Verifying PC/ABS Ratio by DSC and TGA Before You Commit to a Batch is a thermal identity gate for acrylonitrile butadiene styrene (ABS) and PC/ABS blends. ABS is a two-phase terpolymer composed of an SAN matrix and grafted polybutadiene rubber particles; typical acrylonitrile content ranges from 15 wt% to 35 wt%, butadiene from 5 wt% to 30 wt%, and styrene from 40 wt% to 60 wt%. These compositional windows are not fixed by a single ISO or ASTM grade callout, and the same melt flow index can be achieved with different monomer ratios, SAN molecular weight, or additive packages. In PC/ABS blends, a common PC content range is 40 wt% to 70 wt%, but the exact ratio is supplier-specific and often protected by compounding know-how. A batch of ABS or PC/ABS that has drifted from its intended ratio may still pass incoming melt flow index screening under ISO 1133-1:2022 or ASTM D1238, because melt flow index is a single-point viscosity indicator and cannot resolve phase composition. For this reason, thermal analysis before silo acceptance or extrusion start-up uses DSC to assign glass transition events and TGA to resolve the high-temperature mass-loss envelope.

When a PC/ABS batch is accepted on melt flow index alone, a PC-rich drift may be masked by a lower-viscosity SAN or by a processing aid. The result appears downstream as a viscosity shift during filling, not as an out-of-spec melt flow value. DSC and TGA provide orthogonal data: DSC detects the polybutadiene rubber phase and the SAN-rich/PC-rich glass transition separation, while TGA detects the degradation mass-loss envelope and char residue. The two methods together are used to verify PC/ABS ratio before batch commitment because neither method alone provides a complete ratio in filled, flame-retardant, or regrind streams.

Why does an undefined PC/ABS ratio alter molded part failure modes before melt appearance changes?

An undefined PC/ABS ratio changes molded part failure modes before melt appearance changes because the ratio controls low-shear viscosity, shear thinning, and phase continuity. On a 1500 kN hydraulic clamp injection molding machine with a multi-cavity hot-runner tool, a PC-rich shift at fixed barrel settings reduces flow length and moves the filling pressure requirement upward. Short shots, gloss variation, and knit-line strength loss can appear before the melt visually degrades or before the barrel temperature alarm triggers. Notched Izod impact measured under ASTM D256 or ISO 180 is especially sensitive to the ABS rubber phase and PC phase morphology; a blend that retains room-temperature impact can lose low-temperature ductility if the PC-rich phase becomes co-continuous or if the rubber phase is diluted. Heat deflection temperature measured at 1.82 MPa under ASTM D648 or ISO 75-1 rises with PC content, but the exact shift depends on SAN acrylonitrile content, rubber particle size, and flame-retardant package. Tensile yield stress under ISO 527-2 or ASTM D638 is a less sensitive indicator of moderate ratio drift; published data for the exact mechanical shift across all pigment and additive grades is limited.

PC/ABS processing melt temperature is commonly 250 °C to 280 °C, while ABS alone processes at 220 °C to 250 °C. A ratio error forces the molder to operate closer to the PC degradation threshold at 280 °C, increasing the risk of hydrolysis if the resin is not dried below 0.02 wt% moisture. The result is splay, silver streaks, and molecular weight loss in the PC-rich phase. In compounding, a twin-screw extruder with 40:1 L/D and side-feeder for rubber or fillers has a residence time distribution that changes when PC content shifts, because PC increases melt viscosity and alters the screw fill length. If the ratio is not verified, the compounder may adjust screw speed or barrel temperature to a false optimum, producing pellets that pass pellet size and melt flow tests but carry a latent morphology defect into molding.

Rheological confirmation is often performed after the DSC/TGA screen because the zero-shear viscosity of PC/ABS rises non-linearly with PC content. At shear rates between 100 s⁻¹ and 1000 s⁻¹, the difference between a 45 wt% PC blend and a 55 wt% PC blend may be small in melt flow index but large in pressure drop through a hot runner. The phase morphology also changes: at low PC content, ABS forms the continuous phase; at high PC content, PC becomes continuous or co-continuous. This transition affects solvent resistance, thermal expansion, and notched impact. Published data for the exact phase inversion concentration in commercial PC/ABS blends is limited because rubber particle size, SAN acrylonitrile content, and compatibilizer loading move the inversion point.

The operational boundary for drying and melt temperature should be treated as part of the ratio verification workflow. PC-rich blends absorb moisture more rapidly than ABS; at relative humidity above 60%, open storage of PC/ABS pellets for more than 4 h can raise moisture above the 0.02 wt% threshold. Pre-drying at 80 °C to 100 °C for 2 h to 4 h is standard, but drying time must be extended for cold warehouse stock or regrind. Avoid amine-based antistatic or nucleating packages in PC-rich blends because they can catalyze hydrolysis and depolymerization at melt temperatures above 260 °C. The DSC/TGA ratio screen should be repeated after drying if splay or gas evolution is observed during processing.

DSC transition assignment for ABS, SAN, and PC-rich phases in a PC/ABS blend

A representative pellet sample is cut to 5 mg to 10 mg and sealed in an aluminum pan for DSC. A heat-cool-heat protocol from 30 °C to 250 °C at 10 °C/min, quench to -80 °C, and second heating at 10 °C/min is used to erase pelletizing and molding history. The instrument is calibrated with indium and the heat flow signal normalized by sample mass. In ABS, the polybutadiene rubber phase appears as a step transition between -80 °C and -50 °C, depending on rubber crosslink density and cooling rate. The SAN matrix Tg appears between 105 °C and 115 °C, shifted by acrylonitrile content and molecular weight. In PC/ABS, the PC-rich phase Tg is commonly observed between 138 °C and 150 °C, but partial miscibility between PC and SAN can lower the PC-rich Tg and broaden the transition. DSC cannot report a mass ratio because the heat capacity step at each Tg is not linearly proportional to phase mass in a partially miscible blend; however, the presence of the polybutadiene Tg confirms ABS, and the separation between SAN-rich and PC-rich transitions indicates whether the blend is sufficiently phase-separated for the intended mechanical response. Modulated DSC with a ±0.5 °C modulation amplitude every 60 s can separate reversing heat capacity from enthalpy relaxation and is useful when the SAN-rich and PC-rich transitions overlap between 120 °C and 140 °C.

One practical failure occurs when a batch is cooled only to -20 °C; the polybutadiene Tg at -80 °C is missed, and the operator misinterprets the SAN transition as the only amorphous event. For that reason, liquid nitrogen cooling or an intracooler capable of reaching -80 °C is required for ABS and PC/ABS screening. A second practical limitation is that molded parts may contain residual orientation or moisture; the first heating curve is not used for Tg comparison because it contains enthalpy relaxation at the SAN Tg. Only the second heating after controlled quench is reported, and the thermal history should be recorded in the batch test report.

Sample preparation matters because a pellet surface that has absorbed moisture can show an evaporation endotherm below 150 °C that obscures the SAN Tg. Pellets should be cut with a clean blade at room temperature, and the cut surfaces should be representative of the bulk, not the outer oxidized layer. For highly pigmented grades, pigmentation does not normally shift Tg but can alter the baseline if the pigment undergoes a thermal event. In those cases, a solvent extraction or ashing step is not performed before DSC because it would remove the phase morphology that is the target of the verification.

The DSC method can be extended to 300 °C if the PC-rich phase contains a high-temperature additive package, but standard aluminum pans require a lid and a pinhole to avoid deformation. The upper temperature should not exceed the onset of thermal degradation in the first heating, because the first heating curve still contains volatile loss; degradation exotherms or endotherms in the first heating are not used for Tg identification.

Attenuated total reflectance FTIR can confirm the presence of the PC carbonyl at about 1775 cm⁻¹, but it is not a primary ratio method because the carbonyl absorbance depends on surface contact, filler, and pigment. DSC and TGA are preferred for batch verification because they respond to bulk thermal transitions and mass-loss events rather than surface composition.

When TGA separates ABS degradation events from PC char residue before ratio release

TGA complements DSC by measuring the mass-loss envelope during programmed heating. A representative cut pellet mass of 10 mg to 20 mg is loaded in a platinum or alumina pan and heated from 40 °C to 800 °C at 10 °C/min under nitrogen at 50 mL/min to 60 mL/min, followed by an air switch to oxidize residual char. Under nitrogen, neat ABS degrades mainly between 350 °C and 480 °C with a residue below 5 wt%; neat PC degrades mainly between 450 °C and 550 °C and leaves approximately 20 wt% to 25 wt% char. In a PC/ABS blend, the derivative weight loss curve shows overlapping events because SAN and polybutadiene decomposition overlaps early PC chain scission. The ratio estimation therefore requires a calibration set of known PC/ABS ratios run under the same purge gas, heating rate, sample mass, and pan type. Char residue at 700 °C in nitrogen is a practical screening output for neat unfilled and unflame-retarded blends. If the blend contains glass fiber, mineral filler, carbon black, brominated flame retardants, antimony oxide, or phosphate esters, a simple char-to-PC correlation is invalid. In those cases, TGA curve subtraction or chemometric modeling is required, and published data for the specific additive configuration may be limited.

TGA method precision is affected by sample geometry, pan contact, buoyancy, and baseline drift. A blank pan subtraction and a buoyancy correction are required before the mass-loss curve is interpreted. If the sample is too large, self-heating or diffusion-limited decomposition delays the onset temperature and distorts the DTG peak shape. Conversely, a sample below 5 mg may have poor signal-to-noise for minor mass-loss steps. For a lot with unknown regrind, TGA alone cannot distinguish a PC-rich blend from a blend containing high-char ABS grade or a flame-retardant synergist; the DSC polybutadiene and SAN/PC glass transition data must be integrated with the TGA residue.

TGA derivative peak deconvolution uses a constrained peak fitting algorithm, but the number of fitted peaks must be justified by the known chemical composition. A neat ABS may present two overlapping derivative peaks, while a filled ABS may present five or more events. Without a reference lot of the same grade, assigning individual peaks to PC or SAN is unreliable. The ratio verification uses the whole curve shape, not a single inflection point.

At the scale of a production plant, verification before batch commitment is executed on a composite sample from top, middle, and bottom of the bulk container or silo because ratio stratification can occur in regrind streams with different particle sizes. The pellets are dried in a desiccant dryer with a -40 °C dew point until moisture is below 0.02 wt%; PC is hygroscopic, and moisture above 0.02 wt% in PC-rich phases causes hydrolysis and splay at melt temperatures above 260 °C. DSC is run first because the glass transition fingerprint detects gross contamination such as styrene-rich SAN substitution or PC-rich regrind drift within a short cycle time. TGA is run second on the same pellet stock because the high-temperature mass-loss envelope and char residue correlate with the PC/ABS ratio in unfilled grades. If the TGA char value falls outside the release band, the batch is held for capillary rheometry under ISO 11443 or ASTM D3835 and mechanical testing before molding. A batch that passes DSC and TGA can be cleared for extrusion or injection molding only when the additive package is known and the char calibration is confirmed for that specific grade.

In recycling plants where ABS and PC/ABS parts are mixed, a single DSC run on regrind can show a polybutadiene Tg and a PC-rich Tg, but the PC-rich Tg may be broad because the PC came from a different molecular weight grade. TGA char residue then identifies whether the regrind is ABS-rich or PC-rich enough to be routed to a PC/ABS compounding line. If the char residue is below 10 wt% in nitrogen at 700 °C, the stream is treated as ABS-rich and is not committed to a PC/ABS blend without correction. If the char residue is above 15 wt%, the stream may be PC-rich and is separated for PC/ABS reuse. The exact thresholds are plant-specific and depend on the target blend ratio and the acceptable property window. Batches containing flame-retardant or filled ABS require additional ash or elemental analysis; TGA char residue alone is not a release criterion.