Recent advances in nucleophile-triggered CO₂-incorporated cyclization leading to heterocycles

Abstract

Carbon dioxide (CO₂) has emerged as a sustainable, feasible, abundant one-carbon synthon and displays great potential in the synthesis of heterocycles such as lactones, lactams, and 2-oxazolidinones, which are privileged motifs in pharmaceutical chemistry demonstrating bioactivities. Although the fixation of CO₂ is restricted due to its thermodynamic stability and kinetic inertness, multiple breakthroughs have been realized in annulation chemistry. This review concentrates on the advances made in the last five years in CO₂-incorporated cyclization triggered by N-, O-, and C-nucleophiles. Three transformation modes of CO₂ including carboxylative cyclization, carbonylative cyclization, and reductive cyclization have been summarized. Moreover, typical mechanisms and significant applications of these reactions are also described.

---

Sheng Wang

Sheng Wang obtained his BSc degree in Chemistry from Shandong University in 2014. He is currently a PhD candidate under the supervision of Professor Chanjuan Xi at Department of Chemistry, Tsinghua University. His research interests include CO2 activation and its application in synthetic chemistry.

Chanjuan Xi

Chanjuan Xi is a Professor of Department of Chemistry, Tsinghua University. She graduated from Lanzhou University in 1986. She worked at Lanzhou Institute of Chemical Physics, CAS (1986–1995). In the course of her work, she earned a Master's degree in Chemistry in 1994. In 1996, she joined Hokkaido University and earned her PhD in Pharmaceutical Sciences in 1999, and spent one year as a postdoctoral fellow in Catalysis Research Center. In 2000, she joined Tsinghua University as Associate Professor and then she was promoted to a full Professor in 2005. Her research interests include organometallics and sustainable organic chemistry.

---

1. Introduction

The utilization of CO₂ as a one-carbon source to prepare value-added chemicals has received considerable attention.¹ Such interests initially arose as a consequence of fossil fuel depletion and the continuously increasing emission of CO₂ caused by human behavior. In this context, CO₂ has been regarded as a promising one-carbon source to release the environmental pressure and moreover provide various benefits in organic synthesis such as safety, effectiveness, low toxicity, abundance, and sustainability. Despite these merits, recycling CO₂ into valuable molecules effectively is restricted due to its thermodynamic stability and kinetic inertness, which makes the majority of CO₂ lost as waste. Therefore, investigations concentrating on the utilization of CO₂ in chemical transformations are worthwhile and highly desirable.

Up to now, several strategies to overcome the thermodynamic stability and kinetic inertness of CO₂ have been developed. It is well known that CO₂ is a non-polar molecule with a delocalized π bond containing three centers and four electrons, which stabilizes the CO₂ molecule. To realize CO₂ transformation, the utilization of high-energy substrates such as organometallic reagents and 3-membered ring compounds is a straightforward method, while the addition of high-energy additives to trigger the reaction and provide energy for a catalytic cycle is also effective. Moreover, the application of catalysts, which could activate CO₂ or starting materials and accelerate the reaction by lowering the energy barrier of the transition state, is also key to achieve a thermodynamically favored reaction.

Based on the strategy mentioned above, diverse reactions of CO₂ have been reported. The central carbon serves as a weak electrophilic site due to the polar C=O double bond and displays excellent affinity towards strong nucleophiles, while the Lewis basic oxygen can donate electrons. As a result, reducing CO₂ in a straightforward manner to prepare formic acid, formaldehyde, methanol or methane by 2-electron to 8-electron reduction processes using suitable reductants has been developed. Both homogeneous metal catalysts with different ligands² and heterogeneous catalysts with controlled nano-structures³ realize reduction smoothly. More importantly, value-added carbonates, carbamates, carboxylic acids, and various sophisticated heterocycles containing “CO₂”, “CO”, “CH₂” or “CH” moieties are obtained in the presence of CO₂ and different substrates with amino groups, hydroxyl groups, and carbon nucleophiles. Among the diverse products, heterocycles are of key importance due to their synthetic applications and biological properties as significant motifs (Scheme 1).⁴ Consequently, CO₂-incorporated cyclization triggered by nucleophiles has been explored continuously in recent years as an effective and green approach to heterocycles.

Scheme 1 Selected biologically active heterocycles possessing different one-carbon moieties.

According to reported literature, CO₂-incorporated cyclization could be divided into three modes (Scheme 2): (I) nucleophilic attack on CO₂ followed by intramolecular cyclization to afford carboxylative cycles; (II) sequential nucleophilic attack on CO₂ with substrates containing two nucleophilic sites to give cyclic carbonyl derivatives; (III) nucleophilic attack on the reduced CO₂ followed by cyclization to generate cyclic derivatives. Nucleophiles such as amines, alcohols, and some carbon nucleophiles could be carboxylated by CO₂ in the presence of base, but the problem is that these unstable carboxylated products would decarboxylate and return back to starting materials after workup. Therefore, the key to success lies in fixation of CO₂, for instance, transformation of unstable carboxylated intermediates to a stable product or reducing stable CO₂ to more applicable one-carbon sources. To solve the problem, the substrate in mode I provides an electrophilic site to accept excess electrons from oxygen while mode II offers another nucleophile to attack the CO₂. In these two modes, the valence of central carbon is invariant. In mode III, depending on the electron acceptor properties of the central carbon, CO₂ could be switched to a more flexible and applicable one-carbon synthon serving as “CH”, “CO” or “CH₂” in the construction of heterocycles. Based on these reaction modes, various well-designed substrates and catalysis systems have achieved transformations of CO₂ into molecules as one-carbon building blocks to generate different heterocycles. This review endeavors to illustrate different elegant CO₂-incorporated cyclizations in the last 5 years initiated by appropriate nucleophiles to yield various heterocycles: (1) nucleophile-triggered CO₂-incorporated carboxylative cyclization; (2) nucleophile-triggered CO₂-incorporated carbonylative cyclization; (3) CO₂-incorporated reductive cyclization. The cyclization of aziridines and epoxides with CO₂ is not included as it is discussed in detail in previous works.⁵

Scheme 2 Reaction modes of CO₂-incorporated annulation triggered by nucleophiles.

2. Nucleophile-triggered CO₂-incorporated carboxylative cyclization

As an electrophile, CO₂ could be easily trapped by different strong nucleophiles with the formation of C–N, C–O or C–C bonds with the negative charge being transferred to the O atom of CO₂. Taking advantage of the resulting nucleophilic oxygen anions, well-designed substrates undergo tandem cyclization assisted by a catalyst or base to give various valuable chemicals such as 2-oxazolidinones, cyclic carbonates, and lactones.

2.1 N-Nucleophile-triggered carboxylative cyclization

Amines bearing unsaturated bonds such as allylamines and propargylamines are significant substrates for CO₂-incorporated synthesis of attractive 2-oxazolidinones.⁶ Although great progress has been made in the preparation of 2-oxazolidinones, the synthesis of functionalized 2-oxazolidinones is still attractive considering their versatility in pharmacy and materials science. To facilitate the annulation, strategies involving activation of unsaturated bonds and CO₂ have been fully applied.

In 2016, Yu and co-workers developed copper-catalyzed oxytrifluoromethylation of allylamine 1 with CO₂ (1 atm) and Togni's reagent (Scheme 3).⁷ CF₃-Containing 2-oxazolidinones 2 were prepared chemo-, regio-, and diastereoselectively with moderate to good yield. Notably, bulky trisubstituted alkenyl amines could also undergo oxytrifluoromethylation, which was extremely challenging. Although azotrifluoromethylation of allylamine was competitive, the great affinity between CO₂ and the amine controlled the reactivity of the amine and avoided undesirable byproduct formation. The addition of radical scavengers such as TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl) and BHT (butylated hydroxytoluene) generated TEMPO-CF₃ and BHT-CF₃, which proved the existence of ˙CF₃. Based on experiments and DFT calculations,⁸ a mechanism involving Cu(I)–Cu(II) catalytic cyclization was proposed. Initially, the amine coordinates with Cu(I) followed by deprotonation with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU). Subsequently, intermediate I-2 attacks CO₂ to afford copper carboxylated complex I-3. Then, the I-CF₃ bond undergoes homolysis to generate a ˙CF₃ radical and a Cu(II) species followed by the addition of a ˙CF₃ radical to the C–C double bond. Finally, intramolecular radical–radical cross-coupling takes place diastereoselectively to deliver spirocyclic product 3 with the release of the Cu(I) catalyst.

Scheme 3 Cu-Catalyzed carboxylative cyclization of allylamines with CO₂.

Recently, Yu and co-workers further developed copper-catalyzed trifluoromethylative dearomatization of indoles 4 and furans 6 with Togni's reagent and CO₂ (1 atm) in the presence of 1,5-diazobicyclo[4.3.0]non-5-ene (DBN) (Scheme 4).⁹ A variety of CF₃-containing spirocyclic indolines 5 and spiroacetals 7 were constructed. The observation of TEMPO-CF₃ in the ¹⁹F NMR spectrum after the addition of TEMPO as a radical scavenger indicated that the radicals ˙CF₃ played an important role in the reaction. Supported by DFT calculations, the reaction starts with the addition of an amine–copper complex to CO₂ in the presence of base followed by radical addition of ˙CF₃ to the C=C double bond.

Scheme 4 Copper-catalyzed carboxylative cyclization of indoles and furans with CO₂.

Light-driven reactions have been developed in recent years due to their high efficiency and mild conditions. He and co-workers disclosed photo-initiated perfluoroalkylation of allylamines 8 with CO₂ (1 atm) and perfluoroalkyl iodides at room temperature to afford oxazolidinones 9 in good yield (Scheme 5).¹⁰ Controlled experiments and ¹H NMR spectroscopy suggested a possible mechanism as shown in Scheme 5. The allylamine initially attacks CO₂ with the assistance of 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD). The perfluoroalkyl iodide undergoes homolytic cleavage to generate perfluoroalkyl radicals. Subsequently, addition of the perfluoroalkyl radicals to the alkene gives secondary carbon radicals I-7, which could deliver iodoperfluoroalkylated carbamate I-8via coupling with radical I˙ (path I). However, path II involving radical propagation of I-7 and another molecule of perfluoroalkyl iodide is feasible as well.

Scheme 5 Photo-initiated carboxylative cyclization of allylamine with CO₂.

More recently, Yu and co-workers realized visible-light photoredox-catalyzed oxy-difluoroalkylation of allylamines 10 with CO₂ (1 atm) to afford difluoroalkylated 2-oxazolidones 11 (Scheme 6).¹¹ This reaction showed good tolerance to functional groups and steric hindrance. Apart from bromodifluoroacetates, a variety of difluoroalkyl reagents such as bromodifluoroacetamides, triisopropylsilyl-protected gem-difluoropropargyl bromides and phosphonyldifluoromethyl bromides gave the corresponding product under suitable conditions (longer time or iridium catalyst). Radical inhibition experiments and radical-clock experiments were conducted, which indicated the existence of ˙CF₂CO₂Et in the cyclization step. A possible mechanism was proposed as shown in Scheme 6. Initially, 1,4-diazabicyclo[2.2.2]octane (DABCO) reductively quenches the excited ruthenium catalyst to give the vital Ru(I) species, which reduces BrCF₂COOEt to afford the radical ˙CF₂COOEt I-11. Then, radical addition of the intermediate I-11 to carbamate I-10 generates the benzyl radical I-12 followed by oxidation of excited Ru(II) and intramolecular cyclization to yield the product 11.

Scheme 6 Visible-light photoredox-catalyzed oxy-difluoroalkylation of allylamines with CO₂.

Yu and co-workers further realized visible-light-driven oxy-alkylation of allylamines 12 with CO₂ (1 atm) and unactivated alkyl bromides 13 (Scheme 7).¹² Various functional groups could be tolerated in this process except some electron-withdrawing groups such as Bz, Boc, and Ts groups. A variety of primary, secondary, and tertiary alkyl bromides were applied in this reaction as well. Radical inhibition experiments and radical clock experiments proved the existence of alkyl radicals which were also detected by radical trapping experiments with TEMPO. A plausible mechanism was proposed as shown in Scheme 7. First, the Pd(0) is excited by photons and undergoes a single electron transfer (SET) process with alkyl bromide to generate Pd(I) and an alkyl radical. Then, radical addition to the carbamate takes place and gives benzylic radical I-13, which subsequently regenerates Pd(0) and delivers the stabilized cation I-14via SET. Intramolecular cyclization rapidly occurs to give the corresponding 2-oxazolidinone 14.

Scheme 7 Visible-light photoredox-catalyzed oxy-alkylation of allylamines with CO₂.

Although the prolinol scaffold is attractive in medical chemistry as a valuable heterocycle, the synthetic methods are limited for protected allylamines due to the undesirable reaction between free amines and oxidants. In 2017, our group demonstrated a concise iodine-mediated oxidative bicyclization of pent-4-en-1-amines 15 with CO₂ (1 atm) via the oxyamination process to prepare prolinol carbamates 16, which could be easily transformed to various functionalized prolinols (Scheme 8).¹³ A mechanism was proposed as shown in Scheme 8 based on controlled reactions. The amine first adds CO₂ with the formation of carbamate salts I-15. In the second step, the alkene is activated by I₂ and attacked by a CO₂⁻ protected amine. The resulting iodo-carbamic acid I-16 affords the oxyamination product 16 after intramolecular substitution.

Scheme 8 I₂-Mediated carboxylative cyclization of pent-4-en-1-amines with CO₂.

Cyclization of propargylamines to prepare 2-oxazolidinones with CO₂ has been reported with various catalysts such as metal catalysts,^14a–e organic catalysts,^14f,g ionic liquids,^14h CO₂-adduct catalysts,^14i,j and metal–organic frameworks (MOFs).^14k,l

Moreover, atom-economical and step-economical multicomponent reactions to afford highly functionalized 2-oxazolidinones are still desirable. Nevada and co-workers disclosed a palladium-catalyzed multicomponent reaction with propargylamines 17, aryl halides and CO₂ (1 atm) (Scheme 9).¹⁵ In the presence of atmospheric pressure CO₂, highly functionalized 5-methylene-1,3-oxazolidin-2-ones 18 were generated stereoselectively in high efficiency. In this reaction, Pd(0) is first oxidized by aryl iodide to Pd(II), which activates the alkyne to facilitate cyclization giving vinyl palladium complex I-17 by π-activation. Subsequently reductive elimination delivers the product 18. In the case of a terminal alkyne, with the addition of excess aryl iodide, CuI (5 mol%), and DABCO (1,4-diazabicyclo[2.2.2]octane) as a base, Sonogashira reaction would occur before cyclization. The palladium catalyst serves as not only a π-Lewis acid but also a bridge connecting the carboxylative cyclization with cross coupling, which provides an approach to functionalized 2-oxazolidinones 20.

Scheme 9 Palladium-catalyzed carboxylative cyclization of propargylamines with CO₂.

Li and co-workers reported an efficient copper-catalyzed four-component coupling with terminal alkynes, aldehydes, amines, and CO₂ (1 atm) to prepare 2-oxazolidinones 21 bearing exocyclic alkenes.¹⁶ Moreover, Zhou and co-workers realized the asymmetric synthesis of 2-oxazolidinones 21 in a similar pathway through the tandem asymmetric coupling carboxylative cyclization (Scheme 10).¹⁷ Copper in the presence of the chiral PYBOX ligand L₁ catalyzes the three-component coupling reaction to provide chiral N-aryl propargylaniline I-18. Downstream cyclization catalyzed by AgOBz/DPG (1,3-diphenylguanidine) affords 2-oxazolidinones 21. Notably, copper was reused as a catalyst in the cyclization step.

Scheme 10 Tandem asymmetric carboxylative cyclization with alkynes, aldehydes, aryl amines and CO₂.

Besides propargylamines and allylamines, 2,3-allenyl amines were also well-investigated in CO₂ chemistry. In 2013, Ma and co-workers reported a stereoselective palladium-catalyzed synthesis of 5-vinyloxazolidin-2-ones 23 from CO₂ (1 atm), aryl iodides or alkenyl iodides, and 2,3-allenyl amines 22 (Scheme 11).¹⁸ Furthermore, inspired by previous palladium-catalyzed cyclization, Ikariya and co-workers demonstrated a carboxylative cyclization of 2,3-allenyl amines 24 with CO₂ (10–70 atm) and Ag(OAc)(IPr) as a catalyst to form product 25 in 32 to 78% yields (Scheme 11).¹⁹ Owing to the silver catalyst, only trace amounts of intramolecular hydroamination products or 6-membered ring cyclic products were observed. Detailed comparisons with silver catalyst and gold catalyst were made by experiments and DFT calculations. The alkenyl carbon in the alkenyl-silver complex possesses more natural atomic charges than the alkenylgold complex as shown by DFT calculations. In the ¹³C NMR spectra, the chemical shift of the alkenyl carbon adjacent to the Au(I) is larger than Ag(I). According to theoretical and experimental results, the better performance of the silver catalyst might be attributed to the more polarized metal–carbon bond in the alkenyl–metal complex.

Scheme 11 Palladium-catalyzed and silver-catalyzed carboxylative cyclization of 2,3-allenyl amines with CO₂.

When allylamines or propargylamines were used, 5-membered heterocycles were obtained in most cases via a 5-exo-dig or 5-exo-trig process. Longer carbochains between amino groups and unsaturated bonds might give larger rings in the presence of CO₂. Triggered by amino groups, less nucleophilic anilines were also utilized to prepare diverse attractive heterocycles with CO₂ undergoing the 6-exo-trig process. In 2013, Yamada and co-workers described the silver-catalyzed synthesis of benzoxazin-2-one 27 derivatives from o-alkynylanilines 26 and CO₂ (10 atm) assisted by DBU (Scheme 12).²⁰ The silver catalyst serves as a Lewis acid to catalyze the cyclization and gives only the Z-type product via a 6-exo-dig process. Primary anilines as well as secondary anilines also gave the corresponding product smoothly.

Scheme 12 Silver-catalyzed carboxylative cyclization of o-alkynylanilines with CO₂.

Apart from the addition to the unsaturated bond, the substitution reaction has displayed application in fixing the CO₂ into heterocycles. Oxazolidinedione is a versatile scaffold in diverse molecules demonstrating biological activities such as anti-bacterial and anti-cancer. In 2015, Kim and co-workers reported a copper/Cs₂CO₃-catalyzed reaction of 2-bromo-3-phenylacrylic acids 28, amines, and CO₂ (1 atm) to afford various oxazolidinediones 29 (Scheme 13).²¹ The hypothesized mechanism is presented in Scheme 13. First, insertion of CO₂ to amine affords the carbamic acid intermediate I-20. Then, 2-bromo-3-phenylacrylic acid coordinates with Cu(I) followed by the intramolecular oxidative addition with the C–Br bond. In the third step, the copper complex was coordinated by carbamic salt I-20 to form the new complex I-21. Subsequently reductive elimination generates the C–O bond. Finally, intramolecular nucleophilic attack of the amino group yields oxazolidinedione releasing the free copper catalyst. Isotope-labelling experiments with C¹⁸O₂ also support this mechanism.

Scheme 13 Copper-catalyzed carboxylative cyclization of 2-bromo-3-phenylacrylic acids and amines with CO₂.

1,3,4-Oxadiazole-2(3H)-one is a privileged skeleton widely existing in numerous bioactive molecules. Previous facile approaches to this structure mostly involve toxic CO or phosgene, while CO₂ as an environmental-friendly feedstock exhibited superiority. Lu and co-workers disclosed a CsF/18-crown-6-mediated 1,3-dipolar cycloaddition of nitrile imine precursors 30 to synthesize 1,3,4-oxadiazole-2(3H)-ones 31 under CO₂ (20 atm) (Scheme 14).²⁴ Some variations in the substituents were made as shown in Scheme 14 and further application in synthesizing potential drugs was conducted. Controlled experiments and NMR detection indicated that 18-crown-6 promotes the formation of the nitrile imine intermediate and CsF/18-crown-6 enhances the reactivity of CO₂ as a 1,3-dipolarophile.

Scheme 14 CsF/18-crown-6-promoted carboxylative cyclization of nitrile imines with CO₂.

Another approach to oxazolidinediones with CO₂ (1 atm) was reported by Lu and co-workers in 2015 (Scheme 15).²³ The α-ketoester 29 reacts with P(NMe₂)₃via Kukhtin–Ramirez addition reaction affording the electrophilic α-ketoester-tris(dimethylamino)phosphine adduct I-22. Then, the adduct traps the carbamic acid generated in situ to yield the carbamic ester intermediate I-23, which gives the corresponding product 30 in the presence of NaOMe.

Scheme 15 Phosphorus-mediated carboxylative cyclization of α-ketoesters and amines with CO₂.

More recently, a similar strategy using an additive to enhance the ability of the functional group to leave was utilized by Repo and co-workers to synthesize the cyclic carbamates 35 from amino alcohols 33 with the assistance of p-toluenesulfonyl chloride (TsCl) as an activating reagent (Scheme 16).²² 5-Membered and 6-membered rings could be obtained with the corresponding amino alcohols while attempts to prepare 7-membered or 8-membered rings failed. Moreover, chiral substrates inverted the configuration at the stereocenter to deliver the product stereoselectively. DFT calculations suggested that the ring closure step was an intramolecular S_N2 step with leaving of the TsO⁻ group, which is in accordance with conversion of configuration when chiral substrates were used.

Scheme 16 TsCl-promoted carboxylative cyclization of amino alcohols with CO₂.

Besides addition and substitution, other miscellaneous methods to accept electrons from oxygen could realize carboxylative cyclization as well. In 2017, Zhang and co-workers disclosed palladium-catalyzed carboxylative annulation of o-iodoanilines 36 with CO₂ (10 atm) and CO (5 atm) to generate isatoic anhydrides 37 (Scheme 17), which possess great potential in heterocycle synthesis and RNA structure probing.²⁵ The reaction is compatible with both electron-donating groups and electron-withdrawing groups from moderate to good yield. Experiments utilizing labeled CO₂ and CO revealed a possible reaction pathway. CO inserts into a C–Pd(II) bond generated by oxidative addition of ArX to give acyl-palladium intermediate I-24, which was transformed to the 7-membered acyl-palladium carbamate I-25 after addition of CO₂. Then, reductive elimination affords the corresponding isatoic anhydrides with the release of Pd(0). Another possible route involving the formation of a 6-membered palladium carbamate I-26 from CO₂ insertion cannot be excluded.

Scheme 17 Palladium-catalyzed carboxylative cyclization of o-iodoanilines with CO₂ and CO.

2.2 O-Nucleophile-triggered carboxylative cyclization

O-Nucleophiles such as alcohols are capable of inserting CO₂ with the formation of carbonic acid species, and the subsequent intramolecular addition or substitution would fix CO₂ into the heterocycles. Alcohols containing unsaturated bonds or leaving groups could undergo carboxylative cyclization and afford useful cyclic carbonates. Methods utilizing metal catalysts,^26a,b NHC-carbenes,^26c organic bases,^26d,e and ionic liquids^26f have been well-investigated, which provide multiple ways to synthesize promising cyclic carbonates. However, highly functionalized cyclic carbonates are rarely reported.

Alcohols containing C–C double bonds are reactive in the carboxylative cyclization via the addition process. Minakata and co-workers achieved ^tBuOI-mediated cyclization of allyl alcohols and homoallylic alcohols to afford cyclic carbonates bearing an iodomethyl group.^6c However, the synthesis of chiral cyclic carbonates is still challenging. In 2016, Johnston and co-workers applied the dual Brønsted acid/chiral base organocatalyst to construct chiral centers and succeeded in preparing the 6-membered iodo-substituted cyclic carbonate 39 (Scheme 18).²⁷ In the presence of the catalyst, I⁺ generated from the interaction of alkene and NIS (N-iodosuccinimide) induced the cyclization of carbonic acid, which is generated from the alcohols and CO₂. Moreover, the resulting iodocarbonate was a versatile starting material and further transformations such as reduction and hydrolysis were demonstrated. Water exhibited a potent effect on the yield of the product due to the reversible combination with CO₂, which is a competitive reaction to addition of alcohols to CO₂. It is believed that the high enantioselectivity derives from the hydrogen bond between the transient carbonic acid intermediates and the dual Brønsted acid/chiral base catalysts.

Scheme 18 NIS-mediated carboxylative cyclization of homoallylic alcohols with CO₂.

Carboxylative cyclization involving propargylic alcohols and CO₂ (1 atm) has been studied thoroughly. Recently, Cheng and co-workers realized a palladium-catalyzed one-pot carboxylative cyclization of propargylic alcohols 40, organic halides and CO₂ (1 atm) under mild conditions (Scheme 19) via Pd(II) species I-27.²⁸ This reaction features excellent step-economy, a broad substrate scope, good functional group tolerance, and high stereoselectivity, affording various functionalized α-alkylidene cyclic carbonates.

Scheme 19 Palladium-catalyzed carboxylative cyclization of propargylic alcohols and organic halides with CO₂.

In most cases, epoxides are initiated by external nucleophiles to capture CO₂ and give the corresponding cyclic carbonates via an S_N2 mechanism. Different from conventional mechanisms, Kleij and co-workers realized CO₂-incorporated carboxylative cyclization of epoxy alcohols and amines 42 in MEK (methyl ethyl ketone) via an S_Ni mechanism (Scheme 20).²⁹ The nucleophilic oxygen of in situ generated carbonates I-28 attacked epoxides in the presence of aluminum catalyst to generate cyclic carbonate 43 in a 5-exo-tet process. Moreover, this reaction was employed in organic synthesis, and some natural products containing an epoxy alcohol scaffold could give different carbonates with high chemoselectivity in different modes by the regulation of reaction conditions.

Scheme 20 Aluminum-catalyzed carboxylative cyclization of epoxy alcohols with CO₂.

In 2016, the NHC-copper-catalyzed multicomponent coupling of CO₂ (5 atm), bis(pinacolato)diborons, and aldehydes was disclosed by Hou and co-workers (Scheme 21).³⁰ A wide range of lithium boracarbonate ion pairs bearing different functional groups were prepared efficiently, which possess great potential as electrolytes in lithium ion batteries. Based on the isolated intermediates, a multi-component reaction mechanism was proposed. The reaction starts with ligand exchange between LiO^tBu and [(SIMes)CuCl] to generate the copper alkoxide [(SIMes)Cu(O^tBu)] followed by its reaction with B₂(pin)₂ to afford the copper complex [(SIMes)Cu(Bpin)]. Then, addition of Cu–B to the aldehyde occurs generating a new copper complex I-29, which subsequently attacks CO₂ to afford the cyclic boracarbonate derivative I-30 with the formation of a B–O bond. Transmetalation between LiO^tBu and the copper complex provides the final product releasing the copper catalyst.

Scheme 21 Copper-catalyzed carboxylative cyclization of aldehydes and bis(pinacolato)diborons with CO₂.

2.3 C-Nucleophile-triggered carboxylative cyclization

Strong C-nucleophiles, such as Grignard reagents and organic lithium reagents, are able to insert CO₂ with the formation of C–C bonds to give the corresponding carboxylic acids. However, weak C-nucleophiles could only attack CO₂ reversibly to generate unstable carboxylates, which tend to be decarboxylated. To fix CO₂ into the molecules, further intramolecular addition to the unsaturated bond is necessary. Additionally, the resulting lactones are vital scaffolds in pharmacy.

Obviously, enolates possess sufficient nucleophilicity to capture CO₂, but the corresponding β-ketocarboxylic acids would revert to starting materials due to their thermodynamic instability. In 2012, Yamada's group realized the silver-catalyzed intramolecular cyclization of β-ketocarboxylic acids to construct 5-membered lactones.³¹ Furthermore, in 2017, they applied a similar strategy for the synthesis of tetronic acids 46 in good yields. The MTBD (7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-ene)/AgOAc system achieved the capture of CO₂ with conjugated ynones 45via a 5-exo-dig cyclization selectively (Scheme 22).³² Application of this reaction in the total synthesis of Aspulvinone E exhibited enormous potential in synthetic chemistry.

Scheme 22 Silver-catalyzed carboxylative cyclization of conjugated ynones with CO₂.

Organosilane compounds are versatile tools to construct C–C bonds as potential nucleophiles. In the presence of F⁻ or Lewis acids, carboxylation of the C–Si bond with CO₂ (10 atm) to yield carboxylic acids or esters has been investigated. In 2015, Yamada and co-workers reported the silver-catalyzed and CsF-promoted carboxylative cyclization of trimethyl(2-methylenebut-3-yn-1-yl)silane derivatives 47 with CO₂ (10 atm) to form the 2-furanones 48 and 2-pyrones 49 in satisfactory yield regioselectively (Scheme 23).³³ When aryl-substituted alkynes were used, the 2-furanone derivatives 48 were obtained via a 5-exo-dig cyclization; single crystal X-ray diffraction performed on the products showed that they exist as Z-isomers. However, when alkyl-substituted alkynes were used, 2-pyrones 49 were formed as the main products.

Scheme 23 Silver-catalyzed and CsF-promoted carboxylative cyclization of trimethyl(2-methylenebut-3-yn-1-yl)silane derivatives with CO₂.

The C3-position of indoles is a promising C-nucleophile to react with electrophiles, and carboxylation has been established.³⁴ In 2015, Skrydstrup and co-workers described the 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD)-catalyzed carboxylation of alkynyl indoles 50 at 100 °C (Scheme 24).³⁵ 30 mol% TBD catalyzed the cyclization efficiently when R² was an aryl group, and 1.0 equivalent of TBD was necessary when R² was an alkyl. This work provides an effective path to prepare tricyclic indole-containing rings 51 bearing different functional groups. A regular mechanism involving the indole-C3 position attacking the TBD-CO₂ adduct to form the carboxylate and then delivering the lactone by a 6-endo-dig process was proposed. However, more recently, DFT calculation³⁶ conducted by Wong and co-workers revealed that the TBD might not only activate CO₂ to facilitate carboxylation of indole-C3 position but also stabilize the carboxylate to assist cyclization. This theoretical calculation did not contradict the demonstrated experimental result.

Scheme 24 TBD-catalyzed carboxylative cyclization of alkynyl indoles with CO₂.

Benzyne is a highly reactive intermediate with great electrophilicity. The addition of nucleophiles to benzynes could generate C-nucleophiles to trap electrophiles, which provide an access to 1,2-disubtituted arenes. In 2014, Kobayashi and co-workers demonstrated the copper-catalyzed tandem cyclization reaction of terminal alkynes, CO₂ (15 atm) and benzynes to prepare a series of functionalized isocoumarins 53 (Scheme 25).³⁷ Assisted by NHC-copper catalysts and three equivalents of Cs₂CO₃, isocoumarins were obtained in moderate to good yield bearing electron-withdrawing and electron-donating functional groups. The mechanism proposed is shown in Scheme 25. The terminal alkyne is basified and generates an alkyne copper complex in the presence of copper catalyst. Assisted by CsF, 2-(trimethylsilyl)phenyl triflate 52 is converted to benzyne, which subsequently reacts with the alkynyl copper species to give o-alkynyl aryl copper intermediate I-33. Then, reversible nucleophilic addition to CO₂ occurs and a subsequent 6-endo-dig cyclization produces isocoumarin copper complex I-34. Transmetalation regenerates the copper catalyst. Finally, the resulting cesium salt I-35 gives the product 53 after hydrolysis, which is also supported by controlled experiments with deuterium-enriched alkynes.

Scheme 25 Copper-catalyzed carboxylative cyclization of benzynes and terminal alkynes with CO₂.

The addition to alkynes is also an effective route to generate C-nucleophiles. Elegant substrate design might lead to different heterocycles. In 2018, Wu and co-workers disclosed the first visible-light-driven carbocarboxylation of alkynes 54 with the Ir/Co dual catalyst system and CO₂ (1 atm) to prepare 2-benzoxepinones 55 (Scheme 26).³⁸ When ortho-ester substituted aryl alkyne 54 was applied, an unprecedented cobalt-carboxylation/acyl migration occurred to deliver γ-hydroxybutenolides efficiently. Initially, the Co(II) oxidizes Ir(II) which was reduced by ⁱPr₂NEt and visible-light to give vital Co(I) species. Alkyne 54 reacts with CO₂ and the cobalt catalyst to generate a 5-membered cobaltacycle intermediate I-36. Addition to the ester group occurs to generate carboxylate I-37. Then, the cobalt complex I-37 undergoes photoredox-catalyzed reduction and transmetalation with the photocatalyst and ZnBr₂ to release the Co(I) species and carboxylate I-38. Tautomerization spontaneously takes place to deliver γ-keto acryliczinc complex I-39, which is followed by intramolecular cyclization to generate γ-hydroxybutenolide 55. When terminal alkynes were used, 2-pyrones were obtained (Scheme 27). The terminal alkyne inserts 5-membered cobaltacycle intermediate I-40 to give the 7-membered cobaltacycle I-41. Further reductive elimination releases the 2-pyrones and regenerates the Co(I) species.

Scheme 26 Visible-light-driven cobalt-catalyzed carboxylative cyclization of ortho-ester substituted aryl alkynes with CO₂.

Scheme 27 Visible-light-driven cobalt-catalyzed carboxylative cyclization of terminal alkynes with CO₂.

Apart from lactones, other heterocycles such as maleic anhydrides can be easily prepared with CO₂ (1 atm) by miscellaneous methods. Tsuji and co-workers demonstrated the nickel-catalyzed double carboxylation of internal alkyne 58 at room temperature with Zn as a reducing reagent and MgBr₂ as an additive (Scheme 28).³⁹ Controlled experiments and DFT calculations were conducted to elucidate the mechanism. Initially, Ni(II) is reduced by Zn to afford Ni(0), which is followed by the oxidative cycloaddition with an internal alkyne and CO₂ to generate Ni(II) metallacycle I-42. Then, the Ni(I) species I-43 is formed from the Ni(II) species with the assistance of Zn and MgBr₂via a single electron transfer. Addition of the C–Ni bond to CO₂ takes place with the formation of the dicarboxylated acid I-44 that leads to maleic anhydride after acidification.

Scheme 28 Nickel-catalyzed carboxylative cyclization of internal alkynes with CO₂.

The CO₂-incorporated carboxylative cyclization has been developed to afford a diversity of heterocycles such as 2-oxazolidinones, cyclic carbonates, 1,3,4-oxadiazole-2(3H)-ones, and lactones based on different nucleophiles and electrophilic sites. Among these transformations, substrate design is a dominant factor. The nucleophilic site of the substrate traps CO₂ while its electrophilic site fixes CO₂ as part of a heterocycle. Transition metal catalysts often play a vital role in cyclization to enhance the nucleophilicity of nucleophiles or electrophilicity of electrophiles.

3. Nucleophile-triggered CO₂-incorporated carbonylative cyclization

In contrast to the rapid development of carboxylative cyclization, carbonylative cyclization utilizing CO₂ has been rarely reported until recently. Compared with traditional carbonyl sources such as CO⁴⁰ and phosgene, CO₂ is undoubtedly more promising due to it being non-toxic, renewable, and safe. Theoretically, there are three pathways to incorporate CO₂ as a carbonyl source into a heterocycle. To reduce CO₂ to CO with a reductant represents an effective means to exploit CO₂ as a CO surrogate.⁴¹ However, when it comes to carbonylative cyclization, the employment of a reductant makes the transformation less atom-economical. Obviously, a more straightforward method is the carbonylative cyclization with CO₂ in substrates containing two nucleophilic sites in a redox-neutral process. This reaction is initiated by the addition to CO₂ followed by a C–O bond cleavage to deliver the carbonylative product and oxygen-containing species. The main difficulty identified is the poor leaving ability of oxygen after the addition to CO₂. Additionally, the rearrangement of unstable carbamate cycles deriving from CO₂ is also a possible mode. Although various challenges exist, the displacement of toxic CO and phosgene with CO₂ is still promising, which may extend the application of CO₂ as “CO” and “O” synthon in a redox-neutral pathway.

3.1 N-Nucleophile-triggered carbonylative cyclization

Carbonylative cyclization triggered by N-nucleophiles with CO₂ may produce multiform heterocycles such as lactams and imidazolidin-2-ones, which depend on the substitution in the second step forming the new bonds.

Oxindoles are important lactams widely existing as skeletons in natural products and medicinal chemistry. In 2015, Hirao and co-workers described the CO₂-incorporated carbonylative cyclization with ketoamines 60 to prepare 3-aryl-3-hydroxy-2-oxindoles 61, which are one of the most important compounds in oxindole derivatives (Scheme 29).⁴² Magnesium served as a reductant to reduce the carbonyl group. Moreover, in the absence of Me₃SiCl, alcohols were the main product and no desired products were observed, which suggested the assistance of Me₃SiCl in the CO₂ transformation. Notably, the bromo and ester group could survive in this cyclization. In addition, N-methyl substrates also produced the corresponding lactams smoothly. However, the mechanism remains unclear up to now. Single-electron reduction of the ketone to attack CO₂ was proposed as a reasonable path. However, a two-electron reduction process cannot be excluded.

Scheme 29 Me₃SiCl-promoted carbonylative cyclization of ketoamines with CO₂.

In 2016, Yu and co-workers developed the NaO^tBu promoted carbonylation of o-alkenyl- or o-arylanilines 62 with the incorporation of CO₂ (1 atm), which provided a useful method for the formation of important 2-quinolinones and polyheterocycles 63via the lactamization of alkenyl or heteroaryl C–H bonds (Scheme 30).⁴³ This reaction features a broad substrate scope, scalability, and environmental friendliness and presents a feasible approach to various significant lactam derivatives such as the Tipifarnib precursor. Moreover, compared with CO, application of CO₂ in the carbonylation avoided the addition of oxidants and realized the carbonylation of sp² C–H bonds in redox-neutral conditions. Experiments concentrating on the mechanism showed that isocyanates I-45 deriving from CO₂ and anilines might be vital intermediates to give the corresponding lactam after cyclization.

Scheme 30 NaO^tBu-mediated carbonylative cyclization of o-alkenyl- or o-arylanilines with CO₂.

Our group developed MeOTf and TBD mediated carbonylative cyclization of o-arylanilines 64 leading to phenanthridinones 65 with CO₂ (1 atm) in o-DCB (1,2-dichlorobenzene) at 140 °C (Scheme 31).⁴⁴ Although o-alkenylanilines failed to be transformed to the corresponding lactams in this reaction, the conversion of o-arylanilines exhibited good performance owing to the great reactivity of MeOTf, which provided a complementary route to that shown in Yu's work.⁴³ The electronic effect was obvious and the electron-donating functional group enhanced the yield. The isolation of methyl carbamate and the observation of isocyanate in high temperature demonstrated the possible mechanism. First, aniline reacts with the TBD-CO₂ adduct to give methyl carbamate I-46 with the aid of MeOTf. In the second step, after losing HOTf at high temperature, the methyl carbamate intermediate is converted to isocyanate I-47, which is induced by MeOTf to generate 6-methoxyphenanthridine I-48. Finally, phenanthridinone is obtained after acidification.

Scheme 31 TBD and MeOTf-mediated carbonylative cyclization of o-arylanilines with CO₂.

Recently, Cheng's group reported a palladium-catalyzed multicomponent tandem reaction involving N-tosylhydrazones 66, o-iodoanilines 67, and CO₂ (1 atm) to prepare a series of 4-aryl-2-quinolinones 68 (Scheme 32).⁴⁵ The reaction might start with palladium-catalyzed cross-coupling of N-tosylhydrazones and o-iodoanilines followed by the carbonylation of CO₂. However, a pathway generating 1-iodo-2-isocyanatobenzene before cross-coupling cannot be excluded.

Scheme 32 Palladium-catalyzed carbonylative cyclization of N-tosylhydrazones and o-iodoanilines with CO₂.

Compared with lactamization mentioned above, carbonylation with stronger nucleophiles forming C–N or C–O bonds is less challenging. In 2015, Suen and co-workers accomplished KOH-promoted carbonylation of hydrazides 70 with CO₂ (bubble) (Scheme 33).⁴⁶ The hydrazides 70 could be prepared from acid chlorides 69 and hydrazine monohydrate in high efficiency affording biologically significant 5-substituted-3H-[1,3,4]-oxadizole-2-one 71 under mild conditions.

Scheme 33 KOH-promoted carbonylation of hydrazides with CO₂.

2-Benzimidazolone is an attractive heterocycle for application in pharmaceutical chemistry. Liu and co-workers displayed DBU-based ionic-liquid-catalyzed carbonylative cyclization with the formation of two C–N bonds utilizing CO₂ (1 atm) at 120 °C (Scheme 34).⁴⁷ When 2-aminothiophenols were used, the corresponding benzothiazolones could also be obtained efficiently. Notably, [DBUH][OAc] is a dual-active catalyst. [DBUH]⁺ activates CO₂ with tertiary nitrogen, while [OAc]⁻ activates o-phenylenediamine through hydrogen bonding. Furthermore, the azole-anion-based aprotic ionic liquids were also applied in this transformation and delivered 2-benzimidazolones in a similar process.⁴⁸

Scheme 34 [DBUH][OAc]-catalyzed carbonylative cyclization of phenylenediamines with CO₂.

In 2017, Cheng's group reported the NaO^tBu-mediated dearomative carbonylation of N-2-pyridylamidines 74 with the formation of two C–N bonds under 1 atm of CO₂ (Scheme 35).⁴⁹ This annulation provided a range of pyrido[1,2-a]-1,3,5-triazin-4-ones 75 bearing different functional groups, and the iodo-substituted product was extended to synthesize artificial nucleosides in a straightforward manner. The 6π-electron cyclic reaction from isocyanate intermediates I-49 was described as a possible process.

Scheme 35 NaO^tBu-mediated carbonylative cyclization of N-2-pyridylamidines with CO₂.

Although heteroatoms possess sufficient nucleophilicity, some carbonylation using CO₂ is not thermodynamically favorable. For instance, the reaction between 2-aminoethanols and CO₂ to yield 2-oxazolidinones is restricted thermodynamically. In 2015, He and co-workers provided an approach to overcome this difficulty by adding additives to trigger the reaction. With the addition of propargylic alcohols 77 and silver catalyst to avoid the conventional dehydration step, 2-aminoethanol derivatives 77 were transformed to a wide range of 2-oxazolidinones 78 with CO₂ (10 atm) (Scheme 36).⁵⁰ Controlled experiments and DFT calculations were conducted to elucidate the mechanism, and a key intermediate I-50 was also isolated when the reaction was conducted over two hours. Based on the evidence, a plausible mechanism was proposed. Initially, the propargylic alcohol attacks CO₂ to afford cyclic carbonate. The resulting carbonate reacts with 2-aminoethanol via the ring-opening reaction to afford carbamate intermediate I-50,⁵¹ which affords the corresponding 2-oxazolidinone 78 and α-hydroxyl ketone 79 through intramolecular nucleophilic substitution. Furthermore, a similar strategy was also applied in the synthesis of cyclic carbonates.⁵²

Scheme 36 Silver-catalyzed carbonylation of 2-aminoethanol derivatives with CO₂.

Carbonylation of inert C–H bonds utilizing CO₂ is still the most attractive approach due to the atom-economy and high efficiency. In 2018, Yu and co-workers revealed the palladium-catalyzed carbonylation of unactivated C–H bonds in benzamides 80 with CO₂ (1 atm) to give the phthalimides 81 under redox-neutral conditions (Scheme 37).⁵³ The nature of substituents on the nitrogen of the benzamides was crucial, with only a methyl group giving good yield which might arise from the less steric hindrance and better directing ability. It is noteworthy that the carbonylation only occurred at the less-hindered ortho position when meta-substituted substrates were used. The kinetic isotope effect (KIE) value suggested that the C–H activation step might be the rate-determining-step. Although the precise mechanism of the carbonylation step was not clear, the palladacycle was believed as the significant intermediate for detection of I-51 in ESI-HRMS when N,N,N′,N′-tetramethylethylenediamine (TMEDA) was applied as the ligand.

Scheme 37 Palladium-catalyzed carbonylation of benzamide derivatives with CO₂.

Isotope labeling has demonstrated remarkable application in early diagnosis of diseases and pharmaceuticals. In 2018, Audisio and co-workers applied the Staudinger/aza-Wittig reaction in the late-stage carbon labelling with o-azidoanilines 82 or 3-azidopropan-1-amines 84 and labelled CO₂ (1 equiv.) replacing conventional toxic labelled phosgene or CO (Scheme 38).⁵⁴ All isotopes of carbon (¹¹C, ¹³C, ¹⁴C) could be utilized in this method. Moreover, this reaction features excellent tolerance to different functional groups and practicability. Various drug candidates such as Domperidone 86 and even an unprotected peptide could be labelled utilizing this method.

Scheme 38 Carbonylation of o-azidoaniline derivatives with labelled CO₂.

Utilizing CO₂ as the carbonyl source via rearrangement reactions has also been well established.⁵⁵ 2-Aminobenzonitriles 88 are typical substrates that undergo carboxylative cyclization to afford quinazoline-2,4(1H,3H)-diones 89 after rearrangement (Scheme 39). Various strategies were applied to realize efficient transformation via similar rearrangement processes. The amino group reacts with CO₂ and the resulting carbamic acid attacks the C–N triple bond to give a carbamate I-54. Subsequently, the carbamate undergoes C–O bond cleavage with the formation of isocyanate intermediate I-55, which rearranges to quinazoline-2,4(1H,3H)-diones 89 by the nucleophilic addition due to its thermodynamic instability.

Scheme 39 NaO^tBu-mediated carbonylative cyclization of N-2-pyridylamidines with CO₂.

When primary o-alkynylaniline 90 was utilized in the silver-catalyzed carboxylative cyclization, Yamada's group observed 4-hydroxyquinolin-2(1H)-one 91 as a by-product, which was identified by ¹H NMR, ¹³C NMR, and single crystal X-ray diffraction. Isotopic labelling experiments with C¹⁸O₂ and in situ IR measurements strongly suggested the possible mechanism as shown in Scheme 40.⁵⁶ First, carboxylative cyclization occurs to generate benzoxazin-2-one I-56 as in previous work.²⁰ Assisted by DBU, deprotonation followed by C–O bond cleavage delivers isocyanate I-57, which is subsequently transformed to the corresponding product by a nucleophilic addition of the enolate to the C–O double bond. They further extended this intramolecular rearrangement to propargylic amines with CO₂ (1 atm) and a variety of tetramic acid derivatives were obtained in good yield.⁵⁷ In 2014, Lu and co-workers found similar rearrangement using CuI as catalyst (Scheme 41).⁵⁸

Scheme 40 Silver-catalyzed carbonylative cyclization of o-alkynylanilines with CO₂.

Scheme 41 Copper-catalyzed carbonylative cyclization of o-alkynylanilines with CO₂.

Furthermore, this strategy was extended to multicomponent reactions incorporating CO₂ into heterocycles as a carbonyl source. In 2017, Cheng and co-workers developed a palladium-catalyzed three-component reaction between o-alkynylanilines 94, aryl iodides, and CO₂ (1 atm), which provided a sustainable method to construct 3,3-diaryl 2,4-quinolinediones 95 (Scheme 42).⁵⁹ The mechanism is shown in Scheme 42. Firstly, o-alkynylaniline captures CO₂ to give carbamate. Then, ArPdX produced by oxidative addition of ArI and Pd(0) activates the carbamate, which undergoes a trans-oxopalladation to produce intermediate I-58. Subsequently, reductive elimination of the intermediate I-58 takes place to afford heterocycles I-59 with the regeneration of Pd(0). Finally, deprotonation followed by rearrangement delivers the corresponding product 95.

Scheme 42 Palladium-catalyzed carbonylative cyclization of o-alkynylanilines and aryl iodides with CO₂.

The synthesis of N₃-substituted quinazoline-2,4(1H,3H)-diones has attracted great attention due to their wide existence in bioactive molecules such as ketanserin and nitraquazone as well as their utilization in synthetic chemistry. In 2017, Beller and co-workers realized the three-component reaction involving o-bromoanilines 96, isocyanides 97, and CO₂ (10 atm) in the catalysis of Pd(OAc)₂ and BuPAd₂ (butyldi-1-adamantylphosphine) (Scheme 43).⁶⁰ This reaction demonstrated good chemoselectivity, regioselectivity, and tolerance to tertiary, secondary, and primary isocyanides. The post-functionalization of the product could be conducted conveniently. Moreover, as an economical ¹³C-carbonylating reactant, ¹³CO₂ could also deliver the desired heterocycles under optimized reaction conditions, which exhibits application in preparing selective-labelled products. When N-benzyl-2-bromoaniline was used, N-tert-butyl-N′-benzyl-2-amino-benzamide was obtained in 68% yield instead of the formation of the desired heterocycle, which strongly suggested the existence of isocyanate intermediates in the rearrangement. A possible reaction mechanism is shown in Scheme 43. Intermediate I-60 was formed by the oxidative addition of Pd(0) with o-bromoaniline followed by the migratory insertion of isocyanide to a C–Pd(II) bond. In the presence of Cs₂CO₃, a nucleophilic addition to CO₂ followed by reductive elimination generated intermediate I-61 with the release of Pd(0). Finally, cyclic carbamate intermediate I-61 is treated with base to afford quinazoline-2,4(1H,3H) diones 98via isocyanate intermediate I-62. Notably, Ji and Lu's group also independently realized a similar palladium-catalyzed reaction with o-iodoanilines or o-bromoanilines.⁶¹

Scheme 43 Palladium-catalyzed carbonylative cyclization of o-bromoanilines and isocyanides with CO₂.

3.2 O-Nucleophile-triggered carbonylative cyclization

As potent nucleophiles, O-nucleophiles are expected to trap CO₂ and generate lactones after consecutive intramolecular substitution with one C–O bond formation and one C–O bond cleavage. When the o-heteroarylphenols 99 were employed, important coumarin derivatives were obtained utilizing CO₂ (1 atm) (Scheme 44).⁶² Based on controlled experiments, a possible mechanism involving addition of carbonate I-63 to CO₂ followed by substitution of nucleophilic sp² C–H was proposed as shown in Scheme 44. In addition, treatment of o-alkenylphenols 101 with CO₂ also afforded compounds 102 (Scheme 45).

Scheme 44 KO^tBu-mediated carbonylative cyclization of o-heteroyl phenols and o-alkenylphenols with CO₂.

Scheme 45 KO^tBu-mediated carbonylative cyclization of o-alkenylphenols with CO₂.

Although direct carboxylation of C–H bonds with CO₂ is difficult due to the low reactivity of CO₂, the C–H bond with a directing group possesses sufficient nucleophilicity to many electrophiles. Yu and co-workers applied this property in the lactonization of tertiary benzylic alcohols 103 with CO₂ (1 atm) to prepare important phthalides 104 (Scheme 46).⁶³ Primary and secondary benzylic alcohols failed due to β-H elimination. When a ¹⁸O-labeled substrate was used, ¹⁸O was fully retained in the product which supported the possible carbonylative process. DFT calculations were conducted to elucidate the mechanism. The reaction starts with coordination of the benzylic alcohol with the palladium catalyst in the presence of LiO^tBu, which subsequently generates the palladacycle intermediate I-64via a concerted-metalation–deprotonation (CMD) process. Then, the CO₂ insertion to the Pd–O bond occurs successively to generate 9-membered palladacycle intermediate I-65. Subsequently, intramolecular nucleophilic addition gives the intermediate I-66. Finally, decarbonization releases lactone and the palladium catalyst.

Scheme 46 Palladium-catalyzed carbonylative cyclization of tertiary benzylic alcohols with CO₂.

Apart from the straightforward sequential addition to CO₂, rearrangement gave the carbonyl as well when O-nucleophiles were applied. In 2018, Yu and co-workers developed the practical synthesis of appealing tetronic acids 106 with propargylic alcohols 105 and CO₂ (1 atm) in 1,3-dimethyl-2-imidazolidinone (DMI) (Scheme 47).⁶⁴ Substrates with electron-withdrawing groups exhibited better reactivity which might be explained by the higher electrophilicity of the C–C triple in the cyclization step. Carbonate I-67 was considered as an intermediate based on transformation from the carbonate to the product without CO₂. A mechanism involving rearrangement of the cyclic carbonates was proposed. Propargylic alcohol is basified to attack CO₂ and then undergoes intramolecular cyclization to give cyclic carbonate. Nucleophilic attack of the Cs₂CO₃ to the carbonyl promotes the ring-opening process and delivers the intermediate I-68. Then, the Dieckmann-type condensation generates Cs₂CO₃ and the re-cyclized product 106 which could be transformed to tetronic acid.

Scheme 47 Cs₂CO₃-mediated carbonylative cyclization of propargylic alcohols with CO₂.

3.3 C-Nucleophile-triggered carbonylative cyclization

Unlike the N-nucleophiles and O-nucleophiles, C-nucleophiles only exist in a specific position, which are unfavorable to attack CO₂. Furthermore, considering the difficulty of unactivated C–H carboxylation with CO₂ up to now, C–H carbonylative cyclization is still challenging. In 2013, Iwasawa and co-workers disclosed the direct carboxylation of an alkenyl C–H bond in catalysis using palladium with CO₂ (1 atm) to afford a variety of coumarins 108 in good yields (Scheme 48).⁶⁵ Importantly, an alkenyl palladium complex coordinated by cesium salt was isolated and confirmed by single crystal X-ray diffraction, which indicated that C–H bond cleavage had occurred. Additionally, the carboxylation step turns out to be reversible, which displays a tendency to decarboxylation side. Coumarins were obtained by the reaction of I-70 with another molecule o-alkenylphenol 107.

Scheme 48 Palladium-catalyzed carbonylative cyclization of o-alkenylphenols with CO₂.

Triggered by various nucleophiles, benzynes could be transformed to 1,3-zwitterionic intermediates. In 2014, Biji and co-workers demonstrated the isocyanide-triggered cyclization of benzynes with CO₂ (1 atm) (Scheme 49).⁶⁶ Transformation of I-71 to the product 111 could be realized in the presence of F⁻. Consequently, CsF turns out to be crucial not only in the activation of benzyne precursors but also in the rearrangement step.

Scheme 49 CsF-mediated carbonylative cyclization of benzynes and isocyanides with CO₂.

In 2015, Lu and co-workers reported the carbonylation of o-hydroxyacetophenones 112 and o-aminoacetophenones 114via the carboxylation of enolates with DBU (Scheme 50).⁶⁷ The formation of stable heterocycles is the driving force that promotes reversible reaction between enolates and CO₂ towards the carboxylation.

Scheme 50 DBU-mediated carbonylative cyclization of o-hydroxy- and o-acetamido-acetophenones with CO₂.

This practical protocol was further applied by Lu's group in 2016 (Scheme 51).⁶⁸ CsF-promoted carbonylation of propenyl ketones 117 with CO₂ (30 atm) at 100 °C was shown to afford α-pyrones 118, which possessed a significant biological activity. It was assumed that the high pressure of CO₂ and low concentration of reaction inhibited dimerization of substrates. C¹⁸O₂ was also utilized in this reaction to gain more insight into the mechanism and to demonstrate that CO₂ only acts as a CO moiety in this transformation. The reaction might be initialized by carboxylation followed by intramolecular substitution while another mechanism similar to Yu's work⁶² that was triggered by O-nucleophiles could not be excluded.

Scheme 51 CsF-mediated carbonylative cyclization of propenyl ketones with CO₂.

Recently, Yu and co-workers reported a one-pot transition-metal-free lactonization of enamides 119 to prepare 1,3-oxazin-6-ones 120 (Scheme 52).⁶⁹ EtI was used to quench the reaction and the corresponding esters were obtained. N-Methyl substituted enamides failed to be carboxylated in the presence of base. Moreover, the utilization of ¹³CO₂ gave the selective-labelled product. On the basis of these experiments, it is believed that the reaction is initiated by the carboxylation of electron-rich alkenes.

Scheme 52 NaO^tBu-promoted carbonylative cyclization of enamides with CO₂.

Transition-metal catalyzed C–H carbonylation with CO₂via C–H bond activation has rarely been reported since Iwasawa's work.⁶⁵ The main challenge lies in the fact that the acidic and oxidative condition for C–H activation is incompatible with basic and reductive condition for CO₂ activation. In 2018, Li and co-workers displayed an exciting redox-neutral rhodium(II) catalyzed carbonylation of 2-arylphenols with CO₂ (1 atm) (Scheme 53).⁷⁰ Phosphine ligands were extremely important to enhance the reactivity of the rhodium catalyst and to avoid the well-known Kolbe–Schmitt type reaction. KO^tBu was considered to enhance the concentration of CO₂ in the solution by strong interactions with CO₂. Due to steric hindrance, the carboxylation tends to occur at the less hindered C–H bond. The novel complex A prepared was characterized by X-ray crystallography, which could be used as a catalyst to realize the carbonylation of 2-arylphenols efficiently. A tentative mechanism was proposed as shown in Scheme 53. First, 121 is basified with KO^tBu and reacts with complex A to generate complex I-73. Then, assisted by the base ROK (such as KO^tBu), the aryl C–H bond activation of I-73, via a proton abstraction process, generates the metallacycle I-74 reversibly. CO₂ insertion into the C–Rh bond occurs reversibly to give the 8-membered rhodium carboxylate I-75. Subsequently, lactonization rapidly delivers the product 122 and regenerates complex A. More recently, Li's group extended this reaction to more challenging electron-deficient 2-pyridylphenols by tuning ligands, which provide an access to pyrido-heterocycles.⁷¹

Scheme 53 Rhodium-catalyzed carbolyative cyclization of 2-arylphenols with CO₂.

Without reductants, there are two pathways to realize the carbonylation reaction utilizing CO₂. The first method is the successive nucleophilic attack on CO₂ with the substrate possessing two nucleophilic sites. The difficulty of this approach lies in the generation of the small molecule, which makes the reaction thermodynamically unfavorable. As a result, additives or high temperature to provide energy is necessary. Another mode is the rearrangement reaction, which has also been developed with different substrates. The initial cyclization reaction generates an unstable cyclic intermediate, which rearranges immediately to a stable heterocycle.

4. CO₂-Incorporated reductive cyclization

Considering that the carbon center in CO₂ is at its highest valence state, great effort has been made with various reduction systems involving different catalysts and reductants such as hydrogen, boranes, and silanes. Based on the number of electrons donated by reductants, CO₂ could be reduced to different oxidation states to give the corresponding product such as formates and acetals, which could be captured by nucleophiles to afford value-added products.^2,72 Substrates with two nucleophilic sites are able to attack reductive intermediates in sequence to deliver diverse heterocycles.

4.1 2-Electron reductive cyclization

2-Electron reduction could reduce CO₂ to formates, which deliver the amides after formylation with amines.² In 2013, Cantat and co-workers designed a cascade process utilizing substrates with two nucleophilic sites and combined formylation with condensation to produce various heterocycles with a formamidine structure.⁷³ With the aid of NHC-carbenes, 1,2-diamines, anthranilamides, and aminobenzylamines could conduct the cascade reaction successfully with atmospheric CO₂ to generate benzimidazoles, quinazolinones, formamidines and their derivatives in the presence of poly(methylhydrosiloxane) (PMHS) (Scheme 54). This work provided an effective strategy to realize the deoxygenation of CO₂ and made CO₂ more flexible in the construction of heterocycles.

Scheme 54 IPr-catalyzed reductive cyclization of 1,2-diamines, anthranilamides, and aminobenzylamines with CO₂.

A similar strategy was utilized by Liu and co-workers. In the presence of a mixture of gases (150 atm, CO₂ : H₂ = 3 : 2), o-phenylenediamines were catalyzed by RuCl(dppe)₂ to give a series of benzimidazoles (Scheme 55).⁷⁴ Formic acid could react with o-phenylenediamine immediately at 120 °C to provide benzimidazole via formamide I-77. On the basis of this result, a mechanism was proposed. Initially, CO₂ is reduced by H₂ in the presence of a ruthenium catalyst and o-phenylenediamine as a base. Afterwards, N-formylation occurs followed by condensation to give the desired product.

Scheme 55 Ruthenium-catalyzed reductive cyclization of o-phenylenediamines with CO₂.

In 2015, Liu and co-workers reported that o-aminothiophenols and o-phenylenediamines 129 could be used for the reductive cyclization with CO₂ (5 atm) in the presence of an ionic liquid and (EtO)₃SiH as reductant (Scheme 56).⁷⁵ Substrates bearing electron donating groups such as –OMe and –NR₂, or electron withdrawing groups such as –F, –Cl, –Br, and –NO₂ were accommodated in this reaction. In this reaction, the Si–H bond in (EtO)₃SiH was activated by [Bmim]⁺, while the substrates were activated by [OAc]⁻via a hydrogen bond. The cyclic products were formed by a successive substitution from a formoxysilane intermediate.

Scheme 56 [Bmim][OAc]-catalyzed reductive cyclization of o-aminothiophenols and o-phenylenediamines with CO₂.

Frustrated Lewis pairs (FLPs) have demonstrated tremendous potential in activating small molecules. Liu and co-workers applied FLPs as catalysts in the CO₂ transformation to realize N-methylation of amines in 2015.⁷⁶ Subsequently, Sun's group extended FLPs as catalysts to the construction of heterocycles. In the presence of B(C₆F₅)₃, o-phenylenediamines 131 reacted with CO₂ (10 atm) and four equivalents of PhSiH₃ at 120 °C affording benzimidazoles 132,⁷⁷ which has attracted attention due to the relevance to natural products and biologically active molecules (Scheme 57). The amine-CO₂-B(C₆F₅)₃ adduct was detected by NMR spectroscopy. Based on this evidence and further investigations, a plausible mechanism was proposed as displayed in Scheme 57. Initially, amine-CO₂-B(C₆F₅)₃I-78 is formed. Activated by B(C₆F₅)₃, PhSiH₃ reduces this adduct via a 2-electron process to release the catalyst and generates the formamide intermediate I-79, which affords the desired product after dehydration.

Scheme 57 B(C₆F₅)₃-catalyzed reductive cyclization of o-phenylenediamines with CO₂.

Recently, our group reported the Lewis base-promoted selective reduction of CO₂ into boryl formates using BH₃NH₃ as a reductant under mild conditions (Scheme 58).⁷⁸ The in situ generated boryl formates 133 were reacted at 80 °C with benzene-1,2-diamine, 2-aminophenol, and 2-aminobenzenethiol to afford 1H-benzo[d]imidazole (60% yield), 2-aminobenzenethiol (53% yield), and benzo[d]thiazole (40% yield), respectively.

Scheme 58 BH₃NH₃-mediated reductive cyclization with CO₂.

The process of 2-electron reductive cyclization mentioned above by fixing CO₂ through reduction, formylation and condensation in sequence is not the only pathway in 2-electron reductive cyclization. In 2017, Cheng and co-workers described the 2-electron reductive carbonylative cyclization of 2-hydrazinopyridines 135 and benzamidrazones 137 to give triazolones in the presence of K₂CO₃ and CO₂ (1 atm) (Scheme 59).⁷⁹ 5 equivalents of HSi(OEt)₃ served as the reductant to reduce CO₂ in a 2-electron process. To elucidate the mechanism, controlled experiments were conducted. When phenyl hydrazine was used, only the formylated product was obtained under standard conditions, which supported a 2-electron reduction of CO₂. Moreover, the utilization of the pre-prepared compound I-80 as a substrate afforded triazolone 136 along with H₂ in the absence of base and HSi(OEt)₃. This result indicated I-80 as a possible intermediate in the reductive cyclization. Consequently, the reaction might start with formylation of the substrate in the presence of HSi(OEt)₃. Then, nucleophilic attack takes place to generate intermediate I-81. Finally, the product is obtained with extrusion of H₂.

Scheme 59 Reductive cyclization of 2-hydrazinopyridines and benzamidrazones with CO₂.

4.2 4-Electron reductive cyclization

Acetal species can be obtained from CO₂via a 4-electron reduction process, which serve as methylene groups after successive nucleophilic attacks. Nevertheless, this process is not common due to further reduction taking place leading to a 6-electron reduction by C–O bond cleavage. In 2013, Cantat and co-workers reported the TBD-catalyzed 4-electron reduction of CO₂ to give aminals 140 with the formation of two C–N bonds under atmospheric pressure CO₂ (1 atm) (Scheme 60).⁸⁰ Although this work focused on cross-coupling of CO₂ and two amines, the application of diamino substrates also provided the corresponding heterocycles in excellent yield. The reaction starts with reduction of CO₂ with PhSiH₃ to give a silylacetal species. The formed bis(silyl)acetal undergoes successive nucleophilic attack by amines rapidly to deliver the product. Later, He⁸¹ and Han⁸² achieved the glycine betaine-catalyzed 4-electron reduction of CO₂, which was captured by amines, respectively. Although heterocycles were not synthesized in these works, it possesses great potential for 4-electron reductive cyclization.

Scheme 60 TBD-catalyzed reductive cyclization of diamino substrates with CO₂.

Based on a similar process, Bontemps and co-workers also described iron-catalyzed reductive functionalization of CO₂ with 9-BBN (9-borabicyclo[3.3.1]nonane) (Scheme 61).⁸³ CO₂ was reduced to bis(boryl)acetals and no further reduction product was observed in a short reaction time. Nucleophiles were added to functionalize the resulting intermediates 142in situ. Dihydroimidazoles 143 were obtained after addition of N¹,N²-dimethylbenzene-1,2-diamines while addition of NH₃ gave hexamethylenetetramines 144 straightforwardly.

Scheme 61 Iron-catalyzed 4-electron reductive cyclization with CO₂.

More recently, Xia and co-workers developed reductive CO₂ fixation to construct spiroindolepyrrolidines 146via a 4-electron reduction with the formation of C–N and C–C bonds (Scheme 62).⁸⁴ The mechanistic study revealed a pathway involving CO₂ fixation. CO₂ is reduced in the presence of PhSiH₃ and TBD via a 4-electron process to deliver bis(silyl)acetal I-83. Subsequently, the acetal is captured by the amino group and the C3 carbon in tryptamine 145 with the formation of C–N and C–C bonds to afford spiroindolepyrrolidine 146 following dearomatization.

Scheme 62 TBD-catalyzed reductive cyclization of tryptamines with CO₂.

Besides the N atom, other heteroatoms could also attack the acetal species. Recently, Klankermayer's group reported the Ru- and Co-based co-catalyst for the synthesis of dimethoxymethane with CO₂ as a methylene source.⁸⁵ Even though the S atom possesses excellent nucleophilicity, 4-electron and 6-electron reductive fixations on the S atom were not reported. In 2018, our group developed a 4-electron reduction of S-nucleophiles to prepare dithioacetals from thiophenols or thiols using the NaBH₄/I₂ system with CO₂ (1 atm) (Scheme 63).⁸⁶ When propane-1,3-dithiols 147 were used, 1,3-dithianes 148 could be easily obtained in 60%. NaBH₄ is activated by I₂ to give the reactive borane and reduce CO₂.

Scheme 63 NaBH₄/I₂ mediated reductive cyclization of propane-1,3-dithiols with CO₂.

Research involving CO₂-incorporated reductive cyclization concentrates on catalyst systems, which could reduce CO₂ to different oxidation states via 2-electron or 4-electron reduction. The corresponding intermediates could be captured by substrates containing two nucleophilic sites. Although the reduction of CO₂ has been well investigated, the combination of CO₂ reduction and cyclization still remains undeveloped.

5. Conclusions and perspectives

From the perspective of reaction mode, the core idea in carboxylative, carbonylative, and reductive cyclization involves a nucleophilic attack on CO₂ and transformation of the carboxylate to a stable compound. Based on this idea, and as a result of deeper understanding of the properties of CO₂ along with the development of catalytic systems, three reaction modes have been highlighted that have shown exciting growth.

In the past, CO₂-incorporated carboxylative cyclization was restricted to only a few substrates such as epoxides and aziridines. However, substrates have been extended to transform CO₂ elegantly in recent years due to imaginative design strategies developed by chemists. Consequently, a variety of heterocycles have been prepared such as functionalized 2-oxazolidinones, benzoxazin-2-ones, oxazolidinediones, 1,3,4-oxadiazole-2(3H)-ones, cyclic carbonates, isocoumarins, and even maleic anhydrides. The application of nucleophilic NHC-carbene ligands and organic bases also facilitates these types of cyclization. Although catalysts tend to activate substrates rather than CO₂, the reaction is feasible to provide an approach to highly functionalized heterocycles. Carbonylation utilizing CO₂ remained dormant until these years. Compared with the traditional toxic carbonyl source CO and phosgene, CO₂ is certainly a greener carbon source. Carbonylation conducted with sequential nucleophilic attack or rearrangement generates lactones and lactams. It is noteworthy that carbonylation involving inert sp² C–H bonds was realized, which shows the great potential of CO₂ in synthetic chemistry. In contrast to development in CO₂ reduction, few works have been accomplished in CO₂ reductive cyclization. More substrates with two nucleophilic sites to construct heterocycles are still lacking. The oxidation state of the central carbon is decisive in this transformation, which depends on the reducibility imparted by the hydrogen sources and catalysts. Although efficient catalysts such as FLPs, ionic liquids, and organic superbases have been utilized in reductive cyclization with accessible hydrogen sources, deeper exploration with different catalysts, ligands, and hydrogen sources is required.

In spite of the rapid development in CO₂-incorporated cyclization to prepare multifarious heterocycles in recent years, there still exist large challenges and potential for further CO₂ utilization. Besides the elegant design of more efficient and cheaper catalysts with higher TON and TOF, some points include

(1) Light- and electro-driven CO₂ transformation: light- and electro-chemistry has exhibited wide application, while light- or electro-triggered CO₂-incorporated organic synthesis has been rarely reported. The utilization of light and electricity in CO₂ transformation not only makes the transformations greener, but also broadens the substrate scope.

(2) Oxidation: besides a one-carbon source, CO₂ is a mild oxidant as well. Although carbon was not utilized in the oxidation, the by-products deriving from the deoxygenation of CO₂ are also valuable carbon sources for further functionalization.^51,52

(3) Protecting group: as an electrophile, CO₂ is able to react with amines to give carbamates, which weaken the nucleophilicity and reducibility of amines to facilitate downstream reaction.¹³ This property suggested a possible exploitation of CO₂ as a protecting group like Boc (t-butyloxy carbonyl) in a traceless path.⁸⁷

(4) Lewis acid: the central carbon possesses weak Lewis acidity. The addition of CO₂ to substrates could make the substrates more electron-deficient and reactive to facilitate the reactions such as cross-coupling^88a,b and oxidation.^88c

(5) CO₂ as a carbon source in the synthesis of important smaller and larger heterocycles: 4-membered rings such as β-lactams and larger heterocycles such as macrolides are of significance in medicinal chemistry, while CO₂-incorporated cyclization only focused on building 5-membered and 6-membered rings.

(6) The incorporation of rational substrate design and heterogeneous catalysts such as MOFs^14k,l and nanostructured materials³ may exhibit potential in CO₂ utilization.

We believe that further developments in this field will promote CO₂ as an ideal feedstock capable of turning waste into wealth.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 21472106, 91645120, and 21871163). We thank Bo Zhang and Wei Hang from Tsinghua University for the advice.

References

1. (a) T. Sakakura, J. Choi and H. Yasuda, Chem. Rev., 2007, 107, 2365–2387; (b) M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann and F. E. Kühn, Angew. Chem., Int. Ed., 2011, 50, 8510–8537; (c) K. Huang, C. L. Sun and Z. J. Shi, Chem. Soc. Rev., 2011, 40, 2435–2452; (d) R. Martín and A. W. Kleij, ChemSusChem, 2011, 4, 1259–1263; (e) F. Manjolinho, M. Arndt, K. Gooßen and L. J. Gooßen, ACS Catal., 2012, 2, 2014–2021; (f) L. Zhang and Z. Hou, Chem. Sci., 2013, 4, 3395–3403; (g) C. Maeda, Y. Miyazaki and T. Ema, Catal. Sci. Technol., 2014, 4, 1482–1497; (h) A. Pinaka and G. C. Vougioukalakis, Coord. Chem. Rev., 2015, 288, 69–97; (i) D. Yu, S. P. Teong and Y. Zhang, Coord. Chem. Rev., 2015, 293–294, 279–291; (j) Q. Liu, L. Wu, R. Jackstell and M. Beller, Nat. Commun., 2015, 6, 5933–5947; (k) B. Yu and L. He, ChemSusChem, 2015, 8, 52–62; (l) K. Sekine and T. Yamada, Chem. Soc. Rev., 2016, 45, 4524–4532; (m) S. Wang, G. Du and C. Xi, Org. Biomol. Chem., 2016, 14, 3666–3676; (n) M. Börjesson, T. Moragas, D. Gallego and R. Martin, ACS Catal., 2016, 6, 6739–6749; (o) A. Kleij and J. Rintjema, Synthesis, 2016, 3863–3878; (p) J. Vaitla, Y. Guttormsen, J. K. Mannisto, A. Nova, T. Repo, A. K. Bayer and H. Hopmann, ACS Catal., 2017, 7, 7231–7244; (q) Q. Song, Z. Zhou and L. He, Green Chem., 2017, 19, 3707–3728; (r) Z. Zhang, T. Ju, J. Ye and D. Yu, Synlett, 2017, 741–750; (s) L. Zhang and Z. Hou, Curr. Opin. Green Sustain. Chem., 2017, 3, 17–21; (t) T. Janes, Y. Yang and D. Song, Chem. Commun., 2017, 53, 11390–11398; (u) B. M. Bhanage and M. Arai, Transformation and Utilization of Carbon Dioxide, Springer-Verlag, Berlin, Heidelberg, 2014; (v) X. Lu, Carbon Dioxide and Organometallics, Springer International Publishing, Switzerland, 2016; (w) M. Aresta, Carbon Dioxide as Chemical Feedstock, Wiley-VCH, Weinheim, 2010.
2. K. Dong, R. Razzaq, Y. Hu and K. Ding, Top. Curr. Chem., 2017, 375, 23.
3. (a) S. Xie, Q. Zhang, G. Liu and Y. Wang, Chem. Commun., 2016, 52, 35–59; (b) L. Zhang, Z. Zhao and J. Gong, Angew. Chem., Int. Ed., 2017, 56, 11326–11353.
4. (a) M. Patel, S. S. Ko, R. J. McHugh, J. A. Markwalder, A. S. Srivastava, B. C. Cordova, R. M. Klabe, S. Erickson-Viitanen, G. L. Trainor and S. P. Seitz, Bioorg. Med. Chem. Lett., 1999, 9, 2805–2810; (b) P. R. Angibaud, M. G. Venet, W. Filliers, R. Broeckx, Y. A. Ligny, P. Muller, V. S. Poncelet and D. W. End, Eur. J. Org. Chem., 2004, 479–486; (c) X. Fan, J. Zi, C. Zhu, W. Xu, W. Cheng, S. Yang, Y. Guo and J. Shi, J. Nat. Prod., 2009, 72, 1184–1190; (d) L. Kumar, N. Lal, V. Kumar, A. Sarswat, S. Jangir, V. Bala, L. Kumar, B. Kushwaha, A. K. Pandey, M. I. Siddiqi, P. K. Shukla, J. P. Maikhuri, G. Gupta and V. L. Sharma, Eur. J. Med. Chem., 2013, 70, 68–77.
5. (a) X. Lu and D. J. Darensbourg, Chem. Soc. Rev., 2012, 41, 1462–1484; (b) N. Kielland, C. J. Whiteoak and A. W. Kleij, Adv. Synth. Catal., 2013, 355, 2115–2138; (c) B. Xu, J. Wang, J. Sun, Y. Huang, J. Zhang, X. Zhang and S. Zhang, Green Chem., 2015, 17, 108–122; (d) C. Martín, G. Fiorani and A. W. Kleij, ACS Catal., 2015, 5, 1353–1370.
6. (a) V. A. Peshkov, O. P. Pereshivko, A. A. Nechaev, A. A. Peshkov and E. V. Van der Eycken, Chem. Soc. Rev., 2018, 47, 3861–3898; (b) E. García-Egido, I. Fernández and L. Muñoz, Synth. Commun., 2006, 36, 3029–3042; (c) Y. Takeda, S. Okumura, S. Tone, I. Sasaki and S. Minakata, Org. Lett., 2012, 14, 4874–4877.
7. J. Ye, L. Song, W. Zhou, T. Ju, Z. Yin, S. Yan, Z. Zhang, J. Li and D. Yu, Angew. Chem., Int. Ed., 2016, 55, 10022–10026.
8. L. Zhu, J. Ye, M. Duan, X. Qi, D. Yu, R. Bai and Y. Lan, Org. Chem. Front., 2018, 5, 633–639.
9. J. Ye, L. Zhu, S. Yan, M. Miao, X. Zhang, W. Zhou, J. Li, Y. Lan and D. Yu, ACS Catal., 2017, 7, 8324–8330.
10. M. Wang, Y. Cao, X. Liu, N. Wang, L. He and S. Li, Green Chem., 2017, 19, 1240–1244.
11. Z. Yin, J. Ye, W. Zhou, Y. Zhang, L. Ding, Y. Gui, S. Yan, J. Li and D. Yu, Org. Lett., 2018, 20, 190–193.
12. L. Sun, J. Ye, W. Zhou, X. Zeng and D. Yu, Org. Lett., 2018, 20, 3049–3052.
13. S. Wang, X. Zhang, C. Cao, C. Chen and C. Xi, Green Chem., 2017, 19, 4515–4519.
14. (a) S. Hase, Y. Kayaki and T. Ikariya, Organometallics, 2013, 32, 5285–5288; (b) S. Yoshida, K. Fukui, S. Kikuchi and T. Yamada, Chem. Lett., 2009, 38, 786–787; (c) X. Liu, M. Wang, S. Wang, Q. Wang and L. He, ChemSusChem, 2017, 10, 1210–1216; (d) Y. Zhao, J. Qiu, L. Tian, Z. Li, M. Fan and J. Wang, ACS Sustainable Chem. Eng., 2016, 4, 5553–5560; (e) B. Yu, B. Cheng, W. Liu, W. Li, S. Wang, J. Cao and C. Hu, Adv. Synth. Catal., 2016, 358, 90–97; (f) B. Yu, D. Kim, S. Kim and S. H. Hong, ChemSusChem, 2017, 10, 1080–1084; (g) Y. Zhao, J. Qiu, Z. Li, H. Wang, M. Fan and J. Wang, ChemSusChem, 2017, 10, 2001–2007; (h) J. Hu, J. Ma, Q. Zhu, Z. Zhang, C. Wu and B. Han, Angew. Chem., Int. Ed., 2015, 54, 5399–5403; (i) H. Zhou, W. Zhang, C. Liu, J. Qu and X. Lu, J. Org. Chem., 2008, 73, 8039–8044; (j) Y. Wang, Y. Wang, W. Zhang and X. Lu, J. Am. Chem. Soc., 2013, 135, 11996–12003; (k) H. Yang, X. Zhang, G. Zhang and H. Fei, Chem. Commun., 2018, 54, 4469–4472; (l) G. Zhang, H. Yang and H. Fei, ACS Catal., 2018, 8, 2519–2525.
15. P. García-Domínguez, L. Fehr, G. Rusconi and C. Nevado, Chem. Sci., 2016, 7, 3914–3918.
16. W. J. Yoo and C. Li, Adv. Synth. Catal., 2008, 350, 1503–1506.
17. X. Gao, C. Gan, S. Liu, F. Zhou, H. Wu and J. Zhou, ACS Catal., 2017, 7, 8588–8593.
18. S. Li, J. Ye, W. Yuan and S. Ma, Tetrahedron, 2013, 69, 10450–10456.
19. K. Yamashita, S. Hase, Y. Kayaki and T. Ikariya, Org. Lett., 2015, 17, 2334–2337.
20. T. Ishida, S. Kikuchi, T. Tsubo and T. Yamada, Org. Lett., 2013, 15, 848–851.
21. S. Sharma, A. K. Singh, D. Singh and D. Kim, Green Chem., 2015, 17, 1404–1407.
22. T. Niemi, I. Fernandez, B. Steadman, J. K. Mannisto and T. Repo, Chem. Commun., 2018, 54, 3166–3169.
23. W. Zhang, T. Xia, X. Yang and X. Lu, Chem. Commun., 2015, 51, 6175–6178.
24. C. Guo, W. Zhang, N. Zhang and X. Lu, J. Org. Chem., 2017, 82, 7637–7642.
25. W. Zhang, N. Zhang, Y. Sun, Y. Ding and X. Lu, ACS Catal., 2017, 7, 8072–8076.
26. (a) X. Tang, C. Qi, H. He, H. Jiang, Y. Ren and G. Yuan, Adv. Synth. Catal., 2013, 355, 2019–2028; (b) Q. Song, B. Yu, X. Li, R. Ma, Z. Diao, R. Li, W. Li and L. He, Green Chem., 2014, 16, 1633–1638; (c) Y. Kayaki, M. Yamamoto and T. Ikariya, Angew. Chem., Int. Ed., 2009, 48, 4194–4197; (d) N. D. Ca, B. Gabriele, G. Ruffolo, L. Veltri, T. Zanetta and M. Costa, Adv. Synth. Catal., 2011, 353, 133–146; (e) Y. Kayaki, M. Yamamoto and T. Ikariya, J. Org. Chem., 2007, 72, 647–649; (f) Y. Gu, F. Shi and Y. Deng, J. Org. Chem., 2004, 69, 391–394.
27. B. A. Vara, T. J. Struble, W. Wang, M. C. Dobish and J. N. Johnston, J. Am. Chem. Soc., 2015, 137, 7302–7305.
28. S. Sun, B. Wang, N. Gu, J. Yu and J. Cheng, Org. Lett., 2017, 19, 1088–1091.
29. J. Rintjema, R. Epping, G. Fiorani, E. Martín, E. C. Escudero-Adán and A. W. Kleij, Angew. Chem., Int. Ed., 2016, 55, 3972–3976.
30. B. Carry, L. Zhang, M. Nishiura and Z. Hou, Angew. Chem., Int. Ed., 2016, 55, 6257–6260.
31. S. Kikuchi, K. Sekine, T. Ishida and T. Yamada, Angew. Chem., Int. Ed., 2012, 51, 6989–6992.
32. Y. Sadamitsu, K. Komatsuki, K. Saito and T. Yamada, Org. Lett., 2017, 19, 3191–3194.
33. K. Sekine, Y. Sadamitsu and T. Yamada, Org. Lett., 2015, 17, 5706–5709.
34. W. Yoo, M. G. Capdevila, X. Du and S. Kobayashi, Org. Lett., 2012, 14, 5326–5329.
35. Z. Xin, C. Lescot, S. D. Friis, K. Daasbjerg and T. Skrydstrup, Angew. Chem., Int. Ed., 2015, 54, 6862–6866.
36. C. W. Kee, K. Q. E. Peh and M. W. Wong, Chem. – Asian J., 2017, 12, 1780–1789.
37. W. Yoo, T. V. Q. Nguyen and S. Kobayashi, Angew. Chem., Int. Ed., 2014, 53, 10213–10217.
38. J. Hou, A. Ee, W. Feng, J. Xu, Y. Zhao and J. Wu, J. Am. Chem. Soc., 2018, 140, 5257–5263.
39. T. Fujihara, Y. Horimoto, T. Mizoe, F. B. Sayyed, Y. Tani, J. Terao, S. Sakaki and Y. Tsuji, Org. Lett., 2014, 16, 4960–4963.
40. Y. Li, Y. Hu and X. Wu, Chem. Soc. Rev., 2018, 47, 172–194.
41. C. Lescot, D. U. Nielsen, I. S. Makarov, A. T. Lindhardt, K. Daasbjerg and T. Skrydstrup, J. Am. Chem. Soc., 2014, 136, 6142–6147.
42. T. Amaya, I. Kurata and T. Hirao, Org. Chem. Front., 2016, 3, 929–933.
43. Z. Zhang, L. Liao, S. Yan, L. Wang, Y. He, J. Ye, J. Li, Y. Zhi and D. Yu, Angew. Chem., Int. Ed., 2016, 55, 7068–7072.
44. S. Wang, P. Shao, G. Du and C. Xi, J. Org. Chem., 2016, 81, 6672–6676.
45. S. Sun, W. Hu, N. Gu and J. Cheng, Chem. – Eur. J., 2016, 22, 18729–18732.
46. M. Brahmayya, S. A. Dai and S. Y. Suen, RSC Adv., 2015, 5, 65351–65357.
47. B. Yu, H. Zhang, Y. Zhao, S. Chen, J. Xu, L. Hao and Z. Liu, ACS Catal., 2013, 3, 2076–2082.
48. Y. Zhao, Y. Wu, G. Yuan, L. Hao, X. Gao, Z. Yang, B. Yu, H. Zhang and Z. Liu, Chem. – Asian J., 2016, 11, 2735–2740.
49. M. Xia, W. Hu, S. Sun, J. Yu and J. Cheng, Org. Biomol. Chem., 2017, 15, 4064–4067.
50. Q. Song, Z. Zhou, M. Wang, K. Zhang, P. Liu, J. Xun and L. He, ChemSusChem, 2016, 9, 2054–2058.
51. Q. Song, Z. Zhou, H. Yin and L. He, ChemSusChem, 2015, 8, 3967–3972.
52. Z. Zhou, Q. Song and L. He, ACS Omega, 2017, 2, 337–345.
53. L. Song, G. Cao, W. Zhou, J. Ye, Z. Zhang, X. Tian, J. Li and D. Yu, Org. Chem. Front., 2018, 5, 2086–2090.
54. A. Del Vecchio, F. Caille, A. Chevalier, O. Loreau, K. Horkka, C. Halldin, M. Schou, N. Camus, P. Kessler, B. Kuhnast, F. Taran and D. Audisio, Angew. Chem., Int. Ed., 2018, 57, 9744–9748.
55. (a) J. Ma, B. Han, J. Song, J. Hu, W. Lu, D. Yang, Z. Zhang, T. Jiang and M. Hou, Green Chem., 2013, 15, 1485–1489; (b) W. Lu, J. Ma, J. Hu, J. Song, Z. Zhang, G. Yang and B. Han, Green Chem., 2014, 16, 221–225; (c) V. B. Saptal and B. M. Bhanage, ChemCatChem, 2016, 8, 244–250; (d) Y. Zhao, B. Yu, Z. Yang, H. Zhang, L. Hao, X. Gao and Z. Liu, Angew. Chem., Int. Ed., 2014, 53, 5922–5925; (e) Y. Xiao, X. Kong, Z. Xu, C. Cao, G. Pang and Y. Shi, RSC Adv., 2015, 5, 5032–5037.
56. T. Ishida, S. Kikuchi and T. Yamada, Org. Lett., 2013, 15, 3710–3713.
57. T. Ishida, R. Kobayashi and T. Yamada, Org. Lett., 2014, 16, 2430–2433.
58. C. Guo, W. Zhang, S. Liu and X. Lu, Catal. Sci. Technol., 2014, 4, 1570–1577.
59. B. Wang, S. Sun, J. Yu, Y. Jiang and J. Cheng, Org. Lett., 2017, 19, 4319–4322.
60. P. Mampuys, H. Neumann, S. Sergeyev, R. V. A. Orru, H. Jiao, A. Spannenberg, B. U. W. Maes and M. Beller, ACS Catal., 2017, 7, 5549–5556.
61. P. Xu, F. Wang, T. Wei, L. Yin, S. Wang and S. Ji, Org. Lett., 2017, 19, 4484–4487.
62. Z. Zhang, T. Ju, M. Miao, J. Han, Y. Zhang, X. Zhu, J. Ye, D. Yu and Y. Zhi, Org. Lett., 2017, 19, 396–399.
63. L. Song, L. Zhu, Z. Zhang, J. Ye, S. Yan, J. Han, Z. Yin, Y. Lan and D. Yu, Org. Lett., 2018, 20, 3776–3779.
64. G. Shen, W. Zhou, X. Zhang, G. Cao, Z. Zhang, J. Ye, L. Liao, J. Li and D. Yu, Chem. Commun., 2018, 54, 5610–5613.
65. K. Sasano, J. Takaya and N. Iwasawa, J. Am. Chem. Soc., 2013, 135, 10954–10957.
66. T. Kaicharla, M. Thangaraj and A. T. Biju, Org. Lett., 2014, 16, 1728–1731.
67. W. Zhang, S. Liu and X. Lu, Beilstein J. Org. Chem., 2015, 11, 906–912.
68. W. Zhang, M. Yang and X. Lu, Green Chem., 2016, 18, 4181–4184.
69. Z. Zhang, C. Zhu, M. Miao, J. Han, T. Ju, L. Song, J. Ye, J. Li and D. Yu, Chin. J. Chem., 2018, 36, 430–436.
70. L. Fu, S. Li, Z. Cai, Y. Ding, X. Guo, L. Zhou, D. Yuan, Q. Sun and G. Li, Nat. Catal., 2018, 1, 469–478.
71. Z. Cai, S. Li, Y. Gao and G. Li, Adv. Synth. Catal., 2018, 360, 4005–4011.
72. A. Tlili, E. Blondiaux, X. Frogneux and T. Cantat, Green Chem., 2015, 17, 157–168.
73. O. Jacquet, C. Das Neves Gomes, M. Ephritikhine and T. Cantat, ChemCatChem, 2013, 5, 117–120.
74. B. Yu, H. Zhang, Y. Zhao, S. Chen, J. Xu, C. Huang and Z. Liu, Green Chem., 2013, 15, 95–99.
75. X. Gao, B. Yu, Z. Yang, Y. Zhao, H. Zhang, L. Hao, B. Han and Z. Liu, ACS Catal., 2015, 5, 6648–6652.
76. Z. Yang, B. Yu, H. Zhang, Y. Zhao, G. Ji, Z. Ma, X. Gao and Z. Liu, Green Chem., 2015, 17, 4189–4193.
77. Z. Zhang, Q. Sun, C. Xia and W. Sun, Org. Lett., 2016, 18, 6316–6319.
78. B. Zhang, G. Du, W. Hang, S. Wang and C. Xi, Eur. J. Org. Chem., 2018, 1739–1743.
79. X. Wu, S. Sun, B. Wang and J. Cheng, Adv. Synth. Catal., 2017, 359, 3855–3859.
80. X. Frogneux, E. Blondiaux, P. Thuéry and T. Cantat, ACS Catal., 2015, 5, 3983–3987.
81. X. Liu, X. Li, C. Qiao, H. Fu and L. He, Angew. Chem., Int. Ed., 2017, 56, 7425–7429.
82. C. Xie, J. Song, H. Wu, B. Zhou, C. Wu and B. Han, ACS Sustainable Chem. Eng., 2017, 5, 7086–7092.
83. G. Jin, C. G. Werncke, Y. Escudie, S. Sabo-Etienne and S. Bontemps, J. Am. Chem. Soc., 2015, 137, 9563–9566.
84. D. Zhu, L. Fang, H. Han, Y. Wang and J. Xia, Org. Lett., 2017, 19, 4259–4262.
85. (a) K. Thenert, K. Beydoun, J. Wiesenthal, W. Leitner and J. Klankermayer, Angew. Chem., Int. Ed., 2016, 55, 12266–12269; (b) B. G. Schieweck and J. Klankermayer, Angew. Chem., Int. Ed., 2017, 56, 10854–10857.
86. Z. Guo, B. Zhang, X. Wei and C. Xi, Org. Lett., 2018, 20, 6678–6681.
87. (a) J. Ye, I. Kalvet, F. Schoenebeck and T. Rovis, Nat. Chem., 2018, 10, 1037–1041; (b) M. Kapoor, D. Liu and M. C. Young, J. Am. Chem. Soc., 2018, 140, 6818–6822.
88. (a) S. B. Lang, T. M. Locascio and J. A. Tunge, Org. Lett., 2014, 16, 4308–4311; (b) G. Pupo, R. Properzi and B. List, Angew. Chem., Int. Ed., 2016, 55, 6099–6102; (c) D. Riemer, B. Mandaviya, W. Schilling, A. C. Götz, T. Kühl, M. Finger and S. Das, ACS Catal., 2018, 8, 3030–3034.

---