Dehydrogenative synthesis of N-functionalized 2-aminophenols from cyclohexanones and amines: Molecular complexities via one-shot assembly

Abstract

Polyfunctionalized arenes are privileged structural motifs in both academic and industrial chemistry. Conventional methods for accessing this class of chemicals usually involve stepwise modification of phenyl rings, often necessitating expensive noble metal catalysts and suffering from low reactivity and selectivity when introducing multiple functionalities. We herein report dehydrogenative synthesis of N-functionalized 2-aminophenols from cyclohexanones and amines. The developed reaction system enables incorporating amino and hydroxyl groups into aromatic rings in a one-shot fashion, which simplifies polyfunctionalized 2-aminophenol synthesis by circumventing issues associated with traditional arene modifications. The wide substrate scope and excellent functional group tolerance are exemplified by late-stage modification of complex natural products and pharmaceuticals that are unattainable by existing methods. This dehydrogenative protocol benefits from using 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO) as oxidant that offers interesting chemo- and regio-selective oxidation processes. More notably, the essential role of in situ generated water is disclosed, which protects aliphatic amine moieties from overoxidation via hydrogen bond–enabled interaction.

A one-shot assembly method to forge 2-aminophenols was developed through TEMPO oxidation and water’s hydrogen bonding protection.

INTRODUCTION

Phenyl rings are one of the most prevalent units in high-valued chemicals like pharmaceuticals and functional materials, making the syntheses of polyfunctionalized arenes of paramount importance. The N-functionalized 2-aminophenols, as versatile reagents, have been widely used for organic syntheses (1), such as starting materials to natural alkaloids (2), bioactive small molecules (3), agrochemicals (4), materials (5), and catalyst ligands (6) (Fig. 1A). Although Pd-catalyzed modification of existing NH₂ group in N-unsubstituted ortho-aminophenols provides a straightforward avenue to N-arylated 2-aminophenols (7), the preparation of parent ortho-aminophenols requires phenol nitration and subsequent nitro-to-aniline reduction, which suffers from mixed regioisomers, harsh reaction conditions, and limited functional group tolerance (Fig. 1B) (8–11). The remarkable development of C─H manipulation techniques enables direct hydroxylation or amination of corresponding aniline or phenol derivatives to functionalized 2-aminophenols. Nevertheless, additional synthetic steps for modulating starting materials, specific aminating reagents, and selectivity problems continue to pose unresolved challenges (Fig. 1C) (12–17). To tackle these issues, there has been a longstanding desire for the development of highly efficient yet selective strategies to streamline the synthesis of N-functionalized 2-aminophenols with complex molecular structures.

Fig. 1. Significance and development of the methods for syntheses of N-functionalized 2-aminophenols via transformation of C─H bond.

(A) The examples of the functional compounds synthesized from reactions of N-functionalized 2-aminophenols. (B) The conventional route to syntheses of N-functionalized 2-aminophenols. (C) The recently developed methods for construction of 2-aminophenols. (D) This work: the dehydrogenation-driven reaction of amines with cyclohexanones to construct N-functionalized 2-aminophenols.

Driven by our interest in the desaturation-triggered functionalization of carbonyl compounds (18–22), we reason that the dehydrogenative aromatization of cyclohexanone derivatives might offer a suitable access to polyfunctionalized 2-aminophenols since a series of transition metal–catalyzed methods has been successfully developed to efficiently construct diversely substituted aromatic compounds through either cyclohexanones (23–25) or cyclohexanone derivatives generated in situ (26–29) under oxidative conditions. Many elegant studies have also demonstrated that merging carbonyl desaturation with subsequent alkene intermediates conversion could offer powerful methods for diverse value-added carbonyl or other hydrocarbon transformations (18–22, 30–43), which showcase versatility of desaturation in organic synthesis. However, construction of N-functionalized 2-aminophenols through desaturation-enabled transformation is a nontrivial task due to the following reasons: (i) 2-Aminophenols are vulnerable to the commonly used oxidative conditions in dehydrogenative reactions, owing to its inherent electron-richness; (ii) 2-aminophenols can serve as bidentate ligands to chelate with transition metals, which may impose deleterious effect on the catalytic activity of metal catalyst; (iii) the 2-aminophenol products generated from dehydrogenative chemical transformations can undergo further reactions with the alkyl-ketone starting materials through a competitive enamine dehydrogenation pathway (19).

To overcome the above-mentioned challenges, we herein report the successful development of transition metal–free syntheses of N-functionalized 2-aminophenols through dehydrogenation-driven couplings between cyclohexanones and primary amines using 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO) as mild oxidant (Fig. 1D) (44–48). In this transformation, the tolerance of primary aliphatic amines as aminating reagent is remarkable considering their coordinating ability and redox susceptibility. Inspired from biochemical system in which water molecules interact with biomolecules through hydrogen bonding to give exclusive chemo- and regioselectivity (49), we discover that water molecules from ketone-amine condensation could lower electron density of the amine moieties and, in the meanwhile, compete with TEMPO for interacting with amino group in products to prevent further undesired oxidation [oxidation potential of TEMPO (0.67 V versus Ag/AgCl) is more positive than that of 2-aminophenol (0.36 V versus Ag/AgCl)] (50, 51). Our unprecedented protocol uses a noncanonical synthetic route using various cyclohexanones as starting materials, achieving the one-shot installation of both amino and hydroxy groups onto phenyl rings. Since diverse substituted cyclohexanones are readily available through convenient manipulations at the α-, β-, and γ-positions of carbonyls (31, 52, 53), polysubstituted 2-aminophenols could be easily obtained while circumventing reactivity and selectivity problems in traditional functionalization of arenes.

RESULTS

We initially investigated the effects of reaction conditions on reaction outcomes using 4-phenylcyclohexanone and 1-methyl-3-phenylpropylamine as mode substrates (Table 1). Pleasingly, 4-phenyl-cyclohexanone reacted with 1-methyl-3-phenylpropylamine in the presence of 3.8 equiv of TEMPO to afford the targeted N-functionalized 2-aminophenol in excellent yield under the optimized conditions (entry 1). Screening of various reaction parameters disclosed how the reaction conditions affected the target reaction outcomes. Replacing 1,4-dioxane solvent with toluene led to a decrease in yield (entry 2). Changing the amounts of 1,4-dioxane also decreased the yield of 2-aminophenol product (entries 3 to 5), illustrating that the reaction concentrations were also a pivotal parameter in controlling the reaction outcome. The influence of varying solvent and reactant concentration on reaction efficiency should result from the discrepancies of interaction modes among different solvent and reactant molecules. Decreasing the amount of 4-phenylcyclohexanone to 1.0 equiv from 1.5 equiv reduced the yield of 2-aminophenol product likely due to the relatively slow condensation between 4-phenylcyclohexanone and primary amine in this reaction system (entry 6).

Table 1. Optimization studies on the syntheses of N-functionalized 2-aminophenols through dehydrogenation-driven reaction between cyclohexanones with amines.

Entry	Variation from standard conditions	Yield*,† (%)
1	None	85 (79)‡
2	Toluene (0.4 ml)	70
3	1,4-Dioxane (0.2 ml)	73
4	1,4-Dioxane (1.0 ml)	68
5	1,4-Dioxane (2.0 ml)	58
6	1.0 Equiv cyclic ketone	63
7	2.0 Equiv TEMPO	51
8	3.0 Equiv TEMPO	78
9	24 hours instead of 36 hours	79
10	48 hours instead of 36 hours	84
11	Adding 4-Å MS (400 mg) to reaction system	5
12	Air atmosphere instead of N2	75

*The reaction was carried out on the 0.2 mmol scale under nitrogen atmosphere.

†Yields were determined by Ultra Performance Liquid Chromatography analysis using pyrazinecarbonitrile as an internal standard.

‡Isolated yield.

Although the multiple desaturations occurring in the model reaction required several equivalents of TEMPO oxidants, decreasing the amount of TEMPO still gave the N-functionalized 2-aminophenol albeit in slightly lowered yields (entries 7 and 8), implicating that the aromatization-driven desaturation was faster than the early oxidative dehydrogenation. Shortening reaction (entry 9) and lengthening reaction (entry 10) time both gave rise to slightly decreased yield compared with the optimized conditions (entry 1). The slightly lowered yield of lengthening reaction time resulted from overoxidation of electron-rich N-functionalized 2-aminophenol product. Although the condensation of ketone with primary amine to form imine and water is a reversible process, introducing 4-Å molecular sieves (MS) greatly hampered the reaction to reduced 5% yield (entry 11). This deleterious effect is consistent with our speculation that water generated from addition of primary amine to ketone would benefit the targeted transformation by protecting electron-rich 2-aminophenol product from oxidation (the experimental evidences were given below). Running this model reaction under air atmosphere still gave good yield (entry 12), illustrating the easy operation of this reaction which favored its further applications.

With the optimized reaction conditions in hand (entry 1, Table 1), we initially explored the substrate scope of this transformation regarding cyclohexanones using aliphatic amines as the aminating reagents (Fig. 2). Notably, owing to our simple conditions without external acid or base additives, the reaction showed impressive tolerance toward a broad range of functional groups. When 1-methyl-3-phenylpropylamine was used as the coupling partner, a wide variety of γ-substituted cyclohexanones (1 to 32) underwent dehydrogenation-driven cross-coupling with amine to form N-functionalized 2-aminophenols in good-to-excellent yields. In this context, phenyls bearing diverse functional groups as well as polycyclic and heterocyclic aromatic compounds were all well tolerated (1 to 15). In addition, no constraint in alkyl substitution patterns was observed (16 to 23). More complex functionalities such as trifluoromethyl (24), esters (25 to 26), protected amines and ketones (27 to 31), and even primary amide (32) were compatible with this transformation. Substituents located at β-position of cyclohexanones as well as disubstituted cyclohexanones had little influence on the reaction efficiency (33 to 38), which offered a concise pathway to rapidly increase product complexity and diversity. The cyclopropyl groups tethered to β- or γ-carbon of cyclohexanones remained intact under current conditions, implying that no radical species formed at the carbon centers adjacent to cyclopropyl groups (39 to 40).

Fig. 2. Substrate scope for syntheses of 2-(alkylamino)phenols from cyclohexanones and aliphatic amines.

Reaction conditions: aliphatic amines (0.2 mmol, 1.0 equiv), cyclohexanone (1.5 equiv), TEMPO (3.8 equiv), 1,4-dioxane (0.4 ml), 120°C, 36 hours, under N₂ atmosphere. Yields are the yields of isolated products. ^aThe corresponding benzoxazole was isolated in 8% yield.

We next evaluated the scope of aliphatic amines as coupling partners. As demonstrated in Fig. 2, α-mono-substituted, α-disubstituted, and α-trisubstituted aliphatic amines were smoothly converted. Briefly, aliphatic amines bearing alkyl and alkoxyl groups (41 to 52), Boc- and benzyl (Bn-) protected amines (53 to 64), ester (65), alcohol (66 to 67), sulfide (68), sulfone (69), and trifluoromethyl (70) groups all underwent the desired transformation to afford the products smoothly. And no erosion was observed when stereocenters were presented (54, 56 to 58, 64 to 66, and 70). It is worth mentioning that this dehydrogenation strategy exhibits a different chemoselectivity and a broader scope compared with our previously developed dehydrogenation-triggered cross-coupling reactions (22). For example, under our previous conditions that contain MS, α-primary aliphatic amines afforded 2-substituted benzoxazoles as major product with small amount of 2-aminophenols, in which uncontrolled overoxidation dominated the process. In addition, α-disubstituted aliphatic amines could act as efficient aminating reagents under current conditions, whereas in our previously developed reactions, they only delivered mixtures of unidentified byproducts upon decomposition. Moreover, it is noticeable to tolerate free alcohol group (67) even under oxidative dehydrogenation circumstances, which highlights the remarkable mild conditions and excellent functional group tolerance of our current protocol.

The 2-arylaminophenols serve as versatile building blocks to construct natural products such as 1-oxygenated-carbazole alkaloids (2) and other types of bioactive compounds (54), which stimulated our further efforts toward establishing dehydrogenation-driven cross-coupling between anilines and cyclohexanones to synthesize 2-arylaminophenols. Considering that anilines have lower nucleophilicity than aliphatic amines, we had to modify the reaction conditions to improve the efficiency of the dehydrogenation-driven reaction between anilines and cyclohexanones.

We tested the feasibility of the reaction between cyclohexanones and anilines and further optimized the reaction on the basis of the above-mentioned conditions applied to aliphatic amines (see the Supplementary Materials for the detailed screening data). Using 3,5-diaminobenzoic acid (5 mol %) as catalyst to facilitate the condensation of cyclohexanones with anilines into imines (55), in combination with employments of TEMPO as oxidant and 4-Å MS as water scavenger, the dehydrogenation-driven reaction between anilines and cyclohexanones successfully occurred in 1,4-dioxane to generate the desired 2-arylaminophenols. As presented in Fig. 3, this newly modified method for 2-arylaminophenols exhibited wide substrate scope. Cyclohexanones bearing various (hetero)aromatic substituents at γ-position proved to be competent substrates for 2-arylaminophenols syntheses (71 to 88 and 104). This reaction is compatible with diphenyl ketone group in which the conversion occurred preferentially at the carbonyl group of cyclohexanone (104). Unsubstituted and a variety of γ-alkylated cyclohexanones also smoothly participated in the reaction with good yields obtained (89 to 96). The polar functional groups located at γ-positions of cyclohexanones, including trifluoromethyl (97), Weinreb amide (98), and Boc-protected amino groups (99 and 100), were tolerable in this reaction. Notably, the reactions of γ-substituted cyclohexanones afforded the meta-substituent–containing 2-arylaminophenol products that are in principle prepared by the classic Buchwald-Hartwig amination methodology (56, 57), which started from the synthetically challenging meta-substituent–containing 2-halogenated phenols. Consequently, our protocol proved a complementary strategy to the Buchwald-Hartwig amination in the synthesis of 2-arylami-nophenols. The positions of substituents on the ring of cyclohexanones exerted little effects on the reaction, as evidenced by β-substituted cyclohexanones (101, 103, and 105) and 3,5-diphenyl-cyclohexanone (102).

Fig. 3. Substrate scope for syntheses of 2-(arylamino)phenols from cyclohexanones and anilines.

Reaction conditions: anilines (0.2 mmol), cyclohexanone (1.5 equiv), 3,5-diaminobenzoic acid (5 mol %), TEMPO (2.8 equiv), 4-Å MS (200 mg), 1,4-dioxane (0.4 ml), 120°C, N₂, 36 hours. Yields are the yields of isolated products.

Furthermore, an array of anilines adorned with diverse functional groups was tested (Fig. 3, 106 to 135). Aryl halides (107 to 109, 71, 129 to 130, F, Cl, Br, and I), boronic ester (111), and carboxylic ester (114) were kept intact, allowing downstream elaborations of the generated products by way of orthogonal transition metal–catalyzed transformations. In addition, the tolerance of anilines bearing electron-donating groups, such as thioether (116), methoxy (122 to 124 and 131), and amino groups (120 and 121), was impressive despite the oxidative conditions, implicating that α-oxygenation and dehydrogenative aromatization on the six-membered ring of cyclohexanone occurred in preference to oxidation of aniline moiety. Positions of substituents on phenyl ring of anilines slightly affected the reaction outcomes as showcased by comparison of 4-methoxy-aniline (122) with 3-methoxy-aniline (123) or 2-methoxy-aniline (124). Unexpectedly, naphthalene-1-amine (125) gave higher yield than the less sterically hindered naphthalene-2-amine (126) presumably due to electronic factor. Fascinatingly, this reaction was compatible with various medicinally relevant motifs that exist in aniline partners, including amide (127), oxazolidinone (128), pyridine (132), pyrazole (133), and indazole (134). Last, the aniline derived from the natural product estrone with fused multiring scaffold could also be used as suitable aminating reagent (135).

To further demonstrate the potentials of this method to prepare bioactive or pharmaceutically relevant compounds, the formation of 2-(arylamino)phenols was combined with the well-known methylation of hydroxyl to ether (58) and the recently developed Pd-catalyzed intramolecular oxidative biaryl formation (59) to constitute the synthetic sequence that enables efficient syntheses of carbazole-based natural products (Fig. 4A). As shown in Fig. 4A, the reactions through these synthetic sequences could be scaled up to gram-scale to synthesize a series of 1-oxygenated-carbazole–based alkaloids including murrayafoline A (antitumor activity, Sigma-Aldrich, $213.00/1 mg) (60), glycozolicine, and mukonine acid ester (61). More notably, our protocol enabled sophisticated amino acid derivatives (139 to 142), structurally complex medicine precursors (143 and 144), pharmaceutical compounds (146 to 148), natural products (145 and 149), and natural product derivatives (150 to 152) to participate in the direct transformations smoothly with high yields obtained (Fig. 4B). As demonstrated by these transformations, our method can be used to efficiently modify the structures of bioactive compounds to improve their biological functions, illustrating the great potentials in accelerating related pharmaceutical discovery studies.

Fig. 4. Synthetic applications of the dehydrogenation-driven cross-coupling between cyclohexanones and primary amines.

(A) Scale-up syntheses of naturally occurring carbazoles in 10 mmol scale. (B) Late-stage diversification of bioactive and medicinally relevant compounds. ^aThe corresponding benzoxazole was isolated in 7% yield. ^bThe corresponding benzoxazole was isolated in 5% yield. TBS, tert-butyldimethylsily.

DISCUSSION

Having demonstrated the synthetic utility of this method, additional investigations were carried out to give mechanistic implications about this transformation. When the suspected reaction intermediate α-enaminone (153) was submitted to a reaction with TEMPO as the sole oxidant, the proposed 2-arylaminophenol (71) was formed with its yields depending on the amounts of TEMPO (Fig. 5A), which disclosed that α-enaminone was the reaction intermediate which eventually formed 2-arylaminophenol from cyclohexanone and aniline. Notably, our current reaction conditions successfully prevented overoxidation of 2-(alkylamino)phenols to 2-substituted benzoxazoles. This distinct and superior selectivity has aroused our interests in confirming our previous speculation that the water molecules generated from cyclohexanone and aliphatic amine condensation might protect 2-(alkylamio)phenol product from oxidation through hydrogen bonding interaction. To get experimental evidences for this speculation, the reaction between 2-(n-hexylamino)-5-phenyl-phenol (46) and 2.0 equiv of TEMPO oxidant was conducted with varying amounts of water as additive. As illustrated in Fig. 5B, 2-aminophenols were completely oxidized by TEMPO in the absence of water to produce 2-substituted benzoxazoles in excellent yield, but external water as additive substantially retarded the oxidation of 2-aminophenols, which decreased the yield of 2-substituted benzoxazoles to 50 from 99%. This observation supported our speculation that water could prevent the overoxidation of 2-aminophenols. This beneficial effect of water should stem from the hydrogen bonding interactions which reduced the electron density of amine moieties as well as impeding the approach of oxidizing TEMPO to product (62). The ¹⁸O incorporation ratio of 2-aminophenols was in consistent with that of TEMP¹⁸O from HRMS analysis, clarifying that the oxygen atom of 2-aminophenol products stemmed from TEMPO (Fig. 5C).

Fig. 5. Mechanistic investigations and the proposed reaction pathway.

(A) Validation of α-enaminone as possible reaction intermediate. (B) Investigation on the effect of water. (C) Investigation on the source of oxygen atom in 2-aminophenol product. (D) Proposed plausible reaction mechanism.

On the basis of an overall analysis of the above observations and our previous work, the detailed reaction pathway was proposed. As showcased by Fig. 5D, initially, condensation between ketone and amine occurs to form imine (I) and enamine (II). Subsequent TEMPO-mediated α-oxyamination (63) and tautomerization form the α-TEMPO–substituted imine (III), which then undergoes elimination of tetramethylpiperidine and tautomerization to give thermodynamically more favored α-enaminone intermediate (IV). Afterward, TEMPO-enabled multiple dehydrogenations of α-enaminone intermediate (IV) delivered N-functionalized 2-amino-phenols as the final product (VII). In this reaction pathway, TEMPO plays double roles, that is, oxygen transfer reagent and desaturation mediator. Notably, these two roles of TEMPO operate preferentially at the specific positions of reaction intermediates, which ensures the exclusive selectivity in formation of desired 2-aminophenol products.

In conclusion, we have achieved the unified syntheses of N-functionalized 2-aminophenols by reasonable design and fine tuning of reaction parameters. This transformation is operated under transition metal–free conditions by using inexpensive TEMPO as mild oxidant, which enables highly selective constructions of N-functionalized 2-aminophenols from readily available cyclohexanones and primary amines in good-to-excellent yields. By using this strategy, concise access to precursors of valuable natural products such as carbazole alkaloids becomes available. In addition, this method is applicable to late-stage modification of complex natural products and pharmaceuticals. The crucial role of water in situ generated from ketone-amine condensation has been identified to protect aliphatic amine moieties from overoxidation by alleviating electron density of amine part and impeding the interaction of TEMPO with vulnerable 2-aminophenol products. The noteworthy observation of water’s protective role in this study is anticipated to substantially contribute to the advancement of highly selective synthetic methodologies. By virtue of its high efficiency, operational simplicity, and remarkable functional group tolerance, this synthetic protocol should be appealing for further applications in pharmaceutical industry and paves the way for the invention of more efficient protocols to generate polyfunctionalized arenes.

MATERIALS AND METHODS

General information

All reactions were conducted under an atmosphere of nitrogen with dry solvents. Unless otherwise noted, chemical reagents were purchased from commercial supplies and used directly without further purification. Dioxane and toluene were distilled from metal Na and stored under nitrogen atmosphere. Four-angstrom MS were purchased from Alfa Aesar [3 to 5 mm (0.12 to 0.20 inch), beads, CAS no. 70955-01-0, LOT: 5002F30N] and activated at 150°C overnight and stored in the nitrogen-filled glove box.

General procedure for the syntheses of 2-(alkylamino)phenols

In a nitrogen-filled glove box, the 35-ml pressure tube equipped with a stir bar was charged with primary amine (0.20 mmol), cyclic ketone (0.30 mmol), and TEMPO (120.0 mg, 0.76 mmol), and then dioxane (0.4 ml) was added to the pressure tube via syringe. Later, the vial was capped with a screw cap fitted with a Poly tetra fluoroethylene (PTFE)–faced silicone septum, removed from the glove box, and stirred at 120°C for 36 hours with a speed of 250 rounds per minute (rpm). After the reaction was finished, the mixture was filtered through a short plug of silica gel and washed with 10 ml of ethyl acetate. The filtrate was concentrated under reduced pressure and purified by flash chromatography on silica gel to provide the corresponding product.

General procedure for the syntheses of 2-(arylamino)phenols

In a nitrogen-filled glove box, a 35-ml pressure tube equipped with a stir bar was charged with primary aniline (0.2 mmol), cyclic ketone (0.28 mmol), 3,5-diaminobenzoic acid (0.01 mmol), MS (4-Å MS, 0.2 g), and TEMPO (88.8 mg, 0.56 mmol), and then dioxane (0.4 ml) was added to the pressure tube via syringe. Later, the reaction mixture was firstly stirred at room temperature for 1 hour and then stirred at 120°C for 36 hours with a speed of 250 rpm. After the reaction was finished, the mixture was filtered through a short plug of silica gel and washed with 10 ml of ethyl acetate. The filtrate was concentrated under reduced pressure and purified by flash chromatography on silica gel to provide the corresponding product.

Supplementary Materials

This PDF file includes:

Supplementary Text

Figs. S1 to S3

Tables S1 to S4

References