Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Feb 17;13(5):3201–3210. doi: 10.1021/acscatal.2c05790

Rh(I) Complexes with Hemilabile Thioether-Functionalized NHC Ligands as Catalysts for [2 + 2 + 2] Cycloaddition of 1,5-Bisallenes and Alkynes

Jordi Vila †, Miquel Solà †, Thierry Achard ‡, Stéphane Bellemin-Laponnaz ‡,*, Anna Pla-Quintana †, Anna Roglans †,*
PMCID: PMC9990073  PMID: 36910871

Abstract

graphic file with name cs2c05790_0011.jpg

The [2 + 2 + 2] cycloaddition of 1,5-bisallenes and alkynes under the catalysis of Rh(I) with hemilabile thioether-functionalized N-heterocyclic carbene ligands is described. This protocol effectively provides an entry to different trans-5,6-fused bicyclic systems with two exocyclic double bonds in the cyclohexene ring. The process is totally chemoselective with the two internal double bonds of the 1,5-bisallenes being involved in the cycloaddition. The complete mechanism of this transformation as well as the preference for the trans-fusion over the cis-fusion has been rationalized by density functional theory calculations. The reaction follows a typical [2 + 2 + 2] cycloaddition mechanism. The oxidative addition takes place between the alkyne and one of the allenes and it is when the second allene is inserted into the rhodacyclopentene that the trans-fusion is generated. Remarkably, the hemilabile character of the sulfur atom in the N-heterocyclic carbene ligand modulates the electron density in key intermediates, facilitating the overall transformation.

Keywords: [2 + 2 + 2] cycloaddition, allene, rhodium, N-heterocyclic carbene, DFT calculation

Introduction

Transition-metal-catalyzed [2 + 2 + 2] cycloaddition reactions are a useful tool for the synthesis of six-membered carbo- and heterocyclic compounds in a one-step highly atom economic process.1 The different types of unsaturation that can be involved in these processes open the door to a wide range of cyclic derivatives with different functionalities. Among the unsaturations that can be involved in the [2 + 2 + 2] cycloaddition reaction, allenes are particularly versatile as their two cumulated double bonds can individually participate in the cycloaddition. However, with this increased diversity comes greater difficulty in controlling chemoselectivity.2 As the number of allenes involved in the [2 + 2 + 2] cycloaddition increases, the number of possible regioisomers also increases. After the pioneering studies by Benson and Lindsey3a in 1958 based on the cyclotrimerization of allene by a Ni(0) catalyst, only a couple of studies by Ma and co-workers involve three allenes in a [2 + 2 + 2] cycloaddition to obtain steroid-like scaffolds.3b−3d Initially, the group described a bimolecular [2 + 2 + 2] cycloaddition of bisallenes under rhodium catalysis giving a diene, which then underwent a Diels–Alder reaction with the remaining allene, giving precursors of steroidal structures.3b,3c A complementary approach to this process was the cycloaddition between a 1,5-bisallene and monoallenes (Scheme 1a).3d The reaction took place chemoselectively between an internal double bond and a terminal double bond of the bisallene and the terminal double bond of the monoallene under rhodium catalysis, giving a bicyclic derivative with two exocyclic double bonds. On the other hand, if we consider the participation of two allenes with an alkyne, we can find different processes based on the nature of the allenes and the nature of the transition metal used as the catalyst. Tanaka and co-workers4 described the cross-cyclotrimerization of two monosubstituted allenes with one alkyne to afford 3,6-dimethylenecyclohex-1-ene derivatives resulting from the cycloaddition of the terminal double bond of the allene (Scheme 1b). In contrast, when starting with di- or trisubstituted allenes and using the same catalytic system, a β-hydrogen elimination on the rhodacyclopentene intermediate took place instead of the insertion of the second allene, affording dendralene derivatives. Arai and co-workers5 reported the cycloaddition between two different allenes and an alkyne under nickel catalysis and demonstrated by density functional theory (DFT) calculations that the selectivity of the reaction was affected by steric effects around the π-bonds (Scheme 1c). There are also examples involving an alkene with two allenes that have been reported by both Alexanian and co-workers6 and our own group.7 Alexanian described the stereoselective and enantioselective rhodium-catalyzed [2 + 2 + 2] cycloaddition of ene-allenes and allenoates, affording trans-fused carbocycles with four stereogenic centers (Scheme 1d). The authors postulated an initial oxidative coupling of the two allenes involving the two internal double bonds of both allenes. The alkene was then inserted into the rhodacyclopentane establishing the trans ring fusion in this step. We also studied the stereoselective cycloaddition of linear allene–ene–allene substrates to afford tricyclic systems.7 The Wilkinson complex promoted cycloaddition involving the internal double bonds of both terminal allenes, affording the corresponding exocyclic dienes (Scheme 1e). DFT calculations showed that initial oxidative coupling took place between the internal double bond of one of the allenes and the alkene, affording a cis ring fusion. The internal double bond of the second allene was then inserted delivering this time a trans ring fusion to afford the exocyclic hexadiene after reductive elimination. In the same study, we also reported an analogous reaction on an allene–yne–allene substrate.

Scheme 1. Metal-Catalyzed [2 + 2 + 2] Cycloaddition Reactions of Allenes.

Scheme 1

Open in a new tab

Following on from our interest in allenes in cycloaddition reactions, we became interested in involving 1,5-bisallenes in the [2 + 2 + 2] cycloaddition reactions. This special class of allenes has shown a wide range of reactivities under transition-metal catalysis.8 However, in our initial studies, both by reaction with alkenes and alkynes, we were unable to trigger a [2 + 2 + 2] cycloaddition reaction. Instead, very interesting reactivities were observed (Scheme 2). In the reaction of 1,5-bisallenes with alkenes, dihydroazepine- and dihydrooxepine-fused ring systems were obtained in good yields.9 Further mechanistic study by DFT calculations showed that the reaction took place through a rhodium-catalyzed cycloisomerization/Diels–Alder cascade encompassing oxidative coupling of the rhodium to the central carbon atoms of both allenes (intermediate I) followed by a β-hydride elimination (intermediate II) and reductive elimination of the rhodium to afford a non-isolable cycloheptatriene derivative III, which gave a further Diels–Alder reaction with the alkene (Scheme 2a). In contrast, when an alkyne was used as a third component using the same catalytic system as before, 1,5-bisallenes reacted with two molecules of the alkyne to afford cis-3,4-arylvinyl pyrrolidines and cyclopentanes in a totally diastereoselective manner.10 Here again, DFT calculations allowed us to unveil the course of the reaction, which involves a [2 + 2 + 2] cycloaddition between two molecules of the alkyne and the terminal double bond of one of the two allenes of the bisallene (intermediate IV) followed by a cycloisomerization reaction involving the internal double bond of the second allene unit (intermediate V). Finally, a β-hydride elimination step followed by reductive elimination afforded the final product (Scheme 2b).

Scheme 2. Rh(I)-Catalyzed Cyclization Reactions of 1,5-Bisallenes with Alkenes and Alkynes Performed in Our Group.

Scheme 2

Open in a new tab

Still with the idea of developing a [2 + 2 + 2] cycloaddition reaction of 1,5-bisallenes, we came across the study by the Alexanian group11 on the cycloaddition of ene-allenes and alkenes under nickel catalysis. The study nicely showed that the selectivity can be controlled by fine-tuning the catalytic system, as the use of P(OTol)3 triggered a [2 + 2 + 2] cycloaddition, an alkenylative cyclization occurred when PBu3 was used, and finally, a [2 + 2] cycloaddition of ene-allene occurred when Xantphos was the ligand of choice. In addition, 1,5-bisallenes also follow different reaction pathways under rhodium catalysis depending on the nature of the ligands. Whereas Ma and co-workers3d described a bimolecular [2 + 2 + 2] cycloaddition of the bisallene to afford steroid-type scaffolds when the Wilkinson catalyst was used (Scheme 1a), our group, using DTBM-SEGPHOS as the ligand, described a cascade process encompassing a cycloisomerization of the 1,5-bisallene followed by a selective Diels–Alder homodimerization affording spirocyclic compounds.12 Inspired by the results of Alexanian and our own experience, we thought of testing ligands other than bisphosphines to influence the course of the process. An alternative class of ligands are the N-heterocyclic carbenes (NHCs) that often give reactivities that are complementary to those of phosphines. Several examples of transition-metal-catalyzed [2 + 2 + 2] cycloadditions with NHC ligands are described in the literature. Among these, complexes of cobalt13 and, especially, nickel14 are the most used. Louie’s research group has conducted several studies of [2 + 2 + 2] cycloadditions involving mainly hetero-unsaturations, such as isocyanates, ketenes, nitriles, aldehydes, and ketones, using Ni–NHCs ligands.15 To the best of our knowledge, in the case of rhodium, there are only two previous studies reported by our group in which Rh–NHC complexes, both in homogeneous16 and heterogeneous versions,17 are tested for [2 + 2 + 2] cycloaddition reactions of alkynes.

NHCs are strong σ-donor ligands and hence form strong metal–ligand interactions that can prevent catalyst decomposition resulting in loss of ligands under reaction conditions. Moreover, their simple synthetic procedures and the ability to tune their steric and electronic properties by modifying the nitrogen or backbone substituents make them very attractive for rapid ligand screening. Rhodium(I) bearing monodentate NHCs have been successfully applied to several metal-catalyzed transformations such as hydrothiolation, hydrophosphination, cross-coupling reaction, and dimerization reactions.18,19 Although most NHCs are monodentate ligands, a strong interest has been shown for the past 10 years and still continues today for the chemistry of functionalized NHC carbenes in which a Lewis base moiety (often rooted on: N, P, or O atom) is attached to the strongly bonded imidazolyl ring.20 However, although the S-functionalized N-heterocyclic carbenes are less represented,21 they are also an interesting class of ligands because they can potentially provide an “on and off” dynamic chelating effect for the metal complex during a catalytic cycle. This hemilability has been particularly demonstrated for the thioether function with different metals.22 NHC-SR metal complexes have demonstrated their catalytic activity in several transformations such as hydrosilylation of aldehydes22e and ketones,23 the Suzuki–Miyaura cross-coupling reaction,24 hydrogenation of double bonds,25 the click reaction,26 dehydrogenation of amine,27 and the reaction of amidation.28 Only two examples were reported for NHC-SR rhodium(I) complexes by the group of Poli22 and the group of Lassaletta,29 and none of them were applied to cycloaddition reactions.

In this work, a variety of NHC-imidazole ligand precursors (monodentate and bidentate) were evaluated in the Rh(I)-catalyzed [2 + 2 + 2] cycloaddition of 1,5-bisallene and alkynes. Through extensive screening of various conditions, a catalytic system with an S-functionalized NHC–Rh complex, with a catalyst loading of 5 mol %, was developed, which for the first time promoted the efficient [2 + 2 + 2] cycloaddition of 1,5-bisallene and alkynes.

Results and Discussion

Synthesis of S-Functionalized Imidazolium Salts

All S-functionalized imidazolium salts L2-5 (Figure 1) were synthesized according to our previously reported procedure by direct reaction of the 2-bromoethyl-imidazolium derivative with the corresponding sodium thiolate.22d,27 Considering that the size of the R thioether group can influence its ability to coordinate the rhodium center and therefore its hemilability, groups with different steric hindrances were attached to the sulfur atom in order to evaluate their influence in catalysis.

Figure 1.

Figure 1

Open in a new tab

S-functionalized imidazolium salt precursors used in the cycloaddition of bisallene 1a and alkyne 2a.

The stability of a complex formed with a bidentate ligand depends also on the size of the chelate rings. Our NHC-SR ligands L2-5 possess a flexible ethylene organic backbone, which can generate six-membered chelate rings. To evaluate this effect, the ligand L1 with a shorter N-arm was also prepared according to a literature procedure (Figure 1).22b

Cycloaddition Catalysis

We started our study with the reaction of N-tosyl-tethered bisallene 1a and dimethylacetylenedicarboxylate (DMAD) 2a (Table 1). S-functionalized imidazolium salts L1-L5 (Figure 1) were tested as ligands. In the initial tests, the rhodium complexes were generated in situ by treatment of the corresponding imidazolium salt with tBuOK as a base and the dimeric rhodium complex [Rh(cod)Cl]2. The reaction with L1 afforded three products, which were identified by NMR spectroscopy. In striking contrast to our previous studies with the same substrates (Scheme 2b),10 a mixture of diastereoisomers trans and cis3a in 35% yield with a ratio of about 5:1 was formed by a [2 + 2 + 2] cycloaddition reaction between the two internal double bonds of the two allenes of 1 and the alkyne (entry 1), also in contrast to the case of Ma (Scheme 1a) in which both an internal and an terminal double bond were involved in the reaction.3d The molecular structure and stereochemistry of the major diastereoisomer 3a, which was found to be trans, was confirmed by X-ray crystallographic analysis (Figure 2).30 Of note, 3a bears structural similarities to the product obtained through the intermolecular [2 + 2 + 2] reaction of two allenes and one alkyne described by Tanaka and co-workers (Scheme 1b).4 A second product 4 was also obtained, although with a very low yield. This was formed by our previously described process9 based on a cycloisomerization/Diels–Alder cascade reaction (see for instance Scheme 2a), in which in this case the dienophile is an alkyne (entry 1, Table 1). Interestingly, the same reaction with the monodentate 1,3-bis(2,6-diisopropylphenyl)imidazolium chloride (IPr·HCl) provided lower selectivity and reactivity (entry 2, Table 1). In addition, the monodentate IMes and a bidentate OH-functionalized NHC or a bis-NHC have been less successful (see Scheme S2 in Supporting Information for details).

Table 1. Optimization of the Rhodium(I)–NHC-Catalyzed [2 + 2 + 2] Cycloaddition of 1a and 2a.

graphic file with name cs2c05790_0010.jpg

entry ligand variation of reaction conditions yield (%) 3 (trans/cis)/4a
1 L1   35 (83:17)/4
2 IPr·HCl   23 (78:22)/13
3 L2   51 (86:14)/10
4 L2 80 °C, 2 h 66 (86:14)/13
5 L3 80 °C, 2 h 61 (87:13)/10
6 L4 80 °C, 2 h 74 (89:11)/8
7 L5 80 °C, 2 h 61 (90:10)/7
8 L4 80 °C, 2 h, toluene 73 (93:7)/4
9 L4 80 °C, 2 h, toluene, [1a] = 36 mM 76 (93:7)/0
10 L4 80 °C, 2 h, toluene, 2a (2 equiv) 66 (91:9)/8
11 L4 80 °C, 2 h, toluene, 2a (10 equiv) 77 (92:8)/0
12 RhL4 80 °C, 2 h, toluene, RhL4 (1 mol %) [1a] = 36 mM 31 (91:9)/0
13 RhL4 80 °C, 2 h, toluene, RhL4 (2.5 mol %) [1a] = 36 mM 55 (91:9)/0
14 RhL4 80 °C, 2 h, toluene, RhL4 (5 mol %) [1a] = 36 mM 82 (92:8)/0
15 RhL4 80 °C, 2 h, toluene, RhL4 (10 mol %), [1a] = 36 mM 81 (92:8)/0
Open in a new tab
a

Yields and ratios calculated by 1H NMR from the reaction crude.

Figure 2.

Figure 2

Open in a new tab

ORTEP representation of 3a-trans at 50% of the probability level (CCDC 2209948).

Encouraged by the selective formation of [2 + 2 + 2] cycloadducts 3 using this new catalytic system, we explored the other NHC-SR ligands and tested different reaction conditions to improve the efficiency and selectivity of the cycloaddition. Using imidazolium salt L2 with longer N-wing chain the yield of 3 improved to 51%, but the selectivity decreased (entry 3, Table 1). As a series of unidentified compounds were observed in the crude mixture in these experiments, the reaction was performed again by decreasing the temperature to 80 °C and the yield of 3 improved to 66% (entry 4, Table 1). Under these conditions, we proceeded to test the less sterically demanding NHC-SR ligands L3-5 (entries 5–7, Table 1). The use of ligand L4 provided the best yield and selectivity (entry 6, Table 1) and was thus selected for further tests. Performing the reaction in the absence of DCE led to an increase in the trans/cis ratio and a decrease in the formation of 4 (entry 8, Table 1). To our delight, formation of byproduct 4 could be fully suppressed by increasing the concentration of 1a from 18 to 36 mM (entry 9, Table 1). Increasing and decreasing the amount of 2a (2 equiv and 10 equiv, respectively) did not improve the results (entries 10–11, Table 1). Once the ligand was optimized, we proceeded to prepare a preformed and well-defined Rh complex RhL4, which we used as a catalyst.

The NHC–rhodium(I) complex RhL4 was prepared in a one-step process by the reaction of 2 equiv of imidazolium salt L4 with 1 equiv of [Rh(cod)Cl]2 in the presence of 2.2 equiv of tBuOK in THF (Scheme 3) following a procedure previously described by us.31 The corresponding N-heterocyclic carbene complex RhL4 was obtained in a high yield (above 95%) as a yellow shiny solid, which was fully characterized. In the presence of the halogen onto the metal center, no direct evidence of the coordination of the sulfur atom to the metal center was noticed by 1H NMR.32

Scheme 3. Synthesis of Rhodium Complex RhL4.

Scheme 3

Open in a new tab

The molecular structure of RhL4 was confirmed by X-ray diffraction studies (Figure 3).30 The rhodium–carbene bond distance, 2.033(4) Å, is within the range reported for [RhCl(cod)(imidazol-2-ylidene)] complexes.32,33 The Rh–C bonds located in a relative trans position to the carbene are significantly elongated (mean Rh–C 2.217 Å) compared to those in trans to the bromide ligand (mean Rh–C 2.1055 Å) due to the strong trans influence of the NHC ligand. The imidazole-2-ylidene ring is almost perpendicular to the coordination plane of the rhodium center (dihedral angle 88.64°), and this geometry is consistent with the 1H NMR spectra.

Figure 3.

Figure 3

Open in a new tab

ORTEP representation of RhL4 at 50% of the probability level (CCDC: 2213801). Selected bond lengths [Å] and angles [deg]: C(1)-Rh(1), 2.033(4); Rh(1)-Br(1), 2.5026(6); Rh(1)-C(17), 2.102(4); Rh(1)-C(18), 2.109(4); Rh(1)-C(21), 2.207(4); Rh(1)-C(22), 2.227(5); C(1)-Rh(1)-Br(1), 86.5(1); C(21)-Rh(1)-Br(1), 91.8(1); C(22)-Rh(1)-Br(1), 95.4(1); C(18)-Rh(1)-C(1), 94.2(2); C(17)-Rh(1)-C(1), 90.6(2).

Different catalytic amounts of RhL4 were tested, from 1 to 10 mol % (entries 12–15, Table 1). Quantities below 5 mol % reduced the yield of 3a (entries 12–13, Table 1) and the use of 10 mol % of RhL4 (entry 15, Table 1) did not improve the results compared to the use of 5 mol % (entry 14, Table 1). It should be noted that in no case the formation of byproduct 4 was observed. Therefore, the optimal reaction conditions for further studies were defined as 1a ([1a] = 36 mM), 2a (5 equiv), RhL4 (5 mol %) in toluene at 80 °C for 2 h (entry 14, Table 1). In addition, two blank tests were performed. The reaction was first carried out in the presence of the rhodium dimer [Rh(cod)Cl]2 excluding the NHC ligand and second in the absence of both the transition metal and the ligand. The reaction did not work in either case, and only the two starting products 1a and 2a were recovered.

It should be noted that bisphosphine ligands such as BINAP, Tol-BINAP, BIPHEP, DPEphos, Xantphos, and XPhos were not able to provide a defined product upon reaction of 1a and 2a as investigated in a previous study of our group.10

The scope of the reaction was then evaluated (Scheme 4). Methyl, ethyl, and tert-butyl acetylenedicarboxylates 2a, 2b, and 2c were tested in the cycloaddition affording excellent yields and high trans/cis ratios34 of the corresponding cycloadducts 3a–3c. However, terminal alkynes, such as monoacetylenecarboxylates and phenylacetylene analogues, did not react under these conditions. The nature of the substituents at the phenyl ring of the sulfonamide tether in bisallene 1 was then explored. The reaction proceeded efficiently with both electron-donating (3d) and electron-withdrawing groups (3e), as well as with substituents at the ortho position of the phenyl ring (3f, 3g). A bisallene bearing the 5-methyl-2-pyridinesulfonyl group provided 3h in a 60% yield, indicating that the presence of a potentially coordinating nitrogen atom did not poison the catalyst. Sulfonamide tethers with aliphatic substitution (tert-butyl and trimethylsilylethyl) were also efficient, delivering cycloadducts 3i and 3j in 79 and 58% yields, respectively. Changing the sulfonamide protecting group of the nitrogen tether to a carbamate (N-Boc bisallene), the reaction took place, although cycloadduct 3k was obtained in a moderate yield and with low diastereoselectivity. Bisallenes with a carbonyl group attached to the quaternary carbon atom of the tether also participated in the cycloaddition, affording 3l and 3m with 61 and 81% yields, respectively. When the carbonyl group was substituted for arylsulfonyl groups, lower yields of the cycloadducts 3n and 3o were obtained, but the diastereoisomeric ratios were better. In the case of 3o, an inseparable mixture of diastereoisomers was obtained due to the chiral center in the tether making it difficult to determine the diastereoisomeric ratio. Finally, oxygen-tethered bisallene participated in the process affording cycloadduct 3p with a 61% yield and an excellent diastereoisomeric ratio.

Scheme 4. Scope of the [2 + 2 + 2] Cycloaddition Reaction of Bisallenes 1 and Alkynes 2.

Scheme 4

Open in a new tab

Computational Analysis

Intrigued by the role of the hemilabile NHC-SR ligand in the chemoselectivity of the rhodium-catalyzed cycloaddition of 1,5-bisallenes and alkynes, we performed DFT calculations on the entire reaction. The Gibbs energy profile computed at 353.15 K and 1 atm with the ωB97X-D/cc-pVTZ-PP/SMD(toluene)//B3LYP-D3/cc-pVDZ-PP method is depicted in Figure 4, and the molecular structures of all intermediates and transition states (TSs) are available in the Supporting Information (see Supporting Information for a complete description of the computational methods and Table S1 for a justification of the density functional employed).

Figure 4.

Figure 4

Open in a new tab

Gibbs energy profile (in kcal·mol–1) for the cycloaddition of 1,5-bisallene 1a and dimethylacetylenedicarboxylate 2a leading to 3a (E = CO2Me).

The reaction starts with the coordination equilibrium of 1,5-bisallene and DMAD with the rhodium catalyst to give Rh(I) 16 e– square planar complexes A1 and A1′ (Figure 4) (see Figure S1 for the whole set of coordination complexes). The coordination of the two internal double bonds of the 1,5-bisallene (A1) is the most exergonic process, releasing 6.6 kcal·mol–1. For the upcoming oxidative cyclometallation, all possible orientations have been evaluated along with the dynamic chelating effect of the sulfur (see Figures S2 and S3), resulting in two possible TSs for each orientation: one with the sulfur chelating the rhodium (TSs superindexed with S in Figure 4) and the other without such a chelation.35 Since A1 and A1′ (and the rest of the possible coordination complexes) are in equilibrium, according to the Curtin–Hammett principle,36 the major rhodacyclopentene or rhodacyclopentane intermediate formed is the one generated through the lowest in energy TS, in this case TSS(A1′A2). This TS involves an oxidative cyclometallation of the central carbon of one of the allenes of the 1,5-bisallene 1a with DMAD 2a with the sulfur chelating the rhodium. Formation of rhodacyclopentene intermediate A2 takes place with an affordable barrier of 16.7 kcal·mol–1 [A1 to TSS(A1′A2), being 7.6 kcal·mol–1 lower than its non-chelating counterpart (TS(A1′A2), ΔG‡ = 24.3 kcal·mol–1]. This step A1 ⇌ A1′ → A2 is exergonic by 31.6 kcal·mol–1. Alternative ways to generate A2 through other coordination complexes including A1 have higher energy barriers (see Figures 4 and S2 and S3 in Supporting Information). For this path, the sulfur-assisted TS [TSS(A1′A2)] is lower in energy than its analogue in which the sulfur is not coordinated to the rhodium [TS(A1′A2)] due to the S electron-donating character, which adds electronic density to the rhodium and facilitates its oxidation [NPA charges on Rh are −0.101 e in TS(A1′A2) and −0.443 e in TSS(A1′A2)]. However, in the oxidative cyclometallation from A1, coordination of S leads to a pyramidalization of the rest of the ligands coordinated to Rh. Pyramidalization brings the ligands closer to one another and, consequently, the two coordinated double bonds in the 1,5-bisallene become perpendicularly arranged to one another. This perpendicular arrangement is destabilizing37 and the trend is inverted. As a result, the non-chelated TS(A1A2) is lower in energy than TSS(A1A2) by 5.3 kcal·mol–1. A2 is a Rh(III) 18 e– complex and exhibits an octahedral geometry in which the three carbons from the reacted allene are η3-coordinated to the rhodium (dRh–C = 2.217, 2.129, and 2.218 Å). This type of π-allyl metallacycle intermediates have previously been postulated in cycloaddition reactions.38 From this point, A2 needs to rearrange to set a coordination position free for the second allene unit, giving either A3trans or A3cis at the cost of 18.0 and 24.0 kcal·mol–1, respectively. For the formation of the trans-fused rhodabicyclo intermediate A4trans through A3trans, the insertion (via the Schore mechanism39) of the internal double bond of the second allene takes place in the Rh–C sp3 bond through TSS(A3A4)trans. This process A2 → A4trans has a total Gibbs energy barrier of 24.4 kcal·mol–1 and is exergonic by 3.4 kcal·mol–1. In contrast, for the formation of A4cis, the insertion is found to occur in the Rh–C sp2 bond, surpassing a Gibbs energy barrier of 27.0 kcal·mol–1 [TSS(A3A4)cis] and releasing 12.4 kcal·mol–1. The coordination of a second DMAD unit to A2, which would lead to our previously reported cis-3,4-arylvinyl pyrrolidine derivative (Scheme 2b),10 was also considered. However, it was found to be disfavored at the experimental concentration of DMAD (see Figure S6 in Supporting Information).

For the final reductive elimination step, the S-adamantyl functionality is dissociated from the rhodium center to remove electronic density, drastically reducing the Gibbs energy barriers by 8.1 kcal·mol–1 [TS(A4A5)trans vs TSS(A4A5)trans] and 3.9 kcal·mol–1 [TS(A4A5)cis vs TSS(A4A5)cis]. Finally, ligand exchange from A5trans and A5cis releases 3a-trans and 3a-cis and gives A1 to restart the catalytic cycle. The formation of byproduct 4 (Table 1) was also evaluated computationally, and the results account for its formation in minor quantities at low concentrations of 2a (see Scheme S6 and Figures S4 and S5 in Supporting Information for the complete discussion).

In summary, the reaction follows the typical [2 + 2 + 2] cycloaddition mechanism40 and has an overall reaction energy of −98.6 kcal·mol–1 (ΔG = G3a – [G1a + G2a]), almost identical for both diastereoisomers. For 3a-trans, the energetic span between the turnover-frequency-determining intermediate (TDI, A2) and the turnover-frequency-determining TS (TDTS, TS(A3A4)trans) is 24.4 kcal·mol–1, and for 3a-cis, the energetic span between TDI (A4cis) and the TDTS [TS(A4A5)cis] is 29.0 kcal·mol–1.41 The allene–allene oxidative cyclometallation is found to be much higher in energy than the allene–alkyne oxidative cyclometallation (ΔΔG‡ = 16.0 kcal·mol–1), indicating that the ring fusion stereochemistry comes from the latter insertion6b of the second allene, in which the insertion leading to the trans isomer [TS(A3A4)trans] is preferred over the cis [TS(A3A4)cis] by 2.6 kcal·mol–1. This difference in energy is translated into a 98:2 trans/cis ratio using the Eyring equation, in perfect agreement with the experimental data.

In conclusion, the hemilability of NHC-SR ligands enabled an efficient rhodium-catalyzed [2 + 2 + 2] cycloaddition of 1,5-bisallenes and alkynes. A number of N-, C-, and O-tethered 1,5-bisallenes as well as variously substituted alkynes were successfully used in the reaction. The methodology developed gives access to bicyclic 3,6-dimethylenecyclohex-1-ene derivatives. Importantly, the exocyclic alkenes in the product scaffold provide ample opportunities of synthetic manipulation to build more complex molecules. A mechanistic investigation by means of DFT calculations has been carried out to unravel that a canonical [2 + 2 + 2] reaction manifold accounts for the observed transformation and that the hemilabile character of the sulfur in the ligand precisely modulates the electron density in key intermediates and facilitates the overall transformation. This hemilability in the catalytic system is expected to be useful for further development of demanding cycloaddition reactions.

Acknowledgments

We are grateful for financial support from the Ministerio de Ciencia e Innovación (PID2020-113711GB-I00 MCIN/AEI/10.13039/501100011033), the Generalitat de Catalunya (project 2017-SGR-39 and 2021-SGR-623), and the UdG for an IF predoctoral grant to J.V. We also thank Corinne Bailly and Dr. Lydia Karmazin from the University of Strasbourg and Xavier Fontrodona from the University of Girona for the X-ray structural determinations. This work of the Interdisciplinary Thematic Institute QMat, as part of the ITI 2021 2028 program of the University of Strasbourg, CNRS and Inserm, was supported by IdEx Unistra (ANR 10 IDEX 0002) and by SFRI STRAT’US project (ANR 20 SFRI 0012) under the framework of the French Investments for the Future Program. Open Access funding provided thanks to the CRUE-CSIC agreement with ACS.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acscatal.2c05790.

  • Experimental procedures, compound characterization data including NMR spectra for all new compounds, and computational data (PDF)

  • Crystallographic data of 3a (CIF)

  • Crystallographic data of RhL4 (CIF)

  • Crystallographic data of [RhL4][PF6] (CIF)

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.

Supplementary Material

cs2c05790_si_001.pdf (7.1MB, pdf)
cs2c05790_si_002.cif (526.4KB, cif)
cs2c05790_si_003.cif (1.5MB, cif)
cs2c05790_si_004.cif (6MB, cif)

References

  1. a Tanaka K.Transition-Metal-Mediated Aromatic Ring Construction, 1st ed.; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2013. [Google Scholar]; b Stará I. G.; Starý I. Helically chiral aromatics: the synthesis of helicenes by [2 + 2 + 2] cycloisomerization of π-electron systems. Acc. Chem. Res. 2020, 53, 144–158. 10.1021/acs.accounts.9b00364. [DOI] [PubMed] [Google Scholar]
  2. a Lledó A.; Pla-Quintana A.; Roglans A. Allenes, versatile unsaturated motifs in transition-metal-catalysed [2+2+2] cycloaddition reactions. Chem. Soc. Rev. 2016, 45, 2010–2023. 10.1039/c5cs00535c. [DOI] [PubMed] [Google Scholar]; b Domínguez G.; Pérez-Castells J. Alkenes in [2+2+2] cycloadditions. Chem.—Eur. J. 2016, 22, 6720–6739. 10.1002/chem.201504987. [DOI] [PubMed] [Google Scholar]
  3. a Benson R. E.; Lindsey R. V. Chemistry of allene. I. Cyclopolymerization. Synthesis and chemistry of 1,2,4 and 1,3,5-trimethylenecyclohexane and 1,3,5,7-tetramethylenecyclooctane. J. Am. Chem. Soc. 1959, 81, 4247–4250. 10.1021/ja01525a033. [DOI] [Google Scholar]; b Ma S.; Lu P.; Lu L.; Hou H.; Wei J.; He Q.; Gu Z.; Jiang X.; Jin X. What can a metal catalyst do with allenes? One-step formation of steroid scaffolds from readily available starting materials. Angew. Chem., Int. Ed. 2005, 44, 5275–5278. 10.1002/anie.200501350. [DOI] [PubMed] [Google Scholar]; c Ma S.; Lu L. Rh(I)-catalyzed bimolecular cyclization between two different 1,5-bisallenes: a combinatorial one-step approach to heterosteroids and mechanistic implications. Chem.—Asian J. 2007, 2, 199–204. 10.1002/asia.200600274. [DOI] [PubMed] [Google Scholar]; d Lu P.; Ma S. Rh-catalyzed triple allene approach to bicycle[4.4.0]decene derivatives and its application for the stepwise synthesis of steroid-like tetracyclic skeletons. Org. Lett. 2007, 9, 5319–5321. 10.1021/ol702557s. [DOI] [PubMed] [Google Scholar]
  4. Sakashita K.; Shibata Y.; Tanaka K. Rhodium-catalyzed cross-cyclotrimerization and dimerization of allenes with alkynes. Angew. Chem., Int. Ed. 2016, 55, 6753–6757. 10.1002/anie.201602155. [DOI] [PubMed] [Google Scholar]
  5. Arai S.; Izaki A.; Amako Y.; Nakajima M.; Uchiyama M.; Nishida A. Regioselective [2+2+2] cycloaddition reaction using allene-ynes with simple allenes under nickel catalysis. Adv. Synth. Catal. 2019, 361, 4882–4887. 10.1002/adsc.201900719. [DOI] [Google Scholar]
  6. a Brusoe A. T.; Alexanian E. J. Rhodium(I)-catalyzed ene-allene-allene [2+2+2] cycloadditions: stereoselective synthesis of complex trans-fused carbocycles. Angew. Chem., Int. Ed. 2011, 50, 6596–6600. 10.1002/anie.201100272. [DOI] [PubMed] [Google Scholar]; b Brusoe A. T.; Edwankar R. V.; Alexanian E. J. Enantioselective intermolecular [2+2+2] cycloadditions of ene-allenes with allenoates. Org. Lett. 2012, 14, 6096–6099. 10.1021/ol303024q. [DOI] [PubMed] [Google Scholar]
  7. Haraburda E.; Torres Ò.; Parella T.; Solà M.; Pla-Quintana A. Stereoselective rhodium-catalyzed [2+2+2] cycloaddition of linear allene-ene/yne-allene substrates: reactivity and theoretical mechanistic studies. Chem.—Eur. J. 2014, 20, 5034–5045. 10.1002/chem.201304463. [DOI] [PubMed] [Google Scholar]
  8. a Alcaide B.; Almendros P.; Aragoncillo C. Cyclization reactions of bis(allenes) for the synthesis of polycarbo(hetero)cycles. Chem. Soc. Rev. 2014, 43, 3106–3135. 10.1039/c3cs60462d. [DOI] [PubMed] [Google Scholar]; b Chen G.; Jiang X.; Fu C.; Ma S. The diversified reactivities of 1,5-bisallenes. Chem. Lett. 2010, 39, 78–81. 10.1246/cl.2010.78. [DOI] [Google Scholar]
  9. a Artigas A.; Vila J.; Lledó A.; Solà M.; Pla-Quintana A.; Roglans A. A Rh-catalyzed cycloisomerization/Diels-Alder cascade reaction of 1,5-bisallene for the synthesis of polycyclic heterocycles. Org. Lett. 2019, 21, 6608–6613. 10.1021/acs.orglett.9b02032. [DOI] [PubMed] [Google Scholar]; b Artigas A.; Castanyer C.; Roig N.; Lledó A.; Solà M.; Pla-Quintana A.; Roglans A. Synthesis of fused dihydroazepine derivatives of fullerenes by a Rh-catalyzed cascade process. Adv. Synth. Catal. 2021, 363, 3835–3844. 10.1002/adsc.202100644. [DOI] [Google Scholar]
  10. Vila J.; Vinardell R.; Solà M.; Pla-Quintana A.; Roglans A. A Rh(I)-catalyzed cascade cyclization of 1,5-bisallenes and alkynes for the formation of cis-3,4-arylvinyl pyrrolidines and cyclopentanes. Adv. Synth. Catal. 2022, 364, 206–217. 10.1002/adsc.202100934. [DOI] [Google Scholar]
  11. Noucti N. N.; Alexanian E. J. Stereoselective nickel-catalyzed [2+2+2] cycloadditions and alkenylative cyclizations of ene-allenes and alkenes. Angew. Chem., Int. Ed. 2013, 52, 8424–8427. 10.1002/anie.201303211. [DOI] [PubMed] [Google Scholar]
  12. Vila J.; Solà M.; Pla-Quintana A.; Roglans A. Highly selective Synthesis of seven-membered azaspiro compounds by a Rh(I)-catalyzed cycloisomerization/Diels-Alder cascade of 1,5-bisallenes. J. Org. Chem. 2022, 87, 5279–5286. 10.1021/acs.joc.2c00065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. a Saino N.; Kogure D.; Okamoto S. Intramolecular cyclotrimerization of triynes catalyzed by N-heterocyclic carbene-CoCl2/Zn or FeCl3/Zn. Org. Lett. 2005, 7, 3065–3067. 10.1021/ol051048q. [DOI] [PubMed] [Google Scholar]; b Geny A.; Gaudrel S.; Slowinski F.; Amatore M.; Chouraqui G.; Malacria M.; Aubert C.; Gandon V. A straightforward procedure for the [2+2+2] cycloaddition of enediynes. Adv. Synth. Catal. 2009, 351, 271–275. 10.1002/adsc.200800646. [DOI] [Google Scholar]
  14. a Hoshimoto Y.; Ohata T.; Ohashi M.; Ogoshi S. Nickel-catalyzed synthesis of N-aryl-1,2-dihydropyridines by [2+2+2] cycloaddition of imines with alkynes through T-shaped 14-electron aza-nickelacycle key intermediates. Chem.—Eur. J. 2014, 20, 4105–4110. 10.1002/chem.201304830. [DOI] [PubMed] [Google Scholar]; b Zhao J.-P.; Chan S.-C.; Ho C.-Y. Substituted 1,3-cyclohexadiene synthesis by NHC-Nickel(0) catalyzed [2+2+2] cycloaddition of 1,n-enyne. Tetrahedron 2015, 71, 4426–4431. 10.1016/j.tet.2015.05.016. [DOI] [Google Scholar]; c Xue F.; Loh Y. K.; Song X.; Teo W. J.; Chua J. Y. D.; Zhao J.; Hor T. S. A. Nickel-catalyzed facile [2+2+2] cyclotrimerization of unactivated internal alkynes to polysubstituted benzenes. Chem.—Asian J. 2017, 12, 168–173. 10.1002/asia.201601469. [DOI] [PubMed] [Google Scholar]; d Sánchez I. G.; Sámal M.; Nejedlý J.; Karras M.; Klívar J.; Rybácek J.; Budesínský M.; Bednárová L.; Seidlerová B.; Stará I. G.; Starý I. Oxahelicene NHC ligands in the asymmetric synthesis of nonracemic helicenes. Chem. Commun. 2017, 53, 4370–4373. 10.1039/c7cc00781g. [DOI] [PubMed] [Google Scholar]
  15. Thakur A.; Louie J. Advances in Nickel-catalyzed cycloaddition reactions to construct carbocycles and heterocycles. Acc. Chem. Res. 2015, 48, 2354–2365. and references cited therein 10.1021/acs.accounts.5b00054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Roglans A.; González I.; Pla-Quintana A. Rhodium N-heterocyclic carbene complexes as effective catalysts for [2+2+2] cycloaddition reactions. Synlett 2009, 2009, 2844–2848. 10.1055/s-0029-1217982. [DOI] [Google Scholar]
  17. Fernández M.; Ferré M.; Pla-Quintana A.; Parella T.; Pleixats R.; Roglans A. Rhodium-NHC hybrid silica materials as recyclable catalysts for [2+2+2] cycloaddition reactions of alkynes. Eur. J. Org. Chem. 2014, 2014, 6242–6251. 10.1002/ejoc.201402649. [DOI] [Google Scholar]
  18. For a recent review see:Lee J.; Hahm H.; Kwak J.; Kim M. New aspects of recently developed rhodium(N-heterocyclic carbene)-catalyzed organic transformations. Adv. Synth. Catal. 2019, 361, 1479–1499. 10.1002/adsc.201801085. [DOI] [Google Scholar]
  19. César V.; Gade L. H.; Bellemin-Laponnaz S.. NHC-Cobalt, -Rhodium, and -Iridium Complexes in Catalysis. In N-Heterocyclic Carbenes, 2nd ed.; Diez-Gonzales S., Ed.; RSC Catalysis Series, 2017; pp 302–335. [Google Scholar]
  20. a Kühl O. The chemistry of functionalised N-heterocyclic carbenes. Chem. Soc. Rev. 2007, 36, 592–607. 10.1039/b603765h. [DOI] [PubMed] [Google Scholar]; b Zhang W.-H.; Chien S. W.; Hor T. S. A. Recent advances in metal catalysts with hybrid ligands. Coord. Chem. Rev. 2011, 255, 1991–2024. 10.1016/j.ccr.2011.05.018. [DOI] [Google Scholar]; c Hameury S.; de Frémont P.; Braunstein P. Metal complexes with oxygen-functionalized NHC ligands: synthesis and applications. Chem. Soc. Rev. 2017, 46, 632–733. 10.1039/c6cs00499g. [DOI] [PubMed] [Google Scholar]; d Peris E. Smart N-heterocyclic carbene ligands in catalysis. Chem. Rev. 2018, 118, 9988–10031. 10.1021/acs.chemrev.6b00695. [DOI] [PubMed] [Google Scholar]; e Neshat A.; Mastrorilli P.; Mobarakeh A. M. Recent advances in catalysis involving bidentate N-heterocyclic carbene ligands. Molecules 2022, 27, 95. 10.3390/molecules27010095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Fliedel C.; Braunstein P. Recent advances in S-functionalized N-heterocyclic carbene ligands: From the synthesis of azolium salts and metal complexes to applications. J. Organomet. Chem. 2014, 751, 286–300. 10.1016/j.jorganchem.2013.09.036. [DOI] [Google Scholar]
  22. a Huynh H. V.; Yeo C. H.; Tan G. K. Hemilabile behavior of a thioether-functionalized N-heterocyclic carbene ligand. Chem. Commun. 2006, 3833–3835. 10.1039/b608325k. [DOI] [PubMed] [Google Scholar]; b Fliedel C.; Schnee G.; Braunstein P. Versatile coordination modes of novel hemilabile S-NHC ligands. Dalton Trans. 2009, 2474–2476. 10.1039/b902314n. [DOI] [PubMed] [Google Scholar]; c Huynh H. V.; Yeo C. H.; Chew Y. X. Syntheses, structures, and catalytic activities of hemilabile thioether-functionalized NHC complexes. Organometallics 2010, 29, 1479–1486. 10.1021/om9010966. [DOI] [Google Scholar]; d Egly J.; Bouché M.; Chen W.; Maisse-François A.; Achard T.; Bellemin-Laponnaz S. Synthesis, structural characterization and anti-proliferative activity of (κ1-C)- and (κ2-C,S)-PtII complexes bearing thioether-functionalized N-heterocyclic carbenes. Eur. J. Inorg. Chem. 2018, 2018, 159–166. 10.1002/ejic.201701183. [DOI] [Google Scholar]; e Ulm F.; Poblador-Bahamonde A. I.; Choppin S.; Bellemin-Laponnaz S.; Chetcuti M. J.; Achard T.; Ritleng V. Synthesis, characterization, and catalytic application in aldehyde hydrosilylation of half-sandwich nickel complexes bearing (κ1-C)- and hemilabile (κ2-C,S)-thioether-functionalised NHC ligands. Dalton Trans. 2018, 47, 17134–17145. 10.1039/c8dt03882a. [DOI] [PubMed] [Google Scholar]; f De Marco R.; Dal Grande M.; Baron M.; Orian L.; Graiff C.; Achard T.; Bellemin-Laponnaz S.; Pöthig A.; Tubaro C. Synthesis, structural characterization and antiproliferative activity of Gold(I) and Gold(III) complexes bearing thioether-functionalized N-heterocyclic carbenes. Eur. J. Inorg. Chem. 2021, 2021, 4196–4206. 10.1002/ejic.202100495. [DOI] [Google Scholar]
  23. Wolf J.; Labande A.; Daran J. C.; Poli R. Nickel(II), Palladium(II) and Rhodium(I) complexes of new NHC-thioether ligands: efficient ketone hydrosilylation catalysis by a cationic Rh complex. Eur. J. Inorg. Chem. 2007, 2007, 5069–5079. 10.1002/ejic.200700670. [DOI] [Google Scholar]
  24. Sharma K. N.; Satrawala N.; Joshi R. K. Thioether–NHC-ligated PdII complex for crafting a filtration-free magnetically retrievable catalyst for Suzuki–Miyaura coupling in water. Eur. J. Inorg. Chem. 2018, 2018, 1743–1751. 10.1002/ejic.201800209. [DOI] [Google Scholar]
  25. Gandolfi C.; Heckenroth M.; Neels A.; Laurenczy G.; Albrecht M. Chelating NHC Ruthenium(II) complexes as robust homogeneous hydrogenation catalysts. Organometallics 2009, 28, 5112–5121. 10.1021/om900356w. [DOI] [Google Scholar]
  26. Szadkowska A.; Zaorska E.; Staszko S.; Pawłowski R.; Trzybiński D.; Woźniak K. Synthesis, structural characterization and catalytic activities of sulfur-functionalized NHC–copper(I) complexes. Eur. J. Org. Chem. 2017, 2017, 4074–4084. 10.1002/ejoc.201700523. [DOI] [Google Scholar]
  27. Chen W. G.; Egly J.; Poblador-Bahamonde A. I.; Maisse-Francois A.; Bellemin-Laponnaz S.; Achard T. Synthesis, characterization, catalytic and biological application of half-sandwich ruthenium complexes bearing hemilabile (κ2-C,S)-thioether-functionalised NHC ligands. Dalton Trans. 2020, 49, 3243–3252. 10.1039/c9dt04825a. [DOI] [PubMed] [Google Scholar]
  28. Egly J.; Chen W. H.; Maisse-François A.; Bellemin-Laponnaz S.; Achard T. Half-sandwich ruthenium complexes bearing hemilabile k2-(C,S)-thioether-functionalized NHC ligands : application to amide synthesis from alcohol and amine. Eur. J. Inorg. Chem. 2022, 2022, e202101033 10.1002/ejic.202101033. [DOI] [Google Scholar]
  29. Ros A.; Alcarazo M.; Monge D.; Álvarez E.; Fernández R.; Lassaletta J. M. Stereoselective synthesis of cationic heterobidentate C(NHC)/SR rhodium(I) complexes using stereodirecting N,N-dialkylamino groups. Tetrahedron: Asymmetry 2010, 21, 1557–1562. 10.1016/j.tetasy.2010.04.041. [DOI] [Google Scholar]
  30. CCDC-2209948, CCDC-2213801, and CCDC-2232591 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_-request/cif.
  31. César V.; Bellemin-Laponnaz S.; Wadepohl H.; Gade L. H. Designing the “search pathway” in the development of a new class of highly efficient stereoselective hydrosilylation catalysts. Chem.—Eur. J. 2005, 11, 2862–2873. 10.1002/chem.200500132. [DOI] [PubMed] [Google Scholar]
  32. The 1H NMR spectra showed sharp resonances at room temperature and the splitting observed as two multiplets for both CH2–N and CH2–S protons at 5.00 and 4.11 and 3.44 and 2.69 ppm respectively is only due to the lack of an effective plane of symmetry in the molecules due to the impeded rotation already observed for NHC-Rh complexes. For similar behaviour see:Jiménez M. V.; Pérez-Torrente J. J.; Bartolomé M. I.; Gierz V.; Lahoz F. J.; Oro L. A. Rhodium(I) complexes with hemilabile N-heterocyclic carbenes: efficient alkyne hydrosilylation catalysts. Organometallics 2008, 27, 224–234. 10.1021/om700728a. [DOI] [Google Scholar]
  33. a Peñafiel I.; Pastor I. M.; Yus M.; Esteruelas M. A.; Oliván M. Preparation, hydrogen bonds, and catalytic activity in metal-promoted addition of arylboronic acids to enones of a rhodium complex containing an NHC ligand with an alcohol function. Organometallics 2012, 31, 6154–6161. 10.1021/om300498e. [DOI] [Google Scholar]; b Yu X. Y.; Patrick B. O.; James B. R. New Rhodium(I) carbene complexes from carbene transfer reactions. Organometallics 2006, 25, 2359–2363. 10.1021/om060030w. [DOI] [Google Scholar]; c Warsink S.; Venter J. A.; Roodt A. NHC-amide donor ligands in rhodium complexes: Syntheses and characterisation. J. Organomet. Chem. 2015, 775, 195–201. 10.1016/j.jorganchem.2014.10.020. [DOI] [Google Scholar]
  34. For sulfonamide cases, 3a–3k, diastereoisomeric ratios were calculated from the reaction crude by NMR spectroscopy. After purification by column chromatography, the corresponding cis diastereoisomer was partially eluted with the cyclotrimerized [2 + 2 + 2] adduct of the acetylene dicarboxylate 2, making it difficult to measure the diastereoisomeric ratio accurately.
  35. As an experimental evidence of ligand hemilability, bromide abstraction on the neutral Rh complex RhL4 generated a cationic square planar complex with NHC acting as a bidentate ligand, as deduced from X-ray diffraction experiments (CCDC-2232591) (see Supporting Information for details).
  36. Seeman J. I. Effect of conformational change on reactivity in organic chemistry. Evaluations, applications, and extensions of Curtin-Hammett Winstein-Holness kinetics. Chem. Rev. 1983, 83, 83–134. 10.1021/cr00054a001. [DOI] [Google Scholar]
  37. Dachs A.; Osuna S.; Roglans A.; Solà M. A Density functional study of the [2+2+2] cyclotrimerizaton of acetylene catalyzed by the Wilkinson’s catalyst, RhCl(PPh3)3. Organometallics 2010, 29, 562–569. 10.1021/om900836b. [DOI] [Google Scholar]
  38. a Inglesby P. A.; Bacsa J.; Negru D. E.; Evans P. A. The isolation and characterization of a rhodacycle intermediate implicated in metal-catalyzed reactions of alkylidenecyclopropanes. Angew. Chem., Int. Ed. 2014, 53, 3952–3956. 10.1002/anie.201310232. [DOI] [PubMed] [Google Scholar]; b Teng Q.; Mao W.; Chen D.; Wang Z.; Tung C.-H.; Xu Z. Asymmetric synthesis of a fused tricyclic hydronaphthofuran scaffold by desymmetric [2+2+2] cycloaddition. Angew. Chem., Int. Ed. 2020, 59, 2220–2224. 10.1002/anie.201911071. [DOI] [PubMed] [Google Scholar]; c Hong X.; Stevens M. C.; Liu P.; Wender P. A.; Houk K. N. Reactivity and chemoselectivity of allenes in Rh(I)-catalyzed intermolecular (5 + 2) cycloadditions with vinylcyclopropanes: allene-mediated rhodacycle formation can poison Rh(I)-catalyzed cycloadditions. J. Am. Chem. Soc. 2014, 136, 17273–17283. 10.1021/ja5098308. [DOI] [PMC free article] [PubMed] [Google Scholar]; d Casitas A.; Krause H.; Lutz S.; Goddard R.; Bill E.; Fürstner A. Ligand exchange on and allylic C-H activation by iron(0) fragments: π-complexes, allyliron species, and metallacycles. Organometallics 2018, 37, 729–739. 10.1021/acs.organomet.7b00571. [DOI] [Google Scholar]; e Schobert R. Chelated η2-alkene- and η3-allyl-carbene complexes of late transition metals: structure-reactivity relations and preparative use. Part 10. The chemistry of metallacyclic alkenylcarbene complexes. J. Organomet. Chem. 2001, 617–618, 346–359. 10.1016/s0022-328x(00)00604-5. [DOI] [Google Scholar]
  39. Schore N. E. Transition-metal-mediated cycloaddition reactions of alkynes in organic synthesis. Chem. Rev. 1988, 88, 1081–1119. 10.1021/cr00089a006. [DOI] [Google Scholar]
  40. Roglans A.; Pla-Quintana A.; Solà M. Mechanistic studies of transition-metal-catalyzed [2 + 2 + 2] cycloaddition reactions. Chem. Rev. 2021, 121, 1894–1979. 10.1021/acs.chemrev.0c00062. [DOI] [PubMed] [Google Scholar]
  41. a Kozuch S.; Shaik S. Kinetic-quantum chemical model for catalytic cycles: the Haber–Bosch process and the effect of reagent concentration. J. Phys. Chem. A 2008, 112, 6032–6041. 10.1021/jp8004772. [DOI] [PubMed] [Google Scholar]; b Kozuch S.; Shaik S. How to conceptualize catalytic cycles? The energetic span model. Acc. Chem. Res. 2011, 44, 101–110. 10.1021/ar1000956. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

cs2c05790_si_001.pdf (7.1MB, pdf)
cs2c05790_si_002.cif (526.4KB, cif)
cs2c05790_si_003.cif (1.5MB, cif)
cs2c05790_si_004.cif (6MB, cif)

Articles from ACS Catalysis are provided here courtesy of American Chemical Society

RESOURCES