Mechanistic insights into NHC/Cu catalyzed asymmetric synthesis of spirooxindoles: origins of enantioselectivity and diastereoselectivity

Abstract

Density functional theory (DFT) calculations were conducted to elucidate the mechanism of NHC/Cu catalyzed enantioselective annulation between isatin-derived enals and ethynyl carbonates, enabling the asymmetric synthesis of spirooxindole δ-lactones with vicinal all-carbon quaternary stereocenters. The catalytic process involves four key stages: (i) generation of an azolium homoenolate intermediate IM3via NHC-mediated nucleophilic addition to enal 1a; (ii) [Cu]-catalyzed decarboxylation of ethynyl carbonate 2a to afford the copper-alkynyl intermediate IM7; (iii) stereoselective C–C bond formation between IM3 and IM7, followed by a two-water-mediated enol–keto tautomerization yielding the ketone intermediate IM10; and (iv) NEt₃-promoted deprotonation, intramolecular cyclization, and proton transfer affording the product spirooxindole δ-lactone 3a. Notably, both the enantio- and diastereoselectivity-determining step and the rate-determining step occur in Stage III. Furthermore, DIAS and QTAIM analyses of four stereoisomeric transition states identified TS5(S,R) as the most favorable, exhibiting the lowest free energy barrier and multiple stabilizing non-covalent interactions (C–H⋯π, C–H⋯O, C–F⋯π, and Cu⋯H), rationalizing the observed high stereoselectivity. Water molecules are shown to play a crucial role in lowering the barrier of the rate-determining step by enhancing electrostatic, orbital, and dispersion interactions. This study not only deepens the mechanistic understanding of cooperative NHC/Cu catalysis but also provides valuable theoretical guidance for rational designing of next-generation asymmetric annulation reactions.

---

Introduction

The spirooxindole skeleton, featuring multiple contiguous stereocenters, represents a privileged structural motif that is widely found in a variety of natural products and biologically active molecules.^1–8 Due to its architectural complexity and pharmacological relevance, this framework has attracted significant attention from both medicinal and organic synthetic chemists. Consequently, a range of synthetic strategies for constructing the spirooxindole core have been extensively developed in recent years.^9–14 In particular, the incorporation of two chiral centers within the heterocyclic backbone plays a crucial role in modulating the stereochemical configuration and extending bioactive molecules and natural products, which remains an area of active research to date.¹⁵

Nitrogen heterocyclic carbenes (NHCs) have emerged as versatile organocatalysts due to their unique electronic and steric effects.^16–18 Their strong nucleophilicity, derived from a lone pair of electrons on the carbene carbon, enables NHCs to react with carbonyl compounds,¹⁹ thereby converting these electrophiles into nucleophiles via the formation of nucleophilic Breslow intermediates.²⁰ Over the past two decades, substantial progress has been achieved in asymmetric transformations by introducing chirality into the NHC framework.^21–23 However, single NHC catalyst systems often encounter limitations in achieving high levels of chemo-, regio-, and stereoselectivity.

To overcome these limitations, a cooperative catalytic approach that combines NHCs with transition metal catalysis has been developed.^24,25 For example, several hybrid catalytic systems incorporating NHC and transition metal (e.g., Co, Rh, Ir, Ru, Ni, and Pd²⁶) combinations have demonstrated enhanced reactivity, improved stereoselectivity, and access to previously inaccessible transformations.²⁷ Notably, in 2016, Glorius et al. reported a cooperative NHC/Pd-catalyzed enantioselective umpolung annulation, which efficiently avoided mutual catalyst deactivation and delivered a series of benzazepine derivatives with excellent enantioselectivity.²⁸ Building upon this concept, Gong and colleagues in 2019 introduced an NHC/Cu cooperative catalytic system for the asymmetric synthesis of spirooxindole derivatives via enantioselective [3 + 3] and [3 + 4] annulation reactions.²⁹ This strategy has since attracted widespread interest and has been expanded upon in subsequent studies.^30–36 As exemplified in Scheme 1, under the cooperative catalytic influence of the NHC/Cu catalyst (i.e., NHC/[Cu]), isatin-derived enal 1a reacts efficiently with ethynyl cyclic carbonate 2a, generating the spirooxindole δ-lactone 3a. This transformation affords the desired product in 75% yield and delivers excellent enantioselectivity (99% ee) and diastereoselectivity (95 : 5 dr), constructing highly congested vicinal all-carbon quaternary stereocenters. Although the authors proposed a reaction course in the original study,²⁹ the catalytic roles of NHC and [Cu] throughout the whole reaction pathway, as well as the origins of the observed enantiomeric and diastereomeric selectivity, remain elusive.

Scheme 1 NHC/[Cu] cooperative catalysis for the asymmetric synthesis of spirooxindole δ-lactone 3a.²⁹

To address these knowledge gaps, we conducted a detailed mechanistic^37,38 investigation on this unprecedented reaction (Scheme 1), explicitly incorporating the experimentally used solvent (PhF) and base (NEt₃) under the relevant catalytic conditions. Additionally, we elucidated the origins of the high enantioselectivity and diastereoselectivity observed experimentally. By analyzing the energetics and geometries of key intermediates and transition states, we aimed to identify the rate-determining step (RDS) and uncover the structural and electronic factors governing the selectivity. We believe that these insights will not only deepen the mechanistic understanding of cooperative NHC/Cu catalysis but also provide valuable guidance for improving existing catalytic systems and designing new NHC/Cu catalyzed asymmetric transformations.

Computational details

Geometry optimizations without symmetry restriction were carried out using the Gaussian 16 program,³⁹ and the structures were generated using CYLview.⁴⁰ Geometry optimizations were carried out without symmetry constraints using the BP86^41,42 functional with Grimme's D3BJ dispersion correction,⁴³ in conjunction with the def2-SVP basis set.⁴⁴ Harmonic vibrational frequency analyses were conducted at the same level of theory to verify each stationary point as either a minimum (no imaginary frequency) or a transition state (one imaginary frequency) and to obtain thermochemical corrections. Intrinsic reaction coordinate (IRC) calculations were performed to confirm the connection between transition states and the corresponding minima.⁴⁵ Single-point energy refinements were conducted using the BP86-D3BJ/def2-TZVPP level of theory on the optimized geometries, with solvation effects modeled using the SMD implicit solvation model (solvent = fluorobenzene).⁴⁶ The employment of the BP86 functional is well supported by benchmark studies reported for similar Cu-catalyzed reactions.^47–50 Unless otherwise stated, Gibbs free energies reported in the text refer to BP86 + D3BJ/def2-TZVPP (SMD, solvent = fluorobenzene) //BP86 + D3BJ/def2-SVP(SMD, solvent = fluorobenzene) values. To further analyze noncovalent interactions and electron density topology, IGMH (independent gradient model based on Hirshfeld partition) and AIM (atoms in molecules) analyses were conducted using the Multiwfn 3.8⁵¹ software package. Noncovalent interaction plots were rendered using PyMOL.⁵²

The bonding situation in the diatomic molecules was further analyzed by means of an energy decomposition analysis (EDA) which was introduced by Morokuma⁵³ and by Ziegler and Rauk⁵⁴ in conjunction with the natural orbitals for chemical valence (NOCV)^55,56 method. The EDA-NOCV^57,58 calculations were carried out with the ADF 2024 program package⁵⁹ at the BP86 + D3(BJ) level with the Slater-type basis function of TZ2P quality⁶⁰ using the BP86 + D3(BJ)/def2-SVP optimized geometries. In this analysis, intrinsic interaction energy (ΔE_int) between two fragments can be divided into four energy components as follows:

ΔE_int = ΔE_elstat + ΔE_Pauli + ΔE_orb + ΔE_disp (1)

The electrostatic term, ΔE_elstat, represents the quasi-classic electrostatic interaction between the unperturbed charge distributions of the prepared fragments, and the Pauli repulsion ΔE_Pauli corresponds to the energy change associated with the transformation from the superposition of the unperturbed electron densities of the isolated fragments to the wavefunction, which properly obeys the Pauli principle through explicit antisymmetrization and renormalization of the production wavefunction. Since we used D3(BJ), this method also gives the dispersion contribution (ΔE_disp). The orbital term ΔE_orb comprises the mixing of orbitals and charge transfer and polarization between the isolated fragments. The energy change involved in the latter step, which is the main difference between the Morokuma⁵³ and Ziegler/Rauk⁵⁴ approaches, is calculated with an extension of Slater's transition state method for energy differences. It is often referred to as the ETS method. The orbital term ΔE_orb can be further decomposed into contributions from each irreducible representation of the point group of the interacting system as follows:

The combination of the EDA with NOCV enables the partition of the total orbital interactions into pairwise contributions of the orbital interactions which is very vital to get a complete picture of the bonding. The charge deformation Δρ_k(r), resulting from the mixing of the orbital pairs Ψ_k(r) and Ψ_−k(r) of the interacting fragments, presents the amount and the shape of the charge flow due to the orbital interactions (eqn (3)). The associated energy term ΔE_orb gives the size of the stabilizing orbital energy originating from the interaction (eqn (4)).

More details about the EDA-NOCV method and its application are given in recent review articles.^61,62

Results and discussion

We herein present a comprehensive mechanistic investigation of the target reaction (Scheme 1), using the experimentally employed isatin-derived enal 1a and ethynyl cyclic carbonate 2a as model substrates. Importantly, the full catalytic contributions of both the N-heterocyclic carbene (NHC) and [Cu] catalysts were rigorously considered without any structural simplification. As outlined in Scheme 2, the catalytic process proceeds through four key stages. Stage I involves the generation of the azolium homoenolate intermediate (IM3) through the initial activation of substrate 1a. In Stage II, under the cooperative catalysis of NHC/Cu, ethynyl cyclic carbonate 2a is activated and undergoes decarboxylation generating the key copper–alkynyl cationic intermediate (IM7). Stage III features the stereoselective C–C bond formation between IM3 and IM7, leading to the ketone intermediate (IM10). Finally, in Stage IV, the catalytic cycle is completed by the formation of the spirooxindole δ-lactone product 3a, bearing two adjacent all-carbon quaternary stereocenters, along with the regeneration of both the NHC and [Cu] catalysts for the next catalytic cycle.

Scheme 2 NHC/Cu cooperative catalysis via four stages.

Stage I: formation of the azolium homoenolate intermediate IM3 under NHC catalysis

The reaction is initiated by the nucleophilic attack of the NHC catalyst on the substrate 1a, forming a weak complex IM1 (Fig. 1). This intermediate then proceeds through transition state TS1A, with an overall free energy barrier of 11.2 kcal mol⁻¹ relative to the initial NHC and 1a, to afford the (S)-configured Breslow intermediate IM2A. Given that this step establishes a new stereogenic center, an alternative pathway via transition state TS1B affording the (R)-configured Breslow intermediate IM2B, was also explored. However, TS1B is 0.8 kcal mol⁻¹ higher in free energy than TS1A, indicating that this pathway is kinetically less favorable. In addition, two additional transition states, TS1C and TS1D, were identified, with even higher barriers of 13.2 and 14.3 kcal mol⁻¹, respectively. These higher barriers are primarily attributed to increased steric repulsion from the fluorophenyl substituent on the NHC catalyst, which introduces significant hindrance and destabilizes the corresponding transition states. These findings clearly establish TS1A as the most favorable transition state, making it the dominant pathway for the formation of the key Breslow intermediate IM2A. To gain insight into the stability of TS1A, IGMH and EDA-NOCV analyses were performed. As shown in Fig. S1 and Table S1, TS1A exhibits pronounced π–π stacking, strong electrostatic and orbital interactions, and the lowest preparation energy, thereby achieving the most favorable balance between intrinsic interactions and structural reorganization. In contrast, TS1B and TS1C lack effective π–π interactions, while TS1D, despite displaying some degree of noncovalent stabilization, requires substantially higher preparation energy. Consequently, TS1A presents the lowest barrier and emerges as the most favorable transition state.

Fig. 1 Computed free energy profiles (in kcal mol⁻¹) for Stage I. The optimized geometries of transition states are provided in Fig. S2 and S3 of the SI.

Subsequently, IM2A undergoes a [1,2]-proton transfer to form the azolium homoenolate intermediate IM3. Notably, the two-water-assisted pathway, in which IM2A associates with two water molecules to generate IM2A′ and proceeds viaTS2A (IM2A′ → TS2A), displays the lowest energy barrier of 8.8 kcal mol⁻¹. This pathway clearly outperforms all alternatives, including the NEt₃-mediated proton transfer (TS2C), the direct intramolecular [1,2]-proton transfer (TS2E), and the intramolecular [1,2]-proton transfer to the adjacent carbon center (TS2D), each of which presents significantly higher barriers ranging from 14.1 to 35.7 kcal mol⁻¹. Attempts to locate a transition state for a single water molecule-mediated proton transfer were unsuccessful. A comparable two-water-assisted [1,2]-proton transfer from the (R)-configured intermediate IM2B′ proceeds viaTS2B, but with a slightly higher barrier of 11.0 kcal mol⁻¹, making it slightly less favorable than the Si-face pathway viaTS2A.

Stage II: formation of the copper-alkynyl cationic intermediate IM7 under [Cu] catalysis

In this stage (Fig. 2), the active [Cu] catalyst is generated in situ from Cu(CH₃CN)₄PF₆ and the NHC ligand, followed by coordination of ethynyl cyclic carbonate 2a to the Cu center, forming the 2a-[Cu] complex.

Fig. 2 Computed free energy profiles (in kcal mol⁻¹) for Stage II. The optimized geometries of transition states are provided in Fig. S4 of the SI.

Upon addition of NEt₃, a slightly more stable intermediate IM4 is formed, wherein the terminal alkyne hydrogen of 2a engages in a weak interaction with the NEt₃ species. After overcoming a low barrier of 6.4 kcal mol⁻¹via the transition state TS3, the terminal alkyne hydrogen can be successfully transferred to the N-atom of NEt₃, leading to the formation of the thermodynamically more stable copper–alkynyl complex IM5 and the release of Et₃N-H⁺ species. The resulting Et₃N-H⁺ then rearranges to form a less stable intermediate IM6, which undergoes decarboxylation of the carbonate moiety through transition state TS4via a low barrier of 2.0 kcal mol⁻¹ (i.e., IM6 → TS4), releasing CO₂ and yielding the key copper–alkynyl cationic intermediate IM7. When measured from the more stable IM5, the overall barrier for this decarboxylation step (IM5 → TS4) becomes 16.1 kcal mol⁻¹, which is still accessible under mild experimental conditions. The formation of IM7 is a crucial event in the catalytic cycle, as it generates the highly electrophilic species required for the subsequent stereoselective C–C bond-forming reaction, thereby playing a central role in enabling the annulation process.

Stage III: formation of intermediate IM10 bearing vicinal all-carbon quaternary stereocenters under cooperative NHC and [Cu] catalysis

The following pivotal step involves the formation of a C–C bond between the azolium homoenolate intermediate (IM3) generated in stage I and the copper–alkynyl cationic intermediate (IM7) formed in stage II which determines both the enantioselectivity and diastereoselectivity of the overall reaction. As illustrated in Fig. 3, the initial association of IM3 and IM7 generates a more stable intermediate IM8 with the reaction exergonic by 3.1 kcal mol⁻¹, providing a strong thermodynamic driving force for the reaction to proceed. Subsequent C–C bond formation proceeds through four competing transition states, TS5(S,R), TS5(R,S), TS5(S,S), and TS5(R,R), each corresponding to a distinct enantio- and diastereomeric pathway toward the spirocyclic products. Among them, TS5(S,R) is clearly the most favorable, exhibiting the lowest free energy barrier of 11.6 kcal mol⁻¹. In contrast, the alternative pathways are energetically less favorable, with TS5(R,S), TS5(S,S), and TS5(R,R) lying 3.5, 4.9, and 9.0 kcal mol⁻¹ higher in free energy, respectively, than TS5(S,R).

Fig. 3 Computed free energy profiles (in kcal mol⁻¹) for Stage III. The optimized geometries of transition states are provided in Fig. S5 and S6 of the SI.

Subsequent to the formation of the enol intermediate IM9, the reaction undergoes an enol–ketone isomerization to generate the ketone intermediate IM10. Three distinct transition states, TS6A, TS6B, and TS6C were identified for this transformation. Among them, TS6A exhibits the lowest free energy barrier of 24.9 kcal mol⁻¹ (IM9 → TS6A), making it the most kinetically favorable pathway. This transition state is characterized by the cooperative involvement of two explicit water molecules, which establish a hydrogen-bonding network that facilitates an efficient proton relay mechanism, thereby significantly reducing the barrier. We also attempted models with three water molecules; however, the calculations failed to yield stable structures due to steric congestion and unfavorable geometric arrangements. In comparison, TS6B, assisted by only one water molecule, has a higher barrier of 35.5 kcal mol⁻¹ (IM9 → TS6B), while the unassisted pathway viaTS6C displays the highest barrier of 58.8 kcal mol⁻¹ (IM9 → TS6C), making both routes unfavorable under standard conditions. Additionally, we examined the intramolecular hydroxyl-assisted tautomerization pathway viaTS6D. However, its significantly higher barrier (32.8 kcal mol⁻¹) makes it much less favorable than the two-water-assisted pathway viaTS6A (24.9 kcal mol⁻¹). These results highlight the dual role of water molecules, not only as proton shuttles but also as crucial facilitators that reduce the kinetic barrier of the transformation. Notably, TS6A is identified as the rate-determining step (RDS) of the entire catalytic cycle, exerting a critical influence on the overall reaction rate and efficiency.

Stage IV: cyclization to form spirooxindole δ-lactone 3a and catalyst regeneration

In the final stage (Fig. 4), the addition of NEt₃ to intermediate IM10 initially generates a weak complex IM11, which proceeds through the transition state TS7 with a barrier of 11.3 kcal mol⁻¹ (relative to IM10 and NEt₃), affording intermediate IM12 and liberating an Et₃N-H⁺ species. IM12 subsequently undergoes an intramolecular cyclization via transition state TS8, with a barrier of 22.4 kcal mol⁻¹, in a process that coincides with the dissociation of the NHC catalyst and leads to the formation of the copper–alkynyl spirooxindole intermediate IM13. The addition of Et₃N-H⁺ to IM13 gives a more stable intermediate IM14, which undergoes proton transfer via transition state TS9, with a barrier of 9.9 kcal mol⁻¹, to afford the weak π-complex IM15, comprising the final product 3a and the [Cu] catalyst. Finally, IM15 undergoes ligand exchange with one equiv. of the ethynyl cyclic carbonate substrate 2a, releasing the spirooxindole δ-lactone product 3a with vicinal all-carbon quaternary stereocenters, and regenerating the more stable 2a–[Cu] complex for the next catalytic cycle. The overall transformation is highly exergonic by 42.7 kcal mol⁻¹, providing a strong thermodynamic driving force that ensures efficient forward progression of the reaction.

Fig. 4 Computed free energy profiles (in kcal mol⁻¹) for Stage IV. The optimized geometries of transition states are provided in Fig. S7 of the SI.

Based on the comprehensive mechanistic investigation discussed above, we integrated all four stages to construct the most favorable reaction pathway, as illustrated in Scheme 3. Under the cooperative catalysis of NHC and [Cu], the reaction is initiated through two parallel channels. In one branch, the NHC catalyst adds to the substrate 1a, followed by a two-water-assisted proton transfer to generate the azolium homoenolate intermediate IM3 (Stage I). In parallel, the in situ formed [Cu] catalyst coordinates with ethynyl cyclic carbonate 2a and, upon NEt₃-assisted deprotonation and decarboxylation, furnishes the copper–alkynyl cationic intermediate IM7 (Stage II). These two reactive intermediates, IM3 and IM7, then associate in an exergonic step (ΔG = −3.1 kcal mol⁻¹) and undergo stereoselective C–C bond formation via transition state TS5(S,R), yielding the (S,R)-configured enol intermediate IM9. A subsequent enol–keto tautomerization converts IM9 into the corresponding ketone intermediate IM10 (Stage III). NEt₃ then re-engages in the catalytic cycle, facilitating the sequential deprotonation, intramolecular cyclization and proton transfer, ultimately affording the spirooxindole δ-lactone 3a (Stage IV). Notably, the two-water-assisted enol–keto tautomerization via transition state TS6A, with a free energy barrier of the 24.9 kcal mol⁻¹, was identified as the rate-determining step (RDS) of the overall catalytic cycle. The whole reaction is highly exergonic (ΔG = −42.7 kcal mol⁻¹), which affords a strong thermodynamic driving force that ensures the efficient progression of the reaction in the forward direction.

Scheme 3 Complete catalytic cycle of the reaction.

Further discussion

Origin of the observed enantio- and diastereoselectivity

To elucidate the origin of the observed enantio- and diastereoselectivity, the four stereoisomeric transition states, TS5(S,R), TS5(R,S), TS5(S,S), and TS5(R,R), were fully optimized. Their geometries, relative Gibbs free energies, and the key C–C bond distances are summarized in Fig. 5. Boltzmann population analysis indicates that TS5(S,R) overwhelmingly dominates the located transition states, accounting for 99.70% of the population. In contrast, TS5(R,S), TS5(S,S) and TS5(R,R) contribute only 0.27%, 0.026%, and 0.000026%, respectively. These results clearly identify TS5(S,R) as the stereo-determining transition state, determining the absolute configuration of the final product. The predicted diastereomeric ratio (>99 : 1) and enantiomeric excess (>99%) align very well with the experimental observations (dr > 95 : 5, ee = 99%), thereby proving strong support to the proposed mechanism and confirming the validity of the computational model.

Fig. 5 Optimized structures of TS5 of the four diastereomeric transition states (ΔG relative to TS5(S,R)). Non-covalent interaction sites are in TS5(S,R), TS5(R,S), TS5(S,S) and TS5(R,R).

To further elucidate the origin of enantio- and diastereoselectivity, we performed a distortion–interaction activation strain (DIAS) analysis, in which the potential energy surface of the reaction system (ΔE) is decomposed into two key components: the distortion energy (ΔE_strain), representing the energy required to deform the reactants to their transition-state geometries, and the intrinsic interaction energy (ΔE_int) between the distorted fragments. As summarized in Table 1, TS5(S,R) exhibits the most favorable intrinsic interaction energy (ΔE_int = −75.3 kcal mol⁻¹) while maintaining a moderate distortion energy (ΔE_strain = 32.1 kcal mol⁻¹), resulting in the lowest relative Gibbs free energy (ΔG = −16.0 kcal mol⁻¹). In comparison, TS5(R,S), TS5(S,S) and TS5(R,R) exhibit higher ΔG values of −9.2, −7.6, and −6.1 kcal mol⁻¹, respectively, primarily due to increased distortion penalties and less favorable fragment interactions.

Table 1 Quantitative analysis of the four enantio- and diastereoselectivity-determining transition states with the distortion/interaction-activation (DIAS) model. The energy values are in kcal mol⁻¹

Transition state	TS5(S,R)	TS5(R,S)	TS5(S,S)	TS5(R,R)
a The distortion energy of IM4. b The distortion energy of IM7.
ΔEstrain^a	17.8	24.3	16.9	23.1
ΔEstrain^b	14.3	17.0	13.4	9.8
ΔEstrain	32.1	41.2	30.3	32.9
ΔEint	−75.3	−73.4	−64.1	−63.8
ΔG	−16.0	−9.2	−7.6	−6.1

---

Additionally, quantum theory of atoms in molecules (QTAIM) analysis was conducted on the four key transition states. This topological analysis enabled the identification and quantification of critical non-covalent interactions that influence the relative stabilities of the transition states. As shown in Fig. 5 and Table 2, TS5(S,R) is the most stable transition state, with the strongest non-covalent interactions, including multiple C–H⋯π, C–F⋯π contacts, weak Cu⋯H interaction, and dual C–H⋯O hydrogen bonds (one unusually strong and short-range, the other weaker). These interactions act synergistically to enhance its structural stability. In contrast, TS5(R,S) (Fig. 5 and Table S2) displays fewer and weaker interactions (C–H⋯π, limited C–H⋯O, O–H⋯O), resulting in lower stability. TS5(S,S) (Fig. 5 and Table S3) is even less stable, supported only by sparse and weak interactions (C–H⋯π, C–F⋯π, O–H⋯π, Cu⋯H, O–H⋯O), which contribute minimally to its stabilization.

Table 2 Summary of distances (in Å), type of interaction, electron densities (ρ_bcp × 10⁻²), Lagrangian kinetic energies (G_b × 10⁻²), potential energy densities (V_b × 10⁻²), energy densities (H_b × 10⁻³), and Laplacian of electron densities (∇₂ρ × 10⁻¹) at the bond critical points (BCPs) along the bond paths in TS5(S,R)

BCP index	Distance	Type of interaction	ρ bcp	G b	V b	H b	∇2ρ
a	2.42	C–H⋯π	0.473	0.341	−0.240	1.002	0.176
b	2.57	C–H⋯π	1.032	0.716	−0.515	2.009	0.367
c	2.32	C–H⋯π	0.546	0.402	−0.309	0.923	0.198
d	2.51	C–H⋯π	0.459	0.339	−0.217	1.220	0.185
e	2.72	C–H⋯π	0.745	0.545	−0.377	1.684	0.285
f	3.15	C–F⋯π	0.604	0.454	−0.346	1.083	0.225
g	3.18	C–F⋯π	0.232	0.327	−0.215	1.115	0.175
h	3.42	C–F⋯π	0.337	0.257	−0.183	0.742	0.133
i	2.81	C–F⋯π	0.735	0.852	−0.743	1.088	0.384
j	2.81	C–F⋯π	0.719	0.844	−0.731	1.126	0.383
k	2.51	Cu⋯H	0.128	0.757	−0.859	−0.102	0.262
l	1.59	C–H⋯O	5.937	4.804	−5.409	−6.057	1.679
m	2.67	C–H⋯O	0.667	0.520	−0.445	0.750	0.238

---

The critical role of water in the rate-determining step

To gain deeper insight into the rate-determining step (RDS), we performed EDA-NOCV analysis of the key transition states TS6A, TS6B, and TS6C. As shown in the comparison in Table 3, TS6A exhibits the most stabilizing interactions, with a total intrinsic interaction energy (ΔE_int) of −96.0 kcal mol⁻¹, significantly more favorable than those of TS6B (−85.2 kcal mol⁻¹) and TS6C (−44.5 kcal mol⁻¹). This substantial stabilization is attributed to the presence of a two-water cluster in TS6A, which enhances interfragment interactions and stabilizes the transition state. Further decomposition of ΔE_int reveals that electrostatic interaction (ΔE_elstat = −108.2 kcal mol⁻¹, accounting for 37.7% of total attractive interactions) and orbital interaction (ΔE_orb = −167.7 kcal mol⁻¹, contributing 58.5%) are the primary contributors to the overall stabilization, indicating strong coulombic attraction and efficient orbital overlap in TS6A.

Table 3 EDA-NOCV results of TS6A, TS6B and TS6C at the BP86+(D3BJ)/TZ2P level of theory. Energy values are given in kcal mol⁻¹

Orbital interaction	TS6A	TS6B	TS6C
a The values in parentheses give the percentage contribution to the total attractive interactions ΔEelstat + ΔEorb + ΔEdisp.
Fragments	H–2H2O(S) + rest(S)	H–1H2O(S) + rest(S)	H(S) + rest(S)
ΔEint	−96.0	−85.2	−44.5
ΔEPauli	190.9	210.6	193.4
ΔEdisp^a	−10.9 (3.8%)	−8.1 (2.7%)	−1.8 (0.8%)
ΔEelstat^a	−108.2 (37.7%)	−115.3 (39.0%)	−88.4 (37.1%)
ΔEorb^a	−167.7 (58.5%)	−172.4 (58.3%)	−147.8 (62.1%)

---

Although TS6B exhibits stronger electrostatic (−115.3 kcal mol⁻¹) and orbital (−172.4 kcal mol⁻¹) contributions compared to TS6A, it also suffers from significantly greater Pauli repulsion (ΔE_Pauli = 210.6 kcal mol⁻¹), which offsets these attractive interactions and results in a less favorable ΔE_int. In contrast, the unassisted pathway TS6C exhibits the weakest intrinsic interaction energy (−44.5 kcal mol⁻¹), indicating that the absence of a water-mediated hydrogen-bonding network significantly diminishes interfragment stabilization and results in a higher barrier. While dispersion interactions (ΔE_disp) contribute weakly in all three systems (0.8%–3.8%), TS6A benefits from a relatively strong dispersion stabilization (−10.9 kcal mol⁻¹), which is much stronger than in TS6B and TS6C and aligns with IGMH analysis (Fig. 6). The EDA-NOCV results clearly demonstrate that the water molecules in TS6A are critical for lowering the reaction barrier, as they enhance electrostatic, orbital, and dispersion interactions, thereby stabilizing the rate-determining transition state.

Fig. 6 IGMH analysis of TS6A, TS6B and TS6C.

Conclusions

In summary, we have performed a comprehensive mechanistic investigation on the NHC/Cu-catalyzed asymmetric synthesis of the spirooxindole δ-lactone product 3a. The catalytic cycle proceeds through four distinct stages. Stage I involves the nucleophilic addition of the NHC catalyst to enal 1a, forming the homoenolate intermediate IM3 bearing a nucleophilic carbon center. In parallel, Stage II features [Cu]-catalyzed decarboxylation of ethynyl cyclic carbonate 2a, affording the electrophilic copper–alkynyl cationic intermediate IM7. In Stage III, these two reactive species, IM3 and IM7, undergo stereoselective C–C bond formation to furnish the (S,R)-configured enol intermediate IM9. This is followed by a two-water-assisted enol–keto tautomerization viaTS6A to yield the ketone intermediate IM10. Stage IV begins with NEt₃-mediated deprotonation, followed by intramolecular cyclization and proton transfer, generating the π-complex IM15. Subsequent ligand exchange with another substrate 2a releases the final product 3a and regenerates the catalytically active Cu–alkynyl species for the next catalytic cycle. Notably, the C–C bond formation process in Stage III is the enantio- and diastereoselectivity determining step. The two-water-assisted enol–keto tautomerization in Stage III, with an activation barrier of 24.9 kcal mol⁻¹, was identified as the rate-determining step (RDS). The overall transformation is highly exergonic, thus providing a strong thermodynamic driving force for the reaction.

To elucidate the origins of the enantio- and diastereoselectivity, we conducted both quantitative and qualitative analyses on the four key transition states, TS5(S,R), TS5(R,S), TS5(S,S), and TS5(R,R). Among them, TS5(S,R) emerged as the most favorable, exhibiting the lowest activation free energy, moderate distortion energy (ΔE_strain = 32.1 kcal mol⁻¹), and the most stabilizing intrinsic interaction energy (ΔE_int = −75.3 kcal mol⁻¹). Its enhanced stability is further supported by multiple non-covalent interactions, including strong C–H⋯π, C–H⋯O hydrogen bonds, C–F⋯π contacts, and Cu⋯H interactions, which collectively contribute to its enhanced stability. These factors together account for the kinetic and thermodynamic preference for the (S,R)-configured spirooxindole product 3a, which agrees very well with the experimentally observed high diastereo- and enantioselectivity. Additionally, we highlight the essential role of water molecules in lowering the energy barrier of the rate-determining step (RDS). By enhancing electrostatic, orbital, and dispersion interactions, water molecules significantly stabilize the transition state and accelerate the reaction. Overall, this study offers a robust theoretical framework for understanding cooperative NHC/metal catalysis. The mechanistic insights not only deepen the fundamental understanding of stereocontrol in complicated annulation reactions but also provide valuable guidance for designing next-generation asymmetric catalytic systems based on NHC/Cu cooperative catalysis.

Author contributions

All authors have given approval to the final version of the manuscript. Q. M., S. Z., and Q. Z. performed the DFT calculations. X. Y. and L. Q. discussed the project and gave suggestions. L. Z. conceived and supervised the project. L. Z. wrote the original draft, which was reviewed and edited with input from all authors. All authors discussed the results and commented on the paper.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

The data supporting the findings of this study are available within the article and its supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5qo01235j.

Acknowledgements

We acknowledge the financial support from the National Natural Science Foundation of China (No. 22373050), the Natural Science Foundation of Jiangsu Province (BK20250032), the State Key Laboratory of Materials-Oriented Chemical Engineering (No. SKL-MCE-23A06), the Postgraduate Research & Practice Innovation Program of Jiangsu Province (No. SJCX25-0582), the Cultivation Program for The Excellent Doctoral Dissertation of Nanjing Tech University (2024-10), and Nanjing Tech University (No. 39837123, 39837132). We also appreciate the high performance center of Nanjing Tech University for supporting with the computational resources.

References

1. B. Yu, D. Q. Yu and H. M. Liu, Spirooxindoles: Promising scaffolds for anticancer agents, Eur. J. Med. Chem., 2015, 97, 673–698.
2. S. S. Panda, A. S. Girgis, M. N. Aziz and M. S. Bekheit, Spirooxindole: A Versatile Biologically Active Heterocyclic Scaffold, Molecules, 2023, 28, 618.
3. N. Ye, H. Chen, E. A. Wold, P. Y. Shi and J. Zhou, Therapeutic Potential of Spirooxindoles as Antiviral Agents, ACS Infect. Dis., 2016, 2, 382–392.
4. J. S. Wu, X. Zhang, Y. L. Zhang and J. W. Xie, Synthesis and antifungal activities of novel polyheterocyclic spirooxindole derivatives, Org. Biomol. Chem., 2015, 13, 4967–4975.
5. S. S. Babu, N. S. Devi, R. M. Devi, N. D. Kannan and S. Sivakumar, Synthesis and Characterization of Spirooxindole β-Ketothiolester and their Fluorescence Imaging on PC3 Cells, ChemistrySelect, 2024, 9, e202303884.
6. C. V. Galliford and K. A. Scheidt, Pyrrolidinyl-Spirooxindole Natural Products as Inspirations for the Development of Potential Therapeutic Agents, Angew. Chem., Int. Ed., 2007, 46, 8748–8758.
7. T. Wu, Z. Pan and C. Xia, Construction of Spirooxindole Skeleton Through Intramolecular Dieckmann Cyclization, Nat. Prod. Bioprospect., 2017, 7, 275–281.
8. Y. Tang, C. Wang, H. Xie, Y. Xu, C. Wang, C. Du, Z. Wang and L. Wang, Green Synthesis of Spirooxindoles via Lipase-Catalyzed One-Pot Tandem Reaction in Aqueous Media, Catalysts, 2023, 13, 143.
9. Z. Cao, F. Hu, Y. Chu, J. Pei and X.-P. Hui, NHC-catalyzed stereoselective synthesis of spirooxindole lactones from in situ generated oxindole-embedded o-quinone methides and aldehydes, Org. Chem. Front., 2024, 11, 3326–3332.
10. X. Wang and W. Ren, Catalytic Asymmetric [4 + 2] Cyclization of Hydroxyphenyl Indolinone with Azlactone to Construct Spirooxindole δ-Lactone, Chin. J. Chem., 2024, 42, 3362–3366.
11. J. Panda, B. P. Raiguru, S. Nayak, S. Mohapatra, S. Mohapatra, S. Rout, P. Mohanty, H. S. Biswal, D. Bhattacharya and C. R. Sahoo, [3 + 2] Cycloaddition mediated synthesis of spirooxindole pyrrolidine/piperidine fused chromene aldehyde derivatives: DFT studies, antibacterial evaluation and molecular docking investigation as DNA gyrase inhibitors, New J. Chem., 2024, 48, 13609–13630.
12. A. A. Alayyaf, M. Ali, M. A. Alwehaibi, M. K. Al-Muhanna, S. M. Soliman, M. Ríos-Gutiérrez, M. Haukka and A. Barakat, Utilizing MEDT analysis of [3 + 2] cycloaddition reaction: X-ray crystallography of spirooxindole linked with thiophene/furan heterocycles and triazole framework, BMC Chem., 2024, 18, 229.
13. X.-W. Qian, H. Lin, T. Hu, X. Yang and X.-W. Sun, Synthesis of enantioenriched spirocyclic oxindoles catalyzed by bifunctional thiourea, Chem. Commun., 2025, 61, 6348–6351.
14. X. Li, Z. Li, R. Zhang, Z. Zhou, Y. Zhang, Y. Xia, W. Yao and Z. Wang, Asymmetric Synthesis of Multicyclic Spirooxindole Derivatives Bearing Stereogenic Quaternary Carbon Atoms, Org. Lett., 2023, 25, 2642–2646.
15. Y. Pu, M. Wang, W. Tian, X. Ge, D. Zhu, C. Wang, Y. Zeng, F. Tao, Y. Deng and J. Lu, N-heterocyclic carbene catalyzed [2 + 3] annulation reaction for the synthesis of trifluoroethyl 3,2′-spirooxindole γ-lactam, RSC Adv., 2024, 14, 18453–18458.
16. D. Enders, O. Niemeier and A. Henseler, Organocatalysis by N-Heterocyclic Carbenes, Chem. Rev., 2007, 107, 5606–5655.
17. A. J. Arduengo III, R. L. Harlow and M. Kline, A stable crystalline carbene, J. Am. Chem. Soc., 1991, 113, 361–363.
18. C. Guo, M. Fleige, D. Janssen-Müller, C. G. Daniliuc and F. Glorius, Switchable selectivity in an NHC-catalysed dearomatizing annulation reaction, Nat. Chem., 2015, 7, 842–847.
19. H.-W. Wanzlick and H.-J. Schönherr, Direct Synthesis of a Mercury Salt-Carbene Complex, Angew. Chem., Int. Ed. Engl., 1968, 7, 141–142.
20. R. Breslow, On the Mechanism of Thiamine Action. IV.1 Evidence from Studies on Model Systems, J. Am. Chem. Soc., 1958, 80, 3719–3726.
21. C. Costabile, S. Pragliola and F. Grisi, C2-Symmetric N-Heterocyclic Carbenes in Asymmetric Transition-Metal Catalysis, Symmetry, 2022, 14, 1615.
22. J. Han, Y.-M. He, Y. Pan, F. Li, D. Li, W. Hao and Q.-H. Fan, Tunable Unsymmetrical Chiral N-Heterocyclic Carbene Ligands for Highly Diastereo- and Enantioselective Copper-Catalyzed Tandem Borylative Cyclization: Ligand Controlled Diastereoselectivity Reversal, CCS Chem., 2023, 5, 2088–2100.
23. X. Chen, S. Xia, X. Hu, G. Zhong and L. Yang, Chiral NHC-Catalyzed [4 + 2] Cycloadditions: Highly Diastereo/Enantioselective Access to Spiro[cyclohex-4-ene-1,3′-indoles] and DFT Calculations, Chem. – Eur. J., 2023, 29, e202203818.
24. C. Guo, M. Fleige, D. Janssen-Müller, C. G. Daniliuc and F. Glorius, Cooperative N-Heterocyclic Carbene/Palladium-Catalyzed Enantioselective Umpolung Annulations, J. Am. Chem. Soc., 2016, 138, 7840–7843.
25. S. Singha, T. Patra, C. G. Daniliuc and F. Glorius, Highly Enantioselective [5 + 2] Annulations through Cooperative N-Heterocyclic Carbene (NHC) Organocatalysis and Palladium Catalysis, J. Am. Chem. Soc., 2018, 140, 3551–3554.
26. J. Huo, K. Zhong, Y. Xue, M. Lyu, Y. Ping, Z. Liu, Y. Lan and J. Wang, Palladium-Catalyzed Enantioselective Carbene Insertion into Carbon–Silicon Bonds of Silacyclobutanes, J. Am. Chem. Soc., 2021, 143, 12968–12973.
27. C. Guo, D. Janssen-Müller, M. Fleige, A. Lerchen, C. G. Daniliuc and F. Glorius, Mechanistic Studies on a Cooperative NHC Organocatalysis/Palladium Catalysis System: Uncovering Significant Lessons for Mixed Chiral Pd(NHC)(PR3) Catalyst Design, J. Am. Chem. Soc., 2017, 139, 4443–4451.
28. S. Singha, E. Serrano, S. Mondal, C. G. Daniliuc and F. Glorius, Diastereodivergent synthesis of enantioenriched α,β-disubstituted γ-butyrolactones via cooperative N-heterocyclic carbene and Ir catalysis, Nat. Catal., 2020, 3, 48–54.
29. Z.-J. Zhang, L. Zhang, R.-L. Geng, J. Song, X.-H. Chen and L.-Z. Gong, N-Heterocyclic Carbene/Copper Cooperative Catalysis for the Asymmetric Synthesis of Spirooxindoles, Angew. Chem., Int. Ed., 2019, 58, 12190–12194.
30. J. Gao, J. Zhang, S. Fang, J. Feng, T. Lu and D. Du, Synergistic N-Heterocyclic Carbene/Palladium-Catalyzed [3 + 2] Annulation of Vinyl Enolates with 1-Tosyl-2-vinylaziridine, Org. Lett., 2020, 22, 7725–7729.
31. Y.-H. Wen, Z.-J. Zhang, S. Li, J. Song and L.-Z. Gong, Stereodivergent propargylic alkylation of enals via cooperative NHC and copper catalysis, Nat. Commun., 2022, 13, 1344.
32. Z.-J. Zhang, Y.-H. Wen, J. Song and L.-Z. Gong, Kinetic Resolution of Aziridines Enabled by N-Heterocyclic Carbene/Copper Cooperative Catalysis: Carbene Dose-Controlled Chemo-Switchability, Angew. Chem., Int. Ed., 2021, 60, 3268–3276.
33. H. Liu, Y.-F. Han, Z.-H. Gao, C.-L. Zhang, C. Wang and S. Ye, Cooperative N-Heterocyclic Carbene/Nickel-Catalyzed Hydroacylation of 1,3-Dienes with Aldehydes in Water, ACS Catal., 2022, 12, 1657–1663.
34. Y.-F. Han, Y. Huang, H. Liu, Z.-H. Gao, C.-L. Zhang and S. Ye, Photoredox cooperative N-heterocyclic carbene/palladium-catalysed alkylacylation of alkenes, Nat. Commun., 2022, 13, 5754.
35. T. Fan, J. Song and L.-Z. Gong, Asymmetric Redox Allylic Alkylation to Access 3,3′-Disubstituted Oxindoles Enabled by Ni/NHC Cooperative Catalysis, Angew. Chem., Int. Ed., 2022, 61, e202201678.
36. S. Tang, H. Xu, Y. Dang and S. Yu, Photoexcited Copper-Catalyzed Enantioselective Allylic C(sp3)–H Acyloxylation of Acyclic Internal Alkenes, J. Am. Chem. Soc., 2024, 146, 27196–27203.
37. Y. Hu, L. Hu, H. Gao, X. Lv, Y.-B. Wu and G. Lu, Computational study of Cu-catalyzed 1,2-hydrocarboxylation of 1,3-dienes with CO2: Pauli repulsion-controlled regioselectivity of Cu–Bpin additions, Org. Chem. Front., 2022, 9, 2240–2248.
38. L. Hu, H. Gao, Y. Hu, X. Lv, Y.-B. Wu and G. Lu, Origins of regio- and stereoselectivity in Cu-catalyzed alkyne difunctionalization with CO2 and organoboranes, Org. Chem. Front., 2022, 9, 1033–1039.
39. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, Gaussian 16, Revision C.01, Gaussian, Inc., Wallingford, CT, 2016.
40. C. Y. Legault, CYLview20, 2020, https://www.cylview.org.
41. A. D. Becke, Density-functional exchange-energy approximation with correct asymptotic behavior, Phys. Rev. A, 1988, 38, 3098–3100.
42. J. P. Perdew, Density-functional approximation for the correlation energy of the inhomogeneous electron gas, Phys. Rev. B:Condens. Matter Mater. Phys., 1986, 33, 8822–8824.
43. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu, J. Chem. Phys., 2010, 132, 154104.
44. F. Weigend and R. Ahlrichs, Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305.
45. K. Fukui, The path of chemical reactions - the IRC approach, Acc. Chem. Res., 1981, 14, 363–368.
46. A. V. Marenich, C. J. Cramer and D. G. Truhlar, Universal Solvation Model Based on Solute Electron Density and on a Continuum Model of the Solvent Defined by the Bulk Dielectric Constant and Atomic Surface Tensions, J. Phys. Chem. B, 2009, 113, 6378–6396.
47. L. E. English, T. M. H. Downie, C. L. Lyall, M. F. Mahon, C. L. McMullin, S. E. Neale and D. J. Liptrot, Selective hydroboration of electron-rich isocyanates by an NHC-copper(I) alkoxide, Chem. Commun., 2023, 59, 1074–1077.
48. T. Weymuth, E. P. Couzijn, P. Chen and M. Reiher, New benchmark set of transition-metal coordination reactions for the assessment of density functionals, J. Chem. Theory Comput., 2014, 10, 3092–3103.
49. Q. Wei, H. He, J. Zhang and C. Zhu, Theoretical study on cooperation catalysis of chiral guanidine/Cu(I) in asymmetric azide–alkyne cycloaddition/[2 + 2] cycloaddition, J. Org. Chem., 2023, 88, 9973–9986.
50. O. E. Okoromoba, T. A. Chen, E. S. Jang, C. L. McMullin, T. R. Cundari and T. H. Warren, Copper-catalyzed C(sp³)–H α-acetylation: Generation of quaternary centers, Angew. Chem., 2025, 137, e202418692.
51. T. Lu and F. Chen, Multiwfn: A multifunctional wavefunction analyzer, J. Comput. Chem., 2012, 33, 580–592.
52. L. D. W. Schrödinger, The PyMOL Molecular Graphics System, 2020, https://www.pymol.org/pymol.
53. K. Morokuma, Molecular Orbital Studies of Hydrogen Bonds. III. C=O⋯H–O Hydrogen Bond in H₂CO⋯H₂O and H₂CO⋯2H₂O, J. Chem. Phys., 1971, 55, 1236–1244.
54. T. Ziegler and A. Rauk, On the calculation of bonding energies by the Hartree Fock Slater method, Theor. Chim. Acta, 1977, 46, 1–10.
55. J. C. Slater, Statistical Exchange-Correlation in the Self-Consistent Field, Adv. Quantum Chem., 1972, 6, 1–92.
56. G. Frenking, M. Hermann, D. M. Andrada and N. Holzmann, Donor-acceptor bonding in novel low-coordinated compounds of boron and group-14 atoms C-Sn, Chem. Soc. Rev., 2016, 45, 1129–1144.
57. G. Frenking and F. Matthias Bickelhaupt, The EDA Perspective of Chemical Bonding, in The Chemical Bond: Fundamental Aspects of Chemical Bonding, 2014, pp. 121–157.
58. G. Frenking, R. Tonner, S. Klein, N. Takagi, T. Shimizu, A. Krapp, K. K. Pandey and P. Parameswaran, New bonding modes of carbon and heavier group 14 atoms Si–Pb, Chem. Soc. Rev., 2014, 43, 5106–5139.
59. G. te Velde, F. M. Bickelhaupt, E. J. Baerends, C. F. Guerra, S. J. A. van Gisbergen, J. G. Snijders and T. Ziegler, Chemistry with ADF, J. Comput. Chem., 2001, 22, 931–967.
60. E. Van Lenthe and E. J. Baerends, Optimized Slater-type basis sets for the elements 1–118, J. Comput. Chem., 2003, 24, 1142–1156.
61. L. Zhao, M. Hermann, W. H. E. Schwarz and G. Frenking, The Lewis electron-pair bonding model: modern energy decomposition analysis, Nat. Rev. Chem., 2019, 3, 48–63.
62. L. Zhao, M. Hermann, N. Holzmann and G. Frenking, Dative bonding in main group compounds, Coord. Chem. Rev., 2017, 344, 163–204.

---