Copper-Catalyzed Amination of Aryl Chlorides under Mild Reaction Conditions

Abstract

We report a mild method for the copper-catalyzed amination of aryl chlorides. Key to the success of the method was the use of highly sterically encumbered N¹,N²-diaryl diamine ligands which resist catalyst deactivation, allowing reactions to proceed at significantly lower temperatures and with a broader scope than current protocols. A sequence of highly chemoselective C−N and C−O cross-coupling reactions were demonstrated, and mechanistic studies indicate that oxidative addition of the Cu catalyst to the aryl chlorides is rate-limiting. We anticipate that the design principles disclosed herein will help motivate further advances in Cu-catalyzed transformations of aryl chlorides.

N-Aryl amines are prevalent in a wide variety of functional organic molecules, and methods for their synthesis continue to benefit the field of organic chemistry.^1–6 While palladium catalyzed methods have proven particularly enabling for accessing N-aryl amines,⁷ the relatively high toxicity and cost of Pd has motivated the development of Cu based alternatives.^8–10

The venerable Ullmann-type Cu-catalyzed C−N coupling reactions have recently regained attention. The well-known difficulties associated with oxidative addition of aryl halides to Cu, however, often necessitates the use of aryl iodides as coupling partners.¹¹ In an effort to increase Cu reactivity toward oxidative addition, pioneering studies from Ma have resulted in the development of anionic ligands—particularly those based on the oxalamide scaffold—that increase the electron richness of Cu and hence accelerate this process.^12–15

Guided by DFT calculations, our group has recently introduced a new class of anionic N¹,N²-diarylbenzene-1,2-diamine ligands for Cu.^16–18 By simultaneously increasing the electron density on Cu and stabilizing the active anionic catalyst through Cu-π interactions, these ligands enable rapid amination of aryl bromides at room temperature.

Compared to aryl bromides and iodides that are typically employed for C−N coupling reactions with Cu catalysts, aryl chlorides are more attractive due to their greater commercial availability and lower cost (Scheme 1A).^19,20 Despite this, Cu catalysts for C−N coupling reactions that operate on aryl chlorides are particularly rare. This is due largely to the slow rate of oxidative addition.²¹ While Cu catalysts supported by oxalamide ligands have demonstrated effective C−N coupling of aryl chlorides with primary amines (Scheme 1B),²² limitations remain. We felt that there was an opportunity to expand the scope and improve the robustness of aryl chloride amination chemistry in a meaningful way through the use of catalysts based on appropriate anionic N¹,N²-diarylbenzene-1,2-diamine ligands. Herein, we describe the successful realization of this goal, resulting in a catalytic system capable of promoting efficient coupling of complex aryl chlorides and amines under mild reaction conditions (Scheme 1C).

Scheme 1. Contemporary Cu-Catalyzed C−N Coupling of Aryl Chlorides.

In our original report,¹⁶ we speculated that the intramolecular C−H amination of L4 was responsible for the inefficiency of L4-based Cu catalyst in the amination of hindered aryl bromides. This catalyst deactivation could be at least partially mitigated by installing large substituents proximal to the site of undesired ligand modification, as in the case of L5 that possessed tert-butyl groups at the 3′ and 5′ positions of both distal aromatic rings. Subsequently, we wondered whether L5 would facilitate using aryl chlorides as substrates for the amination reaction. Employing the coupling reaction of 4-chloroanisole and morpholine as our test case, we found that L5 was entirely resistant to the carbazole formation pathway (Scheme 2A, D-2). Another ligand modification observed is the N-arylation of the ligand (D-1), a common side process in Cu-catalyzed C−N coupling reactions with diamine ligands.^11d,23 This undesired pathway resulted in consuming 6% of the aryl chloride (see Supporting Information for details and characterization of isolated degradation products). Collectively, the installation of ^tBu groups prevented complete catalyst deactivation, with 40% of L5 remaining after the reaction. Moreover, the use of L5 in the model coupling reaction gave product 3a in excellent yield (94%, entry 1).

Scheme 2. Ligand Evaluation and Reaction Optimization.

Various analogs of L5 were investigated in the model coupling reaction to produce 3a. Substituting the ^tBu groups of L5 with CF₃ groups to produce L6 resulted in a substantially lower yield of 3a (entry 3). Similarly, the use of L8, which we have recently shown to be an excellent ligand for coupling base-sensitive aryl bromides,¹⁸ also led to diminished yield (entry 5). These results suggest that decreasing the electron richness of Cu through the use of less electron-donating ligands results in a diminished ability of the resulting complexes to undergo oxidative addition to aryl chlorides, leading to lower yields of 3a. Consistent with our previous DFT calculations that support that the pendant aromatic ring of the ligand is involved in a Cu-π interaction to stabilize the active anionic Cu(I) catalyst,¹⁶ a poor yield of 3a was obtained when L7, an analog possessing a cyclohexyl group in place of the phenyl of L4, was employed (entry 4).

While conducting the model reaction in DMSO proved to be optimal, using alternative dipolar aprotic solvents, including DMAc and sulfolane, provided comparable yields (entries 6 and 7). An efficient C−N coupling reaction could also be carried out using binary solvent mixtures of DMSO and PhMe (e.g., entry 8). The best conditions were found to involve performing reactions at 55 °C. However, only slight decreases in yield were observed at either 40 °C or room temperature (entries 9 and 10), a feature which may be useful to promote the amination of substrates with thermally sensitive functional groups. Control experiments in which CuBr or L5 was omitted or Pd(OAc)₂ was used in lieu of CuBr, resulted in no detectable yield of 3a (entries 11 and 12), strongly supporting our view that any trace Pd from the preparation of L5 cannot be solely responsible for the observed reactivity.

With optimized reaction conditions in hand, we assessed the substrate scope of the process. The amination of electron-rich and electron-deficient chloroarenes with primary, cyclic, or acyclic secondary amines produced the desired products in good-to-excellent yields (Figure 1). The reaction exhibited tolerance of many common functional groups, including nitriles (3d, 3j), tertiary amides (3g), esters (3h), Bocprotected secondary amines (3i), alcohols (3m), and olefins (3v). Note: NaOMe combined with DMSO promotes the condensation of esters with amines. In Figure 1, example 3h, for the reaction of methyl 3-chlorobenzoate and (S)-2-methylpiperidine—approximately 15% of the amidation product derivatized from 3h was formed as the byproduct.

Figure 1.

Representative substrate scope. All yields are the average of two isolated yields. Standard reaction conditions: aryl chloride (0.5 mmol), amine (0.7 mmol), CuBr (5 mol %), L5 (10 mol %), NaOMe (0.75 mmol), DMSO (0.5 mL), 40 °C, 24 h. ^aCuBr (7.5 mol %) and L5 (15 mol %). ^bAt 55 °C. ^cAt 70 °C. ^dDMSO/PhMe (1:1, v/v) as solvent. ^eDMSO (2 M).

Cyclic aliphatic amines are among the most prevalent heterocycles in FDA-approved drugs.^24,25 However, examples of Cu-catalyzed C−N coupling of aryl chlorides with this important substrate class remain rare.^22,26–28 Our protocol enabled the combination of a variety of cyclic amines, including five-(3s, 3t), six- (3a–3e, 3h–3j, 3r, 3u), and seven-membered (3p) azacycles with a number of aryl chlorides. While the reaction of α-tertiary amines and ortho-substituted aryl chlorides is well-known to be challenging,^26,29 the successful coupling of tert-butylamine (3k), 1-adamantylamine (3l), 2-chloroanisole (3m), 3-chloro-4-methylpyridine (3n), and 2-chlorobiphenyl (3o) underscores the efficacy of this catalytic system in the coupling of hindered substrates. Molecules that contain important N-heterocycles including pyridine (3d, 3p), pyrimidine (3c), pyrrole (3k), pyrazole (3q), quinoline (3r, 3s), isoquinoline (3t), benzothiadiazole (3u) and benzothiophene (3v) were well-tolerated. We note that reactions of highly base-sensitive five-membered heteroarenes (thiazole, oxazole, and imidazole)^30,31 or acidic-proton-containing substrates led to little-to-no formation of desired product 3w–3y (see Supporting Information for more details).

To further explore the synthetic utility of the method, several complex aryl chlorides and amines derived from pharmaceuticals were examined (Figure 2). Perphenazine, an antipsychotic drug featuring an unprotected primary hydroxyl group, was coupled to morpholine in 77% yield (4a). Furthermore, etoricoxib, a COX-2 inhibitor with two pyridyl groups—which frequently pose challenges for cross coupling protocols—underwent successful high-yielding amination with dimethylamine (4b). The antidepressant paroxetine was subjected to the coupling protocol, producing 4c in moderate yield. We presume that the relatively low yield of 4c is due to the increased steric hindrance arising from the 3,4-disubstitution of the piperidine substrate. The analogous reaction with morpholine gives a 98% yield (see the SI for details). Lastly, the amination of an antifungal drug terconazole with morpholine was achieved in an excellent yield (91%, 4d). Together, these results highlight the protocol’s efficacy in elaborating complex pharmaceuticals.

Figure 2.

Diversification of pharmaceuticals. All yields represent the average of two isolated yields. Standard reaction conditions: aryl chloride (0.5 mmol), amine (0.7 mmol), CuBr (5 mol %), L5 (10 mol %), NaOMe (0.75 mmol), DMSO (0.5 mL), 40 °C, 24 h. ^aCuBr (7.5 mol %) and L5 (15 mol %). ^bAt 55 °C. ^cDimethylamine solution (2 M in THF, 350 μL) and DMSO (150 μL) were used. ^dDMSO/PhMe (1:1, v/v) was used as solvent to enhance the solubility of terconazole.

We further investigated a series of reactions to demonstrate the capacity of the N¹,N²-diarylbenzene-1,2-diamine ligand family to selectively couple highly functionalized substrates in a chemoselective manner (Figure 3). We next carried out the sequential amination of 1-bromo-3-chlorobenzene (5a) first with morpholine at room temperature to afford 6a (quantitative isolated yield) and then with n-butylamine, resulting in the formation of the bis-aminated aryl 7a in 92% yield. Similarly, desloratadine (5b), a secondary amine with a C(sp²)−Cl bond, could be coupled to 4-bromoanisole and then subsequently with morpholine to provide 7b in high yield. These sequential amination reactions could also be conducted in one-pot processes, albeit in lower yields (60% for 7a and 32% for 7b). We have recently shown that an L8-based Cu catalyst facilitates C−O coupling of alcohols and aryl bromides.¹⁷ Using consecutive chemoselective C−N and C−O coupling processes to treat perphenazine, we obtained 7c in good yield (Figure 3B).

Figure 3.

Chemoselective sequential C−N and C−O couplings. All yields represent the average of two isolated yields. (A) Sequential amination of the substrates containing both an aryl bromide and an aryl chloride, (B) sequential C−N and C−O cross-coupling.

We next determined the experimental rate law for the coupling of 3-chloropyridine with morpholine (Figure 4A). Using the method of initial rates, the reaction was determined to obey a positive first-order dependence on the concentration of 3-chloropyridine and the L5-based Cu catalyst. These dependencies are consistent with oxidative addition of the Cu catalyst to the aryl halide being at least partially rate-limiting. In support of this, the concentration of morpholine did not impact the initial rate of the transformation. Further, a Hammett study of the reaction rates of 4-substituted aryl chlorides revealed a positive ρ value (ρ = +0.68, Figure 4B). Compared to our previous study on the Cu-catalyzed C−N coupling of aryl bromides using L4 (ρ = +0.27),¹⁶ the higher observed ρ value in this study reflects the increased sensitivity of Cu to electronic effects, presumably due to the comparatively later transition state of aryl chloride oxidative addition relative to that of aryl bromides.³²

Figure 4.

Mechanistic studies of the Cu-catalyzed C−N coupling of aryl chlorides. (A) Initial rates kinetics studies, (B) Hammett analysis of the reaction, (C) proposed catalytic cycle of the C−N coupling of aryl chlorides.

A catalytic cycle consistent with the above observations is depicted in Figure 4C. Following deprotonation of the diamine ligand, the active monoanionic Cu(I) complex reacts with aryl chloride to generate an oxidative addition complex. Subsequent amine binding and deprotonation steps leads to the formation of a formally Cu(III) amido which is poised to undergo reductive elimination, producing the C−N coupled product and regenerating the active Cu catalyst.^33,34

In summary, we have developed a practical and general method for the Cu-catalyzed amination of aryl chlorides under mild reaction conditions. Central to this advance was using an L5-based Cu catalyst that mitigates several catalyst deactivation pathways. This method was successfully applied to the C−N coupling of (hetero)aryl chlorides with a number of structurally diverse amines. Various complex aryl chlorides and amines derived from pharmaceuticals were highly successful substrates. We anticipate that these results will motivate further developments of the Cu-catalyzed transformation of aryl chlorides.