How ABS Viscosity Is Actually Measured: A Lab's Point of View

Laboratory characterisation of acrylonitrile butadiene styrene viscosity is a tiered sequence rather than a single measurement. ABS is a heterogeneous graft copolymer with a continuous styrene-acrylonitrile matrix and dispersed polybutadiene rubber particles; its melt is pseudoplastic, temperature-sensitive, and easily influenced by residual moisture and thermal history. A testing laboratory therefore selects methods according to the shear rate of the relevant conversion process. Compression molding may operate near 10¹ s⁻¹, extrusion from 10² s⁻¹ to 10³ s⁻¹, and injection-molding gates can exceed 10⁴ s⁻¹. The resulting data must be reported with the test geometry, moisture status, and correction state because viscosity values without these qualifiers are not comparable across laboratories. For ABS, the routine sequence includes Karl Fischer moisture analysis, melt flow index or melt volume-flow index, capillary rheometry, and, when low-shear structure is relevant, oscillatory parallel-plate rheometry.

Capillary Rheometry as the Primary High-Shear Reference Method

Capillary rheometry is the usual high-shear method because it can generate shear rates that overlap injection-molding gate and runner conditions. Dried ABS pellets are compacted in a heated barrel of 12 mm or 15 mm diameter and extruded through a tungsten carbide capillary die at controlled piston speed. ISO 11443:2021 covers round-hole capillary and slit-die rheometers; the commonly used round-hole die has an L/D ratio of 20:1 or 30:1 and a 180° entrance angle. ASTM D3835-16 is the corresponding U.S. method. The measured pressure drop includes entrance pressure loss, friction in the reservoir, and wall shear stress. Bagley correction uses orifice dies or a series of dies with different L/D ratios to remove entrance effects. Because ABS is strongly non-Newtonian, the apparent shear rate at the wall must be corrected by the Rabinowitsch-Weissenberg relationship; the local slope of log wall shear stress versus log apparent shear rate supplies the correction factor. Viscosity reported without these corrections is an apparent value and is geometry-dependent.

In a typical capillary test on an extrusion-grade ABS at 220 °C, the operator records pressure at apparent shear rates of 10² s⁻¹, 10³ s⁻¹, and 10⁴ s⁻¹. The measured true shear viscosity may decrease by one to two decades across this range, but the absolute values are grade-specific and cannot be transferred between rubber-modified and unfilled ABS formulations without verification. Thermal stability controls are essential because the polybutadiene phase can crosslink and the SAN matrix can degrade above roughly 250 °C. Testing laboratories typically use a nitrogen purge, preheat the charge for 5 min to 10 min, and discard any run showing pressure drift. The corrected viscosity points are fitted to the Cross-WLF or Carreau-Yasuda model for downstream injection-molding simulation. Without a pressure sweep, the pressure-dependent viscosity coefficient may be estimated from literature, but published data for a specific ABS formulation are frequently unavailable.

Specimen preparation and reporting discipline are inseparable from the measured values. Pellets are dried and charged directly into the capillary barrel; regrind and trimmed parts require the same drying and homogenisation. Non-uniform particle packing changes the thermal history during preheat, so laboratories use a defined number of compression strokes and a fixed preheat interval. When the data are reported, the correction state must be explicit. A true viscosity value obtained after Bagley and Rabinowitsch correction is not interchangeable with an apparent viscosity from a shorter die. Interlaboratory comparisons of the same ABS grade can differ by 10 % to 30 % when the correction state is omitted.

Why Melt Flow Index Alone Cannot Resolve Shear Thinning in ABS

Melt flow index remains the most common release test because it is rapid and requires no curve-fitting. In ISO 1133-1:2022, the extrusion plastometer consists of a heated cylinder with a 9.55 mm bore and a die of 2.095 mm diameter and 8.000 mm length. A defined piston load forces molten ABS through the die after a specified preheat period, and the extrudate mass or volume is recorded over timed intervals. For ABS extrusion grades, the commonly referenced condition is 220 °C with a 10.0 kg load; higher-flow injection-molding grades may be tested at 230 °C with a 3.8 kg load. ASTM D1238-23 provides equivalent Procedure A and Procedure B instructions, but the condition must appear in the reported value because MFR values at different loads cannot be interconverted without a viscosity curve.

The result is expressed in g/10 min or cm³/10 min. It is not a shear viscosity and does not provide the shear-rate dependence required for mold-filling calculations. Two ABS batches may have identical MFR at the low shear stress of the melt index instrument but diverge at 10³ s⁻¹ because of different SAN molar mass distribution, rubber particle size, or grafting efficiency. On production injection presses, this divergence may appear as variable gate freeze-off, pressure-limited filling, or inconsistent weld-line strength despite stable incoming MFR. Consequently, MFR is treated as a lot-to-lot consistency indicator rather than a design value.

Parallel-plate oscillatory rheometry is used when the zero-shear viscosity, terminal relaxation behavior, or linear viscoelastic structure is under investigation. A compression-molded ABS disk of 25 mm diameter and 1.0 mm to 1.5 mm thickness is loaded between nitrogen-purged plates at 220 °C or 230 °C. A strain sweep is performed first to identify the linear viscoelastic region; a strain amplitude below 5 % is common for ABS, but pigments, fillers, or high rubber content can lower the limit. Frequency sweeps from 0.01 rad/s to 100 rad/s record complex viscosity, storage modulus G′, and loss modulus G″. The Cox-Merz rule permits complex viscosity to be compared with steady shear viscosity for homogeneous melts; for ABS, the rule may fail at high strain or frequency because the grafted polybutadiene phase introduces an apparent yield-like response. The crossover frequency where G′ equals G″ shifts to lower frequency with broader SAN molar mass distribution and is used as a comparative structural indicator. This method does not replace capillary rheometry at injection-molding shear rates but can resolve batch differences hidden by single-point MFR.

Moisture, Thermal Residence, and Rubber-Phase Architecture Factors

Moisture is the most common cause of non-reproducible ABS viscosity data. In high-humidity storage, ABS can absorb more than 0.2 % moisture by mass. During capillary or melt-index testing, the moisture vaporises and reduces the observed pressure drop, producing bubble-filled extrudate and an apparent viscosity that is lower than the true value. ABS is therefore dried in a desiccant dryer at 80 °C to 90 °C for 2 h to 4 h to a residual moisture below 0.05 %; a desiccant-bed dew point of −40 °C or lower is typical on production and laboratory dryers. Coulometric Karl Fischer analysis per ISO 15512:2019 verifies the moisture level before high-precision rheological work. Thermal residence is an equal constraint. Above 240 °C, prolonged heating can crosslink the rubber phase and degrade the SAN matrix, producing either an increase or decrease in measured viscosity depending on the dominant mechanism. Rheologists limit preheat time to 5 min to 10 min and confirm thermal stability by repeated runs on the same charge.

Rubber-phase architecture also changes the measured value independently of the SAN matrix. Larger polybutadiene particles can reduce low-shear viscosity and modify the zero-shear plateau; higher graft density tends to increase melt elasticity and die swell. These effects appear in capillary entrance pressure loss, oscillatory phase angle, and extrudate swell, but they are not visible in a single-point MFR. A complete ABS viscosity investigation therefore includes SAN molar mass, rubber content, grafting efficiency, and particle-size data when batch variation is suspected.

If Processors Replace Capillary Data with MFR Only, Which Processing Boundaries Are Missed?

If a molding operation relies only on MFR, pressure loss across the sprue, runner, and gate cannot be calculated from a shear viscosity curve. Injection-molding simulation software requires Cross-WLF parameters from capillary data, including zero-shear viscosity, critical shear stress, power-law index, and temperature dependence. Without these parameters, the predicted fill pressure for a thin-wall part with wall thickness below 1.2 mm can be understated because the shear rate at the gate frequently exceeds 10⁴ s⁻¹. A production-scale press with clamp force above 1,000 kN can still short-shot if the ABS grade has acceptable MFR but excessive high-shear viscosity. Hot-runner pressure drop, shear heating, and gate freeze-off are also invisible in a melt index test. Therefore capillary rheometry at the conversion shear rate is the minimum data set for process design, while MFR serves only as an incoming quality check.

On a compounding line, the laboratory capillary curve is supplemented by in-process slit-die or gear-pump rheometer data. A slit die mounted downstream of a twin-screw extruder with L/D 40:1 and screw diameter 40 mm to 75 mm records melt pressure at two positions and melt temperature at the die exit. The shear viscosity is calculated from the slit pressure drop after correcting for wall shear stress and non-Newtonian flow. The shear-rate window in this configuration is usually 50 s⁻¹ to 2,000 s⁻¹. Pressure rise across a melt-filtration screen pack is simultaneously monitored to detect rubber-particle agglomerates or unmelted SAN domains. When the slit-die data are compared with laboratory capillary results, the offset between the two shear-rate ranges is used to verify the consistency of the viscosity correction factors rather than to merge the data into a single curve.