The control effects of different scaffolds in chiral phosphoric acids: a case study of enantioselective asymmetric arylation

Lihan Zhu , Hend Mohamed , Haiyan Yuan * and Jingping Zhang *
Faculty of Chemistry, Northeast Normal University, Changchun, Jilin 130024, P. R. China. E-mail: jpzhang@nenu.edu.cn; yuanhy034@nenu.edu.cn

Received 17th July 2019 , Accepted 10th October 2019

First published on 11th October 2019


Chiral phosphoric acid (CPA) catalysts with BINOL or SPINOL backbones have attracted immense attention in recent years. Despite a number of successful studies, the theoretical elucidation on the role of the CPA scaffold has rarely been considered, and its effects on enantioselectivity are still far from being understood. Structure inspection shows that axially chiral scaffolds induce a stereogenic central phosphate, and the phosphoric acid functional groups in the (R/S)-BINOL skeleton are identical to those in the (S/R)-SPINOL skeleton. Therefore, a hypothesis is proposed that the orientation of each phosphoric acid functional group regulated by different (R/S) BINOL- and SPINOL-derived scaffolds may control the signs of enantioselectivity (positive or negative ee value) by changing the combination modes of the substrates. In other words, the sign of enantioselectivity (positive or negative ee value) can be tuned with an (R or S) BINOL-derived backbone or an (S or R) SPINOL-derived backbone, respectively. Thus, we elaborate an in-depth mechanistic case study of newly reported CPA-catalyzed enantioselective asymmetric arylation reactions (a–c) catalyzed by CPAs with two different backbones. We found that although the origin of the high enantioselectivity of these three case reactions a–c can be ascribed to favorable C–H⋯O interactions, catalyst and substrate distortion interactions, and electrostatic interactions in the preferential TSs leading to the major products, the signs of enantioselectivity still follow our hypothesis. Our general applicability finding encourages the further development of more effective CPA catalysts and can be used to guide the strategic choice of CPA skeletons.


1. Introduction

In recent years, chiral phosphoric acid (CPA) catalysis has received wide attention;1 two different scaffolds, BINOL (1, Scheme 1) and SPINOL (2, Scheme 1), have evolved as powerful tools for various enantioselective transformations.2 In general, BINOL-based CPAs have been demonstrated to be capable of imparting exceptionally high enantioselectivities in dearomatization,3 arylation,4 the Passerini reaction,5 and Ugi reactions.1a,6 CPAs containing SPINOL scaffolds have also been shown by various research groups to afford products with high selectivity.7 It is noteworthy that CPAs generally catalyze reactions through a “bifunctional” activation mode2d,8 in which the Brønsted basic and acidic sites of the CPA simultaneously activate both nucleophiles and electrophiles.2c Currently, significant progress in the understanding of CPA-catalyzed reactions has been achieved, particularly in terms of the origins of stereoselectivity.9 In 2013, Toast's group10 presented theoretical studies of chiral phosphate anion catalysis, demonstrating that highly stereoselective catalysts can be achieved through favorable noncovalent interactions between the bulky aryl substituents at the 3,3′-positions of the catalyst and substrates. Sunoj et al.1f also identified that the origin of reversal in the sense of seteroinduction can be ascribed to changes in the pattern of noncovalent interactions in the stereocontrolling transition states. In 2016, Zimmerman and Nagorny11 showed that the enantiocontrol of asymmetric spiroketalization reactions arises in large part from steric interactions. In particular, Seguin and Wheeler12 found that the enantioselectivity of asymmetric ring-opening reactions of meso-epoxides stems from favorable electrostatic interactions with phosphoric acid functionality. Despite the immense popularity of theoretical work on the stereoselectivities of individual CPA-catalyzed reactions,1h,8c,13 the roles of different (R/S)-CPA scaffolds and their governmental effects on enantioselectivity remain an unmet challenge. Thus, a more complete understanding of how BINOL- and SPINOL-derived backbones affect the enantioselectivities of asymmetric organocatalysts is computationally challenging and must be assessed. The stereogenic conformation of the central phosphate is induced by the orientation of different backbones, ((R/S)-BINOL or (R/S)-SPINOL); thus, the orientations of the phosphoric acid functional groups connected with the stereogenic phosphate are reversed when the enantioselectivity senses (R or S) of the BINOL- and SPINOL-derived backbones are the same (Scheme 1). In other words, the orientations of the oxygen atoms around the central P atom in (R/S)-BINOL are identical to those in the (S/R)-SPINOL skeleton. Thus, we surmise that the orientations of phosphoric acid functional groups regulated by different (R/S) BINOL- and SPINOL-derived backbones may control the signs of enantioselectivity (positive or negative ee values) by changing the combination modes of the substrates.
image file: c9cy01420a-s1.tif
Scheme 1 (R/S) BINOL- and (R/S) SPINOL-derived CPAs 1 and 2.

In order to identify our hypothesis, a series of efficient enantioselective organocatalytic asymmetric arylation reactions enabled by azo groups catalyzed by CPA catalysts with two different backbones proposed by Tan's group14 are used as case reactions. It is well known that the N[double bond, length as m-dash]N bond of the azo group is used as an excellent directing group in various transition-metal catalyzed aryl C–H bond activation reactions, and a high diversity of ortho-substituted azo compounds can be reached.15 To date, metal catalysts such as palladium,16 rhodium,17 ruthenium,18 iridium,19 and cobalt complexes20 have been successfully employed. However, organocatalytic arylation by azo groups remained a major challenge until a powerful strategy to obtain a wide range of axially chiral arylindoles with good yields and excellent enantioselectivities was reported recently (Scheme 2). The azobenzene derivative (1a) reacts effectively with 2,3-dimethyl indole (2a), 2-tert-butyl-indole (2b), and 2-methylindole (2c), then undergoes a cascade enantioselective formal nucleophilic aromatic substitution–cyclization process catalyzed by (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3 to generate the axially chiral arylindoles. Meanwhile, according to the conventional stereochemical models of CPA reactions, a key general “bifunctional” activation model of selectivity was proposed (Scheme 2).21 In this model, both the hydroxy and phosphoryl oxygen groups of the CPA catalyst activate the substrates via N–H⋯O and O–H⋯N hydrogen bonding interactions (HBs). In the case reactions, the (R)-BINOL and (R)-SPINOL backbones of the CPAs have different influences on the signs of enantioselectivity, with positive and negative ee values. The reactions b and c catalyzed by BINOL-derived (R)-CPA-2 and SPINOL-derived (R)-CPA-3 will generate the target products with high positive ee values (97% and 99%). Alternatively, a reversal of the sense of the product stereoinduction was noted when reactions b and c were respectively catalyzed by (R)-CPA-3 (ee = −90%) and (R)-CPA-2 (ee = −82%). This mechanistic case study provides a very clear understanding of our hypothesis on the governmental effects of the different scaffolds, and we also clarify the preferred activation mode of CPA catalysis and the origin of stereoselectivity by calculating the CPA-catalyzed asymmetric arylation reactions of 1a with 2a, 2b, and 2c. Herein, we report our journey toward the novel discovery of the control of the signs of enantioselectivity by the scaffold configuration and the results of our comprehensive mechanistic case study.


image file: c9cy01420a-s2.tif
Scheme 2 Asymmetric arylation of indole reactions and the possible activation TS mode proposed by Tan and co-workers14 catalyzed by (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3.

2. Results and discussion

2.1. A hypothesis of CPA catalyst scaffold effects

By exploring the structures of the most commonly used (R)-BINOL and (R)-SPINOL scaffolds, we found that when the right sides of the two catalysts are placed on a horizontal plane (black arrow, Fig. 1), the left skeleton and the 3,3′-substituents are facing down in the (R)-BINOL backbone and up in the (R)-SPINOL backbone (red arrow, Fig. 1). Also, eclipsed and staggered conformations of the P[double bond, length as m-dash]O and P–OH groups are observed in the (R)-BINOL and (R)-SPINOL backbones, respectively (the O–P–O–H dihedral angles are 4.3° and 67.4°, respectively, Fig. 1). Thus, different orientations (in-plane or out-of-plane) of the two oxygen atoms directly connected with the carbon atoms of the framework for the (R)-BINOL and (R)-SPINOL skeletons can induce stereogenic phosphates, and the associated phosphoric acid functional groups are distinct. Therefore, it is assumed that each orientation of the phosphoric acid functional groups controls the signs of enantioselectivity (positive or negative ee values) by changing the combination modes of the substrates. Additionally, the chirality information transfers from the axially chiral backbones to the central or axial chirality products via the stereogenic phosphate. In this context, we firstly investigated the activation modes and origins of enantioselectivity in the case ractions a–c. Then, reactions b and c were selected as model reactions to verify the role of the CPA scaffold and its governmental effects on the signs of enantioselectivity. Finally, we found that related representative reactions remarkably confirmed our general conclusion.
image file: c9cy01420a-f1.tif
Fig. 1 Different alignments of 3,3′-substituents and phosphate functional groups for BINOL and SPINOL backbones.

2.2. Activation mode

In order to understand the origins of the stereoselectivity of these three reactions, different activation modes based on the interplay of numerous noncovalent interactions between the CPA catalysts and substrates were considered.1g,2d,10,12,22 First, we calculated all possible conformations of substrate 1a, and the optimized structures are shown in Fig. S1. The calculations revealed that conformer 1a is the most stable among all ten cases in the three solvent phases; thus, it was further considered. Then, two possible activation modes (M1 and M2 in Fig. 2) were identified according to the positions of the biphenyl and indole rings of the two substrates (same side or opposite sides). In the M1 mode (M1-TS1aR, M1-TS1aS, M1-TS1bR, M1-TS1bS, M1-TS1cR, and M1-TS1cS) for these three reactions, the biphenyl and indole rings are oriented on the same side and may achieve moderately good π⋯π overlap; meanwhile, in the M2 mode, these two rings in each TS are on opposite sides (M2-TS1aR, M2-TS1aS, M2-TS1bR, M2-TS1bS, M2-TS1cR, and M2-TS1cS). These TSs were stabilized by two strong N–H⋯O HBs between the catalyst and the substrates; the vital difference between M1 and M2 is the orientation of the substrates. Surprisingly, these computational results indicated that the relative Gibbs free energies of TSs in M1 mode are lower than the corresponding energies in M2 mode by 1.3 to 8.2 kcal mol−1 in reactions a–c; thus, we applied distortion/interaction analysis23 to explain why mode M1 is preferred over M2. The free energy difference between the competing modes (ΔΔE) can be written as ΔΔE = ΔΔEdist(sub) + ΔΔEdist(cat) + ΔΔEint, wherein ΔΔEdist(sub) and ΔΔEdist(cat) are the differences in energy required to distort the substrates (ΔΔEsub) and catalyst (ΔΔEcat) from the ground-state geometry to the transition-state geometries. ΔΔEint corresponds to the energy difference between the catalyst plus the substrates and the complex in the TS structure. As shown in Table 1, the energy separation between M1 and M2 mode in TS1aS and TS1bS mainly arises from differences in the distortion of the catalysts (ΔΔEcat = 3.6 and 5.2 kcal mol−1, Table 1). Both ΔΔEcat and ΔΔEint are positive, which indicates that the two components contribute synergistically to the overall energy difference in these two TSs. Inspection of the structures in TS1aS and TS1bS reveals that the greater distortion of catalysts (R)-CPA-1 and (R)-CPA-2 in M2 mode can be ascribed almost entirely to the larger proton transfer degrees and angle deviations, as observed in Fig. S2.
image file: c9cy01420a-f2.tif
Fig. 2 Optimized structures of the stereocontrolling transition state structures of activation modes M1 and M2 for reactions a, b, and c catalyzed by (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3 to generate major and minor products. Relative Gibbs free energies are provided in kcal mol−1. Distances are provided in Å. For clarity, the catalyst backbones (BINOL and SPINOL) are depicted as simple curves.
Table 1 Predicted solvent-phase energy barrier height differences (ΔΔE, in kcal mol−1) between M1 and M2 modes for reactions a–c catalyzed by (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3; the decompositions of ΔΔE into ΔΔEdist(cat), ΔΔEdist(sub), and ΔΔEint
TS ΔΔEdist(total) ΔΔEdist(cat) ΔΔEdist(sub) ΔΔEint ΔΔE
TS1aR −4.3 −3.6 −0.7 5.0 0.7
TS1aS 2.7 3.6 −0.9 0.5 3.2
TS1bR −0.9 −0.5 −0.4 3.7 2.8
TS1bS 4.1 5.2 −0.9 0.6 4.7
TS1cR −1.7 −0.7 −1.0 3.1 1.4
TS1cS −1.9 −1.3 −0.6 7.8 5.9


For the other four cases, i.e., TS1aR, TS1bR, TS1cR, and TS1cS, the net effect of the distortions (ΔΔEdist(total) = −4.3, −0.9, −1.7, and −1.9 kcal mol−1) is to decrease the energy differences between the M1 and M2 modes. However, these effects are overshadowed by the 5.0, 3.7, 3.1, and 7.8 kcal mol−1 differences in ΔΔEint, which lead to total energy differences of 0.7, 2.8, 1.4, and 5.9 kcal mol−1 for TS1aR, TS1bR, TS1cR, and TS1cS, respectively. These results suggest that mode M1 is more favorable than M2. NCI analysis was conducted to pinpoint the differences in the interaction energies (ΔΔEint) between the substrates and catalyst (Fig. S3). The strong N–H⋯O and O–H⋯N interactions (in blue) between the phosphate oxygen of the CPA catalyst and the azo and imino groups are common to all eight TSs. However, it is clear that the number of C–H⋯π, C–H⋯O, C–H⋯F, and π-stacking interactions in M1 mode is larger than that in M2. Thus, the M1 modes in TS1aR, TS1bR, TS1cR, and TS1cS are much more stabilized by noncovalent interactions than the M2 modes.

Moreover, in the (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3-catalyzed cases of M1, the energetically most preferred transition states M1-TS1aR, M1-TS1bR, and M1-TS1cR were found to be 1.6, 2.9, and 3.1 kcal mol−1 lower in energy than the corresponding diastereomeric M1-TS1aS, M1-TS1bS, and M1-TS1cS transition states. The ee values predicted on the basis of these values of ΔΔG are 92%, 99%, and 99%, respectively. Therefore, the theoretical ee values are highly consistent with the experimental stereoselectivities, as confirmed in Table 2.14

Table 2 Experimental and theoretical ee values and corresponding relative free energies (kcal mol−1) for catalysts (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3a
Experiment Theory
Catalyst ee ΔΔG ΔΔG ee
a All reactions are performed at 298.15 K except for that with catalyst (R)-CPA-1 (253.15 K).
(R)-CPA-1 86% 1.3 1.6 92%
(R)-CPA-2 97% 2.5 2.9 99%
(R)-CPA-3 99% 3.0 3.1 99%


2.3. Origin of the enantioselectivity

To determine the origin of the selectivity between the stereocontrolling TSmajor (TS1aR, TS1bR, and TS1cR) and TSminor (TS1aS, TS1bS, and TS1cS) in reactions a, b, and c catalyzed by (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3, distortion/interaction analysis was considered to unravel the origin of the stereoinduction for these three reactions (Table 3). Clearly, the contributions to ΔΔE show that the interaction energies (ΔΔEint = 3.5 and 5.2 kcal mol−1, reactions a and c) and distortions of the substrates and catalyst (ΔΔEdist = 6.9 kcal mol−1, reaction b) strongly favor TSmajor. There are numerous noncovalent interactions that control the enantioselectivity in TS1aR/TS1aS (reaction a), TS1bR/TS1bS (reaction b), and TS1cR/TS1cS (reaction c); thus, we further employed Wheeler's strategy22f,24 (truncated model systems A–C, Table 3 and Fig. 3) to understand the contributions of individual noncovalent interactions to ΔΔEint. The interaction energy was approximately decomposed into contributions from the interactions of the substrates with three different fragments of the catalyst, i.e., two aryl groups and the phosphoric acid “core” (phosphoric acid functionality). We considered the effects of noncovalent interactions of the substrates with the upper and lower aryl groups in models A and B. ΔΔEint(C) is defined as the interaction energy between the protonated substrate and the catalyst phosphate oxygens in the TS structure.
Table 3 Solvent-phase ΔΔE, ΔΔEcat, ΔΔEsub, and ΔΔEint values for reactions a, b, and c and the differences in the interaction energies for truncated model systems A–C in kcal mol−1
Reaction ΔΔE ΔΔEdist ΔΔEcat ΔΔEsub ΔΔEint ΔΔEint(A) ΔΔEint(B) ΔΔEint(C)
a 2.0 −1.5 −0.8 −0.5 3.5 1.3 0.5 1.7
b 2.9 6.9 4.8 2.1 −4.0 0.8 0.7 −5.5
c 3.0 −2.2 −0.8 −1.4 5.2 0.5 1.4 3.3



image file: c9cy01420a-f3.tif
Fig. 3 Representative TS structure (TS1aR) and the corresponding truncated models (A–C) used to isolate the interactions of the substrates with 3,3′-aryl groups (A and B) and the phosphoric acid “core” (C).

For reaction a, models A–C confirm that the interaction energy difference (ΔΔEint = 3.5 kcal mol−1) mainly arises from the favorable HBs between the substrates and the “core” of (R)-CPA-1 (ΔΔEint(C) = 1.7 kcal mol−1) in TS1aR. Further examination of these two structures reveals that the triple C–H⋯O interactions in TS1aR are stronger than those in TS1aS, which is demonstrated by the HB distances (2.39 Å, 2.39 Å, and 2.42 Å vs. 2.37 Å, Fig. 4). Thus, these C–H⋯O interactions preferentially stabilize TS1aR significantly, enhancing the enantioselectivity. This effect is compensated slightly by the catalyst and substrate distortion energy (ΔΔEdist = −1.5 kcal mol−1, which favor TS1aS), leading to the appreciable level of enantioselectivity for reaction a.


image file: c9cy01420a-f4.tif
Fig. 4 Transition states for the stereoselective nucleophilic attack of azobenzene derivative 1a (highlighted in pink), 2a (highlighted in green), 2b (highlighted in yellow), and 2c (highlighted in blue) catalyzed by (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3. Distances are given in Å.

For reaction b, the contributions to ΔΔE show that the distortions of the substrates and (R)-CPA-2 (ΔΔEdist = 6.9 kcal mol−1) strongly favor TS1bR. The differences in the deviation of the catalyst framework in TS1bR and TS1bS were examined, as shown in Fig. 4. In order to bind the substrates, the angular deviations of (R)-CPA-2 in TS1bS are 2.75° (O1–P–O2), 2.57° (P–O1–H1), and 1.04° (C4′–C5′–C6′) larger than those in TS1bR. These changes indicate a large deviation for the catalyst in TS1bS. Also, the distortion of substrates mainly arises from the interactions with the upper and lower aryl groups of (R)-CPA-2 (ΔΔEint(A) = 0.8 kcal mol−1 and ΔΔEint(B) = 0.7 kcal mol−1). This can be attributed to the favorable C–H⋯F interactions between the substrates and the upper (2.37 Å) or lower (2.34 Å and 2.75 Å) aryl groups in TS1bR compared to the closest interactions in TS1bS (2.41 Å and 2.37 Å). The truncated model C indicates that the largely negative ΔΔEint(C) value (−5.5 kcal mol−1) is primarily due to the stronger HBs in TS1bS (1.72 Å and 1.73 Å) than in TS1bR (1.83 Å and 1.88 Å). This is a consequence of the shorter distances between the substrates and functional groups of (R)-CPA-2 in TS1bS in comparison to those in TS1bR. Thus, the enantioselectivity in reaction b arises from the more favorable deformations of (R)-CPA-2 and the substrates in TS1bR in comparison to those in TS1bS.

The energy separation between TS1cR and TS1cS in reaction c arises primarily from the differences in the noncovalent interactions (ΔΔEint = 5.2 kcal mol−1). Analyses of the truncated models show that the interactions of the substrates with the lower phenanthryl group (ΔΔEint(B) = 1.4 kcal mol−1) as well as the “core” of (R)-CPA-3 (ΔΔEint(C) = 3.3 kcal mol−1) strongly favor TS1cR. The similar C–H⋯π interactions between the substrates and the lower phenanthryl group are approximately 1.4 kcal mol−1 more favorable in TS1cR than in TS1cS (2.78 Å and 2.96 Å vs. 2.86 Å and 3.07 Å, Fig. 4), partially contributing to the high degree of enantioselectivity. Furthermore, examination of these two structures reveals that the double N–H⋯O interactions in TS1cR are stronger than those in TS1cS, which is demonstrated by the HB distances (1.77 Å vs. 1.94 Å and 1.62 Å vs. 1.92 Å). Also, the charges of the H1 and H1c atoms in TS1cR (+0.432 and +0.444) are more positive than those in TS1cS (+0.413 and +0.442) (Fig. S4). We applied Espinosa's equation25 and electrostatic potential (ESP)12,24b,26 analyses to quantify the strengths of the hydrogen bonds in the hydrogen-bond network and the electrostatic effects. The estimated EHB values are given in Fig. S4; the N–H⋯O HBs (a and b) contribute 14.9 kcal mol−1 (6.5 + 8.4 = 14.9 kcal mol−1) to the stabilization of TS1cR but only 11.6 kcal mol−1 (4.4 + 7.2 = 11.6 kcal mol−1) to the stabilization of TS1cS. This difference in electrostatic stabilization results in a 3.3 kcal mol−1 energy gap between the stereocontrolling TSs, favoring TS1cR; this accounts for both short N–H⋯O interactions. Moreover, the electron density between the catalyst and substrates in TS1cR (Vs = −39.17 kcal mol−1) is greater than that in TS1cS (Vs = −26.64 kcal mol−1); thus, the electrostatic interaction is more favorable in TS1cR.27 Consequently, the two N–H⋯O HBs in TS1cR are electrostatically preferred to those in TS1cS. These differences in electrostatic stabilization depend on the orientation of the substrate within the stereogenic electrostatic environment of the catalyst. An analogous electrostatic mode of stereoinduction was demonstrated by Chong28 and Wheeler12,24b,29 recently. Therefore, the enantioselectivities of these three CPA-catalyzed asymmetric arylation reactions a–c are mainly induced by favorable C–H⋯O interactions, distortion interactions, and electrostatic interactions in the preferential TSs leading to the major products.

2.4. The enantioselectivity is mainly induced by the CPA backbone configuration

To validate our hypothesis, we also calculated the enantioselectivities of TS1bR′ and TS1cS′ catalyzed by SPINOL-based (R)-CPA-3 in reaction b as well as those of TS1cR′ and TS1cS′ catalyzed by BINOL-based (R)-CPA-2 in reaction c. The computed relative free energy barriers and ee values are shown in Table 4; they are in remarkable agreement with the experimental results. The space-filling model2d in Fig. 5 demonstrates that in the (R)-CPA-2 catalyzed TSmajor, the substrates are positioned somewhat perpendicular with respect to the 3,3′-substituents (TS1bR and TS1cR′), while in the minor TSs, a parallel orientation of the substrates with respect to the 3,3′-substituents was noted (TS1bS and TS1cS′). However, for (R)-CPA-3, a parallel orientation is exhibited in the major TSs (TS1bS′ and TS1cS), and the substrates and 3,3′-substituents are almost vertical in the minor TSs (TS1bR′ and TS1cR).
Table 4 Experimental38 and corresponding theoretical ee values (%) for reactions b and c catalyzed by (R)-CPA-1, (R)-CPA-2, and (R)-CPA-3
Reaction Entry Cat. Backbone 3,3′-Group Solvent ee (exptl) ΔΔG ee (theor)
b 1 (R)-CPA-2 BINOL C6H5 Toluene 97% 2.9 99%
2 (R)-CPA-3 SPINOL 9-Phenanthryl DCM −90% −1.3 −82%
c 3 (R)-CPA-3 SPINOL 9-Phenanthryl DCM 99% 2.0 94%
4 (R)-CPA-2 BINOL C6H5 DCM −82% −1.2 −77%



image file: c9cy01420a-f5.tif
Fig. 5 Space-filling models for (a) TS1bR, TS1bS, TS1bR′, and TS1bS′; (b) TS1cR, TS1cS, TS1cR′, and TS1cS′.

Moreover, the 2-position tert-butyl substituent in substrate 2b has greater steric hindrance than the methyl substituent in substrate 2c. Therefore, these variances will lead to differences in the distortion and interaction energies between the TSmaj and TSmin catalyzed by (R)-CPA-2 and (R)-CPA-3 for these two reactions. We used the distortion/interaction model to further analyze the origin of the energy differences between TS1bR′ and TS1bS′ as well as TS1cR′ and TS1cS′ for reactions b and c (Table 5).

Table 5 Differences in solvent-phase energies (ΔΔE) between the stereocontrolling TS structures of reactions b and c catalyzed by (R)-CPA-3 and (R)-CPA-2. Decomposition of ΔΔE into distortion (ΔΔEcat and ΔΔEsub) and interaction (ΔΔEint) energies in kcal mol−1
Reaction Catalyst ΔΔE ΔΔEdist ΔΔEcat ΔΔEsub ΔΔEint
b CPA-3 −1.1 −7.6 −3.6 −4.0 6.5
c CPA-2 −1.3 7.2 4.1 3.1 −8.5


For reaction b catalyzed by (R)-CPA-3, the interaction energy (ΔΔEint = 6.5 kcal mol−1) favors the formation of the minor stereoisomer (TS1bR′); however, these effects are overshadowed by the total distortion of the catalyst and substrates (ΔΔEdist = −7.6 kcal mol−1, Table 5), which favors TSmaj (TS1bS′). Thus, the relative energy of parallel conformation TS1bS′ is found to be 1.1 kcal mol−1, which is slightly more stable than the vertical conformation TS1bR′. For reaction c, our calculation results suggested that the ΔΔG value of TS1cS′ is 1.2 kcal mol−1 lower than that of the counterpart TS1cR′. This is because the ΔΔEint (−8.5 kcal mol−1) strongly favors TS1cS′. As shown in Fig. 6, double and quintuple C–H⋯F interactions are found in TS1cR′ (Cc1–Hc1⋯F1 2.75 Å and C2–H5⋯F2 2.42 Å) and TS1cS′ (Cc1–Hc1⋯F4 2.48 Å, C2–H5⋯F3 2.38 Å, Cc2–Hc2⋯F5 2.46 Å, and C4–H4⋯F1 2.43 Å), respectively. Furthermore, the HBs in TS1cS′ are stronger than those in TS1cR′, which is illustrated by the distances of N–H⋯O between the phosphoric acid functionality and substrates (1.66 Å and 1.75 Å for TS1cS′vs. 1.81 Å and 1.91 Å for TS1cR′, respectively). This effect is compensated by the distortions of the substrate and catalyst (ΔΔEdist = 7.2 kcal mol−1). Thus, this strong reversal of enantioselectivity for the different backbone-catalyzed reactions can be identified by the different combination modes between the substrates and catalysts controlled by the orientations of the phosphoric acid functional groups.


image file: c9cy01420a-f6.tif
Fig. 6 Transition state structures leading to the major and minor products of reaction c catalyzed by (R)-CPA-2 (TS1cR′ and TS1cS′).

More importantly, the experimental results of related representative reactions catalyzed by both BINOL and SPINOL-derived phosphoric acids remarkably confirmed our general conclusion (Scheme 3).1a,2e,5,7c,13a,30 When the catalysts with BINOL or SPINOL skeletons are in the same configuration (R/R or S/S), the enantioselectivity of the products is inverted. However, the same signs of enantioselectivity can be obtained by using backbones with converse configurations (R/S or S/R). Therefore, our general conclusion suggests that the same sense (R or S) of BINOL and SPINOL backbones can control the signs of enantioselectivity (positive or negative ee values), which can be used to guide the strategic choice of CPA skeletons. Without a detailed mechanistic investigation, only a correlation can be established. However, the combination of our calculation results and these experimental data supports our hypothesis. We hope to develop a more comprehensive predictive model to aid optimal catalyst choice.


image file: c9cy01420a-s3.tif
Scheme 3 Related representative reactions (a–j) catalyzed by BINOL and SPINOL scaffolds and their corresponding ee values.

Conclusions

In conclusion, the governmental roles of the BINOL and SPINOL scaffolds of CPA catalysts in determining the signs of enantioselectivity (positive or negative ee values) have been investigated. Examination of the CPA catalysts shows that the central phosphorus has stereogenicity which is induced by the BINOL and SPINOL scaffolds; also, for the same sense enantioselectivity (R or S) of different backbones, the orientations of the phosphoric acid functional groups are reversed. Therefore, we assume that each orientation of the phosphoric acid functional groups may have different substrate combination modes and may further control the enantioselectivity. In other words, the (R/S) BINOL- or (S/R) SPINOL-derived backbone can tune the sign of enantioselectivity (positive or negative ee value) of the final products, respectively. In order to verify our hypothesis, we chose efficient enantioselective organocatalytic asymmetric arylation case reactions a–c catalyzed by CPAs with two different backbones. The BINOL and SPINOL backbones of CPAs with the same (R/S) configuration have different influences on the enantioselectivity, with high positive and negative ee values. Two activation modes (M1 and M2) have been identified according to the position of the biphenyl and indole rings of the two substrates (same side or different sides). The calculations show that M1 mode is better than M2 mode due to the preferential distortion of the catalyst and noncovalent interactions of the substrates with the 3,3′-aryl substituents of the catalyst in the stereocontrolling TSs. The enantioselectivities predicted on the basis of the M1 mode are in good agreement with the experimental data, lending strong support to the plausibility of this mode. Moreover, although the high enantioselectivities in reactions a, b, and c are driven primarily by favorable C–H⋯O interactions, catalyst and substrate distortion interactions, and electrostatic interactions in the favored TS, the signs of enantioselectivity still follow our hypothesis. More importantly, the experimental results of related representative reactions catalyzed by both BINOL and SPINOL-derived phosphoric acids remarkably confirmed our general conclusion. Our general applicability insights into the nature of the roles of different scaffolds are expected to be used as a guide for catalyst selection for a given transformation based on the reaction pathway and to develop powerful related organocatalytic reactions.

Computational details

The computational calculations were carried out using the Gaussian 09 suite of the Quantum Chemistry Program.31 The geometries were optimized using the B3LYP-D3 (ref. 32) density functional in conjunction with the 6-31G*33 basis set. Grimme's D3 correction34 provides better treatment of dispersion interactions. Furthermore, harmonic vibrational frequency analysis35 was performed at the same level of theory to characterize the stationary points as minimum energy structures (no imaginary frequency) or transition states (only one imaginary frequency). Additionally, calculations of the intrinsic reaction coordinate (IRC)36 were investigated to verify that the transition states can connect with the appropriate reactant and product. Based on the B3LYP-D3/6-31G* optimized geometries, the single-point energies for all stationary points were calculated at the M06-2X/6–311+G**37 level using the SMD38 solvation model. Single-point energies were also evaluated within the SMD model using the b97-D3 (ref. 39) functional to compare the stereoselectivities. The ee values were calculated by the Boltzmann distributions of the diastereomeric transition states.40 The charge density analysis was performed using the natural bond orbital (NBO) approach41 at the M06-2X/6-311+G** wave function. The atoms in molecule (AIM) formalism42 and non-covalent interaction index (NCI)43 were studied to characterize the important weak interatomic interactions present in the stereocontrolling transition states. Optimized structures were generated using CYLview.44

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support by the National Natural Science Foundation of China (21873018, 21573036, and 21603028), the Fundamental Research Funds for the Central Universities (2412019FZ010), and the open project of the Jilin Province Key Laboratory of Organic Functional Molecular Design & Synthesis (130028655) are greatly acknowledged.

Notes and references

  1. (a) M. Terada, Chem. Commun., 2008, 4097 RSC; (b) A. Rahman and X. Lin, Org. Biomol. Chem., 2018, 16, 4753 RSC; (c) T. Akiyama, J. Itoh, K. Yokota and K. Fuchibe, Angew. Chem., Int. Ed., 2004, 43, 1566 CrossRef CAS PubMed; (d) D. Uraguchi and M. Terada, J. Am. Chem. Soc., 2004, 126, 5356 CrossRef CAS PubMed; (e) F. Xu, D. Huang, C. Han, W. Shen, X. Lin and Y. Wang, J. Org. Chem., 2010, 75, 8677 CrossRef CAS PubMed; (f) X. Tian, N. Hofmann and P. Melchiorre, Angew. Chem., Int. Ed., 2014, 53, 2997 CrossRef CAS PubMed; (g) J. Zhang, P. Yu, S. Y. Li, H. Sun, S. H. Xiang, J. J. Wang, K. N. Houk and B. Tan, Science, 2018, 361, 1072 CrossRef PubMed; (h) P. A. Champagne and K. N. Houk, J. Am. Chem. Soc., 2016, 138, 12356 CrossRef CAS PubMed; (i) T. Yamanaka, A. Kondoh and M. Terada, J. Am. Chem. Soc., 2015, 137, 1048 CrossRef CAS PubMed; (j) J. P. Reid and J. M. Goodman, Chem. – Eur. J., 2017, 23, 14248 CrossRef CAS PubMed; (k) A. Changotra, S. Das and R. B. Sunoj, Org. Lett., 2017, 19, 2354 CrossRef CAS PubMed; (l) P. Maity, R. P. Pemberton, D. J. Tantillo and U. K. Tambar, J. Am. Chem. Soc., 2013, 135, 16380 CrossRef CAS PubMed; (m) J. Y. See, H. Yang, Y. Zhao, M. W. Wong, Z. Ke and Y.-Y. Yeung, ACS Catal., 2018, 8, 850 CrossRef CAS.
  2. (a) Y. Kwon, J. Li, J. P. Reid, J. M. Crawford, R. Jacob, M. S. Sigman, F. D. Toste and S. J. Miller, J. Am. Chem. Soc., 2019, 141, 6698 CrossRef CAS PubMed; (b) D. Qian, L. Wu, Z. Lin and J. Sun, Nat. Commun., 2017, 8, 567 CrossRef PubMed; (c) T. Akiyama, Chem. Rev., 2007, 107, 5744 CrossRef CAS PubMed; (d) D. Parmar, E. Sugiono, S. Raja and M. Rueping, Chem. Rev., 2014, 114, 9047 CrossRef CAS PubMed; (e) G. Jindal and R. B. Sunoj, Angew. Chem., Int. Ed., 2014, 53, 4432 CrossRef CAS PubMed; (f) Y. B. Wang, S. C. Zheng, Y. M. Hu and B. Tan, Nat. Commun., 2017, 8, 15489 CrossRef PubMed; (g) H. Wu, Q. Wang and J. Zhu, Angew. Chem., Int. Ed., 2016, 55, 15411 CrossRef CAS PubMed; (h) H. Wu, Q. Wang and J. Zhu, Angew. Chem., Int. Ed., 2017, 56, 5858 CrossRef CAS PubMed; (i) F. Zhou and H. Yamamoto, Angew. Chem., Int. Ed., 2016, 55, 8970 CrossRef CAS PubMed; (j) Y. Shimoda and H. Yamamoto, J. Am. Chem. Soc., 2017, 139, 6855 CrossRef CAS PubMed; (k) S. Narute, R. Parnes, F. D. Toste and D. Pappo, J. Am. Chem. Soc., 2016, 138, 16553 CrossRef CAS PubMed.
  3. Z. L. Xia, C. Zheng, S. G. Wang and S. L. You, Angew. Chem., Int. Ed., 2018, 57, 2653 CrossRef CAS PubMed.
  4. (a) Y. H. Chen, D. J. Cheng, J. Zhang, Y. Wang, X. Y. Liu and B. Tan, J. Am. Chem. Soc., 2015, 137, 15062 CrossRef CAS PubMed; (b) Y. H. Chen, L. W. Qi, F. Fang and B. Tan, Angew. Chem., Int. Ed., 2017, 56, 16308 CrossRef CAS PubMed; (c) J. Z. Wang, J. Zhou, C. Xu, H. Sun, L. Kurti and Q. L. Xu, J. Am. Chem. Soc., 2016, 138, 5202 CrossRef CAS PubMed.
  5. J. Zhang, S. X. Lin, D. J. Cheng, X. Y. Liu and B. Tan, J. Am. Chem. Soc., 2015, 137, 14039 CrossRef CAS PubMed.
  6. Y. Zhang, Y. F. Ao, Z. T. Huang, D. X. Wang, M. X. Wang and J. Zhu, Angew. Chem., Int. Ed., 2016, 55, 5282 CrossRef CAS PubMed.
  7. (a) S. Li, J. W. Zhang, X. L. Li, D. J. Cheng and B. Tan, J. Am. Chem. Soc., 2016, 138, 16561 CrossRef CAS PubMed; (b) L. Zhang, J. Zhang, J. Ma, D. J. Cheng and B. Tan, J. Am. Chem. Soc., 2017, 139, 1714 CrossRef CAS PubMed; (c) Y. Yu, J. Li, L. Jiang, J.-R. Zhang and L. Zu, Angew. Chem., Int. Ed., 2017, 56, 9217 CrossRef CAS PubMed.
  8. (a) A. Changotra and R. B. Sunoj, Org. Lett., 2016, 18, 3730 CrossRef CAS PubMed; (b) L. Simon and J. M. Goodman, J. Org. Chem., 2011, 76, 1775 CrossRef CAS PubMed; (c) J. P. Reid and J. M. Goodman, J. Am. Chem. Soc., 2016, 138, 7910 CrossRef CAS PubMed.
  9. (a) T. Lu and S. E. Wheeler, Science, 2015, 347, 719 CrossRef CAS PubMed; (b) J. W. Bloom, R. K. Raju and S. E. Wheeler, J. Chem. Theory Comput., 2012, 8, 3167 CrossRef CAS PubMed; (c) S. E. Wheeler, Acc. Chem. Res., 2013, 46, 1029 CrossRef CAS PubMed; (d) S. E. Wheeler and J. W. G. Bloom, J. Phys. Chem. A, 2014, 118, 6133 CrossRef CAS PubMed; (e) S. E. Wheeler, T. J. Seguin, Y. Guan and A. C. Doney, Acc. Chem. Res., 2016, 49, 1061 CrossRef CAS PubMed.
  10. A. J. Neel, J. P. Hehn, P. F. Tripet and F. D. Toste, J. Am. Chem. Soc., 2013, 135, 14044 CrossRef CAS PubMed.
  11. Y. Y. Khomutnyk, A. J. Arguelles, G. A. Winschel, Z. Sun, P. M. Zimmerman and P. Nagorny, J. Am. Chem. Soc., 2016, 138, 444 CrossRef CAS PubMed.
  12. T. J. Seguin and S. E. Wheeler, ACS Catal., 2016, 6, 2681 CrossRef CAS.
  13. (a) Z. Wang, Y. F. Wong and J. Sun, Angew. Chem., Int. Ed., 2015, 54, 13711 CrossRef CAS PubMed; (b) Z. Wang, W. K. Law and J. Sun, Org. Lett., 2013, 15, 5964 CrossRef CAS PubMed; (c) F. Li, T. Korenaga, T. Nakanishi, J. Kikuchi and M. Terada, J. Am. Chem. Soc., 2018, 140, 2629 CrossRef CAS PubMed; (d) L. Simon and R. S. Paton, J. Am. Chem. Soc., 2018, 140, 5412 CrossRef CAS PubMed; (e) K. Mori, R. Isogai, Y. Kamei, M. Yamanaka and T. Akiyama, J. Am. Chem. Soc., 2018, 140, 6203 CrossRef CAS PubMed.
  14. L. W. Qi, J. H. Mao, J. Zhang and B. Tan, Nat. Chem., 2018, 10, 58 CrossRef CAS PubMed.
  15. T. H. L. Nguyen, N. Gigant and D. Joseph, ACS Catal., 2018, 8, 1546 CrossRef CAS.
  16. (a) H. Li, P. Li and L. Wang, Org. Lett., 2013, 15, 620 CrossRef CAS PubMed; (b) B. Majhi, D. Kundu, S. Ahammed and B. C. Ranu, Chemistry, 2014, 20, 9862 CrossRef CAS PubMed.
  17. (a) X. Jia and J. Han, J. Org. Chem., 2014, 79, 4180 CrossRef CAS PubMed; (b) H. Deng, H. Li and L. Wang, Org. Lett., 2015, 17, 2450 CrossRef CAS PubMed; (c) H. Deng, H. Li and L. Wang, Org. Lett., 2016, 18, 3110 CrossRef CAS PubMed.
  18. J. Hubrich, T. Himmler, L. Rodefeld and L. Ackermann, ACS Catal., 2015, 5, 4089 CrossRef CAS.
  19. (a) T. Wakaki, M. Kanai and Y. Kuninobu, Org. Lett., 2015, 17, 1758 CrossRef CAS PubMed; (b) M. T. Blyth and M. L. Coote, J. Org. Chem., 2019, 84, 1517 CrossRef PubMed; (c) N. S. Hill and M. L. Coote, J. Am. Chem. Soc., 2018, 140, 17800 CrossRef CAS PubMed.
  20. N. Muniraj and K. R. Prabhu, J. Org. Chem., 2017, 82, 6913 CrossRef CAS PubMed.
  21. D. Parmar, E. Sugiono, S. Raja and M. Rueping, Chem. Rev., 2014, 114, 9047 CrossRef CAS PubMed.
  22. (a) M. J. Ajitha and K. W. Huang, Org. Biomol. Chem., 2015, 13, 10981 RSC; (b) M. N. Grayson and J. M. Goodman, J. Am. Chem. Soc., 2013, 135, 6142 CrossRef CAS PubMed; (c) M. N. Grayson, S. C. Pellegrinet and J. M. Goodman, J. Am. Chem. Soc., 2012, 134, 2716 CrossRef CAS PubMed; (d) K. Kanomata, Y. Toda, Y. Shibata, M. Yamanaka, S. Tsuzuki, I. D. Gridnev and M. Terada, Chem. Sci., 2014, 5, 3515 RSC; (e) S. S. Meng, Y. Liang, K. S. Cao, L. Zou, X. B. Lin, H. Yang, K. N. Houk and W. H. Zheng, J. Am. Chem. Soc., 2014, 136, 12249 CrossRef CAS PubMed; (f) T. J. Seguin, T. Lu and S. E. Wheeler, Org. Lett., 2015, 17, 3066 CrossRef CAS PubMed; (g) L. Simon and R. S. Paton, Org. Biomol. Chem., 2016, 14, 3031 RSC.
  23. (a) W.-J. van Zeist and F. M. Bickelhaupt, Org. Biomol. Chem., 2010, 8, 3118 RSC; (b) F. M. Bickelhaupt and K. N. Houk, Angew. Chem., Int. Ed., 2017, 56, 10070 CrossRef CAS PubMed; (c) A. Duan, P. Yu, F. Liu, H. Qiu, F. L. Gu, M. P. Doyle and K. N. Houk, J. Am. Chem. Soc., 2017, 139, 2766 CrossRef CAS PubMed; (d) D. H. Ess and K. N. Houk, J. Am. Chem. Soc., 2007, 129, 10646 CrossRef CAS PubMed; (e) I. Fernandez and F. M. Bickelhaupt, Chem. Soc. Rev., 2014, 43, 4953 RSC; (f) R. S. Paton, S. Kim, A. G. Ross, S. J. Danishefsky and K. N. Houk, Angew. Chem., Int. Ed., 2011, 50, 10366 CrossRef CAS PubMed; (g) Y. Wang, B. Wu, L. Zheng, D. Wei and M. Tang, Org. Chem. Front., 2016, 3, 190 RSC; (h) Y.-F. Yang, Y. Liang, F. Liu and K. N. Houk, J. Am. Chem. Soc., 2016, 138, 1660 CrossRef CAS PubMed; (i) P. Yu, Z. Yang, Y. Liang, X. Hong, Y. Li and K. N. Houk, J. Am. Chem. Soc., 2016, 138, 8247 CrossRef CAS PubMed.
  24. (a) T. J. Seguin and S. E. Wheeler, ACS Catal., 2016, 6, 7222 CrossRef CAS; (b) R. Maji and S. E. Wheeler, J. Am. Chem. Soc., 2017, 139, 12441 CrossRef CAS PubMed.
  25. E. Espinosa, E. Molins and C. Lecomte, Chem. Phys. Lett., 1998, 285, 170 CrossRef CAS.
  26. (a) V. M. Lau, C. F. Gorin and M. W. Kanan, Chem. Sci., 2014, 5, 4975 RSC; (b) V. M. Lau, W. C. Pfalzgraff, T. E. Markland and M. W. Kanan, J. Am. Chem. Soc., 2017, 139, 4035 CrossRef CAS PubMed; (c) K. Lee, D. L. Silverio, S. Torker, D. W. Robbins, F. Haeffner, F. W. van der Mei and A. H. Hoveyda, Nat. Chem., 2016, 8, 768 CrossRef CAS PubMed; (d) T. Lu, R. Zhu, Y. An and S. E. Wheeler, J. Am. Chem. Soc., 2012, 134, 3095 CrossRef CAS PubMed; (e) E. Lyngvi, J. W. Bode and F. Schoenebeck, Chem. Sci., 2012, 3, 2346 RSC; (f) H. Yang and M. W. Wong, J. Am. Chem. Soc., 2013, 135, 5808 CrossRef CAS PubMed.
  27. A. Bauza, S. K. Seth and A. Frontera, J. Comput. Chem., 2018, 39, 458 CrossRef CAS PubMed.
  28. R. C. Johnston, D. T. Cohen, C. C. Eichman, K. A. Scheidt and P. H. Cheong, Chem. Sci., 2014, 5, 1974 RSC.
  29. (a) A. C. Doney, B. J. Rooks, T. Lu and S. E. Wheeler, ACS Catal., 2016, 6, 7948 CrossRef CAS; (b) T. J. Seguin and S. E. Wheeler, Angew. Chem., Int. Ed., 2016, 55, 15889 CrossRef CAS PubMed; (c) R. Maji, P. A. Champagne, K. N. Houk and S. E. Wheeler, ACS Catal., 2017, 7, 7332 CrossRef CAS.
  30. (a) J. S. Lin, T. T. Li, G. Y. Jiao, Q. S. Gu, J. T. Cheng, L. Lv and X. Y. Liu, Angew. Chem., Int. Ed., 2019, 58, 1 CrossRef; (b) Y. Ota, A. Kondoh and M. Terada, Angew. Chem., Int. Ed., 2018, 57, 13917 CrossRef CAS PubMed; (c) J. S. Lin, P. Yu, L. Huang, P. Zhang, B. Tan and X. Y. Liu, Angew. Chem., Int. Ed., 2015, 54, 7847 CrossRef CAS PubMed; (d) D. Qian, M. Chen, A. C. Bissember and J. Sun, Angew. Chem., Int. Ed., 2018, 57, 3763 CrossRef CAS PubMed; (e) H. Li, R. Tong and J. Sun, Angew. Chem., Int. Ed., 2016, 55, 15125 CrossRef CAS PubMed.
  31. M. J. T. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Gaussian, Inc., Wallingford, CT, 2009 Search PubMed.
  32. (a) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785 CrossRef CAS PubMed; (b) A. D. Becke, J. Chem. Phys., 1993, 98, 5648 CrossRef CAS; (c) A. D. Becke, J. Chem. Phys., 1993, 98, 1372 CrossRef CAS.
  33. (a) P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 213 CrossRef CAS; (b) W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys., 1972, 56, 2257 CrossRef CAS.
  34. (a) S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104 CrossRef PubMed; (b) S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456 CrossRef CAS PubMed.
  35. P. J. Stephens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, J. Phys. Chem., 1994, 98, 11623 CrossRef CAS.
  36. (a) C. Gonzalez and H. B. Schlegel, J. Chem. Phys., 1989, 90, 2154 CrossRef CAS; (b) C. Gonzalez and H. B. Schlegel, J. Phys. Chem., 1990, 94, 5523 CrossRef CAS; (c) K. Fukui, Acc. Chem. Res., 1981, 14, 363 CrossRef CAS.
  37. (a) Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2007, 120, 215 Search PubMed; (b) J. P. Lovie-Toon, C. M. Tram, B. L. Flynn and E. H. Krenske, ACS Catal., 2017, 7, 3466 Search PubMed.
  38. (a) A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B, 2009, 113, 6378–6396 CrossRef CAS PubMed; (b) A. Moran, A. Hamilton, C. Bo and P. Melchiorre, J. Am. Chem. Soc., 2013, 135, 9091 CrossRef CAS PubMed.
  39. (a) S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104 CrossRef PubMed; (b) S. Grimme, J. Comput. Chem., 2006, 27, 1787 CrossRef CAS PubMed; (c) S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456 CrossRef CAS PubMed.
  40. M. Pareek and R. B. Sunoj, ACS Catal., 2016, 6, 3118 CrossRef CAS.
  41. (a) A. E. Reed, R. B. Weinstock and F. Weinhold, J. Chem. Phys., 1985, 83, 735 CrossRef CAS; (b) E. D. Glendening, C. R. Landis and F. Weinhold, J. Comput. Chem., 2013, 34, 1429 CrossRef CAS PubMed; (c) F. Weinhold, J. Comput. Chem., 2012, 33, 2363 CrossRef CAS PubMed; (d) E. D. Glendening, C. R. Landis and F. Weinhold, Comput. Mol. Sci., 2012, 2, 1 CrossRef CAS.
  42. (a) R. F. W. Bader, Acc. Chem. Res., 1985, 18, 9 CrossRef CAS; (b) R. F. W. Bader, Chem. Rev., 1991, 91, 893 CrossRef CAS; (c) V. Tognetti and L. Joubert, Phys. Chem. Chem. Phys., 2014, 16, 14539 RSC.
  43. (a) E. R. Johnson, S. Keinan, P. Mori-Sánchez, J. Contreras-García, A. J. Cohen and W. Yang, J. Am. Chem. Soc., 2010, 132, 6498 CrossRef CAS PubMed; (b) J. Contreras-García, E. R. Johnson, S. Keinan, R. Chaudret, J.-P. Piquemal, D. N. Beratan and W. Yang, J. Chem. Theory Comput., 2011, 7, 625 CrossRef PubMed; (c) W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics Modell., 1996, 14, 33 CrossRef CAS; (d) T. Lu and F. Chen, J. Comput. Chem., 2012, 33, 580 CrossRef CAS PubMed.
  44. C. Y. Legault, CYLview, 1.0b; Université de Sherbrooke, Sherbrooke, Quebec, Canada, 2009, http://www.cylview.org Search PubMed.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cy01420a

This journal is © The Royal Society of Chemistry 2019