Aerobic Oxidative N‐Heterocyclic Carbene Catalysis

The ease with which simple starting materials can be transformed into highly functionalized products has made oxidative N‐heterocyclic carbene (NHC) catalysis an area of significant interest. However, the use of stoichiometric amounts of high molecular weight oxidants in most reactions generates an undesired equivalent amount of waste. To address this issue, the use of oxygen as the terminal oxidant in NHC catalysis has been developed. Oxygen is attractive due to its low cost, low molecular weight, and ability to generate water as the sole by‐product. However, molecular oxygen is challenging to use as a reagent in organic synthesis due to its unreactive ground state, which often requires reactions to be run at high temperatures and results in the formation of kinetic side‐products. This review covers the development of aerobic oxidative carbene catalysis, including NHC‐catalyzed reactions with oxygen, strategies for oxygen activation, and selectivity issues under aerobic conditions.


Introduction
N-Heterocyclic carbenes (NHCs) have recently surfaced as versatile homogeneous catalysts and are best known for their efficiency in umpolung (inversion of polarity) reactions of aldehydes such as the Benzoin and Stetter reactions. [1] In recent years, the scope of applications for NHCs has expanded significantly, encompassing a wide range of diverse reactions. [2][3][4][5][6][7][8][9][10] In oxidative NHC catalysis, the NHC is combined with an oxidant generally resulting in a reversal of the umpolung reactivity. The oxidative reactivity of NHC catalysis has received widespread attention and several unique reaction types have been developed under this reaction manifold. [11][12][13][14] However, the oxidants used are normally stoichiometric high molecular weight quinone-based oxidants [15] that negatively impacts the atom economy (AE), purification protocols, and operational simplicity. In aerobic oxidative NHC catalysis, which is the focus of this review, oxygen (O 2 ) is used as terminal oxidant instead of high molecular weight oxidants. Oxygen is an ideal oxidant as it is abundant, inexpensive, releases water as byproduct and has a relatively low molecular weight, thus improving the AE of the reaction. However, utilization of oxygen as oxidant is an arduous task as it is unreactive and requires high temperatures to work often resulting into unwanted side-products. [16,17] Nevertheless, a number of important advancements have been made in this area which will be highlighted in this review. The review is divided into sections based on the key intermediates (Scheme 1): oxidation via the Breslow intermediate (section 3), oxidation via the aza-Breslow intermediate (section 4) and oxidation via the deoxy-Breslow intermediate (section 5). Each of these sections is further divided into categories based on the oxidation method employed. These categories include oxida-tion with air alone, aerobic oxidation via a metal-based redox active catalyst, aerobic oxidation with organocatalytic redox active species, utilization of electron transfer mediators (ETMs) and photocatalyzed oxidation combined with air.

Structure and Reactivity of NHC
A carbene is a compound with a neutral bivalent carbon atom and is regarded as a very reactive species in organic synthesis with a short lifetime. However, it is possible to stabilize carbenes by the introduction of flanking heteroatoms enabling resonance stabilization through the corresponding ylide (Scheme 2A) as first reported by Bertrand. [18] A strategy to further increase the persistence of the carbene is to install the carbene into a heterocyclic structure as described by Arduengo and co-workers (Scheme 2B). [19] These types of carbenes are commonly known as NHCs. The nitrogen atoms adjacent to the carbene serves to make the NHCs more stable than the general alkyl based carbene. The nitrogen lone pairs stabilize the empty p-orbital and another stability factor is the sigma electron withdrawing effect of the adjacent electronegative atoms. As a result, the energy of the highest occupied molecular orbital (HOMO) is decreased while the energy of the lowest unoccupied molecular orbital (LUMO) is increased, hence favoring the singlet state over the triplet state of the carbene (Scheme 2C). Furthermore, the inclusion of the carbene into a cyclic structure imposes an sp 2 like hybridization on the carbene carbon further favoring the singlet state. Having both electrons in the exposed vacant p-orbital make the carbene nucleophilic. An additional feature commonly seen for NHCs is that they usually contain bulky R-groups adjacent to the carbene carbon. The bulky R-groups help stabilize the carbene and prevent the dimerization, known as the Wanzlick equilibrium (Scheme 2D). [20][21][22] Utilization of NHCs in synthesis demands generation of the NHC in situ from a stable precursor, the most common method is by deprotonation of a precursor salt. The NHC precursor salts can be classified by their cyclic skeleton which determines their reactivity, namely: imidazolium, thiazolium and triazolium (Scheme 2E). The base that is used for deprotonation, depends on the pK a of the corresponding salt. [23] The pK a value is around 16.5 for thiazolium, 17-28 for imidazolium and 12-15.5 for triazolium precursors (in DMSO). [24] When the NHC 2 reacts with an aldehyde 1 it forms the Breslow intermediate 3. The Breslow intermediate is the common starting point for several NHC-catalyzed reactions (Scheme 3). [20] Examples of pioneering reactions are the Benzoin condensation and the Stetter reaction. The NHC catalyzed benzoin condensation was developed by Ukai et al, [25] however the mechanism was discovered by Breslow in 1958 who used vitamin B1 as the carbene precursor (4, Scheme 4). [26] The Breslow intermediate 3 is nucleophilic in character and adds to the carbonyl carbon of an aldehyde to give intermediate 5.
Intermediate 5 then fragments to give benzoin 6 and the carbene is regenerated. During this process the aldehyde carbonyl carbon is converted from being electrophilic to nucleophilic in the Breslow intermediate. This is the archetypical example of an umpolung.
In the Stetter reaction the Breslow intermediate attacks Michael acceptors such as alkenes containing electron withdrawing substituents. [27] Although the Breslow intermediate is primarily known for its participation in umpolung chemistry, it can under oxidative conditions be transformed into different reactive intermediates leading to other reaction pathways. [28]

Oxidative NHC Catalysis
In the presence of an oxidant the electron rich Breslow intermediate can be oxidized into an acyl azolium intermediate 7.
The process can be viewed as a double umpolung. Several external oxidants are known to promote this reaction, but the most common oxidant is the Kharasch oxidant 19. The acyl azolium is electrophilic at the carbonyl carbon and undergoes acyl substitutions with nucleophiles (Scheme 5).
Alternatively, the acyl azolium can be converted to an azolium enolate 8 by α-deprotonation of the acyl azolium, turning into a reactive nucleophile at the α-position (Scheme 6A). [29,30] When the substrate reacting with the carbene is an α,βunsaturated aldehyde 9 an extended Breslow intermediate is formed also known as the homoenolate 10. The subsequent oxidation of the homoenolate generates the α,β-unsaturated acyl azolium 11 which is electrophilic at the acyl and β-carbon (Scheme 6B). The α,β-unsaturated acyl azolium is a central reactive intermediate in NHC catalysis and several reactions can be derived from this intermediate. [31] If the α,β-unsaturated acyl azolium contains an enolizable proton at the γposition a nucleophilic azolium dienolate 12 can be generated after deprotonation with a base enabling γ-carbon activation that can be used in for example formal (4 + 2) and (3 + 3) annulation reactions. [32][33][34] Although oxidative carbene catalysis is widely used in many transformations, it requires a stoichiometric amount of a high molecular weight oxidant, which results in poor atom economy protocols and potentially difficult purifications. However, these negative aspects can be avoided by utilizing oxygen as the terminal oxidant. Oxygen is a low molecular weight, non-hazardous, and inexpensive oxidant, making it a preferable alternative.

Aerobic Oxidation in NHC-Catalysis -Oxidative and Oxygenative Pathway
Aerobic oxidation of the Breslow intermediate can proceed through two routes -the oxidative pathway and the oxygenative pathway (Scheme 7). [16] Mechanistically, both routes start with single electron transfer (SET) from the Breslow intermediate 3 to oxygen to form a radical ion pair 13 that recombines to either form a hydroperoxide 14 or peroxide ion 16. The hydroperoxide 14 proceeds to the oxidative path and fragments to form the acyl azolium 7 that can react with a nucleophile to give the carbonyl substitution product 15 and the carbene. In the oxygenative pathway the peroxide ion 16 reacts with a second aldehyde to form a carboxylate ion 17. In the next step 17 undergoes a Criegee-type fragmentation to produce a carboxylic acid and an oxo-Breslow structure 18 that fragments to form a second equivalent of carboxylic acid and the carbene. [35] When an external oxidant is used under anaerobic conditions the oxidation of the Breslow intermediate proceeds directly to the acyl azolium. External oxidants often used are stoichiometric oxidants azobenzene, [36] TEMPO [37] or metals such as MnO 2. [38,39] In 2010 Studer showed that the quinone 19 works as a new mild stoichiometric oxidant (first described by Kharasch in 1957) [15a,b] in the oxidative esterification of unsaturated aldehydes (Scheme 8). [40] The Kharasch oxidant 19 is now the most common oxidant in NHC-catalysis, it is highly selective in the oxidation of the Breslow intermediate but when it is used in equimolar amounts the reactions will suffer from low AE due to its high molecular weight (408.63 g/mol). [41]

Oxygen as Terminal Oxidant
In the search for environmentally benign oxidants, molecular oxygen (O 2 ) and hydrogen peroxide (H 2 O 2 ) are good candidates as they have a high efficiency of weight per oxidant and lead to non-hazardous by-products (typically water). Hydrogen peroxide is used as an oxidant on the industrial level but is explosive and corrosive which makes it precarious to handle. Molecular oxygen on the other hand can be an attractive alternative, especially if it is used in its aerial form. Aerial oxygen, as a reagent in organic synthesis is abundant, non-toxic and free of charge. However, large scale reactions with aerobic oxidations are linked to safety issues, nevertheless risk mitigation is possible with for instance flow reactor techniques [42] or by running reactions with low volatile solvents such as deep eutectic solvents or ionic liquids. [43] Molecular oxygen has a triplet ground state making it a diradical molecule. Radicals are typically very reactive and usually requires stabilization by either steric or electronic means to be isolable but molecular oxygen is stable enough to constitute about 21 % of the earth's atmosphere. The stability of molecular oxygen is kinetic rather than thermodynamic as it reacts exothermically with every element except gold. [44,45] Aerobic oxidations are thus generally characterized by favorable thermodynamics and high reaction barriers needing high reaction temperatures to be efficient. This makes aerobic oxidations in synthetic organic chemistry difficult as high reaction temperatures usually result in kinetic side products. In catalysis this is known as the oxidation problem, i. e. the electron transfer between the catalysts and oxygen is too slow compared to the fast deactivation of the catalysts. [46] Scheme 7. Oxidative and Oxygenative pathway. When oxygen is used as an oxidant the oxidative path proceeds through an acyl azolium whereas in the oxygenative through peroxide anion ion and then a carboxylate ion, producing carboxylic acids. The Breslow intermediate can also transform into the acyl azolium directly with an external oxidant.

The Development of Electron Transport Mediators (ETMs)
Nature has on the other hand engineered a way to perform mild oxidation in a controlled manner such as in the electron transport chain in the respiratory system. Electrons are transported via redox reaction through a series of redox donors and acceptors which ultimately leads to the oxidation of NADH by oxygen to generate energy in the form of ATP (Scheme 9). This strategy has been mimicked by chemists and used today for the oxidation of compounds in organic synthesis, the most prominent example being the Wacker Oxidation. [47] The Wacker process, created 1956, is one of the early examples of aerobic oxidation using a system of electron transfer mediators, ETMs, and is used in large-scale industrial synthesis of acetaldehyde 24 from ethylene 23 (Scheme 10). The catalyst in this aerobic oxidation is a metal (M n + 2 ) which oxidizes a substrate and transforms into its reduced form (M n ), the metal is coupled to one or several ETMs which reoxidize the catalyst to its M n + 2 state. The reduced ETM ultimately gets oxidized by oxygen. In the Wacker-Process CuCl 2 serves as an ETM between Pd 0 and oxygen, regenerating Pd 2 + (which can react with another ethylene in the cycle). [48] With the Wacker process acetaldehyde, an important bulk chemical, can be produced in multi tone scale. However, the presence of chloride ions causes formation of unwanted side-products so alternative processes have been developed.
Bäckvall and co-workers developed a milder Wacker oxidation for the creation of terminal ketones which were chloride free and 16 times faster than the original Wacker Process (Scheme 11). In this process the substrate specific catalyst, Pd(OAc) 2 , is linked to two ETMs -a hydroquinone 27 and a metal-cyclic complex (iron(II) phthalocyanine 26 (FePc) or Co(salophen)). [49] 3. Aerobic Oxidative NHC-Catalysis via the Breslow Intermediate

Oxidation of the Breslow Intermediate Using Air
In 2006, Chen and co-workers performed the NHC catalyzed chemoselective ring opening of N-tosyl aziridines 28 with aldehydes under aerobic conditions (Scheme 12). [50] The idea was to investigate whether the N-tosyl bicyclic aziridine could be selectively attacked by the aldehyde in competition with self-benzoin condensation.
The acyl anion was expected to open the ring and generate a β-amino ketone, but after attack by the harder Breslow oxygen anion 31 on the aziridine carbon of 28 and oxidation of the Breslow intermediate 32 by oxygen, the unexpected 1,2amino alcohol derivative 29 was formed with a yield of 15 %. After further optimization, yields up to 90 % were reported. Air alone was not sufficient to oxidize the aldehydes, the Scheme 9. The electron transport chain in the respiratory system.

R e v i e w T H E C H E M I C A L R E C O R D
aziridines were needed, which ruled out that the mechanism was first the formation of carboxylate and then aziridine ring opening. Hazra and co-workers (2009) performed the oxidative esterification of heterocyclic aldehydes 35 with thiamine hydrochloride 4 (vitamin B1) as the pre-NHC catalyst (Scheme 13). [51] Under aerobic refluxing conditions in the presence of triethylamine and dry methanol or ethanol the esters 36 could be obtained in high yield. Only highly electron withdrawing aldehydes were able to react, and the reaction times are short (2-3 hours). The authors propose a mechanism which proceeds via the oxidative pathway -the Breslow intermediate gets oxidized to the acyl azolium with oxygen, which ultimately is attacked by the alcohol to give the carbonyl substitution product.
In 2010, Deng and co-workers reported the NHCcatalyzed synthesis of α-aryl ketones under nitrogen atmosphere. When tested under aerobic conditions to elucidate the mechanism, the ester product was unexpectedly seen with yields up to 45 %. [52] One of the early examples of aerobic oxidative NHC catalysis was reported by Liu and co-workers (2011, Scheme 14). [53] The reaction relies on aerial oxygen as the terminal oxidant and primarily proceeds through the oxygenative path providing a carboxylic acid that is transformed into an ester 39 in a subsequent alkylation step. An isotope labelling experiment with 18  In 2011, Xin and co-workers performed a similar esterification as Liu, but using unactivated alkylbromides. [54] Both air and 1 atm pure oxygen atmosphere were tested, where the pure oxygen atmosphere generated a slightly better yield (83 % vs 75 %).
An aerobic oxidative NHC catalyzed reaction was reported by Youn et al (2011) who used the strategy of converting oalkynyl benzaldehydes 45 to carboxylic acids in their intramolecular alkyne cyclization to give phthalides 46 and isocoumarins 47 (Scheme 15A). [55] Anand and co-workers (2012) [56] were able to show that aldehydes could be converted to phenyl esters in the presence of boronic acids (Scheme 15B). In the suggested mechanism, In 2013, Fu and co-workers reported the bis-NHC catalyzed aerial oxidation of aryl aldehydes 58 to carboxylic acids 59 with good yields (up to 91 %). (Scheme 16). [57] The authors reported that the length of the alkyl chain spacer in the pre-NHC 60 exerted a significant influence on the rate of the reaction, but did not specify the purpose of using two imidazolium moieties in the catalyst. The transformation by Fu and co-workers demanded long reaction times (2-3 days).
In 2016, Zhong and co-workers reported an aerobic oxidative annulation of α,β-unsaturated aldehydes 37 with 1,3-dicarbonyl compounds 61 for the asymmetric synthesis of dihydropyranones 62 (Scheme 17), [58] a reaction that was originally reported with the Kharasch oxidant. [59] In their report air was used as a terminal oxidant, providing the dihydropyranones in yields up to 82 %. Drawbacks were the requirement for high NHC-precatalysts loadings (20 mol%), the usage of LiCl as additive, molecular sieves to remove the water and excess aldehyde to take care of the peroxide anion formed in the oxidation of the Breslow intermediate. In their postulated mechanism, the electron rich Breslow intermediate 64 initially forms after reaction between the carbene and the enal. In the presence of oxygen, the Breslow intermediate oxidizes to the acyl azolium 65, which acts as a Michael acceptor. The 1,3-dicarbonyl 66 adds to the acyl azolium 65 via conjugate addition and after enolization and cyclization the dihydropyranone 67 is formed.
In 2017 Bortolini et al reported that position of substituents on the aromatic aldehyde affects weather the reaction goes via the oxidative or the oxygenative path. [35] It was shown that a bromine substituent in the ortho-position to the aldehyde 68 leads to the oxygenative product 69 (Scheme 18A) whereas para-substituted aldehydes 71 give the oxidative product 72 (Scheme 18B). Mechanistically, the position of the bromo substituents leads to two distinct peroxidic tautomeric structures with divergent reactivity and thus different reaction pathways.

R e v i e w T H E C H E M I C A L R E C O R D
A more recent synthesis employing the oxygenative pathway is achieved for the oxidation of furfural 73 to furoic acid 74 by Nakajima (2018) using air as oxidant (Scheme 19). [60] In their report furfural is converted to the acid with a yield of 69 % in less than 20 minutes and 99 % after 4 h at room temperature. Interestingly, while the NHC also promotes parasitic furfural dimerization to furoin 75 the reversibility of the benzoin condensation eventually leads to quantitative formation of the furoic acid.
In 2021, Xu and co-workers used vitamin B1 4 as a pre-NHC catalyst to perform the esterification of aryl aldehydes 58. [61] Aerial oxygen was used as a terminal oxidant (Scheme 20). The reactions were performed at elevated temperatures (60-80°C) and the reaction times between 12 to 24 h. The esters 76 are obtained with moderate to excellent yields.
Worth mentioning in the context of using air as oxidant is the work by Connon and Zeitler (2013, Scheme 21). [62] In their aerobic NHC-catalyzed oxidative esterification of aldehydes 53 they proposed that the oxidation of the Breslow intermediate 80 does not proceed via an acyl azolium intermediate. Instead, the Breslow intermediate reacts with another aldehyde molecule to form benzoin 81, which is the active species to get oxidized by oxygen in the presence of base to give benzil 82. In the next step the electrophilic diketone 82 is attacked by the NHC to give tetrahedral intermediate 83 that reacts with the alcohol and fragments to ester product 77 plus another molecule of the Breslow intermediate 80. Thus, the esterification does not proceed via the oxidative or the oxygenative pathway but instead a third benzoin pathway. Both alcohols and water can be used as the nucleophile in the reaction to afford esters and acids respectively in excellent yields (up to 94 % for the esters and 95 % for acids).

Aerobic Oxidation of the Breslow Intermediate Using a Metal Redox Active Catalyst
In 2010, Gois and co-workers reported an aerobic oxidative aromatic esterification of aldehydes with boronic acids 84 using iron(II) and air as oxidant with yields up to 97 % (Scheme 22A). [63] In 2011, Gois and co-workers also performed esterification of aryl aldehydes to generate phenyl esters 85 via oxidative NHC catalysis using iron(II) and air as terminal oxidant (Scheme 22B). [64] Good yields (up to 90 %) were observed for Scheme 18. Position of substituents on aromatic aldehyde affects the oxidation pathway.

Scheme 19. Creation of Furoic acid via the oxygenative pathway
Scheme 20. Vitamin B1 as pre-NHC-catalyst together with air as terminal oxidant.

Scheme 21.
Connor and Zeitler suggested that it is benzoin 81 that is reacting with oxygen, not the Breslow intermediate.

R e v i e w T H E C H E M I C A L R E C O R D
the combination of aromatic aldehydes and substituted phenols 86.
A similar aerobic oxidation using air with Pd(OAc) 2 instead of iron were performed by Chen and Cheng in 2011 (Scheme 22C). [65] In 2012, Pan and co-workers performed the ruthenium/ NHC catalyzed tandem oxidation/esterification of benzylic alcohols with phenols 86 using air as the terminal oxidant (Scheme 22D). [66] In the postulated mechanism it was suggested that the alcohol is first oxidized to the aldehyde by ruthenium followed by an NHC-catalyzed oxidative esterification using air.
Studer and co-workers (2013) performed a tandem NHC/ metal oxidation with aerial oxygen as terminal oxidant in their synthesis of esters 87 from aldehydes, using a ruthenium catalysts (Scheme 23). [67] In the postulated mechanism the Breslow intermediate 89 first gets oxidized by Ru(II) via SET to give the radical cation 90. Oxygen acts as terminal oxidant and reoxidizes the Ru(I) to Ru(II). Deprotonation of 90 with base or with the superoxide radical anion gives radical 91, which after further oxidation yields the acyl azolium 92.
In 2018, Huang and co-workers applied a similar concept to synthesize α,β-unsaturated esters and chiral lactones from enals, by oxidizing the homoenolate to the acyl azolium with RuCl 3 as the redoxactive catalyst and oxygen as the terminal oxidant (Scheme 24A). [32] In the oxidative esterification of α,β-unsaturated aldehydes the methyl ester 21 could be obtained with very good yield (80 %). The oxidant was however pure molecular oxygen and not air. The authors reported that if the oxidation with oxygen was slow, protonation of the β carbon leads to formation of the saturated ester 93 which might be the reason for not using aerial oxygen. In the same report the authors also reported that RuCl 3 can be used as the redox active catalyst for the aerobic oxidative NHC catalyzed synthesis of dihydropyranones 94 via 3 + 3 annulation, if the acyl azolium 98 reacts with a dicarbonyl 61 (Scheme 24B) and in a formal 4 + 2 annulation between enals and activated ketones (Scheme 24C). In the (4 + 2) annulation the acyl azolium 98 gets deprotonated by a base at the γcarbon to form the azolium dienolate 12, which reacts with the activated CF 3 ketone 96 to form lactone 97.
Inspired by Fu and co-workers protocol from 2013 (Scheme 16), Lee and co-workers (2020) performed a bis-NHC 99 catalyzed aerial oxidation of aryl aldehydes 58 to carboxylic acids 59 (Scheme 25). [68] Fu and co-workers reported long reaction times (2-3 days) and high catalyst loading (5 mol%). Lee and co-workers were able to show that by adding 1 mol% Pd(OAc) 2 as an additive in combination with oxygen the reaction time could be decreased to 24 hours and the NHC catalyst loading to 1 mol %. The reaction rate was suggested to be increased by utilizing two imidazolium moieties in the catalyst. Both Fu and Lee referenced to the work by Bode (2011), [69] who argued that carbon dioxide was involved in the oxidation by reducing the amount of sideproducts. Therefore, Lee also incorporated four equivalents of

R e v i e w T H E C H E M I C A L R E C O R D
formic acid in the reaction. Moreover, water was used as a solvent to increase the solubility of the bis-NHC.

Aerobic Oxidation of the Breslow Intermediate Using an Organocatalytic Redox Active Catalyst
The oxidation of the Breslow intermediate is also possible to perform with non-metal-based redox catalysts. The benefits of using metal-free catalysis are that there is no presence of tracemetals in the products, they are potentially cheaper and mining resources are not required for their production. [70] In 2011, Iida & Yashima reported the aerial enantioselective oxidative esterification of aldehydes mediated by a chiral NHC catalyst 95 and a redox active flavin 102 derived from riboflavin B2 (Scheme 26). [71] Excellent kinetic resolution of trans-cyclohexane-1,2-diol (> 99 % ee of unreacted diol) with 78 % conversion of alcohol to the ester was obtained. Desymmetrization of meso-diols was also performed with moderate selectivity (30 % yield, 64 % ee). The procedure is also applicable to the formation of amides and thioesters.
Another example of organoredox catalysis was developed by Jang et al (2014) who utilized the combination of air and TEMPO for oxidative NHC-catalysis in the esterification of allylic alcohols 103 (Scheme 27). [72] The reaction is performed in a one-pot reaction, where TEMPO first oxidizes the allylic alcohol to cinnamaldehyde 20, which then participates in the carbene-catalyzed oxidative esterification. In the NHC catalytic cycle TEMPO comes in a second time and oxidizes the Breslow intermediate 105 and intermediate 106 via SET mechanism to form the α,β-acyl azolium 107, which reacts

R e v i e w T H E C H E M I C A L R E C O R D
with the allylic alcohol 103 to form the ester product 104. TEMPOH is reoxidized by oxygen and hexafluoroisopropanol (HFIP) helps to accelerate the formation of cinnamaldehyde.

Oxidation Using a System of ETMs
As described in section 3.1, the selective oxidation of the Breslow intermediate or homoenolate by oxygen to the acyl azolium is a challenging process due to the competition between oxidative and oxygenative reaction pathways, which typically leads to poor selectivity and formation of by-products such as acids. These reactions are characterized by elevated temperatures or long reaction times (Scheme 28A). However, the addition of a redox active catalyst (section 3.2 & 3.3) can improve the situation, although these reactions typically require elevated temperatures or have a limited substrate scope (Scheme 28B). The elevated temperatures needed for these reactions are a direct result of the high energy barrier associated with oxygen's triplet ground-state, which is nature's way of preventing uncontrolled oxidation of organic compounds. In the context of NHC catalysis, this usually means that the oxidation of the Breslow intermediate is inefficient. A possible way to circumvent this issue is to mimic the mitochondrial electron transport chain and create a low energy path for electrons to flow from the substrate to oxygen. Synthetically, this is usually done by performing the oxidation in a stepwise manner by introducing several electron transport mediators (ETMs) between the substrate specific catalyst and oxygen (Scheme 28C). [48] In 2016, inspired by the work by Bäckvall, Sundén and coworkers developed a system of ETMs for the aerobic oxidative NHC-catalyzed esterification of α,β-unsaturated aldehydes 20 (Scheme 29A). [73] The group showed that 2,6-di-tert-butylphenol 108 in combination with FePc can be used as ETMs in the combination with NHCs, under aerial conditions to convert the homoenolate 64 to acyl azolium 65 which transforms a number of aldehydes to esters. It is important to mention that 108 is converted to the Kharasch oxidant 19 insitu. Compared to Studers oxidation in 2010 where a full equivalent of the stoichiometric Kharasch oxidant 19 was used as the only oxidant source, Sundén (2016) used 2 mol% of Kharasch oxidant as the first ETM and 0.6 mol% of FePc as the second ETM. The method has the advantages of good atom economy and waste elimination.
The same year, Sundén and co-workers employed a similar ETM system for the enantioselective synthesis of dihydropyranones 67 (Scheme 29B). [41] Showing that it is possible to carry out enantioselective aerobic oxidative NHC catalysis with ETMs. An exceedingly difficult task as the reaction conditions (solvents, temperature concentration and additives) must be designed to consider promoting both high enantioselectivity and conversion.
ETMs were also used in the aerobic oxidative NHC catalyzed N-acylation of oxazolidinones 109 (Scheme 29C). [74] The N-acylation of oxazolidinones (2018) was done at room temperature. The catalytic system is highly efficient and the reaction can be conducted with 1 mol% loading of the NHC 22, 3 mol% 19 and 2 mol% FePc at room temperature.

R e v i e w T H E C H E M I C A L R E C O R D
solketal 120 with aldehydes (Scheme 31B). [77] In their batch reaction, optimal results were obtained with an aerial oxidation system of electron transport mediators (ETMs) compared to using 100 mol% Kharasch oxidant alone. Furthermore, two pre-catalysts, silica, and polystyrene-supported pre-catalyst were synthesized and tested, with polystyrene pre-NHC 119 showing better performance. In the batch experiment with 119 both glycerol and solketal could be converted to the acetylated counterpart 118 and 121 with good to excellent yields. In the continuous-flow experiment, the utilization of ETM and air did not render good yields (15 %), due to low aerial oxygen concentration in the liquid phase inside the reactor. Instead, the 100 mol% Kharasch oxidant recycled by air alone gave the best result. In 2018, Massi and co-workers used the same reaction to synthesize poly-HMFCA 123 (Scheme 32) as a precursor for the synthesis of HMFCA 124. [78] HMFCA is a biomassderived furan derivative and is a versatile chemical platform as it can be used to synthesize polyesters, it is also an intermediate for the synthesis of a promising interleukin inhibitor. In Massi's previous study HMFCA was synthesized from HMF 122 with 100 mol% Kharasch oxidant 19 as the sole oxidant to a yield of 50 % HMFCA and 42 % poly-HMFCA. In their new paper, in the search for a greener alternative, they utilized Sundéns conditions with ETMs and synthesized poly-HMFCA 123 to a yield of 91 % with air as the terminal oxidant, a polystyrene pre-NHC 119 and Me-THF as a green solvent. The HMFCA, synthesized in the oxygenative pathway, was used as a nucleophile to target the acyl azolium 127 to generate poly-HMFCA. The poly-HMFCA could then easily be transformed with basic hydrolysis to HMFCA 124, or ester and amides derivatives which could be used in other transformations. In this study 5 mol% FePc alone worked equally well as 20 mol%

R e v i e w T H E C H E M I C A L R E C O R D
2,6-DTBP plus 5 mol% FePc. As for the previous study, the continuous flow NHC approach was also studied for various HMF oxidation products with good results, although using the Kharasch oxidant. In 2019, Massi and co-workers also tried the reaction conditions on synthesis of polyester oligomers 131 (Scheme 33) [79] and in 2020, for the asymmetric N-acylation of dihydropyrimidines 132. [80] Both experiments utilized air as terminal oxidant in combination with a ETM system, but with poor yields as the outcome (10 % of the polymer 131 and 20 % of the wanted R enantiomer 133).
Similar to Huang and co-workers (2018) who utilized Ru(III) metal, air and base to activate the gamma position of α,β-unsaturated aldehydes 9, [32] Sundén and co-workers (2022) developed an aerobic method for synthesizing tetrasubstituted benzene derivatives 136 in a [3 + 3] reaction using an ETM system consisting of FePc, Kharasch oxidant and NHC catalyst 30 (Scheme 34). [81] The reaction performs well already at the loading of 10 mol% of the ETMs. This loading of the oxidant can be considered low as the previously reported methods needed two stoichiometric equivalents of oxidant to synthesize the benzene derivatives. [33,82,83] The reaction includes an NHC catalyzed γ-carbon activation of the α,β-unsaturated aldehyde substrate, being the second example of this kind of reactivity under

R e v i e w T H E C H E M I C A L R E C O R D
aerobic conditions. The scope of the reaction is broad and allows to include ketone and ester functional groups on the products, resulting into useful scaffolds for further synthesis.

Oxidation Using Air as Terminal Oxidant under NHC and Photocatalysis
In recent years the combination of NHC and photo redox catalysis has emerged as a powerful way to perform organic synthesis. [84,85] An early example using a metal free photocatalyst to perform a direct oxidation of the Breslow intermediate was reported by Blechert in 2013. [86] By irradiating the photocatalyst, a mesoporous graphitic carbon nitride (mpg-C 3 N 4 ), aldehyde 137 could be converted to carboxylic acid 138 (Scheme 35). However, the photo/NHC catalytic protocol was less efficient than using NHC with air for the oxidation of cyclohexanecarbaldehyde to cyclohexanecarboxylic acid, consequently oxygen was solely used as the oxidant for the substrate scope for the transformation of aldehydes 139 to acids 140. The authors propose that the oxidative pathway is predominant. Furthermore, little difference in reaction yield could be detected when comparing the pure oxygen vs air. An observation that was attributed to the low solubility of oxygen into organic solvents and that the oxygen concentration is roughly the same regardless of what is used (air or pure oxygen) -assuming vigorous stirring of the reaction vessel. In 2020, Ye and co-workers showed the use of O 2 as oxidant under cooperative NHC/photoredox catalysis in the oxidative Smiles rearrangement of 141 to 142 (Scheme 36). [87] Mechanistically the reaction is divided into two stages. First, the Breslow intermediate 144 is oxidized by the photocatalyst 9-Mesityl-10-methylacridinium perchlorate (Mes-Acr-Me + ClO 4 À ), to generate the carboxylic acid 145. The oxidized form of the photocatalyst is then reduced by O 2 . The authors propose that this part of the mechanism is light promoted. In the second part the carboxylic acid 145 is oxidized by the excited form of the photocatalyst *MesÀ Acr + to generate radical 146 that initiates the aryltransfer via spirocyclic intermediate 147 to form the aryl ester radical 148. SET from 148 to Mes-Acr * gives anion 149 that is protonated in the final step to give the salisylate 150. The NaI is a prerequisite additive to accelerate the electron transfer.
Ye and co-workers (2020) have also reported a NHC catalyzed intramolecular oxidative cross dehydrogenative coupling reaction of tetrahydroisoquinoline-tethered aldehydes 151 (Scheme 37). [88] The reaction performs best under an atmosphere of pure oxygen. With the model substrate air was tested but gave lower yield (32 %) of the cyclized product as compared to pure molecular oxygen that gives the product in 77 % yield. The reaction mechanism is quite intricate and proceeds via an oxidative formation of iminium ion 153 that is formed in a catalytic fashion with iodine as the redox active species. In the next step an aerobic oxidation of the Breslow

Oxidation of the Aza-Breslow Intermediate Using Air
In addition to aldehydes, imines serve as versatile starting materials in NHC-catalysis, exhibiting dual roles as both acceptors and donors. [89] When functioning as an acceptor, imines display reactivity towards Breslow intermediates, homoenolates, azolium enolates, or azolium dionolates. Conversely, imines can be transformed into donors upon their reaction with an NHC. [90] For instance, the addition of a In the oxygenative pathway the peroxointermediate 158 reacts with an imine 156 to generate two equivalents of the amide product 161. To promote the oxidative pathway a common strategy is to employ a peroxide scavenger in the form of sodium pyruvate (SP). The oxidative pathway goes via the imidoyl azolium 159 that is reactive towards nucleophiles. The oxidative pathway can also be accessed by the addition of an high molecular weight oxidant such as the Kharasch oxidant, [91] or using a system of ETM in combination with air. [92] The bifurcated reaction pattern of the aza-Breslow intermediate was nicely illustrated by Huang and co-workers (2017) who utilized NHC, air as oxidant and LiCl as a cooperative Lewis acid to synthesize amides 163 from aldimines 162 containing the benzothiazole (BTA) scaffold. (Scheme 39A). [93] The heterocyclic BTA scaffold is common in many natural or synthetic products with pharmalogical and biological properties. [94] The reaction starts with formation of

R e v i e w T H E C H E M I C A L R E C O R D
Three years later, in 2020, Huang and co-workers introduced sodium pyruvate (SP) in the reaction conditions for the oxidation of aldimines 162 to imidates 164 (Scheme 39B). [95] SP functions as a peroxide scavenger and makes it possible to direct the reaction to the oxidative pathway opposed to the previous study. The authors could show that imidoyl azolium 168 can be reacted with nucleophilic species such as alcohols. The addition of SP is critical for the imidate formation -with SP 92 % of the imidate 164 is formed whereas in its absence only 34 % is formed.
In 2018, Huang and co-workers used aerial oxygen as the sole oxidant in the NHC-catalyzed synthesis of isoquinolinone derivatives 171 (Scheme 40). [96] The oxidation of isoquinolinium salts 170 was performed with good to excellent yields (62-96 %). After addition of the isoquinolinium salt to the NHC, followed by deprotonation, the aza-Breslow intermedi-ate 172 forms. After SET mechanism between oxygen and the aza-Breslow intermediate 172, followed by radical recombination a peroxide anion intermediate 173 forms. In the next step, the authors suggest that the aza-Breslow intermediate 172 might act as a reducing agent by reacting with intermediate 173 to form two molecules of 174 and regenerates the carbene. This is in accordance with the mechanism proposed by Chen [50] in 2006 (Scheme 12) where the Breslow intermediate 32 attacks the electrophilic dioxygen of 33. However, this is in discrepancy with the paper from Huang et al (2017) (Scheme 39A) who propose a mechanism where the peroxy aza-Breslow 167 reacts with the electrophilic imine 165 rather than by the aza-Breslow intermediate. Bortolini and co-workers (2017) showed that the dioxygen bridge of the Breslow intermediate is not electrophilic as it needs to be protonated to react and the Breslow intermediate can thus not act as a reducing agent on such species. [35] However, to solve the controversy, calculations on the oxidation of the aza-Breslow intermediate need to be performed as it might have a different reactivity pattern to that of the Breslow intermediate.
Inspired by the work of Huang in 2018, Suresh and coworkers tried in 2020 similar conditions when synthesizing N-Scheme 39. Aerobic oxidative NHC catalyzed synthesis of amides and imidates from aldimines via aza-Breslow intermediates. In Huang's first experiment the reaction proceeds via the oxygenative reaction. In a later experiment, the addition of sodium pyruvate made it possible to go via the oxidative pathway. In 2020 Song and co-workers reported an oxidative NHC catalyzed reaction for the synthesis of benzoxazole, benzothia-zoles and benzimidazoles 187, using air as oxidant (Scheme 43). [98] Two methylimidazolium moieties together with a tungsten molecule were heated up to 100°C and formed an activated pre-NHC  [WO 4 ] to deliver the product in yield of 22 %, suggesting that there might be a competing reaction mechanism that does not rely on an aza-Breslow intermediate.
Song and co-workers also synthesized flavones in 2020 with the concept of ionic liquids and air as sole oxidant, this time with molybdate-based ionic liquid, with good yields (98 %). [99] In 2022, Wang and co-workers reported the NHCmediated oxidation of imine 156 to various amides 161 (Scheme 44). [100] The design of the study was inspired by the firefly luciferase bioluminescence reaction in which luciferase The amides are formed in yields in the range of 48-98 %. The reaction provides an alternative to traditional amide synthesis but without using coupling reagents. Urea 196 is although formed as a side-product, making the procedure noncatalytic.
A similar oxidation of imines to amides was performed by Suresh and co-workers in 2022, who continued their work on NHC-catalyzed imine oxidation (Scheme 45). [101] While utilizing the non-green solvent acetonitrile and pure molecular oxygen in their previous design (Scheme 42), dimethyl carbonate as green solvent and air alone as oxidant was used in their new green protocol for the oxidation of aldimines to amides 161 via imine umpolung. In the postulated mechanism SET between the aza- Breslow

Oxidation of the Aza-Breslow Intermediate Using a System of ETMs
In 2022 Bortolini and co-workers developed a regiodivergent synthesis of benzothiazole-based isosorbide imidates by aerobic oxidative NHC catalysis (Scheme 46). The synthesis of monoimidate-isosorbides 200 and 201 (MIIs) starting from BTA containing aldimines 162 and isosorbides 200 (IS) performs well and the products (exo-201, endo-202) can be isolated with good regioselectivity and yields (up to 77 %). [92] The best results were obtained when air was used as the oxygen source in the combination with a system of ETMs. The regioselectivity of the reaction i. e., substitution on the exo (2-OH) or endo (5-OH) of the isosorbides, is dependent on solvent and additive effects. Using the combination of air and SP as developed by Huang does not improve the regioselectivity nor yield of the MIIs but leads to a higher amount of the amide byproduct 203. The authors postulate that the MIIs are formed through the oxidative pathway and that the byproduct 203 is formed through the oxygenative pathway.

Oxidation of the Deoxy Breslow Intermediate Using Air
One additional type of intermediate encountered in oxidative NHC catalysis is the deoxy Breslow intermediate, which arises from the reaction between alkyl halides and NHCs. [102][103][104] The deoxy Breslow intermediate is sensitive towards oxygen and moisture as it degrades exposed to air and rapidly deprotonates

R e v i e w T H E C H E M I C A L R E C O R D
under basic conditions, requiring the reaction to run in anaerobic conditions. [105] In 2020, Jiao and co-workers capitalized on the reactivity towards oxygen of the deoxy Breslow  the reaction is run in the presence of an alcohol, the reaction proceeds via the oxidative pathway and the formation of esters 211. In the cross-esterification of halides going through the oxygenative pathway, the yield was between 20-85 % (Scheme 47A), whereas in the cross-esterification of halides with alcohols going through the oxidative pathway, the yield is generally lower (Scheme 47B). Self-esterification of the halides creating symmetrical halides were also reported.  Table 1 summarizes the reactions discussed in this review based on the reaction type, product formed, oxidant used, NHC precatalyst loading, and temperature. In reactions involving the oxidation of the Breslow intermediate by air (section 3.1), esters are the primary products formed, but other compounds like phthalides, isocoumarins, 1,2-aminoalcohols, and dihydropyranones can also be obtained. However, these reactions have long reaction times and/or require elevated temperatures. It is worth noting that carboxylic acids may also be formed as by-products due to the competing oxygenative pathway when using oxygen as an oxidant.

Aerobic Oxidative NHC Catalysis by Numbers
When air is combined with NHC and a redox active catalyst (sections 3.2 and 3.3), the reaction times are around 24 hours, and the temperatures are relatively high (up to 130°C). The main products are esters, but it is possible to extend the reaction scope to include chiral lactones at room temperature by using pure oxygen as a requisite for high reaction turnover, as shown by Huang. [32] Using an ETM system (section 3.4) enables the utilization of aerial oxygen as an oxidant, resulting in considerably lower reaction times and temperatures compared to methods that solely use air or employ a redox active metal catalyst (sections 3.1 & 3.2). Notably, utilizing aerial oxygen yields better results than pure oxygen. For example, in the synthesis of dihydropyranones, pure oxygen led to no product formation, while aerial oxygen as a terminal electron donor resulted in a yield of 79 %. This is because Fe(II)Pc forms peroxodimers at high oxygen concentrations. [106] When NHC is used in combination with a photocatalyst and under radiation of light (section 3.5), aryl salicylate and oxidative cyclization products can be realized. The reactions are carried out at room temperature, and reaction times are 12-48 hours.
In the oxidation of the aza-Breslow intermediate (section 4.1), amides are the primary product formed, but other compounds such as imidates, isoquinolinones, and benzoxazoles can also be obtained. These reactions generally have short reaction times, require moderate pre-NHC loading, and work at ambient temperatures. The aza-Breslow oxidation can also be performed with an ETM system (section 4.2), as in the synthesis of monoimidate-isosorbide. The reaction time is short and performed at room temperature.
Oxidations via the deoxy Breslow intermediate (section 5.1) mainly lead to esters as the primary product. These reactions typically require pre-NHC loading in the range of 20-40 mol%, and the reaction times are typically more than 24 hours.
When reviewing the precatalyst loadings that is used for aerobic oxidative NHC catalysis one can see that the most common is 20 mol%, closely followed by 10 mol% and 1-2 mol%, with a mean value of 15 mol%.

Outlook
The ease with which simple starting materials can be transformed into highly functionalized products has made oxidative N-heterocyclic carbene (NHC) catalysis an area of significant interest. However, the use of stoichiometric amounts of high molecular weight oxidants in most reactions generates an undesired equivalent amount of waste, which makes purification troublesome and prohibits upscaling. The development of protocols that use oxygen as the terminal oxidant can potentially address this issue. As we can see in this review, the number of reactions that can be performed under air is increasing, and potentially, this will make aerobic oxidative NHC catalysis interesting in the eyes of, for instance, process chemists as complex reactions can be performed while generating less chemical waste than traditional oxidative NHC catalysis.
In addition to the environmental benefits of aerobic oxidative NHC catalysis, this approach also holds great potential for the development of more efficient electron transfer mediators (ETMs). Hybrid ETMs, [107] which for example combine FePc and a quinone, or even furthercombine NHC, quinone and FePc, have the potential to make the electron transfer between the two more efficient, ultimately leading to cleaner and faster reactions. Moreover, additional examples on aerobic oxidative NHC catalysis combined with photochemistry needs to be elucidated, utilizing another activation techniques than NaI. [87][88] The oxidation of the aza-Breslow intermediate holds considerable interest due to the ability to perform this transformation at ambient temperature, leading to the formation of two distinct products depending on whether it is generated through an oxidative or oxygenative mechanism. While the exploration of this reaction type is relatively limited, it is emerging as a noteworthy area, as highlighted in the review. There exists significant potential for future endeavors aimed at establishing novel reactions by leveraging the oxidation of the aza-Breslow intermediate as a versatile starting point.

Conclusion
Aerobic oxidative N-heterocyclic carbene (NHC) catalysis has gained significant interest in synthetic organic chemistry due to its ability to transform simple starting materials into highly functionalized products. The use of oxygen as the terminal oxidant has been developed to address the issue of generating high molecular weight waste associated with the use of stoichiometric oxidants. This review provides an overview of the development of aerobic oxidative NHC catalysis and discusses the types of NHC-catalyzed reactions reported with oxygen as the terminal oxidant. The review also covers the strategies that can be used for oxygen activation and the selectivity issues that can arise when running NHC catalysis under aerobic conditions. Overall, this review highlights the potential of aerobic oxidative NHC catalysis as a green and sustainable alternative to traditional methods. However, further research is needed to address the challenges associated with oxygen activation and selectivity issues, and to expand the scope of reactions possible with this methodology. Huang, 2020 [95] Imidate SP, Air 20 6 rt Huang, 2018 [96] Isoquinolinone Air 10 12 rt Hou, 2018 [97] Benzoxazole Air 10 10 120 Suresh, 2020 [108] N-substituted cyclic amide Air 30 12 60 Song, 2020 [98] Benzoxazole Air 20 5 100 Wang, 2022 [100] Amide Air 120 0,5-4 rt Suresh, 2022 [101] Amide Air 20 4 rt 4.2 Oxidation of the aza-Breslow intermediate using a system of ETM Bortolini, 2022 [92] Monoimidate-isosorbide 10 mol% DPQ, 5 mol% FePc, Air 10 2-24 rt 5.1 Oxidation of the deoxy-Breslow intermediate using air Jiao, 2020 [105] Ester from halides Air 20 36 40 Jiao, 2020 [105] Ester from alcohols