Selective Oxidation Examples of Directions to Improve Sustainability

Some main directions towards improving sustainability that should be highlighted are: (i) the use of new and clear oxidants, which is exemplified by the use of H2O2, (ii) the use of new feedstocks, which can be illustrated by the substitution of alkenes with alkanes, and (iii) the process development, which can be discussed by showing the developments arising from using new reactor opportunities and an integrated reactor/catalyst design, as well as by substituting air with oxygen. Together with the development of new catalysts, which open up new opportunities to develop innovative routes of transformation, these directions allow a reduction of the environmental impact and use of resources (raw materials, energy) and often also allow a significant process simplification. Some examples are discussed below.

H2O2 as a Clean Oxidant In the area of liquid phase oxidation, perhaps the more relevant development is the substitution of either stoichiometric reagents (e.g., CrVI, but also several other reagents such as permanganate, MnO2, and so on) or of organic peroxides with a clean reagent such as H2O2 in combination with a catalyst able to give selective oxidation such as titanium-silicalite (TS-1). This aspect is discussed in detail in the case of the new processes for the synthesis of propene oxide and phenol, and thus will not be discussed here. A further important development regards the new synthesis of caprolactam, which is discussed above (Figure 2.32). However, the same reaction of ammoxidation catalyzed by TS-1 in the presence of ammonia and H2O2 could be usefully applied in the synthesis of a large variety of substrates of interest for fine chemistry, as exemplified in Figure 2.57.

H2O2/TS-1 also finds application in the hydroxylation of aromatics. A relevant example is the Rhodia process for the manufacture of the flavor ingredient vanillin [229]. The process involves four steps, all performed with a heterogeneous catalyst, starting from phenol (Figure 2.58). Overall, one equivalent of phenol, H2O2, CH3OH,

Selectivity to Oxime 98 (%Mol.)

Figure 2.57 Ammoximation of carbonyl groups with TS-1. Source: adapted from Corma [320].

Selectivity to Oxime 98 (%Mol.)

Figure 2.57 Ammoximation of carbonyl groups with TS-1. Source: adapted from Corma [320].


Figure 2.58 Rhodia vanillin process. Source: adapted from Ratton [229].


Figure 2.58 Rhodia vanillin process. Source: adapted from Ratton [229].

formaldehyde and O2 are converted into one equivalent of vanillin and three equivalents of water.

Titanium-silicalite was the first example of a wider class of materials based on zeolites and micro- or mesoporous materials in which Si and/or Al atoms have been substituted by transition metals. A limitation of these materials was often the easy leaching of the transition metal, making them inapplicable on an industrial scale. A very successful example was, instead, Sn-Beta, which is characterized by a bifunctional active site that involves a Lewis acid tin center and an adjacent oxygen atom capable of accepting hydrogen bonding from water or hydrogen peroxide [321]. In the area of selective oxidation it can be applied to the Baeyer-Villiger oxidation of ketones and aldehydes [322] with aqueous H2O2 (Figure 2.59).

Another very interesting example from the same group is the use of Sn-Beta for the synthesis of melonal (2,6-dimethyl-5-hepten-1-al), a fragrance that is produced industrially by a Darzens reaction from 6-methyl-5-hepten-2-one, with ethyl chlor-oacetate as reagent. A novel halogen-free synthesis involves the chemoselective

Conventional synthesis Using Sn-Beta

Figure 2.59 Baeyer-Villiger oxidation of ketones using Sb-Beta. Source: adapted from Corma [320].

Conventional synthesis Using Sn-Beta

Figure 2.59 Baeyer-Villiger oxidation of ketones using Sb-Beta. Source: adapted from Corma [320].

oxidation of citral (3,7-dimethyl-6-octen-1-al), a common compound in the fragrance industry, with H2O2 and Sn-Beta or Sn-MCM-41 as catalysts [323] (Figure 2.60). Aluminium Bronsted acid sites and zirconium or titanium Lewis acid sites are less efficient and selective than Sn in this Lewis acid site catalyzed reaction.

Notably, there are other types of catalysts active in clean organic reactions using H2O2 [324]. Among the homogeneous catalysts, the use of tungsten-based catalysts (phosphotungstate of the Keggin type [325], silicadecatungstate [326] and

Figure 2.60 Baeyer-Villiger chemoselective oxidation of melonal using Sb-Beta. Source: adapted from Corma [320].

Chemoselective Baeyer-Villiger oxidation with Sn-Beta

Figure 2.60 Baeyer-Villiger chemoselective oxidation of melonal using Sb-Beta. Source: adapted from Corma [320].

R3 R1

R2 [y -SiW10O34(H2O)2](Bu4N)4 (8 jamol) CH3CN (6 mL), 32 °C

Figure 2.61 Epoxidation of alkenes promoted by silicadecatungstate and H2O2. Source: adapted from Goti and Cardona [324].

Na2WO4 [327]) for the selective epoxidation of a large variety of alkenes is notable. Figure 2.61 gives an example with the silicadecatungstate catalyst developed by Mizuno et al. [326].

Some of these W-based catalysts could be transformed into insoluble salts; for example, the phosphotungstate-pyridinium salt in a toluene-tributyl phosphate (4:3) solvent mixture [328]. In this solvent the pre-catalyst is insoluble, but upon reaction with H2O2 it gives the catalytically active W-peroxo complex, which is soluble. The catalytic action is then performed under homogeneous conditions and, at the end of the reaction, H2O2 being completely consumed, the precatalyst precipitates and can be easily filtered off and recovered. The method is smart, but from an industrial point of view the separation and recovery is costly.

Another interesting alternative in the epoxidation of alkenes is the use of methyltrioxorhenium (MTO), originally developed by Herrmann et al. [329]. MTO activates hydrogen peroxide by forming a mono-peroxo complex that undergoes further reaction to yield a bis-peroxorhenium complex. Both complexes are active as oxygen transfer species. Aproblem is the formation of 1,2-diol via ring opening of the epoxide. The addition of urea limits this problem.

Novel Pathways and Reactants This is a very broad area. We will thus restrict discussion to few examples. The first regards the important reaction of phenol synthesis and the possibility to realize it in one step directly from benzene using molecular oxygen as the oxidant. Various aspects of direct phenol synthesis from benzene are discussed in Chapter 13. We highlight here only recent results that exemplify how starting from the previously cited activity of Re complexes in the epoxidation in homogeneous phase could lead to investigation of the behavior of Re complexes when inserted into the channels of zeolites (ZSM-5) and in gas-phase selective oxidations. This has opened a new unexpected direction.

Bai et al. [330] found a remarkable selectivity (88% in the steady-state reaction and 94% in the pulse reaction) in the direct synthesis of phenol from benzene with molecular oxygen over a Re/zeolite catalyst prepared by chemical vapor deposition (Figure 3.62). However, stable performances could be obtained only by continuous feeding of relatively high concentrations of NH3 (around 30%), which is necessary to stabilize the active complex containing interstitial N atoms (see the model of the complex in

Figure 2.62 Direct synthesis of phenol from benzene using molecularoxygen on rhenium complexes in ZSM-5. Source: adapted from Iwasawa et al. [330].

Figure 2.62). This is the main drawback for industrial development, because part of the ammonia is also side converted. Nevertheless, this reaction and catalyst exemplify the novel possibilities opened to develop sustainable processes of selective oxidation.

A second example concerns the use of alkanes instead of alkenes to both use alternative feedstocks and reduce the environmental impact. An interesting example is the oxidation of isobutane to methacrylic acid [331].

The methyl ester of methacrylic acid (CH2=C(CH3)—COOH) is used to produce vinyl polymers used as cast sheets, molding and extrusion powders and coatings, besides being used in various copolymers. Poly(methyl methacrylate) production currently stands at over 2.8 million tons per year on a worldwide scale.

The traditional acetone cyanohydrin (ACH) process is the most widely used in Europe and North America, while other processes are more often used in Asia. In the ACH process (Figure 2.63), acetone and hydrogen cyanide react to yield acetone cyanohydrin; the latter is then reacted with an excess of concentrated sulfuric acid to form methacrylamide sulfate. In a later stage, methacrylamide is treated with excess aqueous methanol; the amide is hydrolyzed and esterified, with formation of a mixture of methyl methacrylate and methacrylic acid. The ACH process offers economical advantages, especially in Europe, where large plants are in use - most ofthem have been in operation for decades. The process also suffers from drawbacks that have been the driving forces for the development of alternative technologies.

Specifically, the process makes use of HCN, a very toxic reactant. Difficulties in its acquisition, though, can be met; in fact, HCN is a by-product of propylene ammoxidation. Integration of acrylonitrile and MMA products requires the balance

Figure 2.63 Acetone cyanohydrin (a) and alternative routes (b) in the synthesis of methacrylic acid.

of the two processes. Alternatively, HCN can be produced on purpose, but this is feasible only for large production capacities. The second major drawback of the process is the disposal of ammonium bisulfate, the co-product of the process. Additional costs are necessary for its recovery or pyrolysis.

The ACH process has been improved by Mitsubishi Gas [332]. Acetone cyanohydrin is first hydrolyzed to 2-hydroxyisobutylamide with a MnO2 catalyst; the amide is then reacted with methyl formate to produce the methyl ester of 2-hydroxyisobutyric acid, with co-production of formamide (this reaction is catalyzed by sodium meth-oxide). The ester is finally dehydrated with an Na-Y zeolite to methyl methacrylate. Formamide is converted into cyanhydric acid, which is used to produce acetone cyanohydrin by reaction with acetone. The process is elegant, since it avoids the co-production of ammonium bisulfate, and no net income of HCN is present. However, there are many synthesis steps, and a high energy consumption.

Other technologies, already commercially applied or under development, are summarized in Figure 2.63b. Alternative routes of synthesis include (i) ethene hydroformylation to propionaldehyde, which then forms methacrolein by condensation with formaldehyde; methacrolein is then oxidized to methacrylic acid (BASF process); (ii) isobuthyraldehyde conversion into isobutyric acid and then oxidative dehydrogenation to methacrylic acid (Mitsubishi Kasei/Asahi process); and (iii) oxidation of tert-butyl alcohol to methacrolein followed by oxidation to methacrylic acid and esterification.

An attractive new route under development is direct (one-step) gas-phase isobutane conversion into methacrylic acid, because of the (i) low cost of the raw material, (ii) simplicity of the one-step process, (iii) very low environmental impact and (iv) absence of inorganic co-products. Several patents claiming the use of polyoxome-talates (POMs, Figure 2.64) as heterogeneous catalysts for this reaction started to appear in the 1980s and 1990s [331]. Rohm and Haas was the first (1981) to claim the use of P/Mo/(Sb) mixed oxides for isobutane oxidation [333]. Later, several patents (issued to Asahi Kasei, Sumitomo Chem, Mitsubishi Rayon, and others) described the use of modified Keggin-type POMs as catalysts. Specifically, most attention has been focused on the possibility of improving the conversion of isobutane and the selectivity to methacrylic acid by developing POMs that contain specific transition metal cations. The major problem is the stability of this catalyst, because the reaction temperatures of operations are close to those necessary for the activation of the alkane. This problem is exacerbated by the fact that the control of the reaction temperature is difficult, due the very high heat of reaction that develops.

A peculiarity of the processes described in the patents is that all of them use isobutane-rich conditions, with isobutane-to-dioxygen molar ratios between 2 (for processes that include a relatively low concentration of inert components) and 0.8, and so closer to the stoichiometric value 0.5 (for those processes where a large amount of inert components is present). Low isobutane conversions are achieved in all cases, and recirculation of unconverted isobutane becomes compulsory.

Figure 2.64 Structure of polyoxometalate catalysts used for the conversion of isobutane into methacrylic acid.

Keggin structure, XM12O40n" Dawson structure, X2M18062n

Figure 2.64 Structure of polyoxometalate catalysts used for the conversion of isobutane into methacrylic acid.

In all cases steam is present as the main ballast. The role of steam is to decrease the concentration of isobutane and oxygen in the recycle loop and thus keep the reactant mixture outside the flammability region. Water can be easily separated from the other components of the effluent stream, playing also a positive role in the catalytic performance of POMs. It is also possible that the presence of water favors the surface reconstruction of the Keggin structure, which decomposes during the reaction at high temperature, and also promotes desorption of methacrylic acid, saving it from unselective consecutive reactions.

Under the reaction conditions described in the patents, methacrolein is always present in non-negligible amounts, and therefore a commercial process necessitates an economical method for recycling it. Figure 2.65 shows a simplified flow-sheet of the Sumitomo process. CO2 is maintained in the recycle loop to act as a ballast component; the desired concentration of CO2 is obtained by combustion of CO, while excess CO2 is separated. Methacrolein is separated and recycled to the oxidation reactor. An overall recycle yield of 52% to methacrylic acid is reported, with a recycle conversion of 96% and a per-pass isobutane conversion of 10%. The heat of reaction produced, mainly deriving from the combustion reaction, is recovered as steam. However, commercialization of this process is still hindered by the actual productivity (about 0.7 mmolh-1 gcat_1), which is still too low [334]. Note that the productivity is limited by the oxygen conversion, the maximum concentration of which is dictated by the flammability limits, and by temperature, since the POM decomposes above 380 °C. Therefore, a possible development is to use microreactor technology, where, due to the high wall-to-volume ratio, operations inside the explosion limits are possible and also where the heat of reaction can be removed efficiently, thus improving catalyst stability.

Recycle gas

Figure 2.65 Simplified flow-sheet for isobutane oxidation to methacrylic acid proposed by Sumitomo. Source: adapted from Cavani et al. [331].

Figure 2.65 Simplified flow-sheet for isobutane oxidation to methacrylic acid proposed by Sumitomo. Source: adapted from Cavani et al. [331].

A recent interesting example of the use of alkane feedstocks to develop more sustainable processes is the direct conversion of ethane into acetic acid developed by Sabic. Acetic acid is the raw material for many key petrochemical intermediates and products, including vinyl acetate monomer (VAM), purified terephthalic acid (PTA), acetate esters, cellulose acetate, acetic anhydride, monochloroacetic acid (MCA), and so on. Acetic acid is produced commercially from several feedstocks and by several different technologies.

Methanol carbonylation technology using syngas accounts for over 60% of global capacity. This share is growing because it is the preferred technology for most new plants. Direct ethane oxidation to acetic acid has been an area of great interest to many chemical companies. Among the many patent holders, the most active players in this area are Hoechst Research and Technology Deutschland GmbH & Company, Saudi Basic Industries Corporation, Mitsubishi Chemical Corp., and BP Chemicals Ltd. In 2006 Sabic began operations for the direct conversion of ethane into acetic acid with an initial 34 000 metric ton per year capacity plant.

The Sabic acetic acid technology is characterized by a novel catalyst (a Mo-V-Nb mixed oxide [335]) and a novel oxidation reactor design, which is different from the conventional methanol-based technology. In the process ethane is mixed with oxygen and compressed, then passed over the catalyst to produce acetic acid and some ethylene, which is then separated and purified for use as a feedstock in other associated plants.

As cited in Chapter 1, the first example of commercial process using an alkane as feedstock, in substitution of the older process starting from benzene, was the synthesis of maleic anhydride from n-butane. Figure 2.66 briefly recalls the reaction scheme on the model surface of the catalyst (vanadyl pyrophosphate) to evidence the

Figure 2.66 Multifunctionality of solid catalysts: synthesis of maleic anhydride from n-butane over a (VO)2P2O7 catalyst.

(VO)2P207: n-butane oxidation to maleic anhydride

Figure 2.66 Multifunctionality of solid catalysts: synthesis of maleic anhydride from n-butane over a (VO)2P2O7 catalyst.

concept that this complex reaction (a 19-electron oxidation that involves the abstraction of eight hydrogen atoms and the insertion of three oxygen molecules in the hydrocarbon substrate) can occur with high selectivity (over 80%) on the catalyst surface without desorption of any intermediate [336]. This is a good example of a sustainable process, not only because it has substituted the older and less sustainable process from benzene but also because it is a prototype of the possibility of realizing very complex reactions in a single stage, which is one of the main objectives of sustainable chemical processes.

Role of Reactor and Process Design Optimization of catalytic performances, in terms of reactant conversion, yield, productivity and selectivity to the desired product, is not only related to a thorough knowledge of the nature of the catalyst and the interactions between reacting components and surface active phases, the reaction mechanism, thermodynamics and kinetics but also to the development and use of a suitable reactor configuration, where all the above-mentioned features can be successfully exploited.

Industrial reactors used in the petrochemical industry for exothermic reactions, with a few exceptions, are either fixed-beds (adiabatic or non-adiabatic) or fluidized-beds when the heat developed is too high to be removed in a fixed-bed reactor. In the last few decades, interest has been mainly directed towards the control of these reactors, which is strictly related to an understanding of the complex phenomena that occur at the interface between the different phases present in the reaction environment, and of the heat and mass transfer influence on the reaction kinetics.

Substantial improvements in the performance of several processes of hydrocarbon selective oxidation can be achieved solely by developing new reactor configurations. An important step in this direction is exemplified by the circulating fluidized bed reactor, which over the years has been proposed for use in several selective oxidation reactions and has, finally, found application in n-butane selective oxidation to maleic anhydride. Although production at the plant (built in Spain) was later stopped, because it was uneconomic, it remains an interesting example that may find application in other reactions.

The principle exploited in this kind of configuration is the decoupling of the classical redox mechanism (which operates in the selective oxidation of most hydrocarbons) into two separate steps, each of which can be optimized, thus improving the overall performance. Besides better control of overall reaction exo-thermicity, further advantages of this operation are: (i) higher selectivity, which is usually achieved, because the hydrocarbon never comes into contact with the molecular oxygen, and (ii) elimination of hazards associated with possible formation of flammable gas mixtures. This reactor configuration also offers the opportunity to work with catalysts in a partially reduced state, which is not possible when oxygen is co-fed, and thus opens a new area of investigation.

Another example is monolithic-type reactors, which have found their main application in the field of combustion. A monolith bed allows better autothermic operations with a minimal pressure-drop. This concept was used to improve performances in commercial methanol into formaldehyde conversion by adding a final monolithic reactor stage. Cross-flow monoliths have been applied to improve performances in highly exothermic oxidation reactions. Examples are the oxidation of ammonia to nitrogen with a Co/a-Al2O3 catalyst, the oxidation of SO2 with a Pt catalyst, the oxidation ofpolychlorinated biphenyls, the oxidative dehydrogenation of light alkanes and the partial oxidation of methane [241-243]. Layers of gauze, stacked one over the other to form a bed several millimeters deep are used in important industrial applications of oxidation reactions such as (i) the oxidation of ammonia to NO and (ii) methane ammoxidation to HCN. Using these reactors, operations at extremely short contact times are possible, which allow significant improvements in selectivity in several oxidation reactions [337] such as (i) methane selective catalytic oxidation to syngas and (ii) alkane oxidative dehydrogenation to alkenes. In all these cases, the new reactor option implies the design of new oxidation catalysts that can operate in the new reactor configuration as well as being able to take advantage of the opportunities offered by the new reactor design.

Membrane technology offers interesting potential advantages in allowing better control of the reaction kinetics [338]. Further aspects of this are discussed in Chapter. The catalytically active component can be either deposited on the membrane or, simply, the catalyst bed is contained in a reactor having membrane walls. Membrane reactors can be used with the aim of distributing oxygen along the catalyst bed. A gradual feeding of oxygen can (i) maintain the optimal O2/hydrocarbon ratio for selectivity, (ii) limit the formation of hot spots and (iii) avoid the occurrence of runaway phenomena. Moreover, a controlled distribution of oxygen may keep the catalyst at the desired average oxidation level.

Conventional fixed-bed reactor operations can also be improved through better integration of catalyst and reactor design. An important commercial example is the new o-xylene oxidation to phthalic anhydride process introduced by Lonza-Alusuisse, which uses a dual-bed configuration. The catalytic beds are arranged in two parts, each containing a catalyst, the formulation of which has been optimized, that is, less active for the first part of the reactor (close to the reactor inlet), where the reaction rate is the highest, and more active for the final part of the reactor, that is, for finishing the reaction. In each section the catalyst is essentially made of a-alumina or steatite pellets, coated with a thin film of V2O5/TiO2-based catalysts. The activity of the catalyst in each section is optimized by controlling the vanadia content, as well as by the addition of dopants (Cs and P, principally). In this way the hot spot temperature is considerably lowered, and the hottest region becomes spread over a longer reactor length, with a considerable improvement in selectivity to the partial oxidation product, as well as longer catalyst life.

Alternative reactor options are also offered by:

• Periodic flow reversal inducing forced unsteady-state conditions [339]. The flow to the reactor is continuously reversed before the steady state is attained. A dual hot-spot temperature profile, characterized by a considerably lower temperature than in the single hot spot that would develop in the traditional flow configuration, forms in exothermic oxidation reactions. An increase in selectivity and better reactor control (lower risk ofrunaway) is possible over fixed-bed reactor operations, but compared to the analogous advantages possible with multi-bed reactors (see case of o-xylene), which have lower reactor and operation costs as well as fewer safety problems, the periodic flow reversal option will be applied to specific cases only.

• Decoupling of the exothermal reaction into two steps. In this reactor configuration, the overall heat of reaction is subdivided into two less exothermic steps. In the first the hydrocarbon is brought into contact with the catalyst, while in the second step the reduced catalyst is reoxidized by contact with gaseous oxygen. Besides better control of the overall reaction exothermicity, further advantages of this operation are: (i) higher selectivity, which is usually achieved, because the hydrocarbon is never brought into contact with molecular oxygen and (ii) elimination of the hazards associated with the possible formation of flammable gas mixtures. This configuration can be carried out either (i) in two (or more) parallel reactors, where one (or more) reactor is at the reaction stage, and the other is at the catalyst reoxidation stage or (ii) in a circulating-bed reactor, where the catalyst is continuously transported from the reaction vessel to the regeneration vessel and vice versa.

All these reactor options allow not only the development of new processes, and the holding of proprietary technologies, but also an improvement in productivity (thus saving energy per ton of product) and often a reduction in the emissions of waste, greenhouse gases and improved safety of operations, for example, improved process sustainability.

A relevant issue in industrial selective oxidation is the substitution of air-based to O2-based processes. Air has been the preferred oxidant for years, but several oxidation processes, both in the liquid and in the gas phase, have been modified over the years to allow the use of pure oxygen [241-243, 340]. These changes have been driven by improvements in productivity and yield, while more recently revamping or modifications aimed at pure oxygen use have been undertaken due to environmental constraints.

The use ofoxygen instead ofair implies that the same partial pressure can be used with a much lower total pressure than with air, thus making it feasible to possibly reduce the total pressure, with obvious energy advantages. Moreover, an increase in the reaction rate makes it possible to reach the same productivity while lowering the reaction temperature, with possible benefits from the selectivity point of view when several products are formed in the reaction. Lower nitrogen contents, a ballast with very poor thermal conduction properties, also allow better control of the temperature profile in the presence of strongly exothermic reactions.

An even more important benefit originates from the considerable decrease in polluting emissions released into the atmosphere, as a consequence of the fact that spent gases can be recycled when oxygen is used in place of air. Less waste gas is produced, with energy savings during incineration. In addition, the heating value of the stream is much higher than for the air-based process (the concentration of nitrogen is lower, while that of hydrocarbons and carbon oxides is expected to be higher). Therefore, the purge stream, instead of being treated, can be used as a fuel to incinerate other wastes.

The following chemical processes make use of oxygen-enriched air or of oxygen, as an alternative to air [243]:

• partial oxidation of oil fractions and coke to synthesis gas;

• oxidation of methanol to formaldehyde (either air or oxygen-enriched air);

• oxidation of ethene to ethene oxide (either air or oxygen; new plants oxygen);

• oxychlorination of ethene to 1,2-dichloroethane (either air or oxygen; new plants mostly oxygen);

• oxidative acetoxylation of ethene to vinyl acetate (oxygen);

• oxidation of n-butane to acetic acid (either air or oxygen);

• oxidation of ethene to acetaldehyde (either air or oxygen);

• oxidation of acetaldehyde to acetic anhydride (either air or oxygen);

• ammoxidation of propene to acrylonitrile (oxygen-enriched air);

• oxidation of cyclohexane to cyclohexanone (either air or oxygen);

• oxidation of isobutane to t-butyl hydroperoxide (the latter is used for propene epoxidation) (oxygen).

Some additional aspects are also discussed in Chapter 5, which is dedicated to accounting for the sustainability of chemical processes.

Guide to Alternative Fuels

Guide to Alternative Fuels

Your Alternative Fuel Solution for Saving Money, Reducing Oil Dependency, and Helping the Planet. Ethanol is an alternative to gasoline. The use of ethanol has been demonstrated to reduce greenhouse emissions slightly as compared to gasoline. Through this ebook, you are going to learn what you will need to know why choosing an alternative fuel may benefit you and your future.

Get My Free Ebook

Post a comment