💡 In This Article
- The relationship between 85/85 testing and an epoxy Tg above 85°C
- Dry Tg vs. Wet Tg and practical considerations for measuring them
- Moisture saturation and the possibility of irreversible degradation
- Why moisture reliability must be designed around the entire adhesive system rather than a single property
In the previous article, “Why Does Epoxy Look Fine When Dry but Fail After Absorbing Moisture? ① – Why 85/85, PCT, and HAST Testing Exist,” I looked at how heat and moisture together cause adhesion loss and reliability failures. This time I want to look more closely at how those failures relate to Tg.
When an adhesive falls short in an 85/85 reliability test, one suggestion comes up constantly during development: “Let’s increase the Tg.” At first glance that seems perfectly reasonable — if the test runs at 85°C, pushing the glass transition temperature above 85°C should logically help the adhesive hold its stiffness and thermal stability, and improve reliability along with it. Tg is genuinely an important property of an epoxy adhesive and a common formulation target.
In actual development work, though, I’ve run into plenty of cases where Tg rose substantially and the 85/85 improvement was far smaller than expected. The reason comes down to one thing: an 85/85 test is not simply an 85°C heat test.
Glass Transition Temperature (Tg)
Glass transition temperature (Tg) marks the range where a polymer’s molecular mobility and mechanical behavior shift significantly. Below Tg, segmental molecular motion stays relatively restricted and the material behaves mostly as a rigid glassy solid; above Tg, molecular mobility increases and the material starts behaving more like a rubber.
So if an adhesive is meant to operate around 85°C, keeping Tg comfortably above that service temperature is a reasonable design objective, and it’s a natural first move when an adhesive underperforms thermally. But there’s an important limit to how far that logic can carry you: 85/85 isn’t just a temperature test.
The Key to 85/85 Is Moisture, Not Just Temperature
What defines the 85/85 test is the combination of elevated temperature and high humidity. Moisture gradually penetrates the epoxy network and changes the interactions within the polymer — it can raise molecular mobility through plasticization, reduce stiffness, and lower the effective Tg of the moisture-conditioned material. At the same time, moisture reaching the adhesive/substrate interface can weaken interfacial interactions and, depending on the chemistry, promote degradation over extended exposure.
Reliability problems under 85/85 conditions, in other words, aren’t governed by temperature alone — they come out of a coupled temperature-and-moisture environment. That’s why raising Tg doesn’t automatically solve the problem; several other factors can matter just as much:
- moisture uptake
- moisture diffusion
- interfacial adhesion
- resistance to hydrolysis or other moisture-induced degradation
- toughener morphology and dispersion
- coupling-agent performance
- residual stress
- crosslink density
- substrate surface chemistry
Tg is only one variable in a much larger reliability system.
Dry Tg vs. Wet Tg: The Plasticization Effect
One of the most common traps in adhesive development is leaning too heavily on the Dry Tg measured on a fully dried specimen or listed on a datasheet. An epoxy exposed to moisture doesn’t necessarily behave like the same epoxy in its dry state — water molecules entering the polymer network interact with polar groups within the epoxy, and these interactions can increase molecular mobility in what amounts to a plasticizing effect. As a result, the Tg of a moisture-conditioned epoxy can end up substantially lower than its dry-state Tg.
How much lower is highly formulation-dependent. For some epoxy systems the gap between Dry Tg and Wet Tg stays fairly small; for others, especially those with significant moisture uptake, the drop can be large enough to matter for reliability.
Take a simplified example: an epoxy with a Dry Tg of 110°C looks comfortably above an 85°C test temperature. But if prolonged moisture exposure lowers the conditioned material’s effective Tg to, say, 75°C, the picture changes completely. Under 85/85 conditions, the adhesive may now be operating in a range where molecular mobility runs substantially higher than the dry-state data would suggest — reducing stiffness, changing stress-relaxation behavior, and leaving the interface exposed to continued moisture effects all at once.
So the question that matters isn’t “is the Dry Tg higher than 85°C?” It’s “what happens to the adhesive after it has absorbed moisture?” That distinction between Dry Tg and Wet Tg is essential to interpreting moisture-reliability data correctly.
Practical Considerations When Measuring Wet Tg
Measuring Wet Tg is considerably trickier than measuring Tg on a dry specimen, and the biggest practical problem is moisture loss during sample preparation and measurement itself.
Take DSC as an example. If a moisture-conditioned specimen sits in a conventional open pan while heated, absorbed water can evaporate during the temperature scan, and the resulting endothermic signal can interfere with the thermal transition being measured. Worse, if enough moisture leaves the specimen before the glass transition is captured, the result may no longer reflect the original wet condition — the specimen can effectively dry out partway through the measurement.
That’s why the measurement procedure has to be designed to preserve the specimen’s moisture state as much as possible. Depending on the analytical method and objective, that can mean:
- using a hermetic pan for DSC to minimize moisture loss
- minimizing the time between conditioning and measurement
- using DMA with an appropriately controlled measurement procedure
- comparing dry and moisture-conditioned specimens under the same measurement conditions
- confirming the actual moisture content of the conditioned specimens
There’s an important practical caveat here, though: a hermetic DSC pan preserves moisture better, but the presence of absorbed water can also complicate the thermal signal itself. The result should be read alongside the measurement configuration and conditioning history rather than treated as an absolute material constant. For DMA, a relatively rapid first temperature scan can sometimes help capture a moisture-conditioned specimen’s behavior before substantial moisture loss sets in — though again, the exact heating rate and conditions should be chosen based on the material and the purpose of the test.
What Is the Moisture Saturation Point?
Comparing Dry Tg and Wet Tg meaningfully requires knowing how much moisture the material has actually absorbed. A polymer exposed to humid conditions initially gains weight as water diffuses in, and eventually that weight gain approaches an equilibrium level — commonly called moisture saturation or equilibrium moisture uptake.
A practical approach is to periodically weigh conditioned specimens on a sufficiently sensitive balance and track the mass change over time — for example, at regular intervals of 24 or 48 hours, until the change becomes negligible relative to the measurement precision and whatever acceptance criterion has been set. There’s nothing universal about the 24-or-48-hour figure itself; the conditioning time needed to reach equilibrium depends on specimen thickness, geometry, temperature, humidity, resin chemistry, filler content, and diffusion behavior. Thin specimens can approach equilibrium fairly quickly, while thick adhesive layers or highly filled systems can take much longer.
This matters because a specimen that hasn’t reached moisture equilibrium may not represent the true worst-case Wet Tg or moisture-induced property change.
When Weight Changes After Saturation
Weight monitoring can also flag something more serious than ordinary moisture uptake. If a specimen appears to reach equilibrium weight and then starts losing mass, that shouldn’t automatically be read as simple moisture desorption. Depending on the material system, it can point to other processes:
- leaching of low-molecular-weight components
- chemical degradation
- formation of microcracks
- loss of volatile species
- physical damage to the specimen
An unexpected weight change after apparent moisture equilibrium is worth investigating rather than averaging out as measurement noise, particularly when the material is being evaluated for long-term reliability. Moisture uptake itself isn’t necessarily the failure mechanism — it can just be the start of a chain of changes that eventually reaches the bulk material and the interface.
Things Become Even More Complicated with Toughened Epoxies
Interpreting Wet Tg gets more complicated once a toughener enters the epoxy system. If the epoxy resin and toughener are fully miscible, the system can often be described fairly simply through a single glass transition. But many practical toughened epoxy systems show some degree of phase separation, with the cured material containing an epoxy-rich continuous phase alongside a toughener-rich dispersed phase.
The Tg that DMA picks up may end up dominated by the epoxy-rich phase’s behavior, while the toughener-rich phase can have a much lower transition temperature — or one that’s weak, broad, or simply outside the temperature range of interest. So a rise in measured Tg doesn’t necessarily mean the entire multiphase system has improved across the board. A single Tg value should never be treated as a complete description of the adhesive.
Reliability Must Be Designed as a System
It’s tempting, during formulation development, to assume that improving one property will lift the whole product. Reliability rarely works that way. Moisture reliability comes out of the combined behavior of the bulk polymer, the interface, moisture transport, chemical stability, and mechanical stress together.
A formulation with high Tg can still fail if it absorbs too much moisture. A formulation with low moisture uptake can still fail if its interface is weak. A strong interface can still fail if the adhesive layer can’t accommodate the stresses generated during environmental exposure.
That’s why, when I’m evaluating an 85/85 result, I find “what is the Tg?” a far less useful question than “why does this adhesive maintain its adhesion after prolonged exposure to heat and moisture?” The second question forces you to look at the whole system instead of a single number.
Designing for Moisture Reliability
Improving moisture reliability takes more than raising the Dry Tg. Depending on the formulation and application, useful approaches can include:
- reducing moisture uptake through appropriate resin and molecular design
- optimizing hydrophobicity without compromising adhesion
- improving the stability of the adhesive/substrate interface
- optimizing the selection and treatment of silane coupling agents
- controlling toughener morphology and dispersion
- optimizing crosslink density and network structure
- minimizing defects such as voids and microcracks
- evaluating the change in properties after moisture conditioning rather than relying only on dry-state data
The practical goal isn’t maximizing initial Tg — it’s minimizing the change in material and interfacial behavior that moisture exposure causes. Seen that way, the gap between Dry Tg and Wet Tg can tell you more than the Dry Tg value on its own.
Final Thoughts
It’s easy to get attached to a single number when developing an epoxy adhesive, and Tg is one of the usual suspects — useful, important, and often highly informative, but not the whole story. An adhesive with a Dry Tg of 110°C can look perfectly safe for an 85°C application on paper, yet after absorbing moisture its molecular mobility, stiffness, stress-relaxation behavior, and interfacial stability can look very different. That’s how an adhesive passes initial evaluation and still fails 85/85 testing.
The real challenge isn’t hitting a high Tg — it’s keeping the behavior of the entire adhesive system stable once moisture has worked its way in. That’s also why reliability-oriented formulation has to look past any single material property and weigh moisture uptake, diffusion, interface chemistry, crosslink structure, toughener morphology, and environmental stress together.
So far, we’ve looked at how heat and moisture cause adhesive failure, and why Tg alone can’t explain moisture reliability. The natural next question is: if Tg matters this much, how should we actually measure it? That leads into the next series of articles, where I’ll look at DSC, DMA, TMA, and other analytical techniques — and why these instruments can report different Tg values for what looks like the same epoxy material..
