Sustainable amidation through acceptorless dehydrogenative coupling by pincer-type catalysts: recent advances

Water and ammonia as amidation reagents

Water is the ultimate environmentally-friendly reagent and solvent, since it is nontoxic, widely available on Earth, and renewable. In 2013, we introduced a new process for converting alcohols into carboxylates in an alkaline aqueous solution, with no added oxidants nor hydrogen acceptors, and with generation of H₂ gas as the only byproduct (Scheme 3a) [36]. This was catalyzed by complex 3, obtained in situ from precatalyst 5 under the basic conditions, in a manner analogous to complex 1. In this unprecedented acceptorless dehydrogenative oxidation process, water itself assumed the role of formal oxidant, contributing an oxygen atom incorporated into each of the carboxylate products. Capitalizing on this discovery, we have been seeking to apply this new water-based synthetic methodology for the oxidation of other substrates.

Scheme 3:

First examples of acceptorless dehydrogenative oxidation of organic substrates using water as a formal oxidant: (a) Oxidation of alcohols into carboxylates in an alkaline aqueous solution, catalyzed by complex 3 that was generated in situ from precatalyst 5 and base, as reported in 2013 [36]. (b) Oxidation of cyclic amines into lactams in an aqueous solution, catalyzed by the dearomatized acridine-based PNP-Ru(II) complex 7 that was generated in situ from precatalyst 6 and catalytic base, as reported in 2014 [37]; later work showed that catalyst 7 can be isolated and used directly under neutral conditions [38].

In 2014, we reported the synthesis of lactams from cyclic amines in a water/dioxane mixture, accompanied by H₂ liberation (Scheme 3b) [37]. This acceptorless dehydrogenative amidation process was catalyzed by the acridine-based PNP-type Ru(II) complex 6 in the presence of a catalytic amount of base (NaOH; 1–1.5 equiv per complex). It was later shown that this complex undergoes in situ conversion into the catalytically-active dearomatized complex 7, through a base-promoted amine-mediated reaction [38]. In fact, directly using the presynthesized complex 7 allowed the reaction to proceed under base-free conditions. The catalytic amidation reaction, involving 6 and catalytic base, was carried out with catalyst loadings of 1–5 mol %, at a nominal temperature^[3] of 150 °C under an N₂ atmosphere, inside a closed reaction vessel. In this manner, a series of cyclic amines exhibiting 5- and 6-membered rings, in the absence and presence of various exocyclic substituents and endocyclic heteroatoms, were converted into the corresponding γ- and δ-lactams in 41–85 % yield, at 59–99 % conversion, within 48–89 h. An ¹⁸O-labeling experiment confirmed that water is the source of the oxygen atom in the generated amide functionality. This transformation represented a fundamentally new type of amide synthesis directly from amines and water under oxidant-free conditions. This kind of methodology, utilizing water as a formal oxidant through acceptorless dehydrogenative coupling, has since been applied for the oxidation of various other substrates [39–45].

It should be noted that when the same catalyst was applied to acyclic primary and secondary amines under similar conditions, deamination was observed, converting the amines into the respective alcohols and free NH₃ [46]. This entropically-driven reaction was prevented in the case of the secondary cyclic amines. Interestingly, the reverse process, namely, dehydrogenative coupling of NH₃ and alcohols into amines, could also be promoted by the very same catalyst, as also described in an earlier report [47]. Thus, when 1,4-butanediol was subjected to the catalytic conditions under pressurized ammonia, a mixture of pyrrolidine and 2-pyrrolidone was obtained [37].

Ammonia, unlike water, is not a significant natural resource on Earth, but is nonetheless industrially ubiquitous, since it is produced globally on an immense scale (e.g., ∼180 million tons in 2020 [48]) through the Haber–Bosch process, primarily for agricultural use. Processes that convert ammonia directly into amides are highly desirable, since they would obviate the need for more complex nitrogen-containing precursors like amines or nitriles, which are much less readily-available and are themselves typically derived from ammonia.

Primary amides are commonly-found in pharmaceuticals, natural products, agrochemicals and biologically-active molecules. In principle, the same acceptorless dehydrogenative methodology used for coupling alcohols and amines into amides should also be applicable for primary amide synthesis, through the use of primary alcohols and ammonia. Nevertheless, catalytic processes that employ the latter substrates, and involve acceptorless dehydrogenation, have typically resulted in amines, imines or nitriles, rather than amides [47, 49–54]. This can be attributed to the underlying mechanisms of these reactions, all of which share the same initial steps, namely, alcohol dehydrogenation into aldehyde, followed by nucleophilic attack of NH₃ on its carbonyl, generating a hemiaminal intermediate (Scheme 2). In order to access the amide product, this hemiaminal must be dehydrogenated, but instead it usually undergoes facile dehydration into imine, which can then react further to afford various N-containing species, depending on the exact reaction conditions.

Very recently, we showed that complex 2 can efficiently and selectively catalyze the unprecedented synthesis of primary amides from alcohols and gaseous ammonia, with liberation of H₂ (Scheme 4) [55]. This catalyst, which was selected from a number of PNN- and PNP-Ru(II) complexes, had also been used in our original dehydrogenative coupling of alcohols and amines into amides [11], and was generated in situ from complex 1 and 2 equiv of base (KOtBu). The optimal catalytic conditions included use of the catalyst at 1 mol % loading in a solvent mixture comprising toluene and tert-amyl alcohol (tAmOH) in a 2:1 volumetric ratio, designed to ensure high ammonia solubility. Furthermore, a dual-step procedure was employed, involving two heating sessions, initially at 120 °C and then at 150 °C (nominal temperatures), with intervening removal of the accumulated H₂ gas to avoid product hydrogenation. When a large series of aliphatic and benzylic primary alcohols were subjected to these reaction conditions, in a sealed reactor under pressurized NH₃ (7 bar), primary amides were obtained in 51–92 % yield, but generally above 80 %, within 36 h.

Scheme 4:

Acceptorless dehydrogenative coupling of primary alcohols and ammonia into primary amides catalyzed by complex 2, as reported in 2022 [55]. The active catalyst was generated in situ by treating complex 1 with base.

Advances in Ru-catalyzed secondary and tertiary amide synthesis

In an effort to expand the scope of amide targets, the acceptorless dehydrogenative coupling methodology was applied for the conversion of ethylene glycol and monoamines into a series of symmetrically-substituted oxalamides (Scheme 5) [56]. The oxalamide moiety is an important structural motif that is present in many biologically-active compounds and pharmaceuticals, but its synthesis has traditionally involved the use of harmful reagents with generation of large amounts of chemical waste, such as in the condensation of amines and oxalyl chloride. By contrast, ethylene glycol is a relatively-benign, inexpensive and readily-available industrial feedstock material that can be derived from biomass, and its coupling with amines, with liberation of H₂ as the only byproduct, represents a sustainable and atom-economical alternative for the synthesis of oxalamides.

Scheme 5:

Dehydrogenative synthesis of symmetrically-substituted oxalamides from ethylene glycol and monoamines, catalyzed by the pyridine-based PNN-Ru(II) complex 8 in the presence of base, as reported in 2020 [56]. The base was necessary to convert this complex in situ into the active catalyst, 9 or 10.

A series of PNN- and PNP-type Ru(II)-pincer complexes were studied as catalysts for the above transformation in various solvents, at a nominal temperature of 135 °C under N₂, inside a sealed vessel. The best catalytic performance was obtained with the PNN-type complex 8 in toluene (Scheme 5), at a catalyst loading of 1 mol %, in the presence of a catalytic amount of base (KOtBu; 2 equiv per complex). It was shown that 1 equiv of base deprotonates the N−H moiety of 8, thereby affording complex 9, which was proposed to be the catalytically-active species. As we reported elsewhere, treatment of 8 with 2 equiv of base leads to double deprotonation at its amine arm, both at the N−H moiety and benzylic position, thereby forming the anionic dearomatized complex 10 (Scheme 5) [57, 58]. Under the catalytic conditions, in the presence of excess alcohol, complexes 9 and 10 are expected to behave similarly, if not identically. The substrate scope of this catalytic system was explored by using a large number of assorted amines, mostly primary and secondary aliphatic amines and benzylamines, as coupling partners for ethylene glycol. In the vast majority of cases, moderate to high yields were obtained within 24 h, ranging from 53 to 97 %, with most being well over 70 %. Notable exceptions were α-branched primary amines and aniline, which afforded much lower yields (12–32 %), attributable to steric interference and low nucleophilicity, respectively.

The same catalytic system, involving complex 8 and base, was also found to efficiently promote the reverse reaction, namely, hydrogenation of oxalamides into ethylene glycol and the corresponding amines. For a given oxalamide, this was carried out in a toluene solution containing the catalyst at 1 mol % loading, together with 4 equiv of base per catalyst, inside a sealed vessel under pressurized H₂ (40 bar), at a nominal temperature of 135 °C. In this manner, a large array of different oxalamide substrates were hydrogenated to afford the respective amines in very high yields, i.e., 85–99 %, within 24 h.

Further progress in Ru-catalyzed dehydrogenative amidation was achieved very recently, when we succeeded in significantly decreasing the required catalytic reaction temperature, thus making the process more energy-efficient [58]. Since our 2007 report of the acceptorless dehydrogenative coupling of alcohols and amines into amides, we have continued developing this type of reactions, as described in the present review, and such reactions have also been accomplished by other research groups, utilizing different homogeneous and heterogeneous catalysts under various reaction conditions. However, all of the reported systems required heating, typically above 100 °C, in order to achieve significant turnovers. In 2021, we demonstrated that this transformation can be realized at near-ambient temperatures, namely, in refluxing Et₂O (35 °C) or methyl tert-butyl ether (MTBE; 55 °C), by using complex 8 (Scheme 6), chosen from a collection of various PNN-Ru(II) and -Mn(I) complexes [58]. This precatalyst was employed at 1 mol % loading in the presence of a catalytic amount of base (KOtBu; 2 equiv per complex), in an open system under argon. As mentioned above, under these conditions, complex 8 converts into 10, which is the active catalyst.

Scheme 6:

Near-ambient-temperature dehydrogenative synthesis of amides from alcohols and amines, catalyzed by the dearomatized pyridine-based PNN-Ru(II) complex 10 that was generated in situ from precatalyst 8 and 2 equiv of base, as reported in 2021 [58].

The above catalytic system was applied on a range of alcohol and amine coupling partners, mainly including aliphatic and benzylic primary alcohols and amines, and the best yields obtained for each pair ranged from 58 to 97 % within 22–60 h, but were mostly above 80 %. Some substrates required refluxing in MTBE rather than Et₂O to achieve high yields. It should be noted that chiral substrates were also employed, and it was found that their coupling can be carried out in this catalytic system without substantial loss of optical activity. Moreover, a few commercial pharmaceuticals (Fig. 2) were synthesized in 92–95 % yield, each from its corresponding alcohol and amine partners, in the first reported use of acceptorless dehydrogenative amidation for the synthesis of pharmaceutical compounds.

Fig. 2:

Commercial pharmaceuticals synthesized using the method outlined in Scheme 6 (MTBE as solvent, 55 °C) [58].

As described above, and throughout this review, the Ru- and Mn-catalyzed dehydrogenative synthesis of amides has primarily relied on alcohols as coupling partners for amines, with esters and water also being employed, but to a lesser extent. Very recently, we expanded this family of oxygen-containing substrates to include epoxides (Scheme 7) [59]. A priori, epoxides may appear ill-suited for dehydrogenative amidation, since the reaction of an epoxide with an amine typically affords a β-amino alcohol rather than the hemiaminal intermediate necessary for amide generation. Nevertheless, after examining a number of Ru(II)-, Mn(I)- and Co(II)-pincer complexes, it emerged that two PNN-Ru(II) complexes, 1 and 5, are able to dehydrogenatively couple aryl epoxides and amines into amides in the presence of catalytic amounts of base. It should be noted that direct amidation of epoxides has no known precedent, except for one uncatalyzed Willgerodt-type reaction reported in the 1950s, which involved harsh conditions and toxic chemicals [60].

Scheme 7:

Dehydrogenative synthesis of amides from aryl epoxides and amines, as reported in 2022 [59]: (a) Reactions involving primary amines, catalyzed by complex 2 that was generated in situ from precatalyst 1 and base. (b) Reactions involving secondary linear and cyclic amines, catalyzed by complex 3 that was generated in situ from precatalyst 5 and base.

The optimal catalytic conditions for dehydrogenative epoxide amidation involved using a toluene solution containing the precatalyst, 1 or 5, at 1 mol % loading, together with 1 equiv of base (KOtBu) per complex, which was required for converting it into the catalytically-active dearomatized species, 2 or 3, respectively. The aryl epoxide and amine were added into this solution, and the resulting mixture was heated for 12 h at a nominal temperature of 150 °C, in a sealed reactor under N₂. In this manner, the in-situ-generated complex 2 was employed to couple styrene oxide with a few primary amines (Scheme 7a), thereby affording the corresponding secondary amides in 44–66 % yield. Coupling of the same epoxide with a series of secondary linear and cyclic amines was accomplished using the in-situ-prepared complex 3 (Scheme 7b), instead of 2, and the resulting tertiary amides were obtained in generally much higher yields, i.e., 62–95 %. In a similar fashion, catalyst 3 was used to couple a range of ring-substituted styrene oxides with dipropylamine, giving the respective amides in 80–92 % yield. It should be noted that the ability to selectively transform epoxides into amides, as demonstrated above, can be combined with well-established procedures for alkene epoxidation, thereby opening the way for new two-step strategies for converting alkenes into amides.

Amidation-based liquid organic hydrogen carriers

Liquid organic hydrogen carrier (LOHC) systems are intended for the high-density storage and on-demand release of H₂ gas, to be applied in hydrogen-based fuel installations, as part of the broader development of a future hydrogen-based economy [61, 62]. In principle, such a system should contain an appropriate LOHC material, which is able to efficiently undergo repeated dehydrogenation/rehydrogenation (H₂ discharge/recharge) cycles, and preferably employ a single catalyst that can promote both the forward and reverse reactions. As mentioned above, our earlier work had shown that complex 2 efficiently catalyzes the coupling of alcohols into esters, as well as alcohols and amines into amides, with concomitant generation of H₂ gas. The same complex was also found to promote the reverse processes, i.e., hydrogenation of esters into alcohols [63], and of amides into alcohols and amines [64]. Similar reactivity involving amide formation and hydrogenation was also observed for the dearomatized complex 3 [33, 34, 64]. These findings prompted us to investigate such complexes as potential catalysts for use in LOHC systems, eventually leading to the development of several such systems, three of which involve amide bond formation.

The first amidation-based LOHC system, reported in 2015, utilized the reversible conversion of 2-aminoethanol (AE) into its cyclic homodimer glycine anhydride (GA), with concomitant liberation of two molecules of H₂ per molecule of AE (Scheme 8a) [65]. This aminoalcohol is a prototypical example of an LOHC material, being inexpensive and readily-available, and at the same time providing a high theoretical hydrogen storage capacity of 6.6 wt % (excluding solvent), which surpasses the target capacity of 5.5 wt % for light-weight vehicles set by the United States Department of Energy (DOE) for the year 2025 [66]. After examining a few PNN-Ru(II) precatalysts, it was found that the pyridine-based complex 11 (Scheme 8b) can achieve high conversions in both AE dehydrogenation and GA hydrogenation. This complex was activated in situ by the addition of 2 equiv of base (KOtBu), which transformed it into the catalytically-active dearomatized variant, namely, the anionic complex 12 (Scheme 8b) [57], which is analogous to complex 10. Attempting the catalytic dehydrogenation under solvent-free conditions, using the neat aminoalcohol substrate, resulted in only trace amounts of GA. This outcome, together with the fact that GA is a high-melting solid, necessitated the use of solvent in order to facilitate both AE dehydrogenation and GA hydrogenation. Of the various solvents examined, dioxane provided the best results for both reactions.

Scheme 8:

(a) Reversible conversion of 2-aminoethanol (AE) into glycine anhydride (GA) and H₂, employed as a basis for the LOHC system reported in 2015 [65]. (b) The pyridine-based PNN-Ru(II) precatalyst 11 used in this system was activated with 2 equiv of base to generate the catalytically-active dearomatized complex 12. (c) Experimental dehydrogenative coupling of AE into GA and linear peptides, and hydrogenation of the latter back into AE, catalyzed by complex 12 that was generated in situ from precatalyst 11.

The dehydrogenation process was carried out at 105 °C, in an open system under argon, whereas hydrogenation was performed in a sealed reactor under pressurized H₂ (50–70 bar), at a nominal temperature of 110 °C. With catalyst loading at 0.5 mol %, the in-situ-generated 12 attained a maximal conversion of 89 % within 12 h, and maximal H₂ and GA yields of 77 and 60 %, respectively, with the remaining products being linear peptides (Scheme 8c). In GA hydrogenation, catalyst 12 achieved an essentially quantitative yield of AE within 48 h, at 1 mol % loading under 50 bar of H₂. It should be noted that this catalytic reaction represented the first reported hydrogenation of any diketopiperazine. Furthermore, 12 was also able to efficiently hydrogenate a mixture of GA and linear peptides into AE, with the amount of this aminoalcohol reaching 85 % of the reaction mixture within 48 h, thereby demonstrating that generation of linear peptides does not pose a substantial impediment for this catalytic system.

The ability of the catalytic system to promote recurring H₂ discharge/recharge cycles was demonstrated by running three consecutive cycles on a single dioxane solution, initially containing AE, 11 (1 mol %) and 2 equiv of base per complex. This system was found to maintain high conversions for both aminoalcohol dehydrogenation and peptide hydrogenation throughout all cycles, albeit with decreasing performance as the cycles progressed, i.e., from 86 to 76 % for dehydrogenation and 97 to 81 % for hydrogenation. It should be noted that although the practicality of this system is limited by various aspects, such as the use of solvent, it constituted the first proof-of-concept demonstration of an LOHC system that is based on reversible acceptorless dehydrogenative coupling reactions involving abundant raw materials.

Having shown that reversible acceptorless dehydrogenative coupling can be utilized to construct a single-component aminoalcohol-based LOHC system, we set to expand the scope of this work by developing a second, dual-component, system based on ethylenediamine (ED) and ethanol (Scheme 9a), which was reported in 2016 [67]. Both of these substrates are inexpensive and readily-available, and their corresponding LOHC system would exhibit a hydrogen storage capacity of 5.3 wt %, provided they can be fully converted into N,N′-diacetylethylenediamine (DAE) with liberation of four molecules of H₂ per molecule of ED. A few PNN-Ru(II) complexes were examined as catalysts for the dehydrogenative coupling of ED and ethanol, as well as the hydrogenation of their reaction products (Scheme 9b and c). In both processes, the best catalytic performance was obtained with the bipyridine-based complex 5 in the presence of a catalytic amount of base (KOtBu; 1–2 equiv per complex), which transformed it into the catalytically-active dearomatized complex 3. Furthermore, the use of dioxane as a solvent was necessary to ensure high conversions and yields in both dehydrogenation and hydrogenation.

Scheme 9:

(a) Reversible coupling of ethylenediamine (ED) and ethanol into N,N′-diacetylethylenediamine (DAE) with generation of H₂, employed as a basis for the LOHC system reported in 2016 [67]. (b) Experimental dehydrogenative coupling of ED and ethanol into DAE and N-(2-aminoethyl)acetamide, catalyzed by complex 3 that was generated in situ from precatalyst 5 and base. (c) Experimental hydrogenation of DAE into ED and ethanol, catalyzed by the in-situ-prepared complex 3.

The dehydrogenative coupling of ED and ethanol was carried out by heating the reaction mixture at 105 °C for 24 h, in an open system under argon. Using a catalyst loading of only 0.2 mol %, the highest conversions achieved for both substrates were quantitative, and were associated with a very high DAE yield of 92 %, accompanied by only 8 % of N-(2-aminoethyl)acetamide as the sole organic byproduct. Importantly, an excellent hydrogen yield of 95 % was also obtained. The reverse process, namely, hydrogenation of DAE, was conducted in a sealed vessel under pressurized H₂ (40–70 bar), at a nominal temperature of 115 °C. The best result achieved under these conditions, with a catalyst loading of 0.4 mol % under 70 bar H₂, was quantitative yield of ED, together with 89 % of ethanol, within 10 h. This catalytic system was also shown to successfully endure three consecutive H₂ discharge/recharge cycles, using a single dioxane solution, initially containing 5 (0.4 mol %), 1 equiv of base per complex, ED and ethanol. The performance of this system was nearly constant throughout this trial, with over 90 % conversion of ED being observed during the dehydrogenation segments, and quantitative yields of this diamine being obtained during the hydrogenation segments.

The third LOHC system, reported in 2018, was also based on ED, but utilized 1,4-butanediol as its dehydrogenative coupling partner (Scheme 10a) [68]. In this system, the target coupling product is N,N′-ethylenedisuccinimide (EDS), the formation of which would be accompanied by eight molecules of H₂ per molecule of ED, thereby providing a theoretical hydrogen storage capacity of 6.7 wt %, which is higher than the aforementioned DOE objective. Upon screening a number of potential PNN-Ru(II) catalysts, the highest dehydrogenative coupling performance was achieved with complex 13 (Scheme 10b), an analog of 11 bearing phenyl P-substituents instead of tert-butyls. The catalytic process that provided the best results involved a dioxane solution containing the substrates, complex 13 at 1 mol % loading, and 2 equiv of base (KOtBu) per complex, and was carried out at a nominal temperature of 120 °C under an N₂ atmosphere, inside a closed vessel. Under these conditions, virtually full conversion of the substrates was achieved within 24 h, affording EDS in 70 % yield, together with γ-butyrolactone (12 %) and oligoamides, as well as hydrogen gas in 84 % yield (Scheme 10b). The reverse process, i.e., hydrogenation of EDS (Scheme 10c), was found to be most effective when complex 11 was employed, in the presence of 3 equiv of base, and the catalytic reaction was conducted with dioxane as solvent, inside a sealed reactor under pressurized H₂ (40 bar), at a nominal temperature of 135 °C. Under these conditions, and with catalyst loading at 1 mol %, the conversion of EDS was essentially quantitative after 40 h, and the products, ED and 1,4-butanediol, were obtained in ≥90 % yield.

Scheme 10:

(a) Reversible coupling of ethylenediamine (ED) and 1,4-butanediol into N,N′-ethylenedisuccinimide (EDS) with generation of H₂, employed as a basis for the LOHC system reported in 2018 [68]. (b) Experimental dehydrogenative coupling of ED and 1,4-butanediol into EDS, γ-butyrolactone and oligoamides, catalyzed by the pyridine-based PNN-Ru(II) complex 13 in the presence of base. (c) Experimental hydrogenation of EDS into ED and 1,4-butanediol, catalyzed by complex 11 in the presence of base.

The ability of the latter LOHC system to sustain repeated H₂ discharge/recharge cycles was demonstrated by conducting two consecutive cycles on a single dioxane solution, initially containing EDS. The appropriate precatalyst, 11 or 13, was loaded into this mixture prior to each hydrogenation or dehydrogenation segment, respectively, together with 3 equiv of base per complex. The system was found to maintain its performance, with high product yields of 85–93 % during the EDS hydrogenation segments, and practically full conversions of ED and 1,4-butanediol during the dehydrogenation segments, alongside EDS yields of 64–68 %.

Methanol and ammonia as amidation reagents

In 2016, we reported the acceptorless dehydrogenative coupling of alcohols and amines into imines, catalyzed by the dearomatized pyridine-based PNP-Mn(I) pincer complex 14 (Scheme 11a) [69]. This represented the first known use of a homogeneous manganese complex as catalyst for any type of dehydrogenation reaction. Soon thereafter, we were able to show that another Mn(I) complex, 15 (Scheme 11b), featuring a PNP-type pincer ligand with an aliphatic amine backbone, catalyzes the acceptorless dehydrogenative N-formylation of primary and secondary amines, with methanol serving as the formylating agent [70]. This transformation, which constituted the first reported example of a manganese-catalyzed dehydrogenative amidation, was carried out with the catalyst at 2 mol % loading in neat methanol, at a nominal temperature of 110 °C under N₂, in a closed vessel. In this manner, an array of secondary cyclic amines, benzylamines and primary aliphatic amines could be N-formylated, affording the respective products in moderate to high yields, ranging from 50 to 86 %, within 12–24 h.

Scheme 11:

First examples of the use of Mn-based catalysts for acceptorless dehydrogenative coupling reactions: (a) Conversion of alcohols and amines into imines, catalyzed by the dearomatized pyridine-based PNP-Mn(I) complex 14, as reported in 2016 [69]. (b) N-formylation of primary and secondary amines using methanol, catalyzed by the PNP-Mn(I) complex 15, as reported in 2017 [70].

In 2019, we reported the synthesis of secondary amides by acceptorless dehydrogenative coupling of benzyl alcohols and ammonia, catalyzed by the dearomatized acridine-based PNP-Mn(I) complex 16 (Scheme 12) [71]. This transformation constituted the first known example of an acceptorless dehydrogenative synthesis of amides from alcohols and NH₃, and preceded our Ru-catalyzed synthesis of primary amides from the same substrates, as discussed above. This method is step-economical in that it avoids amines as coupling partners, and relies solely on alcohols as the source of the N-substituents. The optimal catalytic conditions involved a toluene solution containing complex 16 at 2 mol % loading, a given benzyl alcohol substrate, and base (KH) in an amount equimolar to the alcohol. This mixture was heated at a nominal temperature of 150 °C inside a sealed reactor under pressurized NH₃ (7 bar). Under these conditions, a number of benzyl alcohols were converted into the corresponding secondary amides in 47–90 % yield, but mostly over 75 %, within 24 h.

Scheme 12:

Acceptorless dehydrogenative coupling of benzyl alcohols and ammonia into secondary amides catalyzed by the dearomatized acridine-based PNP-Mn(I) complex 16, as reported in 2019 [71].

Advances in Mn-catalyzed secondary and tertiary amide synthesis

Cyclic imides, which were discussed above in the context of LOHC systems, are also important building blocks in natural products, pharmaceuticals and polymers [72]. However, conventional methodologies for their synthesis usually involve harsh conditions and generate stochiometric amounts of waste, and recently-developed catalytic systems have been largely based on expensive noble metals like ruthenium, rhodium, iridium and palladium [72]. Following the successful application of the pyridine-based PNN-Mn(I) complex 17 (Scheme 13) to ester hydrogenation [73], we were also able to demonstrate that it can promote the acceptorless dehydrogenative coupling of diols and amines into cyclic imides [74].

Scheme 13:

Dehydrogenative synthesis of cyclic imides from diols (1,4-butanediol and 1,5-pentanediol) and monoamines, catalyzed by the pyridine-based PNN-Mn(I) complex 18 that was generated in situ from precatalyst 17 and base, as reported in 2017 [74].

The synthesis of a given cyclic imide was conducted in an open system under argon, by refluxing a toluene solution containing an equimolar mixture of diol and amine, together with complex 17 and 2 equiv of base (KH) per complex. The added base deprotonated the N−H moiety of this complex, thereby generating complex 18 (Scheme 13), which is the active catalyst. Employing a catalyst loading of 5 mol %, 1,4-butanediol was coupled with a series of primary amines, mainly exhibiting benzylic and aliphatic N-substituents, to give the corresponding succinimides in 42–92 % yield, at 60–99 % conversion, within 40 h. Similarly, 1,5-pentanediol was coupled with a subset of these amines, in the presence of the catalyst at 10 mol % loading, to afford glutarimides in 49–74 % yield, with essentially quantitative conversion, within the same time frame. It should be noted that results obtained with precatalyst 17 were comparable to those of the Ru(II) precatalyst 1 under identical reaction conditions, indicating that the corresponding Mn(I) and Ru(II) catalysts were of similar activity.

Attempts were subsequently made to expand the scope of Mn-catalyzed amidation beyond cyclic imides by applying precatalyst 17 to the coupling of primary monoalcohols and monoamines in the presence of base. However, these attempts suffered from low selectivity, producing significant amounts of imine and ester side-products. Nevertheless, when 17 was replaced by the closely-related complex 19, bearing phenyl instead of tert-butyl P-substituents, efficient and selective amidation was achieved (Scheme 14) [75]. This constituted the first example of a general acceptorless dehydrogenative coupling of alcohols and amines into amides catalyzed by a base metal complex.

Scheme 14:

Dehydrogenative amidations catalyzed by the pyridine-based PNN-Mn(I) complex 20 that was generated in situ from precatalyst 19 and base, as reported in 2017 [75]: (a) Coupling of alcohols and amines. (b) Coupling of esters and amines.

The catalytic process that yielded the best results involved a toluene solution containing the substrates, complex 19 (5 mol %) and 2 equiv of base (KOtBu) per complex, and was conducted at a nominal temperature of 110 °C under N₂, inside a sealed vessel. In analogy to complex 17, it was proposed that the added base deprotonates the N−H moiety of 19, thereby forming complex 20 that is the catalytically-active species. Under these reaction conditions, a variety of aliphatic alcohols were coupled with benzylamines (Scheme 14a), achieving practically full conversions and affording the corresponding amides in good to high yields, ranging from 71 to 86 %, within 48 h. It should be noted that using benzyl alcohols resulted in much lower yields of 20–40 %, due to generation of substantial quantities of imines, and a similar observation was made for the aliphatic amine 1-butylamine.

Complex 19 was also found to catalyze the coupling of esters and amines to give amides (Scheme 14b), which was unprecedented for a base metal catalyst. By applying the same catalytic conditions used for coupling alcohols and amines, a series of mostly aliphatic esters were coupled with various amines, mainly benzylamines and secondary cyclic amines, to give the respective amides in 30–95 % yield (typically ≥75 %), with generally high conversions of 75–99 %, after 48–72 h.