Off Spec Lldpe Handling When To Reject Blend Or Return
Linear low-density polyethylene (LLDPE) is an ethylene-α-olefin copolymer with a short-chain branching architecture that depresses density and crystallinity relative to high-density polyethylene. Prime-grade LLDPE specifications are defined by density, melt index, comonomer type, gel content, ash, and additive package. Density is measured by ISO 1183-1:2019 or ASTM D792-20; melt index is determined at 190 °C/2.16 kg using ISO 1133-1:2022 or ASTM D1238-20. Commercial film-grade LLDPE typically falls between 0.915 g/cm³ and 0.940 g/cm³ density and between 0.5 g/10 min and 2.0 g/10 min MI₂. A lot is considered off-spec when any certifiable parameter falls outside the grade’s agreed tolerance, but disposition is a separate engineering calculation. Off-spec LLDPE handling—when to reject, blend, or return—depends on whether the out-of-range parameter can be diluted, whether the deviation indicates a molecular architecture problem rather than a simple flow shift, and whether the material has been contaminated post-reactor.
In continuous polymerisation processes, off-spec LLDPE is generated during catalyst transitions, comonomer feed disruptions, bed-level excursions, and emergency stop events. On a gas-phase fluidised bed line with a bed residence time of 2 h to 4 h, a grade change can produce transition material for several bed turnovers. This material often shows a bimodal comonomer distribution rather than a uniform density deviation. Low-density fractions depress stiffness and film modulus; high-density fractions raise haze and reduce dart impact. The technical decision should therefore use both bulk density and comonomer distribution data. A single density reading of 0.911 g/cm³ on a nominal 0.918 g/cm³ LLDPE does not automatically indicate return; it may be fully salvageable by blending if gel count and MI remain within secondary limits.
Can melt-index deviation alone trigger rejection, or should comonomer distribution govern the decision?
Melt-index deviation alone can trigger rejection when the logarithmic blending rule cannot restore the target MI without exceeding the maximum permitted off-spec fraction on a given line. For a bin-blend of two LLDPE lots, the blend MI₂ follows logarithmic additivity: log(MIblend) = woff log(MIoff) + wprime log(MIprime), where w is mass fraction. Density, by contrast, is linear with mass fraction. A lot with an MI₂ of 3.2 g/10 min against a prime target of 1.0 g/10 min requires a prime fraction of 0.79 to return to 1.5 g/10 min, assuming a prime MI₂ of 1.0 g/10 min. The calculation places the off-spec addition near 21 wt%, which is acceptable only if the film line can tolerate the resulting comonomer heterogeneity. Polymerisation records from metallocene-catalysed LLDPE lots with MI excursions above 2.5× nominal frequently show a depleted high-molecular-weight tail and reduced melt elasticity. Such a lot may still be used in injection moulding but should not be returned as film-grade solely on the basis of MI correction.
On a 90 mm single-screw grooved-feed extruder with 30:1 L/D and a 250 mm blown-film die, replacement of 30 wt% prime LLDPE with off-spec lot having MI₂ 1.8 g/10 min instead of 1.0 g/10 min reduces die pressure and can lower the bubble frost-line height, increasing gauge variability. The same substitution in a cast-film line with a flat die and chill-roll air knife has less effect on bubble stability but may reduce draw resonance. Prime-grade blending should be validated on a small-scale ribbon blender or silo homogenizer before full silo blending. The lower-bound blend ratio is determined by extruder pressure, while the upper-bound ratio is determined by dart impact and tear propagation in oriented film.
Blend window calculation for single-screw film lines
The blend window for off-spec LLDPE is bounded by four parameters: melt-index mismatch, density mismatch, gel population, and comonomer type. Butene-based LLDPE blended into hexene-based film beyond 15 wt% often produces a measurable loss in Elmendorf tear resistance even when density remains unchanged; the difference arises from tie-molecule population distribution, not from bulk density. For non-critical geomembrane or industrial liner applications, this tear loss may be acceptable. For high-speed food-packaging film with a 25 µm target thickness, the same blend may fail ASTM D1922-15 tear values and ASTM D1709-16a dart impact. Thus a blend ratio can be technically on-spec for MI and density but still be unusable in the intended converting operation.
| Parameter | Test method | Return to supplier | Blend only if | Reject |
|---|---|---|---|---|
| Melt index MI₂ | ISO 1133-1:2022 / ASTM D1238-20 | Batch average outside contractual temperature-dependent tolerance with no post-reactor cause | MI₂ ≤ 5.0 g/10 min for thick sheet or injection moulding; film uses ≤ 2.5 g/10 min | MI₂ < 0.10 g/10 min due to excessive viscosity or > 10 g/10 min for film-grade blending |
| Density | ISO 1183-1:2019 / ASTM D792-20 | COA dispute; density below 0.905 g/cm³ or above 0.950 g/cm³ outside LLDPE class | Δρ ≤ 0.003 g/cm³ from prime for film; ≤ 0.006 g/cm³ for injection moulding | Density deviation > 0.008 g/cm³ from prime target |
| Gel count > 75 µm | Film inspection with image analyzer per ISO 18553:2002 adapted to LLDPE | Gels indicate crosslinked or degraded fraction; lot not returnable if contamination is post-reactor | ≤ 150 particles/1000 cm² in 25 µm film for general-purpose film; ≤ 50 particles/1000 cm² for high-clarity film | > 500 particles/1000 cm² in 25 µm film or visible gels in 25 µm blown film |
| Ash | ISO 3451-1:2019 / ASTM D5630-13 | Residual catalyst or additive error > 0.15 wt% with no customer-induced contamination | ≤ 0.08 wt% after dilution for film; ≤ 0.20 wt% for non-critical moulding | > 0.25 wt% after dilution, indicating filler or metal contamination |
| Oxidative induction time | ISO 11357-6:2018 at 200 °C | OIT < 10 min indicates additive underdosing in original lot | OIT > 15 min after blending with additive masterbatch | OIT < 5 min because processing stabilizers are depleted |
The values in the table are screening thresholds, not universal dispositions. For critical film, food-contact, or medical applications, each blend must be re-qualified on the intended line because comonomer distribution and stabilizer depletion are not captured by bulk MI and density alone.
Return to supplier is contractually viable only when the off-spec condition is identified before the lot is transferred from a dedicated silo, the original package seal is intact, and no post-shipment blending has occurred. Once the resin is conveyed into a plant silo or combined with a dust-control additive, the supplier cannot identify the root cause from retained samples because contamination history is lost. In those conditions, return is usually refused; the lot is dispositioned as reject or controlled blend. For railcar shipments of LLDPE, retained sample splits are taken from three horizontal zones of the railcar; top-to-bottom density segregation is common in bimodal LLDPE due to pellet mass differences. The difference between top and bottom density can reach 0.002 g/cm³ if pellet temperature during loading varies by more than 20 °C. A top-silo composite certificate of analysis is therefore insufficient for return decisions.
When off-spec LLDPE is diverted to food-contact packaging, the overall migration limit of 10 mg/dm² under EU Regulation 10/2011 and the extraction limits in FDA 21 CFR 177.1520 must be retested on the finished blend. Blending off-spec with prime does not automatically inherit the prime food-contact status, particularly if the off-spec lot contains unknown oxidative by-products or higher oligomer fractions. Additive incompatibility should also be evaluated: erucamide-loaded film-grade LLDPE diverted into injection moulding can cause screw slip, plate-out, and inconsistent densification under high back pressure.
When gel count exceeds 75 µm in thin-film grades, blending is seldom viable
Gels in LLDPE film originate from incompletely melted high-molecular-weight species, crosslinked particles, oxidized resin accumulations on die lips, and catalyst support agglomerates. In a blown-film line running 25 µm monolayer at 120 m/min line speed, a gel population above 75 µm can generate optical defects and weak spots. Blending off-spec gel-containing LLDPE with clean prime at 10 wt% does not reduce the number of gels per square meter; it dilutes only the concentration. The film defect density remains proportional to the gel concentration divided by the blend ratio. For a lot with 300 gels/1000 cm² in 25 µm film, a 90% prime blend reduces the average to 30 gels/1000 cm², which may satisfy industrial film but not hygiene-film specifications. Published data for the exact gel threshold in cast film using butene-based LLDPE is limited; inline validation is therefore mandatory before accepting a gel-containing lot for film conversion.
A screen pack with 120 mesh or finer is often added ahead of the die for such trials, and pressure drop across the screen pack should be monitored. A rise above 3 MPa above baseline indicates that gel particles are not melting or passing through; continued operation risks screw wear and screen rupture. For critical thin-film applications, lots with gel counts above 150 particles/1000 cm² at 75 µm should be rejected or downgraded to thick injection-moulding flow lines rather than blended.
Off-spec LLDPE with low MI, meaning high melt viscosity, is often diverted to injection moulding of crates, caps, or housewares, but only if density remains below 0.940 g/cm³ and MI₂ under 190 °C/2.16 kg is above 1.0 g/10 min. On an injection moulding machine with a 350 kN clamp force and a single screw of 22:1 L/D, a lot with MI₂ 1.2 g/10 min may fill a multi-cavity tool adequately at 220 °C melt temperature; the same lot at 0.6 g/10 min raises injection pressure and short-shot risk in sections below 1.5 mm. The use of off-spec film-grade LLDPE in injection moulding is governed by short-shot testing according to ISO 294-1:2017 and tensile testing according to ISO 527-2:2012. Without such validation, transfer to a different conversion process is not a guaranteed disposition; it is a new specification exercise.
