Hu
Li
*a,
Haixin
Guo
b,
Zhen
Fang
*a,
Taku Michael
Aida
c and
Richard Lee
Smith
Jr
*b
aBiomass Group, College of Engineering, Nanjing Agricultural University, 40 Dianjiangtai Road, 210031, Nanjing, Jiangsu, China. E-mail: hli17@njau.edu.cn; zhenfang@njau.edu.cn
bGraduate School of Environmental Studies, Tohoku University, 6-6-11, Aoba, Aramaki, Aoba-ku, Sendai, 980-8579, Japan. E-mail: smith@scf.che.tohoku.ac.jp
cDepartment of Chemical Engineering, Faculty of Engineering, Fukuoka University, 8-19-1, Nanakuma Jonan-ku, Fukuoka 814-0180, Japan
First published on 9th January 2020
Biomass resources have infinite possibilities for introducing nitrogen, sulfur, or phosphorus heteroatoms into their structures by virtue of controllable carbon–heteroatom bond formation. In this review, cycloamination approaches for thermal (catalyst-free) and catalytic transformation of biomass feedstocks into N-heterocyclic molecules including mechanistic pathways are analyzed. Bottom-up (small molecule substrates) and top-down (large molecule substrates) are considered. Sustainable routes for synthesis of five-membered (pyrroles, pyrrolidones, pyrazoles, imidazoles), six-membered (pyridines, pyrazines), fused (indoles, benzimidazoles), and other relevant azaheterocycles are critically assessed. Production of biomass-derived six-, seven-, and eight-membered as well as fused N-heterocyclic compounds with present approaches have relatively low selectivities. Attention to methods for forming analogous sulfur or phosphorus heteroatom compounds from biomass resources using either bottom-up or top-down strategies appear to have been greatly overlooked. Synthetic auxiliaries (heating modes, nitrogen sources) that enhance reaction efficiency and tunability of N-heterocyclic ring size/type are considered and plausible reaction mechanisms for pivotal pathways are developed.
Nitrogenous compounds hold a privileged position in the preparation of drugs, agrochemicals, polymers, and other functional materials.27,28 In particular, nitrogen species are presented in more than 80% of the top 200 pharmaceuticals,29,30 and two thirds of these N-containing medicines contain N-heterocyclic skeletons.31 Given the above considerations, urgent attention is needed to develop green and efficient approaches to task-specific azaheterocycles from renewable biomass feedstocks via sustainable chemistry. The employed nitrogen sources can be derived from ammonia, amines/amides pre-prepared by C–N coupling reactions (e.g., reductive amination, aminolysis, and amidation),32–36 or natural N-reservoirs (e.g., chitin, proteins) for producing amines and derivatives (e.g., glucosamine, N-acetyl-D-glucosamine, amino acids).37–41 Beginning from the accessible bio-based functional molecules, N-heterocycles can be constructed by C–N and/or C–C bond formation typically via three dominant synthetic routes including intramolecular cyclization, cycloaddition (e.g., aza-Diels–Alder), and multicomponent condensation reactions (Fig. 2).42,43 Relevant reaction processes and pathways in the presence or absence of designed catalysts have been investigated under optimized thermal conditions,42,43 for which these cycloamination strategies are attractive for the sustainable synthesis of bio-based N-heterocycles, with much development needed to further expand substrate scope and generality.
Previously published review articles have focused on design of functional catalytic materials and conversion means for efficient production of specific/desired platform molecules and biofuels from lignocellulosic biomass.44–55 However, increasing attention is on the synthesis of heteroatom-containing chemicals via green and sustainable routes, with emphasis on the exploitation of approaches to acyclic amines and correlative N-containing commodities or fine chemicals.56–58 In view of the much higher versatility of azaheterocycles, this review aims to critically assess the cycloamination strategies developed for the preparation of five-membered (section 2), six-membered (section 3), fused (section 4), and other (section 5) biomass-derived N-heterocyclic products. Auxiliaries such as heating modes and solvent or substrate types optimized to overcome reaction barriers are discussed, with reaction mechanisms and pathways being outlined for representative processes and systems.
Beginning from 1,4-dicarbonyl compounds 1, especially since 2,5-hexadione is derivable from hexose sugars,80N-substituted pyrroles 4 can be synthesized by Paal–Knorr condensation with primary amines 2 over acid catalysts or with nitro compounds 3 over metal catalysts (Scheme 1). The combination of a heterogeneous cobalt–nitrogen catalyst (Co–Nx/C-800-AT) with formic acid serving as both hydrogen donor and an acid catalyst is efficient for one-pot synthesis of 2,5-dimethyl-1-phenylpyrrole (95.2% yield, 110 °C, 12 h) from nitrobenzene and 2,5-hexadione in ethanol.81 Nitrobenzene is subjected to transfer hydrogenation with formic acid over the Co–Nx/C-800-AT catalyst, affording aniline (Scheme 1). The carboxide species of 2,5-hexadione is protonated by formic acid to yield the intermediate (IM), which is then attacked by aniline, followed by cascade cyclization and dehydration to afford the product.81 The reaction system is applicable to heterocyclization of substituted nitrobenzenes with 2,5-hexadione to produce corresponding N-substituted pyrroles with yields of up to 100%. Notably, the developed cobalt–nitrogen catalyst is tolerable to the acidic liquid H-donor (HCOOH), which may be ascribed to the highly dispersed metal particles that are coordinated and stabilized by nitrogen species of the solid carbonaceous supports. This unique heterogeneous feature of the non-noble metal catalyst is able to not only remarkably reduce the loss of metal species in the reaction processes and the overall production cost, but also provides an example of using sustainable HCOOH instead of flammable hydrogen gas as H-donor.
Scheme 1 Synthesis of N-substituted pyrroles 4 from 1,4-dicarbonyls 1 with amines 2 or nitro compounds 3. Adapted with permission from ref. 81 and 82, Copyright © 2017 Royal Society of Chemistry; Copyright © 2010, American Chemical Society. |
Lignocellulosic biomass-derived alcohols are another green feedstock for synthesis of pyrroles, although complex byproducts (e.g., cyclic imides, pyrrolidines, and lactones) were often frequently obtained by coupling of carbonyl intermediates 1in situ formed from catalytic dehydrogenation of alcohols.82–85 With hydrogen and water as co-products, catalytic acceptorless dehydrogenative coupling of 1,4-butanediol or 1,4-substituted 1,4-butanediols 5 and amines 2 over base-metal complexes (e.g., cobalt or manganese pincer complex) generate 1 or 1,2,5-substituted pyrroles 4, respectively, at 150 °C for 24 h with 90% yield (Scheme 2).86,87 In the catalytic process, an aldehyde or ketone intermediate 1 is initially formed by dehydrogenation of alcohol to liberate H2, followed by coupling with the primary amine 2 to produce the N-substituted pyrrole 4 and water via Paal–Knorr condensation (Scheme 2).86,87 A synergic effect between the metal and ligand species is observed in the dehydrogenative coupling reaction, which mainly contributes to the enhanced selectivity towards the product pyrrole.87 Although these homogeneous catalytic systems exhibit pronounced performance in the Paal–Knorr condensation reaction, difficulty in the catalyst recovery will lead to additional cost and negative impact on the environment.
Scheme 2 Catalytic coupling of primary or secondary diols 5 with primary amines 2. Adapted with permission from ref. 86 and 87, Copyright © 2016 Wiley-VCH; Copyright © 2019, American Chemical Society. |
Unsaturated diols, such as butene-1,4-diol or butyne-1,4-diol, can be used to construct substituted pyrroles by reaction with primary amines (Scheme 3) over precious metals (e.g., Ru, Pd) in the presence of phosphine ligands, although they afford relatively low selectivities.88,89 The application of earth-abundant base metals (e.g., Ni, Co, Fe, and Mn) to catalysis with specific nitrogen ligands is able to overcome the long-standing problem of high catalyst cost and low productivity.90,91 It is postulated by the authors that the nonprecious metal (e.g., Fe and Ni) catalyze dehydrogenation or isomerization of cis-butene-1,4-diol or butyne-1,4-diol (6) to afford 7 that is then condensed with primary amine 2, leading to the formation of allylic amine 9via hydrogenation of imine 8.92,93 In a one-pot operation, subsequent dehydrogenation of 9 gives intermediate 10, which is further subjected to cascade intermolecular cyclization and dehydration to construct N-substituted pyrroles 4 (Scheme 3). In parallel, N-substituted pyrroles may also be generated by sequential metal-catalyzed isomerization and intermolecular cyclization of the imine 8via intermediate 10.92,93 Due to the presence of CC or CC bond in the diols, the reaction system is tolerant of free halides and alcohols, leading to pyrrole yields of up to 90%.90,91
Scheme 3 Intermolecular cyclization of butene-1,4-diol or butyne-1,4-diol 6 with primary amines 2 to pyrroles 4 catalyzed by base metal (e.g., Ni). Adapted with permission from ref. 92 and 93, Copyright © 2017, American Chemical Society; Copyright © 2016 Wiley-VCH. |
Metal-catalyzed hydrogen autotransfer or the borrowing of hydrogen is capable of converting alcohols to carboxides together with the liberation of H2, followed by condensation to give imine species that could be further reduced to amines by the in situ formed hydrogen (Scheme 3).94,95 Michlik and Kempe reported on the synthesis of 2,5-disubstituted pyrroles 4 from sustainable secondary alcohols 11 and amino alcohols 12 initiated by alcohol dehydrogenation in the presence of sodium tert-butoxide (NaO-t-Bu) and an organoiridium catalyst via sequential formation of C–N and C–C bonds (Scheme 4).96 This synthetic protocol was implemented under relatively mild reaction conditions (90 °C, 24 h) with ketone 13 and imine 14 as key intermediates, and could tolerate a wide range of functional groups like olefins, Cl, Br, NH2, OH, and organometallic moieties with moderate to high yields of pyrroles (42–97%).96 Other types of N/P-ligands stabilized Ir and Ru complexes are demonstrated to be efficient for the synthesis of pyrrole derivatives,97–101 although it would be desirable to develop catalysts based on earth-abundant base metals in view of recycle and reuse requirements. Kallmeier et al. reported that pyrrole could be synthesized in isolated yields of up to 93% over Mn PN5P-pincer catalysts, whereas no significant activity was observed for related Fe and Co complexes.102 The Mn-mediated reaction smoothly progressed at moderate reaction temperatures (78 °C) with 2-methyltetrahydrofuran as solvent, which is a lower temperature than that used for Ir- and Ru-catalyzed processes (≥90 °C),96–102 and is an inspiring example of nonprecious metal-based catalysis for replacing noble metal complexes in the synthesis of heterocycles. The release of H2 from the used reagent alcohols over the metal catalysts can eliminate the use of flammable and high-pressure hydrogen gas, while the development of active heterogeneous counterparts will be helpful for practical applications.
Scheme 4 Catalytic synthesis of pyrroles from secondary alcohols and amino alcohols via cascade dehydrogenation (step 1), imine formation (step 2), intramolecular C–C coupling and isomerization (step 3). Adapted with permission from ref. 96, Copyright © 2013, Springer Nature. |
Another method to synthesize pyrroles (Scheme 5) is via catalytic amination of bio-derived furanic compounds 15 with primary amines 2 in the presence of an acid catalyst (e.g., Al2O3 and TiO2) that affords pyrrole derivatives at 20–60% yields at high-temperature (250 °C to 400 °C).103,104 Under relatively mild reaction conditions (150 °C, 5 bar N2, 0.7 h), Tao et al. demonstrated that a solid acid H-form zeolite H-Y(2.6) gave N-(m-tolyl)-2,5-dimethylpyrrole 4a in nearly quantitative yields (99%) from the condensation of 2,5-dimethylfuran 15a and m-methylaniline 2a.105 Besides the porous structure, moderate Lewis/Brønsted acid strength of H-Y(2.6) as determined by NH3-TPD and pyridine-adsorbed FT-IR, contribute to the optimized pyrrole yield. In contrast, H-MOR(12.5) and H-ZSM-5(18) with strong acidity significantly inhibit the desired condensation reaction (9–34% yields), possibly due to deactivation of the active sites by a strong binding interaction with N-containing intermediates.105 Rather than from the direct reaction between aniline and 2,5-dimethylfuran, 2,5-hexanedione 16 was identified to be the key intermediate that reacts with different anilines via Paal–Knorr reaction to afford pyrrole derivatives. With water as the only co-product, this atom-economic H-Y(2.6)-catalytic system (Scheme 5) shows good generality for synthesis of polysubstituted pyrroles 4 (62–93% yields) from anilines 2 and bio-derived furans 15 (e.g., 2,5-dimethylfuran, 2-methylfuran, and furan) at 150 °C or 180 °C for 0.5 h to 6 h.105 Unlike conventional Paal–Knorr reactions that typically use 1,4-diketones as substrates, this catalytic approach based on biofurans and commercial zeolites is a promising method to construct pyrrole-based scaffolds.
Scheme 5 Catalytic condensation of furans 15 with anilines 2 to give pyrroles 4. Reaction conditions: 1 mmol furan, 1 mmol aniline, 2 mL toluene, 150 mg of H-Y(2.6), pN2 = 5 bar. Adapted with permission from ref. 105, Copyright © 2017, American Chemical Society. |
Starting from biomass-derived 5-hydroxymethyl-furfural (HMF, 17), N-substituted 2-hydroxymethyl-5-methylpyrroles 4 (74–99% yields) can be synthesized using ethanol as solvent at room temperature after reaction for 10 min to 48 h via a two-step catalyst-free process, involving the hydrogenation of HMF (17) to 1-hydroxyhexane-2,5-dione (HHD, 18) over Ir-complex 19, followed by Paal–Knorr reaction with amines and anilines (2) bearing both electron-withdrawing and electron-donating groups (Scheme 6).106 The presence of hydroxyl functionality has the possibility for further modification to afford bioactive compounds and to introduce desired ligating groups, with no detailed purification steps being required to obtain the desired products, except for simple evaporation of the solvent (ethanol) and vacuum drying.
Scheme 6 Synthesis of N-substituted 2-hydroxymethyl-5-methylpyrroles from 5-hydroxymethyl-furfural (HMF) via 1-hydroxyhexane-2,5-dione (HHD). Adapted with permission from ref. 106, Copyright © 2018, Wiley-VCH. |
Conventionally, reaction of hexose sugars (e.g., glucose and fructose) directly with amines 2 are able to proceed either Amadori rearrangement of glycosylamine 20 or Maillard reaction, affording the equilibrium mixture of furanose 21a, pyranose 21b, and the open chain isomer 21c in solution107 or a complex mixture of poorly characterized molecules at 200 °C or higher,108 respectively (Scheme 7). In the presence of an organic acid (oxalic acid), Adhikary et al. found that glucose directly reacting with N-benzylamine gives 1-benzyl-5-(hydroxymethyl)pyrrole-2-carbaldehyde with a maximum yield of 40% in DMSO at 90 °C for 0.5 h.109 In contrast, acetic acid (pKa 4.7), trifluoroacetic acid (pKa 0.23), and sulfuric acid (pKa −3.0) afford the pyrrole in yields of 8%, 28%, and 16%, respectively. The optimal acidity of oxalic acid (pKa 12.5) most likely contributed to its superior reactivity in the cascade process involving nonaqueous Maillard reaction, ring-opening, dehydration, cyclization, and hydrolysis to produce the pyrrole 4 and regenerate the amine 2 (Scheme 7).109 The hydroxyl configuration of sugars does not remarkably affect the reaction efficiency, and corresponding N-substituted 5-(hydroxymethyl)pyrrole-2-carbaldehydes in comparable yields (21–53%) is obtained from galactose, mannose, ribose, or xylose by reacting with primary amines. Due to the existence of active species (–OH, –CHO), the pyrrole-2-carbaldehyde skeleton shows great potential for production of pyrrole alkaloid natural products (Scheme 7). The use of more easily available bio-based feedstocks can remarkably reduce production cost, and improve development of correlated reaction routes with recoverable and lost-cost catalysts in the presence of low-pressure liquid H-donor sources (e.g., HCOOH, alcohols) and is thus highly desirable for the synthesis of pyrroles.
Scheme 7 Direct conversion of hexose sugars by reacting with amines 2 to N-substituted 5-(hydroxymethyl)pyrrole-2-carbaldehydes 4. Adapted with permission from ref. 109, Copyright © 2015, American Chemical Society. |
Scheme 8 Catalytic reductive amination of levulinic acid (LA, 22a) with primary amines 2 to pyrrolidones via amination–reduction–cyclization (path 1) or amidation–cyclization–dehydration–reduction (path 2). Adapted with permission from ref. 112 and 113, Copyright © 2017 & 2018, Royal Society of Chemistry. |
Molecular hydrogen gas is frequently used as H-donor in schemes for cascade reductive amination and amidation of LA (22a) or its ester to synthesize pyrrolidones. Among the developed catalytic systems, noble metal catalysts (e.g., Pt/P-TiO2, Pt-MoOx/TiO2, Pt/TiO2D, Ru/TiO2, Ir/SiO2-SO3H, and iridium complexes) are highly efficient for the cascade reaction process (up to 99% yield of pyrrolidones) under mild conditions (room temperature to 120 °C, P(H2) ≲1 MPa).114–120 Apart from LA (22a) and primary amines, carbohydrates (e.g., glucose) and nitro compounds can be employed as the precursor of LA (22a) and nitrogen sources for producing pyrrolidones via additional acid-catalyzed hydrolysis and in situ hydrogenation steps, respectively.121,122 The replacement of precious elements in Scheme 8 with abundant and low-cost metals is highly desirable, for which several well-designed base metal catalysts, such as Cu15/Al2O3 doped with Pr (Cu15Pr3/Al2O3),123 carbon-supported FeNi nanoparticles,124 and carbon nanotubes supported porous-carbon-coated Ni (CNFx@Ni@CNTs)125 have been explored for synthesis of N-substituted pyrrolidones (up to 99% yield) from LA (22a) and amines although they require relatively harsh reaction conditions (130 °C to 175 °C, 1 to 5 MPa H2). Well dispersed metal particles (e.g., Cu and Ni) on porous solid supports (e.g., γ-Al2O3, carbon nanotubes) and the use of continuous fixed-bed reactors enhance catalytic performance of prepared non-noble metals.123–125 Solid supports not only increase metal stability and improve reusability, but also provide acid sites with appropriate acid strength and density that enhances substrate adsorption onto active sites of the catalyst that facilitates tandem condensation and hydrogenation processes.
In comparison with non-noble metals (e.g., Ni), precious metals (e.g., Pt) typically show stronger chemisorption of the imine generated from LA (22a) and an amine (e.g., benzyl amine) via condensation, but they have lower affinity for H2.126,127 With respect to non-noble metal catalysts, weak chemisorption towards the intermediate leads to difficulty in activation of the imine, while strong H2 adsorption results in excessive blocking of the metal surface. These two factors might eventually result in the non-noble metal catalyst to have relatively low hydrogenation activity based on the Sabatier principle.128 In good agreement with the above statement, Gao et al. elaborated an unconventional pathway for the reductive amination of LA (22a) over a base metal catalyst, in which Ni/C first promotes the formation of amides 26 followed by undergoing cascade cyclization, dehydration, and reduction to yield the target product pyrrolidone 23 (path 2, Scheme 8),125 in accordance with the Sabatier principle.
Instead of using flammable H2 gas, air-stable and safe hydrosilane (e.g., PhSiH3 and (EtO)3SiH) can be used as a mild reductant for reductive cycloamination of LA (22a) with primary amines to produce pyrrolidones 23 in moderate to good yields (up to 99%) over B(C6F5)3,129 Fe-complex,130 or transition metal salts such as In(OAc)3 and AlCl3·6H2O131,132 at 30 to 120 °C for 1 to 24 h, as in Scheme 9. Particularly, by changing the transition metal salt from In(OAc)3 or AlCl3·6H2O to either InI3 or RuCl3·3H2O that bear stronger Lewis acidity, affords pyrrolidines 27 in good yields (55–93%) under similar reaction conditions to those used in the synthesis of pyrrolidones 23.131,132 Namely, the removal of oxygen from the lactam (pyrrolidone, 23) with hydrosilane (PhSiH3) to give the cyclic amine (pyrrolidine, 27) and corresponding siloxane needs to be activated by a Lewis acidic metal salt that is stronger than In(OAc)3 or AlCl3·6H2O. In other words, the selectivity toward pyrrolidones can be controlled by design of the catalyst with appropriate Lewis acidity or basicity. Similar to metal salts, a pharmaceutically acceptable and nontoxic ionic liquid 1-butyl-3-methylimidazolium lactate ([BMIm][Lac]), in combination with (EtO)3SiH, is efficient for production of pyrrolidones 23 (36–96% yields) at 80 °C within 1 h to 3 h.133 Serving as a multifunctional catalyst, [BMIm][Lac] with both acidic (–OH) and basic (–COO−) sites concurrently promotes activation of hydrosilane and cycloamination of LA (22a) or keto acids (22) successively to 24 and 28, to finally afford pyrrolidones 23 (Scheme 9B). However, the use of hydrosilanes as H-donor may co-produce waste due to the formation of silicon resins, and the spent homogeneous catalysts suffer from issues regarding reusability and potential environmental contamination. Exploration of benign and eco-friendly alternatives for both aspects is thus highly desirable for sustainable synthesis of pyrrolidones and other relevant azaheterocycles.
Scheme 9 Schematic illustration for (A) reductive cycloamination of LA (22a) or keto acids (22) to pyrrolidones/lactams 23 or pyrrolidines/cyclic amines 27 (30 °C to 120 °C for 1 h to 24 h) and (B) detailed reaction mechanism. Adapted with permission from ref. 131–133, Copyright © 2016, Wiley-VCH; Copyright © 2017, Royal Society of Chemistry; Copyright © 2017, American Chemical Society. |
Formic acid (HCOOH) was a co-product of LA (22a) in the synthesis of LA from biomass sugars, and its use as a hydrogen source minimizes the generation of chemical waste in biorefineries.134,135 Metal catalysts such as Au/ZrO2-VS, Ir-complex, Ru-complex, Ru3(CO)12, and Fe3(CO)12 allow selective decomposition of HCOOH to CO2 and H2, while they were normally unable to tolerate CO that forms in the dehydration of HCOOH.136–140 In the reductive cycloamination of LA (22a), either straightforward hydrogenation with H2 generated in situ from HCOOH or via transfer hydrogenation process occurs as the rate-determining step in the synthesis of pyrrolidones.136–140 Instead of using high-pressure hydrogen or acidic HCOOH and basic ammonia or amines for the reductive amination reaction, ammonium formate (HCOONH4) is a convenient nitrogen and hydrogen source. Amarasekara and Lawrence found that RANEY® Ni efficiently catalyzes cycloamination of LA (22a) with HCOONH4 in water, affording 94% yield of 5-methyl-2-pyrrolidone (MPD) at 180 °C for 3 h,141 and their protocol has authentic safety advantages over the conventional methods using high-pressure H2–NH3 gas mixture. Sun et al. showed that NHC (N-heterocyclic carbene)-based Ru(II)-coordination polymer is active for the synthesis of MPD (up to 99% yield) from LA (22a) and HCOONH4 at 80 °C for 12 h, with TON value of 6.7 × 104 being reported,142 with both catalysts being recyclable without apparent activity loss, and being applicable to producing N-substituted 5-methyl-2-pyrrolidones from amines and LA (22a).141,142
Under non-catalytic conditions, Wei et al. reported that LA (22a) reacts with benzylamine to give 1-benzyl-5-methylpyrrolidin-2-one (BMP) to afford a moderate yield of 72% using DMSO as solvent at 100 °C for 4 h.143 Upon addition of equivalent triethylamine relative to LA (22a), BMP yield increases to 89% under otherwise identical conditions. Thus, an appropriate balance of system acidity and basic additive (e.g., triethylamine) is used to obtain favorable reaction progress, and the catalyst-free system is suitable for reductive amination of LA (22a) with electron-deficient and -rich amines to afford a variety of N-substituted 5-methyl-2-pyrrolidones (up to 93% yield).143 The use of a continuous-flow microreactor heated at 140 °C remarkably accelerated the reaction rate, and comparable yields of pyrrolidones could be achieved with a residence time of 10 min.144 To further improve the greenness of the reaction, Ledoux et al. shows that the thermal treatment of LA (22a), HCOOH and amine mixtures allow access to a series of 5-methylpyrrolidone derivatives with >80% yields in most cases.145 Due to the absence of catalysts, solvents or additives, this sustainable and efficient reaction system has an exceptionally low E-factor of 0.2 and has good potential for application on the industrial scale.
Li et al. developed a generic strategy that does not require catalyst or external hydrogen by involving in situ controlled-release of HCOOH from N-formyl species (e.g., HCONH2) with H2O for the cycloamination of LA (22a) and other keto-acids (22), that provides access to 5-methylpyrrolidones and relevant N-(un)substituted lactams,146 that is elucidated by a combination of model experiments and density functional theory calculations. An unconventional pathway via cyclic imines (5-methyl-3,4-dihydro-2-pyrrolone (29) and its tautomeric structures) as key intermediates is elucidated by density functional theory (DFT) calculations show yields of 5-methyl-2-pyrrolidone (30) from LA (22a) and HCONH2, which is different from the conventional approaches encompassing cascade reductive amination and cyclization (Scheme 10).146 The simple and eco-friendly protocol of Li et al.146 may open an avenue for direct synthesis of N-(un)substituted pyrrolidones/lactams by cyclic diamination of keto acids without external hydrogen gas. Besides the above-discussed reductive amination of LA (22a) and N-containing compounds, a number of other renewed synthetic routes such as ketoamides proceeding via cascade cyclization and ionic hydrogenation mediated by Al(OTf)3 and Et3SiH,147 reductive transformation of itaconic acid and NH3 over Ru/C and H2,148 reductive N-alkylation and decarboxylation of glutamic acid catalyzed by Pd/Al2O3 with and H2,149 and Zr-catalyzed N-acylation of lactams150 have been explored for efficient construction of pyrrolidone-type motifs. The catalyst-free reaction system seems more sustainable and economic for producing pyrrolidones, while its reaction rate is much lower than metal- or acid-catalyzed processes. In this regard, the design of appropriate continuous flow reactors may facilitate the rapid thermal conversion routes to the synthesis of pyrrolidones.
Scheme 10 Reaction pathways for cycloamination of LA (22a) and H2NCHO in water with computed free energies and enthalpies in parentheses (kJ mol−1). IM: intermediate; MDPY: 5-methyl-3,4-dihydro-2-pyrrolone; MPD: 5-methyl-2-pyrrolidone. Adapted with permission from ref. 146, Copyright © 2019, Wiley-VCH. |
Scheme 11 Representative synthetic routes to pyrazole derivative 32 from glucose and phenylhydrazine via glucose phenylosazone 31. Adapted with permission from ref. 157, Copyright © 2007, Taylor & Francis. |
As shown in Scheme 12, starting from phenylhydrazine 33, β-ketoesters 34, and aromatic aldehydes 35 (e.g., vanillin and syringaldehyde) pre-prepared from lignin by thermal treatment with 2 mol L−1 NaOH and nitrobenzene (170 °C, autogenous pressure 1 MPa, 2 h) in a stainless steel autoclave, a diversity of 4,4′-arylmehtylenebis(1H-pyrazole-5-ol)s 36 with 79–88% yields are obtained in water under microwave irradiation (300 W, 100 °C, 10 min).160 In comparison with standard drug (Trolox), all the prepared bispyrazoles exhibit good radical scavenging activities of 2,2′-azino-bis(3-ethylenzothiazoline-sulphonic acid) diammonium salt (ABST+) and N,N-diphenyl-N′-picrylhydrazyl (DPPH). Given that the environmentally benign synthetic procedures developed by Yang et al.,160 these bispyrazoles derivatives have promising potential in the therapy of free radical-related diseases (e.g., tumors).
Scheme 12 Synthetic routes to 4,4′-arylmehtylenebis(1H-pyrazole-5-ol)s 36 from phenylhydrazine 33, β-ketoesters 34, and lignin-derived aromatic aldehydes 35. Adapted with permission from ref. 160, Copyright © 2012, Wiley-VCH. |
Other bio-based molecules (e.g., 1,3-dicarbonyl compounds, vinylogous formamides, and 1,3-diols) besides sugars or lignin derivatives can be used to produce pyrazoles.161,162 The 1,3-dicarbonyls are commonly employed as feedstocks to construct pyrazoles by condensation with a aryl or alkyl hydrazine, while their low stability restricts their applicability for obtaining substituted pyrazoles.163,164 On the other hand, 1,3-diacetals or vinylogous formamides 37 by reaction with hydrazines 38 allow access to substituted pyrazole motifs 39 (path A, Scheme 13), while multistep synthesis procedures were prerequisite due to the existence of sensitive functional groups.165 Much attention has been paid to the development of more efficient routes for substituted pyrazoles. For instance, Schmitt et al. reported that 1,3-diols 40 used instead of the typical masked dialdehydes enable the production of pyrazoles via hydrogen transfer process promoted by a Ru-complex, RuH2(PPh3)3CO/xantphos (path B, Scheme 13).166 The scope of both the diol and hydrazine components can be extended with good compatibility, and satisfactory pyrazole yields (84%) from scaled-up isopropyl diol (1 g, 8.5 mmol) and phenylhydrazine (8.5 mmol) in toluene at 110 °C after 24 h,166 however, unreliable regioselectivity of the obtained 3-alkyl pyrazoles needs to be resolved in follow-up studies. Also, the initial acquisition of starting materials from biomass generally involves complex processes, which would be another primary task needing to take into consideration.
Scheme 13 Synthetic routes to pyrazoles from masked dialdehydes 38 or diols 40. Adapted with permission from ref. 166, Copyright © 2015, American Chemical Society. |
Scheme 14 Synthetic routes to 4-tetrahydroxy-butyl imidazole 41 from glucose and formamidine. Adapted with permission from ref. 172, Copyright © 2016, Thieme Chemistry. |
In a one-pot process, a variety of mono- and disaccharides react with amidines 42 in molten ammonium carbonate to give tetrahydroxybutyl substituted imidazoles 43 with varying glycosylation patterns, depending on the type of saccharide substrates (Table 1).173 After purification by silica gel chromatography (eluent, 3:1 (v/v) ethanol:ammonia), moderate isolated yields (25–50%) of the imidazole products 43 are obtained from sugars like fructose, glucose, isomaltulose, melibiose, leucrose, maltose, cellobiose, and lactose (Table 1) after heating at 65 °C to 80 °C until reaction completion as monitored by thin-layer chromatography.173 The reaction system uses a benign solvent (molten ammonium carbonate), it does not require protection groups in its chemistry, and is thus sustainable and has great potential to replace methods based on petroleum resources.
Substrate | R | Product | R4 | R3 | R2 | R1 | Yield (%) |
---|---|---|---|---|---|---|---|
a Formamidine acetate. b Ethylacetimidate. c Benzamidine hydrochloride. | |||||||
Fructose | Ha | 2 | H | H | H | H | 50 |
Glucose | H | 2 | H | H | H | H | 47 |
Isomaltulose | H | 3 | α-D-Glc | H | H | H | 49 |
Isomaltulose | Meb | 4 | α-D-Glc | H | H | H | 30 |
Isomaltulose | Phc | 5 | α-D-Glc | H | H | H | 5 |
Melibiose | H | 6 | α-D-Glc | H | H | H | 38 |
Melibiose | Me | 7 | α-D-Glc | H | H | H | 27 |
Leucrose | H | 8 | H | α-D-Glc | H | H | 38 |
Leucrose | Me | 9 | H | α-D-Glc | H | H | 26 |
Maltose | H | 10 | H | H | α-D-Glc | H | 28 |
Cellobiose | H | 11 | H | H | β-D-Glc | H | 25 |
Lactose | H | 12 | H | H | β-D-Glc | H | 40 |
Several other synthetic approaches were also developed to produce substituted imidazoles from sugars (Scheme 15).174–176 For example, arabinose-derived aldehyde 44 undergoes condensation with glyoxals 45 in methanol/NH3 mixtures to yield linear imidazole sugars that can be further converted to the imidazole 46 after the removal of the trityl group with HCl in dioxane (Scheme 15A).174 The reaction of O-acetylated glucoseamine 47 with o,o′-disubstituted arylisothiocyanates 48 gives 49, which when followed by treatment with acetic anhydride in pyridine and subsequent elimination of acetic acid affords the imidazoline thion 50 (Scheme 15B).175 In another manner, direct introduction of imidazole ring is achieved through adding 2-lithio1-[(dimethylamino)methyl]-1H-imidazole 52 to 2,3,4,6-O-benzyl-glucono-1,4-lactone 51, yielding sugar-derived imidazole 53 (Scheme 15C).176
Scheme 15 Synthetic routes to imidazole derivatives from sugars: (A) L-arabinose, (B) O-acetylated glucoseamine 47, and (C) 2,3,4,6-O-benzyl-glucono-1,4-lactone 51. Adapted with permission from ref. 174–176, Copyright © 1995 & 2006, Wiley-VCH; Copyright © 2015, Elsevier. |
The 2-haloenones 54 derivable from sugars were deemed as crucial intermediates for construction of heterocycles.177,178 In the absence of any ligand or metal catalyst at ambient temperature, Mal and Das showed that a diversity of chiral hydroxy imidazoles 55 and furans 56 were able to be synthesized from 2-haloenones 54 with amidines or 1,3-dicarbonyl compounds, respectively, over K2CO3 in DMSO at room temperature (Scheme 16) through sequential Michael addition, cyclization, and sugar-ring opening reactions.179 Notably, the benign reaction system is tolerant to the substrates with variable substituents, affording chiral substituted imidazoles 55 with moderate to good yields (43–88%) and excellent regioselectivity (single in most cases) at room temperature within 45–120 min. It is worth mentioning that the yields of imidazoles obtained from sugars are relatively low (typically <50%), possibly due to the complicated reaction processes, which has to be improved by developing more efficient catalyst systems.
Scheme 16 Synthetic routes to either chiral hydroxy imidazoles 55 or furans 56 from 2-haloenones 54. Adapted with permission from ref. 179, Copyright © 2016, American Chemical Society. |
Similar to the construction of the furan-thiazole conjugated scaffold, furyloxazole 63 can be obtained (52% yield) via a one-pot cascade dehydrative condensation and oxidation of bio-based furfural and serine methyl ester 62 (Scheme 18).182 At room temperature, the condensation of furfural with serine methyl ester 62 mediated by K2CO3 and MgSO4 in N,N-dimethylacetamide (DMA) affords oxazolidine 64 in ring-chain tautomers (64a and 64b), followed by oxidation over BrCCl3/DBU (1,8-diazabicyclo(5.4.0)undec-7-ene) to give 2,5-dihydrooxazole 65. The resulting intermediate is then subjected to isomerization and a second oxidation, ultimately yielding the furyloxazole 63.182
Scheme 18 Synthetic routes to furyloxazole 63 from serine methyl ester 62 and furfural. Adapted with permission from ref. 182, Copyright © 2010, American Chemical Society. |
The readily available feedstock 2,5-furandicarboxylic acid (FDCA, 66) can be derived from cellulosic biomass-derived HMF (17) via oxidation.183 Selvakumar et al. demonstrated that FDCA (66) can be used for producing chiral bisoxazolines 69 (Scheme 19).184 Initially, FDCA (66) is transformed into its acyl chloride 67, followed by condensation with chiral substituted amino alcohols to yield relevant amides 68 (32–69% yields), which are eventually cyclized to generate the bisoxazolines 69. Furthermore, other five-membered N-heterocycles such as 1,2,3- and 1,2,4-triazoles, oxadiazoles, thiadiazoles, and tetrazoles can be synthesized from sugars in modest yields.185 However, most reaction systems involved lack stereo- or enantioselectivity, and need to be updated to use green synthetic methods (e.g., photo- and electroinduction) as such pathways are highly desirable for concise synthesis of these N-heterocycles.186–188
Scheme 20 Synthetic routes to pyridines (71 and 72) from acrolein 70 and NH3. Adapted with permission from ref. 198 and 199, Copyright © 2019, Springer Nature; Copyright © 2011, Wiley-VCH. |
Via a three-step sequential reaction process, 3-pyridinol 74 can be synthesized from xylose or even pentosan involving acid-catalyzed hydrolysis/dehydration to furfural, reductive amination to furfurylamine 73, and succedent 30% H2O2-mediated oxidation in 3 mol L−1 HCl (Scheme 21A).208 In contrast, more laborious procedures are required for the upgrading fructose (or hexosan) to 6-hydroxymethyl-3-pyridinol 77 (Scheme 21B), where N- and O-blocking (75 → 76) of HMF 17 are prerequisite prior to proceeding with electrocatalytic oxidation in methanol, thus leading to tandem deacetylation, acetal-hydrolysis, and cyclization to form the pyridine framework (85% yield of 77).208 In related ways, 2-(hydroxymethyl)-5-(aminomethyl)-furan 75 can be converted to 6-methyl-3-pyridinol 78 with a high yield of 88% via cascade hydrolysis and cyclization catalyzed by HCl under reflux (Scheme 21C).209
Scheme 21 Synthetic routes to pyridinols (74, 77 & 78) from furfuryl amines: (A) 73, (B) 75 & 76, and (C) 75 derived from pentose and hexose sugars. Adapted with permission from ref. 208, Copyright © 1998, Elsevier. |
Scheme 22 Synthetic routes to pyrazines 81via (A) inter- and intramolecular dehydrogenative self-coupling of amino alcohols and (B) intramolecular cyclization dehydrogenation of hydroxyl diamines. Adapted with permission from ref. 214, Copyright © 2003, Elsevier. |
Starting from glyceraldehyde 83 or 1,3-dihydroxyacetone 84 which can be derived from biomass, Song et al. found that 2-hydroxymethyl-5-methylpyrazine 87 is obtained in good yields (ca. 72%) by reacting 87 in the presence of diammonium phosphate at 90 °C for 1 h in basic mixture solvent of water and dioxane (pH = 8.0–9.1).218 The 2-imino-1,3-propanediol 85 was identified to be a key reaction intermediate, which might be formed by ketimine condensation of 1,3-dihydroxyacetone 84 with NH3 generated in situ from diammonium phosphate, followed by cyclization to 86 and final dehydration to yield the pyrazine product 87 (Scheme 23).218 Among these three elemental steps, the rate-determining step is the cyclization reaction, which is possibly ascribed to the lack of active catalytic sites.
Scheme 23 Plausible pathways for synthesis of 2-hydroxymethyl-5-methylpyrazine 87 from diammonium phosphate and 1,3-dihydroxyacetone 84 at hydrothermal conditions. Adapted with permission from ref. 218, Copyright © 2017, Royal Society of Chemistry. |
In the presence of a tungsten-based catalyst (ammonium metatungstate), Chen et al. reported that glucose in 25% aqueous ammonia is transformed directly into 2-methyl pyrazine 89 in moderate yields (ca. 25.6%) within 15 min at 180 °C via tandem fragmentation and cyclization in a single pot (Scheme 24).219 Prior to fragmentation, β-D-glucopyranosylamine 88 formed from condensation of glucose and NH3 was identified as one of the important intermediates. The reaction system was remarkably facilitated by tungsten clusters (e.g., [HW2O7]− and [W4O13]2−), with 7.2–23.3% yields of 2-methyl pyrazine 89 being achieved from monosaccharides (e.g., fructose, xylose, chitin monomer, glucosamine) and several disaccharides (e.g., maltose and cellulose) under otherwise identical reaction conditions.219 From the viewpoint of sustainability, it would highly desirable to further develop more robust and active catalysts for using tandem fragmentation and cyclization methodologies.
Scheme 24 Possible pathways for the synthesis of 2-methyl pyrazine 89 from glucose and NH3 over ammonium metatungstate. Adapted with permission from ref. 219, Copyright © 2017, American Chemical Society. |
Under basic conditions (MeONa in methanol), nucleophilic attack of the amidinium reagent (i.e., acetamidinium, benzamidinium and guanidinium salts) takes place at the C1 position of 2-formyl-galactal 93, followed by cyclization to give the substituted 5-(1,2,4-tri-O-benzyl-D-lyxo-1,2,3,4-tetrahydroxy-butyl)pyrimidines 94 (Scheme 26A).222 Likewise, substituted triazines 97 can be prepared at 37 °C by condensation of aminoguanidine 96 with two 3-deoxy-1,2-dicarbonyl derivatives 95, including 4-O-acetyl-1-deoxy-5,6-O-isopropylidene-2,3-D-threo-hexodiulose 95a, and 4-O-acetyl-1-deoxy-5,6-O-isopropylidene-2,3-D-erythro-hexodiul 95b (Scheme 26B).223 In contrast to their significance, development of more pragmatic strategies for the sustainable production of six-membered N-heterocycles from biomass resources is urgently required.
Scheme 26 Synthesis of (A) pyrimidines 94 from 93, and (B) triazines 97 from dicarbonyl derivatives 95. Adapted with permission from ref. 222 and 223, Copyright © 1995 & 2002, Taylor & Francis. |
Scheme 27 Representative pathways for upgrading furans to azaheterocycles via (A) Yuriev reaction, (B) Butin reaction, (C) Achmatowicz reaction, and (D) oxidative rearrangement. Adapted with permission from ref. 231, Copyright © 2016, American Chemical Society. |
Zeolite HZSM-5 (Si/Al = 25) was illustrated to efficiently catalyze direct gas-phase conversion of furfural via cascade thermal conversion and ammonization with NH3 to give indoles (yield 20.79%) at 650 °C with a weight hourly space velocity (WHSV) of 1.0 h−1 and NH3/furfural molar ratio of 2. It was proposed that the reaction between furfural and ammonia gives furfuryl imine 100, followed by cracking reaction to furan 101 which is considered as the key intermediate leading to pyrrole 102 and ultimately to indoles 103, with pyridines, aniline, and benzenes being concurrently generated as byproducts (Scheme 28).232 At moderately high temperatures (500 °C), moderate yields of indoles 103 (ca. 32%) are obtained from furan 101 under otherwise identical reaction conditions,233 indicating that the elemental step involving decarbonylation of furfural to furan 101 is crucial to the overall reaction. It is interesting that NH3 diluted by N2 is able to inhibit the formation of coke derived from 2-furonitrile 104, thus significantly increasing the yield of indoles 103 (33%) through enhancement of the HZSM-5 stability.234
Scheme 28 Possible pathways toward indoles 103 directly from furfural over HZSM-15. Adapted with permission from ref. 232, Copyright © 2015, Elsevier. |
Bio-based diols and anilines are applicable to the production of indoles under thermal conditions (e.g., 350 °C) or with metal catalysts at relatively low temperatures (e.g., 175 °C).235,236 The combined use of Pt/Al2O3 and ZnO catalyzes the dehydrogenation of ethylene glycol 105 to glycolaldehyde 106, followed by condensation with anilines to the imines 107 that undergo tautomerism and acylation/elimination reaction to afford relevant pyrrole-ring unsubstituted indoles 108 (Scheme 29).236 In view of H2 and H2O being the sole co-products, this type of atom-efficient reaction systems, in combination with stable and lost-cost catalysts, represents an efficient and promising synthetic approach toward indoles.
Scheme 29 Synthetic routes to indoles from anilines and ethylene glycol 105 involving acceptorless dehydrogenative condensation. Adapted with permission from ref. 236, Copyright © 2018, American Chemical Society. |
Scheme 30 Representative synthetic routes to benzimidazoles 111via (A) classical methods or (B) acceptorless dehydrogenative coupling strategies with solvent or (C) acceptorless dehydrogenative coupling strategies without solvent. Adapted with permission from ref. 247, Copyright © 2013, American Association for the Advancement of Science. |
Among the developed strategies for condensation of diamines 109 with alcohols 112 to construct azaheterocycles like benzimidazoles 111, acceptorless dehydrogenative coupling (ADC) has emerged as an attractive protocol with only molecular H2 and innocuous H2O formed as the co-products (Scheme 30B & C).247,248 However, the requirement of a homogeneous noble-metal-based complex with the co-addition of a strong base are prerequisite to ensure smooth reaction progress.249,250 To make the ADC chemistry sustainable, several earth-abundant metal catalysts such as copper-doped porous metal oxides, cobalt–pincer complexes, and magnetic nanofibers confined in carbon nanotubes have been explored, and are capable of promoting efficient production of 2-substituted benzimidazoles 111 (up to 99% yield) from dehydrogenative coupling of primary alcohols 112 and aromatic diamines 109 under base-free conditions.251–253 Sun et al. disclosed that readily available 2-nitroanilines 113 can be used as nitrogen sources despite the requirement of an additional hydrogenation step prior to cyclization, such that comparable yields of benzimidazoles 111 are obtained by reaction with primary alcohols 112 at 250 °C for 3 h to 5 h reaction time.251
Besides monohydric alcohols, polyhydric alcohols have the possibility to form azaheterocycle frameworks together with the benzimidazole ring. Climent et al. showed that glyceraldehyde 83 (or glycerol) undergoes oxidation–cyclization twice with o-phenylenediamine derivatives 109 to give various substituted benzimidazoylquinoxalines 115 (24–80% yields) catalyzed by ceria supported gold nanoparticles (Au/CeO2) using air as the oxidant at 140 °C for 24 h (Scheme 31).254 In the reaction pathway, glyceraldehyde 83 (possibly in situ generated from glycerol by oxidation) first proceeds coupling and cyclization with o-phenylenediamine derivatives 109 to produce the intermediate benzimidazol 114, followed by cascade oxidation–cyclization with another o-phenylenediamine 109 to afford the substituted benzimidazoylquinoxalines 115. The electron-donating substituents in the phenyl ring favor the competitive oxidative cleavage of the diol 114, and lead to a decrease in the product selectivity by formation of other azaheterocycles 116–119 (Scheme 31).254
Scheme 31 Possible pathways for the formation of benzimidazoylquinoxalines 115 and byproducts 116–119. Adapted with permission from ref. 254, Copyright © 2013, Wiley-VCH. |
Scheme 32 Synthetic routes to quinolines from amino alcohol 120 reacting with (A) ketone 121 or secondary alcohol 122, (B) nitrile 124, or (C) nitrile 124 and alcohol 112. Adapted with permission from ref. 258, 259 and 270, Copyright © 2018 & 2019, Wiley-VCH; Copyright © 2007, Elsevier. |
Synthesis of quinoxalines was traditionally performed by double-coupling of 1,2-carbonyls with 1,2-phenylenediamines, while the employed 1,2-carbonyls were highly reactive and cause undesired self-condensation.271,272 Several other synthetic approaches such as oxidative coupling of α-hydroxycarbonyls and 1,2-diamines,273,274 1,4-addition of diazenylbutenes with 1,2-diamines,275 oxidation trapping of epoxides and ene-1,2-diamines,276 and cyclization–oxidation of phenacyl bromides and 1,2-phenylenediamines277 can provide substituted quinoxalines, but with moderate or low yields in most cases. Using biomass-derived glycols or vicinal diols 128 with 1,2-phenylenediamine derivatives 109 as starting materials, Climent et al. reported that quinoxalines 130 (35–91% yields) are efficiently synthesized in a one-pot two-step oxidative coupling reaction process over ceria supported gold nanoparticles (Au/CeO2) at 140 °C after 24 h in the absence of any homogeneous base with air as oxidant (Scheme 33).278 Notably, the oxidative cleavage of the substrate diol 128 or the hydroxycarbonyl intermediate 129 inevitably took place to form aldehydes 131 and 132, which further undergo condensation with the phenylenediamine 109 affording the benzimidazole derivatives 133, 134, 135, and 136. Similarly, starting from 1,2-dinitrobenzene and 1,2-propanediol, yields of up to 83% of 2-methylquinoxaline are realized, where the nitro-to-amino reduction is performed at 80 °C under 1 MPa H2, followed by oxidative coupling at 140 °C for 30 h in the presence of atmospheric air.278 The quinoxaline skeleton can also be established by tandem cyclization/hydrosilylation of 1,2-phenylenediamines and keto esters,279 and coupling of 2-(1H-pyrrol-1-yl)anilines with DMSO.280 Both of these latter methods provide sustainable approaches to quinoxalines since they use bio-based feedstocks although the protocols could be improved by possibly using earth-abundant metal catalysts or safe or renewable solvents.
Scheme 33 Reaction pathways for the formation of quinoxalines 130 and byproducts 133–136 from vicinal diols 128 and 1,2-phenylenediamines 109. Adapted with permission from ref. 278, Copyright © 2012, Elsevier. |
As illustrated in Scheme 34, carbohydrate substrates (e.g., glucose, galactose, arabinose, xylose, maltose, mannose, lactose) combined with an amine (e.g., anilines, glucosamine) and barbituric acid 137 or thio-barbituric acid 138 as reagents can be used to form pyrimidine-fused heterocycles 139 (with up to >90% yield) via a one-pot multicomponent condensation reaction in the presence of an acid catalyst (e.g., para-toluenesulfonic acid, nanocrystalline cellulose sulfuric acid) under mild conditions (e.g., 50 °C).281–283 The active aldehyde group of the sugar takes part in the condensation reaction, which gives the resulting pyrimidine-fused heterocycles hydrophilicity and potentially good bioactivity.281–283 All together with glucosamine, aldehyde, and barbituric acid, addition of malononitrile is able to allow a four-component condensation reaction, giving polyhydroxy-substituted pyrido[2,3-d]pyrimidines in good yields (89–94%) catalyzed by para-toluenesulfonic acid in ethanol heated at 50 °C.284
Scheme 34 Synthesis of pyrimidine-fused heterocycles 139via multicomponent condensation of sugar, amine, and barbituric acid 137 or thio-barbituric acid 138. Adapted with permission from ref. 281, Copyright © 2018, Springer Nature. |
Multicomponent condensation reactions can be used for the synthesis of pyridine-fused heterocycles. For example, Shpuntov et al. showed that a three-component Mannich-type reaction of 2-alkylfuran, methyl 2-formylbenzoate, and carbamate in the presence of iodine at 0 °C for 2 h affords N-Boc-1-[2-(carbomethoxy)aryl]furfurylamines in moderate yields (51–89%), and further application of a two-step reaction process of oxidative furan-ring cleavage and N-Boc deprotection catalyzed by meta-chloroperoxybenzoic acid and HCl, respectively, gives 6H-isochromeno-[4,3-b]pyridin-6-ones in moderate yields (61–68%).285 The coupling of chitin-derived 3-acetamido-5-acetylfuran 140 with ketones 141 catalyzed by HCl provides dihydrodifuropyridines (up to 64% yield) after reaction at 70 °C for 16 h (Scheme 35).286 The ketone 141 initially reacts with 140 to form difurylmethane 142, followed by sequential acid-mediated acetamide hydrolysis, tautomerism, and intramolecular cyclisation, eventually giving dihydrodifuropyridine 143 after removal of ammonia (Scheme 35), as proposed by those authors.286
Scheme 35 Reaction pathways for the synthesis of pyridine-fused heterocycles 143 from chitin-derived 3-acetamido-5-acetylfuran 140 and aliphatic ketones 141. Adapted with permission from ref. 286, Copyright © 2018, Springer Nature. |
Considering that aryl aldehydes are available from both lignin and carbohydrate components, efforts have been made to prepare dihydropyrano[2,3-c]pyrazoles from biomass-derived aldehydes via multicomponent condensation reactions. For example, Yang et al. reported a one-pot, two-step catalyst-free protocol for synthesis of 2H,4H- and 1H,4H-dihydro-pyrano[2,3-c]pyrazoles from lignin-derived aromatic aldehydes 35 (Scheme 36A).287 In the first stage, microwave-assisted condensation of hydrazines 37 and β-ketoesters 144 in water at 80 °C in 2 min affords intermediates (pyrazolones, 145). To the resulting mixture, malononitrile 146 and aromatic aldehydes 35 promptly added and heated under the identical irradiation conditions for another 3 min gives dihydropyrano[2,3-c]pyrazoles 147 (48–95% yields) by direct precipitation from the reaction mixture after cooling to room temperature.287 In the replacement of β-ketoesters with acetylene ester, Ambethkar et al. illustrated that dihydropyrano[2,3-c]pyrazoles (65–93% yields) are directly synthesized through four-component condensation of diethylacetylene dicarboxylate, hydrazine hydrate, aryl aldehydes, and malononitrile in the presence of L-proline under solvent-free and mechano (grinding) conditions.288 In the single pot one-step reaction process (Scheme 36B), diethylacetylene dicarboxylate 148 is initially condensed with hydrazine hydrate 149 to give intermediate 150, followed by Michael addition to intermediate 151 that is in situ formed from Knoevenagel condensation between aryl aldehyde 131 and malononitrile 146, while the dihydropyrano[2,3-c]pyrazole product 152 is afforded after undergoing subsequent intramolecular cyclization and tautomerization.288 On the whole, this synthetic protocol seems to be facile, efficient, and eco-friendly for producing dihydropyrano[2,3-c]pyrazole derivatives.
Scheme 36 Synthesis of dihydropyrano[2,3-c]pyrazoles 152 from biomass-derived aldehydes 35via a (A) two- or (B) one-step approach. Adapted with permission from ref. 287 and 288, Copyright © 2014 & 2015, Elsevier. |
Besides the above-mentioned synthetic methods employed for synthesis of fused azaheterocycles, several other strategies have been developed with the aim of constructing more specific N-heterocyclic skeletons amendable to biomass feedstocks. For instance, a furan ring opening-pyrrole ring closure strategy has been adopted for the synthesis of pyrrole-fused heterocycles such as pyrrolo[1,2-a][1,4]diazepines and 1,2,3,4-tetrahydropyrrolo[1,2-a]pyrazin-3-ones.289,290 In view of the feasibility of furan-ring opening, biofuranics and their derivatives (e.g., levulinic acid) are extensively used for the construction of complex azaheterocycles like indolo[3,2-c]quinolines and bicyclic heterocycles.291,292 Direct transformation of biomass sugars to azaheterocycles has been investigated with profound results,293,294 while most studies of upgrading lignin are more focused on its model molecules such as 2-phenoxy acetophenone and aromatic aldehydes via modified approaches based on conventional reactions.295,296
Scheme 37 CO2-incorporated annulation triggered by C, N, or O nucleophiles. Adapted with permission from ref. 299, Copyright © 2019, American Chemical Society. |
In the presence of a hydrogen source (e.g., H2, boranes, and silanes), CO2 can be subjected to in situ reduction followed by cyclization with N-containing nucleophiles to afford a wide range of azaheterocycles.299 The 2-electron reduction is able to promote the hydrogenation of CO2 to formates, followed by condensation with substrates containing two nucleophilic sites to furnish quinazolinones, benzimidazoles, formamidines and their derivatives.305–307 On the other hand, CO2 may also serve as the methylene species via 4-electron reduction, giving saturated N-heterocycles after successive nucleophilic attacks.299 Although studies on CO2 reduction have been investigated extensively, controllable CO2 hydrogenation combined with successive cyclization still remains a challenge.
Fig. 3 Schematic illustration for the synthesis of N-heterocycles with controllable ring size from imine esters 153. |
Cycloamination strategies are commonly used for the construction of azaheterocycles that can be well controlled by adjusting the type of nitrogen source. Starting from 1,4-dicarbonyl compounds 1 that are readily available from oxidative opening of the furan ring, pyrroles 154, pyridazines 155, and diazepines 156 can be selectively obtained by diamination with NH3, N2H4, and o-phenylenediamine, respectively (Scheme 38).316 In a similar way, reduced sugars proceeding via 1,2-dicarbonyl intermediates by reaction with different nitrogen sources furnish the quinoxaline, 1,2,4-triazine, pyrazine and pyrazolo[3,4-b]quinoxaline skeletons.317 In these strategies, protection group chemistry is not needed for the in situ multi-step transformations. With respect to the synthesis of azaheterocycles via multicomponent reactions, selection of subcomponents allows control of the product distributions. For instance, coupling cyclization of amino alcohols and alcohols with desired functionalities with suitable spatial distances is able to selectively afford pyrrole, pyridine, quinoline, pyrazine, carbazole, or acridine derivatives.318–322
Scheme 38 Reaction routes to pyrroles 154, pyridazines 155 and diazepines 156 from sugars by adjusting the type of nitrogen source. Adapted with permission from ref. 316, Copyright © 2001, Royal Society of Chemistry. |
Chiral N-heterocyclic compounds have a privileged role in pharmaceuticals.323 A specific chiral source of asymmetric induction (i.e., chiral catalyst) is typically required for the enantioselective synthesis of N-heterocycles,324,325 through which stereoselective upgrading of thermodynamically stable CO2 to optically pure azaheterocycles can be realized.326 Biomass derivatives (e.g., carbohydrates, amino acid, and levoglucosenone) are inherently in enantiopure form, thus they have high feasibility for use in simple production of N-heterocycles with appropriate configuration. For example, in the absence of any chiral catalyst, cellulose-derived (−)-levoglucosenone can undergo well-controlled reactions like 1,3-dipolar cycloaddition, aza-Michael addition or isomerization, being enantioselectively converted into polysubstituted pyrrolidines or 1,2,3-triazoles.327–329 A versatile protocol via multicomponent Ugi-type reactions of unprotected saccharides and L- or D-configured amino acids is accessible to 1,2-syn or anti configured products, respectively.330 In this regard, the exploitation of synthetic protocols to biomass-derived chiral molecules, especially those with potent bioactivities, is promising with respect to green manufacturing processes.
Pyrolysis or hydrothermal liquefaction of lignocellulosic biomass with NH3 has been developed for the production of nitrogenous heterocyclic compounds, while obtained products are mixtures of pyrrole, pyridine, indole and other nitrogen-containing compounds with relatively low overall yields (<30%).336–343
Direct thermal conversion of nitrogenous biopolymers (e.g., chitin and proteins) can be implemented by sub- and supercritical water treatment, where the overall hydrolysis process including surface hydrolysis and destruction of hydrogen and glycosidic bond is controlled by solvent conditions.344 Mechanochemical treatment (e.g., ball mill) of chitin enhances its solubility in supercritical water (400 °C) allowing rapid (1 min) transformation, while raw chitin with high crystallinity shows poor dissolution in high-temperature water (HTW).345 After pretreatment, chitin could be transformed into N-acetyl-D-glucosamine (the monomer) and/or N,N′-diacetylchitobiose (the dimer) with yields of no more than 8%.346,347 At relatively high temperatures of (200 to 260) °C, treatment of proteins with HTW and fast heating rates (135 to 180) K s−1 inhibits initial protein aggregation that occurs during the reaction process at slow heating rates (ca. 0.25 K s−1), affording high-molecular-weight peptides (1500 to 8300) Da as dominant products with the transformation advantageously occurring as random-scission.348
On the other hand, with hydrothermal liquefaction (HTL) at temperatures between 250 °C and 350 °C, several heterocycles could originate from the components of proteins, such as piperidines and quinolines from lysine in bio-oils, while pyrazines form from the mixture of disaccharides and lysine.349 Besides organic matter derived from HTL with liquid-N accounting for the majority of nitrogenous products,350–353 the HTL conditions can affect the concentrations of metals (e.g., Fe, Na, Mg, Ca) and inorganic species (e.g., P) in algal biocrude.354 It is worth noting that N-containing heterocycles partition undesirably into bio-crude oils after hydrothermal treatment.355,356 leading to a future challenge for the reaction system or product separation.
The use of catalysts (e.g., HZSM-5) in the reaction system remarkably increases the selectivity toward the N-containing compounds composed of up to 79% pyrroles, 63% pyridines, and 57% indoles, respectively.357 By selecting crude shrimp shells as the starting feedstock, pyrrole is detected as the dominant nitrogen-containing product after hydrothermal treatment with NH3 in the presence of NaOH,358 while the co-production of N-heterocyclic bio-char and other aromatic amines are unavoidable in most cases.359 Appropriate design of reaction processes and functional catalysts opens avenues to other elegant N-containing chemicals like β-lactam, nitrile, and isoxazole.360–362 Last, but not least, adopting new strategies for C–N bond formation are required for sustainable and efficient valorization of raw biomass materials.
(1) Organic synthesis and synthetic methodologies may provide efficient references for rationally devising effective and renewed reaction routes toward specific N-heterocycles with satisfactory stereo- or enantioselectivity, while the reaction processes can be further simplified by following the principles of green engineering and green chemistry. In contrast, studies on sustainable production of other heteroatom (S, P)-containing compounds are relatively scarce, indicating the necessity to explore synthetic approaches in the construction of S–C or P–C bonds.
(2) The presence of a suitable catalyst with designated chiral induction center can remarkably lower reaction barriers with enhanced overall rate and optical purity. Basic nitrogen sources or used together with acidic substrates can themselves act as catalyst, and most of biomass derivatives naturally bearing absolute/spatial configurations have the greatest potential to furnish the desired N-containing enantiomers without any catalyst.
(3) Hydrothermal treatment of raw biomass materials with ammonia has been demonstrated to be capable of giving N-containing compounds but with low selectivity, let alone the desired relevant azaheterocycles. In this regard, biomass pretreatment with sustainable solvents and/or ecofriendly catalysts can be a promising auxiliary to enhance reaction efficiency with respect to both selectivity and yield of azaheterocycles.
(4) Definite reaction mechanisms have been explicitly illustrated for some of the above-mentioned reactions, while a majority of reaction schemes have been revealed without neither elucidation of the reaction pathway nor the catalyst structure–activity relationship. There is no doubt that the disclosure of detailed reaction routes using modern characterization techniques will be helpful for both catalyst preparation and design of reaction processes for producing N-heterocycles.
(5) Raw biomass materials are solid-state and are not soluble in thermal reaction processes, which often results in the formation of plentiful solid biochars rather than the desired N-heterocyclic compounds. In this respect, the design of compatible reactors is needed to facilitate production in overall cycloamination processes.
In conclusion, cycloamination coupled with C–C bond formation processes is efficient for the synthesis of N-heterocycles especially five-membered azaheterocycles from biomass feedstocks. Many endeavors have been made with great achievements being made that improve the chemical avenues to bio-based N-heterocycles. Studies on the diversity of the obtained N-containing compounds need to be further strengthened. Development of efficient and renewed synthesis approaches to task-specific N-heterocyclic compounds with emphasis on sustainable chemistry will be required for comprehensive biomass valorization.
This journal is © The Royal Society of Chemistry 2020 |