In continuous firelog extrusion lines that densify mixed softwood and hardwood sawdust into combustible logs, the binder phase functions as both a processing lubricant and a post-die structural matrix. Its viscosity at the extruder wall and die temperatures determines whether it forms a coherent film over discrete sawdust particles or channels through interstitial voids before solidification. Sawdust conditioned to a dry-basis moisture content of 8–12 wt%, measured by ASTM D4442-20, presents a dynamic wetting substrate because the hydrophilic lignocellulosic surface is partially saturated with adsorbed water. This moisture changes the apparent contact angle of non-polar paraffin wax from approximately 15–25° on oven-dried fibre to more than 70° on saturated fibre. The resulting binder distribution is governed by a coupled temperature–viscosity–time window in which the wetting front must penetrate the packed particulate bed before wax freezes at the die exit or steam generated from residual moisture disrupts film continuity. A production line equipped with a single-screw extruder having an L/D of 16:1 and a constant-taper compression screw operating at 40–80 rpm cannot, in practice, compensate for a binder whose viscosity at the die inlet deviates outside the narrow range that balances pressure generation and particle wetting. A binder that wets too slowly leaves unwetted sawdust pockets that act as stress concentrators in the finished log and produce surface cracks after cooling. A binder that wets too aggressively but lacks adequate film strength can saturate the die wall boundary layer, reduce the wall friction required to build extrusion pressure, and cause slip-related surging at the die. These conditions are not adequately captured by a single Brookfield or melt-flow measurement; rotational viscosity of the neat binder at an arbitrary temperature underestimates the shear thinning and wall slip that occur when wax-filled sawdust compacts pass through a converging die.

What limits binder penetration into sawdust cell wall structure?

Penetration of a molten binder into a packed sawdust bed is governed by the Washburn equation for capillary rise, written as L = (γ cos θ r t / 2η)^0.5, where L is penetration length, γ is liquid–vapour surface tension, θ is the apparent contact angle at the solid–liquid interface, r is effective pore radius, t is contact time, and η is dynamic viscosity. In extruded firelog mixtures, the effective pore radius is not a single capillary dimension; it is a distribution generated by particle size range, packing density, and the elastic recovery of compressed sawdust under pressure. When sawdust moisture content exceeds approximately 10 wt%, the polar wood surface adsorbs water into cell-wall micropores, and the apparent contact angle of non-polar paraffin wax increases from below 25° on oven-dried fibre to above 70° on saturated fibre, as reported in published contact-angle studies on wood surfaces using an adaptation of ASTM D7490. This shift reduces the cosine of the contact angle from above 0.9 to below 0.35, effectively lowering penetration depth at constant viscosity and time. The practical consequence is that a paraffin wax with a melt viscosity of 5–10 mPa·s at 100°C, determined by ASTM D3236-15, may penetrate uniformly through dry sawdust but forms discrete wax-rich lenses when feedstock moisture approaches 12 wt%. Published data for the specific capillary radius distributions of compressed mixed sawdust in firelog extruders remain limited. Direct application of the Washburn equation to production-scale wetting therefore requires measured pore-size data from mercury intrusion porosimetry on the actual compressed feedstock, rather than reliance on generic biomass porosity values from other densification processes.

At shear rates representative of the metering section of a firelog extruder, measured by capillary rheometry according to ASTM D3835-16, the apparent viscosity of wax-filled sawdust compacts diverges significantly from the neat binder viscosity because the suspended wood particles act as a high-solids filler. A capillary rheometer equipped with a 1 mm diameter die and a length-to-diameter ratio of 20:1, operated at 80–110°C and shear rates from 10 s⁻¹ to 500 s⁻¹, shows that a paraffin wax binder with a neat viscosity of 8 mPa·s at 100°C produces a sawdust-wax mixture with an apparent viscosity between 150 mPa·s and 1 200 mPa·s, depending on sawdust volume fraction and particle aspect ratio. At low shear rates below 50 s⁻¹, the mixture exhibits yield-like behaviour because mechanical interlocking of acicular wood particles must be overcome before bulk flow can occur. At shear rates above 200 s⁻¹, shear thinning becomes pronounced, and the apparent viscosity approaches a high-shear plateau that is more dependent on the continuous binder phase than on particle-particle friction. For production lines, a rotational viscosity reading on the neat binder at 100°C does not predict how the mixture will flow. Extruder screw torque correlates more closely with the high-shear plateau viscosity and the wall slip velocity at the barrel interface; wall slip itself is promoted by migration of low-viscosity wax to the metal surface. The shear rate at the die wall is approximated as 4Q/πR³ for a round die of radius R and volumetric flow rate Q. In a 60 mm die operating at 120 kg/h, the wall shear rate can exceed 80 s⁻¹, placing the process in the shear-thinning regime where small changes in binder molecular weight distribution produce disproportionately large changes in die pressure.

Rheological constraints in binder selection for densified firelog extrusion

Binder selection for a continuous firelog extruder must be treated as a rheological operating window rather than a single specification. For a paraffin wax, the congealing point, viscosity at barrel temperature, and oil content determine the lower and upper process limits. A fully refined paraffin wax with a congealing point of 52–54°C and an oil content below 0.5 wt% typically exhibits a kinematic viscosity of 5–8 mm²/s at 100°C when tested to ISO 3104. A semi-refined slack wax with an oil content of 2–5 wt% may show a viscosity at 100°C of 4–10 mm²/s, but the oil fraction migrates to the sawdust surface and can cause post-extrusion exudation. The lower viscosity boundary is set by die pressure: if the molten binder viscosity is too low, the mixture cannot maintain the back pressure necessary to compact sawdust into a dense log with an apparent density of 0.9–1.1 g/cm³ after cooling. The upper viscosity boundary is set by torque and wetting: if the mixture apparent viscosity exceeds approximately 1 500 mPa·s at the die shear rate, extruder motor demand rises, and the binder may not spread sufficiently over the particles within the available residence time. These boundaries are not fixed across equipment because a longer extruder with an L/D of 24:1 can generate more pressure at lower die temperature than an L/D 16:1 machine. Firelog plants using paraffin wax binders often control the die head temperature in a band of 70–95°C; below this band, wax viscosity climbs steeply as the congealing point is approached, and above it, the risk of steam generation from residual sawdust moisture increases sharply. The processing window narrows to ±5°C when sawdust moisture is at the upper specification limit of 12 wt%, because steam formation and wax film rupture occur simultaneously at die temperatures near 100°C.

On production-scale single-screw firelog extruders, the interaction between moisture, die temperature, and binder viscosity produces specific failure modes that are not apparent in laboratory melt blending. A line running mixed spruce-pine sawdust at 100 kg/h through a heated die with a land length of 120 mm and a die aperture of 55 mm may exhibit periodic pressure oscillations of ±0.5 MPa when feedstock moisture content varies by as little as ±1 wt%. These oscillations arise because injected steam from heated sawdust creates a compressible phase at the die inlet, collapsing the pressure profile and interrupting adhesion of the wax film to the die wall. The resulting logs show alternating dense and porous bands, and the failure is frequently misattributed to binder viscosity when it actually reflects moisture heterogeneity in the feed. Pre-drying sawdust to 7–9 wt% dry-basis moisture, measured on line with ASTM D4442-20 as the reference method, reduces steam generation but can make the sawdust surface less receptive to aqueous binders and can increase dusting at the feed throat. Excessive drying below 5 wt% can cause static charge accumulation and uneven wax adhesion because wood extractives migrate to the surface during heated storage. Waste heat from mechanical compression raises the compact temperature along the screw; in a compression screw with a compression ratio of 3:1, local temperature can rise from 25°C at the feed throat to 95°C at the die without external heating. The actual binder temperature is therefore not simply the barrel heater setpoint but a convolution of barrel heating, frictional heating, and evaporative cooling from moisture. A production-scale extruder with an installed motor power of 37 kW can tolerate only short overtorque excursions above 80% rated load; repeated excursions caused by high mixture viscosity lead to premature gearbox wear and screw flight breakage.

Batch qualification of binder-sawdust mixtures using a torque rheometer provides a more direct indication of extruder motor load than a viscometric curve. A Brabender Plastograph EC equipped with a 30 cm³ mixer head and roller rotors, preheated to 90°C and operated at 40 rpm, can differentiate between two paraffin waxes that have the same rotational viscosity at 100°C but different oil contents or molecular weight distributions. The torque trace during the first 10 min after adding 85 wt% sawdust at 9 wt% moisture to 15 wt% molten wax shows an initial fusion peak, a steady-state mixing torque, and a temperature rise due to friction. A typical fully refined paraffin wax formulation reaches a steady-state torque of 12–18 N·m, whereas a semi-refined slack wax of identical neat viscosity may produce 20–28 N·m because the oil component softens the sawdust surface and promotes compaction before dispersion. Torque rheometry also reveals dry-powder lubrication effects: adding 0.5–1.0 wt% of calcium stearate lowers mixing torque by reducing wall friction, but it can also reduce water absorption and create a hydrophobic layer on the sawdust that inhibits aqueous binder wetting in mixed-binder systems. This torque method is not prescribed by a single ISO standard for firelogs, so its results are used as an internal specification correlated to the production extruder rather than as a release test. The critical limitation is that a torque rheometer operates at low shear relative to the extruder die; therefore it cannot replace capillary rheometry for die pressure prediction but is effective for ranking formulations and detecting batch-to-batch changes in feedstock particle size or moisture.

When starch-based binders replace paraffin wax in low-smoke formulations

When starch or molasses-based binders replace paraffin wax to reduce visible smoke and petroleum-derived content, the extrusion process shifts from a melt-coated particulate flow to a water-film-bound paste flow. A pregelatinized corn starch binder in water at 30–40 wt% solids develops an apparent viscosity between 500 mPa·s and 2 000 mPa·s at 60–80°C, depending on shear rate and starch source, as measured by ISO 3104 or rotational rheometry with parallel plates. This viscosity is one to two orders of magnitude higher than paraffin wax at the same temperature, requiring lower sawdust moisture and higher screw torque to extrude. Gelatinized starch acts as an adhesive film that wets hydrophilic sawdust surfaces more effectively than paraffin wax because the contact angle of a starch paste on wood is typically below 35°, but the water in the starch must be removed after extrusion. Firelogs bound with starch therefore require a separate curing or drying step, typically at 60–80°C for 4–8 h in a forced-air tunnel, to reduce final moisture to below 8 wt% for acceptable combustion. The drying step is the capacity bottleneck; a line originally designed for paraffin wax at 120 kg/h may be limited to 50–70 kg/h when starch binders are introduced, not because of the extruder but because of downstream dryer residence time. Molasses binders contain sugars that plasticize with water but can become brittle when over-dried, leading to surface spalling; they also have a viscosity that rises steeply below 25°C, causing cold-start problems in unheated hoppers. Compatibility with wood extractives must be verified because tannins and acetic acid released from hardwoods can reduce starch paste viscosity over time. Published data for this specific firelog configuration are limited, and the stability of starch-sawdust blends should be tested with a torque rheometer over a 4 h hold at 60°C before production qualification.

Table 1 summarises comparative binder behaviour for firelog extrusion.

Wetting envelopes for hardwood and softwood fines at 8–12 wt% moisture

Wetting behaviour differs between softwood and hardwood sawdust fractions because hardwood vessels and fibres present a more polar surface with higher hemicellulose content than softwood tracheids. A hardwood mix containing oak, maple, and cherry sawdust at 10 wt% moisture, screened through a 2 mm screen to a d50 of 0.6 mm when tested to ISO 17827-2:2016, shows a greater capacity for aqueous binder absorption than a spruce-pine-fir mix at the same particle size. The difference is observed in higher wetting agent demand in particleboard and wood-fibre insulation processes and cannot be transferred to firelog extrusion without adjusting binder addition. For non-polar paraffin wax, the apparent contact angle on softwood can be lower than on hardwood at the same moisture because softwood extractives include hydrophobic resin acids that compete with the wax for surface coverage. This creates a feedstock-dependent wetting envelope: a softwood mix at 8 wt% moisture may wet uniformly with 6–8 wt% paraffin wax, while a hardwood mix at 12 wt% moisture can require up to 10–12 wt% of the same binder or a wetting additive to avoid dry pockets. The addition of fatty acid-based wetting agents, such as vegetable-oil-derived distilled fatty acids with an acid value of 190–205 mg KOH/g and an iodine value below 120 g I₂/100 g, can reduce interfacial tension at the wax-water-wood boundary and restore wetting at higher moisture. However, these polar additives lower the wax congealing point and can increase smoke if addition exceeds 2 wt% of the binder. Production experience shows that batch-to-batch variance in sawdust species and moisture creates a larger shift in wetting than the nominal viscosity tolerance of the binder; therefore the standard operating procedure should specify moisture and particle size more tightly than wax viscosity. Published data for the specific contact angles of production sawdust blends are limited, but the trend is consistent with wood surface energy measurements under ASTM D7490 and with practical requirements for wax emulsions used in wood composite manufacturing.

Before a sawdust-binder mixture is released to continuous extrusion, incoming raw material and compounded mixture should be verified against the standardised methods shown in the compliance matrix below. Moisture content is determined gravimetrically by ASTM D4442-20, with an upper acceptance limit of 12 wt% for paraffin-bound firelogs and a lower limit of 8 wt% except where post-drying is part of the process. Binder viscosity is measured by ASTM D3236-15 at 100°C for wax-based binders and by ISO 3104 at the specified aqueous process temperature for starch or lignosulfonate binders. Ash content is tested according to ISO 18122:2022 and should not exceed 1.5 wt% on a dry basis for premium firelogs, because ash-forming minerals contribute to sintered residue and visible smoke. Net calorific value is measured by ISO 18125:2017 with a typical acceptance threshold of 18.5 MJ/kg dry basis. Finished log density is verified by geometry and mass, with an acceptance target of 0.9–1.1 g/cm³ for continuous screw-extruded logs. The compliance matrix does not replace inline process rheometry because sampling a hot wax-sawdust mixture from a running extruder is hazardous and can disturb die pressure; therefore the matrix is applied to raw binder and conditioned sawdust before extrusion, while extruder parameters are monitored separately for motor load, barrel pressure, and die temperature. A production lot should be rejected if binder viscosity at 100°C exceeds 25 mPa·s unless the extruder has been specifically designed for high-viscosity paste binders.

These acceptance bands are not universal; they are specific to continuous screw extrusion with a paraffin wax binder and should be revalidated when binder chemistry changes. Starch-bound firelogs, for example, can be extruded at higher mixture viscosity but require a final moisture content below 8 wt% after drying and therefore fall outside the paraffin wax matrix. Recycled sawdust from construction and demolition waste may contain resins, paints, or chlorinated compounds that require chemical screening under REACH or local waste regulations before use. Avoid combining wax-bound sawdust with amine-based additives because such additives can catalyse wax oxidation at hot die surfaces and produce varnish-like deposits in the extruder barrel. Process water used for starch or molasses binders should be tested for hardness because divalent calcium and magnesium ions can crosslink lignosulfonates and raise paste viscosity unexpectedly.