Telling Polyethylene From Polypropylene: Field Tests and Recycling Codes

Under ASTM D7611/D7611M-20, polyethylene (PE) is assigned to two separate Resin Identification Code categories: RIC 2 for high-density polyethylene and RIC 4 for low-density and linear low-density polyethylene. Polypropylene (PP) is assigned RIC 5. The distinction is operationally consequential because mixed PE and PP in a reprocessing stream creates immiscible domains that produce visible surface defects and lower mechanical ductility. A PP contamination level as low as 5 wt% in PE is known to generate weld-line weakness and gloss variation in injection molded articles, although published data for specific grade interactions is limited. Because both materials are non-polar polyolefins with densities spanning 0.895–0.970 g/cm³ according to ISO 1183-1:2019, water floatation alone cannot differentiate them. Field identification therefore requires a combination of density, thermal, flammability, and spectroscopic checks linked to the recycling codes.

At 23 °C, density column checks using ethanol-water mixtures are only partially decisive. An ethanol-water solution adjusted to 0.925 g/cm³ floats PP and sinks most HDPE and a portion of LDPE, but the LDPE density interval of 0.910–0.940 g/cm³ overlaps the upper PP range. Operators at a recycling line with a 1,000 mm sink-float tank and 15 rpm paddle wheel observe PP carryover in the sinking HDPE fraction when shredded flakes retain air pockets or labels. Hot xylene at 140 °C dissolves both PE and PP at comparable rates, and cold xylene immersion swells PP more slowly than LDPE; however, the difference is not sufficiently repeatable for forensic sorting. The field method is therefore limited to screening, with confirmatory DSC or FTIR required under ISO 11357-3:2018 or ASTM E1252-98(2021).

A density gradient column per ASTM D1505-18 provides stronger resolving power than simple sink-float because the column is calibrated with glass floats covering a range of 0.90–0.98 g/cm³ at 23 °C. A flake sample is wetted with a low-surface-tension wetting agent and placed in the column; the equilibrium position after 60 min is read to the nearest 0.001 g/cm³. On a column filled with isopropanol-distilled water, HDPE settles at 0.94–0.97 g/cm³ and PP bands at 0.90–0.92 g/cm³, but LDPE may overlap with PP unless the column is operated with a slope no greater than 0.005 g/cm³ per 10 cm. Temperature control is critical; a drift of 0.5 °C can shift the apparent density by roughly 0.0003 g/cm³. Published interlaboratory reproducibility for density gradient columns is tight for homogeneous unfilled resins but degrades when flake carries print inks or paper fiber.

Can a Bunsen-Style Flame Test Replace Spectroscopic Identification?

Flame testing is rapid but destructive. PE burns with a blue flame core, yellow luminous tip, and molten dripping while emitting a paraffin-like odor similar to candle wax. PP burns with a comparable blue-to-yellow flame but produces a sharper, mildly acrid odor and more pronounced melt dripping. Odor differentiation is operator-dependent and is not recognized as a release criterion under EN 15343:2007 for recycled content traceability. Ignition temperature measurement under ASTM D1929-20 occasionally separates material types, but published interlaboratory repeatability for PE and PP olefin mixtures is limited. On a production-floor bench torch with a 20 mm flame height, the destructive nature of the test and the accumulation of low-molecular-weight combustion residues in ventilation filters restrict the method to spot checks. Density and DSC melting point remain the quantitative confirmatory measurements.

For plant-laboratory differentiation, differential scanning calorimetry under ISO 11357-3:2018 resolves PE and PP through melting endotherms separated by roughly 30–55 °C depending on comonomer content. HDPE homopolymer typically exhibits a melt peak at 125–135 °C, LDPE at 105–115 °C, and PP homopolymer at 160–170 °C. This thermal interval provides a decisive separation when flake surfaces are free of mineral oil, labels, and crosslinked gel. In recycling qualification, a single DSC run at 10 K/min under nitrogen is often paired with melt flow-rate measurement according to ISO 1133-1:2022. PE grades are tested at 190 °C with 2.16 kg load, while PP grades are tested at 230 °C with 2.16 kg; direct comparison of MFR values between the two temperatures is invalid. A 75 mm single-screw recovery extruder with L/D 30:1 and a melt temperature set point of 220 °C can extrude both materials, but PP contamination in PE manifests as fluctuating melt pressure and poor strand pelletization because the higher-viscosity PP phase is not uniformly homogenized under the same shear conditions.

When equipped with diamond attenuated total reflectance crystals, handheld Fourier transform infrared spectrometers identify PE through the characteristic 2,915 cm⁻¹ and 2,850 cm⁻¹ asymmetric and symmetric C–H stretching bands of the methylene chain. PP adds the methyl bending absorption near 1,377 cm⁻¹, which is absent from unmodified PE homopolymers with negligible branching. The intensity ratio of the 1,377 cm⁻¹ band to the 1,460 cm⁻¹ C–H bending envelope provides a semi-quantitative field indicator; however, pigment loadings above 1 wt% carbon black attenuate mid-infrared signal unless the surface is microtomed or the sample is solvent-cast. Portable Raman systems operating at 785 nm also differentiate PE and PP through skeletal C–C stretching, but fluorescence from colored additives can overwhelm the Raman response. According to ASTM E1252-98(2021), spectral identification requires reference libraries generated from the same sample form and thickness.

When Automated Sorting Lines Misclassify Mixed Olefin Flake

Near-infrared sorters operating in the 1,100–1,700 nm range discriminate PE and PP through second overtones of C–H stretching bands; however, black masterbatch-loaded articles absorb strongly and are often not classified by primary NIR cameras. A recycling line running at 2.5–5.0 m/s belt speed with a 1,000 mm working width can misdirect PP into LDPE/HDPE streams when flake thickness is below 10 mm and surface moisture alters reflectance. Electrostatic separation after NIR can recover some PP because tribocharging differs, but relative humidity above 60% collapses charge retention and decreases separation efficiency. Published data for the exact recovery rate of mixed PE/PP flakes on commercial sorters is limited; equipment manufacturers typically recommend bench-scale tribocharging trials before specifying a line.

Recycling Code Boundaries and Contamination Tolerance

RIC 2 applies to HDPE; RIC 4 applies to LDPE and LLDPE; RIC 5 applies to PP. ASTM D7611/D7611M-20 defines the code geometry and assignment, not compositional purity. A bale of code 2 HDPE may contain 2–8 wt% PP closures and labels if the stream is co-collected. PP contamination in HDPE reprocessing produces immiscible inclusions at the micrometer scale; the resulting extruded sheet can display “fish-eye” gels under ASTM D7310-21 film-gel counting. Melt filtration with screens of 100–250 µm does not remove dissolved PP; it only removes solid contaminants above the screen aperture. A PP concentration above 5 wt% is known to shift the shear viscosity of the blend and can reduce Izod impact strength under ISO 180:2023 by a measurable amount, although reported values vary with grade and processing history.

On a co-collected recycling stream, code markings under ASTM D7611/D7611M-20 are insufficient for automatic sorting because labels are often missing, painted, or covered by shrink sleeves. A code 2 bottle may include a PP cap with no marking, and a code 4 flexible pouch may be laminated with ethylene-vinyl acetate or polyamide layers. The physical recycling process must therefore rely on the full set of density, thermal, and spectral observations rather than code presence alone. When mixed PE/PP flake is converted on a co-rotating twin-screw extruder with L/D 40:1 and 700 rpm screw speed, PP islands tend to persist at barrel set points below 230 °C, producing die-lip buildup at the strand die. Raising the set point to 250 °C reduces islands but accelerates low-molecular-weight degradation in the PE phase; published data for the exact residence time threshold in mixed olefin extrusion is limited.

In injection molding, screw recovery time and cushion stability are contaminated by PP in PE. A 40 mm injection screw with L/D 20:1 and a shot weight of 120 g typically shows a slower recovery and more injection pressure variability when PP concentrations exceed 5 wt%. The melt viscosity mismatch between PE and PP is not eliminated by raising back pressure; above 80 bar back pressure, high-shear heating intensifies degradation in the PE matrix and can cause yellowing. Processing windows are therefore bounded by a practical barrel-temperature ceiling of 240 °C for mixed PE/PP moldings; beyond this, oxidative degradation consumes stabilizer packages and reduces long-term thermal stability under ISO 4577:1983.

Cast film and blow film lines provide another operational distinction. LLDPE blown film typically processes at a blow-up ratio of 2.0:1–3.0:1 and a frost-line height of 150–350 mm on a 45 mm extruder with L/D 25:1. PP cast film processes at higher melt temperatures of 230–250 °C and narrower die-lip openings. A PE/PP mixed feed on a blown film line produces haze bands across the web and unstable bubble geometry; gel-count increases are measurable under ASTM D7310-21, but the threshold for a specific film grade is not constant. Published data for the exact gel-count threshold in mixed olefin film is limited.

Regulatory limitations reinforce the need for correct resin identification. FDA 21 CFR 177.1520 lists olefin polymers and specifies migration limits that differ by material and food type; a misidentified PP stream sold as PE may fail hexane extractables under the prescribed test conditions. Under EU Regulation 10/2011, food-contact polyethylene and polypropylene are authorized with specific migration limits, but processing aids and recycled feedstock require appropriate quality assurance. The presence of non-olefin contaminants and mixed-resin degradation products changes overall migration and must be assessed under EN 1186-1:2002 rather than assumed from RIC labeling alone.

Injection-molded parts exhibit enough mechanical dissimilarity to support field identification. PP homopolymer typically shows tensile modulus values of 1,300–1,800 MPa under ISO 527-2:2012, while HDPE falls near 700–1,200 MPa. The difference corresponds to a characteristic flexural stiffness and a harder “snap” on bending; PP sample plaques show stress whitening at bends more readily than PE. Shrinkage in the mold also differs: PP typically shrinks 1.0–2.0%, whereas HDPE shrinkage is often 1.5–3.0%, although these ranges overlap and depend on wall thickness and nucleating agents. Consequently, part stiffness and mold shrinkage are not sufficient to establish resin identity without the thermal and density checks described above. Published data for dimensional response of recycled mixed olefin blends is limited, and operational boundaries remain set by feedstock variability, surface contamination, and processing history.