Palladium‐Catalyzed Intramolecular Asymmetric Allylic Amidation: A General Route to 5‐Substituted Chiral Oxazolines

ABSTRACT

A general and modular catalytic enantioselective route to 5‐substituted chiral oxazolines has been developed via palladium‐catalyzed intramolecular asymmetric allylic amidation. This strategy employs readily available racemic 2‐substituted‐5‐styryl oxazolidinones as substrates. Key to the success is the identification of a sterically hindered biphenol‐derived phosphoramidite ligand, which, in combination with Pd₂(dba)₃, promotes the cyclization with high efficiency and enantiocontrol. The reaction demonstrates broad substrate scope and excellent functional group tolerance, accommodating a wide range of sterically diverse 2‐substituents and electronically varied aryl allylic units, delivering the desired 5‐vinyl oxazolines in good yields with high enantioselectivity (up to 99% ee).

A highly efficient palladium‐catalyzed asymmetric intramolecular allylic amidation of racemic 2‐substituted‐5‐styryl oxazolidinones is developed for 5‐substituted chiral oxazolines. With a sterically hindered biphenol‐derived phosphoramidite/Pd₂(dba)₃ catalyst, the reaction delivers products with up to 99% ee and broad substrate scope, enabling gram‐scale synthesis and representing the general catalytic enantioselective route to these N,O‐heterocycles.

1. Introduction

Chiral N,O‐heterocycles represent a privileged structural class commonly found in natural products and contemporary pharmaceuticals [1, 2]. In particular, oxazolines and their derivatives have garnered sustained interest owing to their broad range of biological activities, such as antifungal, antiviral, antitumor, anti‐inflammatory, and analgesic properties (Figure 1), as well as their valuable role as chiral ligands in asymmetric catalysis [3, 4, 5, 6, 7, 8, 9, 10]. While 4‐substituted chiral oxazolines have been extensively studied and are widely represented in bioactive molecules and catalytic systems, their 5‐substituted counterparts have received comparatively less attention in terms of general and modular asymmetric access [11]. In particular, the enantioselective construction of 5‐substituted oxazolines independent of chiral amino alcohols derived from natural amino acids remains a synthetically attractive but challenging objective. Consequently, the development of catalytic enantioselective routes to such heterocyclic architectures has become an active area of investigation. To date, established strategies encompass organocatalytic asymmetric halocyclization of unsaturated amides [12, 13, 14, 15, 16, 17, 18] and transition‐metal‐catalyzed approaches as shown in Scheme 1a [19, 20]. Representative examples comprise the ruthenium‐catalyzed asymmetric hydrogenation of substituted oxazoles reported by the Kuwano group [19], and the palladium‐catalyzed asymmetric coupling–cyclization of N‐allenamides with aryl or alkenyl iodides developed developed by Ma and coworkers [20], which provide efficient access to various 5‐substituted oxazoline derivatives. Despite these advances, a general, catalytic, and stereocontrolled method for assembling structurally diverse 5‐substituted chiral oxazolines from readily available starting materials remains highly desirable.

FIGURE 1.

Selected natural products and drugs containing oxazoline with chiral centers at 5‐position.

SCHEME 1.

(A) Known methods for the catalytic enantioselective synthesis of 5‐oxazolines. (B) This work: Palladium‐catalyzed intramolecular enantioselective synthesis of 5‐Substituted oxazolines.

Recent advances in transition‐metal‐catalyzed asymmetric allylic alkylation (AAA) have furnished key mechanistic understanding and reliable synthetic methodologies, demonstrating considerable potential for the assembly of complex chiral molecules, especially heterocyclic frameworks [21, 22, 23, 24, 25, 26, 27, 28, 29]. In this context, Feringa and coworkers reported in 2014 the pioneering enantioselective iridium‐catalyzed intramolecular allylic substitution using an amide oxygen as the nucleophile, enabling efficient synthesis of benzoxazine derivatives along with a single example of a 5‐vinyl oxazoline with high stereocontrol [30]. More recently, Dong and colleagues described an iridium‐catalyzed kinetic resolution of racemic benzoxazinones via intramolecular asymmetric allylation, delivering a range of chiral products in moderate to good yields with excellent enantioselectivity [31]. Despite these notable achievements, palladium‐catalyzed allylic substitution approaches toward chiral oxazolines have predominantly focused on diastereoselective transformations; to the best of our knowledge, a general catalytic enantioselective route to 5‐substituted chiral oxazolines via palladium‐catalyzed AAA has not been reported [32, 33, 34, 35, 36, 37]. To address this gap, we envisioned that 5‐styryl oxazolidinones could serve as suitable precursors for conversion into the target 5‐oxazolines through an intramolecular palladium‐catalyzed asymmetric allylic amidation (Scheme 1b). This strategy would not only establish a direct and versatile catalytic enantioselective entry to this valuable heterocyclic scaffold, but also leverage the broad synthetic applicability of palladium catalysis for further molecular elaboration and diversification.

2. Results and Discussion

Using (E)‐1‐isobutyryl‐4‐styrylpyrrolidin‐2‐one 1a as the model substrate in the presence of Pd(OAc)₂ in toluene at 0°C, various chiral phosphine ligands were examined (Table 1, entry 1−12). Phosphoramidite‐type ligands showed promising catalytic performance (see Supporting Information). Subsequent systematic evaluation of chiral phosphoramidite ligands revealed that the binaphthol‐derived phosphoramidite bearing a chiral dibenzylamine moiety afforded moderate enantioselectivity (36% ee, entry 5). Modification of the 8H‐binaphthol backbone did not improve the enantioselectivity of product 2a. Replacing the 8H‐binaphthol unit with biphenol‐derived phosphite (ligand L6) significantly enhanced the enantioselectivity (entry 6). Further structural tuning of the chiral diamine component demonstrated that sterically hindered, electron‐rich variants led to better outcomes. Optimization of phosphoramidite ligands L7–L10 highlighted the importance of steric bulk: the bulkiest ligand L10 provided 2a in 68% yield with 83% ee at 0°C (entry 12). Further screening of reaction solvents and palladium sources identified toluene and Pd₂(dba)₃ as optimal, delivering the asymmetric allylic amidation product 2a in 74% yield and 88% ee (entry 18).

TABLE 1.

Optimization of reaction conditions ^a .

Entry	Ligand	T (°C)	Solvent	[Pd]	Yield (%) b	ee (%) c
1	L1	30	Toluene	Pd(OAc)2	40	5
2	L2	30	Toluene	Pd(OAc)2	21	20
3	L3	30	Toluene	Pd(OAc)2	30	17
4	L4	30	Toluene	Pd(OAc)2	65	12
5	L5	30	Toluene	Pd(OAc)2	48	36
6	L6	30	Toluene	Pd(OAc)2	45	56
7	L6	0	Toluene	Pd(OAc)2	41	66
8	L6	−10	Toluene	Pd(OAc)2	23	72
9	L7	0	Toluene	Pd(OAc)2	12	59
10	L8	0	Toluene	Pd(OAc)2	38	41
11	L9	0	Toluene	Pd(OAc)2	23	75
12	L10	0	Toluene	Pd(OAc)2	68	83
13	L10	0	PhF	Pd(OAc)2	36	82
14	L10	0	PhCl	Pd(OAc)2	56	80
15	L10	0	CH2Cl2	Pd(OAc)2	62	63
16	L10	0	Acetone	Pd(OAc)2	60	61
17	L10	0	Toluene	[PdCl2(CH3CN)]2	Trace	—
18	L10	0	Toluene	Pd2(dba)3	74	88
19	L10	0	Toluene	Pd(acac)2	70	86
20	L10	0	Toluene	Pd(tBuCO2)2	69	70

^a Reaction comditions: 1a (0.1 mmol), [Pd] (5 mol%), Ligand (10 mol%), in 1.0 mL solvent, 0°C, 48 h.

^b Isolated yield.

^c The ee values were determined by chiral HPLC.

With the optimized conditions for the palladium‐catalyzed intramolecular asymmetric allylic amidation to construct 5‐substituted oxazolines established, we next explored the substrate scope of this transformation as depicted in Scheme 2. Systematic evaluation of a series of racemic 2‐substituted oxazolidinones revealed that increasing steric bulk at the 2‐position (Scheme 2a), which arises from substituents such as 1,1‐diethylmethyl, phenylisopropyl, tert‐butyl, and 1,1‐dimethylpropyl, significantly improved both the yield and enantioselectivity of the cyclization, furnishing the corresponding 5‐substituted oxazolines with excellent enantioselectivity (up to 97% ee). Similarly, oxazolidinones bearing other sterically demanding cyclic substituents at the 2‐position also underwent efficient cyclization to furnish the desired chiral 5‐substituted vinyl oxazolines (2g–2i, 2k–2m) in high yields (59%–88% yield) with excellent enantiocontrol (91%–98% ee). In contrast, substrates with less hindered 2‐substituents gave slightly lower enantioselectivities, as exemplified by products 2f and 2j. The absolute configuration of 2m was established by single‐crystal X‐ray diffraction analysis, enabling the stereochemical assignment of the remaining analogues by analogy.

SCHEME 2.

Palladium‐catalyzed intramolecular asymmetric allylic amidation: A general route to 5‐Substituted chiral oxazolines. ^areaction conditions: 1 (0.1 mmol), Pd₂(dba)₃ (2.5 mol%), L10 (10 mol%), in 1.0 mL toluene, 0°C, 72 h. ^bIsolated yield. ^cThe ee values were determined by chiral HPLC.

We next examined a diverse array of aryl‐substituted allylic moieties in this allylic amidation/cyclization reaction (Scheme 2b). Substrates bearing para‐ or meta‐substituted aryl groups, irrespective of their electronic nature (encompassing electron‐withdrawing substituents such as NO₂, F, Cl, and Br as well as electron‐donating groups such as tert‐butyl and methyl), underwent efficient cyclization to furnish the corresponding chiral 5‐substituted oxazolines 2n–2u and 2w–2z in 59%–81% yield with 86%–96% ee. Furthermore, naphthyl‐functionalized allylic oxazolidinones proved compatible, delivering products 2v and 2aa in moderate yields with 82%–92% ee. Despite the broad generality demonstrated by the substrates described above, the replacement of the alkenyl substituent with heteroaromatic groups such as furan and thiophene proved challenging. For the furan‐ and thiophene‐containing substrates, the desired cyclization proceeded, albeit with low enantioselectivities, giving products 2ab and 2ac in 30% ee and 35% ee, respectively. Likewise, when the alkenyl group was substituted with aliphatic moieties such as cyclohexyl, only trace amounts of the desired product 2ad were detected, and the starting materials remained largely unconsumed. To further demonstrate the synthetic utility of this method, several commercially available carboxylic acid‐containing pharmaceuticals, including valproic acid, gemfibrozil, and fenofibric acid, were derivatized into the corresponding 2‐substituted oxazolidinones. Subjecting these substrates to the standard catalytic conditions delivered the target products 2ae−2ag in good yields (69%–90%) with high to excellent enantioselectivities (82%–99% ee) (Scheme 2c).

To evaluate the synthetic utility of the Pd‐catalyzed intramolecular allylation for the enantioselective synthesis of 5‐substituted oxazolines, a gram‐scale transformation of racemic model substrate 1c was carried out, delivering the desired product 2c in 69% yield with 92% ee (Scheme 3a). As illustrated, the chiral 5‐substituted oxazoline product can undergo diverse downstream functionalization, highlighting its synthetic versatility (Scheme 3b). Reduction of 2c with aluminum hydride (AlH₃, generated in situ from LiAlH₄ and AlCl₃) proceeded smoothly to afford β‐amino alcohol 3 in 52% yield without loss of enantiopurity (92% ee) [38]. The olefin moiety of 2e could be converted via ozonolysis/reduction to furnish hydroxymethyl derivative 4 in 66% yield with full retention of stereochemical integrity. Furthermore, the oxetane substituent at the 2‐position of 2 h was selectively hydrolyzed under acidic conditions to give diol 5 in 75% yield, while the oxazoline ring remained intact under these conditions.

SCHEME 3.

Scale‐up reaction and synthetic transformation.

Given the use of racemic 5‐arylvinyloxazolidinones in this study, we explored the potential involvement of kinetic resolution as shown in Scheme 4. Time‐dependent analysis of racemic 1c under the standard conditions was conducted by quenching the reaction at 24 and 48 h, respectively. Notably, the enantiomeric excess (ee) of product 2c remained excellent and unchanged in both cases (92% ee), while the recovered starting material exhibited ee values of 31% and 42%, respectively. These results demonstrate that a dynamic kinetic resolution (DKR) of the racemic starting material operates, allowing the product enantiopurity to be preserved throughout the process. This behavior establishes robust stereocontrol within the system, highlighting its potential for the asymmetric synthesis of challenging targets.

SCHEME 4.

Kinetic resolution studies.

To predict the absolute configuration of the product, we propose a catalytic cycle for the model reaction as depicted in Scheme 5. Oxidative addition of substrate 1a to the Pd⁰ L10 complex, which is generated in situ by ligand exchange from Pd₂(dba)₃ and L10, affords intermediate A. Decarboxylation of this intermediate can lead to two intermediates, B and B′. Intermediate B is sterically less congested than the allyl‐Pd‐oxacyclic intermediate B′; in the latter, the amide substituent experiences significant steric repulsion with the amine moiety of the chiral ligand. Moreover, both B and B′ benefit from π–π stacking interactions between the aromatic ring and the biphenol unit of the ligand, which help stabilize the intermediates and favor the formation of B. Consequently, intermediate B preferentially undergoes trans nucleophilic attack to give the palladium–olefin complex C. Ligand dissociation from C regenerates the Pd(0)/L10 catalyst and delivers the desired product 2a with the observed R‐configuration. The critical role of π–π stacking interactions is further supported by substrate scope studies. For heteroaromatic substrates such as furan and thiophene, these interactions are weakened, resulting in lower enantioselectivities. In the case of aliphatic substrates, the absence of such interactions likely disfavors the transition state, leading to poor reactivity. These mechanistic insights will guide future ligand design to accommodate more challenging substrate classes.

SCHEME 5.

Proposed mechanism.

3. Conclusion

In summary, we have developed a general, modular catalytic enantioselective route to 5‐substituted chiral oxazolines via palladium‐catalyzed intramolecular asymmetric allylic amidation of readily accessible racemic substrates. Key to this success is the identification of a sterically hindered biphenol‐derived phosphoramidite ligand, which, together with Pd₂(dba)₃, enables efficient cyclization with broad substrate scope and excellent functional group tolerance, delivering products in good yields and high enantioselectivity (up to 99% ee). Gram‐scale synthesis confirms the method's practicality, proceeding efficiently without enantiocontrol erosion and providing a stereoselective platform for advancing chiral N,O‐heterocycle synthesis and related catalytic strategies.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Additional supporting information can be found online in the Supporting Information section. Additional references cited within the Supporting Information [39, 40, 41, 42, 43, 44, 45, 46].

Supporting File 1:chem71046‐sup‐0002‐SuppMat.pdf.

Supporting File 2: chem71046‐sup‐0002‐Data.zip.

Dedicated to Professor Jieping Zhu on the occasion of his 60th birthday