Chiral Spiro‐Scaffolded C(sp³),P‐Chelated Iridacycles: Direct Asymmetric Hydrogenation of β,β‐Disubstituted Acrylate Salts

ABSTRACT

Rational design of chiral ligands with precisely tailored steric and electronic properties is pivotal to advancing asymmetric transition‐metal catalysis. Despite their promising strong σ‐donating ability, sp³‐hybridized carbanion ligands have long been underexplored owing to the inherent instability of C(sp³)‐metalated complexes. In this study, we report a modular design of chiral spiro monophosphine ligands ((R)‐SciPhos) integrating an electronic‐withdrawing group, a quaternary carbon moiety, and a rigid spiro scaffold, which efficiently addresses this long‐standing challenge. This judicious design enables the synthesis of bench‐stable neutral C(sp³),P‐chelated iridium complexes via intramolecular C─H activation. These novel iridacycle catalysts exhibit high efficiency, excellent enantioselectivity, and a broad substrate scope in the asymmetric hydrogenation of challenging β,β‐disubstituted sodium acrylates, achieving up to 99% yield and 97% ee with a turnover number (TON) of 5000. Mechanistic studies reveal an Ir(III)/Ir(V) catalytic cycle involving an olefin dihydride iridium(III) intermediate, wherein the migratory insertion step dictates enantioselectivity via noncovalent interactions. This work establishes a general platform for C(sp³)‐metalated complex‐based catalysis, opens new avenues for the development of innovative asymmetric catalytic systems, and unlocks access to previously inaccessible enantioselective transformations.

A novel class of stable chiral spiro‐scaffolded C(sp³),P‐chelated iridium catalysts has been developed to address the long‐standing challenge of the inherent instability of C(sp³)‐metalated complexes. These iridacycle catalysts exhibit high efficiency and enantioselectivity in the asymmetric hydrogenation of challenging β,β‐disubstituted acrylate salts.

1. Introduction

Asymmetric transition‐metal catalysis has revolutionized modern synthetic chemistry, providing powerful and unparalleled tools for the enantioselective construction of chiral molecules essential to pharmaceuticals, agrochemicals, and bioactive natural products [1, 2, 3]. Advances in this field rely critically on the discovery and development of novel chiral ligands, which directly govern the reactivity and enantiocontrol of the metal center by the virtue of their precisely tuned steric and electronic properties [4, 5]. Although numerous efficient chiral ligands are now available, the pursuit of highly efficient asymmetric catalysis continues to demand innovative ligand architectures [6, 7]. Consequently, the rational design of new chiral ligands that enable fine‐tuning the steric and electronic environment surrounding the metal center remains a central and enduring challenge in synthetic chemistry [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]. Progress in this direction will foster the development of novel catalytic systems and open new avenues to previously unattainable asymmetric transformations [21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31].

Chiral ligands derived from classical heteroatomic σ‐donors, such as phosphorus, nitrogen, and sulfur, have been extensively explored in asymmetric catalysis [32, 33, 34, 35] (Scheme 1A). Among these, chiral phosphine ligands stand out due to their strong σ‐donating ability and tunable steric properties, driving substantial advances in both academic research and industrial applications [36]. N‐heterocyclic carbenes (NHCs), another class of potent σ‐donors featuring sp²‐hybridized coordinating carbons, have also become valuable ligands [37, 38]. However, the spatial separation between the chiral center and the sp²‐coordinating carbon within NHCs hinders precise modulation of the steric environment around the metal center, restricting further optimization of catalytic performance. In recent years, metal catalysts coordinated with chiral cyclopentadienyl anions [39, 40, 41, 42, 43, 44, 45, 46] and aryl (sp²)‐carbanions [47, 48, 49, 50] have undergone rapid development and exhibited excellent catalytic performance across diverse asymmetric transformations. In stark contrast, sp³‐hybridized carbanions, despite their structural analogy to phosphines and potentially stronger σ‐donor capability, remain largely underexplored in chiral ligand design [51]. The primary challenge stems from the inherent instability of C(sp³)‐metalated complexes, which readily undergo protonation, reductive elimination, and β‐hydride elimination (Scheme 1B, left). This long‐standing limitation has hindered their application in robust catalytic systems for decades [52, 53, 54, 55, 56, 57].

SCHEME 1.

From classical ligand donors to sp³‐carbanionic donors and the design of C(sp³),P‐chelated iridacycles. (A) From classical heteroatom and carbon donors to sp³‐carbanion donors. (B) Addressing the instability of C(sp³)‐metalated complexes: our stabilization strategy. (C) This work: synthesis of chiral spiro C(sp³),P‐chelated iridacycles and their application in asymmetric hydrogenation.

Building upon our ongoing efforts in chiral spiro ligands and catalysts [58, 59, 60, 61, 62, 63, 64, 65, 66, 67], we hypothesized that a rationally designed ligand architecture could mitigate this instability and unlock the synthetic potential of C(sp³)‐metalated complexes (Scheme 1B, right). Our design integrates three key structural elements: (a) an electron‐withdrawing group to stabilize the sp³‐hydridized carbanion and modulate its nucleophilicity for catalytic compatibility; (b) an adjacent quaternary carbon to suppress β‐hydride elimination, the primary deactivation pathway of late‐transition‐metal alkyl complexes; and (c) a rigid spiro scaffold to embed these functional motifs, enhancing thermodynamic stability and locking the spatial configuration of donor atoms—essential for enantiocontrol. Guided by this strategy, we synthesized a new family of chiral monophosphine ligands, exemplified by (R)‐SciPhos ((R)‐1), derived from the chiral spiro monophenol (R)‐SCIOL featuring a spirochroman‐2‐one‐4,1′‐indane scaffold [63] (Scheme 1C). These ligands undergo lactone‐positioned intramolecular C─H activation [68] to afford bench‐stable neutral C(sp³),P‐chelated iridium complexes Ir‐(R)‐SciPhos ((R,S)‐2). The strong σ‐donation of sp³‐hybridized carbanion is expected to increase the electron density at the iridium center, enhancing the catalytic reactivity, while the rigid spiro backbone creates a well‐defined chiral pocket, enabling exceptional stereocontrol.

Herein, we report the synthesis of C(sp³),P‐chelated chiral iridacycles and their application in the direct asymmetric hydrogenation of β,β‐disubstituted acrylate salts. Chiral carboxylate salts are significant in pharmaceutical development, offering superior aqueous solubility and bioavailability compared to their corresponding carboxylic acids and esters [69]. Despite the well‐established hydrogenation protocols for acrylic acids and esters [70, 71, 72], the transformation of acrylate salts remains largely underexplored. This is primarily due to two key hurdles: (a) catalytic deactivation caused by strong coordination of carboxylate to the metal center, and (b) reduced reactivity resulting from the delocalized negative charge of the carboxylate anion, which diminishes the electrophilicity of the C═C double bond. Prior research includes only one rhodium‐catalyzed example, yielding a modest 39% ee [73], while existing acrylic acid hydrogenation methods remain largely limited to α‐substituted and α,β‐disubstituted substrates, and additive dependency, typically requiring the addition of bases and/or counteranions for optimal activity and enantioselectivity [71]. Our protocol distinguishes itself from conventional enantioselective C─C bond formation [74, 75] through its superior atom economy, unique substrate compatibility, and excellent scalability. The specially designed iridacycle scaffold effectively addresses inherent catalytic limitations, achieving up to 97% ee and a turnover number (TON) of 5000. Mechanistic studies clarify that the hydrogenation proceeds via an Ir(III)/Ir(V) catalytic cycle involving an olefin dihydride iridium(III) intermediate. This work significantly expands the utility of C(sp³)‐metalated complexes and establishes a novel platform for the asymmetric synthesis of chiral carboxylates.

2. Results and Discussion

2.1. Synthesis and Structural Characterization of C(Sp³),P‐Chelated Iridacycles

We initiated our study with the synthesis of the chiral spiro monophosphine ligands (R)‐SciPhos ((R)‐1) and their corresponding C(sp³),P‐chelated iridium complexes Ir‐(R)‐SciPhos ((R,S)‐2) from spiro monophenol (R)‐SCIOL [63], following a previously established protocol [59] (Scheme 2A). Treatment of (R)‐SCIOL with triflic anhydride (Tf₂O) in the presence of pyridine afforded triflate (R)‐5 in an almost quantitative yield (99%). Subsequent palladium‐catalyzed phosphinylation of (R)‐5 with diarylphosphine oxides using a Pd(OAc)₂/dppp catalyst system yielded phosphine oxide intermediates (R)‐6 with excellent yields (89%−98%). Reduction of (R)‐6 with trichlorosilane yielded chiral spiro monophosphine ligands (R)‐1 in 89%−95% yields. Finally, lactone‐directed C─H activation of (R)‐1 with iridium precursor [Ir(COD)Cl]₂ afforded chiral C(sp³),P‐chelated iridacycles Ir‐(R)‐SciPhos ((R,S)‐2) quantitatively, completing the efficient synthesis of the target chiral ligands and their corresponding iridium catalysts.

SCHEME 2.

Synthesis and structural characterization of chiral C(sp³),P‐chelated iridacycles. (A) Synthesis of chiral C(sp³),P‐chelated iridacycles. (B) X‐ray structure of (R,S)‐2d (30% probability ellipsoids). (C) Second‐order perturbation analysis: BD(Ir–C2) → BD(C1 = O). (D) Weak interaction analysis of (R,S)‐2d (by an independent gradient model based on Hirshfeld partition (IGMH))].

Single crystal x‐ray diffraction analysis of the representative C(sp³),P‐chelated iridium complex (R,S)‐2d revealed its key structural features [76] (Scheme 2B). The ligand (R)‐1d coordinates with the iridium center through both the phosphorus atom and the C2 carbon atom of the spiro scaffold, thereby forming a conformationally constrained six‐membered heterometallic ring. This cyclometallation process generates a new chiral center at the carbanionic atom, which adopts an (S)‐configuration. Notably, the P─Ir─C2 bond angle is 91.56°, and the dihedral angle of the spiro scaffold is 90.3°—a value narrower than that of (R)‐SCIOL (99.3°) [63]. These structural features indicate a more compact spatial arrangement of the two phenyl rings in the spiro skeleton upon complex formation, which is presumably responsible for the excellent catalytic performance of the complexes (R,S)‐2. Most importantly, these neutral C(sp³),P‐chelated iridacycles demonstrate excellent bench stability and facile handling, which are highly advantageous for their practical application in asymmetric catalysis.

To further elucidate the role of the lactone moiety in the stability and reactivity of the C(sp³),P‐chelated iridium complexes, density functional theory (DFT) calculations were performed on (R,S)‐2d. Second‐order perturbation analysis reveals a pronounced hyperconjugative interaction between the Ir─C2 bond and the C═O group of the lactone unit, with a stabilization energy of 35.11 kcal/mol, which plays a crucial role in stabilizing the Ir─C2 bond (Scheme 2C). Independent gradient model based on Hirshfeld partition (IGMH) analysis indicates the presence of weak C─H···O interactions between the lactone carbonyl oxygen and two C─H bonds of the COD ligand, which may facilitate substrate recognition and modulation during the catalytic reaction (Scheme 2D). Collectively, these electronic features highlight the crucial role of the lactone moiety in tuning the stability and reactivity of the iridium complexes. Combined with the confined chiral environment imposed by the rigid spiro scaffold, this synergistic effect may enable C(sp³),P‐chelated iridium complexes to exhibit excellent catalytic activity and remarkable enantioselectivity in asymmetric catalytic reactions.

2.2. Optimization of Reaction Conditions for Asymmetric Hydrogenation

The catalytic performance of chiral iridacycles (R,S)‐2 was first explored in the asymmetric hydrogenation of β‐alkyl‐β‐aryl substituted acrylate salts. Sodium (E)‐3‐phenylbut‐2‐enoate (3a) was selected as the model substrate and the catalytic activity of iridium catalysts (R,S)‐2a–d was evaluated in MeOH under 20 atm H₂ at room temperature for 24 h (Table 1). All complexes effectively catalyzed the conversion of 3a to (R)‐4a, achieving 70%−99% conversion and 66%−80% ee (entries 1–4). Among these, (R,S)‐2d, featuring 3,5‐tert‐butyl and 4‐methoxy groups on the P‐phenyl rings, exhibited the best performance (>99% conversion, 80% ee, entry 4) and was thus selected as the optimal catalyst for subsequent condition optimization.

TABLE 1.

Optimization of the reaction conditions for the asymmetric hydrogenation of (E)‐3‐phenylbut‐2‐enoate. ^a

Entries	(R,S)‐2	Solvents	Temp. (°C)	Conv. (%) b	ee (%) c
1	(R,S)‐2a	MeOH	rt	70	66
2	(R,S)‐2b	MeOH	rt	80	71
3	(R,S)‐2c	MeOH	rt	>99	79
4	(R,S)‐2d	MeOH	rt	>99	80
5	(R,S)‐2d	EtOH	rt	43	77
6	(R,S)‐2d	nPrOH	rt	9	64
7	(R,S)‐2d	tBuOH	rt	5	71
8	(R,S)‐2d	THF/MeOH (80:20)	rt	78	80
9	(R,S)‐2d	Toluene/MeOH (80:20)	rt	54	88
10	(R,S)‐2d	CH2Cl2/MeOH (80:20)	rt	20	84
11 d	(R,S)‐2d	Toluene/MeOH (80:20)	rt	>99	88
12 d	(R,S)‐2d	Toluene/MeOH (95:5)	65	>99	91
13 d	(R,S)‐2d	Toluene/MeOH (98:2)	65	>99	95
14 d	(R,S)‐2d	Toluene/MeOH (99:1)	65	>99	93
15 d , e	(R,S)‐2d	Toluene/MeOH (98:2)	65	>99	91
16 d , f	(R,S)‐2d	Toluene/MeOH (98:2)	65	>99	94

^a Reaction conditions: 0.3 mmol scale, 3a/(R,S)‐2 = 200, 1.0 mL, 20 atm H₂, room temperature (20‒25°C), 24 h.

^b Determined by ¹H NMR.

^c Determined by transferring the salt to an ethyl ester and analyzing it by HPLC using a chiral column.

^d 4.0 mL solvent.

^e Lithium salt.

^f Potassium salt.

Solvent screening revealed a profound effect on catalytic efficiency: replacing methanol with other protic solvents (EtOH, nPrOH, and tBuOH) drastically reduced conversion and slightly decreased enantioselectivity (entries 5–7), while aprotic solvents (THF, toluene, and CH₂Cl₂) were ineffective due to the poor solubility of the sodium acrylate salt in these nonpolar media. To enhance enantioselectivity, we investigated MeOH‐based mixed solvent systems with various aprotic solvents (entries 8–10). Gratifyingly, the toluene/MeOH (80:20) mixture afforded a markedly improved enantioselectivity (88% ee) compared to pure MeOH, albeit with a moderate conversion (54%), justifying further optimization of this system. Increasing the total solvent volume to 4.0 mL to ensure complete substrate dissolution maintained both conversion and enantioselectivity (>99% conversion, 88% ee, entry 11). Raising the reaction temperature to 65°C and fine‐tuning the toluene/MeOH ratio further enhanced the enantioselectivity: a 98:2 ratio delivered (R)‐4a with >99% conversion and excellent 95% ee (entry 13), which was identified as the optimal reaction condition. These observations suggest that while MeOH ensures the solubility of sodium acrylates, an excess of MeOH solvates the iridium‐center, weakening substrate–catalyst interactions and compromising enantioselectivity. Conversely, toluene establishes a nonpolar microenvironment that fosters stabilizing secondary interactions, such as C─H···π interaction between the substrate and the chiral catalyst, which in turn enhances enantiocontrol. The optimal 98:2 toluene/MeOH ratio balances substrate solubility and the preservation of the chiral environment, enabling both full conversion and excellent enantioselectivity. Furthermore, control experiments with lithium and potassium salts of the substrate (entries 15 and 16) confirmed the generality, with the potassium salt yielding nearly identical results to the sodium salt (94% ee, >99% conversion). In contrast, the corresponding acrylate ester (CO₂Et) displayed negligible reactivity (<5% conversion), and the acrylate acid (CO₂H) exhibited both low conversion (15%) and poor enantioselectivity (12% ee). This control experiment confirms that the sodium carboxylate group is essential for achieving high reactivity and enantioselectivity in this system. Notably, chiral PHOX‐ and SIPHOX‐derived iridium catalysts (N,P‐ligands), which are highly efficient for the asymmetric hydrogenation of acrylate esters and acids [70, 71, 72], together with the diphosphine BINAP ligand and diamine‐based NHC‐ligand, showed no or only moderate conversions without enantiocontrol under the standard conditions (see Table S2 for details). These results further highlight the unique effectiveness of our C(sp³),P‐chelated iridium catalyst in tolerating strongly coordinating carboxylate salts and enabling efficient and enantioselective hydrogenation of this challenging class of substrates.

2.3. Substrate Scope of the Asymmetric Hydrogenation

Under the optimal reaction conditions, we explored the substrate scope of the asymmetric hydrogenation for a series of β‐alkyl‐β‐aryl substituted sodium acrylates 3a–w (Scheme 3). All substrates were effectively hydrogenated to the corresponding chiral carboxylates (R)‐ or (S)‐4a–w with excellent yields (88%‒99%) and high enantioselectivities (82%‒97% ee), demonstrating the broad applicability of the catalytic system. The electronic properties of the aryl ring substituents exerted only a minor influence on both reactivity and enantioselectivity, with both electron‐donating and withdrawing groups well tolerated. Meta‐ and ortho‐substituted aryl substrates 4g–o exhibited a slight reduction in enantioselectivity (86%‒97% ee), which was attributed to the increased steric hindrance of the substituents around the C═C double bond. Consistent with this steric trend, the 2,6‐dimethyl disubstituted substrate 3p required an increased catalyst loading of 3.0 mol% and a longer reaction time to furnish (R)‐4p in 88% yield with 96% ee. Heterocycle‐containing substrates bearing furan (3q) and thiophene (3r) moieties also under went hydrogenation smoothly, affording products (R)‐4q and (R)‐4r with comparable enantioselectivity, albeit requiring a higher catalyst loading (1.0 mol%) and a longer reaction time (48 h). Replacing the β‐methyl group with bulkier alkyl substituents (n‐pentyl (3s), benzyl (3t), cyclohexyl (3u), and cyclopropyl (3v) did not compromise catalytic efficiency, with all reactions proceeding to >99% conversion and affording 90%‒95% ee. Notably, the ester‐containing substrate 3w was hydrogenated to (S)‐4w in 96% yield and 86% ee, highlighting the excellent functional group tolerance of the catalyst (R,S)‐2d. A stereochemical probe with the Z‐isomer of 3a yielded (R)‐4a in 98% yield but with significantly diminished enantioselectivity (47% ee), indicating that the E‐geometry of the C═C double bond is a key factor governing reaction enantiocontrol.

SCHEME 3.

Asymmetric hydrogenation of β,β‐disubstituted sodium acrylates catalyzed by (R,S)‐2d. [a] Reaction conditions: 0.3 mmol scale, 3/(R,S)‐2 = 200, 4.0 mL toluene/MeOH (98:2), 20 atm H₂, 65°C, 24 h. [b] Isolated yields. [c] Enantiomeric excess (ee) values were determined after conversion of the sodium salts to ethyl ester or anilide derivatives, followed by HPLC analysis using a chiral column. [d] S/C = 33, 48 h. [e] S/C = 100, toluene/MeOH (95:5), 48 h. [f] toluene/MeOH (95:5), 30°C, 24 h. [g] 36 h.

Encouraged by the outstanding performance with β‐alkyl‐β‐aryl substituted sodium acrylates, we extended the substrate scope to β,β‐diaryl substituted sodium acrylates, another challenging class of substrates for asymmetric hydrogenation. Successful examples for such substrates remain limited, despite the substantial advances in the hydrogenation of α,β‐unsaturated carboxylic acids [77, 78, 79]. Following a brief optimization of the reaction conditions (see Table S2 for details), (R,S)‐2d efficiently catalyzed the hydrogenation of a diverse range of β,β‐diaryl sodium acrylates, affording (R)‐ or (S)‐4t‐ag with 95%‒99% yields and 84%‒97% ee (Scheme 3). In contrast to the β‐alkyl‐β‐aryl series, the enantioselectivity of β,β‐diaryl substrates was sensitive to electronic effects: electron‐donating substituents (e.g., Me and MeO) at para‐ or ortho‐positions of the aryl rings resulted in excellent enantioselectivities (90%–97% ee), while electron‐withdrawing groups (e.g., halogens and CF₃) led to a moderate reduction in enantioselectivity (78%‒90% ee). Meta‐substituted β,β‐diaryl substrates also exhibited lower enantioselectivity (84%‒88% ee), consistent with the steric effects observed in the β‐alkyl‐β‐aryl series. Finally, β,β‐dialkyl and α‐alkyl‐β‐aryl derivatives were also successfully hydrogenated to 4ah (98% yield, 67% ee) and 4ai (98% yield, 81% ee), respectively, demonstrating the generality of the catalyst for a wide range of disubstituted acrylate salts—even the most challenging dialkyl‐substituted substrates.

2.4. Scale‐Up Experiments and Synthetic Applications

To assess the practical applicability of the method, we conducted scale‐up experiments using model substrate 3a, reducing the catalyst loading to 0.1 mol% (Scheme 4A). Under the optimized conditions, the reaction afforded only 27% conversion with 82% ee after 48 h, indicating that catalyst turnover was the limiting factor at low loading (entry 1). We postulated that trace HCl, generated during active iridium hydride species formation from the chloride‐containing iridium precursor, might facilitate the dissociation of product from the metal center and enhance catalytic turnover. To verify this hypothesis, several acids, including HCl, HCO₂H, PhCO₂H, and the conjugate acid of 3a (3a‐H), were added to the reaction system. Gratifyingly, adding 2 mol% of these acids led to a striking improvement in conversion (90%−>99%) along with a pronounced enhancement in enantioselectivity (92%−94% ee, entries 2–5). Further reducing the catalytic loading to 0.02 mol% (S/C = 5000) and increasing 3a‐H to 4.0 mol% still afforded >99% conversion with 93% ee after 4 days (entry 6). This result demonstrates the exceptional catalytic efficiency and robustness of (R,S)‐2d, achieving a turnover number (TON) of up to 5000, which is among the highest reported for the asymmetric hydrogenation of β,β‐disubstituted acrylic acids and esters [70, 71, 72].

SCHEME 4.

Gram‐scale asymmetric hydrogenation and its applications. (A) The influence of acid addition on the asymmetric hydrogenation of 3a. (B) Gram‐scale synthesis of key intermediates for pharmaceuticals containing β‐aryl β‐cyclopropyl propanoic acids.

Chiral β‐cyclopropyl β‐arylpropanoic acids are privileged scaffolds in medicinal chemistry, serving as key intermediates for the synthesis of pharmaceuticals, including AM‐1638 [80], SCO‐267 [81] (GPR40 full agonists for diabetes treatment), and runcaciguat [82] (a novel soluble guanylate cyclase activator for cardiovascular diseases) (Scheme 4B). Current synthetic methods for these chiral acids suffer from significant limitations, such as narrow substrate scope, stringent catalyst requirements, or complex workup procedures [83, 84, 85, 86]. To address these drawbacks, we developed a practical and scalable synthetic route to these valuable intermediates using our established hydrogenation method. First, 3‐aryl‐3‐cyclopropylacrylates 5a–c were synthesized from aryl cyclopropyl ketones via the Horner‐Wadsworth‐Emmons reaction, followed by hydrolysis with sodium hydroxide to afford the corresponding 3‐aryl‐3‐cyclopropyl sodium acrylates 6a–c in 85%‒90% yield. Gram‐scale asymmetric hydrogenation of 6a–c with 0.2 mol% (R,S)‐2d under the optimal conditions (in the presence of 4 mol% of conjugate acids 6‐H) proceeded to completion after 48 h, affording the carboxylates (R)‐7a–c in 98‒99 yields and 81‒94% ee. These salts can be readily converted to the corresponding free β‐aryl β‐cyclopropyl propanoic acids via acidification, establishing a practical and scalable route to these valuable pharmaceuticals.

2.5. Mechanistic Investigations

To gain mechanistic insights into the asymmetric hydrogenation, we conducted deuterium‐labeling experiments with the model substrate 3a and catalyst (R,S)‐2d (Scheme 5A). When the reaction was performed with D₂ in MeOH, product (R)‐d ₃‐4a was obtained with partial deuterium incorporation at the α‐ (85%) and β‐positions (35%) of the carboxylate, alongside with H/D exchange at the ortho‐positions on the phenyl ring (average 35%). This ortho‐deuteration indicates reversible, substrate‐induced transient cyclometalation during the catalytic cycle. Nevertheless, sterically encumbered substrates 3a and 3b, bearing 3,5‐ and 2,6‐dimethyl substituents that suppress ortho‐C─H activation, still exhibit excellent enantioselectivity (Scheme 3), demonstrating that transient cyclometalation is not essential for achieving high stereocontrol. In contrast, the reaction with H₂ in CD₃OD afforded (R)‐d ₂‐4a with low deuterium incorporation at the α‐ (13%) and β‐positions (15%). These labeling patterns are consistent with dynamic H/D exchange between the active iridium hydride species and MeOH, supporting an olefin iridium dihydride intermediate in the catalyst cycle [87].

SCHEME 5.

Deuterium‐labeling and investigation of iridium dihydride species studies. (A) Deuterium‐labeling of (E)‐3a with D₂ and CD₃OD. (B) Investigation of the iridium dihydride intermediates for (R,S)‐2d.

We further probed the nature of the active catalytic species via NMR spectroscopy and high‐resolution mass spectrometry (HRMS) (Scheme 5B). Treatment of (R,S)‐2d in d ₈‐toluene/CD₃OD (1:2) under 40 atm H₂ at 50°C for 24 h yielded a pair of doublet‐of‐doublets signals at δ −14.27 (d, J = 26.4 Hz) and −14.91 (d, J = 25.2 Hz) in ¹H NMR spectrum, assigned to an iridium dihydride species 8 (the formation of a dimeric species cannot be entirely excluded). Upon addition of one equivalent of the substrate 3a and stirring for 2 h, two new sets of doublet‐of‐doublets signals appeared at δ −14.41/−17.79 and −14.43/−17.85 (J = 25–27 Hz), consistent with the coordination of 3a to the iridium dihydride center and formation of substrate‐bound iridium dihydride complexes 9a and/or 9b. These assignments were further supported by ³¹P NMR spectroscopy, which exhibited characteristic shifts for the phosphorus atom upon substrate coordination. HRMS analysis confirmed the formation of the iridium dihydride species 8 (Calcd for C₄₇H₆₁IrNO₄P⁺ [M + H]⁺: 913.3931; found: 913.3932), while no signals corresponding to the substrate‐coordinated complexes 9a/b were observed. This indicates that substrate coordination is reversible and that the substrate dissociates readily from the metal center, consistent with the high turnover rate of the catalyst.

Based on the experimental observations and the reported Ir(III)/Ir(V) catalytic pathway for iridium/phosphine‐oxazoline‐catalyzed asymmetric hydrogenation of unsaturated carboxylic acids [88, 89], we performed DFT calculations to elucidate the reaction mechanism and the origin of enantioselectivity, using (E)‐3‐phenylbut‐2‐enoate anion 3a′ as the model substrate. Consistent the mechanistic studies (Scheme 5B), the iridium(III) dihydride complex 8 was used as the active catalyst species of the reaction. The computations show that the formation of intermediate (R)‐IM1 via the coordination of 3a to the iridium center of 8 is exergonic by 13.5 kcal/mol (see Supporting Information for details). Intermediate (R)‐IM1 adopts the octahedral configuration, wherein the O atom of 3a′ occupying the apical position and the C─C double bond situated in the equatorial plane trans to the alkyl ligand. On the other hand, intermediate (S)‐IM1, where the C─C double bond is positioned trans to the phosphine ligand, is 5.8 kcal/mol less stable than (R)‐IM1. Therefore, the sum of the free energies of (R)‐IM1 and the substrates was set as the zero point on the relative free energy scale (Figure 1A).

FIGURE 1.

DFT calculations of the reaction pathways. (A) Calculated energy profile. (B) Optimized geometries. The bond distances are given in Å.

The reaction begins with the migratory insertion of the C═C double bond into the Ir─H bond. Starting from (R)‐IM1, the migratory insertion was found to proceed through transition state (R)‐TS1, leading to Ir(III) intermediate (R)‐IM2. Then, the coordination of the incoming H₂ to the Ir center gives intermediate (R)‐IM3. The ensuing H─H oxidative addition was calculated to occur through transition state (R)‐TS2, delivering Ir(V) intermediate (R)‐IM4. The catalytic cycle is completed by the C─H reductive elimination through transition state (R)‐TS3, affording the final product (R)‐4a′ together with the Ir(III) complex (R)‐IM1. It should be noted that the σ‐metathesis pathway from (R)‐IM3 was also examined, but it consistently converged to the C─H reductive elimination transition state (R)‐TS3. The computations indicate that the migratory insertion step is irreversible and therefore constitutes the enantioselectivity‐determining step of the reaction. The transition state (R)‐TS1 was calculated to be lower in energy than (S)‐TS1 by 2.0 kcal/mol, in agreement with the experimentally observed enantioselectivity. The optimized geometries (Figure 1B) show that in (R)‐TS1, C─H···π interactions between the Ph and Me substituents of 3a′ and the tBu and aryl groups of the chiral ligand are present, whereas in (S)‐TS1 only a C─H···O hydrogen‐bonding interaction is observed. These noncovalent interactions render (R)‐TS1 more stable than (S)‐TS1, thereby accounting for the experimentally observed enantioselectivity. It is worth mentioning that the E‐geometry of the C═C double bond is a key factor governing reaction enantiocontrol, whereas the Z‐isomer of 3a affords significantly diminished enantioselectivity (47% ee, Scheme 3). To further understand this effect, the key migratory insertion step involving the Z‐isomer of 3a was also investigated. The calculated energy difference between the competing transition states is only 0.5 kcal/mol, consistent with the experimentally observed poor enantioselectivity (see Figure S4 for details).

3. Conclusion

In summary, we have developed a new class of cyclometalated iridium(III) complexes from chiral spiro monophosphine ligands with a unique spirochroman‐2‐one‐4,1′‐indane scaffold. A rational ligand design, integrating an electronic‐withdrawing group, a quaternary carbon, and a rigid spiro scaffold, efficiently mitigates the inherent instability of C(sp³)‐metalated complexes via intramolecular C─H activation, resolving a longstanding challenge in asymmetric catalysis. These iridacycle catalysts exhibit exceptional activity and enantioselectivity in the direct asymmetric hydrogenation of β,β‐disubstituted acrylates. The optimal catalyst (R,S)‐2d enables the hydrogenation of a broad scope of β,β‐disubstituted acrylates (β‐alkyl‐β‐aryl, β,β‐diaryl, β,β‐dialkyl) to chiral carboxylates with up to 99% yield, 97% ee, and an ultrahigh TON of 5000. Gram‐scale experiments confirm the scalability, and the protocol has been successfully applied to the efficient synthesis of key intermediates for pharmaceuticals AM‐1638, SCO‐267, and Runcaciguat, overcoming the limitations of existing synthetic routes. Notably, this catalytic system avoids the use of expensive external counterions (e.g. BAr^F ₄₋) commonly required in Crabtree‐type systems [90, 91, 92], greatly enhancing its practical value. Mechanistic studies, including deuterium‐labeling experiments and DFT calculations, reveal an Ir(III)/Ir(V) catalytic cycle involving an olefin‐iridium dihydride intermediate. The migratory insertion is the enantioselectivity‐determining step, where noncovalent C─H···π interactions govern the high enantiocontrol of the reaction. This work establishes a general platform for C(sp³)‐metalated catalysts and advances chiral sp³‐carbanionic donor ligand design. Our rational ligand design strategy is anticipated to extend to other transition metals and asymmetric transformations, expanding the utility of C(sp³)‐metalated complexes in synthetic chemistry. Ongoing research focuses on broadening the catalytic scope of these C(sp³),P‐chelated iridium catalysts to other challenging asymmetric reactions.

Author Contributions

Yang‐Ming Zhang: methodology, data curation, formal analysis, and validation. Yong‐Rui Li: methodology, validation, data curation, and formal analysis. Yue Shi: software and data curation. Xiao‐Jie Zheng: methodology, validation, and data curation. Shu‐Min Guo: writing – review and editing and supervision. Zhenbo Mo: software, data curation, and writing – original draft. Genping Huang: software and data curation. Jian‐Hua Xie: conceptualization, supervision, writing – original draft, writing – review and editing, investigation, funding acquisition, resources, project administration, and visualization. Qi‐Lin Zhou: funding acquisition, and conceptualization.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: The authors have cited additional references within the Supporting Information.

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information of this article.