Keywords: heat shrink tubing, dimension change, shrinkage, recovery, shape memory, TGA, DSC, MDSC, TMA, MTMA, DMA, thermal analysis
TA498
Abstract
This note describes the use of thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), thermomechanical analysis (TMA), and dynamic mechanical analysis (DMA) to characterize commercially available heat shrink tubing materials. This material is explored using high resolution TGA methods to help quantify the composition, and Modulated DSC™ Techniques and TMA are used to investigate the recovery of the material as it shrinks.
With the rising numbers of electric vehicles, data centers to support AI, and other electronic applications, cables and wires are being increasingly used. For safe and reliable operation in different end use environments, it’s critical for connections to be insulated and sealed to prevent shorting. Heat shrink tubing is expected to provide reliable performance in a variety of demanding conditions. Shape memory materials such as heat shrink have been utilized for many years but increasing volumes and changing use conditions have led to renewed interest in this area. Newer systems may require multiple connections leading to varying size across the covered area. Additionally, there are now a variety of heat shrink tubing options that can include mixed matrix materials of multilayer polymers or adhesives. More complex formulations, greater exposure to thermal gradients, UV, and humidity along with an expectation for a long service lifetime means there are plenty of opportunities for additional research to understand product performance and determine areas where further development is needed.
A variety of advanced instrumentation techniques can be utilized to determine characteristic properties of these materials, such as thermal stability and expansion. In this work, the utility of TGA, DSC, TMA, and DMA will be demonstrated on single-wall heat shrink tubing. Measurements will primarily be conducted on the material as received (expanded), but the same techniques can be used for analyzing the recovered (shrunk) material as well.
The sample of commercially available heat shrink tubing was labeled as a flame retardant polyolefin in a white color. The listed shrink ratio on the packaging was 2:1 with a shrinkage temperature of 125 °C.
A TA Instruments™ Discovery™ TGA 5500 was calibrated with alumel and nickel with nitrogen purge at 25 mL/min. Heat shrink tubing was cut into small pieces and 5 mg placed in high temperature platinum pans (100 μl). A standard temperature ramp was conducted at 20 °C/min from room temperature to 600 °C with a gas switch to air from 600 to 1000 °C. Additionally, a Hi‑Resolution Dynamic test was completed in nitrogen with a base heating rate of 50 °C/min with the resolution number 5 and sensitivity value 1.
The Discovery DSC 2500 with RCS 90 was calibrated with indium. Samples of 5-7 mg were prepared in TZero™ Aluminum Pans with TZero Lids, loaded into the cell at 21 °C and tested under nitrogen purge at 50 mL/min. A rate of 10 °C/min was used to ramp the sample from -90 to 180 °C, cool to -90 °C, and reheat. A modulated DSC (MDSC™ Method) Method was created to heat the sample from -40 to 180 °C using a 60 second period and 2 °C/min ramp rate using heat only amplitudes (0.318 °C for selected conditions populated by TRIOS™ Software). An additional method utilizing the DSC Microscope Accessory will be discussed in the results section.
The Discovery TMA 450EM with MCA 70 was used to measure expansion and contraction of the samples with the Film/Fiber probe. Probe calibration was completed along with temperature calibration using indium wire and CTE calibration with aluminum foil. All tests were completed under nitrogen purge at 100 mL/min. The included sample preparation block was used to mount 8 mm cross sections of heat-shrink tubing in the stainless-steel film clamps. A standard temperature ramp from -40 to 180 °C was completed using a linear heating rate of 5 °C/min and force of 0.05 N. Subsequently, a modulated temperature ramp with amplitude of 5 °C and modulation period 300 seconds was conducted at a rate of 0.5 °C/min from 10 to 140°C.
The Discovery DMA 850 with Gas Cooling Accessory was used with the dual screw film/fiber clamp. An iso-strain temperature ramp was conducted with 0.1% constant strain while ramping at 1 °C/min from 25 to 150 °C. A small-amplitude linear oscillation of 1 Hz and 10 μm (amplitude in LVR) was used with preload force of 6 N and force track for a 3 °C/min temperature ramp.
Thermal Stability
In a thermal analysis workflow, TGA is regularly the first test conducted since this will provide information on sample stability and inform future test method development. Preliminary observations using thermogravimetric analysis show the material has extrapolated onset of decomposition at 325 °C with 5% weight loss occurring by 334 °C (Figure 1). There are two obvious degradation events visible in the raw weight loss data. By further evaluating the weight loss derivative curve, more than four steps become apparent. In addition to the listed flame retardant, it is likely there are a variety of other materials present in the formulation, including process aids, stabilizers, and fillers.
There was a slight weight change after switching the gas to air, suggesting a very small amount of pyrolytic carbon present, likely from the degradation of the additives. There is 25% residue remaining at the end of the ramp which is likely inorganic fillers that did not burn off with the gas switch to air. The material that remains in the pan after this test was light gray.
Figure 1. TGA temperature ramp at 20 °C/min in nitrogen to 600 °C and in air to 1000 °C (inset: residue after experiment)
When multiple weight losses occur around the same temperature, it can be challenging to quantify some of the components due to overlapping decomposition. The most common adjustment to yield better resolution is to slow the heating ramp rate. However, doing so comes at the expense of longer experiments. For users looking to optimize both their data resolution and the test time, TA Instruments provides advanced high-resolution (Hi-Res) method options.
The TGA 5500 includes the Hi-Res TGA methods as a standard feature. This advanced method automatically adjusts the temperature profile based on the weight changes occurring in the sample, rather than using a single heating rate throughout the duration of the experiment. In this work, the Hi-Res Dynamic method with a fast base heating rate was selected to expedite the test in regions without any weight change, saving experimental time. The Hi-Res method adapts to the rate of weight change from the sample being tested, adjusting the heating rate control in real time. The rate slows during times of rapid weight change and then speeds back up as the weight change is minimized.
In this experiment, the heating rate was reduced as low as 0.5 °C/min during the closest spaced weight loss events. The results of the sample-controlled heating rate can be seen in the derivative of Figure 2, where the small weight changes can now be more clearly distinguished. The minimum points in the derivative signal are regularly used to help guide selection of endpoints for the weight change analysis. The Hi-Res tests were conducted in nitrogen alone and yielded a black powder remaining in the pan, with only small differences (0.5%) in residue percentage at 700 °C.
Figure 2. TGA Hi-Res temperature ramp at 50 °C/min in nitrogen with resolution number 5 and sensitivity value 1 (inset: residue after experiment)
While not explored in this paper, other advanced uses of the TGA 5500 can include multi-run kinetics experiments or singlerun modulated TGA tests to compare activation energy of decomposition. From these experiments the estimated lifetime of different formulations can be compared. Examples of these methods are covered in TA125 and TA475 [1] [2].
Recovery Upon Heating
By design, heat shrink tubing is expanded during manufacturing and possibly irradiated or crosslinked to lock it into a known expanded state. Therefore, relaxation and recovery is expected upon heating the sample. A differential scanning calorimeter (DSC) measures the total heat flow of any endothermic and exothermic events occurring in a sample. DSC is regularly used to evaluate polymers for their melting point and glass transition temperature (ASTM D3418) [3].
The first heat in the DSC shows a small endothermic transition at 50 °C and larger endotherm at 88 °C (Figure 3). Both temperatures are much lower than known melting of polyethylene or polypropylene, the most commonly used polyolefins. Additionally, this material has a relatively low observed enthalpy 42 J/g with both peaks included in the integration. This value does not account for the presence of 25% filler; when adjusted to remove the filler contribution, this would be approximately 56 J/g. The relatively low enthalpy could be due to a high crosslink density. The known heat of fusion for 100% crystalline polyethylene and polypropylene are 293 and 207 J/g respectively [4], but 40% crystallinity is common for polyolefins.
There is an apparent glass transition at -28 °C on heating. On cooling, the material crystallizes at 74 °C with observed enthalpy of 37 J/g. On the second heat there is an apparent glass transition at -26 °C. The second heat is the recovered (shrunk) tubing, and the melting peak is at 87 °C with observed heat of fusion 34 J/g.
Modulated DSC Methods (MDSC Methods) are included with the Discovery DSC and enable further investigation into materials of interest. A detailed theoretical background of this technique can be found in prior application notes TP006-TP013 [5]. In brief, by applying a sinusoidal oscillation to the linear temperature ramp, the modulated heat flow signal can be Fourier transformed and deconvoluted into reversing and non-reversing signals, which together make up the total heat flow. Generally, the heat capacity is seen in the reversing heat flow, and things that are kinetic in nature – like enthalpic recovery and crystallization – are seen in the non-reversing heat flow.
Based on MDSC Methods, the thermogram in Figure 4 indicates that the initial endotherm appears in the non-reversing heat flow and is immediately followed by exothermic changes in the sample as well. There are likely several complex kinetic effects ongoing during heating. This is not surprising due to the processing history of this material required for the initial expansion. Modulated DSC Techniques can provide visibility, enabling researchers to quantify these different effects.
Standard test methods for heat shrink tubing discuss storage life testing using a 40 °C isothermal to ensure there are no issues with material remaining in an expanded state (ASTM D2671) [6]. With transitions near 40 °C in the first heating and MDSC, further investigation may be warranted because in extreme storage and transport conditions, elevated temperatures are possible.
If users would like to observe the sample during the test, the DSC Microscope Accessory can be installed prior to the experiment. There are options to record video or capture images by using custom method segments controlled by TRIOS Software, and both segments can be used simultaneously. A picture of the microscope accessory and associated method used in this work are shown in Figure 5. Images were collected every thirty seconds throughout the entire duration of the experiment, and a video was recorded during the heating ramp.
Figure 5. (A) DSC Microscope Accessory, (B) Custom TRIOS Software method incorporating image and video segments
To visualize the sample, an open TZero Aluminum Pan is used. Figure 6 includes images saved at different times throughout the experiment in chronological order. In the leftmost image, the tubing is approximately square, as it was when placed in the pan. There were no obvious differences between room temperature and 63 °C images, confirming the first endotherm is not associated with shrinkage. At 88 °C, the sample has noticeably started to contract in a single direction, with further contraction seen in images at 93 and 103 °C. A video recording of the entire heating ramp was captured during the same test. A 10x speed recording can be found here.
Dimension Change – Simultaneous Shrinkage and Expansion
Using the TMA 450EM, the dimension change is commonly monitored during a constant low force temperature ramp to determine the coefficient of thermal expansion. Most materials expand upon heating, but heat-shrink materials are designed to yield a negative dimension change when heated beyond the designated shrink temperature. From the DSC images and video, it became obvious the shrinkage event is anisotropic, differing dependent on orientation.
The material shrinks 2:1 in the radial direction yet remains similar in size for the length of the tubing. Because this sample increases in thickness during the recovery, a small cross-section of the sample was cut to mount in the stainless-steel film clamp instead of using a full tube. The tube was tested in the two directions, first in the diameter direction, and next in the lengthwise direction. Graphical examples of this are shown in Figure 7. The tubing was cut into cross sections ~12-20 mm in length and 4 mm in width, and the film fiber preparation block was used to mount the samples with a dimension of nominally 8 mm between clamps.
Figure 7. Graphics showing the cross sections in blue that were cut from the heat shrink tube and tested in (A) diameter (radial) and (B) lengthwise (longitudinal) directions; photos show the respective shrinkage directions
From the TMA data in Figure 8, the dimension change shown on the y-axis is initially increasing as the sample increases in length indicated by positive coefficient of thermal expansion (CTE) of 226 ppm. It is also interesting that in this initial test, there is an inflection at 39 °C where the material starts expanding more rapidly (429 μm/m°C) just before the material begins to shrink. The extrapolated onset of shrinkage is calculated at 89 °C. The shrinkage begins earlier than anticipated from the material spec sheet and finishes the recovery process near 125 °C, which is the manufacturer’s listed temperature. The total dimension change observed in this preliminary test is the sum of all expansion and contraction occurring in the sample. Even within a single layer material, there can be simultaneous processes occurring due to the processing history. To further explore this concept, Modulated TMA (MTMA) can be used.
MTMA applies a sinusoidal temperature profile to the heating ramp. From the raw signals of modulated dimension change and modulated temperature, the reversing (rev) dimension change signal is calculated, as well as the non-reversing (non-rev) dimension change. When using MTMA methods, TRIOS Software displays three signals – total, rev, and non-rev dimension change – during the experiment (Figure 9). In any modulated experiment, it’s important to check that the modulation is maintained and that there are a minimum of 4-5 modulations over the region of interest. This is achieved by heating this sample with a rate of 0.5 °C/min.
The reversing signal is proportional to the coefficient of linear thermal expansion, and the non-reversing signal is where kinetic processes such as shrinkage or stretching will be found (TA311) [7]. The tubing is analyzed with samples taken from both directions (radial and longitudinal) to take a closer look at the dimension change anisotropy.
Figure 10 shows the results of the modulated experiment in the radial direction and Figure 11 in the longitudinal direction. There are numerous small changes in the non-reversing expansion from room temperature to 65 °C. Both MDSC and MTMA Methods highlight that when this material is exposed to higher temperatures, the expanded material experiences slight changes leading up to the recovery. It’s possible the sample creeps under applied load and it’s seen in the non-reversing signal. MDSC Methods revealed that the non-reversing signal had endothermic and exothermic events occurring from approximately 30 to 115 °C. Similarly, from the MTMA in the radial direction (Figure 10), there are several changes in the non-reversing dimension change leading up to 50 °C. There is also a slight change in the reversing signal with differences in CTE before and after 50 °C in both dimensions of the tubing sample.
The shrinkage is primarily in the radial direction along the diameter of the tubing, which was expanded by the manufacturer (Figure 10). Along the longitudinal direction of the tubing, there is primarily expansion of the polymer along with the residual stresses (Figure 11). By design, the diameter should shrink to tightly cover any connection, and in this case the lateral direction remains similar, which will help ensure the tubing remains long enough to protect the necessary area once fully recovered.
The MTMA data shows that even with a single filled material, the stresses imparted during expansion lead to unique behavior on heating. With newer multi-layer formulations there is an increase in the number of interfaces, so knowing the coefficient of expansion of different materials is important. Matching the CTE can help prevent delamination in layered structures.
Mechanical Testing
The glass transition temperature (Tg) can be challenging to determine in some materials. In those instances, mechanical testing can provide additional sensitivity to detect the transition through dynamic oscillatory experiments. In addition to finding the Tg, DMA provides additional information about the mechanical strength and modulus of the material. During the oscillatory temp ramp (Figure 12), there is a drop in the storage modulus with onset at -26 °C. This correlates well with the apparent Tg in DSC. Because the material is semi-crystalline and may be crosslinked, the peaks in loss modulus and tan delta are broadened.
An iso-strain experiment was conducted where the tubing was held at a constant strain of 0.1% (a fixed displacement) for the duration of a temperature ramp. The necessary force to maintain the requested strain is applied by the DMA and plotted in Figure 13. From room temperature to 60 °C, the sample is expanding due to the CTE of the polymer and therefore a lower force is required to maintain the same length. There is a minimum force of -0.395 N at 42 °C. As the temperature continues to rise the sample is designed to recover and shrink, but the instrument applies the necessary force to prevent the sample from shrinking. The maximum tensile force required is quantified as 0.709 N at 96 °C to maintain the requested sample displacement.
In this paper, both thermal and mechanical characteristics of commercial heat shrink tubing were explored. Due to the excellent sensitivity of the Discovery TGA and the use of Hi-Resolution methods, users can more easily quantify components of different thermal stabilities. The images from the DSC Microscope Accessory and dimension change from TMA clearly show the anisotropy of shrinkage occurring during the tubing recovery. By using modulated methods on the Discovery DSC and TMA, users can further investigate the changes ongoing during the preliminary heating of their materials by separately evaluating non-reversing events that could be related to the process history. The Discovery DMA provides additional information on the modulus of the material before and after the glass transition and can be used to find the shrinkage force.
Complementary future studies could utilize the Discovery Sorption Analyzer to investigate moisture uptake of heat-shrink tubing. While the uptake for commonly used materials should be minimal, protecting connections from moisture is another key requirement for heat shrink tubing and the Sorption Analyzer (SA) provides users with the ability to study the gravimetric uptake of moisture. The Discovery TMA-RH enables further study the dimension change from hygroscopic expansion.
The sample used in this work was a single layer heat shrink tube. As formulations become more complex with additional layers and adhesives, manufacturers must ensure compatible characteristics between those materials. TA Instruments provides a variety of precision instrumentation on a common acquisition and analysis platform to help researchers explore the thermal and mechanical properties of their materials.
References
- TA Instruments, “TA125 Estimation of Polymer Lifetime by TGA Decomposition Kinetics”.
- A. Janisse and J. Vail, “TA475 Thermal Lifetime Analysis of PET and Recycled PET Fibers”.
- ASTM International, “ASTM D3418 Standard Test Method for Transition Temperatures and Enthalpies of Fusion and Crystallization of Polymers by Differential Scanning Calorimetry”.
- R. L. Blaine, “TN048 Polymer Heats of Fusion”.
- L. C. Thomas, “TP006 Why Modulated DSC? An Overview and Summary of Advanatages and Disadvantages Relative to Traditional DSC”.
- ASTM International, “ASTM D2671 Standard Test Methods for Heat-Shrinkable Tubing for Electrical Use”.
- TA Instruments, “TA311 Modulated Thermomechanical Analysis – Measuring Expansion and Contraction Simultaneously”.
Acknowledgement
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 Jennifer Schott, PhD.
TA Instruments, Discovery, TZero, Modulated DSC, MDSC, and TRIOS are trademarks of Waters Corporation or its affiliates.
Click here to download the printable version of this application note.