Keywords: Polyimide, Iso-stress, Iso-strain, DMA, TGA, DSC
RH153
Abstract
Polyimides (PI) are known for their thermal stability, mechanical integrity with flexibility, and electrical insulation. Their properties make them desirable for electronics applications and an increase in demand is expected as the electronics market experiences continued advancements. While PI films are known to be robust and lack any significant thermal transitions over a wide temperature range, they do possess subtle property changes at elevated temperatures. These secondary transitions are associated with chain-level segmental mobility and can have an impactful role on properties such as flexibility and damping. With the tight material property tolerances associated with electronic assemblies, it is imperative to characterize these nuanced transitions.
In this note, a commercial PI film is characterized by thermogravimetric analysis (TGA), differential scanning calorimetry (DSC) and dynamic mechanical analysis (DMA). The decomposition by TGA helps determine the upper processing temperature limit and DMA can detect a secondary transition in the PI film based upon temperature ramps that probe the dynamic moduli of the material. DSC has been included to show how heat flow analysis gives little insight into PI thermal behavior due to the fully amorphous and rigid molecular structure of the polymer. In addition, transient iso-stress and isostrain DMA tests were performed to identify the samples strain and stress response, respectively, to mimic application stress and strain. A thickness dependence response of the PI film to these stress and strains is also discussed in relation to its application use.
Polyimide (PI) use in the electronics industry is experiencing strong and sustained growth, driven by the exceptional thermal stability, electrical insulation, and mechanical flexibility it offers. As of 2024,the global PI market was valued at approximately 5.46 billion USD, with projections reaching 7.60 billion by 2029, reflecting a Compound Annual Growth Rate (CAGR) of 6.83%. Electronics account for nearly 37% of total PI consumption, primarily in flexible printed circuit boards, Organic Light-Emitting Diode (OLED) displays, and insulation layers. Innovations in transparent and flexible PI films are further accelerating demand, especially in wearable and foldable devices, which have seen growth rates of 44% and 36%, respectively, since 2023 [1].
For many of these applications, PI films are exposed to various stresses, deformations, and elevated temperatures. While PI films are recognized as robust, high temperature application materials, they do possess nuanced thermo-mechanical responses at elevated temperatures. Understanding these subtle properties of PI films is critical as applications such as flexible electronics require small tolerances for dimensional changes. [2] Unexpected film expansion or contraction at higher temperatures can produce unwanted stresses on key components and compromise product performance and lifetime [3].
In this note, the high temperature thermo-mechanical properties of a commercial grade PI film were analyzed by TGA, DSC, and DMA. TA Instruments Discovery Series™ Instruments, shown in Figure 1, were used. TGA was used to determine the upper processing temperature limit of the PI film along with derivative analysis providing insight into the decomposition mechanisms and associated temperature ranges. Temperature ramps were performed on the DSC and DMA to illustrate the importance of thermo-mechanical methods for determining subtle material property changes with temperature. Iso-strain and iso-stress tests were further used with the DMA to determine the PI films’ response to constant strain or stress at elevated temperatures. The response of the PI films with different thicknesses is discussed in relation to their application use.
A commercially available PI film with a thickness of 125 μm was used for all experiments. For iso-strain and iso-stress tests, two additional samples with thicknesses of 75 and 50 μm were tested to examine the thickness-dependent stress and strain response, respectively.
TGA
Decomposition temperature was determined per IPC Test 2.3.40 on a Discovery TGA 5500 Thermogravimetric Analyzer. In brief, the sample was ramped from 0-150°C at 5 °C/min and held for 15 minutes to eliminate moisture. The sample was then ramped to 800 °C at 5 °C/min. The test was performed under air and nitrogen [4].
DSC
A heat/cool/heat experiment was run on a Discovery DSC 2500 Differential Scanning Calorimeter. The temperature range was 0-200 °C with ramp rates of 10 °C/min for all steps. The first heat was utilized to remove thermal history and provide a cooling and second heating curve without hysteresis effects. Therefore, the cooling and second heating only are shown for discussion.
DMA
A Discovery DMA 850 Dynamic Mechanical Analyzer was used for DMA temperature ramps, iso-strain, and iso-stress tests. All tests were performed using the dual screw film tension clamp. The length and width of the sample was 6.0 mm and 8.8 mm, respectively. The TRIOS Software calculates and uses the exact length based upon the clamp separation. The temperature ramp was conducted from room temperature to 380 °C at 0.01% strain with a frequency of 8 Hz.Iso-stress and Iso-strain tests were run from room temperature to 420 °C with a stress of 3 MPa and strain of 1%, respectively.
TGA
Figure 2 shows the TGA temperature ramps for the PI film under air (blue) and nitrogen (green). The decomposition temperature of the film per the ICP standard is the temperature at which 5% decomposition has occurred. This corresponds to a decomposition temperature of 557.2 °C in air and 556.5 °C in nitrogen. This provides an absolute upper limit for processing the film. In practice, the film is used at much lower temperatures; as shown by DMA, the upper functional limit is well below this threshold.
In addition to decomposition temperature determination, the derivative of weight loss can be examined to probe the decomposition process. In Figure 2, the derivative of the weight change as a function of temperature is plotted as the dashed curves. An obvious distinction between the experiments in different atmospheres is the presence of two weight loss events (peaks) in air and a single weight loss in nitrogen. Furthermore, only partial decomposition of ~65% is observed under the inert nitrogen atmosphere compared to full decomposition in air. It is known that the PI decomposition begins around 500 °C with imide ring cleavage followed by chain scission and aromatic ring break down. This is captured in the initial primary weight loss event on both runs. The second major weight loss event observed under air is due to the combustion of char and is not possible under the inert nitrogen atmosphere.
Thermal stability analysis from TGA is useful for determining the upper processing temperature but provides little information about a material’s behavior during application use temperatures. When developing a material’s thermal response profile, the common step after TGA is DSC heat flow analysis. This identifies any thermal transitions in the material such as the glass transition (Tg), crystallization, or melting the material may go through in typical application temperature ranges. Figure 3 shows the cooling and second heating thermograms of the PI film from a heat/cool/heat cycle DSC experiment.
The obvious feature of these DSC curves is that no transitions are present. This PI film is completely amorphous and, therefore, no melting or crystallization transitions are anticipated. In addition to this, polyimides possess highly aromatic and rigid backbones with strong intermolecular interactions which significantly hinder chain mobility [6] [7]. A consequence of this is a very high or non-distinguishable Tg. The absence of thermal transitions within these practical application temperature ranges is a feature that makes PI films useful for many applications where predictable and unchanging behavior is desired upon heating and cooling.
If the thermal behavior of this material was solely based upon heat flow and weight loss, an incorrect conclusion could be drawn that the film undergoes no thermally activated changes until ~500 °C. It is well known – for even highly rigid polymers such as PIs – that there are small (chain level) length scale motions that are activated at elevated temperatures [8]. These manifest as sub-Tg transitions, referred to as secondary transitions such as beta and gamma relaxations, and contribute to temperature-initiated dimensional changes in the material. The heat flow signal from a conventional DSC experiment is generally not sensitive enough to detect these local chain segment mobility transitions. However, these transitions do have implications on material properties such as damping, flexibility, impact resistance, and dimensional stability [9] [10].
DMA
The above listed properties play a role in the processability and performance of the polymer. This necessitates the need for a characterization technique sensitive enough to probe these subtle, but impactful transitions. One of the most common ways to study these transitions is by monitoring the dynamic moduli of the material as a function of temperature. The dynamic moduli are more sensitive compared to heat flow in detecting these subtle material property changes.
Figure 4 shows a temperature ramp of the PI film from room temperature to the upper recommended functional limit of the material (~400 °C). As opposed to the DSC heat flow profile, there are clear mechanical property changes as the material is heated. The storage modulus (E’), representing the elastic component of the dynamic modulus, decreases steadily with temperature. The absence of a significant drop in storage modulus of 2-3 decades in magnitude indicates there is no observable Tg. The loss modulus (E”), representing the viscous component of the dynamic modulus, remains relatively steady with a peak at 141 °C. This peak in the loss modulus with no significant decrease in storage modulus is characteristic of secondary transitions. This is also readily observed as a peak in Tan delta, which is defined as E”/E’ and also termed the loss or damping factor.
The identification of the secondary transition indicates the material behavior will differ when utilized above or below 141 °C. When used in a flexible electronics application, for example, the polyimide film may have non-ideal damping or flexibility properties either before or after the transition that restrict its use to specific temperature ranges.
DMA Iso-stress and Iso-strain
In addition to using dynamic testing for secondary transition identification, the DMA can also be operated in a transient mode. In this mode, a constant stress or strain is applied to the material and the material’s strain or stress response is measured, respectively. When performed at an iso-thermal temperature these tests are referred to as creep for constant stress and stress relaxation for constant strain. These tests can also be performed by applying a linear heating rate to the sample while being held at constant stress or strain. These are referred to as iso-stress and iso-strain tests, respectively. The utility of these tests arises from the ability to mimic application stresses or strains the material may encounter. Studying the materials response will allow design optimization and set temperature range limits on the end-use of the material.
Figures 5A and 5B show the iso-stress and iso-strain results, respectively, of the same PI film grade with thicknesses of 50, 75, and 125 μm. The iso-stress results (Figure 5A) are plotted as change in length (ΔL) versus temperature. The plot shows the films remain constant in length up to the temperature at which the secondary transition occurred in the dynamic temperature ramp. Past this temperature the length, or strain, of the sample increases; that is attributed to sample softening. The change in length, taken as ΔL=end length-start length is more significant in the thinner PI films compared to the 125 μm film. As the temperature exceeds 400 °C, the ΔL begins to increase sharply.
The iso-strain results in Figure 5B are plotted as force versus temperature. The force here refers to that exerted to maintain a constant strain of 1%. For all films, the maximum force occurs at the start of the test. When the sample is heated, a significant drop in force is observed when the temperature of 141 °C is reached. This is like the iso-stress behavior and identifies the importance of the secondary transition. The curves subsequently level off with a decreasing force with increasing temperature. When the upper functional limit of the films is approached, near 400 °C, all films are softened significantly.
The DMA results identify a secondary transition at 141 °C but no apparent Tg. This transition correlates to inflection points in the ΔL versus temperature and force versus temperature plots for the transient iso-stress and iso-strain tests, respectively. The transient tests also show significant dimension change and softening when 400 °C is surpassed. This identifies ~400 °C as the upper functional limit of the film. Above this temperature the mechanical integrity is rapidly lost. The same grade of PI film also has a thickness dependence; thinner films experience greater dimensional change and softening during constant force and strain tests. From an application standpoint these results suggest a thicker film should be used when mechanical integrity is needed at elevated temperatures. When choosing the thickness, there is a tradeoff between high temperature mechanical stability and low temperature rigidity.
The global market for PIs in the electronics industry is anticipated to continue growing due to technological advancements and increased use of flexible printed circuit boards, OLEDs, and wearable/foldable devices. PIs are known for their thermal stability and mechanical integrity but do possess subtle property changes with temperature.
Weight loss analysis by TGA identified the decomposition temperature of the PI film under air and nitrogen to be ~557 °C. This provides an absolute upper processing limit for the PI film. Derivative analysis of the TGA weight loss curves provided insight into the degradation behavior under the various atmospheres. The initial breakdown of the material is similar, but under the oxidative air atmosphere the char formation undergoes combustion, and complete weight loss is observed – unlike the partial weight loss under nitrogen.
DSC heat flow analysis of the PI film showed no thermal transitions within typical application temperature ranges. This is attributable to the PI film being completely amorphous and the underlying rigid molecular structure hinders chain mobility to an extent a Tg is not expected within the range or undefinable. The lack of transitions in the DSC thermograms does not mean the PI films lack any property changes with temperature.
A DMA temperature ramp monitoring the dynamic moduli as a function of temperature shows a secondary transition at 141 °C along with softening of the film. This secondary transition is known to influence key properties such as flexibility and damping in these types of polymers.
Transient iso-stress and iso-strain can mimic application-relevant stress and strains on the material at elevated temperatures. The significant dimensional change and softening after the secondary transition temperature identifies the relevance of this subtle transition on the polymer’s high temperature behavior. A thickness dependence was also observed in the films where the thicker film maintained its mechanical integrity at higher temperatures compared to thinner films. Also, films showed a rapid softening and dimensional change after 400 °C. This temperature can be taken as the upper functional limit of the material.
Related Resources
References
- “Polyimides (PI) Market SIZE & SHARE ANALYSIS – GROWTH TRENDS & FORECASTS UP TO 2029 Source: https://www.mordorintelligence.com/industry-reports/polyimidesmarket,” Mordor Intelligence, 2025.
- K. Harris, A. Elias and H. Chung, “Flexible electronics under strain: a review of mechanical characterization and durability enhancement strategies,” Journal of Materials Science, pp. 2771-2805, 2016.
- Z. Zhao, H. Fu, R. Tang, B. Zhang, Y. Chen and J. Jiang, “Failure mechanisms in flexible electronics,” International Journal of Smart and Nano Materials, vol. 14, no. 4, pp. 510-565, 2023.
- The Institute for Interconnecting and Packaging Electronic Circuits, “2.3.40,” IPC-TM-650 Test Methods Manual, p. 1, 1995.
- Q. Li, B. Chen, S. Yao, C. Sang, L. Lu, S. Dong, H. Cao, Z. Si and P. Qin, “The thermal behavior and pyrolysis mechanism of a polyimide gas separation membrane,” RCS Sustainability, vol. 2, p. 3916, 2024.
- P. Xiao, X. He and Q. Lu, “Exceptionally High-Temperature-Resistant Kapton-Type Polyimides with Tg > 520 °C: Synthesis via Incorporation of Spirobis(indene)-bis(benzoxazole)-Containing Diamines,” Polymers, vol. 17, no. 7, p. 832, 2025.
- M. Keating, “High glass transitions of high-performance thermoplastics,” Thermochimica Acta, vol. 319, pp. 201-212, 1998.
- K. Menard and N. Menard, “Dynamic Mechanical Analysis in the Analysis of Polymers and Rubbers,” in Encyclopedia of Polymer Science and Technology, 2015, pp. 1-33.
- X. Yu, W. Liang, C. Jianhua and D. Wu, “Mixed Rigid and Flexible Component Design for High-Performance Polyimide Films,” Polymers, vol. 9, no. 9, p. 451, 2017.
- G. Ragosta, M. Abbate, P. Musto and G. Scarinzi, “Effect of the chemical structure of aromatic polyimides on their thermal aging, relaxation behavior and mechanical properties, Journal of Materials Science, vol. 47, pp. 2637-2647, 2012.
Acknowledgement
The author thanks Andrew Janisse for his assistance with the TGA experiments run for this work.
For more information or to request a product quote, please visit www.tainstruments.com to locate your local sales office information.
This paper was written by Mark Staub, PhD.
TA Instruments, Discovery, Discovery Series, and TRIOS are trademarks of Waters Technologies Corporation.
Click here to download the printable version of this application note.