Keywords: TMA-RH, Thermomechanical Analyzer, relative humidity, hygroscopic expansion, polyimide
TA499
Abstract
A material’s dimensional changes as a function of temperature and relative humidity are key factors in product design. In devices with multiple layered materials, such as those in electronics, a mismatch in thermal or hygroscopic expansion can lead to internal stresses and potentially part failure. Polyimides are polymers used heavily in the electronics industry. This note demonstrates the use of a TA Instruments™ Discovery™ TMA 450 RH to measure hygroscopic expansion of polyimide films from relative humidity levels of 0 %RH to 80 %RH at isothermal temperatures of 25 °C, 50 °C, and 75 °C.
A thermomechanical analyzer (TMA) is an analytical instrument that measures dimensional change of a sample as a function of time, temperature, and atmosphere [1]. A probe is placed in contact with a sample, and a small mechanical load is applied while any changes in the linear dimension are recorded. A TMA is commonly used to determine the coefficient of linear thermal expansion (CLTE) of a material. Most materials will expand as they are heated, and a TMA can be used to measure that expansion as a function of temperature. This is a critical property to consider when selecting materials in product design. If multiple materials of dramatically different CLTEs are within the same device, large mismatches in expansion during heating can create internal stresses, which can ultimately compromise performance or result in total part failure [2].
In addition to expanding with temperature, materials can also respond differently to humidity. Some materials are very hydrophilic and will readily take on water, while other, more hydrophobic materials will resist water uptake. This can influence many material properties, including dimensional stability. Dimensional increases because of moisture sorption is called hygroscopic expansion [3]. Hygroscopic expansion strains a material and produces internal stress. Understanding that material response can be crucial for final product performance. In the electronics industry, components combine metals, ceramics, and hygroscopic polymers, all of which can have different dimensional responses to temperature and humidity. Accurate determination of these material properties is key to product performance and longevity. Mismatches in expansion can warp printed circuit boards (PCBs), crack solder joints, delaminate interfaces, and accelerate electrical failures [4].
The Discovery TMA 450 RH (TMA-RH) is a standalone instrument that can measure linear dimensional change, like a traditional TMA, but is equipped with a humidity chamber so the relative humidity (%RH) can be controlled (Figure 1) [5]. Using this instrument, a material’s dimensional properties can be studied as a function of humidity level as well as temperature.
Polyimides (PI) are high-performance aromatic polymers that are widely used in the electronics industry. They have high thermal stability, chemical resistance, mechanical strength, and favorable dielectric properties [4] [6]. These properties make them attractive materials to use as substrates for flexible PCBs, in flexible copper‑clad laminates, and as high‑temperature insulating tapes and wire coatings [7] [8]. PI films can be tailored to tune thermal, mechanical, electrical, and moisture uptake properties. It is important to know the dimensional stability of the PI grade to be used in a high performance electronic, not just as a function of temperature, but also as a function of relative humidity. This work demonstrates how hygroscopic expansion of polyimide films can be measured using a Discovery TMA 450 RH.
Commercially available polyimide samples were studied using the Discovery TMA 450 RH. Films of the same polymer grade but with two different thicknesses were tested to explore the effect of sample bulk volume on the hygroscopic sorption of PI films. The thinner of the two samples had a thickness of 80 μm and is designated as PI‑80. The other film was 130 μm thick and is designated PI-130. Both film samples were cut with a die into rectangular pieces 2.55 mm wide and held in the TMA‑RH Film/Fiber Accessory at a gap length of 24 mm (Figure 2). A constant load of 0.1 N was applied throughout testing on all samples to maintain a slight tension on the film samples and prevent kinking.
For every run, samples were initially conditioned with a drying step. Relative humidity was set to 0 %RH and the chamber temperature equilibrated at 80 °C. Samples were held at that condition for 300 minutes. At the completion of the full 300-minute drying step, the instrument equilibrated at the testing temperature for the given experiment. Experiments were performed at 25 °C, 50 °C, or 75 °C. Once at the target experimental temperature, the samples were conditioned in an isothermal hold for an additional 60 minutes. The preliminary conditioning is programmable in TRIOS™ Software, which controls the TMA-RH (Figure 3). The custom method created for this work included the conditioning steps followed by a “Measure” segment just before the beginning of the experimental step. The Measure step in TRIOS Software records the length of the sample at that point, which is then used as the initial length when determining normalized dimensional change from that point forward in the experiment. This means that the normalized dimensional change for every sample began at 0 μm/m when the experimental steps commenced (step 7).
The experimental steps consisted of isothermal/isohume holds at increasing relative humidity levels. Each sample was held at a single temperature throughout the experiment, and the relative humidity was stepped from 0 %RH to 80 %RH in 10 %RH increments. Samples were held at each isohume step until dimensional change was stabilized. The “Step Humidity Abort” method segment allows you to program the magnitude of %RH steps from the current relative humidity set point to a final level. It also allows for stability criteria to be set which, when met, stops the current isohume step and proceeds to the next relative humidity level. When designing a similar method for a new material, preliminary experimentation is recommended to find appropriate stability conditions. If the criteria is too low, steps will end prematurely before all potential sorption has occurred. Programming stability conditions that are too high can result in longer steps than necessary and overall experiment time. The method segment also requires a maximum duration of each isohume step to be set. However, if appropriate stability conditions have been identified for the samples it should not be expected that the maximum duration of each step will ever be reached. Based on the results from preliminary experimentation, a maximum step duration of 480 minutes was set for each experiment with a stability condition where if less than 1 % dimensional change occurred for 15 minutes the step would end and the instrument would proceed to the next relative humidity level.
New samples of PI-80 and PI-130 were run through the humidity stepped experiment at isothermal conditions. The conditioning steps prior to the experimental humidity step segment took between 450 and 500 minutes depending on the experimental isothermal temperature. All dimensional change data is presented here as normalized to the initial length measured after conditioning and before the first 10 %RH step commenced. Figure 4 presents the normalized dimensional change of PI-80 and PI-130 during the stepped humidity experiment. The figure is separated by isothermal testing temperature for comparison of the two samples and to observe the effect of film thickness. The beginning of every step in dimensional change observed in Figure 4 is when the relative humidity level increased in the experiment. Each step from 10 %RH to 80 %RH is presented.
When comparing the samples at a common temperature, the dimensional changes for PI-80 and PI-130 are equivalent. There is less than 1 μm difference in absolute dimensional change between the two samples at each relative humidity level for each temperature. However, the time for each step to reach stability is clearly affected by film thickness. At every temperature studied, PI- 80 more quickly increases in dimension and reaches stabilization criteria sooner. The total duration of the humidity steps for each experiment is recorded in Table 1.
Higher isothermal temperature conditions decreased the time necessary to reach dimensional stability for both samples. The data demonstrates the effect of film dimensions on experimental time. The difference in thickness between the two film types tested means that the volume of PI-130 was 1.62 times greater than PI‑80. At 25 °C the step duration at each relative humidity level was approximately twice as long for PI-130 as for PI-80. The average step durations were 1.70 and 1.58 times longer for PI-130 at 50 °C and 75 °C, respectively.
Table 1. Total duration of full stepped humidity segment for all experiments
| PI-80 | PI-130 | |
|---|---|---|
| 25 °C | 24 hr 14 min | 47 hr 14 min |
| 50 °C | 13 hr 33 min | 23 hr 5 min |
| 75 °C | 8 hr 16 min | 13 hr 3 min |
A moisture sorption isotherm graph can be generated from the data collected in the experiment. The final normalized dimensional change of each step is plotted against the relative humidity. It provides a snapshot of how a material will behave under different humidity and temperature conditions after sorption equilibrium is reached at those conditions. The coefficient of hygroscopic expansion (CHE) can be determined from sorption data. CHE is calculated from the slope of the sorption curve. It can be used to predict expansion at relative humidity levels which were not directly measured in the TMA-RH.
In Figure 5, the sorption data for each film are grouped and the hygroscopic expansion is compared at each isothermal temperature. For both films the sorption behavior at low relative humidity is very similar. A slight spread in the sorption data is present at 40 %RH and above. The largest dimensional change at high relative humidity for both films is observed at the lowest tested temperature of 25 °C. That indicates a greater strain on the films once sorption equilibrium is reached under those conditions. The largest CHE values for this data set were observed in PI-80 tested at 25 °C. These results can be used to predict performance and help in materials selection and product design.
This work demonstrates the use of the Discovery TMA 450 RH to measure hygroscopic expansion in polyimide under controlled temperature and humidity conditions. Films of two thicknesses of the same PI grade were studied by stepping relative humidity from 0 to 80 %RH at isothermal temperatures of 25 °C, 50 °C, and 75 °C. The two films studied exhibited equivalent hygroscopic expansion at each temperature and humidity step after stabilization was reached. The time required to reach dimensional change stability was strongly affected by film thickness. Sorption isotherms and coefficients of hygroscopic expansion (CHE) were also able to be generated from the stepped humidity isothermal experiments.
References
- TA Instruments, “Determination of the Dimensional Stability of a Thin PET Film,” [Online]. Available: https://www.tainstruments.com/applications-notes/determination-of-the-dimensional-stability-of-a-thin-pet-film-ta420/.
- TA Instruments, “Predicting Printed Circuit Board Delamination,” [Online]. Available: https://www.tainstruments.com/pdf/literature/TS9.pdf.
- TA Instrument, “Measurement of the Coefficient of Hygroscopic Expansion (CHE) Using Controlled Humidity Dynamic Mechanical Analysis,” [Online]. Available: https://www.tainstruments.com/pdf/literature/TA370.pdf.
- S. Diaham, “Polyimide in Electronics: Applications and Processability Overview,” in Polyimide for Electronic and Electrical Engineering Applications, 2021, pp. 3-20.
- TA Instruments, “Investigating the Effect of Humidity on the Glass Transition of Nylon Using Dynamic TMA 450 RH,” [Online]. Available: https://www.tainstruments.com/applications-notes/investigating-the-effect-of-humidity-on-the-glass-transition-of-nylon-using-dynamic-tma-450-rh/.
- Y. Dong, J. Ma, S. Yang and H. Yang, “Polyimides and their diverse applications in multiple optoelectronic devices,” Advanced Devices & Instrumentation, vol. 4, p. 11, 2023.
- S. Lahokallio, K. Saarinen and L. Frisk, “Effect of High-Humidity Testing on Material Parameters of Flexible Printed Circuit Board Materials,” Journal of Electronic Materials, vol. 42, no. 9, pp. 2822-2834, 2013.
- J. Lin, J. Su, M. Weng, W. Xu, J. Huang, T. Fan, Y. Liu and Y. Min, “Applications of flexible polyimide: Barrier material, sensor material, and functional material,” Soft Science, vol. 3, no. 1, 2023.
- TA Instruments, “Thermo-Mechanical Properties of Polyimide Films for High Temperature Applications,” [Online]. Available: https://www.tainstruments.com/applications-notes/thermo-mechanical-properties-of-polyimide-films-for-high-temperature-applications-rh153/.
- K. Sager, A. Schroth, A. Nakladal and G. Gerlach, “Humiditydependent mechanical properties of polyimide films and their use for IC-compatible humidity sensors,” Sensors and Actuators A: Physical, vol. 53, no. 1-3, pp. 330-334, 1996.
- C. Smith, “Thermochemical and physical properties of printed circuit board laminates and other polymers used in the electronics industry,” Polymer Testing, vol. 52, pp. 234-245, 2016.
Acknowledgement
This paper was written by Andrew Janisse, PhD
For more information or to request a product quote, please visit www.tainstruments.com to locate your local sales office information.
TA Instruments, Discovery, and TRIOS are trademarks of Waters Corporation or its affiliates.
Click here to download the printable version of this application note.