Cyclohexanone (CAS 108-94-1; molecular weight 98.15 g/mol) is a saturated cyclic ketone with a normal boiling point of 155.6 °C at 101.3 kPa and a closed-cup flash point of 44 °C. It is consumed principally in caprolactam synthesis via cyclohexanone oxime, in adipic acid production by nitric acid oxidation, and as a low-profile solvent for vinyl chloride copolymers, nitrocellulose, and pesticide formulations. Two industrial pathways dominate supply: direct phenol hydrogenation and the ketone-alcohol (KA) oil route from cyclohexane oxidation followed by cyclohexanol dehydrogenation. The operating choice between Phenol Route vs KA Oil Route to Cyclohexanone is not merely feedstock switching; it changes reactor metallurgy, catalyst deactivation mode, purification burden, hydrogen demand, and the impurity profile reaching downstream caprolactam or adipic acid units.

What Process Conditions and Catalyst Chemistries Distinguish Phenol Route vs KA Oil Route to Cyclohexanone?

In phenol hydrogenation, the net reaction is C₆H₅OH + 2H₂ → C₆H₁₀O. Commercial units operate with supported palladium catalysts, typically Pd on alumina or activated carbon, in fixed-bed reactors. Inlet temperatures are commonly held between 130 °C and 170 °C, while reactor pressure ranges from 0.2 MPa to 1.0 MPa. Hydrogen-to-phenol molar ratios are maintained from 1.5:1 to 10:1 to suppress oligomer formation. Reported phenol conversion exceeds 99%, with cyclohexanone selectivity of 95–98 mol%. The principal selectivity conflict is over-hydrogenation to cyclohexanol, which becomes troublesome when local hydrogen partial pressure rises above the kinetic threshold for carbonyl saturation. Because the reaction is exothermic, interbed hydrogen quench or shell-and-tube reactor designs with coolant on the shell side are used to limit hot spots above 200 °C, beyond which ring hydrogenation and light-end cracking accelerate.

By contrast, the KA oil route is a two-stage sequence. In the first stage, cyclohexane is oxidized with air or oxygen-enriched air in the liquid phase at 150–165 °C and 0.8–1.2 MPa. The reaction is a radical-chain autoxidation catalyzed by soluble cobalt naphthenate or cobalt/chromium carboxylates. Per-pass cyclohexane conversion is deliberately restricted to 4–6% to preserve KA oil selectivity, which typically falls between 80% and 88%. The reactor effluent contains cyclohexane, cyclohexanone, cyclohexanol, water, formic acid, acetic acid, caproic acid, succinic acid, adipic acid, esters, and high-boiling condensation products. Unconverted cyclohexane is recovered by distillation and recycled. In the second stage, cyclohexanol is dehydrogenated over copper-based catalysts, commonly Cu/ZnO or Cu/MgO, at 250–300 °C and pressure near 0.05–0.15 MPa. Per-pass cyclohexanol conversion is typically 60–75%, with cyclohexanone selectivity above 98 mol%.

Mechanistically, phenol hydrogenation proceeds through a surface phenoxy intermediate and enol tautomerization to cyclohexanone. The selectivity to cyclohexanone versus cyclohexanol is governed by the relative rates of enol desorption and surface hydrogen addition. On palladium, electron-deficient or alkali-promoted supports tend to improve ketone selectivity. The KA oil oxidation stage, however, is governed by cyclohexyl hydroperoxide formation and decomposition. At low cyclohexane conversion, hydroperoxide decomposition is selective toward ketone and alcohol; at elevated conversion, secondary oxidation of the desired products produces carboxylic acids and esters. Consequently, the two routes present fundamentally different kinetic risk profiles: phenol hydrogenation is selectivity-limited by surface hydrogen availability, while the KA oil route is yield-limited by radical termination and acid-forming side reactions.

When Cyclohexane Oxidation Is Operated Above 8% Conversion, Selectivity Losses Escalate Nonlinearly

On production-scale bubble columns and stirred oxidation reactors, the constraint on cyclohexane conversion is not arbitrary. Raising per-pass conversion above 8% increases radical collision frequency with already-formed cyclohexanol and cyclohexanone, producing dicarboxylic acids and short-chain organic acids. Field data from air-oxidation trains show that the acid number of the oxidized product can increase from 0.5 mg KOH/g to above 1.5 mg KOH/g when conversion is pushed into the 8–10% range. The elevated acid load transfers to the saponification and distillation sections, increasing caustic consumption and generating sodium carboxylate sludges. Off-gas oxygen concentration is controlled below 2 vol% to remain outside the flammability envelope of cyclohexane-air mixtures, and reactor overhead pressure is regulated to avoid entrainment of unreacted cyclohexane into the vent system.

Bubble-column reactors for cyclohexane oxidation are typically fabricated from stainless steel and equipped with external circulation loops to remove heat and disperse air. Multiple air injection stages are common, with interstage condensers returning liquid cyclohexane to the reactor. The design pressure is generally 1.6 MPa or higher to accommodate nitrogen-blanketed operation. Agitated systems may use gas-inducing impellers, but bubble columns are often preferred because mechanical seals in stirred reactors present air leakage and maintenance bottlenecks. In one reported production configuration, a three-stage bubble column with external coolers maintained a temperature profile of 152 °C, 158 °C, and 162 °C across the stages. The staged profile allowed a slight conversion increase without exceeding the acid formation threshold. Published data for the exact selectivity gain under that specific arrangement is limited, but the operational principle is consistent with radical kinetic models.

Cyclohexanol dehydrogenation downstream of the oxidation train is an endothermic reaction requiring approximately 60–70 kJ/mol. Shell-and-tube fixed-bed reactors with catalyst particle diameters of 2–4 mm and bed length-to-diameter ratios above 5:1 are used to maintain radial temperature uniformity. The feed to dehydrogenation is subject to acid removal because residual acids neutralize the copper surface and accelerate sintering. Specification sheets for KA oil dehydrogenation feeds commonly require total acidity below 0.1 mg KOH/g and water below 0.1 wt%. If acid breakthrough occurs, catalyst activity declines rapidly, and the reactor outlet shows rising unconverted cyclohexanol. The dehydrogenation section is normally operated at liquid hourly space velocities between 0.5 h⁻¹ and 1.5 h⁻¹. At higher space velocity, cyclohexanol slip increases; at lower space velocity, the endothermic heat demand becomes difficult to supply without exceeding tube-wall temperatures that promote catalyst fouling.

In contrast to the KA oil train, phenol hydrogenation units can be brought to high per-pass conversion in a single reactor. However, the adiabatic temperature rise requires careful distribution of hydrogen. Some licensors use a split-bed design in which a portion of the hydrogen is injected at the inlet and the remainder between catalyst beds. The interbed quench suppresses the thermal excursion that would otherwise drive cyclohexanol formation. Phenol feedstock quality also matters: sulfur compounds at part-per-million levels deactivate palladium sites, while high-boiling oxygenates accumulate in recycle hydrogen and deposit on the catalyst. A guard bed containing alumina or zinc oxide is often placed upstream. Production-scale catalyst life for phenol hydrogenation is reported in the range of 12–24 months when feed phenol contains sulfur below 1 ppmw. This is an operational boundary with direct economic impact.

Purification Bottlenecks and Product Quality Standards for Downstream Caprolactam

The product from the phenol route is inherently simpler than KA oil reactor effluent. A phenol hydrogenation reactor outlet typically contains cyclohexanone, cyclohexanol, unconverted phenol, water, and minor high boilers. Distillation requirements are relatively modest: a light-ends column removes water and traces of cyclohexane, and a vacuum rectification column separates cyclohexanone from cyclohexanol and phenol. The KA oil route requires a more elaborate purification sequence. After cyclohexane recovery, the oxidized mixture is washed with water and alkali to remove water-soluble acids, then distilled to separate cyclohexanone from cyclohexanol and high boilers. The cyclohexanol fraction is sent to the dehydrogenation reactor, and the resulting crude cyclohexanone is distilled again. Residual acids, esters, and cyclohexylidene compounds can survive distillation and create color or redox instability in the final product.

Downstream caprolactam producers impose tight specifications because cyclohexanone purity directly affects oxime formation selectivity and final caprolactam color. Commercial cyclohexanone specifications frequently require assay by GC-FID above 99.8 wt%, water below 0.05 wt% by ASTM E203, Pt-Co color below 10 by ASTM D1209, distillation range from 152 °C to 157 °C at 101.3 kPa by ASTM D1078, and acidity below 0.01 wt% as acetic acid. For phenol-derived material, residual phenol is controlled below 0.10 wt% because phenol can interfere with oxime formation and contributes to color in downstream caprolactam. Cyclohexanol is often limited to 0.05 wt% or lower because it consumes oxidant in subsequent adipic acid oxidation and reduces caprolactam yield.

For KA oil-derived material, the purification bottleneck is not only separation efficiency but also trace aldehyde and ester removal. During dehydrogenation, dehydration side products such as cyclohexene and condensation products can form if the catalyst is operated too hot or if the feed contains excessive water. Aldehyde impurities, even at 0.05–0.10 wt%, can reduce cyclohexanone oxime stability and cause haze in downstream caprolactam. Some producers place a nickel guard bed operating at 80–120 °C before the final distillation column to hydrogenate trace aldehydes and olefinic impurities. If aldehyde content is not reduced below 0.05 wt%, downstream oxime formation may be retarded. This guard-bed step is generally unnecessary for phenol-route product, which is another practical advantage of the direct hydrogenation path.

Energy and Hydrogen Accounting Shifts Site-Level Process Economics

The phenol route consumes hydrogen as a primary feedstock, with a stoichiometric requirement of 0.041 kg hydrogen per 1 kg cyclohexanone produced. Actual consumption is higher because of hydrogen dissolution, vent losses, and over-hydrogenation. Published mass balances typically place phenol consumption near 0.97–1.00 kg per 1 kg of refined cyclohexanone. This route is therefore favored at sites with low-cost hydrogen from naphtha reforming, chlor-alkali byproduct, or steam methane reforming with carbon capture. However, the feedstock cost penalty of phenol relative to cyclohexane remains significant. The phenol route is less energy-intensive in distillation but more sensitive to hydrogen pricing and phenol price volatility.

The KA oil route does not require external hydrogen for cyclohexanone synthesis; in fact, the dehydrogenation step generates hydrogen, which can be recovered and used in hydrogenation processes or exported to refinery fuel gas. The overall cyclohexane consumption is typically higher than the theoretical value of 0.858 kg cyclohexane per 1 kg cyclohexanone because of acid byproducts, oxidation to CO_x, and mechanical losses. Industrial yields are commonly placed in the range of 75–85% of theoretical cyclohexanone from cyclohexane. The KA oil train consumes substantial steam in distillation and caustic in acid neutralization. The oxidation reaction also produces water, which must be separated and treated. Energy integration with the adjacent cyclohexane recovery columns and dehydrogenation reactor effluent is necessary to reduce steam consumption.

Operationally, the phenol route allows a shorter startup time and a more direct path to specification-grade cyclohexanone. It is, however, limited by the availability of high-purity phenol. Phenol containing phenolic tars or alkyl phenols above trace levels increases catalyst deactivation and raises the heavy-ends load in the final distillation column. Conversely, the KA oil route tolerates a wider range of refinery cyclohexane feedstocks but requires continuous monitoring of oxidation acid number, dehydrogenation catalyst activity, and trace carbonyl impurities. On production-scale lines, batch-to-batch variance in KA oil acid number from 0.2 mg KOH/g to 1.5 mg KOH/g is observed when upstream oxidation temperatures fluctuate, and this variance translates directly into caustic demand and catalyst life. Therefore, oxidation reactor temperature control is maintained within ±2 °C of the target profile to minimize acid excursions.

Storage of cyclohexanone requires exclusion of copper and copper alloys because trace metal dissolution accelerates color development. Moisture ingress above 0.05 wt% can create haze and reduce solvent power in coatings applications. The product is incompatible with strong oxidizing agents, and storage vessels are typically blanketed with nitrogen to limit peroxide formation. For resin and coating applications, cyclohexanone is incorporated as a high-solvency diluent in vinyl chloride copolymer formulations, where its low water miscibility relative to acetone and its evaporation rate improve film coalescence. In these formulations, the ketone content must be high because residual cyclohexanol can plasticize the film and slow hardness development. The same quality constraints that apply to caprolactam-grade cyclohexanone are therefore relevant to solvent-grade material when downstream color and drying performance are critical.