Cyclohexanone supply for caprolactam and adipic acid remains anchored to liquid-phase air oxidation of cyclohexane, a route in which per-pass conversion is deliberately restricted to 4–6 mol% to preserve selectivity to the cyclohexanol–cyclohexanone mixture known as KA oil at 80–85%. The oxidation is typically conducted at 413–453 K and 0.8–1.6 MPa using either a cobalt naphthenate or boric acid promoter system. The boric acid modification esterifies cyclohexanol in situ, suppressing ring-opening oxidation to adipic and glutaric acid byproducts, but it requires downstream borate hydrolysis and boric acid recovery. The separation train includes a vacuum distillation column operating with more than 80 theoretical stages because the boiling points of cyclohexanone (155.6°C) and cyclohexanol (161.1°C) are separated by only 5.5 K. Recycle ratios for unconverted cyclohexane can exceed 20:1 on a molar basis, making cyclohexane recovery and close-boiling ketone–alcohol separation the dominant energy loads. Polymer-grade cyclohexanone is typically released against assay values of ≥99.8 wt%, water content ≤0.05 wt%, and Pt-Co color ≤10 according to ASTM D1078, ASTM E203, and ASTM D1209.

Bio-Based Routes to Cyclohexanone: Status Check

The status of bio-based production is assessed across two principal pathways: depolymerization of lignin-derived phenolics followed by selective hydrogenation, and direct hydrodeoxygenation of biomass-derived oxygenates. Neither route has reached parity with petrochemical cyclohexane oxidation on installed capacity, liquid hourly space velocity, or catalyst lifetime. The central technical barrier is not cyclohexanone formation itself, but preservation of the ketone intermediate while suppressing further hydrogenation to cyclohexanol and simultaneous removal of oxygen-containing co-products that accelerate catalyst deactivation. Published data for direct cyclohexanone yield from raw bio-oil is limited, because most hydrodeoxygenation studies report total deoxygenated liquid yield rather than ketone-specific selectivity after separation.

Selective phenol hydrogenation is the most developed chemical bridge between bio-based phenolics and cyclohexanone. The reaction sequence requires 2 mol H₂ per mol phenol to form cyclohexanone; an additional 1 mol H₂ per mol cyclohexanone leads to cyclohexanol. The rate ratio between the first and second hydrogenation events controls the maximum ketone yield. Palladium and platinum catalysts can achieve high cyclohexanone selectivity at moderate phenol conversion, but selectivity typically declines as residence time increases because cyclohexanol formation becomes kinetically competitive once phenol is depleted. In batch autoclave and continuous fixed-bed studies, reported operating windows span 353–453 K and 0.5–3.0 MPa hydrogen partial pressure. Lower pressure favors cyclohexanone selectivity by reducing cyclohexanone adsorption strength, but may reduce overall reaction rate and increase phenol slip.

Bio-based phenol is not a single-component feed. Lignin depolymerization produces phenol, cresols, guaiacol, syringol, catechol, and methoxylated oligomers. Water content can range from 5–30 wt% depending on pyrolysis severity and condensation strategy. Acetic acid, formic acid, and sulfur-containing species are also present in many bio-oils. These impurities alter the catalytic surface by competitive adsorption, acid-catalyzed condensation, or metal leaching. Palladium on carbon, for example, is sensitive to acidic process liquors and can lose 30–50% of initial hydrogenation activity over extended runs if the feed pH is not buffered. Pilot hydrotreating of bio-oil has reported pressure-drop excursions in the preheat section when lignin-derived oligomers polymerize at temperatures above 523 K; published data for this specific configuration remains limited.

A fixed-bed hydrotreater processing bio-based phenolics generally requires an internal recycle or quench system because the phenol hydrogenation enthalpy is large and hot-spot formation can drive over-hydrogenation. Industrial hydrogenation reactors with a catalyst bed length-to-diameter ratio above 5:1 tend to develop axial temperature gradients exceeding 20 K at the inlet when undiluted bio-phenol is introduced. The downstream purification burden is then shifted to the vacuum distillation network, where close-boiling cyclohexanol, cresols, and partially hydrogenated methoxyphenols reduce the available tray efficiency. The conventional KA oil purification train is not configured to reject methoxy aromatic impurities without an additional hydrotreating stage.

Direct hydrodeoxygenation of lignin-derived bio-oil uses sulfided NiMo/γ-Al₂O₃ or CoMo/γ-Al₂O₃ catalysts at 573–723 K and 5–15 MPa hydrogen pressure, with weight hourly space velocities of 0.5–2 h⁻¹. The objective is oxygen removal as water, carbon monoxide, and carbon dioxide while retaining aromatic or cycloaliphatic carbon. Cyclohexanone appears as an intermediate when the feed contains guaiacol or phenol, but the ketone is readily hydrogenated to cyclohexanol under the same conditions. Sulfided catalysts require a minimum sulfur partial pressure to maintain the active metal sulfide phase; bio-based feeds with low sulfur content can draw sulfur out of the catalyst lattice, leading to loss of hydrodeoxygenation activity and increased coking. This sulfur requirement is a direct conflict with polymer-grade cyclohexanone specifications, where sulfur contamination must be rigorously excluded from caprolactam and nylon intermediates.

Can Bio-Based Phenol Hydrogenation Achieve Sufficient Selectivity?

The selectivity ceiling is dictated by the adsorption equilibrium between phenol, cyclohexanone, and hydrogen on the metal surface. On palladium and platinum catalysts, cyclohexanone desorption is favored when polar solvents or acidic supports are used. Polar solvents stabilize the enolic transition state and accelerate the tautomerization of the partially hydrogenated intermediate. Cyclohexanone selectivity above 90% at phenol conversion above 85% has been reported for Pd/C in cyclohexane and for Pd on basic supports in continuous operation. However, the same catalyst systems often show selectivity erosion of 5–15 percentage points when the feed contains 0.5–2.0 wt% cresols or methoxyphenols, because these compounds compete for hydrogenation sites and extend the required residence time.

The presence of water in bio-based phenol is not always detrimental. Water can displace cyclohexanone from the catalyst surface and reduce over-hydrogenation, but it also promotes hydrolysis of the support and increases acetic acid dissociation. Aqueous phase processing changes reactor metallurgy requirements, requiring 316L stainless steel or higher alloy construction at low pH. In batch hydrogenation, the reaction rate can be maintained by increasing catalyst loading, but continuous fixed-bed operation requires careful control of the phenol-to-hydrogen ratio. A molar H₂/phenol ratio above 2.2:1 risks cyclohexanol formation, while a ratio below 1.8:1 reduces phenol conversion and leaves residual phenol in the crude ketone.

Process control for selective hydrogenation of bio-based phenol is constrained by the absence of a single on-line analyzer that discriminates cyclohexanone from cyclohexanol without a chromatographic separation. Refinery gas chromatographs with a polar capillary column and thermal conductivity detection can quantify the ketone-to-alcohol ratio, but the cycle time is typically 3–10 min, which is too slow for tight hydrogen ratio control in a high-throughput fixed-bed reactor. Near-infrared spectroscopy calibrated to gas chromatography can reduce measurement delay to 30–60 s, but the calibration must be rebuilt when the bio-phenol feed composition changes by more than 2–5 wt%. This measurement gap is a recurring bottleneck in pilot campaigns that alternate between different lignin sources.

Direct fermentation routes to cyclohexanone have not reached pilot-scale output, and published data for this specific configuration is limited. Yeast and bacterial host platforms can produce phenol via the shikimate pathway, but titers remain economically insufficient for phenol hydrogenation feedstocks. The technology readiness level of bio-based phenol production from fermentation is generally below 3–5, whereas lignin pyrolysis and hydrodeoxygenation operate at pilot scale but with unresolved catalyst stability and separation economics.

Catalyst Deactivation Profiles and Regeneration Windows

Sulfided hydrodeoxygenation catalysts deactivate by coke deposition, sulfur loss, and metal agglomeration. In pilot units, coke accumulation in the preheat zone and first catalyst bed accounts for the majority of pressure-drop increase. Regeneration by oxidative coke burn in diluted air at 573–753 K with 0.5–2.0 vol% O₂ can restore surface area, but repeated regeneration cycles can fracture extrudates and produce fines. Fines from regenerated catalysts contribute to bed plugging and excessive pressure drop across downstream filters. For noble-metal hydrogenation catalysts, regeneration is more limited because high-temperature oxidation can sinter metal crystallites and irreversibly reduce selectivity. Palladium crystallite growth from 2 nm to 6 nm shifts product distribution toward cyclohexanol, even if overall activity is partially recovered.

A production-scale fixed-bed hydrotreater that processes bio-phenol must be designed for catalyst turnover intervals of 6–18 months rather than the 24–36 months typical of petrochemical phenol hydrogenation. The shorter interval reflects higher organic acid concentration, oligomer formation, and trace metals from biomass ash. Feed filtration to 10 µm absolute is usually insufficient for raw bio-oil; a two-stage filtration train with a 1–2 µm guard bed is required to protect the catalyst bed from inorganic particulates. In addition, the hydrotreater should avoid chloride-containing guard beds if the downstream cyclohexanone enters caprolactam production, because residual chloride causes corrosion and ammonium chloride fouling in the oxime and Beckmann rearrangement sections.

Bio-oil hydrodeoxygenation produces a two-phase liquid product containing an organic phase and an aqueous phase. The organic phase contains cyclohexanone, cyclohexanol, cyclic ketones, and unreacted phenolics. The aqueous phase contains acetic acid, methanol, and dissolved ketones. Recovery of cyclohexanone from the aqueous phase by distillation is not economical below 0.5 wt% ketone content. Liquid-liquid extraction with a low-boiling solvent, followed by extractive distillation, has been proposed, but published data for this specific configuration is limited. The extraction solvent must be fully recovered to meet volatile organic compound limits, and any residual solvent must not interfere with caprolactam oxime formation.

The final purification of bio-based cyclohexanone requires the same close-boiling vacuum distillation as petrochemical KA oil. However, the impurity profile changes: bio-based streams contain saturated cyclic alcohols, methoxy aromatics, and traces of sulfur-containing compounds. A conventional distillation train with 80–100 theoretical stages can separate cyclohexanol from cyclohexanone, but methoxyphenol derivatives with boiling points near 478–493 K require an additional hydrotreating stage or a complexation step. Sulfur content must be reduced to below 1 mg/kg for caprolactam-grade use, which is frequently achieved by a nickel guard bed operated at 423–473 K and 2–5 MPa hydrogen.

Downstream caprolactam production places a hard constraint on cyclohexanol and water content in cyclohexanone. Cyclohexanol enters the oxime formation section and can consume hydroxylamine reagent without producing oxime, lowering the effective yield. Water above 0.05 wt% promotes ammonium sulfate byproduct formation and can reduce the conversion of cyclohexanone oxime in the Beckmann rearrangement. In industrial practice, a cyclohexanone feed containing 0.2–0.5 wt% cyclohexanol is typically still processable, but the economic penalty arises from excess reagent consumption and purification load before polymerization.

If a Bio-Based Feed Stream Enters an Existing KA Oil Purification Train

Introduction of bio-based phenol or bio-derived cyclohexanone into an existing KA oil purification train requires an impurity audit, because recycle loops can accumulate non-volatile residues and inorganic salts. The main distillation column reboiler is vulnerable to fouling when trace lignin-derived oligomers exceed 100–500 mg/kg in the feed. Reboiler tube wall temperatures above 473 K then promote polymerization of residual phenolics and cause a reduction in overall heat transfer coefficient. Reboiler cleaning intervals may shorten from 12–24 months to 3–6 months depending on feed cleanliness. This operational boundary must be evaluated with a fouling test on a side-stream reboiler before continuous bio-based feed introduction.

Control of acidity is essential because bio-phenol carries carboxylic acids that concentrate in the aqueous phase and lower the pH in the overhead receiver. Carbon steel overhead lines can corrode when pH drops below 4.5, leading to iron contamination in the distillate. Upgraded metallurgy in the overhead system, typically 316L stainless steel or duplex stainless steel, is required for sustained operation with acidic bio-feeds. Neutralization with dilute sodium hydroxide is not recommended in cyclohexanone service because sodium salts can deposit in the reboiler and promote aldol condensation fouling. Instead, upstream extraction or ion exchange is used to remove organic acids ahead of distillation.

Advanced process control is required to manage the variable impurity profile of bio-based cyclohexanone. On-line gas chromatographs should be placed on the feed, distillate, and bottoms streams to track cyclohexanone-to-cyclohexanol ratio, residual phenol, and light ketone content. A ratio controller linked to reflux and reboiler duties can maintain overhead purity, but the controller must accommodate feed composition changes of 10–20% between biomass lots. Without feed-forward composition control, the column can enter a flooding condition if light organic acids and water are introduced too rapidly. Dynamic simulation with actual plant column hydraulics is recommended before transition to bio-based feed, with a maximum water content of 0.2 wt% specified for the distillation feed.

The material balance around an integrated bio-based cyclohexanone unit is dominated by hydrogen consumption and oxygen rejection. For a lignin-derived phenolic feed containing 10–20 wt% oxygen, hydrodeoxygenation to cyclohexanone consumes roughly 0.8–1.5 kg H₂ per kg oxygen removed, including saturation of aromatic rings. The resulting water is separated as an aqueous phase and requires treatment before discharge. The organic yield of cyclohexanone on a carbon basis is reduced by demethoxylation and decarboxylation reactions that release methane, carbon monoxide, and carbon dioxide. Carbon loss to gas phase of 5–15% is typical in hydrodeoxygenation of guaiacol-based feeds. The ketone retained in the organic liquid must then be separated from cyclohexanol and cyclic hydrocarbons, which increases the distillation energy demand relative to petrochemical KA oil.

The operational limits of bio-based cyclohexanone final purification are most pronounced in the presence of cyclohexene and methylcyclopentane byproducts. Cyclohexene can polymerize in the distillation column and form films on structured packing. Methylcyclopentane has a boiling point near 344.9 K, which is far below cyclohexanone, and accumulates in the light ends. A lights column with a reflux ratio of 3–6:1 is required to reject methylcyclopentane without losing cyclohexanone overhead. The addition of a lights column increases capital cost and steam demand, and is often omitted in lab-scale publications but is unavoidable in a production-scale KA oil or bio-based purification train.

Feedstock trace metals from biomass ash, including potassium, calcium, sodium, and iron, can concentrate in the hydrotreater and distillation bottoms. Potassium and sodium promote aldol condensation of cyclohexanone to high-boiling dimers, especially at reboiler temperatures above 453 K. Calcium salts deposit in reboiler tubes and reduce heat transfer. The feed specification for a bio-based cyclohexanone unit should therefore include total ash content below 50 mg/kg, with sodium and potassium individually below 10 mg/kg. When these limits cannot be met by upstream filtration, a wiped-film evaporator or flash column is inserted ahead of the main distillation train to remove inorganic and high-boiling residues.