What Happens When You Run the Wrong HDPE Grade: Sag, Warpage and ESCR Cracking

In an accumulator-head extrusion blow moulding cell producing a 19 L industrial jerrycan with a target mass of 52 g, the parison is extruded through a diverging die gap of 2.4 mm at a melt temperature of 185 °C. The specified blow moulding HDPE has a melt flow rate of 0.30 dg/10 min at 190 °C/2.16 kg per ISO 1133-1:2022, a high-load melt index at 21.6 kg of 9.0 dg/10 min, and a melt flow ratio of 30. When an injection moulding HDPE with a melt flow rate of 8.0 dg/10 min and a melt flow ratio below 18 is substituted, the parison elongates from 480 mm to 1 050 mm in 8.2 s. The high-molecular-weight tail is insufficient to resist extensional necking. The upper shoulder thins from 2.8 mm to 1.1 mm, while the bottom flash retains 3.0 mm. Top-load compression at a crosshead speed of 25 mm/min on a 5 kN universal testing machine produces buckling at 390 N, below the internal control range of 600 N to 650 N. The failure location corresponds to the thinnest wall section measured with a Hall-effect probe.

This is the practical consequence of selecting the wrong HDPE grade: sag, warpage and ESCR cracking. In the blow moulding case, sag is not caused by melt temperature drift but by the lower zero-shear viscosity of the injection grade. Die swell at the die exit also drops from 28 % to 18 %, so the parison diameter after swell is smaller even before axial draw-down. Extrusion head pressure falls from 22 MPa to 12 MPa at a screw speed of 35 min⁻¹, reducing specific energy input and changing the parison temperature distribution. The resulting container fails on stack height, wall thickness distribution, and dimensional stability.

Melt Tension Deficiency in Accumulator-Head Blow Moulding

Parison sag is governed primarily by the high-molecular-weight tail. For a 0.30 dg/10 min blow moulding HDPE, resin supplier rheology data widely place the zero-shear viscosity at 190 °C in the range of 0.8 × 10⁵ Pa·s to 2.0 × 10⁵ Pa·s. For an 8.0 dg/10 min unimodal injection grade, the value is typically one order of magnitude lower. Capillary rheometry with a haul-off attachment at 190 °C and draw acceleration of 12 mm/s² gives melt tension values of 0.15 N to 0.35 N for the blow moulding grade; the injection grade falls below 0.05 N. A laser curtain at the die exit records sag length between 1.0 s and 8.0 s of 85 mm for the specified grade and more than 260 mm for the wrong grade. A 25-point parison programmer can redistribute wall thickness, but it cannot replace the missing molecular weight.

The high-load melt index at 21.6 kg is not a substitute for direct melt strength measurement. A grade can meet a high-load melt index specification and still exhibit poor sag performance if the molecular weight distribution is narrow. In a single-layer 10 L container line, a lot with a high-load melt index of 8.5 dg/10 min but a melt flow ratio of 24 showed visible necking after 6.0 s; a second lot with the same high-load melt index and a melt flow ratio of 36 held parison length within 75 mm over the same interval. The difference was traced to the high-molecular-weight fraction above 10⁶ g/mol by gel permeation chromatography. Resin lot changes in blow moulding are therefore treated as a process event, not a simple drop-in.

On a 1.2 L rectangular injection moulded container tool with a 0.8 mm hinge section and a 2 000 kN hydraulic clamp, a 20 dg/10 min HDPE fills the cavity at a peak injection pressure of 85 MPa. The part cools in an ambient fixture at 23 °C and 50 % relative humidity. Plaque shrinkage measured per ISO 294-4 shows flow-direction shrinkage of 0.025 mm/mm and cross-flow shrinkage of 0.019 mm/mm; the differential of 0.006 mm/mm exceeds the internal limit of 0.004 mm/mm. Corner-to-corner warp on a granite surface plate is 2.6 mm. The part is assembled into a dispenser with a design tolerance of ±0.8 mm. Reducing hold pressure from 55 MPa to 35 MPa and raising mould coolant temperature from 15 °C to 40 °C lowers warp to 2.0 mm but does not eliminate it. A blow moulding grade with melt flow rate of 0.30 dg/10 min cannot fill the 0.8 mm hinge; an intermediate injection grade with MFR of 9 dg/10 min and density of 0.950 g/cm³ brings warp to 0.9 mm. The residual warpage is anisotropic crystallisation and orientation, not cooling alone.

Why Does Differential Shrinkage Persist After Mould Temperature Is Raised to 60 °C?

When the coolant set-point is raised to 60 °C, part surface temperature at ejection measured with a contact thermocouple increases from 61 °C to 74 °C, and the cycle time extends from 18 s to 26 s. Differential shrinkage remains directionally biased because the 20 dg/10 min injection grade has a narrow molecular weight distribution and rapid stress relaxation after filling. Gate-sealing time at the valve gate is 7.2 s for the 20 dg/10 min grade, compared with 9.4 s for the 9 dg/10 min grade. Early gate freeze reduces packing efficiency and leaves under-packed regions near the end of flow. Under polarised light microscopy, the flow-induced skin layer in the high-MFR part occupies 8 % of the wall thickness, while the core contains larger spherulites. This structure produces a frozen-in bending moment. Mould temperature adjustment above 60 °C is not sufficient because warpage is not solely a thermal phenomenon.

Warpage is further compounded by moisture and regrind. HDPE is not hygroscopic to the same extent as polyamide, but surface condensation on cold pellets or regrind stored at high humidity can introduce volatiles that produce splay and local surface cooling. The use of 25 % regrind of the high-MFR injection grade changes the melt viscosity and widens the shrinkage distribution; notched Charpy impact energy under ISO 179-1 falls from 12 kJ/m² to 7 kJ/m². For a container that must survive a drop at −20 °C, the lower-MFR blow moulding grade may be required even if it does not fill the hinge as easily. The correct response is to modify the gate or the nucleating system, not to run the wrong grade.

The notched constant tensile load test at 50 °C in 10 % by volume Igepal CO-630 according to ASTM D5397 provides a direct comparison. A pipe-grade bimodal PE100 HDPE with a density of 0.949 g/cm³ and a melt flow rate of 0.10 dg/10 min reaches a failure time of 3 200 h at an initial stress of 0.8 MPa, whereas a 0.962 g/cm³ monomodal injection HDPE fails in 6 h. The fracture surface of the injection grade is smooth and brittle, with no large-scale fibrillation; the pipe grade exhibits microductile tearing and stress whitening. In a 315 mm outside diameter PE100 water main, the same substitution appears as low melt pressure at the pipe die, lower haul-off tension, and a pipe that meets wall thickness but not the slow crack growth resistance required by ISO 4427. Butt-fusion joints are produced at a heater plate temperature of 225 °C per ISO 21307; joint strength tests are not sufficient to expose the latent ESCR defect.

Stress cracking is also influenced by cooling rate and residual surface stress. Fast cooling from 230 °C to 80 °C suppresses crystalline perfection but also raises surface tensile stress. For a 0.948 g/cm³ grade with high comonomer content, the tie-molecule concentration is higher and the failure under ASTM D1693 shifts from brittle cracks to ductile tearing. The injection grade with 0.962 g/cm³ has higher crystallinity and lower amorphous tie-chain density, so the ESCR F50 value drops below 30 h in the same test. In an extrusion blow moulded detergent bottle, the wrong grade fails at the pinch-off weld after 3 weeks of contact with a 10 % nonylphenol ethoxylate solution at 50 °C, while the specified grade survives beyond 2 500 h. The failure is environmental stress cracking, not chemical degradation; the polymer chains are disentangled and pulled apart under stress.

Wrong-grade damage is not limited to moulded and pipe products. In a 1.8 m flat-die sheet line running at 280 kg/h, the same 8.0 dg/10 min injection HDPE produces a melt curtain that necks in at the die edges. Edge bead thickness increases from 1.2 mm to 2.0 mm, and the sheet centre thins to 0.7 mm. When the sheet is thermoformed, the uneven thickness creates warped trays and corners that crack at the female mould radii. The sheet line is designed for a fractional-MFI blow moulding grade; the wrong grade lowers melt curtain stability and changes the draw ratio in the calendar stack. This is the same sag mechanism translated to a flat-die process.

When PE100 Pipe Resin Is Replaced by a 0.962 g/cm³ Monomodal Injection Grade

The material identification code on the pipe reads PE100, but the certificate of analysis shows a density of 0.962 g/cm³ and an MFR of 8.0 dg/10 min. This material does not meet ISO 4427 or the ASTM D3350 cell class for PE4710. The long-term hydrostatic strength at 20 °C predicted by ISO 9080 is below the required 10 MPa for 50 years; published data for this specific configuration is limited, but the absence of a bimodal high-molecular-weight component removes the slow crack growth resistance required. The resin fails the notched constant tensile load test before 24 h, whereas many PE100 materials exceed 1 000 h. On a pipe extrusion line with a 60 mm single-screw extruder and a 315 mm die head, the melt pressure is 15 % lower because the injection grade has a higher MFR. The haul-off speed must be reduced to maintain diameter, producing a thicker wall that still does not compensate for brittle failure. The resulting pipe is outside the design window for buried pressure service.