Menu
Ethylene-vinyl acetate copolyolefin foams are manufactured by compounding 18–28 wt% vinyl acetate content EVA resin with 0–30 phr styrene-butadiene-styrene block copolymer, 2–6 phr azodicarbonamide, 0.5–1.5 phr dicumyl peroxide, 0.5–2.0 phr zinc oxide, and 0.5–1.5 phr zinc stearate. The compound is sheeted on a calender with a thickness tolerance of ±0.05 mm and expanded in a continuous hot-air oven with a residence time of 8–12 min. On a co-rotating twin-screw extruder with L/D 44:1 and a gear pump, increasing SBS from 5 phr to 25 phr reduces volumetric expansion ratio from approximately 6.5:1 to 4.1:1 while apparent density rises from 120 kg/m³ to 190 kg/m³; the loss is steepest between 15 phr and 20 phr. The same change raises melt pressure at the gear pump from 7.8 MPa to 10.4 MPa and reduces melt flow index measured at 190°C and 2.16 kg according to ASTM D1238-20 from 2.8 g/10 min to 1.6 g/10 min. Scanning electron microscopy on microtomed cross-sections shows a reduction in mean cell diameter from 0.32 mm at 5 phr SBS to 0.21 mm at 25 phr, and skin density measured by water displacement according to ISO 845:2006 increases by 12–18%. Published data for this specific continuous-line configuration is limited; the values above are line-specific but the trend is consistent with published studies on EVA/SBS blends.
Above 15 phr, the expansion loss is attributed to two competing effects. The styrene end-blocks of SBS form reversible physical crosslinks that increase zero-shear viscosity and extensional viscosity; in a parallel-plate rheometer at 200°C and 0.1 rad/s, complex viscosity for a 22 wt% VA EVA compound rises from 8,500 Pa·s at 5 phr SBS to 18,600 Pa·s at 25 phr, while at 100 rad/s the increase is less than 20%. Loss tangent at 0.1 rad/s decreases from 0.9 to 0.5, reflecting a more elastic response, but at 100 rad/s it remains near 0.7, indicating that the elastic plateau is not uniformly available during high-rate cell growth. The second mechanism involves dicumyl peroxide half-life behaviour; the peroxide has a half-life of 1 min at 171°C and 1 h at 135°C, so at foaming temperatures radical flux is high. The 1,4-butadiene midblock of SBS has allylic hydrogens that undergo preferential hydrogen abstraction, diverting alkoxy radicals from EVA vinyl acetate segments and reducing effective EVA crosslink density. Gel fraction measured by solvent extraction according to ASTM D2765-16 may remain within 55–65% because SBS itself crosslinks, but the EVA network continuity is lower and gas retention during the critical expansion window is impaired. This is visible as reduced cell density and a wider cell-size distribution at 20–25 phr SBS.
| SBS loading (phr) | Complex viscosity at 0.1 rad/s and 200°C (Pa·s) | Gel fraction (%) | Apparent density (kg/m³) | Volumetric expansion ratio (V/V) | Cell density (cells/cm³) |
|---|---|---|---|---|---|
| 0 | 7,500–8,500 | 60–65 | 110–120 | 6.5–7.0 | 1.4–1.6 × 10⁵ |
| 5 | 8,000–9,000 | 58–63 | 112–124 | 6.2–6.7 | 1.3–1.5 × 10⁵ |
| 10 | 8,500–9,800 | 56–61 | 118–130 | 5.8–6.3 | 1.2–1.4 × 10⁵ |
| 15 | 9,500–11,000 | 54–60 | 124–136 | 5.4–5.9 | 1.0–1.2 × 10⁵ |
| 20 | 11,500–13,500 | 52–58 | 132–148 | 4.9–5.4 | 8.5 × 10⁴–1.0 × 10⁵ |
| 25 | 14,000–18,600 | 50–56 | 150–175 | 4.2–4.8 | 6.8–8.5 × 10⁴ |
| 30 | 16,500–22,000 | 48–54 | 165–190 | 3.8–4.3 | 5.5–7.0 × 10⁴ |
On twin-screw compounding lines equipped with side-stuffers, the feeding of SBS crumbs is rate-limited by the low bulk density and high elastic recovery of the block copolymer granules. SBS is pre-dried at 60–70°C for 2–3 h when ambient relative humidity exceeds 60%; moisture above 0.05% creates surface defects in the calendered sheet and disturbs peroxide dispersion. The screw profile uses kneading blocks at L/D 18–24 for distributive mixing and reverse elements at L/D 32–36 for elongational mixing, with barrel temperature settings of 100°C, 110°C, 115°C, 120°C, and 115°C from feed to die. Melt temperature must remain below 125°C to avoid premature azodicarbonamide decomposition and below 140°C to minimize dicumyl peroxide decomposition during compounding. At 25 phr SBS, specific energy input increases to approximately 0.18 kWh/kg compared with 0.14 kWh/kg for the neat EVA compound, and melt temperature rises by 4–6°C at constant screw speed. The increase in melt viscosity also raises feeder amperage and reduces output stability; production batches with more than 3% variation in SBS concentration show visible density striping across the width of the expanded sheet. Because SBS domains remain in a disperse or co-continuous morphology depending on mixing history, the screw configuration and specific energy input are stronger determinants of expansion retention than simple SBS weight fraction.
The expansion window is bounded by two kinetic processes: peroxide-induced crosslinking and azodicarbonamide decomposition. In a moving die rheometer at 180°C and 0.5° arc according to ASTM D5289-19, substitution of 25 phr SBS raises minimum torque from 0.9 dNm to 1.5 dNm and maximum torque from 4.2 dNm to 5.8 dNm, but the scorch time ts2 shortens from 1.2 min to 0.8 min; this compression of the safe processing interval means that the same oven residence time that produced a uniform foam at 10 phr SBS can produce a dense skin and an under-expanded core at 25 phr. Differential scanning calorimetry at 10°C/min according to ISO 11357-1:2016 shows that the decomposition onset of azodicarbonamide shifts from 198°C to 203°C when SBS loading exceeds 20 phr, because the endothermic disruption of styrene domains and higher matrix viscosity delay heat transfer. The resulting mismatch between gel formation and gas evolution displaces cell nucleation away from the optimal viscosity plateau, reducing cell density from 1.2 × 10⁵ cells/cm³ at 10 phr SBS to 6.8 × 10⁴ cells/cm³ at 25 phr by scanning electron microscopy image analysis. The same phenomenon produces a broader cell-size distribution; the ratio of the largest to the smallest cell diameter increases from approximately 2.1 at 10 phr to 3.4 at 25 phr, and the mean cell wall thickness determined from the same micrographs increases from 0.08 mm to 0.14 mm.
When a calendered sheet containing 25 phr SBS enters the hot-air foaming oven, the edge and centre zones do not reach the same temperature at the same time because the thicker and more elastic sheet has lower thermal diffusivity and greater emissivity variation after repeated calender passes. The belt speed is reduced from 4.2 m/min at 10 phr SBS to 3.5 m/min at 25 phr to compensate for slower cell growth, but this increases the outer surface residence time and produces a denser skin. Density mapping across a 1,100 mm wide sheet shows edge-to-centre variation of ±6 kg/m³ at 10 phr SBS and ±14 kg/m³ at 25 phr, which exceeds the footwear midsole tolerance of ±5 kg/m³ typically specified for high-rebound formulations. The sheet also displays anisotropic dimensional recovery after expansion; at 25 phr SBS, machine-direction shrinkage after 24 h at 23°C and 50% relative humidity is 1.8% compared with 0.9% for the 10 phr compound, meaning downstream cutting and skiving must be delayed or tooling compensation must be increased. The high compound elasticity causes die swell at the calender; the die gap is typically opened from 0.8 mm to 1.1 mm to maintain sheet weight, but this reduces molecular orientation and lowers expansion repeatability. Compression set measured after 50% compression for 24 h at 23°C according to ISO 1856:2018 is 8–12 percentage points higher at 25 phr SBS than at 10 phr SBS, which reduces energy return in footwear midsoles. Batch-to-batch variation in expansion ratio is therefore wider at elevated SBS because small changes in calender gap or melt temperature shift the orientation and domain morphology of the SBS phase.
Process limits at loadings above 20 phr are narrower than at 10 phr, particularly when zinc oxide and stearic acid are used to activate azodicarbonamide decomposition. The exotherm of the activated blowing agent is large enough to create local temperature overshoots above 215°C, which can degrade the polybutadiene midblock and produce surface tack, blowholes, and split edges. Avoid combination with amine-based additives because they can accelerate premature crosslinking or alter ADC decomposition pH; this incompatibility is more severe in SBS-rich compounds than in neat EVA. Pre-drying is mandatory above 60% relative humidity, and SBS grades with diblock content above 15 wt% should be avoided for expansion-critical applications because the non-network diblock fraction lowers melt strength and increases gas loss. At loadings above 30 phr, published data for continuous EVA/SBS co-foaming is limited; the available rheological and processing evidence suggests that the expansion transition is discontinuous and depends strongly on styrene block length, vinyl acetate content, and the ratio of peroxide to blowing agent. Compliance considerations for skin-contact applications require low polycyclic aromatic hydrocarbon levels in SBS extender oils under REACH Annex XVII Entry 50, with benzo[a]pyrene equivalent below 1 mg/kg, and the overall foam must meet ASTM D3574-17 or ISO 1798:2008 tensile requirements as specified by the final article.
