Epoxy molding compound analysis: what resin, hardener and filler can a cured EMC or electronic encapsulant reveal?
Short answer
We can investigate retained small molecules and filler-element clues in cured epoxy molding compounds or electronic encapsulants, while resin and reacted-hardener conclusions may remain at family level. Propose only a new unused encapsulated or potted specimen, or a newly prepared cured block, for declaration and SDS review; old devices removed from service are not received. The English report would distinguish supported organic findings, measured elements and uncertain network assignments, without recovering the original resin-to-hardener mixing ratio.
Which EMC ingredients matter beyond the general epoxy resin question?
Which resin and hardener families might the cured network represent?
Electronic molding compounds can use cresol-novolac, biphenyl-type or other multifunctional epoxy resins with phenolic hardeners such as phenol-novolac or phenol-aralkyl families. Electronic potting systems can also use anhydride or amine curing chemistry, but a cured network no longer contains every starting resin and hardener in their original free forms. Our two-part epoxy guide covers the general resin-and-hardener question; this page focuses on highly filled electronic compounds and the limits of tracing their cure chemistry.
Could fillers and low-level additives explain a formulation difference?
Such compounds can contain a high mineral contribution, commonly including fused silica, alongside carbon black, release waxes, silane coupling ingredients and ion scavengers. Silicone-rubber particles can provide low-stress modification, while phosphorus-containing or brominated chemistry with antimony trioxide can form part of a flame-retardant package. These materials serve IC packaging, power devices, transformers and sensor potting, where a thin outer skin or resin-rich pocket may not represent the complete filled compound.
Can a filler or flame-retardant clue answer a separate performance question?
Our thermal-interface guide treats thermal or conductive material composition, while the flame-retardant guide treats chemical ingredient families rather than fire performance.
How would we investigate a filled cured encapsulant without inventing its starting recipe?
1. We would identify the accepted cured compound and target region, then divide representative portions, keeping package metal or other assembly material separate from the encapsulant question.
2. We would prepare accessible extractables and, where supported, releasable small-molecule clues from the network, recording which findings represent retained free material and which reflect transformed chemistry.
3. We would investigate suitable release-wax species, silane-related hydrolysis products, residual monomers and hardener-related breakdown clues with GC-MS or LC-MS/MS, limiting identities to sufficiently specific evidence.
4. We would estimate supported accessible organic contributions semi-quantitatively and investigate silicon, aluminum or antimony with ICP-MS, adding bromine and phosphorus only where the method supports it.
5. We would compare matched cured regions where a reference exists, then review alternative resin or hardener sources for each clue and leave shared or inaccessible network chemistry unresolved.
Why do small-molecule clues not uniquely reconstruct a cured epoxy network?
GC-MS separates suitable gas-phase molecules and reads their mass-spectral patterns, while LC-MS/MS separates suitable dissolved nonvolatile species and examines their ion-fragment relationships. Those routes can support accessible wax, residual or preparation-released molecules, but a fragment shared by several hardeners does not distinguish which original hardener formed the network. Curing joins reactive ingredients into a network, so an absent free hardener is not evidence that the uncured mixture lacked that hardener. ICP-MS supplies relevant elemental evidence after suitable preparation, but the elemental measurement does not establish fused-silica structure or preserve filler particle geometry, and an element shared among ingredients does not reveal its unique molecular source.
What will the electronic-compound report say, and what will remain outside its scope?
The conclusion card, component table, estimated proportions and data appendix would separate supported extractables, measured elements, resin-family evidence and hardener clues, with uncertain network contributions explained. We do not report glass-transition temperature, thermal-expansion coefficient, moisture uptake or filler particle size, and elemental findings alone do not establish filler crystal form or a complete flame-retardant molecule.
Our reports are research-grade analytical reports from a laboratory that is not accredited. They cannot be used for product registration or legal proceedings.
Which cured or uncured electronic compound should I declare?
New unused encapsulated or potted pieces and newly prepared cured blocks need declaration and SDS review; old removed devices are not received. Uncured molding granules and separate potting components are reviewed case by case after declaration, including isocyanate-containing components; non-water-based liquid components are not received as supplied, so propose a cured specimen for review. Use our industrial inquiry page and shipping guide; ship after agreement, payment and an acceptance number.
Which related guides should I read?
FAQ
Can a cured EMC identify one exact hardener?
A cured EMC may support a hardener-family inference, but shared fragments and network incorporation can prevent identification of one exact starting hardener.
Does silicon establish fused silica as the filler?
Silicon is elemental evidence and cannot independently establish the structure of the original silica filler or exclude other silicon-containing ingredients.
Can an old electronic component be sent?
Old components removed from service are not received from overseas; propose a new unused specimen or newly prepared cured block for declaration and SDS review.
Send a sample declaration and tell us which question the sample should answer.