkage change, the lot is tested for thermal stability by DSC before any assembly. The crystallization half-time at 140 °C is recorded; when it lengthens by more than 30% compared with the reference, the crystalline order is reduced and the lot is excluded from precision parts with a tolerance tighter than ±0.15 mm.

Acidic Contaminants and the Threshold for Irreversible Depolymerization

POM is incompatible with acidic or halogenated contaminants. Trace PVC contamination accelerates hydrolytic chain scission and can lower the onset decomposition temperature by 10–20 K compared with the clean reference. If a lot contains visible flecks or if the TGA trace shows an additional mass loss step between 200 °C and 250 °C, the material should not be blended into prime resin. Cross-contamination with polyamide or polyester creates visible delamination and cannot be corrected by reprocessing. The disposition path for this contaminated lot is segregation and specialized waste or energy recovery, not mechanical recycling. Where food-contact or medical classification is required, the relevant grade must meet 21 CFR 177.2470 or 21 CFR 177.2480 as applicable, or EU Regulation (EU) No 10/2011. A lot with unknown contamination history is not reassigned to these applications regardless of mechanical test results.

Infrared spectroscopy in attenuated total reflectance mode is used on microtomed sections. A carbonyl absorption at 1720 cm⁻¹ growing relative to the methylene C–H band indicates oxidation or contamination. Unexpected amide N–H absorptions near 1540 cm⁻¹ indicate polyamide contamination; ester carbonyl indicates polyester contamination. The detection limit for a second polymer by ATR-FTIR is approximately 2–5 wt%. Below this limit, a second melting endotherm in DSC is used. If a second endotherm is present above 220 °C, the lot is not approved for clear or light-colored parts.

At the molding machine, a deviated lot with MFR at the upper end requires a lower injection velocity and a longer cooling time to prevent flash and sink marks. For a 2.5 mm wall thickness, a holding pressure of 45–50 MPa may be sufficient instead of 60 MPa because lower viscosity shortens pressure transfer. The switch-over point is set by peak cavity pressure measured with a cavity pressure transducer, not by screw position, to maintain dimensional stability. If the process window is narrower than 10 °C for melt temperature, the lot is restricted to machines with closed-loop process control. When reclaim is added at 10–20%, the screw rotation speed is reduced by 10–15% to limit shear heating; the back pressure is kept below 0.5 MPa to avoid excessive residence time. This operational boundary determines whether the deviated lot can be molded without losing the lot.