Adipic acid for nylon 66 salt preparation is manufactured industrially by the nitric acid oxidation of cyclohexanone or mixed cyclohexanone–cyclohexanol feeds. The principal stoichiometry for cyclohexanone oxidation is represented as C₆H₁₀O + 2 HNO₃ → C₆H₁₀O₄ + 2 NO + H₂O; however, the actual nitrogen distribution includes NO₂, N₂O, and N₂, with the ratio governed by temperature, nitric acid concentration, catalyst loading, and oxygen partial pressure in the reactor headspace. Cyclohexanone has a molecular mass of 98.15 g/mol, adipic acid has a molecular mass of 146.14 g/mol, and the theoretical cyclohexanone requirement is 0.67 kg per 1.00 kg of adipic acid; actual consumption exceeds this value because side reactions form glutaric acid, succinic acid, and carbon oxides. Published industrial operating envelopes place the nitric acid concentration between 45 wt% and 60 wt% and the oxidation temperature between 60°C and 90°C. The reaction is strongly exothermic, and the liquid phase is a mixture of nitric acid, water, dissolved metal nitrates, and partially oxidized intermediates; the gas phase consists of nitrogen oxides and water vapour. The oxidation is not a single-step transformation but a sequential oxidative cleavage in which the cyclohexanone ring is opened only after enolization, nitrosation, and nitrate ester formation have created activated C–C bonds adjacent to carbonyl centres. This mechanistic sequence explains why the reaction cannot be operated safely as a simple batch nitration: nitrate ester intermediates can accumulate when the reaction temperature is too low or when nitric acid is added too rapidly, creating a delayed-exotherm hazard in downstream concentration and evaporation equipment. Consequently, industrial units use continuous multi-stage reactors with controlled addition of nitric acid and continuous removal of off-gas to keep the concentration of dissolved nitrogen oxides within the range required for stable conversion. The oxidation stage is coupled to an absorber and bleach system that recovers nitric acid from off-gas and controls NO_x emissions; this coupling is not optional because the consumption of nitric acid per tonne of adipic acid is substantially higher than the stoichiometric amount when the absorber pressure and temperature are not optimised. The resulting crude adipic acid is purified to polymer-grade specifications before it is combined with hexamethylenediamine to form the salt used in nylon 66 polycondensation. The boundary between acceptable polymer-grade adipic acid and lower-grade material is determined by impurity thresholds that influence endgroup balance, colour, and thermal stability during melt polymerization and subsequent injection moulding or extrusion.

What Role Does Copper–Vanadium Redox Cycling Play in Selective C–C Bond Preservation?

The oxidation of cyclohexanone to adipic acid is selective only when the nitric acid medium contains dissolved copper and vanadium species that shuttle between oxidation states. Vanadium(V) in nitric acid functions as a one-electron oxidant that abstracts an electron from the enolised cyclohexanone ring or from a nitrosated intermediate; the resulting vanadium(IV) is reoxidised by nitric acid or by copper(II) through a coupled redox cycle. Copper(II) is reduced to copper(I) by reaction with nitric oxide or with organic radical intermediates, and copper(I) is reoxidised to copper(II) by nitric acid or by dissolved oxygen introduced in the off-gas bleach system. Published pilot-plant kinetic studies indicate that the catalyzed reaction is approximately first order in cyclohexanone up to 60% conversion and approximately zero order in nitric acid when the acid concentration remains above 45 wt%. The apparent activation energy for the copper–vanadium-catalyzed oxidation is reported in the range 65–85 kJ/mol, which is lower than the uncatalyzed nitric acid oxidation and directs the reaction toward C–C bond preservation rather than unselective over-oxidation. Without vanadium, the nitric acid oxidation of cyclohexanone can proceed through 2-nitrosocyclohexanone and nitrate ester intermediates that tend to form glutaric acid and succinic acid through decarboxylation and oxidative cleavage at multiple ring positions. When vanadium is present at 0.02–0.10 wt% and copper is present at 0.05–0.30 wt%, the oxidation is biased toward adipic acid because the metal ions accelerate the collapse of cyclic intermediates before unselective C–C scission occurs. Excess vanadium above 0.15 wt% can promote over-oxidation and increase the formation of lower dicarboxylic acids; insufficient copper can slow the reoxidation of vanadium and allow the accumulation of organic reduction products that complex with vanadium and lower the effective catalyst concentration. The Cu/V ratio is therefore controlled not merely as a total metal concentration but as a ratio, commonly between 2:1 and 5:1 on a molar basis. The redox cycle is directly coupled to the gas-liquid mass transfer characteristics of the reactor because nitric oxide formed in the oxidation layer must be removed from the liquid; if the off-gas is not removed rapidly, the reverse reaction between nitric oxide and nitric acid can suppress the reoxidation of the catalyst. This is why staged reactors are operated with deliberate gas-liquid disengagement and why the liquid recirculation rate must exceed the gas evolution rate. Published data for this specific configuration is limited, but the observed yield advantage of the copper–vanadium system over uncatalyzed nitric acid oxidation is consistently reported as 10–15 percentage points across the 45–60 wt% HNO₃ range. The catalyst is added as nitric acid-soluble copper nitrate and ammonium metavanadate; precipitation of hydrated vanadium pentoxide can occur if the local acid concentration falls below 20 wt%, and this is prevented by maintaining the bulk acid concentration above 45 wt% and by continuous catalyst injection into the recirculation loop rather than into the low-acid reactor headspace.

At nitric acid concentrations between 45 wt% and 60 wt%, the oxidation stage is configured as a continuous cascade of two to four stirred vessels or as a staged bubble column with external recirculation loops. Each reaction stage is fabricated from low-carbon 304L stainless steel, titanium grade 2, or titanium grade 7, depending on the process owner's corrosion database; titanium is preferred when the acid concentration exceeds 50 wt% because uniform corrosion in 50 wt% HNO₃ at 80°C is published as less than 0.025 mm/year with no measurable pitting. The reactor heat removal system uses shell-and-tube or plate-and-frame exchangers with cooling water or tempered water on the utility side; published vendor technical bulletins for tantalum and titanium shell-and-tube exchangers in nitric acid service list design pressures up to 1.0 MPa at 150°C and heat transfer coefficients between 1,500 W/m²·K and 2,500 W/m²·K when the tube-side velocity is maintained between 1.5 m/s and 2.5 m/s. The processing window around the temperature setpoint is narrow because pilot-plant data indicate that conversion drops below 85% when the temperature falls below 76°C, while APHA colour of molten crude adipic acid increases from 2 to 12 when the temperature is raised from 82°C to 88°C. Industrial units therefore hold the oxidation temperature within ±3°C of the setpoint, and some catalyst-injection configurations require ±2°C because the oxidation rate and off-gas generation rate are strongly coupled to the liquid-phase redox state of the vanadium species. Coolant temperature is maintained between 10°C and 18°C in the tempered water loop; cooling water at 28–32°C is not sufficient when ambient air temperatures rise above 35°C, and this is a known production bottleneck in plants without chilled water backup. Residence time in the oxidation cascade is typically 25–45 min, reactor liquid volume is between 20 m³ and 40 m³, and gas superficial velocity in the bubble column is 0.02–0.04 m/s to avoid entrainment of acid mist. The agitator in a stirred oxidation vessel is a top-entering turbine with tip speeds below 3.5 m/s, and the vessel is fitted with a gas disengagement zone sized for a liquid velocity below 0.05 m/s so that dissolved nitrogen oxides separate before the liquid enters the recirculation pump. Start-up after line stoppage requires catalyst flushing with hot 50 wt% nitric acid because localized acid depletion below 20 wt% can precipitate vanadium and create a stagnant high-nitrate-ester zone that is susceptible to delayed decomposition. The oxidation cascade is therefore operated with continuous catalyst injection, continuous nitric acid trim, and continuous off-gas pressure control; loss of any one of these three control loops shifts the reactor into a region where either conversion or selectivity deteriorates within 10–15 min.

Conversion losses are highest when the absorber pressure is below 250 kPa absolute

The oxidation off-gas contains nitric oxide, nitrogen dioxide, nitrous oxide, elemental nitrogen, water vapour, residual oxygen, and entrained nitric acid mist; its composition changes with temperature, acid concentration, and the ratio of cyclohexanone feed to nitric acid. The absorber train converts nitrogen dioxide and nitric oxide back into nitric acid by the reaction 3 NO₂ + H₂O → 2 HNO₃ + NO, and the nitric oxide generated in this absorption step must be oxidised back to nitrogen dioxide in a separate bleach column or in an oxidiser section supplied with compressed air. When the absorber pressure is below 250 kPa absolute, the partial pressure driving force for NO₂ absorption is insufficient to achieve satisfactory recovery in a column of practical height, and the tail-gas NO_x concentration rises sharply. Published NO_x recovery data from nitric acid absorber trains indicate that staged absorption at 400–700 kPa absolute and 30–50°C can remove more than 97% of the nitrogen oxides from the oxidation off-gas, while operation at 150–200 kPa absolute requires a substantially larger number of trays and still leaves three to five times more NO_x in the tail gas. The absorber is typically a bubble-cap or sieve-tray column with 20–40 theoretical stages and is fabricated from 316L stainless steel or low-carbon 304L; the recirculating liquid is dilute nitric acid, and the column is fitted with a high-efficiency mist eliminator to prevent acid carryover into the tail-gas preheater. Nitrous oxide is not absorbed in water to any useful extent, and its abatement requires either thermal decomposition at 850–1200°C in the tail-gas unit or catalytic decomposition over a copper-zinc or noble-metal catalyst. The residual NO_x after thermal decomposition and selective catalytic reduction can be maintained below 50 ppmv on a dry basis corrected to 3% O₂, which is the level required in some operating permits. A 100 kPa increase in absorber pressure between 250 kPa and 350 kPa has been reported in published process economics data to reduce nitric acid makeup by 8–12%, but this benefit is offset by higher compression cost and by the need for higher-pressure reaction gas disengagement. The bleach column is operated with 10–20% excess air relative to the stoichiometric oxygen requirement for NO oxidation, and the oxidation of nitric oxide to nitrogen dioxide is kinetically favoured by low temperature but limited by the residence time available in the interstage piping. The absorber train therefore occupies a disproportionately large capital cost in a nitric acid oxidation plant, and its performance determines both variable cost and environmental compliance. Published data for this specific configuration is limited, but the general relationship between absorber pressure and NO_x leakage is well documented in nitric acid plant licences.

Crude adipic acid exiting the oxidation section is recovered through evaporative crystallization followed by continuous pusher centrifugation. The first crystallizer is a draft-tube baffled vacuum crystallizer operated at 65–85°C and 20–40 kPa absolute; cooling is achieved by vacuum evaporation, which concentrates the mother liquor and reduces the solubility of adipic acid. The mother liquor contains glutaric acid, succinic acid, nitrate esters, dissolved copper and vanadium, and monobasic acids; a purge stream is maintained at 3–8 wt% of feed to prevent the concentration of these impurities from reaching levels that would co-crystallise with adipic acid. The crude crystals are washed on a pusher centrifuge with hot deionized water at 70°C; the wash water has a conductivity below 1 µS/cm at 25°C to avoid introducing metal ions that would otherwise remain on the crystal surface. Recrystallization is then carried out from deionized water, and the hot solution is treated with activated carbon having a specific surface area between 900 m²/g and 1,200 m²/g, an iodine number between 800 mg/g and 1,000 mg/g, and an ash content below 5 wt%. The carbon treatment step removes colour bodies and residual organic nitrates; published vendor data show that the APHA colour of a 5 g/100 mL aqueous adipic acid solution can be reduced from 20–30 to below 5 under standard contact conditions of 30 min at 80°C. After filtration through a 0.5–1.0 µm depth filter, the recrystallized adipic acid is crystallized again in a second vacuum crystallizer and dried in a fluid-bed dryer at 60–80°C with a drying air dew point between -20°C and -40°C. Final moisture is controlled below 0.20 wt% as determined by Karl Fischer titration according to ASTM E203-16; residual water above this level can hydrolyse the nylon salt and shift the endgroup balance during subsequent polycondensation. Residual nitrate ester impurities are hydrolysed by holding the crude liquor at 85°C for 30 min before crystallisation; this hold step is critical because nitrate esters that survive into the dryer can decompose on the hot metal surface and locally raise the temperature. The purified material is then sampled for polymer-grade testing, and lots that exceed the metal or colour thresholds are either re-crystallised or diverted to non-polymer applications.

Polymer-Grade Adipic Acid Specifications, Analytical Methods, and Impurity Thresholds

The polymer-grade adipic acid specification is defined by its influence on nylon 66 salt clarity, endgroup balance, thermal stability, and colour development during melt polycondensation. Merchant specifications published by adipic acid producers and nylon 66 salt producers typically require adipic acid purity above 99.7 wt% on a dry basis, with water, iron, copper, vanadium, total nitrogen, and colour controlled because each of these parameters has a direct consequence in the polymerization reactor or in the final moulding resin. Iron is particularly critical: iron above 2 mg/kg in adipic acid can act as a thermal oxidation catalyst during the 270–285°C finishing stage of nylon 66 polymerisation and has been associated with a measurable loss in relative viscosity after 15 min at 280°C. Copper above 1 mg/kg can catalyse hydroperoxide decomposition in the melt and reduce long-term heat ageing performance. Vanadium above 1 mg/kg contributes to yellow colour bodies and can interact with phosphorus-based stabilisers. Total nitrogen above 20 mg/kg indicates residual nitrate or nitro-organics and is monitored because nitrogenous impurities may decompose during finishing and create volatile species that affect polymer consistency. The colour of a 5 g/100 mL aqueous solution is determined according to ASTM D1209 and is typically limited to 5 APHA; higher colour in the acid is carried directly into the salt solution and subsequently into the polymer. Melting point or melting range is measured by differential scanning calorimetry according to ASTM D3418-21; the accepted range for high-purity adipic acid is 151.5–153.5°C. Water content is determined by ASTM E203-16 and is typically limited to 0.20 wt%. For food-contact nylon 66 resins, the salt and polymer must also comply with 21 CFR 177.1500 and applicable migration limits under the food-contact regulatory framework. The table below summarises the polymer-grade parameters, analytical methods, typical numerical limits, and the processing consequence of non-compliance.

Because nylon 66 degree of polymerization is governed by the stoichiometric balance between carboxyl and amine endgroups, the salt preparation step combines adipic acid with hexamethylenediamine at a 1:1 molar ratio in deionized water. Hexamethylenediamine has a molecular mass of 116.21 g/mol; adipic acid has a molecular mass of 146.14 g/mol; and the resulting hexamethylenediammonium adipate salt has a molecular mass of 262.35 g/mol. The salt solution is typically prepared at 45–60 wt% solids and is adjusted to a pH between 7.6 and 7.8 at 25°C as measured by a glass electrode according to ASTM E70. A pH deviation of more than 0.1 pH units corresponds to a measurable excess of one monomer and will limit the ultimate molecular weight achievable in the finishing reactor. The solution is kept under a nitrogen blanket with dissolved oxygen below 5 ppmv to prevent oxidative discolouration; storage temperature is held at 40–50°C to avoid salt crystallisation while minimising thermal hydrolysis. The salt may be isolated as a crystalline flake or spray-dried powder, but many continuous nylon 66 plants use the aqueous salt solution directly in the polycondensation feed system. Pre-filtration through 1–5 µm cartridge filters removes insoluble particulates that could act as nucleation sites or cause spinneret blockage in fibre spinning. The salt solution is assayed for total amine and total acid content, and the ratio of carboxyl endgroups to amine endgroups is maintained within 0.990–1.010. Off-spec salt can be corrected by adding adipic acid or hexamethylenediamine, but repeated corrections increase the concentration of non-volatile salts and can shift the melt viscosity profile. The final salt quality therefore depends not only on adipic acid purity but also on the hexamethylenediamine specification, which typically requires a purity above 99.6 wt% and an APHA colour below 5. The salt solution is then transferred to the polymerization section through jacketed and nitrogen-blanketed piping; residence time in the transfer line is minimised to avoid pre-polymerisation and pressure fluctuations at the autoclave inlet.

If the Nylon 66 Salt Solution Is Autoclaved Below 1.5 MPa

The thermal polycondensation of hexamethylenediammonium adipate salt is a step-growth process in which water is eliminated and the molecular weight rises only when the water is removed under controlled pressure and temperature. If the autoclave pressure is allowed to fall below 1.5 MPa before the reactor contents have reached the target temperature, water is removed too rapidly, the salt can precipitate, and the molten prepolymer may not achieve the required endgroup conversion. The standard batch cycle heats a 45–60 wt% salt solution from 180°C to 240°C under autogenous pressure; the pressure is held at 1.7–2.0 MPa by controlled venting of steam while the temperature is raised. Once the target pressure stage is completed, the reactor is depressurised at a controlled rate to atmospheric pressure while the temperature is increased to 270–280°C, after which vacuum finishing at 0.1–5 kPa absolute removes the final water and drives the molecular weight to the target range. The autoclave is fabricated from 316L stainless steel, and the agitator is a helical ribbon or anchor impeller sized for melt viscosities from 100 Pa·s to 300 Pa·s at 280°C. Continuous polymerisation lines use a two-stage reactor system with a prepolymeriser operating at 1.5–2.5 MPa and a finishing unit operating at 0.1–5 kPa absolute; the residence time in the finisher is controlled to 20–40 min to limit thermal degradation. The relative viscosity of the finished nylon 66 is measured according to ISO 307:2019 and ASTM D789-18; injection moulding grades typically have a relative viscosity between 40 and 50, while extrusion grades may be higher. Molecular weight is controlled by adding a monofunctional acid such as acetic acid or by adjusting the diamine excess; the target amine endgroup concentration for a moulding grade is between 40 mmol/kg and 60 mmol/kg, and the carboxyl endgroup concentration is similarly bounded. Heat stabilisers such as sodium hypophosphite or copper halide systems are added to the salt solution or melt to reduce degradation at 280–300°C; delustrant-grade TiO₂ is added at 0.3–2.0 wt% for fibre and film applications. If oxygen is not excluded, the melt yellows and the melt viscosity increases through crosslinking; therefore the nitrogen purge in the finisher is controlled to an oxygen level below 10 ppmv. Published data for this specific configuration is limited, but the effect of autoclave pressure on molecular weight is well established in step-growth polyamide kinetics.

Melt processing of nylon 66 compounds is performed on co-rotating intermeshing twin-screw extruders with length-to-diameter ratios from 32:1 to 48:1 and segmented screw elements capable of generating specific mechanical energy inputs between 0.15 kWh/kg and 0.25 kWh/kg. The resin must be pre-dried at 80°C to below 0.20 wt% moisture according to ISO 15512:2019; at ambient relative humidity above 60%, desiccant drying with a dew point below -30°C is required because nylon 66 absorbs water rapidly and melt hydrolysis reduces molecular weight. The barrel temperature profile is maintained between 260°C and 290°C, the melt temperature is kept between 280°C and 295°C, and the die pressure is controlled from 5 MPa to 15 MPa depending on throughput and die resistance. Injection moulding uses a mould temperature of 60–90°C and a hold pressure of 80–120 MPa; glass-fibre reinforced compounds with 30 wt% glass fibre exhibit a tensile strength of 170–190 MPa and a tensile modulus of 9–11 GPa when tested according to ISO 527-2:2012. These properties are highly sensitive to moisture, mould temperature, and fibre length retention; a drop in fibre length of 50 µm can reduce tensile strength by 5–10%. The compounder must avoid amine-based additives that shift the endgroup balance and can accelerate thermal oxidation; compatibility of any additive package is verified by melt viscosity stability testing after 15 min at 280°C, with a relative viscosity loss greater than 10% considered unacceptable. Residence time in the melt phase beyond 15 min at 285°C is not recommended because thermal degradation and branching reactions increase the melt viscosity and reduce the elongation at break. The dried resin is conveyed under dry air to the feed throat; regrind levels above 30 wt% require additional predrying and melt filtration because the recycled material carries residual contaminants and retains a different moisture history than virgin pellet. The final moulded article is conditioned to equilibrium moisture before mechanical testing; nylon 66 absorbs 2.5–3.0 wt% water at 50% relative humidity and 23°C, and this moisture plasticises the polymer and reduces tensile modulus by 20–30% relative to the dry-as-moulded state.