A grip zone that peeled off in the field doesn’t tell you why on its own — the fracture surface does. Before changing elastomer grade, mold temperature, or substrate, a short forensic check on the failed part usually narrows the cause to one of four categories, and guessing wrong wastes a full mold-trial cycle.

Step 1: Read the Fracture Surface First

Pull a failed sample apart (or examine one that already separated) and look at where the break actually occurred. Cohesive failure — the elastomer itself tears, leaving a thin layer of it still stuck to the substrate on both sides of the break — indicates the bond interface was actually stronger than the material, meaning the chemistry match was correct and the failure lies elsewhere (usually part geometry or a stress concentration). Adhesive failure — a clean separation at the interface with the substrate showing bare and the elastomer intact — means the interface itself never formed a real chemical bond, which points straight at a chemistry mismatch or a contaminated/under-prepared surface. Confusing these two is the single most common diagnostic mistake: teams sometimes reformulate an elastomer that was never the problem because they didn’t check which side of the interface actually failed.

Step 2: Quantify With a Peel or Cross-Hatch Test

Visual inspection narrows the category; a simple test confirms it. A 90-degree or T-peel pull on a sacrificial sample gives a force-per-width figure — adhesive-failure interfaces on genuinely incompatible pairs (TPU on unprimed polypropylene, for example) typically peel at well under 1 N/mm, while a correctly matched pair holding a cohesive bond usually requires enough force to tear the elastomer before the interface gives at all. Where peel testing isn’t practical on a finished part geometry, a cross-hatch adhesion test (a grid scored through the elastomer down to the substrate, then tape-pulled) gives a fast pass/fail read: any grid squares lifting cleanly indicate an adhesive-failure-prone interface even before the part sees service loading.

Step 3: Rule Out Process Variables Before Blaming Chemistry

A correct chemistry match can still fail to bond if the molding process undermines it. Three process variables account for most “chemistry looked right but bond failed anyway” cases:

• Substrate moisture at time of overmold. Hygroscopic substrates like nylon (PA6/PA66) that absorb ambient moisture between drying and molding can off-gas steam at the mold interface during the overshot, creating microscopic voids at the bond line that never show up until a peel or field-stress event opens them. A substrate re-dried within 2 hours of the second shot, versus one sitting on the shop floor overnight, can show a measurably different bond strength on an otherwise identical formulation.
• Second-shot mold temperature below the elastomer’s flow window. Most TPU and TPE grades need the mold surface warm enough for the polymer to wet out the substrate microtexture before it starts solidifying; a mold running 15–20°C below the recommended window often produces a bond that looks fine on the day of molding but shows reduced peel strength after even mild thermal cycling, because the interface never achieved full molecular contact.
• Residual mold release or degreasing agent on the substrate. Even a substrate with correct base chemistry for the elastomer can fail to bond if a release agent film from the first-shot mold sits on the surface — this produces a false adhesive-failure result that looks like a chemistry mismatch but resolves entirely with a substrate cleaning step before the second shot, no material change required.

Step 4: When the Failure Really Is a Chemistry Mismatch

If the fracture surface shows clean adhesive failure, process variables check out clean, and peel force sits well below expectations, the elastomer sub-class genuinely doesn’t match the substrate’s surface chemistry. This is common when a “default” elastomer gets specified early in a design (often SEBS, because it’s inexpensive and easy to mold) without confirming it against the substrate that ended up in the final bill of materials — a substrate swap late in development is a frequent, quietly overlooked cause. At that point the fix isn’t a process tweak; it requires either switching to a chemically matched elastomer sub-class for the substrate in question, introducing a primer or surface-activation step, or adding a mechanical interlock feature (through-holes or an undercut) that doesn’t depend on chemical adhesion at all.

Step 5: Confirm the Fix With Accelerated Aging, Not Just a Fresh Pull

A corrected bond that passes a peel test immediately after molding can still fail in service if the interface degrades under heat, humidity, or chemical exposure the original test didn’t simulate. Running corrected samples through a boiling-water immersion (typically 1–4 hours, substrate-dependent) or a heat-aging cycle at the assembly’s expected service temperature before signing off on a fix catches degradation modes that a same-day peel test misses entirely — this is especially relevant for parts that will see repeated cleaning-agent exposure or outdoor thermal cycling in the field. Email Us for guidance on an aging protocol matched to your service environment.

Building the Root-Cause Habit Into a Development Program

Treating a bond failure as a single-cause event — “must be the wrong TPE” — skips the fracture-surface step that would have told you in five minutes whether the interface or the elastomer itself gave way. Teams that document fracture mode, peel data, and process conditions on every failed sample build a reference library that shortens root-cause time on the next failure, rather than starting the diagnostic process from zero each time. For the underlying compatibility framework this diagnostic approach assumes, see our companion guide on TPU and TPE compatibility differences for beginners, and for substrate-specific grade selection once a chemistry mismatch is confirmed, see Incure’s Uni-Weld™ plastic bonder guide.

Incure’s adhesion and primer systems support root-cause correction across the polar and non-polar substrate chemistries covered above, whether the fix turns out to be a process change, a primer, or a different elastomer family entirely. Contact Our Team for help running a fracture-surface review on a failed sample before committing to a rework plan.

Visit www.incurelab.com for more information.