Heterogeneously catalyzed lignin depolymerization
- First Online:
- Cite this article as:
- Pineda, A. & Lee, A.F. Appl Petrochem Res (2016) 6: 243. doi:10.1007/s13203-016-0157-y
- 3.7k Downloads
Biomass offers a unique resource for the sustainable production of bio-derived chemical and fuels as drop-in replacements for the current fossil fuel products. Lignin represents a major component of lignocellulosic biomass, but is particularly recalcitrant for valorization by existing chemical technologies due to its complex cross-linking polymeric network. Here, we highlight a range of catalytic approaches to lignin depolymerisation for the production of aromatic bio-oil and monomeric oxygenates.
KeywordsBiomass Lignin Heterogeneous catalysis Solid acid
The concept of sustainability is now embedded throughout modern society, from the design of cities and methods of food production and transport, to the sourcing and recycling of natural resources. Integral to these developments are new technologies able to deliver chemicals and fuels from renewable, i.e., non-fossil, resources, of which biomass has emerged as one of the most abundant and economically attractive feedstocks. Lignocellulosic biomass, in particular, is considered an attractive source of biomass for low carbon energy thermal conversion technologies, such as pyrolysis and gasification, since its use, in this regard, faces only limited competition from the agricultural and paper sector. However, biomass as a replacement for fossil fuel feedstocks is problematic due to the high water and oxygen content of the former, the presence of a water-soluble fraction of alkaline and halogen elements, along with hazardous trace elements, and resultant low-energy density/heating value, pH, and ash-fusion temperatures of biomass. The heterogeneous nature of biomass between different plants and microorganisms, and significant regional and seasonal variations even within the same species, further necessitate the development of versatile processes for the chemical transformation of biomass . However, there are strong political and financial drivers for renewable energy technologies, with the European Union mandating that a 20 % of overall energy consumption must derive from renewable sources by 2020, rising to 27 % by 2030. In regard of transportation fuels, 10 % must originate from renewable resources, and provide reductions in greenhouse gas emissions of at least 35 % as compared with fossil fuels (including emissions arising from cultivation, processing, and transport). Policy directives are also being introduced to ensure that land designated for biomass production for bio-fuels cannot have been previously used for carbon stock, such as wetlands or forest, or impact of regions with high biodiversity, such as primary forests or highly biodiverse grasslands .
Lignin valorization is the key to fuels and chemicals co-production from biomass through the biorefinery concept. Potential process under consideration within biorefineries fall into three broad classes, as categorized by Gallezot : high-temperature thermal conversion via gasification or pyrolysis to deliver light gaseous or condensable molecules for subsequent transformation via established chemical processing, such as Fischer–Tropsch conversion of syngas; lower temperature catalytic or enzymatic conversion of lignin into aromatic building blocks, such as benzene, toluene, xylene, and phenol; or one-pot routes direct to highly functionalised products.
The structure of lignin
Delignification of biomass
Lignin isolation from lignocellulosic biomass is non-trivial due to competing condensation and oxidation reactions. Lignin is obtained commonly through either sulfur-derived technologies, such as those employing sulfites for Kraft lignins, or sulfur-free processes, in which lignin is extracted by organic solvents or alkaline solution. Additional methods employed to isolate lignin are indicated by the following nomenclatures: milled wood lignin, acidolysis lignin, cellulolytic enzyme lignin, enzymatic mild acidolysis lignin, pyrolysis lignin, and steam explosion lignin . Alkaline lignin or Kraft lignin is produced in large quantities by the pulp and paper industry, which are, hence, the most abundant sources of lignin for the subsequent chemical processing . In the Kraft process, lignocellulose is treated with either sodium hydroxide and/or sodium sulfide at 170 °C, facilitating the separation of cellulosic fibers from lignin ; however, this approach is energy intensive, and the sulphite route introduces considerable sulfur into the resulting lignin. The Organosolv process for lignin isolation was developed by Theodor Kleinert , and as its name suggests employs organic solvents to extract lignin, which consequently has a low ash and sulfur content compared with Kraft lignin. Depending on the severity of the treatment, partial dissolution of cellulose and hemicellulose may also occur . The lignin extraction efficiency in an organic solvent, such as ethanol, can be increased by the addition of an inorganic acid, or by replacing ethanol with lower boiling point acetone which facilitates lignin recovery through simple distillation .
Catalytic lignin depolymerization
The decomposition of lignin into its constituent phenolic components is a hindered by the complex interconnectivity within the polymeric structure . Diverse methods have been employed to produce phenolic compounds from lignin, including hydrolysis, hydrocracking (hydrogenolysis), and oxidation [11, 20]. Oxidative cleavage of C–C and C–O–C bonds can afford vanillin and similar compounds accompanied by CO, CO2, and H2O formation. However, oxidative protocols are undesirable, because the resulting free radicals drive recondensation. In contrast, reductive cleavage of such bonds has proven more successful for yielding monomeric compounds, such as phenols, benzene, toluene, and/or xylene.
Homogeneous catalysts have been rarely utilised due to their inherent disadvantages in respect of their difficult separation from the reaction mixture, corrosiveness (in the case of soluble acid catalysts), and, hence, are more favoured for selective bond cleavage within model compounds, but have been implemented in oxidative depolymerisation [21, 22]. The present review focuses on reductive catalytic solutions which favour phenolic compounds suitable for subsequent upgrading by HDO routes  to yield fuels and fuel additives.
Catalytic depolymerization is dominated by thermochemical transformations to monomeric phenols and typically requires harsh conditions due to the stability of the polyphenolic lignin structure. Organosolv lignin is more sensitive to hydrogenolysis due to its high degree of unsaturation and low molecular weight . Bifunctional catalysts, containing both acid and noble metal sites, have attracted interest in this arena, since they can simultaneously cleave ether bonds and deoxygenate the resulting phenolic monomers in the presence of molecular H2 or a H-donor solvent. The use of noble metal catalysts favours a mixture of cyclohexane derivatives and products arising from C–C bond cleavage. Metallic nickel has shown the most promise for the production of phenolics from lignin due to its superior propensity for cleaving aryl–aryl C–O–C bonds and C–OH bonds in side chains containing CH3 or CH2 functions, commonly yielding C1–C3 alkane-substituted guaiacols as the major product . In the case of lignosulfates, Ni catalysts show good resistance to sulfur poisoning, while remaining active and selective for the transformation of lignosulfate into phenolics. While nickel is able to cleavage C–O aliphatic bonds, it is less active for the rupture of aromatic C–O and arene bonds. Lignin conversion to alkane-substituted guaiacols over Ni catalysts is independent of the support type (e.g., activated carbon, zeolite, or MgO) . Higher conversions were obtained in the presence of diol or triol functions, such us glycerol and ethylene glycol. Song and co-workers concluded that for lignosulfates, Ni(0) active sites catalyse hydrogenolysis of C–O–C bonds, C–OH bonds in side alkyl chains, and the reduction of sulfonate groups into hydrogen sulfide (H2S) .
The combination of Raney Ni with an acidic zeolite also exhibited a synergetic effect in the depolymerization of a cellulolytic enzyme lignin using a methanol/water solvent mixture at 250 °C . The monophenol yield increased from 12.9 wt% when the reaction was catalyzed by Raney Ni alone, to 21–27.9 wt% when used in conjunction with an acid catalyst. Lignin depolymerization into aromatic monomers over solid acid catalysts, such as zeolites, clays, and sulphated zirconia, was evaluated by Deepa and Dhepe . It was proposed that the improved yield from combining Raney Ni with a zeolite arose from the latter hindering the reaction between the parent lignin and other more reactive lignin fragments that could together to produce high molecular weight oligomers. Bimetallic NiM (M = Ru, Rh, Pd, Au, Mo) catalysts have also been studied for lignin hydrogenolysis [33, 34],  , wherein NaOH addition proved beneficial, promoting the formation of aromatic compounds and supressing arene hydrogenation. Konnerth et al.  also observed increased Organosolv lignin depolymerization via NaOH addition to the aqueous reaction mixture containing an Ni7Au3 nanoparticle catalyst. It was speculated that the basic conditions favoured the formation of phenolate products, whose superior solvation inhibited their subsequent absorption and hydrogenation.
Molybdenum catalysts have also been studied for lignin depolymerization over different supports, including alumina and carbon [38, 39]. Ma et al. reported lignin depolymerization in supercritical ethanol over Mo/Al2O3 and the optimum reduction temperature evaluated and identified as 750 °C. Lignin was converted into a complex mixture of low molecular weight compounds, such as aliphatic alcohols, C8–C10 esters, monophenols, benzyl alcohols, and arenes . The same authors reported that nanostructured MoxC achieved complete catalytic ethanolysis of Kraft lignin in supercritical ethanol under N2 at 280 °C . Hydrogen pressure exerted a moderate influence on the yield of liquid products, increasing alcohol production at the expense of esters. The principal products were aliphatics (C6 alcohols and C8–C10 esters), and aromatics, including phenols, phenyl alcohols, and C8–C10 arenes, with negligible dimers, oligomers or char observed. Ethanol provided superior a liquid product yield to water, methanol, or isopropanol, with a complex interplay between the solvent, lignin source and reaction conditions. The small molecules produced through this route are amenable to the direct introduction to a petroleum refinery without further upgrading.
Char formation during lignin depolymerization is highly undesirable. Cu-doped porous catalysts in supercritical methanol reportedly suppress chars, with hydrogen obtained by in situ solvent reforming . A similar approach was adopted by Anastas et al. , employing porous metal oxides containing Cu2+, Mg2+, and Al3+, achieving almost complete lignin conversion to catechols under mild conditions of 140–220 °C. The role of Cu as a char inhibitor was attributed to enhanced hydrogenolysis, dehydration, and hydrogenation steps after cleavage of β-O-4 linkers, eliminating product aromaticity [41, 42].
Klamrassamee et al. incorporated ZrO2 in MCM-41 and SBA-15 via wet impregnation as supports for the depolymerization of Organosolv lignin using methanol or water as solvents . Zirconia was selected due to it hydrothermal stability and acidity, with the supports chosen for their mesoporous textural properties. Organosolv lignin was depolymerized at 250 °C under an atmospheric pressure or hydrogen, with MCM-41 and SBA-15 supported zirconia offering superior conversion to other ZrO2 analogues, attributed to the higher acidity of the silica supports.
Lignin transformation into monomeric units, such as aromatic hydrocarbons, can also be achieved via catalytic pyrolysis, wherein the use of a catalyst affords more heavily deoxygenated and less acidic bio-oils than thermal pyrolysis, thus improving their stability and heating value. In catalytic biomass pyrolysis, the vapours generated in the pyrolysis process are passed through a catalyst bed with a view to promoting deoxygenation through, e.g., decarbonylation, decarboxylation, or dehydration reactions. Lignin requires higher temperatures and long residence times than for catalytic pyrolysis of cellulosic materials . Since lignin is the most thermally stable of the lignocellulosic biomass components, reaction temperatures typically exceed those of 500–550 °C required for cellulose and hemicellulose . The major products of lignin pyrolysis are phenolic compounds, such as phenol, guaiacol, and syringol [46, 47, 48].
A range of metal oxides (Co3O4, MoO3, NiO, Fe2O3, MnO3, and CuO), mixed metal oxides, such as CoO/MoO3, and supported metal oxides have also been explored for lignin pyrolysis [48, 53]. In general, the liquid bio-oil yield was higher for all catalysts than for thermal pyrolysis, with phenol alkoxy compounds as the primary products over the metal oxides. Nickel and manganese oxide offered a similar product distribution to thermal pyrolysis, with 2-methoxy phenol favoured by NiO and CuO, and MoO3 favouring vanillin instead of phenols and 2-methoxy phenols. The yield of aromatic hydrocarbons was increased by supporting nickel or cobalt oxides over ZSM-5 , which promoted direct deoxygenation of the aromatic lignin backbone compared to the unsupported metal oxide counterparts.
Lignin depolymerization over base catalysts is an excellent route to monomeric phenols. Aryl–alkyl (β-O-4) ether bonds in lignin are easier to cleave than C–C bonds in base-catalyzed lignin depolymerisation , although mineral bases typically require harsh reaction temperatures of around 340 °C [55, 56, 57], which can generate significant gaseous products via side reactions and consequent lower monomeric hydrocarbon yields . In base-catalyzed depolymerization, the choice of lignin influences the yield of bio-oil, but not the resulting composition ; however, the use of mineral bases may necessitate neutralisation of the subsequent bio-oil, in addition to reactor corrosion. Thring et al. studied the effect of time and temperature on the depolymerization of Alcell lignin catalyzed by NaOH  obtaining guaiacol, syringol, and under certain conditions catechol and its derivatives [57, 60, 61]. Base-catalyzed depolymerization has proven an effective protocol in polar organic solvents, such as ethanol and methanol, suppressing char formation, and hence enhancing monomer and dimer production relative to solid acid catalysts. Toledano et al. [56, 62] reported that base catalysts prevent the repolymerization of monomeric products from lignin, offering improved bio-oil yields. Base strength plays an important role on the depolymerization mechanism, with hydroxides promoting phenol, cresol, guaiacol, catechol, and 4-methylcatecol production, while neither catechol or cresols were observed when using potassium carbonate . The use of capping agents, such as boric acid, to protect hydroxyl groups in phenolic compounds has also been explored in efforts to minimize repolymerization, and results in lower molecular weight products . In alkaline media, boric acid reacts with phenolic monomers to produce their respective esters (NaB(OR)4), which hinders side reactions of phenol products and, in turn, hinders char formation. The formation of borated esters during base-catalyzed depolymerization also offers higher yields as compared with boric acid protection alone under analogous conditions . The use of phenol was also explored as a capping agent to stabilise monomeric products of lignin depolymerization, with catechol and ferulic acid formed, whereas capping by boric acid yielded significant dimers. Toledano et al.  suggested that the different behaviour of these capping agents were due to boric acid trapping intermediates which lowered the yield of monomeric aromatics, whereas phenol capping reduced oligomerization, but not demethoxylation and dealkylation pathways.
MgO is a widely used solid base catalyst for biodiesel synthesis , and has also been employed for lignin depolymerisation in a range of solvents, including water, methanol, ethanol, and tetrahydrofuran (THF) . Maximal lignin conversion was achieved for THF due to its superior solubilisation of lignin; however MgO also catalyzed repolymerization of phenol oligomers. Other solid bases, such as RbCO3 CsCO3, have also shown promise for β-O-4 cleavage of ether bonds in lignin model compounds, but are able to suppress char formation and increase bio-oil yield [65, 66].
Considering that lignin depolymerization can be performed successfully with homogeneous base catalysts, such as KOH and NaOH, with conversions reaching 95 % , , and that heterogeneous Ni catalysts are able to cleave C–O bonds , Sturgeon et al. proposed a combination of these active components . A 5 wt% Ni supported on a MgAl hydrotalcite (a solid base catalyst employed widely for lipid transesterification to biodiesel [68, 69, 70]) was shown to depolymerise a lignin model compound, 2-phenoxy-1-phenethanol, and ball-milled lignin extracted from corn stover, into low molecular weight alkyl aromatics; interestingly, nickel nanoparticles present on the (likely external) hydrotalcite surface required no activation with H2 prior to their catalytic application, suggesting that the active species was a mixed valence nickel oxide. Catalytic properties of nickel hydrotalcites are strongly influenced by the nature of interlayer anion. The presence of NO3− anions, or other basic anions, such as OH− and CO32−, in the interlayers conferred up to a three-fold activity enhancement in the depolymerisation of the model lignin compound 2-phenoxy-1-phenethanol , wherein acetophenone (and phenol) were the principal products. Although the origin of this base promotion remains unclear, the depolymerisation mechanism may involve direct nitration of the phenolic. The same nitrate intercalated Ni hydrotalcite also proved effective for the depolymerization of enzymatically hydrolyzed lignin from corn stover, resulting in a large proportion of low molecular weight products, predominantly phenol, guaiacol, and syringol monomers, when performed in water, as compared with 4-vinylphenol when conducted in 3-methyl-3-pentanol at temperatures <200 °C.
Oxidative depolymerization is often considered undesirable due to the propensity for generating free radicals which may promote repolymerization of the desired mono/dimeric products and consequent char production. However, oxidative protocols can be advantageous, since they permit retention of both aromatic and acyclic organic frameworks without C–C bond scission. Oxidative degradation is the route adopted for lignin decomposition in the natural world , and offers a hydrogen-free pathway to vanillin [11, 73]. A range of organometallics (notably methyl trioxo rhenium), polyoxometalates, and vanadium catalysts are reported for the oxidative depolymerization of lignin. Vanadium–oxo complexes have also been demonstrated as efficient catalysts for C–O bond cleavage in lignin model compounds . Silk et al. showed that the choice of vanadium complex strongly influenced selectivity, directing either the cleavage of C–C bonds between an aryl ring and the contiguous group, or C–O linkages scission . Polyoxometalates are robust redox catalysts, and effective for oxidative degradation; when performed in the presence of molecular oxygen, a two-step process occurs, in which the polyoxoanion is first reduced and, subsequently, reoxidised without any net structural change . Polyoxometalates applied for the decomposition of lignin residues under molecular oxygen often yields CO2 and H2O by-products . Dimethyl succinate and aromatics have also been produced from pyrolytic lignin using a vanadium/molybdenum polyoxometalate in a methanol/water solvent mixture under oxygen . Bio-oil production increased with the amount of methanol (or ethanol) via methanol oxidation to CCH3 and CH3OC radicals which promoted lignin depolymerization. However, the addition of methanol (or ethanol) under alkaline conditions did not favour monomeric products . The major products using a H5PMo10V2O40 catalyst were organic acids and their esters, such as dimethyl fumarate and dimethyl succinate, formed by the breakdown of aromatics .
Salen complexes, particularly of cobalt, have also been tested for oxidative lignin transformation by H2O2 or molecular oxygen. Copper complexes, such as Cu-bipy and Cu-phen, can also catalyse the oxidative transformation of lignin model compounds, such as veratryl alcohol and 2,2-biphenol. Oxidation activity was a function the of steric and electronic ligand effects . Hydrogen peroxide is widely used in organic transformations and, generally, considered as a green oxidant with water as the sole by-product . The in situ production of reactive oxygen species from peroxy species by reaction with Cu or Fe is termed the Fenton reaction , and been utilised for the degradation of veratryl alcohol .
A chemoselective method for the aerobic oxidation of secondary benzylic alcohols, in the presence of unprotected primary alcohols (abundant in lignin), has also been developed by combining 4-acetamido-TEMPO (as a radical oxidant) together with HNO3 and HCl [83, 84]. This approach has been applied to wood pulp delignification, in which oxidation of the Cα alcohol to a ketone favours the breakdown of β-O-4 linkages. Wang et al.  also reported a bifunctional Pd/CeO2 catalyst able to efficiently cleave β-O-4 bonds in the presence of Cα-hydroxyl group, in which synergy between palladium activation of the oxidant and the redox properties of ceria facilitates β-O-4 cleavage; Pd is widely used in heterogeneous catalysis for aerobic alcohol oxidation [86, 87, 88, 89, 90]. This bifunctional catalyst was able to transform a range of Organosolv lignin in ethanol under oxygen to vanillin, guaiacol, and 4-hydroxybenzaldehyde.
Oxidative cleavage of a dilignol model compound, erythro-1-(3,4-dimethoxyphenyl)-2-(2-methoxyphenoxy)-l,3-propanediol, was investigated employing transition metal containing hydrotalcite, and V(acac)3 and Cu(NO3)2.3H2O mixtures, with molecular oxygen as the oxidant . Oxidative depolymerization of the dilignol model was influenced by the solvent selection, with negligible conversion observed in toluene, but good activity obtained in pyridine resulting in veratric acid as the major product. The same catalysts were subsequently applied to both Organosolv and Kraft lignin; however, their performance could not be readily extrapolated from dilignol model due to the complexity of practical lignin sources and presence of zeolite contaminants; harsher reaction conditions (extended reaction duration and higher oxygen pressure) were required to depolymerize Organosolv lignin, even at lower concentrations than used for the dilignol model, although both Cu-V hydrotalcite and V(acac)3/Cu(NO3)2·3H2O successfully cleaved β-O-4 linkages and resinol components yielding dimeric and trimeric products.
Use of ionic liquids
Ionic liquids (ILs) have attracted a great deal of interest in recent years as green solvents suitable for the solubilisation of lignocellulosic biomass, that are insoluble in other media [92, 93]. After dissolution in an IL, biomass is amenable to acid hydrolysis , acylation , and acetylation . Selection of the appropriate IL is key to control the product selectivity in oxidative depolymerization . Lignin depolymerization has proven possible under mild conditions of only 110–150 °C through the use of an acidic IL, 1-H-3-methylimidazolium, via hydrolysis of ether bonds . The most widely used ILs are 1-ethyl-3-methylimidazolium (EMIM) and 1-butyl-3- methylimidazolium (BMIM) with anionic counterions, such as trifluoromethanesulfonate [CF3SO] . Bosmann et al.  evaluated the choice of IL in the oxidative depolymerization of lignin catalyzed by iron, copper, and manganese salts. Highest conversion was found for Mn(NO3)2 in [EMIM][CF3SO] after 24-h reaction at 100 °C and 8.5 bar air, yielding 2,6-dimethoxy-1,4-benzoquinone (DMBQ), and either syringaldehyde or vanillic acid . Cobalt salts favoured the oxidation of benzyl and other alcohol functionalities within lignin after its complete dissolution in an IL, such as 1-ethyl-3-methylimidazolium diethylphosphate [EMIM][DEP], without modifying phenolic functionalities or 5-5′ β-O-4 phenilcoumaran bonds, explaining the absence of monomeric products . [EMIM][DEP], thus provide an efficient medium for lignin oxidative degradation.
Enzymatic lignin depolymerization
Lignin represents an important renewable resource for the production of chemicals and bio-fuels, and hence, its valorisation will play an increasingly important role in the development of sustainable biorefineries. However, the heterogeneous nature of lignin presents significant obstacles for the establishment of a standard methodology for its depolymerization. The development of novel catalytic routes to lignin depolymerisation will facilitate more atom-efficient and energy-efficient processes, and safer approaches, such as the use of hydrogen-donor solvents instead of high-pressure molecular hydrogen. Heterogeneous catalysts will also afford facile catalyst separation/reuse and product purification. Alternative solvents, such as ILs, are promising for low-temperature lignin decomposition in conjunction with the conventional mineral acids or tailored homogeneous/heterogeneous catalysts, although the high price and potential corrosiveness and toxicity of some ILs may limit their industrial scale application. The aromatic fraction produced through catalytic lignin depolymerization still exhibits a comparatively low heating value and high acid content, reflecting the high proportion of oxygenated compounds present in the bio-oil, and concomitant poor miscibility with aliphatic hydrocarbons, in addition to a broad boiling point range. Bio-oil upgrading technologies will, thus, be essential for either fuel or chemical applications; of these, hydrodeoxygenation (HDO) will likely prove most successful for biofuel, affording the complete hydrogenation of oxygenated depolymerisation products to benzene and cyclohexane.
We thank the EPSRC (EP/K021796/1 and EP/K029525/2) for financial support. Support from the European Union Seventh Framework Programme (FP7/2007-2013) under Grant Agreement No. 604307 is also acknowledged.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.