Rubber failure analysis: why did a new O-ring, gasket or hose crack, swell or harden before it went into service?
Short answer
We can investigate chemical differences associated with cracking, swelling or hardening in a new rubber seal, gasket or hose.
We rank possible explanations supported by the evidence, without promising one unique cause or assigning responsibility.
We need the affected new part, an unfailed same-batch or earlier accepted control, and records of storage, assembly and contact exposure.
What are you really asking when a new rubber part changes?
Why did cracks appear during warehouse storage?
Fine surface cracks can suggest ozone attack on susceptible rubber or light-related ageing, with insufficient or depleted protection among the possibilities; storage under stretch can increase ozone-cracking susceptibility.
Candidate families include NBR, HNBR, EPDM, FKM, silicone rubber, chloroprene rubber and natural-rubber/SBR systems, whose responses to exposure are not interchangeable.
A hose can have a different rubber lining and outer cover, so an oil-contact finding from the lining should not be assigned to the cover without separate evidence.
Why did the seal swell after contact with oil?
Swelling can reflect fluid–rubber incompatibility: for example, a rubber family unsuited to mineral oil can absorb it and increase in size without a different filler being the explanation.
An oil-exposed seal can also retain absorbed contact medium, so a larger extract cannot automatically be read as more plasticizer originally added to the compound.
Did the supplier change the compound, or did exposure change it?
Hardening and brittleness can reflect ageing, plasticizer or softening-oil loss/extraction, or changes associated with excessive curing; chemistry alone cannot confirm the cure history.
Comparing the oil-contact face, interior and unexposed reference helps distinguish retained incoming oil from loss of native softening ingredients, although overlapping oil chemistry can leave the source unresolved.
Tackiness or softening can involve degradation, excessive plasticizer or ingredients migrating into the rubber, while batch differences can involve elastomer, filler, plasticizer or cure-system changes.
Our bloom guide addresses surface deposits; here we ask how surface and internal chemistry relate to the failed region.
How would we approach rubber seal or hose failure analysis?
1. We would establish whether inspection, assembly, storage or your bench/soak trial revealed the symptom, record the part, specified rubber, contact medium and temperature, then select unfailed or earlier accepted controls.
2. We would record locations and separate affected, apparently normal and control regions, keeping crack faces and surface material distinct from the interior so each result retains its location.
3. We would compare faces, interior and controls by FTIR for elastomer-family evidence, interpretable oxidation-related carbonyl changes and foreign surface material, yielding a relative chemical comparison.
4. We would examine suitable solvent extracts by GC-MS and GC-MS/MS for plasticizers, softening-oil patterns and accessible antioxidant/accelerator-related residues, comparing relative differences under matching preparation.
5. We would use high-resolution LC-MS and LC-MS/MS for accessible nonvolatile antioxidants and transformation products, checking whether retained protective ingredients differ from the control.
6. We would add XRF and ICP-OES for relevant zinc, silicon or calcium clues, distinguishing elemental differences from an unsupported assignment of a complete filler or activation ingredient.
7. We would combine rubber-family, extractable, protective-additive and foreign-material findings into ranked explanations, identifying missing exposure records rather than treating every changed signal as a formulation change.
Why can these methods reveal chemical clues in failed rubber?
FTIR reads molecular-bond absorption patterns, allowing interpretable family and oxidation-related comparisons, although existing ester additives can also contribute carbonyl signals.
GC-MS separates suitable gas-phase molecules from an extract and reads mass patterns; GC-MS/MS adds selected fragment evidence within the mixed rubber background.
High-resolution LC-MS adds accurate-mass clues for suitable dissolved nonvolatile additives, while LC-MS/MS examines ion-fragment relationships for candidate protective ingredients and their products.
XRF and ICP-OES provide elemental evidence, so zinc or silicon requires chemical context rather than naming a unique rubber ingredient.
Matched regions and histories make these observations interpretable, but unequal extraction or response means raw peak sizes are not original additive dosages.
What will the rubber report say, and what will remain open?
The English PDF would contain a conclusion card, location-specific component table, supported relative comparisons, data appendix and unresolved items, separating observations from ranked possible causes.
Mechanical properties, including resistance to indentation, stretching and permanent deformation after compression, and cure-network measurements are outside this study; no fluid-resistance grade is assigned.
The report does not identify a supplier or commercial grade, determine conformity with drawings, specifications or standards, or provide a unique-cause or liability judgment.
Which rubber specimens should you declare before sending?
New, unused seals, gaskets, hoses and mounts failing at inspection, assembly, storage or bench testing need declaration and SDS review with an unfailed control.
Parts removed from service, used parts, waste and scrap are not received from overseas.
Solvent-based contact liquids, test oils and other non-water-based media are not received as supplied; declare them and propose an alternative specimen for review.
Use our inquiry page and shipping guide; ship after agreement, payment and an acceptance number.
Which related guides should you read?
Our failure-analysis service explains composition evidence and the limits of causal conclusions; Our seal-material guide focuses on elastomer and filler identity rather than investigating a failed region; Our rubber-additive guide examines surviving accelerator residues and antioxidants without restoring the original cure recipe; Our batch-comparison guide explains paired specimens and why unequal preparation or storage histories complicate interpretation.
FAQ
Does a smaller antioxidant signal prove the supplier added too little?
A smaller antioxidant signal in failed rubber can reflect consumption, migration, extraction or response differences; it does not by itself establish the original dosage.
Does no extractable curing agent mean it was never used?
A curing ingredient can react into the rubber network, so absence from an extract is not proof that it was never used.
Can a control from another storage history settle the cracking cause?
An older or differently stored control introduces exposure differences alongside possible formulation differences, so it may leave the rubber-cracking explanation unresolved. A change confined to the outer rubber surface can have a different explanation from a changed elastomer pattern throughout the seal or hose.