Next Article in Journal
PiFM and XPS Studies of Porous TiO2 Films for the Photocatalytic Decomposition of Polystyrene
Next Article in Special Issue
Density Functional Theory Study of CuAg Bimetal Electrocatalyst for CO2RR to Produce CH3OH
Previous Article in Journal
Hollow g-C3N4@Cu0.5In0.5S Core-Shell S-Scheme Heterojunction Photothermal Nanoreactors with Broad-Spectrum Response and Enhanced Photocatalytic Performance
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Catalysts
Volume 13
Issue 4
10.3390/catal13040724
catalysts-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Evidence of a Wheland Intermediate in Carboxylate-Assisted C(sp2)−H Activation by Pd(IV) Active Catalyst Species Studied via DFT Calculations

by
Ji Eun Park
1,2 and
Youn K. Kang
2,*
1
Department of Chemistry, College of Natural Sciences, Seoul National University, Seoul 08826, Republic of Korea
2
Department of Chemistry and Energy Engineering, Sangmyung University, Seoul 03016, Republic of Korea
*
Author to whom correspondence should be addressed.
Catalysts 2023, 13(4), 724; https://doi.org/10.3390/catal13040724
Submission received: 12 February 2023 / Revised: 5 April 2023 / Accepted: 6 April 2023 / Published: 11 April 2023
(This article belongs to the Special Issue Theoretical and Computational Studies of Catalytic Reactions)
Browse Figures
Versions Notes

Abstract

:
Evidence of a Wheland intermediate in carboxylate-assisted C−H activation was found using DFT calculations when the Pd(IV) catalyst species was postulated as the active catalyst species (ACS). In order to delineate the reaction mechanism of Pd-catalyzed bisarylation of 3-alkylbenzofuran, five hypothetical catalyst species, [Pd(OAc)(PMe3)(Ph)] (I), [Pd(OAc)2] (II), [Pd(OAc)2(PMe3)] (III), [Pd(OAc)2(Ph)]+ (IV) and [Pd(OAc)2(PMe3)(Ph)]+ (V) were tested as potential ACS candidates. The catalyst species I, previously reported as an ACS in the context of ambiphilic metal−ligand assistance or a concerted metalation-deprotonation mechanism, was unsuccessful, with maximum activation barriers (ΔG‡max) for the C(sp2)−H and C(sp3)−H activations of 33.3 and 51.4 kcal/mol, respectively. The ΔG‡max values for the C(sp2)−H and C(sp3)−H activations of II−V were 23.8/28.7, 32.0/49.6, 10.9/10.9, and 36.0/36.0 kcal/mol, respectively, indicating that ACS is likely IV. This catalyst species forms an intermediate state (IV_1) before proceeding to the transition state (IV_TS1,2) for C(sp2)−H activation, in which C(2) atom of 3-methylbenzofuran has a substantial σ-character. The degree of σ-character of the IV_1 state was further evaluated quantitatively in terms of geometric parameters, partial charge distribution, and activation strain analysis. The analysis results support the existence of a Wheland intermediate, which has long been recognized as the manifestation of the electrophilic aromatic substitution mechanism yet never been identified computationally.

Graphical Abstract

1. Introduction

Transition-metal-catalyzed direct C−H bond activation has emerged as a crucial topic in modern organometallic chemistry due to its potential for the catalytic functionalization of naturally abundant hydrocarbons [1,2,3,4,5,6,7,8,9]. The ability to activate C−H bonds without the need for prefunctionalization steps offers a more sustainable and atom-economical synthetic route to valuable chemicals and materials. Moreover, it offers a new strategy to acccess complex molecular architectures that are difficult to prepare using traditional synthetic methods. The elucidation of the underlying mechanisms of C−H bond activation reactions is essential for advancing this field and developing more efficient and selective catalysts. The mechanism of transition-metal-catalyzed C−H activation processes at the early stage of its development involves (i) oxidative addition with electron-rich late transition metals, (ii) σ-bond metathesis with early transition metals, (iii) 1,2-addition with both early and late transition metals, and (iv) electrophilic aromatic substitution (SEAr) mainly with late transition metals [3,8,10,11,12,13]. However, recent mechanistic studies have suggested that the major interaction between the transition-metal catalyst and substrate resides within the continuum of electrophilic, ambiphilic, and nucleophilic interactions [8,14,15].
Recently, Chung and coworkers reported the Pd-catalyzed bisarylation of 3-alkylbenzofuran (Scheme 1) [16]. Tandem C(sp2)−H and C(sp3)−H activations occurred even with only one equivalent of phenyl iodide in this reaction. This reaction utilized a catalyst system comprising a transition metal precursor complex aided by bifunctional ligands bearing a Lewis-basic heteroatom. The researchers employed [Pd(OAc)2] as a catalyst precursor and pivalate as a bifunctional ligand. These conditions have been commonly known as “carboxylate-assisted C−H activation”, which has gained widespread popularity in recent years due to its ease of C−H activation [8,17,18,19].
The mechanism of this reaction has been under intense investigation and a comprehensive summary of the scientific efforts has been provided [8,12,18]. Initially, carboxylate-assisted C−H activation was considered to be SEAr, with the Wheland intermediate being prevalent [10]. However, despite extensive efforts, no convincing computational evidence for SEAr has been reported [18]. Recent mechanistic investigations have instead conceptualized these reactions in terms of “ambiphilic metal–ligand assistance” (AMLA) [20,21] or “concerted metalation-deprotonation” (CMD) [22,23,24], with the former emphasizing the presence of an agostic intermediate prior to the main C−H activation. These concepts were validated by computational studies [21,24]. In this context, the term electrophilic CMD (eCMD) has been introduced recently to describe cases in which the mechanism falls between the CMD and SEAr [25].
In Chung and coworker’s report, there were a couple of important experimental results that provided insights to elucidate the reaction mechanism [16]: (i) The reaction proceeds at room temperature, implying that the energetic span (ΔG‡span) [26,27,28,29] for the entire reaction pathway is small. (ii) The kinetic isotope effect (KIE) experiment (KIE values for k1/k1-D = 1.2 and k1/k1-CD3 = 1.1) indicated that the C−H bond cleavage event, whether C(sp2)−H or C(sp3)−H, is not rate-determining.
Based on these experimental clues, we commenced our computational work to understand the reaction mechanism of this peculiar results, assuming that the reaction mechanism was based on AMLA or CMD. We chose the structure of the active catalyst species (ACS) as [Pd(OAc)(PMe3)(Ph)] (I) inspired by Fagnou’s work [23], given the similarity of the reaction conditions between the two groups. The structural motifs of the key transition states for the C−H bond cleavage processes were adopted from related computational studies [23,24,30]. While the details of this computational study will be described in the later part of this manuscript, only the calculation results obtained with I as ACS showed that the ΔG‡ values for the C(sp2)−H and C(sp3)−H activations were 33.3 and 51.4 kcal/mol, respectively, from the initial state. These calculation results appeared to be inconsistent with the experimental results, according to which the reaction proceeded even at 25 °C, implying that other species could work as an ACS.
The identification of the correct ACS and the formation of an appropriate catalyst-substrate adduct conformation are crucial prerequisites in computational studies aimed at understanding the mechanism of transition-metal-catalyzed organic transformations [12,18]. The transient nature of reactive intermediates and transition states as well as the structural changes that ACS undergoes during the catalytic cycle make it challenging to obtain experimental structural information [18]. Therefore, computational studies, which are not limited by these restrictions, have therefore become a valuable research tool in the field of homogeneous catalysis. Advances in density functional theory (DFT) calculations have significantly impacted this area of research, leading to a growing reliance on DFT calculations in investigations of catalytic mechanisms [31,32,33].
In order to identify the species that acts as the ACS in the above mentioned reaction, we considered a series of Pd complexes as potential candidates for DFT calculations. During the investigation of the reaction mechanism in these hypothetical complexes, we found that [Pd(OAc)2(Ph)]+ (IV) could serve as a competent ACS of the reaction, even in the absence of a phosphine ligand. Further analysis of the C(sp2)−H activation pathway revealed that it proceeds via a Wheland intermediate, suggesting that the reaction mechanism may be a SEAr reaction. The results of this computational study are presented in this report.

2. Results and Discussion

When 3-methylbenzofuran approached I (1), intermediate I_1 formed, likely due to the electron-rich nature of the heteroarene substrate (Figure 1). This process was found to be substantially endergonic by 20.7 kcal/mol with an activation barrier of 21.4 kcal/mol. In the I_1 state, the substrate was positioned trans to the phosphine ligand, with Pd−C(2) and Pd−C(3) distances of 2.499 and 2.830 Å, respectively (Figure 2a). Being not perfect, the interaction mode between the ACS and the substrate was an η2 π-bond. The distances between C(2)−H and Pd−H were found to be 1.084 and 2.609 Å, respectively, indicating that the C(2)−H bond was not polarized and the agostic interaction between Pd and H atoms was not present [21]. A more detailed analysis of the structures of the intermediate and subsequent transition states will be provided in the later part of this manuscript.
Catalysts 13 00724 g001 550
Figure 1. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by I. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The phenyl and trimethylphosphine groups are drawn translucent for clarity.
Figure 1. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by I. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The phenyl and trimethylphosphine groups are drawn translucent for clarity.
Catalysts 13 00724 g001
Catalysts 13 00724 g002 550
Figure 2. Optimized geometries of intermediate states prior to the C−H bond cleavage: (a) I_1 (Figure 1), (b) II_1 (Figure 3), (c) III_1 (Figure 4), (d) IV_1 (Figure 5), and (e) V_1 (Figure 6). Interactions between acetate oxygen atom and C(sp2)−H atoms are shown as blue dotted lines. Pd−H interactions are shown as orange dotted lines. Deviation angles (α) defined in the figure of Table 2 are displayed in green. Note that (a–c) are neutral and (d,e) are +1 cationic.
Figure 2. Optimized geometries of intermediate states prior to the C−H bond cleavage: (a) I_1 (Figure 1), (b) II_1 (Figure 3), (c) III_1 (Figure 4), (d) IV_1 (Figure 5), and (e) V_1 (Figure 6). Interactions between acetate oxygen atom and C(sp2)−H atoms are shown as blue dotted lines. Pd−H interactions are shown as orange dotted lines. Deviation angles (α) defined in the figure of Table 2 are displayed in green. Note that (a–c) are neutral and (d,e) are +1 cationic.
Catalysts 13 00724 g002
The intermediate, I_1, was observed to diverge into two distinct transition states, I_TS1,2 and I_TS1,3, that corresponded to the C(sp2)−H and C(sp3)−H activation pathways, respectively. The structures of these transition states were obtained through potential scans by rotating the Pd−C(2) bond in clockwise and counterclockwise directions, leading to the formation of six-membered-ring structural motifs (Figure 1b). The constituting atoms in these six-membered rings differed between the two transition states, with the former featuring Pd−C(2)−H−O−C−κ1−O and the latter having Pd−C(2)−C(3)−C(10)−H−κ1−O. Notably, the C(10)−H bond in I_TS1,3 did not directly interact with the Pd metal and was instead facilitated by the κ1−O atom of the acetate ligand, which has been characterized as allylic C−H activation [34,35,36,37]. Meanwhile, C(sp2)−H cleavage was supported by the displaced carbonyl oxygen. The consideration of a four-membered-ring structural motif involving Pd, C(2), H, and κ1−O for C(sp2)−H activation was also evaluated. However, it was discarded due to its significantly higher free energy compared to its six-membered-ring analog.
The activation energy (ΔG‡) required for the C(sp2)−H and C(sp3)−H activation from the intermediate state I_1 was calculated to be 12.6 and 30.7 kcal/mol, respectively. The differences in energies between the transition state (TS) and the initial state (I_0), in which I and 1 were isolated, were thus the maximum free energy barriers (ΔG‡max) [38] for these processes, which were determined to be 33.3 and 51.4 kcal/mol for C(sp2)−H and C(sp3)−H activations, respectively. The free energies of the product states, I_2 and I_3, were calculated to be 23.7 and 24.7 kcal/mol, respectively, relative to the I_0 state, indicating an overall endergonic nature of these reactions. These results were not in agreement with the experimental observation, as the reaction proceeded even at 25 °C, suggesting the possibility of the existence of another species acting as the ACS. Careful exploration to identify the ACS is thus necessary before further elucidation of the entire reaction mechanism.
As mentioned in the introduction, we considered four additional model complexes as potential candidates for the ACS. The structures of these complexes were speculated based on their generation routes, as outlined in Scheme 2. The starting point was the catalyst precursor [Pd(OAc)2] (II), which could be transformed into [Pd(OAc)2(PCy3)] upon the addition of PCy3. The formation of [Pd(OAc)2(Ph)]+ (IV) could be achieved through the oxidative addition of phenyl iodide to II, followed by the dehalogenation by Ag+. The addition of PCy3 to IV would result in the formation of [Pd(OAc)2(Ph)(PCy3)]+. In order to reduce computational costs, PCy3 was replaced with PMe3 in the model complex, resulting in the establishment of [Pd(OAc)2(PMe3)] (III) and [Pd(OAc)2(Ph)(PMe3)]+ (V) as representative model complexes instead of [Pd(OAc)2(PCy3)] and [Pd(OAc)2(Ph)(PCy3)]+.
The initial computational investigation of the carboxylate-assisted C−H activation was reported by Sakaki et al., involving benzene and methane as model substrates for C(sp2)−H and C(sp3)−H activation, respectively, and [M(κ2−O2CH)2] (M = Pd, Pt) as the ACS [39]. According to this mechanism, the activation process involves the formation of an intermediate state where a formate ligand is displaced from the κ2 to κ1 mode, followed by heterolytic cleavage of the C−H bond. In line with this scenario, we performed calculations on the II ACS (Figure 3).
Catalysts 13 00724 g003 550
Figure 3. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by II. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The spectator acetate group is drawn translucent for clarity.
Figure 3. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by II. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The spectator acetate group is drawn translucent for clarity.
Catalysts 13 00724 g003
The intermediate state II_1, in which the C(2) atom of 1 occupies one coordination site of the pseudo-square planar geometry of II, was located. This process was endergonic by 7.5 kcal/mol and required the κ2-to-κ1 displacement of one carboxylate ligand whose ΔG‡ was 23.8 kcal/mol. The C(2)−H and Pd−H distances of II_1 were 1.078 and 2.645 Å, respectively, indicating a lack of agostic character (Figure 2b) [21]. The structural motif of the transition state for the C−H activation was similar to that obtained for I and featured a characteristic six-membered-ring geometry. However, a different transition state structure was obtained for C(sp3)−H activation compared to that of I. From the II_1 structural configuration, a clockwise rotation of 3-methylbenzofuran around the Pd−C(2) bond facilitates an interaction between the displaced oxygen atom of the acetate ligand and the hydrogen atom of the methyl group. Subsequent potential scan by constricting the distance between these two entities led to the location of a transition state suitable for C(sp3)−H activation. Notably, this transition state structural motif forms an unusual eight-membered-ring configuration comprising Pd−C(2)−C(3)−C(10)−H−O−C−κ1−O atoms. An alternative six-membered-ring transition state structure consisting of Pd−C(2)−C(3)−C(10)−H−κ1−O atoms was also obtained for C(sp3)−H activation. However, this transition state exhibited an energy that was 5.8 kcal/mol higher than that of the eight-membered-ring counterpart. In spite of the revelation that an eight-membered-ring transition state geometry possesses a lower activation energy than the conventional six-membered-ring configuration, C(sp3)−H activation still necessitates a notably higher activation energy than C(sp2)−H activation, similar to the case of I. The ΔG‡ values for the heterolytic cleavages of the C(sp2)−H and C(sp3)−H bonds from the II_1 state were 10.5 and 27 kcal/mol, respectively. As a result, the C(sp2)−H activation was endergonic by 12.6 kcal/mol, while the C(sp3)−H activation was mildly exergonic by −2.6 kcal/mol from the II_0 initial state. Notably, given the fact that the energy of II_TS1,2 is lower than that of II_TS0,1, the overall free energy barrier governing the formation of II_2 is determined by II_TS0,1, whereas that governing the formation of II_3 is dictated by II_TS1,3, with the values of 23.8 and 28.7 kcal/mol, respectively.
We next examined the C−H activation pathway using an active catalytic species in which phosphine is bound to [Pd(OAc)2]. The approach of 1 towards the metal center of III necessitated the displacement of oxygen atom of the κ2-carboxylate ligand. The ΔG‡ of this process was in a similar level with the case II (21.7 kcal/mol). However, the energy of the III_1 state was 16.2 kcal/mol higher than that of III_0 indicating that this process is highly endergonic (Figure 4). The resulting III_1 state showed no agostic character between Pd and H atoms, as revealed by the C(2)−H and Pd−H distances of 1.079 and 2.669 Å (Figure 2c) [21]. The following C(sp2)−H activation required 16.8 kcal/mol of ΔG‡ from the III_1 state. This resulted in a combined ΔG‡ of 32.0 kcal/mol from the III_0 state. For the C(sp3)−H activation process, the ΔG‡ was 34.4 kcal/mol, which was much greater than that of C(sp2)−H activation, even if we considered the eight-membered-ring transition state geometry. The combined ΔG‡ was 49.6 kcal/mol from the III_0 state. As a consequence of the significant stabilization of the III_0 state by 24.2 kcal/mol, the overall free energy barriers for C(sp2)−H and C(sp3)−H activation processes were determined to be 7.8 and 25.4 kcal/mol, respectively. However, the cumulative activation energies from the energy level of the III_0 state were found to be 32 and 49.6 kcal/mol, respectively. Moreover, the C(sp2)−H activation was endergonic by 31.7 kcal/mol from the III_0 initial state. These results suggest that the ACS III confers no clear advantage over the ACS II.
Catalysts 13 00724 g004 550
Figure 4. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by III. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The spectator acetate and trimethylphosphine groups are drawn translucent for clarity.
Figure 4. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by III. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The spectator acetate and trimethylphosphine groups are drawn translucent for clarity.
Catalysts 13 00724 g004
We next considered the Pd(IV) catalyst species. Research on Pd(IV)-catalyzed C−H activation has recently become intense [19,40,41,42,43,44]. The presence of transient reactive Pd(IV) intermediate was detected by electrospray ionization tandem mass spectrometry [45]. Theoretical calculation results that support Pd(IV) species have been reported [46]. Additionally, the formation of 3,3′-dimethyl−[2,3′-bibenzofuran]−2′(3′H)-one, which resembles 3 (Scheme 1) in terms of the benzofuranone structural motif, was confirmed by using an intentionally prepared Pd(IV) catalyst species [47]. Based on this information, we constructed complexes IV and V as potential models for ACS. The generation pathway of catalytic species IV and V has been postulated and depicted in Figure S1. The Gibbs free energies of IV and V, as calculated from the zero value of [Pd(OAc)2] precursor state, are 6.6 and −14.9 kcal/mol, respectively. The ΔG‡max value during the oxidative addition of phenyl iodide to the Pd metal center was found to be 34.6 kcal/mol. It is important to note that this oxidative addition pathway is hypothetical, and thus the ΔG‡max value for the generation of IV is only an initial estimate. It is probable that a more efficient route may be present, which is our current research interest. Despite this, the calculated Gibbs free energies of IV and V are plausible, and it is appropriate to compare all energy levels of reaction pathways conducted by IV and V based on the energy level of the [Pd(OAc)2] precursor state.
Catalysts 13 00724 g005 550
Figure 5. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by IV. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The spectator acetate and phenyl groups are drawn translucent for clarity.
Figure 5. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by IV. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The spectator acetate and phenyl groups are drawn translucent for clarity.
Catalysts 13 00724 g005
Upon substrate addition with its long axis orthogonal to the z-axis of IV, when the axis containing the phenyl group is set as the z-axis, the equatorial coordination site of IV, which was left vacant due to the elimination of the iodide ligand (as seen in Figure S1), was reoccupied by the incoming substrate forming an intermediate state IV_1. It is noteworthy to focus on the structure of IV. The distances between the two oxygen atoms of the acetate ligand on the xy plane and Pd are 1.960 and 2.120 Å, respectively, indicating that they are two κ2−O atoms. However, those on the xz plane and Pd are 1.975 and 2.607 Å, respectively, suggesting that one oxygen atom is clearly displaced from Pd to some extent. This oxygen atom weakly binds to Pd and can readily depart from its position upon substrate approach, placing itself in proximity to the C(2) hydrogen of 3-methylbenzofuran in the IV_1 state. As a result, the formation process of IV_1 differs significantly from the preceding three cases. While the preceding three processes were all endergonic and required more than 20 kcal/mol of activation barrier, this process is not only exergonic by 3.7 kcal/mol, but also has an activation energy of only 10.9 kcal/mol, which is much smoother than the previous processes. Additionally, both C(sp2)−H and C(sp3)−H activations proceed smoothly from IV_1. The geometry of IV_1 is inherently susceptible to C(sp2)−H activation, and rotating the κ1−OAc group counterclockwise around the Pd−O bond by approximately 110 degrees leads to the geometry of IV_TS1,3, in which the characteristic eight-membered-ring structural motif is maintained, without disrupting the position of 3-methylbenzofuran. Consequently, the activation energies required to reach IV_TS1,2 or IV_TS1,3 from IV_1 are only 6.9 and 6.7 kcal/mol, respectively, and the energies of the resulting C(sp2)−H and C(sp3)−H activation product states are also significantly more stable, by 5.8 and 20.9 kcal/mol, respectively, than that of the IV_1 state. The energetic profiles for C(sp2)−H and C(sp3)−H activations exhibit a cascade-type behavior, with final Gibbs free energies of −2.9 and −18.0 kcal/mol, respectively. These results indicate that both C(sp2)−H and C(sp3)−H activation processes mediated by ACS IV are kinetically and thermodynamically favorable (Figure 5a).
Similar to the case of III, the addition of phosphine ligand to IV stabilized its energy significantly, which made the state energy of V_0 −14.9 kcal/mol. Since all six coordination sites around the Pd metal were occupied, the addition of 3-methylbenzofuran to the metal center required a substantial ΔG‡ of 36.0 kcal/mol. Not surprisingly, this process was found to be endergonic by 24.3 kcal/mol. The incoming substrate displaced one κ2−O atom of the acetate ligand and occupied one equatorial position of the Pd coordination sphere. In V_1 state, the displaced oxygen atom positioned in proximity to the C(2) hydrogen of 3-methylbenzofuran with O−H distance of 2.325 Å ready for the next C(sp2)−H activation. The subsequent C(sp2)−H cleavage process occurred exergonically by 18.4 kcal/mol with a ΔG‡ of 0.6 kcal/mol through the conventional six-membered-ring structural motif.
Catalysts 13 00724 g006 550
Figure 6. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by V. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The phenyl and trimethylphosphine groups are drawn translucent for clarity.
Figure 6. (a) Free energy reaction profile for C(sp2)−H (red) and C(sp3)−H (blue) activation by V. Numbers are ΔG in water at 298 °C (energies in kcal/mol). (b) Optimized geometries of transition states with selected bond lengths (Å). The phenyl and trimethylphosphine groups are drawn translucent for clarity.
Catalysts 13 00724 g006
In contrast to ACS II–IV, the C(sp3)−H cleavage process by V did not proceed through the eight-membered-ring transition state structure. The formation of the eight-membered-ring transition state structure necessitates the acetate oxygen atom to be oriented orthogonally to the xy plane. However, the presence of the phosphine ligand in the upper apex obstructed such a conformation. Instead, the C(sp3)−H cleavage process occurred through a six-membered-ring structure comprising Pd−C(2)−C(3)−C(10)−H−κ1−O, similar to I_TS3,4. This process was considerably exergonic by 57.5 kcal/mol with a ΔG‡ of 11.6 kcal.
To assess the feasibility of the five ACSs, Table 1 summarizes the largest free energy barriers for any individual step (ΔG‡i,max, kcal/mol)(ref), as well as the ΔG‡max values for C(sp2)−H and C(sp3)−H activations of each species, along with the Gibbs free energies of the product states. The results indicate that among the five species, IV is the most promising candidate for the ACS of both C(sp2)−H and C(sp3)−H activations of substrate 1, owing to its lowest ΔG‡max and ΔG‡i,max values for both activation processes. The ΔG‡max and ΔG‡i,max values for species IV are all 10.6 kcal/mol, which is consistent with the experimentally observed low reaction temperature of 25 °C.
The presence of the IV_1 intermediate state provided valuable insight into the reaction mechanism of C(sp2)−H activation. The Pd−C(2) bond length was found to be within the normal range (2.134 Å) and the C(2)−H and Pd−H distances were 1.090 and 2.573 Å, respectively. This suggests that the C(2)−H bond is not polarized and the metal-to-Hydride interaction is weak, indicating the absence of agostic character in this intermediate state. Notably, the deviation angle (α, Table 2 and Figure 2d) of the hydrogen atom at C(2) from the benzofuran plane was seemingly large (35.8°) compared to those of I_1, II_1, and III_1 (9.0°, 17.7°, and 9.8°, respectively). The same was true for the intermediate state V_1, with a deviation angle of 45.2°. Since the α value indicates the degree of σ-character in the σ−π continuum of the Pd-substrate adduct [48], this noticeable difference in α values between the Pd(II) and Pd(IV) groups suggests a possible mechanism shift due to changes in the oxidation state of the metal center. The large α value can also be considered as evidence for the existence of a Wheland intermediate and highlights the relationship between the deviation angle and the degree of σ-character in the σ−π continuum of the Pd-substrate adduct.
Table
Table 2. Mayer bond indices (BIMs), n0 values (% in parenthesis), and deviation angles (α, °) of ipso hydrogen in the optimized adduct states a.
Table 2. Mayer bond indices (BIMs), n0 values (% in parenthesis), and deviation angles (α, °) of ipso hydrogen in the optimized adduct states a.
Catalysts 13 00724 i001
BIMDeviation Angle
C(2)–C(3)C(3)–C(4)C(4)–C(9)(α, °)
11.676 (0)1.129 (0)1.261 (0)
I_11.394 (49.6)1.113 (−6.2)1.263 (−1.7)9.0
II_11.086 (103.7)1.054 (−29.1)1.257 (2.6)17.7
III_11.346 (57.9)1.105 (−9.5)1.240 (14.2)9.8
IV_11.083 (104.3)1.338 (80.1)1.183 (53.8)35.8
V_11.079 (104.9)1.337 (79.8)1.114 (102.3)45.2
1●+1.107 (100)1.389 (100)1.117 (100)
ΔBI0.569−0.2600.144
a Calculations were performed at B3LYP/def2TZVP/SDD level. BIM values were calculated by BORDER program [49].
Given the unprecedented computational evidence for the Wheland intermediate, which accompanies SEAr C−H activation mechanism, we scrutinized the IV_1 state more thoroughly. Murahashi and coworkers recently reported a method to estimate the degree of σ-character (n0) in the Pd(II)-indole complex. The n0 values were evaluated using Mayer bond indices (BIM) according to the equation, n0 (complex) [%] = 100 × [BIM (N-methylindole)—BIM (complex)]/ΔBIoM, where ΔBIoM = BIM (N-methylindole)—BIM (N-methylindolium)] [48]. Similarly, n0 values of the IV_1 state along with the other intermediates studied herein were calculated with the corresponding BIM s. In order to select proper bonds for the analyses, we first compared bond lengths of 1 and 1●+ (see figure in Table 2). The large differences of bond lengths (Δd) between 1 and 1●+ were found in C(2)–C(3), C(3)–C(4), and C(4)–C(9) bonds and thus we chose these three bonds as the indicators for the quantitative evaluation of n0 values [48]. The calculated BIM as well as n0 values are listed in Table 2.
The degree of σ-character was prominently demonstrated in the intermediate state V_1. The calculated n0 values for the C(2)–C(3) and C(4)–C(9) bonds in V_1 exceeded 100%, which suggests that V_1 can be considered a σ-complex. The same holds true for the intermediate state IV_1, as the n0 values for the C(2)–C(3) bond in this state was also above 100%, and the n0 values for the C(3)–C(4) and C(4)–C(9) bonds were also significantly high (80.1 and 53.8%, respectively). On the contrary, the n0 values for the three bonds in the Pd(II) catalyst species were lower than those in the Pd(IV) counterpart, indicating that the intermediate states of the former have much less σ-character. The degree of σ-character of these intermediate states was also evaluated in terms of Wiberg bond indices(BIW) as well as bond lengths (d). The details of these evaluation results are summarized in Supplementary Materials (Table S4). The discrimination in σ-characters between Pd(II) and Pd(IV) groups was again confirmed. The n0 values with BIW were −1.0% to 56.2% for the Pd(II) group and 75.9–87.8% for the Pd(IV) group. The corresponding values with bond lengths were 37.7% to 39.6% and 73.7% to 88.7%.
The degree of σ-character in the substrate was derived from the charge transfer from the intrinsic π-bond of the substrate to the metal center. In order to provide further evidence for the assertion that the IV_1 and V_1 states exhibit pronounced σ-character, we performed a natural bond orbital (NBO) population analysis (Table 3). The corresponding charges (q) of the catalyst and substrate moieties were calculated by summing up the individual charges of the atoms in these moieties. Given that the substrate is intrinsically neutral, the q values of the substrate serve as an indicator of the degree of charge transfer (Δq), which is directly proportional to the σ-character of the intermediates. Our NBO analysis revealed a consistently supported our previous assessments based on Mayer bond indices (BIM, n0) or deviation angles (α), with the Pd(IV) group exhibiting higher Δq values (0.782–0.815) than the Pd(II) group (0.151–0.321). Moreover, visual examination of the three-dimensional mapped surfaces of the electrostatic potential of the intermediates further highlights the discernible differences between the Pd(II) and Pd(IV) groups (Figure 7), with the contrast between the catalyst and substrate moieties being particularly pronounced in the Pd(IV) species.
We established a correlation between the parameter α and the averaged n0 values over the C(2)−C(3), C(3)−C(4), and C(4)−C(9) bonds as well as between α and Δq. These relations are co-plotted and graphically displayed in Figure 8. The graphs clearly show that the Pd(II) and Pd(IV) groups occupy distinct regions on the σ−π continuum thereby being characterized as π-complexes and σ-complexes, respectively. It is important to note that since the intermediate state IV_1a generated by the most probable ACS IV belongs to the σ-complex, confirming the Wheland nature of these states, the mechanism of C(sp2)−H activation of 3-methylbenzofuran by catalyst IV can be regarded as SEAr.
In recent years, activation strain analysis (ASA) has become increasingly popular in the field of organic synthesis, as it offers a valuable tool for predicting and rationalizing the outcomes of chemical reactions. ASA is a theoretical method used in computational chemistry to understand the factors controlling the reactivity and mechanism of organic reactions by analyzing the energies to distort isolated reactants to the transition state geometry (the distortion energy) and the energy of interaction between these distorted reactants (the interaction energy). Recently, this method has been successfully applied to analyze the reactivity of oxidative addition [50,51] and C−H bond activation [52]. To cast light on the reactivities of five ACSs studied in our work, we performed ASA for both C(sp2)−H and C(sp3)−H activation processes by five ACSs (Scheme 3). The analysis results are listed in Table 4.
In the present analysis, the distortion energy necessary to achieve a transition state geometry is partitioned into two components: one corresponding to the conversion from the initial state (X_0, X = I–V) where the catalyst and substrate are separated, to the intermediate state (X_1) geometry (EdistI); and the other associated with the transition from the intermediate geometry to the transition state (X_TS1,2) geometry (EdistII). By this way, we were able to discern dissimilar properties of the intermediates that manifest depending on the different ACS. Specifically, in the case of the Pd(II) catalyst (I−III), EdistII is considerably greater than that in step I, whereas for the Pd(IV) catalyst, EdistI and EdistII are similar to each other (IV), or EdistI is even greater than EdistII (V). This can be attributed to the distinctive nature of the intermediate species formed in the case of Pd(IV), wherein the C−H bond undergoes a significant loss of its original sp2 character and gains an sp3 character that closely resembles the structure of the transition state. As a result, EdistII is primarily governed by the lengthening of the C−H bond, rather than the sp2 to sp3 transformation of the C−H bond. As such, our analysis provided further corroboration of the formation of a Wheland intermediate in the case of Pd(IV) catalyst, ascertained through ASA.
Besides this information, ASA enabled us to determine the origin of the relative activation barriers of C(sp2)−H and C(sp3)−H activations through a comparison of the total distortion energy and interaction energy (Table S6). We observed that Edist was the dominant factor for I, II, and V, while interaction energy was the dominant factor for catalyst III. Notably, for catalyst IV, C(sp2)−H bond activation was slightly more disadvantageous than C(sp3)−H counterpart mainly due to the relatively small difference in distortion energy between the two activations. This trend was also observed in III, as the eight-membered-ring transition state structure for C(sp3)−H activation did not require significant distortion of the substrate. However, in the case of II, the significant difference in distortion energy between C(sp2)−H and C(sp3)−H activations is attributed to II_TS1,3, which constitutes an eight-membered-ring transition structure with strong Pd−C(2) sigma bonding, causing the C(2)−H atoms to be severely out of the 3-methylbenzofuran ring plane. Our analysis thus provided a more comprehensive and three-dimensional understanding of the ACS through ASA. Besides the aforementioned information, ASA allowed us to determine the origin of the difference between E‡ values of C(sp2)−H and C(sp3)−H activations by comparing the Edist and Eint (Table S6). Our findings revealed that Edist was the dominant factor for catalysts I, II, and V, while Eint was the dominant factor for catalyst III. Notably, for catalyst IV, C(sp3)−H bond activation was slightly more advantageous than its C(sp2)−H counterpart mainly due to the relatively small difference in Edist between the two activations. This trend was also observed in catalyst III, as the eight-membered-ring transition state structure for C(sp3)−H activation did not require significant distortion of the substrate. However, in the case of catalyst II, the significant difference in Edist between C(sp2)−H and C(sp3)−H activations is attributed to II_TS1,3, which constitutes an eight-membered-ring transition structure with strong Pd−C(2) σ bonding, causing the C(2)−H atoms to be severely bent out of the 3-methylbenzofuran ring plane. Thus, our ASA provided a more comprehensive understanding on the nature of catalytic C−H activation reactions undergo by five ACSs studied in this work.

3. Computational Methods

The calculations were performed using the Gaussian 09 [53] and Gaussian 16 [54] program packages. All results were obtained using a spin-restricted formalism at the DFT using the B3LYP hybrid functional [55,56,57]. All geometries of stationary states were fully optimized without any symmetry restriction. The palladium, silver, and iodine atoms were described by LANL2DZ basis set with the relativistic effective core potential (ECP) of Hay and Wadt for the inner electrons and a double-ζ basis set for the outer electrons. The standard 6–31G(d) basis set was used for the remaining atoms. Single point frequency calculations were then performed to characterize the stationary points as minima or transition states. Transition states (TS) were identified by only one imaginary frequency that corresponds to the vibrational mode of the given bond breaking/bond making. The Karlsruhe basis sets in triple-ζ level were used for all atoms [58,59]. The relativistic ECPs for the inner 28 electrons of palladium [60], silver [61], and iodine [62] atoms were represented by the Stuttgart RSC 1997 (SDD). Solvation corrections were added using a self-consistent reaction field (SCRF) approach [63,64,65] by a conductor-like polarizable continuum model (CPCM) [66,67] in water (εwater = 78.3553). Intrinsic reaction coordinate (IRC) calculations were performed to ensure that the TS actually connects the proposed reactants and products.

4. Conclusions

We conducted a DFT study to elucidate the rection mechanisms of carboxylate-assisted C(sp2)−H and C(sp3)−H activations of 3-methylbenzofurans to 3-methyl-2-phenylbenzofurans or 3-benzylbenzofuran with five hypothetical catalyst species: [Pd(OAc)(PMe3)(Ph)] (I), [Pd(OAc)2] (II), [Pd(OAc)2(PMe3)] (III), [Pd(OAc)2(Ph)]+ (IV), and [Pd(OAc)2(PMe3)(Ph)]+ (V) as potential ACS candidates. The information obtained in this work is as follows: (1) [Pd(OAc)2(Ph)]+ (IV) in the Pd(IV) oxidation state among five candidates was found to be the most favorable. The calculated reaction pathways catalyzed by IV exhibited smooth cascade-type energy profiles with 10.9 kcal/mol of ΔG‡max for both C(sp2)−H and C(sp3)−H activations. This result indicated that the inclusion of a phosphine ligand may not be crucial in the catalytic reaction conditions for the C−H activation of heteroarenes. (2) For the C(sp3)−H activation conducted by catalysts II, III, and IV, the eight-membered-ring transition state structural motif was found to be more favorable than the conventional six-membered-ring counterpart. (3) Catalyst species in the Pd(IV) oxidation states forms a Wheland intermediate. This intermediate has a strong σ-character supported by the facts that (i) the significant partial charge transfer from the substrate to the metal center was determined by an NBO population analysis, (ii) the high degree of n0 value was judged by bond length, Mayer and Wiberg bond indices of the furan ring, and (iii) a significant deviation angle (α) is pronounced between the C(2)−H bond and the benzofuran plane. These findings are of an unprecedented computational study that reveals the evidence of a Wheland intermediate in the C(sp2)−H activation. (4) The existence of the Wheland intermediate suggests that the reaction mechanism of C(sp2)−H activation is likely SEAr, indicating that a given reaction can potentially proceed through distinct mechanisms depending on the employed active catalytic species. As such, consideration of diverse ACS is crucial in understanding more precise and comprehensive insights into the reaction mechanism in the computational study.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/catal13040724/s1. Full reference of Gaussian 09 and 16 program; Table S1: Energies of local minima stationary points; Table S2: Energies and imaginary frequencies of transition states; Table S3: Atomic coordinates for stationary points; Figure S1: Free energy reaction profile for the generation of IV and V from II; Table S4: Optimized structural parameters, Wiberg bond indices, n0 values, and deviation angle of ipso hydrogen of intermediate states; Table S5: Mapped surfaces of electrostatic potential of intermediate states; Table S6: Distortion–interaction analysis.

Author Contributions

DFT calculation, J.E.P. and Y.K.K.; investigation, J.E.P. and Y.K.K.; writing—original draft preparation, J.E.P. and Y.K.K.; writing—review and editing, Y.K.K.; visualization, J.E.P. and Y.K.K.; supervision, Y.K.K.; funding acquisition, Y.K.K. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Research Foundation of Korea (NRF), grant funded by the Korea Government (2018R1D1A1B07049976). This work was also supported by the Korea Institute of Science and Technology Information with supercomputing resources including technical support (Grants KSC-2020-CRE-0369 and KSC-2021-CRE-0594).

Data Availability Statement

The data presented in this study are available in this article and Supplementary Materials.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Bergman, R.G. Activation of Alkanes with Organotransition Metal Complexes. Science 1984, 223, 902–908. [Google Scholar] [CrossRef] [PubMed]
  2. Crabtree, R.H. The Organometallic Chemistry of Alkanes. Chem. Rev. 1985, 85, 245–269. [Google Scholar] [CrossRef]
  3. Shilov, A.E.; Shul’pin, G.B. Activation of C-H Bonds by Metal Complexes. Chem. Rev. 1997, 97, 2879–2932. [Google Scholar] [CrossRef] [PubMed]
  4. Jia, C.; Kitamura, T.; Fujiwara, Y. Catalytic Functionalization of Arenes and Alkanes via C-H Bond Activation. Acc. Chem. Res. 2001, 34, 633–639. [Google Scholar] [CrossRef]
  5. Labinger, J.A.; Bercaw, J.E. Understanding and exploiting C-H bond activation. Nature 2002, 417, 507–514. [Google Scholar] [CrossRef]
  6. Brookhart, M.; Green, M.L.H.; Parkin, G. Agostic interactions in transition metal compounds. Proc. Natl. Acad. Sci. USA 2007, 104, 6908–6914. [Google Scholar] [CrossRef] [Green Version]
  7. Chen, X.; Engle, K.M.; Wang, D.H.; Yu, J.Q. Palladium(II)-catalyzed C-H activation/C-C cross-coupling reactions: Versatility and practicality. Angew. Chem. 2009, 48, 5094–5115. [Google Scholar] [CrossRef]
  8. Ackermann, L. Carboxylate-assisted transition-metal-catalyzed C-H bond functionalizations: Mechanism and scope. Chem. Rev. 2011, 111, 1315–1345. [Google Scholar] [CrossRef]
  9. Rouquet, G.; Chatani, N. Catalytic functionalization of C(sp2)-H and C(sp3)-H bonds by using bidentate directing groups. Angew. Chem. 2013, 52, 11726–11743. [Google Scholar] [CrossRef]
  10. Ryabov, A.D. Mechanisms of intramolecular activation of carbon-hydrogen bonds in transition-metal complexes. Chem. Rev. 1990, 90, 403–424. [Google Scholar] [CrossRef]
  11. Lersch, M.; Tilset, M. Mechanistic Aspects of C-H Activation by Pt Complexes. Chem. Rev. 2005, 105, 2471–2526. [Google Scholar] [CrossRef]
  12. Balcells, D.; Clot, E.; Eisenstein, O. C-H Bond Activation in Transition Metal Species from a Computational Perspective. Chem. Rev. 2010, 110, 749–823. [Google Scholar] [CrossRef]
  13. Gray, A.; Tsybizova, A.; Roithova, J. Carboxylate-assisted C-H activation of phenylpyridines with copper, palladium and ruthenium: A mass spectrometry and DFT study. Chem. Sci. 2015, 6, 5544–5553. [Google Scholar] [CrossRef] [Green Version]
  14. Ess, D.H.; Goddard, W.A.; Periana, R.A. Electrophilic, Ambiphilic, and Nucleophilic C-H Bond Activation: Understanding the Electronic Continuum of C-H Bond Activation Through Transition-State and Reaction Pathway Interaction Energy Decompositions. Organometallics 2010, 29, 6459–6472. [Google Scholar] [CrossRef] [Green Version]
  15. Ess, D.H.; Nielsen, R.J.; Goddard, W.A., III; Periana, R.A. Transition-State Charge Transfer Reveals Electrophilic, Ambiphilic, and Nucleophilic Carbon-Hydrogen Bond Activation. J. Am. Chem. Soc. 2009, 131, 11686–11688. [Google Scholar] [CrossRef] [Green Version]
  16. Cho, B.S.; Chung, Y.K. Palladium-catalyzed bisarylation of 3-alkylbenzofurans to 3-arylalkyl-2-arylbenzofurans on water: Tandem C(sp(3))-H and C(sp(2))-H activation reactions of 3-alkylbenzofurans. Chem. Commun. 2015, 51, 14543–14546. [Google Scholar] [CrossRef]
  17. Cannon, J.S.; Zou, L.; Liu, P.; Lan, Y.; O’Leary, D.J.; Houk, K.N.; Grubbs, R.H. Carboxylate-assisted C(sp(3))-H activation in olefin metathesis-relevant ruthenium complexes. J. Am. Chem. Soc. 2014, 136, 6733–6743. [Google Scholar] [CrossRef]
  18. Davies, D.L.; Macgregor, S.A.; McMullin, C.L. Computational Studies of Carboxylate-Assisted C–H Activation and Functionalization at Group 8–10 Transition Metal Centers. Chem. Rev. 2017, 117, 8649–8709. [Google Scholar] [CrossRef] [Green Version]
  19. Maleckis, A.; Kampf, J.W.; Sanford, M.S. A Detailed Study of Acetate-Assisted C–H Activation at Palladium(IV) Centers. J. Am. Chem. Soc. 2013, 135, 6618–6625. [Google Scholar] [CrossRef]
  20. Boutadla, Y.; Davies, D.L.; Macgregor, S.A.; Poblador-Bahamonde, A.I. Mechanisms of C–H bond activation: Rich synergy between computation and experiment. Dalton Trans. 2009, 30, 5820–5831. [Google Scholar] [CrossRef] [Green Version]
  21. Davies, D.L.; Donald, S.M.A.; Macgregor, S.A. Computational Study of the Mechanism of Cyclometalation by Palladium Acetate. J. Am. Chem. Soc. 2005, 127, 13754–13755. [Google Scholar] [CrossRef] [PubMed]
  22. Stuart, D.R.; Fagnou, K. The Catalytic Cross-Coupling of Unactivated Arenes. Science 2007, 316, 1172–1175. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  23. Lafrance, M.; Gorelsky, S.I.; Fagnou, K. High-Yielding Palladium-Catalyzed Intramolecular Alkane Arylation:  Reaction Development and Mechanistic Studies. J. Am. Chem. Soc. 2007, 129, 14570–14571. [Google Scholar] [CrossRef]
  24. Gorelsky, S.I.; Lapointe, D.; Fagnou, K. Analysis of the Concerted Metalation-Deprotonation Mechanism in Palladium-Catalyzed Direct Arylation Across a Broad Range of Aromatic Substrates. J. Am. Chem. Soc. 2008, 130, 10848–10849. [Google Scholar] [CrossRef] [PubMed]
  25. Wang, L.; Carrow, B.P. Oligothiophene Synthesis by a General C-H Activation Mechanism: Electrophilic Concerted Metalation-Deprotonation (eCMD). ACS Catal. 2019, 9, 6821–6836. [Google Scholar] [CrossRef]
  26. Amatore, C.; Jutand, A. Mechanistic and kinetic studies of palladium catalytic systems. J. Organomet. Chem. 1999, 576, 254–278. [Google Scholar] [CrossRef]
  27. Goossen, L.J.; Koley, D.; Hermann, H.L.; Thiel, W. Palladium Monophosphine Intermediates in Catalytic Cross-Coupling Reactions:  A DFT Study. Organometallics 2006, 25, 54–67. [Google Scholar] [CrossRef]
  28. Kozuch, S.; Shaik, S. How to Conceptualize Catalytic Cycles? The Energetic Span Model. Acc. Chem. Res. 2011, 44, 101–110. [Google Scholar] [CrossRef]
  29. Bhaskararao, B.; Singh, S.; Anand, M.; Verma, P.; Prakash, P.; Athira, C.; Malakar, S.; Schaefer, H.F.; Sunoj, R.B. Is silver a mere terminal oxidant in palladium catalyzed C-H bond activation reactions? Chem. Sci. 2020, 11, 208–216. [Google Scholar] [CrossRef] [Green Version]
  30. Rousseaux, S.; Gorelsky, S.I.; Chung, B.K.W.; Fagnou, K. Investigation of the mechanism of C(sp3)-H bond cleavage in Pd(0)-catalyzed intramolecular alkane arylation adjacent to amides and sulfonamides. J. Am. Chem. Soc. 2010, 132, 10692–10705. [Google Scholar] [CrossRef]
  31. Burke, K. Perspective on density functional theory. J. Chem. Phys. 2012, 136, 150901. [Google Scholar] [CrossRef]
  32. Klimeš, J.; Michaelides, A. Perspective: Advances and challenges in treating van der Waals dispersion forces in density functional theory. J. Chem. Phys. 2012, 137, 120901. [Google Scholar] [CrossRef] [Green Version]
  33. Goerigk, L.; Grimme, S. A thorough benchmark of density functional methods for general main group thermochemistry, kinetics, and noncovalent interactions. Phys. Chem. Chem. Phys. 2011, 13, 6670–6688. [Google Scholar] [CrossRef]
  34. Yin, G.; Wu, Y.; Liu, G. Scope and Mechanism of Allylic C-H Amination of Terminal Alkenes by the Palladium/PhI(OPiv)2 Catalyst System: Insights into the Effect of Naphthoquinone. J. Am. Chem. Soc. 2010, 132, 11978–11987. [Google Scholar] [CrossRef]
  35. Engelin, C.J.; Fristrup, P. Palladium catalyzed allylic C-H alkylation: A mechanistic perspective. Molecules 2011, 16, 951–969. [Google Scholar] [CrossRef]
  36. Liron, F.; Oble, J.; Lorion, M.M.; Poli, G. Direct Allylic Functionalization Through Pd-Catalyzed C-H Activation. Eur. J. Org. Chem. 2014, 2014, 5863–5883. [Google Scholar] [CrossRef]
  37. Duarte, F.J.S.; Poli, G.; Calhorda, M.J. Mechanistic Study of the Direct Intramolecular Allylic Amination Reaction Catalyzed by Palladium(II). ACS Catal. 2016, 6, 1772–1784. [Google Scholar] [CrossRef]
  38. The energy difference between the highest energy point during the C-H activation process and the initial state.
  39. Biswas, B.; Sugimoto, M.; Sakaki, S. C-H Bond Activation of Benzene and Methane by M(η2-O2CH)2 (M = Pd or Pt). A Theoretical Study. Organometallics 2000, 19, 3895–3908. [Google Scholar] [CrossRef]
  40. Hull, K.L.; Lanni, E.L.; Sanford, M.S. Highly regioselective catalytic oxidative coupling reactions: Synthetic and mechanistic investigations. J. Am. Chem. Soc. 2006, 128, 14047–14049. [Google Scholar] [CrossRef]
  41. Racowski, J.M.; Ball, N.D.; Sanford, M.S. C–H Bond Activation at Palladium(IV) Centers. J. Am. Chem. Soc. 2011, 133, 18022–18025. [Google Scholar] [CrossRef] [Green Version]
  42. Kawai, H.; Kobayashi, Y.; Oi, S.; Inoue, Y. Direct C-H bond arylation of arenes with aryltin reagents catalysed by palladium complexes. Chem. Commun. 2008, 12, 1464–1466. [Google Scholar] [CrossRef] [PubMed]
  43. Rosewall, C.F.; Sibbald, P.A.; Liskin, D.V.; Michael, F.E. Palladium-catalyzed carboamination of alkenes promoted by N-fluorobenzenesulfonimide via C-H activation of arenes. J. Am. Chem. Soc. 2009, 131, 9488–9489. [Google Scholar] [CrossRef] [PubMed]
  44. Juwaini, N.A.B.; Ng, J.K.P.; Seayad, J. Catalytic regioselective oxidative coupling of furan-2-carbonyls with simple arenes. ACS Catal. 2012, 2, 1787–1791. [Google Scholar] [CrossRef]
  45. Zhu, W.; Wang, H.; Peng, H.; Liu, G.; Guo, Y. Study of the Transient Reactive Pd(IV) Intermediate in the Pd(OAc)2-Catalyzed Oxidative Coupling Reaction System by Electrospray Ionization Tandem Mass Spectrometry. Chin. J. Chem. 2013, 31, 371–376. [Google Scholar] [CrossRef]
  46. Zhao, X.; Sun, C.; Lu, Y.; Xing, Z.; Sun, N.; Chen, D. A DFT study on the difference of C-H bond activation by Pd(II) and Pd(IV) complex. Comput. Theor. Chem. 2015, 1056, 41–46. [Google Scholar] [CrossRef]
  47. Cho, B.S.; Chung, Y.K. Palladium(II)-Catalyzed Transformation of 3-Alkylbenzofurans to [2,3′-Bibenzofuran]-2′(3′H)-ones: Oxidative Dimerization of 3-Alkylbenzofurans. J. Org. Chem. 2017, 82, 2237–2242. [Google Scholar] [CrossRef]
  48. Yamamoto, K.; Kimura, S.; Murahashi, T. σ-π Continuum in Indole-Palladium(II) Complexes. Angew. Chem. Int. Ed. 2016, 55, 5322–5326. [Google Scholar] [CrossRef]
  49. BORDER. Available online: http://occam.chemres.hu/programs/index.html (accessed on 5 April 2023).
  50. De Jong, G.T.; Bickelhaupt, F.M. Transition-State Energy and Position along the Reaction Coordinate in an Extended Activation Strain Model. ChemPhysChem 2007, 8, 1170–1181. [Google Scholar] [CrossRef]
  51. Legault, C.Y.; Garcia, Y.; Merlic, C.A.; Houk, K.N. Origin of Regioselectivity in Palladium-Catalyzed Cross-Coupling reactions of Polyhalogenated Heterocycles. J. Am. Chem. Soc. 2007, 129, 12664–12665. [Google Scholar] [CrossRef]
  52. Gorelsky, S.I.; Lapointe, D.; Fagnou, K. Analysis of the Palladium-Catalyzed (Aromatic)C-H Bond Metalation-Deprotonation Mechanism Spanning the Entire Spectrum of Arenes. J. Org. Chem. 2012, 77, 658–668. [Google Scholar] [CrossRef]
  53. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G.A.; et al. Gaussian 09, Revision B.01; Gaussian, Inc.: Wallingford, CT, USA, 2010. [Google Scholar]
  54. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.; Petersson, G.A.; Nakatsuji, H.; et al. Gaussian 16, Revision A.03; Gaussian, Inc.: Wallingford, CT, USA, 2016. [Google Scholar]
  55. Becke, A.D. Density-Functional Thermochemistry. 3. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648–5652. [Google Scholar] [CrossRef] [Green Version]
  56. Lee, C.; Yang, W.; Parr, R.G. Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B 1988, 37, 785–789. [Google Scholar] [CrossRef] [Green Version]
  57. Stephens, P.J.; Devlin, F.J.; Chabalowski, C.F.; Frisch, M.J. Ab-Initio Calculation of Vibrational Absorption and Circular-Dichroism Spectra Using Density-Functional Force-Fields. J. Phys. Chem. 1994, 98, 11623–11627. [Google Scholar] [CrossRef]
  58. Schäfer, A.; Horn, H.; Ahlrichs, R. Fully optimized contracted Gaussian basis sets for atoms Li to Kr. J. Chem. Phys. 1992, 97, 2571–2577. [Google Scholar] [CrossRef] [Green Version]
  59. Schäfer, A.; Huber, C.; Ahlrichs, R. Fully optimized contracted Gaussian basis sets of triple zeta valence quality for atoms Li to Kr. J. Chem. Phys. 1994, 100, 5829–5835. [Google Scholar] [CrossRef]
  60. Andrae, D.; Häußermann, U.; Dolg, M.; Stoll, H.; Preuß, H. Energy-adjusted ab initio pseudopotentials for the second and third row transition elements. Theor. Chim. Acta 1990, 77, 123–141. [Google Scholar] [CrossRef]
  61. Stoll, H.; Fuentealba, P.; Schwerdtfeger, P.; Flad, J.; Szentpály, L.V.; Preuss, H. Cu and Ag as one-valence-electron atoms: CI results and quadrupole corrections for Cu2, Ag2, CuH, and AgH. J. Chem. Phys. 1984, 81, 2732–2736. [Google Scholar] [CrossRef]
  62. Bergner, A.; Dolg, M.; Küchle, W.; Stoll, H.; Preuß, H. Ab initio energy-adjusted pseudopotentials for elements of groups 13–17. Mol. Phys. 1993, 80, 1431–1441. [Google Scholar] [CrossRef]
  63. Miertuš, S.; Scrocco, E.; Tomasi, J. Electrostatic interaction of a solute with a continuum. A direct utilizaion of AB initio molecular potentials for the prevision of solvent effects. Chem. Phys. 1981, 55, 117–129. [Google Scholar] [CrossRef]
  64. Miertuš, S.; Tomasi, J. Approximate evaluations of the electrostatic free energy and internal energy changes in solution processes. Chem. Phys. 1982, 65, 239–245. [Google Scholar] [CrossRef]
  65. Tomasi, J.; Mennucci, B.; Cammi, R. Quantum mechanical continuum solvation models. Chem. Rev. 2005, 105, 2999–3093. [Google Scholar] [CrossRef] [PubMed]
  66. Adamo, C.; Barone, V. Exchange functionals with improved long-range behavior and adiabatic connection methods without adjustable parameters: The mPW and mPW1PW models. J. Chem. Phys. 1998, 108, 664–675. [Google Scholar] [CrossRef]
  67. Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. Energies, structures, and electronic properties of molecules in solution with the C-PCM solvation model. J. Comput. Chem. 2003, 24, 669–681. [Google Scholar] [CrossRef] [PubMed]
Catalysts 13 00724 sch001 550
Scheme 1. Pd-catalyzed bisarylation of 3-alkylbenzofurans to 3-arylalkyl-2-arylbenzofurans.
Scheme 1. Pd-catalyzed bisarylation of 3-alkylbenzofurans to 3-arylalkyl-2-arylbenzofurans.
Catalysts 13 00724 sch001
Catalysts 13 00724 sch002 550
Scheme 2. Generation of ACS candidates I–V.
Scheme 2. Generation of ACS candidates I–V.
Catalysts 13 00724 sch002
Catalysts 13 00724 g007 550
Figure 7. Mapped sphere of electrostatic potential in the −0.1–0.5 range: (a) I_1, (b) II_1, (c) III_1, (d) IV_1, and (e) V_1.
Figure 7. Mapped sphere of electrostatic potential in the −0.1–0.5 range: (a) I_1, (b) II_1, (c) III_1, (d) IV_1, and (e) V_1.
Catalysts 13 00724 g007
Catalysts 13 00724 g008 550
Figure 8. Correlation diagram of α vs. n0 (avr.) (blue open square) and α vs. |Δq| (dark red closed circle).
Figure 8. Correlation diagram of α vs. n0 (avr.) (blue open square) and α vs. |Δq| (dark red closed circle).
Catalysts 13 00724 g008
Catalysts 13 00724 sch003 550
Scheme 3. Activation strain analysis.
Scheme 3. Activation strain analysis.
Catalysts 13 00724 sch003
Table
Table 1. Maximum free energy barriers (ΔG‡max), largest free energy barrier for any individual step (ΔG‡i,max, kcal/mol), and thermodynamic free energies (ΔG) of C−H activation products for the C(sp2)−H and C(sp3)−H activation processes.
Table 1. Maximum free energy barriers (ΔG‡max), largest free energy barrier for any individual step (ΔG‡i,max, kcal/mol), and thermodynamic free energies (ΔG) of C−H activation products for the C(sp2)−H and C(sp3)−H activation processes.
ΔG‡max (kcal/mol)ΔG‡i,max (kcal/mol)ΔG (kcal/mol)
C(sp2)−HC(sp3)HC(sp2)−HC(sp3)HC(sp2)−HC(sp3)H
I33.351.421.430.723.724.7
II23.828.723.823.812.6−2.6
III32.049.621.734.4−15.17.5
IV10.910.910.910.9−2.9−18.0
V36.036.036.036.0−9.0−58.1
Table
Table 3. Charge distribution of catalyst and substrate moieties of n_1 intermediate states (n = I–V) obtained by NBO population analysis.
Table 3. Charge distribution of catalyst and substrate moieties of n_1 intermediate states (n = I–V) obtained by NBO population analysis.
QaΔq b
CatalystSubstrate
I_1−0.1510.1510.151
II_1−0.3210.3210.321
III_1−0.1720.1720.172
IV_10.2180.7820.782
V_10.1910.8090.809
a Summed charge of the moiety. b The amount of charge transferred from substrate to Pd metal.
Table
Table 4. Activation strain analysis results. Catalyst and substrate distortion energies, metal–arene interaction energies of C(sp2)−H and C(sp3)−H cleavage processes by five different ACSs a.
Table 4. Activation strain analysis results. Catalyst and substrate distortion energies, metal–arene interaction energies of C(sp2)−H and C(sp3)−H cleavage processes by five different ACSs a.
PathEdistI (PdL) bEdistI (ArH) cΔEIEdistII (PdL) dEdistII (ArH) eEintΔE‡EdistI fEdistII g
I (sp2)8.51.66.94.338.5−30.222.710.142.8
I (sp3)8.51.66.98.557.3−35.940.110.165.9
II (sp2)14.25.9−2.42.627.9−42.58.120.130.5
II (sp3)14.25.9−2.48.349.7−63.115.020.158.1
III (sp2)5.21.43.84.239.4−31.518.66.643.6
III (sp3)5.21.43.87.934.8−17.831.56.642.7
IV (sp2)13.216.3−17.36.125.0−71.6−11.129.431.1
IV (sp3)13.216.3−17.310.333.3−85.0−12.029.443.6
V (sp2)30.321.86.43.321.1−69.96.552.124.4
V (sp3)30.321.86.48.337.7−79.818.252.145.9
a All listed energies are electronic energies. b Distortion energy of the catalyst moiety of the pre-C−H cleavage intermediate state. c Distortion energy of the 3-methylbenzofuran moiety of the pre-C−H cleavage intermediate state. d Distortion energy of the catalyst moiety of the C−H cleavage transition state—EdistI (PdL). e Distortion energy of the 3-methylbenzofuran moiety of the C−H cleavage transition state—EdistI (ArH). f EdistI (PdL) + EdistI (ArH). g EdistII (PdL) + EdistII (ArH).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Park, J.E.; Kang, Y.K. Evidence of a Wheland Intermediate in Carboxylate-Assisted C(sp2)−H Activation by Pd(IV) Active Catalyst Species Studied via DFT Calculations. Catalysts 2023, 13, 724. https://doi.org/10.3390/catal13040724

AMA Style

Park JE, Kang YK. Evidence of a Wheland Intermediate in Carboxylate-Assisted C(sp2)−H Activation by Pd(IV) Active Catalyst Species Studied via DFT Calculations. Catalysts. 2023; 13(4):724. https://doi.org/10.3390/catal13040724

Chicago/Turabian Style

Park, Ji Eun, and Youn K. Kang. 2023. "Evidence of a Wheland Intermediate in Carboxylate-Assisted C(sp2)−H Activation by Pd(IV) Active Catalyst Species Studied via DFT Calculations" Catalysts 13, no. 4: 724. https://doi.org/10.3390/catal13040724

APA Style

Park, J. E., & Kang, Y. K. (2023). Evidence of a Wheland Intermediate in Carboxylate-Assisted C(sp2)−H Activation by Pd(IV) Active Catalyst Species Studied via DFT Calculations. Catalysts, 13(4), 724. https://doi.org/10.3390/catal13040724

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop