Epoxy Adhesive Application Guide: Surface Preparation, Mixing Ratio and Cure Schedule Data

Overview #

The specification that procurement teams most consistently get wrong when sourcing epoxy adhesives from China is not the tensile strength — it’s the mix ratio tolerance and pot life under actual shop-floor conditions, not laboratory conditions. A two-part epoxy rated at 100:40 by weight may perform exactly as specified at 23°C in a climate-controlled lab, but deliver 30–40% lower lap shear strength when mixed at ±5% ratio deviation on a production line running at 32°C ambient. We have seen this failure mode repeatedly in incoming qualification programs, and it is almost never caught by the COA alone.

Epoxy adhesives sourced from China span a wide performance range — from commodity bisphenol-A systems curing at room temperature to high-performance cycloaliphatic and novolac-modified formulations for elevated-temperature service. The selection decision is straightforward once you know which three parameters actually govern field performance: mix ratio sensitivity, glass transition temperature (Tg) under service conditions, and adhesion retention after environmental aging. Everything else is secondary.

Material Chemistry and Grade Selection by Operating Environment #

The first question to resolve before issuing an RFQ to Chinese suppliers is whether the application requires a one-part or two-part system — and if two-part, whether the cure mechanism is amine, anhydride, or polyamide. This is not a formulation detail. It determines which test data to request, which storage conditions to specify, and which failure modes to monitor at incoming inspection.

Standard bisphenol-A/epichlorohydrin (DGEBA) systems with amine hardeners represent the majority of epoxy volume sourced from China. These systems typically achieve lap shear strength of 18–25 MPa on grit-blasted steel per ASTM International D1002, with Tg values in the 60–90°C range depending on cure schedule. For applications where service temperature exceeds 120°C continuously, these grades are not adequate — and Chinese suppliers will not always flag this proactively.

Cycloaliphatic epoxy systems and novolac-modified formulations extend the Tg ceiling to 150–180°C and are used in electrical potting, aerospace bonding, and high-temperature structural applications. The performance gap between a standard DGEBA system and a novolac-modified system at 150°C service is not marginal — lap shear retention drops from approximately 85% to under 40% of room-temperature values for DGEBA, while novolac systems retain 70–75% under the same conditions.

Most Western buyers do not realize that SAC China Standards GB/T 7124 governs lap shear testing for structural adhesives in China, and the specimen geometry and surface preparation protocol differ from ASTM International D1002 in ways that can inflate reported values by 10–18%. A Chinese supplier quoting “lap shear 28 MPa” may be reporting GB/T 7124 data on polished aluminum, not grit-blasted steel per ASTM D1002. These are not equivalent. Request the test substrate, surface preparation method, and cure schedule alongside every strength value on the datasheet.

For pump-valve-seals and fluid-contact bonding applications, chemical resistance of the cured epoxy matrix is the governing parameter — not tensile strength. Amine-cured DGEBA systems show acceptable resistance to aliphatic hydrocarbons and dilute acids but degrade measurably in ketones, chlorinated solvents, and concentrated alkalis above pH 12. Anhydride-cured systems generally outperform amine systems in chemical resistance, particularly in aqueous environments above 60°C.

Mix Ratio Sensitivity, Pot Life, and Cure Schedule Verification #

This is where most sourcing decisions go wrong, and it is the section of the technical datasheet that procurement teams read least carefully.

Two-part epoxy systems are stoichiometrically sensitive. The mix ratio — expressed either by weight or by volume — must be held within a defined tolerance to achieve full cure and specified mechanical properties. For a typical amine-cured system at 100:35 by weight, a deviation of +10% hardener (100:38.5) will not improve cure — it will leave excess amine in the matrix, reduce Tg by 8–15°C, and increase moisture absorption. A deviation of -10% hardener (100:31.5) will leave unreacted epoxy groups, reduce crosslink density, and drop lap shear strength by 15–25% compared to the specified value.

In our supplier qualification program, we require suppliers to provide mix ratio sensitivity data: lap shear strength and Tg measured at ±5% and ±10% deviation from nominal ratio. Fewer than half of Chinese epoxy suppliers we have evaluated can provide this data from their own testing. When they cannot, we generate it ourselves during incoming qualification — and the results frequently reveal that the “robust” system described in the datasheet is actually sensitive to the ±5% variation that is normal in manual dispensing operations.

Pot life is the second parameter that is routinely misrepresented. Pot life is temperature-dependent and is almost always reported at 23°C in Chinese supplier datasheets. At 30°C ambient — which is realistic for factories in Guangdong, Zhejiang, or Jiangsu during summer months — pot life for a fast-cure amine system rated at 30 minutes at 23°C may drop to 15–18 minutes. For a production line with a 20-minute assembly cycle, this is a critical failure point. We always request pot life data at both 23°C and 35°C before recommending a system for production qualification.

Cure schedule verification is the third area where incoming inspection adds value. A standard cure schedule for a room-temperature epoxy might specify 24 hours at 23°C for handling strength and 7 days at 23°C for full mechanical properties. Elevated-temperature post-cure — typically 60–80°C for 2–4 hours — can achieve equivalent properties in 2–4 hours and is often necessary to reach the specified Tg. Chinese suppliers frequently list the elevated-temperature cure schedule as optional in the datasheet, but for applications requiring Tg above 80°C, it is not optional — it is mandatory. We have seen assemblies fail in service at 90°C because the post-cure step was omitted during production ramp-up, and the room-temperature-cured Tg of 72°C was insufficient.

Per ASTM International D4065, dynamic mechanical analysis (DMA) is the most reliable method for confirming Tg of a cured epoxy system. DSC (differential scanning calorimetry) per ASTM International E1356 is acceptable but typically reports Tg 5–10°C lower than DMA for the same system. When comparing Tg values across suppliers, confirm which method was used — the difference is not trivial if your service temperature is close to the rated limit.

Surface Preparation Requirements and Adhesion Performance by Substrate #

Epoxy adhesive performance is substrate-dependent in ways that are not always obvious from the datasheet. The adhesive’s bulk mechanical properties — tensile modulus, elongation at break, Tg — are fixed by the formulation. The bond strength in service is determined by the interface, which is controlled by surface preparation. This distinction matters enormously when evaluating Chinese supplier test data.

For steel substrates, grit blasting to Sa 2.5 per ISO Standards 8501-1 followed by solvent wipe is the baseline preparation for structural bonding. This achieves surface roughness Ra of 3–6 µm and removes mill scale, rust, and contamination. Lap shear values on properly prepared steel typically run 20–28 MPa for standard DGEBA/amine systems. On mill-finish steel with only solvent wipe, the same adhesive may achieve only 10–14 MPa — and more importantly, the failure mode shifts from cohesive (within the adhesive) to adhesive (at the interface), which is the failure mode associated with premature bond degradation in service.

For aluminum, chromate conversion coating or anodizing provides the best adhesion durability, particularly in humid environments. Bare aluminum with only abrasion and solvent wipe will show acceptable initial lap shear (typically 16–22 MPa) but significant strength loss after 1,000 hours of salt spray per ASTM International B117 — typically 30–50% reduction for amine-cured systems on untreated aluminum versus 10–15% reduction on anodized aluminum.

For gaskets-sheet-sealing and composite bonding applications, surface energy is the governing parameter. Epoxy adhesives require a minimum substrate surface energy of approximately 38 mN/m for adequate wetting. Low-surface-energy substrates — PTFE, polyethylene, polypropylene — require plasma treatment, corona treatment, or chemical etching to achieve bondable surface energy above 50 mN/m. Chinese suppliers rarely include surface energy requirements in their datasheets, and this omission causes bond failures that are then incorrectly attributed to the adhesive formulation.

The practical threshold we apply in qualification: if peel strength on the target substrate after 72 hours at 40°C/95% RH is less than 80% of the dry peel strength, the surface preparation protocol is inadequate — regardless of what the initial bond strength shows.

Practical Guidance for Buyers #

When sourcing epoxy adhesives from China, the first specification to request is not tensile strength — it is mix ratio sensitivity data and pot life at your actual shop-floor temperature, not at 23°C. Most Chinese suppliers report performance at laboratory conditions. If your production environment runs at 28–35°C, you need data at those conditions before committing to volume.

The most common sourcing mistake we see is accepting a COA that lists lap shear strength without specifying the test substrate, surface preparation method, and cure schedule. A value of “25 MPa lap shear” is meaningless without this context — and Chinese suppliers frequently report values on polished aluminum per GB/T 7124, which will not replicate on your grit-blasted steel assemblies per ASTM D1002.

Before committing to volume order, require three things: first, a mix ratio sensitivity report showing mechanical properties at ±10% deviation from nominal ratio; second, pot life data at both 23°C and your maximum ambient production temperature; third, a DMA or DSC trace confirming Tg after your intended cure schedule — not the supplier’s recommended schedule, yours. If the supplier cannot provide all three, qualify a different supplier. The cost of incoming qualification testing is always lower than the cost of field bond failures.

In our qualification program, we also require three consecutive production batch COAs before recommending a supplier for approved vendor list status. Lot-to-lot consistency in epoxy formulation — particularly hardener stoichiometry and filler loading — is the variable that separates reliable Chinese suppliers from those who pass initial sample approval and then drift in production.

Frequently Asked Questions #

Q1: What is the most important parameter to verify on a Chinese epoxy supplier’s COA?

A: Mix ratio tolerance and the test conditions for reported lap shear strength — specifically the substrate, surface preparation method, and cure schedule. Shore hardness and color are easy to control; stoichiometric consistency across production batches is not.

Q2: How do I select between amine-cured and anhydride-cured epoxy systems for elevated-temperature service?

A: If your continuous service temperature exceeds 120°C, amine-cured DGEBA systems are not adequate — their Tg ceiling of 80–90°C is insufficient. Anhydride-cured systems reach Tg of 100–130°C and are the standard choice for electrical potting and thermal cycling applications. For service above 150°C, specify novolac-modified systems and confirm Tg by DMA per ASTM International D4065, not just by DSC.

Q3: Why do bond failures occur even when the supplier’s datasheet shows adequate strength?

A: This is where most sourcing decisions go wrong. The most common cause is surface preparation that does not match the test conditions used to generate the datasheet data. A supplier reporting 25 MPa on polished aluminum will not replicate on your production substrate without equivalent preparation. The second most common cause is pot life exceedance at elevated ambient temperature — a 30-minute pot life at 23°C may drop to 15–18 minutes at 35°C.

Q4: What certification or test documentation should I require before approving a Chinese epoxy supplier for structural bonding applications?

A: Request DMA traces confirming Tg per ASTM International D4065, lap shear data per ASTM International D1002 on your specified substrate and surface preparation, and salt spray adhesion retention data per ASTM International B117 at 500 hours minimum. For food-contact or potable water applications, also require NSF International certification documentation — not just a supplier declaration.

Q5: Is a higher mix ratio hardener always better for faster cure?

A: No — and this is the most common misconception we encounter. Excess hardener above the stoichiometric ratio does not accelerate cure; it leaves unreacted amine in the matrix, reduces Tg by 8–15°C, and increases moisture sensitivity. Hold the mix ratio within ±5% of nominal.

Published by sinoraw.com Technical Team | Request a sourcing consultation