BlogApplication notes
Why won't my adhesive cure, or cures slowly?
- Type
- Application notes
- Published
- 20 November 2026
- Prepared by
- Technical team
- Product group
- Adhesives
When an adhesive will not cure, a bad batch is the first suspect, but more often one condition the cure needs is missing. The cause can be low temperature, a wrong mix ratio, too little moisture, amine blush on epoxy, trace contaminants that poison the catalyst, or a wide gap and dull surface for anaerobics.
Sometimes a part bonded in the morning still smears by the afternoon, or a bead that feels hard is soft inside when cut. On such days fixtures and clamps stay on, the next step waits, and parts passed on under-cured can fail later under load or heat and come back as rework and claims. So "it feels hard" is not enough.
An adhesive is applied as a liquid and carries load once it turns solid. Cure is a chemical reaction in which small molecules link into a network, and only a complete reaction gives the designed strength and heat resistance. A thermoset first gels, so it no longer flows, and then vitrifies, so its molecules lock in place (Gillham, 1986). If vitrification happens at the cure temperature, the reaction effectively stops, and the unreacted part costs strength and heat resistance (Gillham, 1986; Montserrat, 1991).
So when an adhesive will not cure, a missing condition for the cure reaction is more often to blame than a bad batch. Products that cure by mixing two parts, by moisture from the air or by contact with metal ions each need different conditions, so knowing how the product cures narrows the cause sharply.
In such cases the technical team sets up the six hypotheses below and rules them out one by one. The start is always the same: check the cure type, mix ratio and recommended temperature and humidity on the technical data sheet, the TDS, against the conditions at the time of work. That one comparison screens several hypotheses at once.
Hypothesis 1: too cold
Temperature is the first thing to suspect. It is easy to measure and especially common from late autumn to early spring. Typical cases are assembly areas where heating was off overnight, benches beside loading doors, and adhesive used straight from cold storage. Picture a winter morning: a two-part epoxy goes onto steel brackets just brought in from an outdoor store. The room air warms quickly, but thick metal parts stay cold for a long time, and the adhesive cures against that cold surface. If the same product applied in the afternoon cured on time but only the first batch of the morning is soft, this hypothesis is the strong one.
The cure lags on a cold surface because of reaction speed. In a two-part epoxy, the reactive epoxy groups in the resin bond with the reactive amine groups in the hardener; as temperature falls, the molecules move less and far fewer collisions carry enough energy to react. A study of low-temperature epoxy adhesive cure by differential scanning calorimetry, or DSC, which measures reaction heat, reports that under long cold conditions the cure reaction may not finish (Xia et al., 2014).
It can do more than slow down; it can stop. As the reaction proceeds, the network's glass transition temperature, Tg, the point where the cured resin turns glassy, rises, and once Tg reaches the cure temperature the molecules lock and the reaction nearly stops. This is vitrification (Gillham, 1986). Isothermal cure studies show that the lower the cure temperature, the lower this limiting conversion and limiting Tg (Montserrat, 1991). A joint cured in the cold therefore never finishes, however long it waits, until it is warmed. Batch size plays a part too. The cure reaction gives off heat, so a large batch keeps that heat and cures fast, while a small batch or a thin film loses it and cures slowly. That is why the gel time test method requires the mixed volume to be reported with every result (ASTM International, 1999).
Start by checking whether delays cluster on winter mornings, in cold spaces or on cold metal parts. Measure not only air temperature but the surface temperature of the parts with a surface thermometer, and log the storage temperature of the adhesive. Curing the same mix as a coupon in a warm room alongside settles it quickly. Check cure with a fingernail or a durometer, which gauges surface hardness, and when needed confirm strength with lap-shear specimens: two overlapping strips bonded and pulled apart to measure the breaking force (ISO, 2003).
Once temperature is confirmed, the fix is temperature too. When the working temperature is below the minimum cure temperature on the TDS, warm the parts and adhesive beforehand or heat the work area. For joints already under-cured, check whether a heated post-cure schedule on the TDS can finish the reaction. If the same problem returns every winter, review a grade with a low-temperature hardener with the supplier.
Hypothesis 2: off-ratio mix
If soft spots appear even when the temperature is normal, mixing comes next. It shows up with hand mixing by eye from two tins, with the habit of adding a little extra hardener "to make it cure faster", and with cartridge work where the first shot from the nozzle is not discarded. For example, if only some of the parts bonded by the same operator on the same day are soft, and the soft areas are scattered like stains, look at mixing before temperature. A temperature problem would slow everything evenly.
Soft spots stay patchy because the reactive groups are mismatched. A two-part adhesive has a mix ratio set so that the reactive groups of both parts match; if one side is in excess, groups without a partner are left over as loose chains that do not join the network. For epoxy-amine systems, experiments and models confirm that moving off stoichiometry, the balance of reactive groups, changes Tg and mechanical properties markedly (Foster et al., 2023). Adding extra hardener has the same effect. Even with the right ratio, uneven mixing creates spots where the local ratio is wrong, and those spots stay tacky. Mixes with different amine content also differ in surface chemistry and in how sensitive they are to the air (Bannister et al., 2022).
To check mixing, look first for traces: material left on container walls and bottoms, or colour streaks. Check that you meter on the same basis, weight or volume, as the TDS ratio, and on meter-mix equipment catch each component separately and weigh them to check the ratio. With cartridges, check that the first shot from the mixing nozzle, the tip that blends the two parts, was discarded; for products with a mix indicator colour, check that the colour is fully even.
If mixing is the cause, set fixed ways to meter and mix. Meter weight ratios on a scale and volume ratios with graduated containers or meter-mix equipment, and stop eyeballing. When hand mixing, scrape the walls and bottom, then transfer to a fresh container and mix again so unmixed material from the walls stays behind. Log a ratio check on meter-mix equipment at a fixed interval. A joint with the wrong ratio does not recover with heat, so remove and redo it.
Hypothesis 3: too dry or deep
Look at this hypothesis when a one-part product squeezed straight from the tube, such as a silicone, urethane or silyl-modified polymer sealant, cures slowly. Typical cases are when the air is dry, as in heated rooms in winter, when joints are filled deep or beads run thick, and when the material is spread wide between two non-porous faces such as glass or metal. For example, if a sealant that filled a panel joint deep in one pass is still paste-soft inside when cut two days later, suspect the path for moisture before the product.
The core lags because such products cure by reacting with water from the air. The water enters at the exposed surface and has to diffuse through the cured layer to reach deeper material, so cure starts at the surface and moves inward. A study of a one-part silicone proposes a model in which cure rate depends on both temperature and the water concentration in the air, with cure taking longer in drier air (Elsmore & De Focatiis, 2021). In a moisture-cure urethane, isocyanate groups react with water to form an amine and carbon dioxide, and the amine then reacts with more isocyanate to form urea links that build the network (Engels et al., 2013; Akindoyo et al., 2016). So too little moisture slows cure, while too thick a layer can trap the carbon dioxide formed inside as bubbles (Akindoyo et al., 2016). Between two non-porous faces, the water has no path in at all.
Start by checking for a skinned surface with a soft core when cut. Log the relative humidity at the time of work and check whether the area is dry, as in heated buildings in winter. Surface cure is measured as tack-free time, the point where a light touch no longer picks up material, and a standard test method exists (ASTM International, 2022). For depth of cure, make a coupon of the same depth and cut it at set intervals to measure the cured thickness.
If moisture is the cause, give it time and a path in. If humidity is below the TDS recommendation, allow more cure time or raise humidity. Keep joints and beads within the maximum depth on the TDS by using backer material. To bond large faces with no air access, consider switching from a moisture-cure to a two-part product. If bubbles are the issue, apply thinner and dry the surfaces.
Hypothesis 4: amine blush
This is the hypothesis when the body is hard but the surface alone is tacky or greasy, under a cloudy white film. Typical cases are two-part epoxies used somewhere cold and damp: rainy late autumn or early winter, poorly ventilated basement workshops, or winter shops running kerosene or gas heaters. For example, if an epoxy applied on a damp evening feels slick to the touch the next morning, and the coating applied over it lifts, the hardener at the surface has likely reacted with the air first.
What happened at the surface? Amine hardeners attract water and absorb carbon dioxide readily. At the exposed surface the amine can react with moisture and carbon dioxide to form carbamate, a salt of amine and carbon dioxide; the white layer it leaves is known as amine blush (Bannister et al., 2022). Amine used up this way cannot react with the epoxy, so the surface is under-cured, and the layer becomes a weak boundary that spoils adhesion of anything applied on top. The higher the humidity, the worse it gets. In one study, a carbamate surface layer formed at 80% relative humidity and shrank when the wait after mixing was longer, while at 40% and below no such difference was measured (Bannister et al., 2022). Low temperature slows the main reaction with the epoxy and gives the amine more time to react with the air, so it appears most on cold, damp days (Xia et al., 2014; Bannister et al., 2022).
The first thing to check is whether the tack is on the surface only and whether there is residue that wipes off with water. Log temperature, relative humidity, dew point, the temperature at which moisture in the air starts to condense, and part surface temperature together. A surface near the dew point collects condensation and raises the risk further. Indoors, combustion heaters add both carbon dioxide and water vapour, so check how the space is heated.
If it is blush, the answer lies in the working conditions: bring them inside the TDS range. Work only when humidity and surface temperature are within that range, and use electric rather than combustion heating. If the TDS calls for an induction time, a wait between mixing and applying, keep it; in the study above, a longer wait reduced carbamate at high humidity (Bannister et al., 2022). Before recoating a blushed surface, confirm with the TDS how to wash it with water and detergent and then sand it.
Hypothesis 5: poisoned catalyst
This is the hypothesis when a platinum-catalysed addition-cure silicone used for moulding, potting or coating cures in the body, but stays liquid or tacky only on a certain part or mould, or where a certain operator handled it. Typical cases are where rubber gloves were worn, on faces touched by rubber gaskets or tape, and on master patterns made of epoxy or urethane. For example, if the same silicone stays tacky at the interface only on a newly introduced resin pattern, suspect the pattern material before the silicone.
The reason only the interface stays soft lies in the catalyst. Addition-cure silicones cure by hydrosilylation, in which a platinum catalyst joins Si-H groups and vinyl groups on the silicone chains. The reaction leaves no by-product and cures evenly in thick sections, but because the catalyst works at very low levels, anything that binds to it is a threat (Lukin et al., 2020). Sulfur compounds are reported to deactivate the platinum catalyst completely (Lukin et al., 2020), and a silicone producer's technical guidance lists nitrogen compounds such as amines and amides, sulfides, fatty acid tin salts, phosphorus compounds, and some residual solvents and primers as inhibitors (Dow, n.d.). Common sources on the floor are sulfur-cured rubber gloves and gaskets, residues of amine-cured epoxies or urethanes, tin-catalysed condensation-cure silicones, and some tapes and masking materials.
One comparison settles whether it is inhibition. Put a small amount on the suspect material and the same amount on clean glass or aluminium as a control, and cure them side by side. If only the control cures, it is inhibition. Also log whether the soft areas coincide with certain gloves, parts or tools.
The fix is to keep inhibitors away. Use gloves that do not inhibit, such as nitrile, instead of latex or sulfur-cured rubber, and clean parts with a solvent the TDS allows. Where condensation-cure silicones or amine hardeners share the line, keep tools and benches separate. Make the side-by-side coupon test a standard step before any new part, tape or pattern material is introduced.
Hypothesis 6: wide gap, dull metal
This is the hypothesis when an anaerobic adhesive, such as a thread locker, pipe sealant or retaining compound, does not cure on time. Typical cases are stainless steel and on plated or anodised parts, in fits with wide clearance, and in cold fasteners in winter. For example, if a thread locker that holds well on carbon steel bolts falls short on breakaway torque after the parts switch to zinc-plated bolts, look at the changed surface before the adhesive. Note that adhesive squeezed out of the thread staying liquid is normal, since it is in contact with air.
Anaerobic adhesives need two things at once to cure. They contain acrylic monomers and a peroxide initiator, and while oxygen from the air is present the reaction is held back. Once oxygen is shut out in a tight metal gap, metal ions at the surface break down the peroxide into radicals, highly reactive molecular fragments, which start polymerisation, linking the monomers into chains (Abenojar et al., 2024). Cure therefore needs both oxygen exclusion and an active metal surface. Speed also depends strongly on the metal. In a DSC study, the same adhesive cured faster and more completely on copper than on iron, and the authors attribute this to copper ions lowering the energy needed to start the cure reaction (Abenojar et al., 2024). By the same mechanism, surfaces that give off few metal ions, and joints with wide gaps where oxygen remains, cure slowly or not at all.
The place to look is inside the joint. Check the substrate and its finish, such as plating or anodising, and whether the gap is within the maximum gap on the TDS. After the TDS cure time, measure breakaway torque, the force needed to first loosen the cured fastener, and compare it with the specification.
If the gap or surface is the cause, match them to the adhesive. On surfaces that give off few metal ions or in the cold, check the TDS for an activator primer from the manufacturer. For wide gaps, choose a high-viscosity or large-gap grade. Oil and cutting fluid block contact between adhesive and metal, so degrease and dry fully before assembly.
Testing the hypotheses
Six hypotheses do not mean six full investigations. Rule them out one at a time, starting with what is easy to measure and often at fault.
The first step is to check on the TDS how this adhesive cures. Whether it is a two-part product, a one-part moisture cure, a platinum-catalysed addition-cure silicone or an anaerobic decides which hypotheses to look at. Next, write down the air temperature, part surface temperature, relative humidity and dew point at the time of work, and compare them with the TDS ranges. Temperature and humidity alone screen out much of hypotheses 1, 3 and 4.
Then look at where the soft areas are, because their pattern tells you more about the cause than anything else. If everything is evenly soft, look first at temperature or moisture, hypotheses 1 or 3. Scattered patches point to mixing, hypothesis 2; a tacky surface only, to blush, hypothesis 4; softness only where a certain part or mould touched, to catalyst inhibition, hypothesis 5; and an uncured inside of a fastened joint, to hypothesis 6. If mixing is suspected, check by weight that the metering basis is right and that the equipment actually dispenses the right ratio.
If that still does not settle it, make a control coupon. Cure the same lot on clean glass or metal somewhere warm and clean, and set it beside a coupon made on the floor. If the control cures properly, the cause is the floor conditions, not the product. Only when the control also fails is it time to suspect the product: check storage temperature, time since opening and shelf life, and ask the supplier to check that lot.
Once the cause is found and the conditions fixed, the last step is to confirm strength. It is done only when lap-shear specimens or breakaway torque show the real strength is there (ISO, 2003).
One more thing to remember: the hypotheses often overlap. On a cold, damp day, 1 and 4 arrive together, and cold resin thickens and mixes poorly, so 2 follows as well. In such cases, fix temperature first and then recheck the rest, because temperature tends to amplify the other causes.
Limits and cautions
Recommended temperature, humidity, mix ratio, maximum depth and gap differ by formulation, so numeric decisions must follow the product's TDS. For structural joints, safety parts, or food and medical contact, do not decide on floor judgement alone; get the supplier and the design engineer to confirm.
Safety matters too. Amine hardeners and epoxy resins can irritate skin and, with repeated contact, sensitise it so an allergy develops, so check the product's safety data sheet, the SDS. The isocyanates in urethane adhesives can cause occupational asthma and other lung problems as well as eye and skin irritation (OSHA, n.d.). Under-cured adhesive still holds reactive components, so do not touch it bare-handed, and follow the SDS for gloves, eye protection and ventilation. When heating to speed cure, stay below the maximum temperature on the TDS.
Sources
- Abenojar, J., López de Armentia, S., del Real, J. C., & Martínez, M. A. (2024). Kinetic study of anaerobic adhesive curing on copper and iron base substrates. Materials, 17(12), 2886. https://doi.org/10.3390/ma17122886
- Akindoyo, J. O., Beg, M. D. H., Ghazali, S., Islam, M. R., Jeyaratnam, N., & Yuvaraj, A. R. (2016). Polyurethane types, synthesis and applications – a review. RSC Advances, 6(115), 114453–114482. https://doi.org/10.1039/C6RA14525F
- ASTM International. (1999). ASTM D2471-99, Standard test method for gel time and peak exothermic temperature of reacting thermosetting resins (withdrawn 2008). https://store.astm.org/d2471-99.html
- ASTM International. (2022). ASTM C679-15(2022), Standard test method for tack-free time of elastomeric sealants. https://store.astm.org/c0679-15r22.html
- Bannister, C., Guy, A., Mihaylova, R., Orgill, J., Burg, S. L., Parnell, A., & Thompson, R. L. (2022). The influence of ambient cure chemistry and stoichiometry on epoxy coating surfaces. RSC Advances, 12(44), 28746–28754. https://doi.org/10.1039/d2ra05067f
- Dow. (n.d.). Platinum-catalyzed silicone inhibitors (technical FAQ). Retrieved November 2026. https://www.dow.com/en-us/faq/platinum-catalyzed-silicone-inhibitors.html
- Elsmore, M. T., & De Focatiis, D. S. A. (2021). Combined roles of temperature and humidity on cure of a silicone elastomer. Polymer Testing, 93, 106967. https://doi.org/10.1016/j.polymertesting.2020.106967
- Engels, H.-W., Pirkl, H.-G., Albers, R., Albach, R. W., Krause, J., Hoffmann, A., Casselmann, H., & Dormish, J. (2013). Polyurethanes: Versatile materials and sustainable problem solvers for today's challenges. Angewandte Chemie International Edition, 52(36), 9422–9441. https://doi.org/10.1002/anie.201302766
- Foster, J. C., Yoon, A., Lyons, J., Martinez, E., Leguizamon, S. C., Bezik, C. T., Frischknecht, A. L., & Redline, E. M. (2023). Unexpected thermomechanical behavior of off-stoichiometry epoxy/amine materials. Macromolecules, 56(8), 3183–3194. https://www.ornl.gov/publication/unexpected-thermomechanical-behavior-stoichiometry-epoxyamine-materials
- Gillham, J. K. (1986). Formation and properties of thermosetting and high Tg polymeric materials. Polymer Engineering & Science, 26(20), 1429–1433. https://doi.org/10.1002/pen.760262012
- International Organization for Standardization [ISO]. (2003). ISO 4587:2003, Adhesives — Determination of tensile lap-shear strength of rigid-to-rigid bonded assemblies (confirmed 2021). https://www.iso.org/standard/34852.html
- Lukin, R. Yu., Kuchkaev, A. M., Sukhov, A. V., Bekmukhamedov, G. E., & Yakhvarov, D. G. (2020). Platinum-catalyzed hydrosilylation in polymer chemistry. Polymers, 12(10), 2174. https://doi.org/10.3390/polym12102174
- Montserrat, S. (1991). Vitrification and physical ageing on isothermal curing of an epoxy resin. Journal of Thermal Analysis. https://doi.org/10.1007/BF01912205
- Occupational Safety and Health Administration [OSHA]. (n.d.). Isocyanates. U.S. Department of Labor. Retrieved November 2026. https://www.osha.gov/isocyanates
- Xia, L., Zuo, L., Zha, S., Jiang, S., Guan, R., & Lu, D. (2014). Kinetic research on low-temperature cure of epoxy adhesive. International Journal of Adhesion and Adhesives, 50, 255–264. https://doi.org/10.1016/j.ijadhadh.2014.02.005