๐ก In This Article
- Tg is not simply a single number โ it reflects a complex set of property changes associated with polymer segmental motion
- Why Tg values differ between analytical techniques โ how DSC, DMA, TMA, and other methods observe different physical manifestations of the same glass transition
When developing an adhesive, two properties come up constantly: adhesion strength and Tg (glass transition temperature). Knowing how well an adhesive bonds is obviously important, but because an adhesive is a polymeric material, understanding how its properties shift with temperature matters just as much.
I covered adhesion testing in a previous article, Adhesion Testing โ : Pull, Shear, and Peel Tests Explained. From here on, I want to focus on Tg and the analytical techniques used to measure it.
Developers hear the same questions about Tg over and over: “What Tg does the customer require?” “Can we push the Tg a little higher?” “If the Tg is higher, doesn’t that automatically mean better heat resistance?” When I first started developing epoxy adhesives, I thought of Tg the same way โ as just a number. But the more formulations I developed and the more materials I evaluated, the clearer it became that Tg isn’t just a number; it’s a temperature range associated with a change in material behavior.
What Actually Changes at Tg?
At low temperatures, polymer chains have relatively limited mobility and the material behaves as a rigid glassy state. As temperature rises, molecular mobility gradually increases, and around the glass-transition region the polymer starts behaving more like a rubbery state. This shift is the glass transition, and Tg is the temperature used to characterize it.
The important thing is that the material doesn’t change at one exact temperature โ several physical properties shift over a range. Around the glass-transition region, for instance, we can see changes in heat capacity, modulus, and dimensional behavior. So while Tg often shows up as a single number in a report, the underlying phenomenon involves several material properties changing together. That distinction becomes especially important once you start comparing different analytical techniques.
The Molecular Origin of the Glass Transition
From a molecular perspective, the glass transition tracks the increasing mobility of polymer segments and the growth of free volume. In the glassy state, molecular motion is tightly restricted; as temperature rises, thermal energy and free volume increase, allowing increasingly significant segmental motion and relaxation to occur.
These molecular-scale changes show up macroscopically in different ways depending on how the material is measured โ as an increase in apparent heat capacity, a significant drop in stiffness or modulus, or a shift in the coefficient of thermal expansion. The underlying molecular transition is the same, but the physical property through which we observe it can look completely different. That’s one of the main reasons different analytical instruments can report slightly different Tg values.
Why Do Different Instruments Give Different Tg Values?
The first time I ran into different Tg values from different instruments, I wondered why Tg didn’t have one consistent value. The answer gets much clearer once you look at what each instrument is actually measuring.
| Property being observed | Typical technique |
|---|---|
| Change in heat capacity / heat flow | DSC |
| Change in viscoelastic modulus | DMA |
| Change in dimensional expansion | TMA |
These techniques aren’t measuring three different glass transitions โ they’re observing different physical manifestations of the same underlying phenomenon. That’s why Tg values from DSC, DMA, and TMA don’t always line up. A difference between them isn’t necessarily a sign that one measurement is wrong; it can just mean the same material transition is being picked up through a different physical response.
Measurement Conditions Also Affect the Reported Tg
There’s another variable at play: even with the same epoxy sample tested on the same type of instrument, the reported Tg can shift depending on measurement conditions โ particularly heating rate and, for dynamic mechanical measurements, frequency.
Generally, a faster heating rate gives the polymer less time to respond to the changing temperature, and a higher measurement frequency probes the material over a shorter timescale. Both tend to push the apparent transition toward a higher temperature. This time-scale dependence matters a great deal for DMA and rheological measurements, where frequency is built into the measurement itself.
So simply writing “Tg = 120ยฐC” doesn’t give the full picture โ a meaningful Tg value needs to be read alongside the measurement method and conditions. For reliable comparison, the test method should follow the relevant standard, such as an applicable ASTM or ISO method, and the report should identify not just the instrument but the key measurement conditions and the criterion used to determine Tg. Depending on the technique, the reported value might be based on onset, midpoint, peak, or some other characteristic change in the measured property. When comparing Tg values, the measurement method and interpretation criteria matter just as much as the number itself.
A Development Note: Making the Most of What You Have
I used a rheometer far more often than a DMA, for a simple reason: we didn’t have a DMA. So I had to figure out how to pull as much useful information as I could from the equipment I actually had.
Early on, I used the rheometer to monitor viscosity changes during curing and understand how different formulations cured. For fully cured specimens, I also tracked the temperature dependence of G’ (storage modulus) and G” (loss modulus) to see how viscoelastic behavior changed with temperature. This obviously isn’t a standardized Tg measurement equivalent to DMA, but during formulation development it was genuinely useful โ it let me see when a material’s behavior started to change, compare formulations against each other, and make decisions about where development should go next.
For a formulation developer, that kind of comparative information can sometimes matter more than one additional precise number. The question isn’t always “what is the exact Tg?” โ sometimes the more useful question is “how does this material start behaving differently as temperature rises?”
A Developer Looks at the Change, Not Just the Tg Number
Over time, Tg stopped being a single number to me and became more of a reference point for understanding when and how a material’s behavior starts to shift. That’s why, in formulation work, I don’t think Tg should be evaluated on its own. The questions worth asking are: what property is changing? How fast does it change? Does the material stay stable after the transition?
This matters especially for epoxy adhesives. A higher-Tg formulation isn’t automatically better for every application โ depending on what the adhesive needs to do, modulus, toughness, thermal expansion, stress relaxation, and adhesion retention can all matter just as much.
Material characterization ultimately isn’t about reading numbers off an instrument โ it’s about understanding what those numbers are telling you about the material. Tg is a good illustration of that: the glass transition is one physical phenomenon, but we observe it through different material responses, so DSC, DMA, TMA, and other techniques each offer a different perspective on the same underlying change. Once that clicks, differences in Tg values between techniques become much easier to interpret.
Coming Next: Reading a DSC Curve
In the next article, “How Do You Read a DSC Curve? โ Why Is DSC Used to Determine Tg?,” I’ll take a closer look at DSC (Differential Scanning Calorimetry), one of the most widely used techniques for evaluating Tg โ not just what number shows up on a DSC report, but what actually changes in the DSC curve and why developers use that change to identify the glass transition..
