Thermogravimetric analysis (TGA) is the essential experimental technique for quantifying the thermal stability of PEI-modified PEEK blends used in aerospace, electronics, and other high-performance engineering applications. This article presents a complete experimental framework for PEI/PEEK systems, covering sample preparation, instrument parameters, measured quantities, data interpretation, and typical degradation behavior. Materials engineers and researchers can use this guide to design reproducible TGA experiments and to correlate polyetherimide loading with onset decomposition temperature, peak degradation rate, and char yield.
Technical Overview
PEI-modified PEEK refers to a binary blend of polyetheretherketone (PEEK), a semi-crystalline aromatic thermoplastic, with polyetherimide (PEI), an amorphous high-temperature resin. The two polymers are melt-miscible over a broad composition range, and blending PEI into PEEK raises the glass transition temperature of the amorphous phase, improves dimensional stability above the PEEK crystalline relaxation, and modifies crystallization kinetics during molding. Because these blends are routinely specified for service at elevated temperatures, a rigorous understanding of their thermal decomposition behavior is critical for material qualification, processing window definition, and fire-performance assessment.
Thermogravimetric analysis is the primary method for characterizing this behavior. In a TGA experiment, a small specimen is heated at a controlled rate under a regulated atmosphere while a high-precision balance continuously records mass change. The resulting thermogram reveals the onset of degradation, the number and location of degradation stages, and the mass of carbonaceous residue (char) remaining at the final temperature. For PEI/PEEK systems, the experiment is typically conducted under inert nitrogen to establish intrinsic stability, with complementary runs in air to evaluate oxidative stability.
- Neat PEEK typically exhibits 5% mass-loss temperatures near 550 to 575 degrees Celsius and a single-step degradation profile under nitrogen, with char yields of roughly 45 to 55 percent.
- Neat PEI degrades at slightly lower onset temperatures but forms a highly aromatic, thermally stable char residue.
- PEI/PEEK blends display intermediate degradation characteristics, and the residual char often exceeds the simple rule-of-mixtures prediction, reflecting synergistic char formation between the imide-rich and ether-ketone-rich chains.
- The DTG (derivative thermogravimetric) curve is used to identify the temperature of maximum mass-loss rate and to resolve overlapping degradation events.
Testing Samples
Neat Resin Reference Materials
Two unfilled reference materials are tested to establish baseline degradation behavior: virgin PEEK resin and virgin PEI resin. The reference specimens must be processed under the same thermal history as the blends wherever possible, because prior melt processing, residence time, and cooling rate can measurably shift decomposition temperatures. Reference runs also serve as system suitability checks, confirming that the instrument reproduces literature values for well-characterized polymers before blend specimens are analyzed.
PEI/PEEK Binary Blends
The core experimental matrix consists of PEI/PEEK blends prepared at defined weight ratios, commonly 10/90, 20/80, 30/70, and 40/60. The blends are produced by melt compounding in a co-rotating twin-screw extruder at barrel temperatures between 360 and 400 degrees Celsius, followed by pelletizing. The extrudate should be visually inspected and, where required, sectioned for microscopy to confirm homogeneity, since poorly dispersed domains can produce bimodal or broadened DTG peaks that complicate interpretation.
Sample Conditioning and Mass Requirements
All specimens, including references and blends, are conditioned to remove absorbed moisture prior to testing. PEEK and PEI are only mildly hygroscopic, but even small moisture contents generate an early mass-loss step that can distort baseline correction near the degradation region. A standard conditioning protocol is as follows:
- Vacuum dry pellets or cut specimens at 150 degrees Celsius for a minimum of 4 hours.
- Cool under vacuum in a desiccator and weigh test portions immediately after removal.
- Use a specimen mass of 5 to 10 milligrams, selected so that the sample forms a thin, uniform layer in the platinum or alumina pan.
- Record the exact initial mass and specimen geometry in the test report for traceability.
Testing Items
The experiment is designed to extract a defined set of quantitative metrics from each thermogram. Every reported value should be accompanied by the atmosphere, heating rate, and specimen mass under which it was obtained, because these parameters strongly influence apparent degradation temperatures. The standard testing items are:
- TG curve: percent mass remaining as a function of temperature from ambient to the final temperature.
- DTG curve: first derivative of the TG signal, used to locate degradation rate maxima and resolve multi-step processes.
- T5%: the temperature at which 5% mass loss occurs, conventionally reported as the onset of significant degradation.
- T10%: the temperature at which 10% mass loss occurs, providing a secondary stability index.
- Tmax: the DTG peak temperature, representing the temperature of maximum degradation rate.
- Char yield: the residual mass at 700 and 800 degrees Celsius under nitrogen, an indicator of condensed-phase flame resistance.
- Residual mass in air at the same temperatures, reflecting oxidation and char burn-off behavior.
- Optional kinetic parameters: apparent activation energy as a function of conversion using multi-heating-rate methods such as Kissinger or Flynn-Wall-Ozawa analysis.
Testing Methods
Instrumentation and Calibration
A research-grade thermogravimetric analyzer with a horizontal or top-loading microbalance is used. Before the experimental campaign, the instrument must be calibrated for temperature using Curie-point standards or the melting transitions of certified reference materials, and the balance must be verified with certified weights. The furnace and balance purges are set to high-purity nitrogen at a combined flow of approximately 50 to 60 milliliters per minute, and leak checks are performed to ensure oxygen ingress below instrument specification.
Temperature Program and Atmosphere
The standard program heats the specimen from 30 to 800 degrees Celsius at 10 degrees Celsius per minute under nitrogen. Where moisture removal is performed in situ, an initial isothermal hold at 120 degrees Celsius for 10 to 20 minutes can be included, with the mass at the end of the hold defined as the effective initial mass. For oxidative stability, the identical program is repeated under dry synthetic air, allowing direct comparison of onset temperatures and char oxidation behavior between inert and oxidative atmospheres. For kinetic studies, additional runs are performed at 5, 15, and 20 degrees Celsius per minute using freshly loaded specimens.
Data Processing and Reporting
Raw mass data are baseline-corrected for buoyancy effects using an empty-pan reference run collected under the same program. Onset temperatures are determined by the intersection of the extrapolated pre-degradation baseline with the tangent to the steepest portion of the TG curve, while T5% and T10% are read directly from the mass-loss curve. The DTG curve is computed and smoothed only with documented, non-distorting parameters. Each formulation should be tested in triplicate, and results are reported as mean values with standard deviations. When interpreting PEI/PEEK thermograms, note that a single, merged degradation stage indicates co-degradation of the miscible phases, whereas separated DTG maxima suggest partial immiscibility or composition heterogeneity.
Applications
The thermal stability data generated by this experiment directly support the qualification and deployment of PEI-modified PEEK in demanding environments. Typical application areas include:
- Aerospace structural brackets, clips, and interior components requiring FST (fire, smoke, toxicity) compliance and long-term heat resistance.
- Electrical connectors, high-temperature insulators, and surface-mount parts that must survive solder reflow and reflow-adjacent thermal excursions.
- Automotive under-hood components such as sensor housings, transmission seals, and turbocharger peripheral parts.
- Oil and gas equipment including seal energizers, backup rings, and insulating components exposed to combined heat and pressure.
- Medical and sterilization-relevant hardware where repeated autoclave exposure demands stable thermal margins.
- Additive manufacturing feedstock, where PEI/PEEK-type blends are processed at high nozzle and chamber temperatures and where TGA data define safe processing windows.
In each case, the char yield measured under nitrogen provides a rapid screening parameter for condensed-phase flame resistance, while the oxidative run establishes the practical upper service boundary in air. Formulators use the composition dependence of T5%, Tmax, and char yield to select the PEI fraction that balances stiffness retention, processability, and decomposition margin.
FAQ
Why is the TGA experiment performed under nitrogen first?
Nitrogen provides an inert environment that reveals the intrinsic thermal decomposition behavior of the polymer system, free from oxidative reactions. This isolates the material's fundamental stability and allows meaningful comparison with published literature values. A complementary air run is then used to evaluate thermo-oxidative stability, which is usually the more realistic condition for service in ambient atmospheres.
How does PEI content affect the thermal decomposition of PEEK?
Increasing PEI content generally shifts the degradation profile toward that of neat PEI, with onset temperatures in the blend falling between those of the two neat resins. In many formulations the char yield increases beyond the weighted average of the components because the aromatic imide structure of PEI promotes carbonaceous residue formation. The exact trend depends on compounding quality, thermal history, and the presence of residual moisture or additives, which is why triplicate testing under standardized conditions is essential.
What specimen mass and heating rate should be used?
A specimen mass of 5 to 10 milligrams and a heating rate of 10 degrees Celsius per minute are the widely accepted defaults for comparative polymer TGA. Smaller, thinly spread specimens reduce thermal lag and mass-transfer limitations, sharpening the DTG peak. If kinetic analysis is planned, multiple heating rates, typically 5, 10, 15, and 20 degrees Celsius per minute, must be run so that model-free methods such as Flynn-Wall-Ozawa can be applied.
Why must the samples be dried before testing?
Both PEEK and PEI absorb small quantities of atmospheric moisture, and this moisture desorbs between room temperature and approximately 150 degrees Celsius. Without drying, an early mass-loss step appears in the thermogram, which complicates baseline construction and can shift apparent onset temperatures. Vacuum drying at 150 degrees Celsius followed by testing directly from a desiccator eliminates this artifact.
What does the char yield tell us about the blend?
Char yield, reported as the residual mass at 700 to 800 degrees Celsius under nitrogen, indicates the polymer's tendency to form thermally stable carbonaceous residue rather than volatilize completely. Higher char yields correlate with improved condensed-phase flame retardancy, lower heat release, and better performance in FST-regulated applications. PEI modification is often adopted specifically because it raises the effective char yield of the PEEK matrix.
How are single-step and multi-step DTG curves interpreted?
A single DTG maximum indicates that the blend degrades as one coexistent phase, consistent with good miscibility between PEI and PEEK. Multiple resolved maxima can indicate phase separation, preferential degradation of one component, or overlapping chain-scission and char-forming reactions. When multi-step behavior appears, it should be cross-checked against blend morphology and, if necessary, supported by multi-heating-rate kinetic analysis to separate mass-transfer and reaction-controlled events.