One of the grand challenges of the twenty-first century is the implementation of the transition from an unsustainable economy based on fossil resources–oil, coal, and natural gas—to a sustainable, carbon-neutral economy based on the use of renewable biomass. This switch to a so-called bio-based economy is urgently required in order to mitigate global warming caused by increasing carbon dioxide emissions to the atmosphere. First generation (1G) renewable raw materials, exemplified by corn starch, sugar cane, and sugar beet, are not perceived as sustainable options in the long term as their utilization involves, directly, or indirectly, competition with food production. In contrast, the use of second generation (2G) renewable biomass, in the form of waste polysaccharides, such as lignocellulose (Liguori and Faraco, 2016; Zhang et al., 2017) and pectin, from agricultural and forestry residues and food supply chain waste (Dahiya et al., 2018), is perceived as a sustainable long term option for producing biofuels and commodity chemicals (Sheldon, 2014, 2016, 2018; Horváth et al., 2017). Looking further afield, third generation (3G) aquatic biomass, such as micro- and macro-algae and cyanobacteria, has additional advantages (John et al., 2011; Al Abdallah et al., 2016; Shuba and Kifle, 2018). For example, there is no requirement for arable land and fresh water for their production and they have much higher growth rates than terrestrial plants. On the other hand, there are substantial technical problems associated with their production and conversion which, in the short term, represent a significant hurdle to be overcome for commercial viability.

In a petrochemical refinery the basic chemicals are lower olefins (ethylene, propylene, and butenes) and aromatics (BTX: benzene, toluene, and xylenes), together with carbon monoxide and hydrogen (syn gas). The hydrocarbons are gases or hydrophobic liquids. They are converted with petrochemical catalytic technologies, particularly catalytic oxidation with dioxygen, to a variety of commodity chemicals, usually in solvent-free systems. In contrast, the basic chemicals in a bio-refinery will be C₆ and C₅ sugars produced by hydrolysis of polysaccharide feedstocks, and/or syn gas produced by their gasification. The carbohydrates are hydrophilic, water soluble solids. Several scenarios can be envisaged for further conversion to commodity chemicals (Figure 1):

A key reaction in organic synthesis is the oxidation of primary and secondary alcohols to give the corresponding aldehydes or carboxylic acids and ketones, respectively. Traditionally these transformations were performed with stoichiometric quantities of inorganic oxidants, notably chromium (VI) compounds such as the Jones reagent (CrO₃ and sulfuric acid). However, such procedures are not atom efficient and lead to the formation of copious amounts of toxic, chromium-containing waste, i.e., high E factors and problematic waste disposal issues. Consequently, in the last two decades such methods have been increasingly replaced by atom efficient catalytic alternatives involving dioxygen or hydrogen peroxide as the terminal oxidant.

Interestingly, the history of catalytic oxidations of carbohydrates (Arts et al., 1997) predates the oxidations of lower olefins and aromatics that form the basis of the petrochemical industry. The aerobic oxidation of mannose over a platinum black catalyst, for example, dates from 1861 (von Gorup-Besanez, 1861) and many supported noble metal catalyzed aerobic oxidations of carbohydrates were developed in the first half of the last century. At the turn of the century, we developed an aqueous biphasic system for the aerobic oxidation of primary and secondary alcohols to the corresponding aldehydes and ketones, respectively, in a solvent free system using a water-soluble palladium complex of bathophenanthroline (ten Brink et al., 2000). This system could also be effective in the aerobic oxidation of water soluble alcohols, including carbohydrates. Indeed, there are many examples of the aerobic oxidation of alcohols catalyzed by precious metals such as palladium, platinum and gold (Stahl, 2004; Parmaggiani and Cardona, 2012). However, in the context of the conversion of carbohydrates to large volume, low-priced commodity chemicals, precious metals such as palladium have the disadvantage that the future availability of these scarce, “endangered elements” at cost-effective prices is rather unpredictable. Indeed, in contrast to most materials, their price tends to increase with increasing usage. Another disadvantage of noble metal catalyzed oxidations is their functional group intolerance. First row, more earth abundant metals tend to be more functional group tolerant.

Consequently, alternative methods have been developed that use “earth abundant” metals, such as copper and iron, as catalysts. One method with broad applications in the selective oxidation of primary alcohols to aldehydes, even in the presence of secondary alcohols, involves the combination of a Cu(II)-bipyridine (Cu-bpy) complex with a base, such as potassium hydroxide, a stable nitroxyl radical, exemplified by 2,2,6,6-tetramethyl-1-piperidine-N-oxyl (TEMPO) and its derivatives, at ambient temperature with air in aqueous acetonitrile (Figure 2; Gamez et al., 2003; Sheldon and Arends, 2004; Marais and Swarts, 2019). The generally accepted mechanism involves as the key, rate determining step, abstraction of a hydrogen atom from an alkoxide ligand by a coordinated nitroxyl radical analogous to that involved in the aerobic oxidation of primary alcohols catalyzed by the copper-dependent oxidase, galactose oxidase (Dijksman et al., 2003). An improved procedure, using Cu(I) salts with TEMPO and bipy in combination with N-methylimidazole as a base in acetonitrile as solvent was subsequently described by Stahl and coworkers (Hoover and Stahl, 2011). Furthermore, extensive mechanistic studies confirmed the copper-centered galactose oxidase-like mechanism for these systems (Geiβlmeir et al., 2005; Hoover et al., 2013).

The lack of reactivity of secondary alcohols was attributed to steric hindrance in the abstraction of an α-hydrogen atom from a coordinated alkoxide by a coordinated TEMPO ligand. Consequently, the use of sterically less hindered nitroxyl radicals such as AZADO and ABNO, respectively, in combination with Cu(I) complexes, were developed by the groups of Iwabuchi (Shibuya et al., 2006, 2011; Iwabuchi, 2013) and Stahl (Steves and Stahl, 2013), for the aerobic oxidation of secondary alcohols (Figure 3), including sterically demanding alcohols such as menthol and a variety of unprotected amino alcohols (Sasano et al., 2014).

More recently, Ma and coworkers reported the use of an Fe(III)/4-hydroxyTEMPO/NaCl combination as a catalyst for the aerobic oxidation of both primary and secondary alcohols (Jiang et al., 2016, 2019). Other metals, including manganese, cobalt and vanadium, have also been used in combination with nitroxyl radicals (Cao et al., 2014) Transition metal free nitroxyl systems have also been described, usually involving nitrogen dioxide as the active co-catalyst. The commercially most attractive source of the NO₂ cocatalyst is nitric acid (Kuang et al., 2010). For example, nitric acid or NaNO₂ or a mixture of both was used, in combination with ABNO or keto-ABNO, for the selective aerobic oxidation of secondary alcohols (Figure 4; Lauber and Stahl, 2013). Interestingly, the combination of nitric acid and NaNO₂ catalyzes the aerobic oxidation of alcohols even in the absence of a stable nitroxyl radical. The reaction involves an alkyl nitrite intermediate which decomposes to the carbonyl compound and HNO which is reoxidized by dioxygen (Aellig et al., 2011). Unfortunately, the greenhouse gas, nitrous oxide (N₂O), can be irreversibly formed as a byproduct.

Although these various nitroxyl radical-based catalysts have been widely used in the aerobic oxidations of alcohols they have generally involved relatively simple primary and secondary alcohols in organic solvents, sometimes mixed with water. This was done with the development of green syntheses of, for example, active pharmaceutical ingredients (APIs) and flavors and fragrances, in mind. However, in a bio-based economy it is of interest to use these methodologies in the selective oxidation of renewable carbohydrates or key alcohols, diols, and polyols derived from them, and this involves in many cases aqueous solutions of solid substrates.

Water has both advantages and limitations as a solvent for aerobic oxidations. For example, oxidations with oxygen are much safer as there is no formation of explosive mixtures of oxygen with volatile organic solvents in the gas phase. In a typical process oxygen is supplied by bubbling air through the solution. However, the transfer of oxygen from the gas to the liquid phase is notoriously slow owing to its low solubility in water under typical operating conditions (0.268 mM at 25°C and 1 bar air) which limits the maximum space time yield to 200 mmol/L/h (Pedersen et al., 2015). This problem was alleviated in the Cu/TEMPO system by using air-microbubble techniques to facilitate gas absorption into the liquid phase (Mase et al., 2011). Alternatively, the rates of enzymatic aerobic oxidations were increased by a factor of 100 in continuous flow operation compared to the conventional batch operation (Chapman et al., 2018; Hone and Kappe, 2019). Another disadvantage of water as a solvent is that its relatively high heat capacity, compared to volatile organic solvents, translates to high energy costs for its removal by distillation.

A pertinent example, from the viewpoint of the bio-based economy, is the use of acetylamino-TEMPO (AA-TEMPO) together with nitric acid as the cocatalyst for the aerobic oxidation of primary and secondary alcohols to the corresponding aldehydes and ketones, respectively, in acetic acid or water as the solvent (Dingerdissen et al., 2011). The method was particularly useful for the oxidation of the key biomass-derived diol, isosorbide, to the corresponding diketone (Figure 5; Klasovsky et al., 2015). This is particularly surprising because of the low reactivity of the shielded endo OH group in isosorbide.

Isosorbide is a commercially interesting platform chemical produced by hydrogenation of glucose to sorbitol followed by dehydration (Figure 6). It has interesting features as an industrial monomer based on its rigidity, chirality, and non-toxicity (Fenouillot et al., 2010). For example, reaction with dicarboxylic acids (or esters) affords polyesters. Alternatively, oxidation to the corresponding diketones, followed by reductive amination, affords the corresponding bis-primary amine (Figure 5) which can be converted to polyamides by reaction with dicarboxylic acids (esters).

As was noted elsewhere, biocatalysis is green and sustainable (Hollmann et al., 2011; Sheldon and Woodley, 2018), conforming to 10 of the 12 principles of green chemistry, and the catalyst is non-toxic, biocompatible and biodegradable. Moreover, enzymatic reactions are generally performed in water, a particularly suitable medium for conversions of polysaccharides in a bio-based economy. Furthermore, carbohydrates tend to be ideal substrates and generally have a stabilizing effect on enzymes. In contrast with precious metal catalysts, the long term availability and price stability of enzymes is assured since they are produced from inexpensive, readily available biomass. Moreover, industrial scale oxidations employing precious metal catalysts often involve a costly purification step to remove traces of the metal in the product. In contrast, no costly removal of trace amounts of enzymes are needed in enzymatic oxidations and any enzyme ending up in aqueous effluent undergoes facile biodegradation.

Acid catalyzed dehydration of pentoses and hexoses produces furfural and 5-hydroxymethylfurfural (HMF) (Tong et al., 2010), respectively. Furfural is an important commodity chemical (Lange et al., 2012) and HMF has the potential to become one (van Putten et al., 2013; Kucherov et al., 2018). Initial isomerization of D-glucose to D-fructose is followed by acid catalyzed dehydration (Figure 13) but in yields that are not conducive to commercial viability owing to the limited stability of HMF in the acidic reaction medium (Wang et al., 2017). However, according to a recent report HMF can be obtained in 95% yield by conducting the reaction with D-fructose under continuous flow conditions (Galaverna et al., 2018).

Polyethylene furandicarboxylate (PEF), which was developed by Avantium¹ is produced from furan-2,5-dicarboxylic acid (FDCA) and ethylene glycol It is seen as a renewable alternative for fossil-based polyethylene terephthalate (PET). In addition to reducing CO₂ emissions, PEF has superior mechanical, thermal, and gas barrier properties to PET. The key raw material, FDCA, can be produced in excellent yields by aerobic oxidation of HMF using supported precious metal catalysts (Liu et al., 2015; Zhang and Deng, 2015; Zheng et al., 2017; Motagamwala et al., 2018) or an engineered flavin-dependent alcohol oxidase (Dijkman et al., 2015) or whole cell biocatalysts (Koopman et al., 2010) in aqueous media (Figure 14).

FDCA can also be produced from uronic acids present in various agricultural residues. D-galacturonic acid, for example, is available in large quantities from the pectin in sugar beet pulp (Leijdeckers et al., 2013) and D-glucuronic acid is one of the main constituents of pectin in certain soft- and hardwoods. Aerobic oxidation of uronic acids over gold catalysts (van Es et al., 2013) affords the corresponding aldaric acids that can subsequently be dehydrated to FDCA (Figure 15; Miller et al., 2017). Alternatively, uronic acids can be isomerized to the corresponding 5-keto aldonic acids which can be converted to FDCA dimethyl ester by acid catalyzed cyclodehydration to the methyl ester of 5-formyl-2-furoic acid in methanol followed by Au/C catalyzed aerobic oxidation (van der Klis et al., 2017).

Finally, we note that this review is focused on the conversion of the carbohydrate fractions of feedstocks to commodity chemicals in a biorefinery. In practice, for commercial viability both the carbohydrate and the lignin fractions will be converted to both commodity chemicals and biofuels and this will involve both chemo- and biocatalytic methods in water (Rinaldi et al., 2016; Bugg et al., 2019).

Remarkable progress has been made in the last two decades in the development of green and sustainable catalytic methodologies for the aerobic oxidations of primary and secondary alcohols to aldehydes and ketones, respectively. However, a cursory perusal of the literature reveals that we have hardly scratched the surface with regard to the application of such catalytic methodologies to the valorization of bio-based feedstocks and key platform chemicals in biorefineries. Recent developments in the engineering of oxidative enzymes, such as copper- and flavin-dependent alcohol oxidases, using directed evolution techniques, strongly suggest that industrially viable methods for catalytic oxidations relevant to a bio-based economy will be forthcoming in the near future.

One could say that glucose is the new ethylene, and possibly propylene and butenes all rolled into one. In the words of Primo Levi: “It is the destiny of wine to be drunk and it is the destiny of glucose to be oxidized.”

The author confirms being the sole contributor of this work and has approved it for publication.