Ethylbenzene dehydrogenation in adiabatic radial-flow reactors is operated with superheated dilution steam to supply a substantial fraction of the endothermic reaction heat, lower the ethylbenzene partial pressure, suppress thermal cracking of the aromatic ring, and gasify amorphous carbon from the iron-oxide catalyst surface. The steam-to-oil mass ratio is conventionally maintained in the range 1.2–2.0 kg steam/kg ethylbenzene, with dilution steam superheated to 700–800 °C and injected before the first catalyst bed. When the ratio is reduced below 1.0 kg/kg, the combined feed heat capacity declines by 20–35% compared with the 1.5 kg/kg baseline, the equilibrium conversion of ethylbenzene per pass is lowered, and the water vapour partial pressure available for carbon gasification falls below the threshold at which the gasification reaction C + H₂O ⇌ CO + H₂ can keep pace with carbon-forming side reactions such as dealkylation and styrene condensation. These effects are non-uniform through the bed. The first bed sees the highest ethylbenzene partial pressure and the lowest steam concentration, while downstream beds receive hydrogen and carbon monoxide produced by earlier dehydrogenation and water-gas shift, shifting the local redox potential. Low steam-to-oil operation therefore changes not only the quantity of coke but also its chemical structure and axial location. Short-cycle operation is usually limited by the pressure drop increase across the first bed rather than by the loss of total conversion, because front-bed coke deposits consolidate into a dense annulus that disturbs radial gas flow. This operational mode is relevant to production sites seeking reduced steam consumption and lower energy costs, but it imposes stricter monitoring of catalyst bed temperature profiles and requires catalyst formulations with higher resistance to carbon accumulation and potassium volatilization. The principal process constraint is that the reaction heat input and dilution effect are coupled through the same steam stream, so any reduction in steam-to-oil ratio reduces both the endothermic heat supply and the equilibrium shift simultaneously.

Does a Reduced Steam-to-Oil Ratio Alter Coke Gasification Kinetics?

The gasification of deposited coke is a heterogeneous reaction whose rate depends on the surface carbon structure, the partial pressure of water vapour, and the bed temperature. At steam-to-oil mass ratios below 1.0–1.2 kg/kg, the water vapour partial pressure at the reactor inlet falls from approximately 0.35–0.45 bar to below 0.25 bar, depending on total pressure and pressure drop. This reduction moves the gasification equilibrium toward carbon deposition and lowers the net removal rate of graphitic overlayers. Coke morphology changes from loosely bound aliphatic and alkylaromatic films to condensed polyaromatic structures with lower H/C ratios, typically below 0.4. The Fe-K-Ce-Mo oxide catalyst relies on potassium ferrite phases, KFeO₂ and K₂Fe₂₂O₃₄, to maintain surface oxygen mobility and facilitate carbon gasification. When the steam partial pressure is insufficient, carbon species condense into polyaromatic sheets and the iron oxide surface becomes partially reduced from Fe³⁺ to Fe²⁺, which weakens the interaction between the active phase and the potassium promoter. The cycle time penalty is not linear. A decrease in steam-to-oil ratio from 1.5 kg/kg to 1.0 kg/kg may increase the coking rate by a factor of 1.5–3.0, depending on bed age, inlet temperature, and promoter loading. Published data for specific commercial catalyst configurations at steam-to-oil ratios below 0.9 kg/kg are limited, but spent-catalyst analysis from pilot rigs shows front-bed carbon concentrations exceeding 8–12 wt% after 30 days when the ratio is held at 0.8 kg/kg. The increased carbon inventory accelerates the formation of hard coke species that require oxidation temperatures above 500 °C for complete removal, narrowing the safe operating window for regeneration.

In the first catalyst bed, the water-gas shift reaction simultaneously consumes water and generates hydrogen, further reducing the local steam-to-carbon ratio and increasing the hydrogen partial pressure. Hydrogen is a product of the main dehydrogenation reaction, and its accumulation suppresses the equilibrium conversion of ethylbenzene to styrene. At an inlet steam-to-oil ratio of 0.8 kg/kg, the hydrogen partial pressure at the first-bed outlet can approach 0.10–0.15 bar compared with 0.05–0.08 bar at 1.5 kg/kg. This higher hydrogen partial pressure promotes the re-hydrogenation of styrene precursors and favours the formation of heavy alkylbenzene condensation products that remain on the catalyst surface. The resulting carbon species have a higher thermal stability and require higher regeneration temperatures. The coupling between water-gas shift and carbon gasification means that a low steam-to-oil ratio does not simply reduce the steam available for carbon removal; it also makes the remaining steam less effective because the reverse water-gas shift can convert CO and H₂ back to carbon and water under locally hydrogen-rich conditions. The net effect is that the steady-state carbon inventory on the catalyst increases until either the gasification rate balances the coking rate or the pressure drop forces a regeneration cycle.

Cycle time in a three-bed radial-flow reactor is typically defined as the elapsed time between the start of hydrocarbon feed and the point at which the first-bed outlet temperature must be raised to the upper operating limit to maintain styrene selectivity. In a conventional arrangement with interstage steam reheat, low steam-to-oil operation causes the front-bed pressure drop to rise earlier, because coke deposition increases the pressure drop coefficient of the packed annulus. At total superficial mass fluxes of 8–15 kg/m²·s and catalyst pellet equivalent diameters of 3.0–4.8 mm, an increase in bed solids volume fraction from 0.42 to 0.55 due to carbon agglomeration can raise the bed ΔP by 40–70%. The resulting radial maldistribution creates hot spots in the central region and cool zones near the reactor wall; wall-adjacent catalyst may remain below 600 °C while the central portion exceeds 640 °C. These thermal nonuniformities accelerate potassium migration from the catalyst surface in the hot zones, producing a permanent decline in selectivity that cannot be recovered by decoking. This is a critical operational boundary: low steam-to-oil operation may be sustainable for short campaigns but becomes self-amplifying once the front-bed coke level reaches the threshold where radial gas redistribution is compromised. Process control based only on average bed temperature fails to capture this spatial spread, so operators must rely on multilevel thermocouples in the radial bed and on pressure-drop trending. When the first-bed pressure drop exceeds 0.20–0.30 bar, the reactor usually must be switched to regeneration or the steam-to-oil ratio raised to avoid irreversible damage.

Thermal Degradation Pathways in Fe-K-Ce-Mo Oxide Dehydrogenation Catalysts

Potassium promotion is essential for the catalytic activity of ethylbenzene dehydrogenation over iron oxide. The working surface contains a dynamic equilibrium between KFeO₂, K₂Fe₂₂O₃₄, and free Fe₂O₃, with cerium and molybdenum functioning as structural stabilizers. Under low steam-to-oil conditions, the reducing potential of the atmosphere increases because less steam is present relative to generated hydrogen. This reduction favours the transformation of Fe³⁺ to Fe²⁺ in the near-surface region, weakening the potassium ferrite framework and lowering the catalyst's resistance to attrition. Attrition measurements on commercial extrudates using the rotating drum method of ASTM D4058-15 show that catalyst fines can increase from 0.5–1.0 wt% to 2.0–3.5 wt% after 100 hours when the inlet steam-to-oil ratio is reduced from 1.5 kg/kg to 0.8 kg/kg. Pore structure changes are similarly significant. Mercury intrusion porosimetry by ISO 15901-1:2016 shows a loss of pore volume in the 10–50 nm diameter range from 0.18–0.24 cm³/g to 0.09–0.14 cm³/g in spent front-bed samples. Single-pellet crush strength measured by ASTM D4179-22 can decline from 2.5–3.5 N/mm to 1.4–2.0 N/mm in the coked front-bed layer, though the strength partially recovers after controlled decoking. These physical changes increase the effective diffusion resistance for ethylbenzene and styrene, decreasing observed selectivity. The degradation is partially reversible through oxidation at 450–500 °C with air diluted to keep bed temperatures below 650 °C, but potassium loss induced by high-temperature excursions is permanent. Cerium and molybdenum promoters reduce the irreversible attrition loss by inhibiting the growth of large Fe₂O₃ crystallites, but their effectiveness depends on maintaining the steam-to-oil ratio above the promoter-specific stability limit.

The operating data in Table 1 indicate that the first-bed pressure drop and carbon inventory are the primary cycle-time constraints under low steam-to-oil operation. The measurement methods guard against process variability, but the values should not be extrapolated beyond the stated catalyst family and bed geometry. In axial-flow pilot reactors the same steam-to-oil reduction produces a different deposition pattern, because axial temperature gradients are more pronounced and the catalyst bed lacks the radial flow distribution that commercial reactors use. Published data for commercial cycle times at steam-to-oil ratios below 0.9 kg/kg are limited, and the values reported above are screening observables rather than performance guarantees.

When the S/O Mass Ratio Falls Below Unit Stoichiometric Gasification Demand

When the inlet steam-to-oil mass ratio is deliberately reduced from 1.5 kg/kg to 0.8 kg/kg, the process approaches the stoichiometric threshold at which the water vapour supply is barely sufficient to gasify the carbon formed in parallel coking reactions. At this point, the partial pressure of water vapour at the first-bed outlet is further reduced by consumption in the water-gas shift reaction, and the remaining steam may be less than 0.10 bar at the bed exit. This creates a coke deposition front that moves downstream faster than in baseline operation; the front-bed carbon accumulation rate can exceed 1.5–3.0× the baseline rate, while the second bed begins to accumulate measurable carbon within 7–14 days. The low steam-to-oil condition also reduces the total mass flux through the radial bed, which initially lowers the pressure drop but later amplifies flow maldistribution because the lower steam flow is less able to sweep carbon precursors out of the catalyst pores. Ethylbenzene diffusion into the catalyst pores becomes limited by the accumulation of heavy alkylaromatic species, and the intraparticle temperature difference increases. Thermal cracking to benzene and toluene becomes more significant when the ethylbenzene partial pressure remains high, lowering styrene selectivity and increasing the downstream separation load. The dehydrogenation reactor effluent contains more condensable oligomers that can foul the feed-effluent heat exchanger and the vent gas compressor suction drum, unless the unit includes a quench oil circulation loop with adequate filtration and tar removal. For these reasons, low steam-to-oil operation is generally restricted to campaigns where the unit has excess styrene selectivity margin, and the reactor heater and downstream recovery train can tolerate higher fouling rates.

Decoking of a spent front bed must be executed with controlled air or steam-air mixtures to avoid a runaway temperature excursion caused by the highly exothermic carbon oxidation reaction. A typical regeneration procedure uses 0.5–1.0 vol% O₂ in nitrogen first, with the bed outlet temperature held below 450 °C, then gradually increases oxygen to 2–4 vol% after carbon monoxide measurements fall below 2000 ppm. The maximum allowed bed temperature during decoking is usually 650 °C, because higher temperatures cause irreversible potassium volatilization and structural rearrangement of the iron oxide crystallites. The cycle time is determined by the time required to return the first-bed pressure drop to baseline and by the post-regeneration activity recovery. A catalyst subjected to repeated low steam-to-oil cycles may lose 10–15% of its fresh surface area after 3–5 cycles, as measured by ISO 9277:2010 BET nitrogen adsorption, even when the pressure drop is fully restored. This loss is attributable to partial collapse of mesoporosity during coke burn-off and to potassium redistribution. In such cases, the unit may require an early change-out of the first-bed catalyst charge, which is operationally preferable to running the whole reactor beyond the selectivity limit. These limitations define the lower steam-to-oil boundary for commercially viable cycle times in fixed-bed ethylbenzene dehydrogenation units.