A tandem extrusion line configured for polyolefin foam with isobutane differs fundamentally from conventional single-screw sheet or profile extrusion because the first-stage extruder is operated as a high-pressure melt pump and mixing device while the second-stage extruder is operated as a heat exchanger that must reduce melt temperature without allowing the dissolved blowing agent to undergo phase separation. Isobutane is injected into the polymer melt after the polymer has formed a sealed melt plug, typically at injection pressures between 12 MPa and 18 MPa for low-density polyethylene and between 15 MPa and 22 MPa for high-melt-strength polypropylene. The gas-laden melt is homogenised in the first extruder and transferred through an adaptor into the cooling extruder, where barrel temperatures are set progressively lower. Foam density is governed by the mass fraction of isobutane dissolved in the melt, the die melt temperature, and the die pressure, while cell morphology is governed by nucleation density, bubble growth rate, coalescence, and the extensional viscosity of the melt during the rapid pressure drop at the die exit. For polyolefin foam boards and extruded plank, apparent density is typically reported according to ISO 845:2006 or ASTM D3575-20, and cell size is characterised by ASTM D3576-20. The term “tandem” refers not merely to the use of two extruders, but to the separation of plasticating and cooling functions so that a polymer melt can be held at high pressure while its temperature is reduced below the normal melt-processing temperature of the resin. Commercial tandem units commonly pair a first-stage extruder with an L/D ratio of 30:1 to 36:1 with a second-stage cooling extruder having an L/D ratio of 25:1 to 30:1. This operational separation is particularly critical with isobutane because the blowing agent acts as a plasticiser, lowering melt viscosity and shifting the zero-shear viscosity and shear-thinning behaviour of the polyolefin. The result is a process window in which die temperature is often only 5 °C to 15 °C above the onset of melt fracture, while die pressure must simultaneously remain above the saturation pressure of the isobutane in the polymer. Any disturbance in first-stage screw speed, second-stage barrel cooling, or die lip heater output can shift density and cell size within seconds, producing off-specification plank with surface defects or internal blowholes. Production-scale tandem lines therefore rely on gear-pump-assisted discharge, static mixers, and pressure transducers at the adaptor and die inlet to maintain the single-phase solution state. Isobutane is a saturated hydrocarbon with a normal boiling point of -11.7 °C, a critical temperature of 134.7 °C, and a critical pressure of 3.64 MPa; above the critical temperature and pressure it exists as a supercritical fluid, but in the cooled second-stage melt it may be a subcritical compressed gas dissolved in the polymer.

What Solubility and Vapour-Liquid Equilibrium Constraints Determine the Minimum Die Pressure?

Isobutane does not behave as an ideal gas at the temperatures encountered in the second-stage cooling extruder. The minimum die pressure required to prevent premature foaming is determined by the vapour-liquid equilibrium of isobutane in the specific polyolefin grade, the melt temperature at the die, and the isobutane mass fraction. Published pressure-decay sorption studies on LDPE report that solubility increases with pressure and decreases with rising temperature; in the temperature range of 105 °C to 150 °C, the solubility of isobutane in LDPE at 6 MPa is sufficient to support apparent densities below 80 kg/m³, but the exact value varies with chain branching, comonomer content, and melt index. Because the critical temperature of isobutane is 134.7 °C, the blowing agent may be injected into the first-stage melt as a supercritical fluid when the melt temperature exceeds that value, but it becomes a subcritical gas or compressed liquid after the cooling extruder reduces the melt temperature below critical. At the first-stage injection point, the melt is typically in the supercritical region for isobutane, but the solubility of the blowing agent still depends on pressure and temperature; a high-pressure LDPE with a melt mass-flow rate of 2 g/10 min under ISO 1133-1:2022 often exhibits a dynamic complex viscosity of 1,500 Pa·s to 3,000 Pa·s at 190 °C and angular frequency 100 rad/s. Dissolving 6 wt% isobutane can lower this viscosity by approximately 20% to 40%, which is beneficial for cooling but harmful for cell stabilisation if the die temperature is too high. The solubilisation process also narrows the safe pressure window: if the adaptor pressure falls below the bubble point, foam forms inside the cooling screw, leading to surging, loss of pumping efficiency, and cell defects at the die. Commercial tandem lines therefore maintain die pressure in the range of 4 MPa to 8 MPa for LDPE and 6 MPa to 12 MPa for HMS-PP, depending on die gap, output rate, and melt temperature. The pressure drop rate at the die lips, not the absolute pressure alone, influences cell nucleation. Classical nucleation theory indicates that nucleation density increases with the degree of supersaturation and the rate of pressure drop; for isobutane-foamed LDPE, a pressure drop rate in the order of 10⁵ kPa/s to 10⁷ kPa/s is frequently cited in extrusion foaming studies, although published data for a specific die geometry may be limited. The practical consequence is that die channel length and lip gap must be matched to output rate so that the melt does not spend excessive time under supersaturated conditions before exiting. A long land length can lower the pressure drop rate and produce coarse cells, while an excessively short land length can generate melt fracture and poor thickness control. The screw speed of the first extruder, the speed of the gear pump, and the second-stage screw speed must be controlled to within ±1% of set point on high-density foam lines to avoid pressure oscillations that are translated directly into density variation.

Second-stage cooling does not proceed as a uniform radial heat-transfer process.

The second cooling extruder is not simply a lower-temperature version of the first extruder; it must remove heat from a viscous polymer melt whose thermal conductivity is only 0.15 W/(m·K) to 0.35 W/(m·K) for polyolefins. The screw flight generates viscous dissipation, while the barrel wall removes heat, creating a radial temperature gradient in which the melt adjacent to the cooled barrel is colder and more viscous than the melt near the screw root. If the second-stage barrel is set too cold, a frozen or highly viscous layer forms along the barrel wall, reducing the effective flow channel and increasing shear heating, which can paradoxically raise the melt temperature at the centre. If the barrel is set too warm, the melt leaves the die at an elevated temperature that lowers melt strength and promotes cell coalescence. This trade-off is especially severe with isobutane because the blowing agent reduces melt viscosity and increases the sensitivity of viscosity to temperature. Processing trials on 90 mm and 120 mm tandem cooling extruders with L/D ratios of 25:1 to 30:1 have shown that the die melt temperature may vary by 3 °C to 6 °C across the die face when the barrel wall temperature is lowered too aggressively. A static mixer installed between the cooling extruder and the die can reduce this gradient by redistributing melt from the wall and core regions, but it also adds pressure drop and residence time. The use of a gear pump after the cooling extruder decouples the pressure generation from the cooling screw and permits more uniform discharge, but it does not eliminate the radial temperature gradient. In some commercial lines, the second-stage screw is designed with deep flights and low compression ratio to minimise viscous heat generation, and the barrel is divided into multiple independently controlled zones with water or oil cooling. The temperature set point of the last barrel zone is typically 5 °C to 10 °C above the die set point to compensate for shear heating through the adaptor and die. When high-melt-strength polypropylene is processed, the cooling range is higher because the melting point of the polymer is around 160 °C, and the die melt temperature must remain above the crystallisation temperature but low enough to provide extensional viscosity. The second-stage extruder therefore operates in a narrow window where the melt is not fully solidified and not excessively fluid. Published resin supplier processing guides for HMS-PP foam report die melt temperatures of 150 °C to 160 °C for isobutane foam, while LDPE foam uses 105 °C to 115 °C, though the optimum depends on screw design and line speed.

Quantification of cell morphology in isobutane-foamed polyolefins requires a consistent definition of cell size, cell density, and open-cell content. The average cell size is often determined by optical microscopy or scanning electron microscopy on sections prepared perpendicular to the extrusion direction. When the cell structure is anisotropic, the chord length is measured along both the machine direction and the transverse direction, and the cell aspect ratio is reported as the quotient of the two mean values. For microcellular and fine-celled foams, image analysis software is used to count cells and to calculate cell density according to N_f = (nM²/A)^3/2, where n is the number of cells in an image area A and M is the magnification. The result is expressed as cells/cm³ and is sensitive to the threshold setting and the minimum cell area used in the image analysis. For extruded polyolefin foam, the cell density can range from 10⁴ cells/cm³ in coarse, low-density LDPE to above 10⁸ cells/cm³ when a nucleating agent and a fast pressure drop are used, but published data for a specific line configuration may be limited because cell density depends on die geometry, nucleator dispersion, and the measured surface area. Thickness and density gradients also influence cell morphology: the surface skin of the plank may have smaller cells due to rapid cooling, while the core may have larger cells due to slower cooling and lower extensional strain. To avoid misleading comparisons, the measurement location must be fixed, and the specimen conditioning must follow the same protocol as the density test. ISO 845:2006 specifies apparent density of cellular plastics by measuring mass and volume of conditioned specimens; the mass is determined on an analytical balance with resolution at least 0.1 mg, and the volume is determined from specimen dimensions or by water displacement for irregular shapes. For cell size, ASTM D3576-20 provides a chord length method for rigid cellular plastics, and although flexible polyolefin foams may require adaptation, the same imaging principles are applied. The cell size distribution, not only the arithmetic mean, is critical for mechanical and thermal performance; a broad distribution with coarse cells in the core can reduce compressive strength and increase thermal conductivity compared with a narrower distribution at the same density.

When die land temperature differs from melt temperature by more than 10 °C, cell anisotropy becomes irreversible.

The die is the last thermal and shear zone before bubble nucleation. The die land temperature is frequently controlled separately from the adaptor and die body because it affects the surface skin, die pressure, and cell orientation. If the die land temperature is more than 10 °C lower than the melt temperature, the surface layer of the extrudate may cool below the melt crystallisation temperature, forming a dense skin that restricts bubble growth near the surface and produces a high-aspect-ratio cell structure. If the die land temperature is more than 10 °C higher than the melt temperature, the surface melt may thin and rupture after exit, leading to gas escape, surface roughness, and open-cell content. The die temperature profile across the width of a flat die is adjusted by heater zones and occasionally by internal oil channels; a temperature variation of ±2 °C across the die face is often considered acceptable for LDPE foam, while tight density tolerances on HMS-PP plank may require ±1 °C uniformity. The pressure drop rate at the die lips is approximately equal to the die pressure divided by the residence time of the melt in the final land, and the residence time is governed by the land length and the melt velocity. In flat dies used for foam board, the land length is typically 10 mm to 30 mm, while the lip gap may be 0.5 mm to 2.5 mm depending on board thickness and expansion ratio. The expansion ratio, defined as the ratio of polymer density to foam apparent density, can exceed 40 for low-density LDPE foam, but at high expansion ratios the cell walls become very thin and are prone to rupturing during bubble growth. High-melt-strength PP grades are specified by resin suppliers with melt tension values in the range 20 cN to 35 cN at 190 °C; this extensional melt strength is required to prevent excessive cell wall thinning during the expansion phase. Cell anisotropy, quantified as the ratio of machine-direction cell diameter to transverse-direction cell diameter, is often between 0.6 and 1.8 for extruded polyolefin foams made with isobutane; values outside this range indicate a process fault, such as excessive draw-down, die lip sticking, or uneven cooling. The anisotropy, once formed, cannot be corrected downstream because the polymer solidifies within seconds after the die exit. For this reason, the die land temperature is not optimised independently but is set through a response surface based on die pressure, output rate, and melt temperature.

The dispersion of talc nucleating agent in the first-stage extruder determines the number of heterogeneous nucleation sites available during the pressure drop at the die, and this effect is stronger than the absolute gas concentration above the solubility limit. Talc is used in isobutane-foamed polyolefins at loadings between 0.2 wt% and 1.0 wt% for LDPE and between 0.3 wt% and 1.2 wt% for HMS-PP. The talc particles provide interfaces that lower the activation energy for bubble nucleation, but they must be adequately dispersed to avoid agglomerates that act as large voids or plate-out on the die lips. Talc is often incorporated as a masterbatch or precompounded resin because direct feeding of fine talc into the main hopper causes segregation and dust. A first-stage screw with a high-shear mixing section, such as a Maddock or Egan section, is used to disperse the talc before isobutane injection. However, excessive shear heating in the first extruder can reduce the solubility of isobutane and accelerate polymer degradation. Melt filtration is sometimes used, but screen packs must be selected carefully because fine screens increase pressure drop and can cause the dissolved blowing agent to phase-separate if the local pressure falls below the bubble point. A gear pump positioned before the cooling extruder or after it can stabilise pressure and reduce the effect of screw beat on cell structure. The cell size distribution narrows as talc dispersion improves, but there is a practical limit because talc itself increases melt viscosity and can promote cell wall rupture at high expansion ratios. The surface chemistry of talc, including particle size and aspect ratio, influences the nucleation efficiency; fine talc with median particle size below 10 µm generally produces smaller cells than coarser talc, but it may agglomerate more readily. The shear history in the first extruder must therefore be sufficient to break agglomerates without reducing the molecular weight of the polyolefin. In some formulations, a secondary nucleating agent such as citric acid-based chemical blowing agent is combined with isobutane, but that introduces decomposition residues and can affect the closed-cell content.

Foam Density Reporting and Cell Structure Standardisation in Commercial Specifications

Commercial quality agreements for extruded polyolefin foam specify density and cell structure by reference to a limited set of test standards, but the choice of standard and specimen preparation can change the reported value more than the actual process variation. The apparent density of cellular plastic is measured by ISO 845:2006 or ASTM D3575-20, both of which require conditioning at 23 °C and 50% relative humidity for a specified period and reporting mass per unit volume in kilograms per cubic metre. The result should not be confused with the density of the unfoamed polymer or with the skin density measured on separate sections. Cell size is measured according to ASTM D3576-20 using a chord length method, but for flexible polyolefin foam the specimen preparation and imaging magnification must be agreed between supplier and customer because the standard was developed for rigid cellular plastics. Melt mass-flow rate is measured by ISO 1133-1:2022 at 190 °C with 2.16 kg for LDPE and at 230 °C with 2.16 kg for PP; this value is used as an incoming resin property rather than a foam property. Thermal conductivity is measured by ASTM C518-21 or ISO 8301:1991 at a mean temperature of 10 °C or 23 °C, and the measured value is affected by foam density, cell size, and aging because isobutane has a gas thermal conductivity of roughly 0.016 W/(m·K) at 25 °C, which is lower than air, but the high vapour permeability of polyolefin foam allows the blowing agent to diffuse out over time. The following table aligns the relevant methods with the measurement target.

The table below compiles representative processing ranges reported in resin supplier processing guides and extruder technical bulletins; they are not universally valid because screw geometry, die design, and resin grade shift the operating window. Each parameter must be validated on the production line using the indicated test method or instrumentation.

Because the second-stage cooling extruder and die geometry generate shear and pressure losses that shift these values, the ranges in the table should be treated as screening targets rather than as guaranteed process settings. Process capability studies on individual lines are required to establish the actual upper and lower control limits for density and cell size.

Safe operation of an isobutane tandem line imposes constraints that are not present with chemical blowing agents. Isobutane is a flammable hydrocarbon with a lower explosive limit of 1.8 vol% and an upper explosive limit of 8.4 vol% in air, so the injection skid, adaptor, and die area must be ventilated and monitored with calibrated combustible gas detectors. Electrical equipment in the blowing-agent storage and injection zone is specified for the relevant hazardous area classification, and the system is purged with nitrogen before maintenance. The polymer melt must not be exposed to oxygen at high temperature for extended periods because oxidative chain scission reduces melt strength and creates gel particles that plug the screen pack and disturb the die pressure. Moisture in talc nucleator or regrind can generate steam at processing temperatures, producing irregular cells and surface pinholes; talc masterbatches exposed to relative humidity above 60% should be pre-dried to below 0.1% moisture before use. Combinations of isobutane with amine-based additives or peroxide crosslinking agents can create exothermic decomposition or odour compounds, so such formulations require separate compatibility testing. The die and adaptor must be equipped with pressure relief paths because a sudden solidification of melt in the cooling extruder can create a confined pressure rise. The process is not suited to resins with low melt strength, such as conventional linear low-density polyethylene, because severe cell coalescence occurs at densities below 200 kg/m³ unless the resin is blended with high-pressure LDPE or a long-chain branching concentrate. These limitations define the operational boundary of the tandem isobutane foam extrusion process and dictate the selection of high-pressure LDPE, HMS-PP, or blends that have sufficiently high extensional viscosity to stabilise growing cells.