N2O From a Waste Compound to a Reactant for Downstream Applications

As an alternative, N2O can be recovered from the off-gas in pure form, either for selling or for use as an oxidant in some downstream processes. Within this context, an innovative solution has been developed by Solutia, together with the Boreskov Institute of Catalysis, in which N2O is the oxidant used for the hydroxylation of benzene to phenol in the presence of a ZSM-5 catalyst exchanged with Fe(III) [11]. Phenol can then be hydrogenated to yield cyclohexanol, hence completing the N2O cycle (Scheme 7.3).

A growing number of gas-phase applications are being investigated in which N2O is the co-reactant, for example, the oxidation of methane to methanol, the epoxidation

Scheme 7.3 Integration in the BIC/Solutia process.

Scheme 7.3 Integration in the BIC/Solutia process.

of olefins, the oxidative dehydrogenation of alkanes to alkenes, amongst others [11h]. Within this context, it is useful to remember that the cost of N2O is more than ten times that of oxygen; clearly, from a cost standpoint, it is difficult to justify new commodity chemical processes using purposely produced N2O as the oxygen source. However, for fine chemical applications such as pharmaceuticals and agrochemicals the cost may well be justified. In addition, any downstream process that may use the N2O-containing stream - possibly without any further treatment, and thus realizing efficient process integration - may constitute a valid alternative to the disposal of that stream [11h].

In the BIC/Solutia process, using a zeolite that contains only small amounts of iron, benzene can be oxidized to phenol with a selectivity of over 95% at around 3000C, but N2O selectivity is lower than 95% [11e].

Despite the remarkable performance of this innovative process, which was scaled up to the pilot unit, the BIC/Solutia process has not yet been put into commercial operation. This may be due to rapid catalyst deactivation caused by tar deposition, to low efficiency with respect to N2O and to the poor economics of such small-sized phenol plants, suited to balance the AA process unit, as compared to bigger, traditional plants for phenol production. An alternative option would be to build large phenol plants and produce N2O separately. For instance, a patent belonging to Solutia [11g] discusses the use of a Bi/Mn/Al/O catalyst for the oxidation of ammonia to nitrous oxide. N2O selectivity is reported to be about 92% at 99.2% conversion. The cost of N2O is projected to be about 25% that of H2O2 [11h].

Concerning other reactions that make use of nitrous oxide as the oxidant, a process developed by BASF is worth mentioning, which involves the synthesis of cyclododecanone, a raw material for Nylon-12 and Nylon 6-12 [12]. Cyclodode-canone is oxidized to 1,12-dodecanedioic acid by oxidative cleavage with nitric acid. It is currently made by a five-step sequence from cyclododecatriene. The first step of the new process is the synthesis of cyclododecatriene from butadiene, which is then reacted with nitrous oxide, to produce cyclododecadienone, which is finally hydrogenated to cyclododecanone. The yield to cyclododecanone is significantly higher than with the conventional process, waste is reduced and investment costs are lower. BASF already has a contracted customer and is building a plant due to come on stream in 2009.

Alternatives for AA Production

Oxidation of KA Oil with Air

The second step of the KA Oil process can be carried out with oxygen as the oxidant, in place of nitric acid, and catalytic amounts of Co and Mn acetate, at 70-80 ° C, in acetic acid solvent 13. The fact that N2O is not generated and that the dedicated HNO3 plant is not necessary (when a large-scale HNO3 plant is not already available) may make the air-based process an alternative to the process currently employed. However, the potential use in industrial applications is still unclear, due to the lower yield achieved than that with nitric acid, and also due to the lower quality of the AA obtained. Moreover, using acetic acid as the solvent leads to severe corrosion problems, particularly in combination with the Mn and Co salts.

In this reaction it is important to achieve 100% conversion of the reactant, because recycling of unconverted KA Oil complicates the process. Yields to AA reported in most of the patent literature are not higher than 70% at high Ol/One conversion, with an overall yield to diacids close to 80%.

AA is probably generated via the intermediate formation of 2-hydroxycyclohexanone and 6-oxohexanoic acid. The mechanism may potentially include the formation of the enol tautomer of cyclohexanone, favored by the presence of an acid (Scheme 7.4; this may also be an alternative mechanism for cyclohexanone activation). The cyclohexen-1-ol formed is then oxidized to 2-hydroperoxycyclohexanone. The hydroperoxide generates 2-hydroxycyclohexanone, which is then cleaved to 6-oxohexanoic acid. The latter is then converted into AA via monoperoxyadipic acid; this step is eventually catalyzed by cobalt. Scheme 7.5 shows the steps in the mechanism of reaction.

Side-reactions lead to the formation of lighter acids. For instance, monoperox-yadipic acid can decarboxylate to yield the pentanoic acid radical, precursor of the byproduct valeric acid. The same C5 radical may react with O2 to yield 5-oxopentanoic acid, which is then oxidized to monoperoxyglutaric acid, a precursor of glutaric acid. An analogous mechanism starting from the butanoic acid radical may yield the byproduct succinic acid. Azelaic acid may form by the coupling of radical species (e.g., between the butanoic acid radical and the pentanoic acid radical), whereas the dimerization of the pentanoic acid radical may yield the by-product sebacic acid.

Scheme 7.4 One possible mechanism of cyclohexanone activation via enol tautomer.

Scheme 7.4 One possible mechanism of cyclohexanone activation via enol tautomer.

Scheme 7.5 Mechanism for the oxidation of cyclohexanone with O2. Source: elaborated from [130].

Researchers from Asahi have ascertained the industrial feasibility of a process based on a Co/Mn catalyst and either pure oxygen at atmospheric pressure or nitrogen-diluted air at 12 atm (to avoid explosion hazards), with water and acetic acid as the solvent [13o]. The authors describe in detail the main features of the process and report a detailed study of the reaction mechanism.

Specifically:

1. The combination of a Mn(OAc)2 catalyst and a Co(OAc)2 catalyst (optimum Mn/Co ratio, 1:1) is effective for improving AA selectivity in the liquid-phase oxidation of cyclohexanone with oxygen. With pure oxygen, at atmospheric pressure, a selectivity to AA as high as 77% at total conversion of cyclohexanone is obtained, which is higher than that previously reported in the patent literature. The selectivity to glutaric acid is 12% and that to succinic acid is 2%, to oxoacids 2%, for an overall selectivity to acids of 93%. Selectivity to AA shows a maximum at 700C.

2. The water concentration is less than 20%, and does not affect the reaction. The use of an acid, such as p-toluenesulfonic acid, accelerates the oxidation reaction (Scheme 7.4).

The process scheme proposed by Asahi includes three in-series high-pressure, continuous-stirred, tank reactors (Figure 7.7).

To avoid explosion hazards, air containing less than 10% oxygen (at best, around 5%) was used. Under these conditions, the selectivity to AA was 73%, with a o2, n2

acetic acid recycle

Figure 7.7 Simplified block-diagram oftheAsahi process for the oxidation of KA Oil with nitrogen-diluted air. Source: elaborated from [130].

acetic acid recycle

Figure 7.7 Simplified block-diagram oftheAsahi process for the oxidation of KA Oil with nitrogen-diluted air. Source: elaborated from [130].

cyclohexanone conversion of 99.3%. The first two reactors operate at 70 °C, the third at over 70 °C. More than 90% of the cyclohexanone is consumed in the first two reactors; the remaining cyclohexanone and the intermediates are completely consumed in the third reactor at elevated temperatures. The reaction solution is fed into the AA recovery section, where AA is purified by recrystallization. Three-stage recrystallization is necessary, because the selectivity of AA is lower than that achieved with nitric acid oxidation, which uses a two-stage recrystallization. Water (from the solvent recovery system) is added in the AA recovery section to wash the AA crystals and to dissolve them for recrystallization. After removing the AA, the solution is moved to the solvent recovery section, where acetic acid is recovered by distillation. Part of the concentrated solution in the solvent recovery section is moved to the DBA (dibasic acid) recovery, aimed at removing the by-products and the remaining AA. The other part of the concentrated solution is moved to the catalyst recovery section, in which part of the by-products is also removed. Acetic acid and the catalysts are recycled to the first reactor.

The technology described by Asahi is very similar to that used for the synthesis of terephthalic acid by p-xylene oxidation. The selectivity of AA and the total selectivity of dicarboxylic acids are still lower than those in the nitric acid process [13o].

Comprehensive reviews on the catalytic homolytic oxidative C—C bond cleavage of ketones with either oxygen or air have been published by Bregeault [13l, m]. Several homogeneous systems are active and selective in this reaction; vanadium-containing heteropolyacids with the Keggin structure are amongst the most efficient for the aerial oxidation of cyclohexanone [14]. These compounds are soluble either in aqueous or in organic mediums, depending on their composition. For example, at 70 °C, with 1 atm O2, an acetic acid-water solvent mixture (volume ratio 9:1) and a Keggin-type H7(PMo8V4O40)12H2O catalyst, after a 7-h reaction time 99% conversion of cyclohexanone was achieved, with 50% yield to AA, 19% to glutaric acid and 3% to succinic acid [14a]. A slightly higher yield to AA (54%) was obtained by using a V2-heteropoly-compound, in acetonitrile-methanol solvent, in 24-h reaction time, at 60 °C, with 98% cyclohexanone conversion. The reactivity of the P/Mo/V

polyoxometalates was greatly affected by the composition, for example, the number of V ions incorporated in the Keggin unit and the cation type. At below 90 0C, the mechanism included O2~ transfer from the compound to the organic substrate [14l]. Owing to their high reactivity, their low-temperature redox reversibility and structural flexibility, catalysts based on heteropoly-compounds are potentially promising for several oxidation reactions with O2 under mild reaction conditions [14i].

Fe(III), Ce(IV), Ru(II) and monomeric V species (e.g., [VO{O-i-Pr}3]) also lead to cyclohexanone conversions that are as good as the heteropoly-compounds in this class of reactions, but with lower selectivity to the diacids. However, better performance is obtained when the reaction is catalyzed by Cu(NO3)2 [13l]. At 1100 C and 8 h reaction time, 95% cyclohexanone conversion with 72% yield to AA, 8% to glutaric and 10% to succinic acid were obtained in an acetic acid-water solvent. A similar performance was reported with Mn(OAc)2 in acetic acid-CF3COOH solvent, at 65 0 C after a 3-h reaction time: 99.8% conversion, 75% yield of AA, 9% to glutaric acid and 1% to succinic acid [14j].

High yields to AA were obtained when a Co/Mn cluster complex was used, which was superior to the individual Co and Mn acetates [14k]; at 900 C, and 37 atm pressure, in acetic acid and water solvents, the oxidation of cyclohexanone with air gave complete conversion and 76.3% yield to AA. The authors suggested that 1-hydro-xo-cyclohexen-2-one, the tautomeric form of 1,2-cyclohexandione, is the precursor for the formation of the glutaric and succinic acid by-products. Excellent yields were also reported [14m] for the oxidation of cyclohexanone using Mn(NO3)2 and Co(NO3)2 (molar ratio 1:1) in the presence of oxygen and catalytic quantities of nitric acid at atmospheric pressure. The conversion was 97.5% and the selectivity to AA was 93.4%.

Table 7.2 summarizes some results obtained with various catalyst types for the KA Oil oxidation with oxygen [13-15].

Table 7.2 Summary of catalysts used for the aerial oxidation of cyclohexanol (Ol) and cyclohexanone (One).

Catalyst

Reactant

Solvent, T

Conversion

Selectivity

Reference

(C)

(%)

AA (%)

Pt/charcoal

Ol

Water, 150 °C

100

50

[15b]

Pt/carbon/monolith

One

Water, 140 °C

100

21

[15d]

Mn(OAc)2/Co(OAc)2

Ol/One

Acetic acid, 700 C

100

77

[13o]

Mn(OAc)2

One

Acetic acid/

99.8

75

[13j]

CF3COOH, 65 °C

Mn(NO3)2/Co(NO3)2

One

Acetic acid + catalytic

97.5

93.4

[14m]

HNO3, 40 °C

H7PMo8V4O40

One

Acetic acid/water,

99

51

[14a]

70 °C

HsPMo!0V2O40

One

Acetonitrile/metha-

98

55

[14h]

nol, 60 0C

Co/Mn cluster

One

Acetic acid + water

97.6

86.6 (wt)

[14k]

(MEK), 100 0C

Scheme 7.6 Mechanism ofcyclohexanol oxidation to AAwith HP. Source: elaborated from [15e].

The oxidation of Ol/One has also been investigated using aqueous HP as the oxidant, with either homogeneous or heterogeneous catalysts. For instance, the oxidation of the ketone with HP under homogeneous conditions using acetic acid or t-butanol as the solvent gave about a 50% yield of AA [15f]. H2WO4 afforded the transformation of the ketone with a 91% isolated yield of AA, and that of cyclohexanol with an 87% AA yield, by using a 3.3-molar 30% HP at 900C for 20h reaction time, without the use of solvents [15e]. Remarkably, there was no unproductive decomposition of HP. The mechanism proposed involves the Baeyer-Villiger type oxidation to e-caprolactone, followed by ring opening to yield 6-hydroxyhexanoic acid (Scheme 7.6).

This ketone conversion into dicarboxylic acid by HP is applicable to five- to eight-membered cyclic ketones. The acidic nature of the catalyst is essential, since Na2WO4 did not catalyze the reaction. H2WO4 was the precursor of the true active species, H2[WO(O2)2(OH)2], which is soluble in water. The crystalline product was recovered by filtration (AA precipitated during cooling) followed by drying in air. The aqueous phase of the reaction mixture could be reused with 60% HP to give AA in 71% yield.

Direct Oxidation of Cyclohexane with Air

Compared to the traditional technology, the direct oxidation of cyclohexane with air or oxygen, also called the One Step AA process, should lower the total investment cost. This is due to the following differences:

• elimination of one oxidation step;

• elimination of nitric acid production, handling, recovery, purification and recycle units;

• simplification of the air abatement system by eliminating N2O and NOx emissions;

• simplification of the wastewater treatment by elimination of nitrates.

7.3.2.1 Homogeneous Autoxidation of Cyclohexane Catalyzed by Co, Mn or Cu

Many patents and papers [16-22] describe catalysts and process configurations for the single-step aerobial oxidation of cyclohexane to AA, catalyzed by homogeneous

Co-, Cu-, Mn- or Fe-based complexes, including biomimetic systems. Therefore, the same catalyst type that is used for the oxidation of cyclohexane to KA-Oil also oxidizes this reactant to AA, depending on reaction conditions and the cyclohexane conversion achieved. Indeed, the Asahi Chem Co had already developed the process in the 1940s, using Co acetate catalyst and acetic acid as the solvent, under 30atm of O2 at 90-100 0C [16]. The best selectivity to AA was 75%, the main by-product being glutaric acid, with a cyclohexane conversion of 50-75%. Asahi succeeded in achieving high conversion of cyclohexane through the use of a relatively high concentration of Co(III) acetate combined with acetaldehyde or cyclohexanone which served as a promoter [16].

A review by Schuchardt et al. thoroughly analyses the various catalytic systems reported in the literature up to 2000, both homogeneous and heterogeneous ones, and those that use oxidants other than oxygen [e.g., HP or t-butyl hydroperoxide (t-BuOOH)] [2c]. The mechanism involves the formation of cyclohexanol via the cyclohexyl radical and cyclohexyl hydroperoxide. According to the Haber-Weiss mechanism, cyclohexyl hydroperoxide decomposes into alkoxy and alkyloxy radicals (Section 7.2.1). Cyclohexanol is finally oxidized to cyclohexanone. A similar mechanism may occur at the a-C, affording 1,2-cyclohexanedione, which is finally cleaved to AA. Oxidation of the intermediately formed cyclohexanone to AA then occurs through a mechanism similar to that illustrated in Scheme 7.5.

Many companies have studied the optimization of catalyst composition and process conditions in order to improve the performance of the reaction and the economics ofthe process. In the Gulf process, the reaction is carried out at 90-1000 C, with a Co(III) acetate catalyst and acetic acid as the solvent [17]. The molar selectivity is around 70-75%, for a cyclohexane conversion that can be as high as 80-85%. The high concentration of Co(III) acetate used also favors the direct reaction ofthe cation with cyclohexane, generating the cyclohexyl radical. In fact, in Gulf patents the reaction is reported to occur in a critical amount of Co(III) (25-150 mmoles per mole of cyclohexane). The catalyst is activated during the initial induction period, and water is also added in the initial stage to enhance the selectivity to AA, but the rate of production decreases because the induction period increases.

In Amoco patents [18b], the addition of controlled amounts of water after the initiation ofthe oxidation reaction is claimed to be a tool to obtain a better yield to AA. The best yield achieved was 88% (based on the identifiable compounds) at 98% cyclohexane conversion, with a Co(II) acetate catalyst, an acetic acid solvent, at 95 0C and 70 atm air pressure. It is reported that water, if present during the induction period, depletes the concentration of free radicals; in the absence of water, the yield was remarkably lower. These results are comparable to those attained by the air/nitric acid two-step oxidation of cyclohexane.

In patents issued by Redox Technologies, oxidation is reported to occur more efficiently ifhigh concentrations of cyclohexane are used (AcOH/cyclohexane weight ratio between 50:50 and 15:85, instead of 80:20), and if the conversion ofthe reactant is kept below 30%, instead of more than 70% [19]. A low concentration of catalyst is recommended. An 88% AA selectivity at 21% cyclohexane conversion is obtained in an acetic acid solvent with a Co(OAc)2 catalyst (cyclohexane/Co ratio higher than 150), at 105 ° C and with 14 atm pressure of O2-enriched air flow, for 45 min reaction time; acetaldehyde is added as an initiator [19a]. In Reference [19b], the purification stage of the final mixture is disclosed. The treatment consists of cooling the product mixture to bring about the precipitation of the diacid followed by separation by filtrating the diacid from the two liquid phases, a nonpolar one that is recycled and a polar one that is also recycled after an optional hydrolysis and separation of an additional amount of diacid.

Various bicomponent catalytic systems have been suggested for this reaction, including: (i) Co-Fe [18c, d], which affords 80% selectivity to AA by restricting the conversion to below 30%; (ii) Co-Mn [20a]; (iii) Co-Zr [18e, f]; and (iv) Co-Cr [20b].

Rhone Poulenc (now part of Rhodia) [20c] disclosed a process for recycling the catalyst, including treating the reaction mixture by extraction of the by-products glutaric and succinic acid. For example, with a Mn/Cr catalyst, at 105 °C, 100 bar air pressure, in acetic acid solvent, 11.3% cyclohexane conversion is obtained in 170 min reaction time, with 10.2% selectivity to cyclohexanol and 65.5% to AA [20n]. One innovative aspect of recent Rhodia patents [20d-l] is represented by a lipophilic catalyst that can be separated easily from the reaction mixture and recycled together with the unconverted cyclohexane. At the end of the reaction, water is added to the solution to solubilize the AA. The reactant, the catalyst and the intermediate products cyclohexanol and cyclohexanone form an immiscible phase and are recovered and recycled (Figure 7.8). AA is crystallized from the aqueous solution by cooling. Thus, the innovation is the use of a lipophilic acid, for example, 4-tert-butylbenzoic acid, in place of acetic acid; the stronger the acid, the faster the oxidation rate. At 130 ° C, with an optimal amount of carboxylic acid of 10-12 wt%, 100 wppm of Mn(II), 1 mol% of cyclohexanone (with respect to cyclohexane) as a radical initiator and 100 bar air, the best performance was obtained in terms of AA selectivity (around 33%) and productivity, with a cyclohexane conversion of 10%. The addition of Co(II) as a co-catalyst remarkably improved the oxidation rate, and the selectivity to AA (about 50%). By-products were glutaric acid, succinic acid, cyclohexanone, cyclohexanol, 3-hydroxyadipic acid and formic acid. Under continuous conditions, 10.4% conversion was obtained with 56% selectivity to AA; after recycling of the organic phase, the selectivity increased to 70.6%, also because of the conversion of intermediates into AA. An outstanding value of 95 gAA L_1 h_1 was obtained.

During the 1990s, the Twenty First Century Corporation in a joint venture with RPC Inc. carried out an extensive laboratory experimental program aimed at the development of a new oxidation technology. The core of the technology is the design

Figure 7.8 General diagram illustrating the Rhodia approach for cyclohexane oxidation. Source: elaborated from [20f].

of a new concept for the oxidation reactor. The unit uses oxygen rather than air (used in conventional units), which significantly reduces the size of the air abatement systems. The solution of cyclohexane and cobalt catalyst dissolved in acetic acid is nebulized through a series of nozzles within the reactor. In the reaction chamber there is a very high liquid-gas interface and condensation can be an excellent factor for controlling the reaction temperature. This is achieved by feeding the solution through the bottom nozzles of the reactor at a temperature lower than the solution fed through the upper line. Conversion is controlled by operating on different variables, such as the concentration of catalyst and reactants, the hold-up time of the liquid feed and the size of droplets. The reactor is equipped with a cooling system for the removal of the heat of reaction. The off-gas containing excess of oxygen, nitrogen and VOCs is cooled, compressed and recirculated. A solvent recovery system removes water, recycles the solvent (acetic acid) and cyclohexane. The reaction also yields over-oxidized product and adipic esters. Esters react with water in the presence of a catalyst, converting them into AA and a small amount of cyclohexanone and cyclohexanol. AA is again separated in recovery crystallization and recycled to an AA recovery section. The combination of the single oxidation and hydrolysis boosts yields to levels comparable with conventional technology.

Many patents have been filed by the Twenty First Century Corporation on reactor technology with reactant atomization [21a, b], AA recovery and solvent separation [21c], catalyst handling and recovery [21d] and the control of temperature and pressure [21e-g]. Figure 7.9 shows a simplified block diagram of the process.

Fluor Daniel [21h, i] have performed a conceptual economic analysis of the Twenty First Century Corporation/RPC Inc technology by comparing a theoretical grass roots facility ofcomparable capacity to an existing global scale unit. The results show that the new process could reduce the capital cost by 33% and the operating costs by nearly 23% with respect to conventional technologies.

An application combining the direct oxidation of cyclohexane with air and the use of HP as a finishing oxidant, to complete the oxidation of by-products separated during alkaline washing (see Figure 7.1), has been proposed by Sumitomo [21j]. The wastewater containing hydroxycaproic acid is made to react with 30% HP in the presence of sulfuric acid and a tungsten catalyst, with a 95% yield to AA. This technology allows an increase in the yield and a reduction in the cost of waste

Figure 7.9 Simplified block diagram of the twenty-first century Co/RPC Inc one-step AA process.

treatment. The consumption of HP is limited to the oxidation of a partially oxygenated by-product, and the incidence of its processing costs is reduced.

Despite the moderate yields to AA achieved by the various technologies developed, the One-Step AA process has not yet been implemented at a commercial level, because of the high corrosivity of acetic acid and the high energy demand for solvent recycle and product workup. Moreover, the per-pass conversion and selectivity is generally lower than the two-step process from cyclohexane. Finally, traces of acetic acid are left in the product even after repeated crystallization steps, which lower the quality of the product and limit its use in polyamide synthesis.

Disaster Preparedness Kit

Disaster Preparedness Kit

Don't get caught out in a disaster, be prepared for the unlikely. You just never know when disaster will strike, but if it does you will be more then prepared for it with the information this ebook gives you.

Get My Free Ebook


Post a comment