Industrial compounding of acrylonitrile-butadiene copolymer (NBR) is a negotiation among cure chemistry, filler morphology, and mixer heat history. The recurring technical inquiry described as NBR Compounding FAQ: Cure Systems, Filler Loading and Processing Window begins with polymer selection variables that dominate downstream decisions: bound acrylonitrile content, Mooney viscosity, and gel level. Commercial NBR grades typically span 18–50% bound ACN and Mooney ML(1+4) at 100°C from 20 MU to 120 MU according to ISO 289-2 and ASTM D1646. Increasing bound ACN from 18% to 34% raises fuel and oil resistance and raises glass transition temperature, but also raises polarity, mixer torque, and melt viscosity. High-Mooney grades improve green strength for mandrel-formed hose but require lower screw speed and tighter temperature control during extrusion and injection molding. The cure system is not an isolated choice. Sulfur, sulfur-donor, and peroxide systems generate distinct crosslink architectures that shift scorch time, modulus, compression set, and the permissible mixing window. Filler loading, carbon black structure, and silica surface chemistry further shift the processing window because they alter compound viscosity and shear heating. When filler loadings exceed 70 phr or when peroxide is used, the margin between complete filler dispersion and incipient scorch can fall below 5°C, requiring production equipment modifications rather than a simple formulation adjustment.

How Do Sulfur, Sulfur-Donor, and Peroxide Cure Systems Alter NBR Vulcanization Kinetics and Service Limits?

Sulfur cure systems for NBR generally contain elemental sulfur 0.5–2.5 phr, sulfenamide or thiuram accelerators 0.8–2.0 phr, zinc oxide 3–5 phr, and stearic acid 1–2 phr. Conventional high-sulfur formulations produce polysulfidic crosslinks with high tensile strength and tear propagation resistance, but the polysulfidic bond is thermally labile. Heat aging at 100°C for 70 h per ASTM D573 can reduce elongation at break to below 50% of the original value in a conventional system, depending on ACN level and antidegradant concentration. Semi-efficient and efficient vulcanization systems reduce free sulfur to 0.5–1.0 phr or replace elemental sulfur with thiuram or dithiocarbamate donors, generating shorter monosulfidic and disulfidic crosslinks. The outcome is improved compression set resistance per ASTM D395 Method B at 125°C and better retention of elongation after aging, but at the cost of reduced tear strength and longer induction time in some accelerator combinations.

Peroxide cure systems use dicumyl peroxide at 1.5–2.5 phr or 2,5-dimethyl-2,5-di(tert-butylperoxy)hexane at 1.0–2.0 phr, often with a co-agent such as triallyl cyanurate at 0.5–1.5 phr. Peroxide decomposes to form carbon–carbon crosslinks that are thermally and hydrolytically stable, giving low compression set and good resistance to hot oil and steam. The limitations are equally specific: peroxide cure produces lower tear strength than high-sulfur systems, and acidic compounding ingredients consume free radicals. Silica, clay, and acidic carbon black surfaces retard peroxide cure unless zinc oxide or magnesium oxide is present. Peroxide-cured NBR compounds must be mixed and processed at lower temperatures than sulfur systems; a practical upper dump temperature is 100–105°C in an internal mixer to avoid scorch, compared with 120–140°C for sulfur compounds. Vulcanization kinetics are monitored by moving-die rheometry per ISO 6502 or ASTM D5289 at 180°C or 190°C; minimum elastic torque S′_min and maximum torque S′_max define the degree of crosslinking, while t_c90 is used for mold dwell time calculation.

Filler Loading, Carbon Black Structure, and Silica Interactions in NBR

Carbon black selection in NBR is driven by the balance among tensile strength, compression set, extrusion die swell, and dynamic stiffness. N550 and N774 with iodine adsorption of 43 g/kg and 29 g/kg respectively per ASTM D1510 are semi-reinforcing and reduce compound Mooney rise compared with N330. The loading window in industrial compounds ranges from 20 phr in low-hardness gaskets to 120 phr in high-hardness, low-shrinkage seals. At N550 loadings above 70 phr, the filler network dominates dynamic modulus, and the Payne effect measured on a rubber process analyzer per ASTM D6601 becomes the primary source of non-linear viscoelastic behavior. The low-strain storage modulus G′ can increase by a factor of 3–6 compared with the same formulation at 30 phr loading, while loss modulus rises less steeply. This has direct consequences for low-temperature sealing force and vibration isolation.

Filler loading does not change cure kinetics proportionally. Carbon black surface area and structure adsorb curatives; high-surface-area N330 at 50–70 phr can increase nonproductive sulfur or accelerator demand by 5–15% compared with N774 at equal loading. The mechanism is not simple dilution: oxygen-containing functional groups on the black surface interact with basic accelerators and delay cure. Zinc oxide dosage therefore must be confirmed by moving-die rheometry per ISO 6502. For silica-filled NBR compounds, precipitated silica at 20–50 phr requires a bifunctional organosilane at 2–8% by weight of silica to buffer the silanol surface; without silane, diethylene glycol at 1–2 phr recovers scorch delay but does not achieve equivalent filler-filler network suppression. Silica-filled NBR also requires pre-drying at 80–100°C for 2 h when ambient relative humidity exceeds 60%, because moisture introduced during mixing can hydrolyze the silane and create porosity in molded parts. In compounds containing both silica and carbon black, the order of addition is critical: silica should be added before carbon black to allow silanization, and the first-stage dump temperature should be maintained at 130–150°C for silane coupling, but only when the final curatives are sulfur-based and added in a later stage.

Table values are indicative ranges from compounder datasheets for sulfur-cured NBR; specific values depend on ACN content, plasticizer level, and mixing quality. Fillers also alter the boundary between viscous and elastic behavior. At filler loading above 80 phr, tan δ at 60°C may fall below 0.8, reducing heat buildup but increasing modulus; however the same compound may exhibit die swell above 25% in high-shear extrusion, causing dimensional instability. In low-hardness formulations, plasticizer migration to the surface can combine with low filler content to reduce coefficient of friction below 0.4; this is undesirable for applications requiring stable stacking. Published data for this specific configuration is limited, and production validation with a capillary rheometer per ISO 11443 or a rubber process analyzer is required.

When Production-Scale Mixing Shrinks the NBR Processing Window

Production-scale mixing of NBR on intermeshing or tangential internal mixers with chamber volumes of 45–270 L is commonly performed in two stages. Stage one combines NBR, carbon black or silica, plasticizers, zinc oxide, and stearic acid at rotor speeds of 30–50 rpm. Filled NBR compounds generate heat through viscous dissipation; the rate of temperature rise in a 120 L mixer at 40 rpm can reach 3–5°C/min above 60 phr filler loading. When the target dump temperature is 120–140°C for sulfur systems, the first-stage mixing window may be limited to 3–5 min after ram closure. For peroxide compounds, the dump temperature limit of 100–105°C decreases available mixing time further and requires cooling water inlet temperatures below 15°C, rotor speed reduction to 20–35 rpm, and possibly side-feed addition of carbon black to manage heat generation. Mixing energy input is quantified by kWh/kg; laboratory scale mixers may require 0.2–0.4 kWh/kg for dispersion, but production mixers are more efficient and can achieve equivalent dispersion at 0.15–0.25 kWh/kg.

Second-stage curative addition is performed on a two-roll mill with friction ratio 1.1:1 to 1.2:1 and roll temperatures held at 50–70°C. The curatives are added at a low bank volume and cut three-quarters rolls to minimize temperature-sensitive accelerator decomposition. Compounds with Mooney scorch t_s5 below 8 min at 125°C per ASTM D1646 should not be stored more than 24 h at 25°C before fabrication; warm storage at 35°C can reduce usable storage life by 50%. These production constraints are frequently overlooked when laboratory mixing is scaled up without accounting for thermal mass and shear rate differences. A laboratory two-roll mill has a much lower shear rate and no adiabatic core, so the same compound can appear safe in the laboratory and scorch on an internal mixer.

Cold-feed extrusion of NBR compounds with Mooney viscosity above 70 MU uses screw L/D 16:1 to 20:1 and barrel zones 60–80°C. At screw speeds above 80 rpm, melt temperature can exceed 110°C in high-filler compounds, triggering scorch. Injection molding of NBR with t_s2 below 8 min at 125°C requires barrel residence time below 3 min and mold fill times below 2 s; otherwise premature crosslinking creates surface roughness and knit-line porosity. Peroxide-cured NBR compounds intended for injection molding should have a minimum t_s2 at 125°C of 5 min, and the barrel profile should not exceed 80°C. Injection molding of high-hardness NBR compounds with Shore A 85–90 may require injection pressure above 120 MPa and clamping force above 150 kN for small mold cavities. The processing window is therefore not a single temperature but a moving boundary defined by shear history, screw geometry, and curative half-life.

Seals for automatic transmission fluid and fuel systems often use sulfur-donor cured NBR with N774 or N750 at 60–80 phr to balance tensile strength, compression set, and methanol-containing fuel resistance. In such compounds, t_s2 at 125°C is held above 12 min to permit hot-runner injection molding, hardness is controlled to Shore A 65–75 per ASTM D2240, and compression set after 22 h at 125°C is specified below 35% per ASTM D395. The filler-processing interplay becomes critical at 70 phr: if the mixer dump temperature is allowed to exceed 140°C, thermal oxidative degradation reduces the polymer molecular weight and lowers tensile strength, even if no visible scorch occurs. Compounds with a required RPA G′ range for low-temperature sealing force must be validated on the production line, not only in the laboratory, because mill orientation and batch-to-batch mixing energy input affect filler dispersion and Payne effect amplitude. Published data for this specific configuration is limited; therefore lot-to-lot Mooney, rheometer, and RPA correlation with the specific mixer is required before release to compression or injection molding. The same principle applies to bonded metal-NBR components: adhesive primers are sensitive to sulfur bloom at free sulfur levels above 2.0 phr and to peroxide decomposition residues, so cure system selection must be reviewed jointly with surface preparation and molding temperature.