The driver is polarity matching: ethanol's Hansen solubility parameters sit close to acrylonitrile's polar contribution, so higher-ethanol fuels pull more plasticizer and free volume into the elastomer network.
For a process engineer selecting seals, hoses, or nitrile-rubber diaphragms in flex-fuel or E85-capable equipment, the practical reading is simple: anything rated for E10 is not automatically rated for E85, and the design margin has to absorb the extra 10–20 percentage points of volumetric change plus the corresponding hardness drop and tensile-loss penalty.
What E85 and E10 actually are, chemically
E10 is pump gasoline with up to about 10% ethanol denatured into the base hydrocarbon stream; E85 is nominally 85% ethanol and 15% gasoline, though real-world U.S. shipments vary from 51% to 83% ethanol depending on region and season to keep cold-start workable [S1][S4]. That 51–83% band is the operative number for elastomer compatibility, not the "85" label on the pump.
Ethanol is polar, hygroscopic, and a mild solvent for many elastomers; gasoline is non-polar and a comparatively mild swelling agent for NBR. When you stack the two, E85's effective polarity is closer to pure ethanol than to gasoline, so any polar rubber (NBR, HNBR, FKM with low fluorine, polyacrylate) will absorb more of it [S2].
Why NBR swells: the solubility-parameter mechanism
NBR is a copolymer of acrylonitrile (ACN) and butadiene; higher ACN content raises polarity, oil resistance, and resistance to non-polar hydrocarbons, but it also raises the rubber's affinity for polar fluids such as ethanol, methanol, and ketones. The Hansen total solubility parameter for NBR (≈19–20 MPa^0.5 depending on ACN level) is closer to ethanol (≈26.5 MPa^0.5) than to gasoline (≈14–15 MPa^0.5), and the closer the polar and hydrogen-bonding components, the deeper the mutual diffusion. [S2]
Three things happen as ethanol content climbs from 10% to 70%+: the fuel penetrates the free volume between crosslink junctions, the rubber's glass-transition temperature drops (plasticization), and any extractable species (plasticizers, oligomers, antiozonants) leach out into the fuel. The net of "fuel in, plasticizer out" is the measured volume change, and for NBR in E85 that net is consistently above the same compound's response in E10 [S2].
Reading the swell numbers across the ethanol window
Typical 70-durometer NBR (≈33% ACN, the most common general-purpose grade) shows volume swell in the following bands at 23 °C / 70 h immersion: Fuel C (the legacy 50/50 toluene/isooctane reference) about 25–35%, E10 about 3–8%, E30 about 8–14%, E85 about 15–25%, and neat ethanol (E100) about 25–35% [S2]. The curve climbs with ethanol content but is not linear; there is a steep rise between E0 and E30, then a more gradual climb to E100, because the diffusion front saturates once the rubber's free volume is filled.
For the same compound at elevated temperature (60–100 °C, the under-hood or process-fluid range), those numbers climb another 3–8 percentage points, and the time to reach equilibrium drops from weeks at room temperature to days. Two side effects matter as much as the swell itself: hardness typically drops 5–15 Shore A points, and tensile strength / elongation-at-break can fall 20–40% in the swollen state, which changes the seal's contact stress in a real gland [S2].
NBR grade selection: pushing ACN is not the answer
The instinctive move, "higher ACN for more fuel resistance," works for non-polar gasoline and diesel but backfires for E85. A 45% ACN NBR will swell less in pure gasoline but swell more in E85 than a 33% ACN NBR, because the extra nitrile groups raise polarity and pull more ethanol in. For high-ethanol service, the more useful direction is HNBR (hydrogenated NBR) with moderate ACN, or FKM (fluoroelastomer) with high fluorine content, both of which decouple polarity from hydrocarbon resistance. [S1]
On a like-for-like 70-Shore-A basis at 23 °C / 168 h immersion, the published ranking in E85 is roughly FKM (high F, ≈65–68%) at 5–10% swell, HNBR at 10–15% swell, NBR (33% ACN) at 15–25% swell, and NBR (45% ACN) at 20–30% swell [S2]. FKM's fluorine content pushes its solubility parameters away from ethanol, which is why E85 fuel-system O-rings, injector O-rings, and pump shaft seals in production FFVs are almost universally FKM or FEPM, not NBR.
What this means for service life and gland design
For static NBR seals in E10 service, the usual gland fill is 85–90% and expected life is measured in years. In E85, the same gland fill is wrong: 15–25% swell closes the gap to 100%+ and the seal extrudes, nibbles, or loses contact stress as it ages. The two practical fixes are to drop gland fill to 70–80% (accepting a brief leak path until swell closes it) or to step up to HNBR / FKM so swell never exceeds the design budget. [S2]
Dynamic applications (rotary shaft seals, diaphragm pumps, flexible couplings) are less forgiving: swell plus friction plus heat accelerates wear, and NBR in E85 typically needs a 25–50% derate on continuous duty cycle versus the same seal in E10. For new FFV-spec fuel rails, OEM drawings from the past decade use FKM or HNBR almost exclusively; NBR survives in non-fuel wetted locations (oil-side seals, air-side gaskets) where the E85 never reaches it [S2].
Verification, sourcing, and standards
The reference immersion tests behind these numbers are SAE J120, ISO 1817, and ASTM D471 (rubber property testing in liquids), with fuel C, fuel CE10, fuel CE20, fuel CE50, fuel CE85, and fuel E100 as the standard liquids. The "CE" prefix means the test fuel is gasoline doped to a target ethanol percent, so CE85 is the lab-grade proxy for pump E85. SAE J1681 and several OEM internal specs (Ford WSS-M2D401, GM 9985979, FCA MS-9616) add cyclic exposure and temperature swings on top of the static immersion. [S1]
For engineers sourcing seals for E85 wetted service, the most reliable datasheet lines are not "compatible with E85" but the actual swell, hardness-change, and tensile-change values in CE85 or E100 at the service temperature, with the test duration stated. Anything quoting a single number without fluid, temperature, and time is incomplete, and anything quoting "E85 compatible" without the underlying test fluid and grade is marketing. For broader context on how elastomer selection sits inside a flex-fuel vehicle's fuel-system bill of materials, the E85 compatibility breakdown covers material-pairing decision logic that applies beyond fasteners, and the same caution about "rated for X" without the underlying test data is the rule for any rubber-to-fuel pairing.
Spec-level background on the components involved: oxy fuel cutter, and pressure transmitter.