Abstract
Sulfur-cured ethylene–propylene–diene monomer (EPDM) sealing scrap cannot be melt-reprocessed; devulcanization recovers rubber only if it cuts crosslinks selectively. This review separates EPDM-direct findings from tire, natural rubber (NR) and styrene–butadiene-rubber transfers. Metrics have converged on swelling, crosslink density, sol fraction, Horikx selectivity analysis, and a rheological proxy, without a grading standard. Two EPDM studies favored milder temperatures; extrusion selectivity depended jointly on thermal profile and screw speed, and excessive severity can cause re-crosslinking. Thermomechanical and agent-assisted extrusion have run continuously on EPDM in more than one laboratory; one group’s agent-assisted devulcanizate met the tensile, compression set and hardness limits of a non-automotive seal specification, the last two only after blending with 25 wt% virgin rubber. In the only direct comparison, ethylidene-norbornene EPDM was much less reactive toward disulfides than NR, which its authors attribute to a denser, more monosulfidic network less prone to main-chain scission. A single EPDM process life cycle assessment (LCA) identifies energy as the main environmental lever. The review contributes maturity classes, scission pathways labeled proposed or demonstrated, and a minimum LCA reporting set. Closing five EPDM-direct gaps (grading standards, green-route data, in-line selectivity sensing, joint throughput–energy–quality reporting, feedstock sorting) would make continuous agent-assisted extrusion a specifiable route.