Insights into the mechanism of active catalyst generation for the Pd^II(acac)₂/PPh₃ system in the context of telomerization of 1,3-butadiene

Abstract

The mechanism by which Pd^II precursors are reduced to catalytically active low-valent Pd species has been a subject of interest for developing better catalysts. This process is well understood for catalytic systems employing a combination of palladium(II) acetate [Pd(OAc)₂] and tertiary phosphines. However, the mechanism of reduction of palladium(II) acetylacetonate [Pd(acac)₂] in the presence of phosphines has not been thoroughly investigated. This is especially important in the context of the Pd-catalyzed butadiene telomerization process, which uses a combination of Pd(acac)₂ and tertiary phosphines in methanol to produce the commercially valuable precursor 1-methoxyoctadiene (MOD-1). In this work, we elucidate the steps for generating the active Pd⁰ species for this reaction using a combination of Pd(acac)₂ and triphenylphosphine (PPh₃). The investigations presented in this study provide the following key insights: (a) unification of the steps involved in the generation of the active precatalyst [Pd^II(acac)(PPh₃)₂]⁺; (b) elucidation of the mechanism of reduction of the precatalyst to Pd⁰ without MOD-1, which parallels the chemistry of the Pd(OAc)₂/PPh₃ system; and (c) the generation of Pd^II-octadienyl species from the reaction between the precatalyst and MOD-1, the product of the telomerization reaction. A reversible C–O bond cleavage process was identified that leads to the formation of the Pd^II π-octadienyl species as the active catalyst in the commercial telomerization process. These studies provide important insights into the reduction of Pd(acac)₂ into active Pd⁰ species or Pd^II π-allyl species, which have wide implications for both cross-coupling catalysis as well as the telomerization reaction.

---

Introduction

Cross-coupling reactions form the cornerstone of modern pharmaceutical and agrochemical industries due to their ubiquity in forging C–C and C–heteroatom bonds. Palladium-catalyzed cross-coupling reactions^1,2 form the most used and well-understood classes of reactions, and were recognized by the 2010 Nobel Prize in Chemistry^3,4 being awarded for the development of this technology. Due to considerations of the ease of use, bench-stable Pd^II salts are used, often in conjunction with tertiary phosphines to generate the active catalysts for a plethora of cross-coupling reactions. Significant advancements have been made in attempts to understand the mechanism of generation of these active catalysts, which are often low-valent Pd species that can oxidatively engage with the electrophilic reaction partners. One of the most commonly used Pd^II sources is palladium acetate, [Pd(OAc)₂], whose reactivity with tertiary phosphines to generate the catalytically active Pd⁰ or Pd^I species has been investigated by Amatore and Jutand^5,6 and revisited recently by Fairlamb and coworkers (Fig. 1).⁷ An important insight obtained from the latter study was that a complicated Pd speciation involving low-valent dinuclear and trinuclear Pd clusters occurred en route to the generation of [Pd⁰(PPh₃)_n] and oxidized phosphine species. Thus, understanding the mechanistic intricacies of in situ reduction of Pd^II sources into catalytically competent, low-valent palladium species is a relevant problem that can give new insights into catalyst design.

Fig. 1 Summary of key insights obtained in the reduction of the Pd(OAc)₂/PPh₃ system to Pd⁰ from the work of Amatore, Jutand, and Fairlamb. Unlike this system, the reduction chemistry of Pd(acac)₂/PPh₃ is less well understood, which is the subject of the present study.

Another common source of Pd^II is the neutral palladium acetylacetonate [Pd(acac)₂], also used in conjunction with tertiary phosphines for various reactions like the hydrogenolysis of allylic formates⁸ and carbonates⁹ and displacement reactions of vinyl sulfones.¹⁰ One of the foremost industrial uses of the Pd(acac)₂/phosphine system is in the telomerization of 1,3-butadiene developed by The Dow Chemical Company. Butadiene telomerization is an important industrial process to generate feedstock precursors used in the preparation of polyolefins and linear low density polyethylenes (LLDPE; via copolymerization with ethylene).¹¹ LLDPE is an important material in the commodity world (Fig. 2A), and has one of the largest global demands of any polymeric material.¹² In the Pd-catalyzed telomerization reaction, two equivalents of butadiene are dimerized to generate a Pd-octadienyl species (Fig. 2B). A capping agent (telogen) which is present in the reaction mixture performs a tail-end functionalization of this Pd-octadienyl species and prevents higher order alkene formation.^13,14 This octadiene derivative can further be processed first by hydrogenation to yield the substituted octane, which upon cracking with an alumina catalyst, generates 1-octene (Fig. 2C). It is important to note that the reaction is proposed to be initiated by the coordination of two butadiene units at a monophosphine Pd⁰ center, followed by their oxidative dimerization. Yet, there have been no studies which have investigated the mechanism of generation of Pd⁰ from the commercially employed Pd(acac)₂/tertiary phosphine mixture in the reaction system.

Fig. 2 (A) Global commodity plastics production by type as of 2019;¹⁵ (B) sketch of the butadiene telomerization reaction to form 1-methoxyoctadiene developed by The Dow Chemical Company using the pre-catalyst solution shown; (C) butadiene telomerization reaction and downstream conversion to polymeric materials.

Among telomers, the most common in the literature and the most practical to be used on an industrial scale is methanol. The products of the telomerization reaction are methoxyoctadienes, yet the selective generation of the desired linear isomer 1-methoxy-2,7-octadiene (MOD-1) is not trivial. Other side products can be generated, such as the branched isomer 3-methoxy-1,7-octadiene (MOD-3), generated from alkoxylation at the C3 position, various unfunctionalized octatrienes, and 4-vinylcyclohexene as a Diels–Alder product.^16,17 The use of Pd catalysts supported by phosphine ligands has been extensively studied as a way to improve selectivity and stability over the course of the reaction occurring typically at elevated temperature and pressure (>60 °C and >60 psi).^18–23

Commercially, this reaction is achieved through the use of Pd(acac)₂ as the palladium precursor, in conjunction with a tertiary phosphine such as triphenylphosphine (PPh₃) as the ligand.²⁴ The consensus mechanism is shown in Fig. 3, and several Pd^II species are important intermediates in the butadiene telomerization process.²⁵ The first compound, abbreviated as Cat-1 in this study, is a cationic acetylacetonato-bisphosphine Pd^II species. The organometallic chemistry of Cat-1 has been studied by Kawaguchi,²⁶ Suslov,²⁷ and others, and serves as the catalyst precursor that enters the catalytic cycle. The second set of compounds, abbreviated as Cat-2 and Cat-2′, are cationic phosphine-ligated Pd^II(octadienyl) complexes that are proposed intermediates in the catalytic cycle. Cat-2 is a monophosphine Pd^II(η³-allyl)(η³-alkenyl) compound, while Cat-2′ is a bisphosphine Pd^II(η³-allyl) species. This speciation might be critical, as it has been suggested by Beller and coworkers that the origin of the MOD-1/MOD-3 selectivity is determined by the ligation number of this active catalyst.¹⁶ The monophosphine adduct results in high MOD-1 selectivity, while the bisphosphine adduct leads to an increase in MOD-3 selectivity. It has been shown by researchers from The Dow Chemical Company that using a MOD-1 modified precatalyst solution eliminates an induction period in catalysis (Fig. 2B).²⁸ It is plausible that the induction period originates from facile generation of the proposed active Pd⁰ complex from the Pd^II precursor, and it has been proposed that on-cycle species such as Cat-2 and Cat-2′ generated in situ eliminate this induction period. However, the mechanism of the generation of these Pd^II species from a reaction between Cat-1 and MOD-1, presumably via coordination of olefinic species to low-valent Pd species, has not been studied in detail.

Fig. 3 Proposed catalytic cycle for butadiene telomerization using Pd(acac)₂ and tertiary phosphines. The key complexes that are discussed herein are highlighted in blue.

In this study, we use the combination of Pd(acac)₂ and PPh₃ as a model system to address some of the gaps in knowledge in the context of the aforementioned chemistry (Fig. 4). First, we build on the foundational studies from Kawaguchi et al.²⁶ to validate the mechanism of generation of Cat-1 from Pd(acac)₂ and PPh₃, which is not a simple, one-step reaction. We use a combination of multinuclear NMR spectroscopies and X-ray crystallography to elucidate this set of reactions and uncover a solvent dependence on the major products formed. We then provide definitive evidence of the generation of Pd⁰ by the in situ reduction of this Pd^II species,^29,30 which has broader thematic relevance in the context of Pd-catalyzed reactions which make use of Pd(acac)₂/phosphine systems. Finally, we show that in the presence of MOD-1, the product of the telomerization reaction, the formation of Pd^II octadienyl species like Cat-2 and Cat-2′ occurs from Cat-1 via a series of steps. These include the reduction of Pd^II to Pd⁰, followed by a reversible cleavage of the C–OMe bond in MOD-1 to generate the Pd^II-octadienyl species Cat-2 and Cat-2′. The existence of Cat-2 and Cat-2′ has been widely noted, but the pathway of generating them from the product of the telomerization reaction has never been investigated. While this reaction is specific for telomerization chemistry, the generation of Pd^II-allyl species from a mixture of Pd(acac)₂, phosphines, and an allylic group has broader relevance as Pd^II-allyl compounds are privileged precatalysts for various cross-coupling reactions.³¹

Fig. 4 The context of the present study: (a) validation of the mechanism of generating Cat-1 and unraveling its mechanism to form Pd⁰ phosphine species; (b) elucidation of the mechanism of formation of the octadienyl Pd^II precatalysts Cat-2 and Cat-2′ when the Pd(acac)₂/PPh₃ system is treated with MOD-1.

Results and discussion

Validation of the mechanism of formation of [(acac)Pd^II(PPh₃)₂]⁺ (Cat-1)

The reaction of Pd(acac)₂ with triarylphosphines in the presence of acetic acid and Lewis bases has been studied extensively by Kawaguchi²⁶ and Suslov,²⁷ respectively (Scheme 1). In this context, we set out to investigate the entire set of reactions involved in the proposed generation of [(acac)Pd^II(PPh₃)₂]⁺ (Cat-1) from Pd(acac)₂. The first step is the addition of one equivalent of phosphine to Pd(acac)₂, leading to the linkage isomerization of one bound acac ligand from the κ²-O,O to the κ¹-C_γ binding mode and formation of compound 1, which was crystallographically characterized by Kawaguchi and coworkers.²⁶ The preference of the κ¹-C_γ binding is governed by the relative stabilities of the keto versus enol forms of the β-dicarbonyl compounds and has been reviewed extensively by Kawaguchi.³²

Scheme 1 Prior work from the Suslov and Kawaguchi groups in identifying the intermediates leading to the formation of Cat-1.

In the presence of acetic acid, the formation of 2 can be readily explained by the protodemetalation of the κ¹-C_γ-bound acac to release one equivalent of free Hacac and subsequent κ¹-O coordination of the acetate.^26,33 This compound has been characterized by NMR and IR spectroscopy by the Kawaguchi group, but no crystallographic evidence was presented. The final step is the ligand substitution of the acetate by another equivalent of phosphine to generate Cat-1.³³ An analogous reaction has been reported by Suslov and coworkers,²⁷ whereby Cat-1 was prepared and crystallographically characterized from the reaction of 1 with two equivalents of the Lewis acid boron trifluoride etherate (BF₃·OEt₂) and one equivalent of phosphine. To the best of our knowledge, the sequence of reactions to get from Pd(acac)₂ to Cat-1 in the presence of acetic acid and PPh₃ has never been consolidated into a single, contiguous set of reactions. During our efforts to do so, we also crystallized several Pd–phosphine species that are along the reaction pathway in Scheme 1 and an off-pathway cis-(PPh₃)₂Pd^II(OAc)₂ byproduct that is in contrast to the expected product from ligand trans influence. Scheme 2 summarizes the key ³¹P NMR signatures observed experimentally in our studies.

Scheme 2 Key ³¹P NMR signatures for identifying the intermediates leading to the formation of Cat-1 in CD₂Cl₂ and CD₃OD described in this study.

In the context of butadiene telomerization, Cat-1 is usually prepared directly by mixing Pd(acac)₂ with 2 equivalents of PPh₃ and 1 equivalent AcOH in methanol.²⁸ This solution exhibits a single resonance of 36.5 parts per million (ppm) in the ³¹P NMR spectrum in CDCl₃ (Fig. S1†). ESI-MS analysis of the methanolic solution reveals a peak at m/z = 729.1317, which corresponds to the mass of the cationic species Cat-1. In order to understand the solution speciation of the sequence of reactions described above, we conducted stepwise reactions shown in Scheme 2. The reaction mixture was monitored by ³¹P and ¹H NMR spectroscopy in CD₃OD (the solvent used for telomerization) and CD₂Cl₂ (a less polar, aprotic solvent). As shown in Fig. 5A (top), addition of one equivalent of PPh₃ to Pd(acac)₂ in CD₂Cl₂ leads to a single peak in the ³¹P NMR spectrum at 33.5 ppm and characteristic shifts for the methine protons of the κ²-O,O and κ¹-C_γ bound acac (Fig. S16†). This compound corresponds to the Pd^II species with the κ¹-C_γ acac in the ¹H NMR spectrum (1), and it has been crystallized before by Kanda and coworkers.²⁶ During the course of our studies, an analog of 1 was crystallized through the use of a bis-(2-methoxyphenyl)phenylphosphine ligand – labelled 1-PPh^2-OMePh₂ (Fig. 7). On adding one equivalent acetic acid to the solution of 1 in CD₂Cl₂, the major species seen in the ³¹P NMR spectrum is a peak at 23.7 ppm (Fig. 5A, middle), which is attributed to 2 in the ¹H NMR spectrum. The disappearance of the methine and methyl protons corresponding to the κ¹-C_γ acac was observed, along with the appearance of the characteristic protons of free Hacac (Fig. S17 and S18†), thus confirming the replacement of the κ¹-C_γ bound acac with κ¹-O bound acetate. This was a fast reaction, and no intermediates were observed in the NMR timescale, suggesting that the formation of 2 from 1 is a simple ligand substitution reaction. An independent synthesis of the same compound in toluene showed the same ³¹P NMR shift (Fig. S2†). We were able to grow single crystals by the diffusion of diethyl ether into a toluene solution of 2, and the solid-state structure confirms the proposed identity of 2.

Fig. 5 (A) ³¹P NMR (202 MHz) spectra of the contiguous set of reactions starting from Pd(acac)₂ and leading to Cat-1 in CD₂Cl₂; (B) ³¹P NMR (202 MHz) spectra of the contiguous set of reactions starting from Pd(acac)₂ and leading to Cat-1 in CD₃OD. For panels (A) and (B), the sequence of reactions is as follows: (a) Pd(acac)₂ + 1 equiv. PPh₃ (top), (b) reaction mixture of (a) + 1 equiv. AcOH, and (c) reaction mixture of (b) + 1 equiv. PPh₃.

Interestingly, when the same set of reactions were carried out in CD₃OD, the importance of solvent polarity became apparent. Addition of one equivalent of PPh₃ to Pd(acac)₂ led to the precipitation of a small amount of yellow solid from the solution mixture. The ³¹P NMR of the solution showed two peaks, an intense peak at 36.5 ppm (Cat-1) and a less intense peak at 33.5 ppm (1, Fig. 5B, top). Adding 1 equivalent of AcOH to the reaction mixture led to the appearance of a peak at 23.6 ppm (2) with the concomitant formation of insoluble yellow solids, but the major species was still retained at 36.5 ppm. The yellow precipitate in these two reactions, which are formed in miniscule amounts, are 1 and 2, along with other minor species (Fig. S19 and S20†). These observations are consistent with the neutral compounds 1 and 2 being soluble in a relatively less polar solvent (CD₂Cl₂) and sparingly soluble in a polar solvent like CD₃OD. Interestingly, some of the monocationic Cat-1 forms even during the first step in methanol, while the formation of Cat-1 is incomplete even after the addition of one equivalent of PPh₃, AcOH, and a further equivalent of PPh₃ to Pd(acac)₂ in dichloromethane (vide infra).

Finally, the last step in the reaction to generate Cat-1 is the replacement of the inner-sphere acetate by a second equivalent of PPh₃. Accordingly, the addition of one equivalent of PPh₃ to the reaction mixture containing 2 in CD₂Cl₂ led to the formation of two new species in the ³¹P NMR (Fig. 5A, bottom). One of them was readily identified as the cationic Cat-1 (36.5 ppm) and a second unidentified species was seen at 30.5 ppm. Our hypothesis about the identity of this species is a bridged Pd^II-acetate compound, which has been discussed later (vide infra). As expected, the same reaction in CD₃OD led to the clean formation of a single peak at 36.5 ppm (Fig. 5B, bottom), confirming that the more polar solvent helps stabilize the cationic species Cat-1. In order to verify this hypothesis, we reasoned that the formation of the cationic species may not need AcOH, as the acac itself can act as the counteranion. Indeed, when two equivalents of PPh₃ are added to Pd(acac)₂ in CD₃OD, clean formation of Cat-1 at 36.5 ppm is observed (Fig. 6A, top). When the same reaction is performed in CD₂Cl₂, the major peak corresponding to 1 is seen at 33.5 ppm, along with free PPh₃ at −4.5 ppm (Fig. 6A, bottom). It has been previously noted by Kawaguchi and coworkers that 1 is formed even when two equivalents of phosphine (or other Lewis bases) are reacted with Pd(acac)₂.³³ The ¹H NMR also verifies the identity of the species in solution as 1 along with free PPh₃ being formed (Fig. S23†).³³ When the precatalyst solution is prepared in one pot by mixing Pd(acac)₂, one equivalent of AcOH, and two equivalents PPh₃ in CD₂Cl₂, several signals are seen in the ³¹P NMR spectrum (Fig. 6B, bottom). Only a small peak is seen at 36.5 ppm, which stands in contrast to the clean formation of Cat-1 in a methanolic solution (Fig. 6A, top, and also confirmed by ¹H NMR, Fig. S26†). Taken together, these results demonstrate the importance of methanol in stabilizing the cationic Cat-1 species.

Fig. 6 (A) ³¹P NMR (202 MHz) spectra of the reaction between Pd(acac)₂ and two equivalents of PPh₃ performed in CD₃OD (top) versus CD₂Cl₂ (bottom); (B) ³¹P NMR (202 MHz) spectra of the reaction between Pd(acac)₂, one equivalent acetic acid, and two equivalents of PPh₃ performed in CD₃OD (top) versus CD₂Cl₂ (bottom).

Fig. 7 Oak Ridge Thermal Ellipsoid Plots (ORTEPs) of complexes 1-PPh^2-OMePh₂, 2, and 3, shown at 50% probability and with H atoms omitted for clarity. Selected bond distances (Å) for 1-PPh^2-OMePh₂: Pd1–P1: 2.2598(8); Pd1–C1: 2.104(3); Pd1–O1: 2.060(2); Pd1–O2: 2.058(2); 2: Pd1–P1: 2.2466(3); Pd1–O1: 1.9915(9); Pd1–O2: 2.0531(9); Pd1–O3: 2.0095(9); and 3: Pd1–P1: 2.255(4); Pd1–P2: 2.288(4); Pd1–O1: 2.058(10); Pd1–O2: 2.068(10).

Some general observations can be made about the differences in reactivity between the two solvents. It is likely that compounds 1, 2, and Cat-1 exist in equilibrium with one another. In CD₂Cl₂, a salt like Cat-1 may form immediately but is unlikely to be stable. In such a case, the outer-sphere counteranion would preferentially displace one of the PPh₃ ligands to form either 1 or 2 from acac or ⁻OAc counteranions. On the other hand, methanol will likely preferentially stabilize the charge separated Cat-1 more strongly than 1 or 2. This is particularly evident in the NMR comparison shown in Fig. 6B.

We then set out to connect the aforementioned chemistry with the Pd(OAc)₂/PPh₃ system, and probe whether we could access similar Pd species starting from Pd(acac)₂. Upon addition of excess AcOH to the reaction mixture in methanol (5 equivalents instead of 1 equivalent), we were able to synthesize and crystallographically characterize an unusual compound cis-(PPh₃)₂Pd^II(OAc)₂ (3) that has a characteristic ³¹P NMR peak of 28.3 ppm in CDCl₃. While 3 does not directly fall along the reactivity scheme, we hypothesize that its formation occurs via a repeat of the earlier reaction steps on Cat-1. The first step is the κ²-O,O to κ¹-C_γ isomerization of the acac ligand, followed by a protodemetalation with a second equivalent of acetic acid, and then coordination of both acetate anions. We propose that this compound links the more well understood Pd(OAc)₂/PPh₃ with the Pd(acac)₂/PR₃ chemistry discussed here. It is known that Pd(OAc)₂ reacts with two equivalents of triphenylphosphine at room temperature to form trans-(PPh₃)₂Pd^II(OAc)₂, which was shown to undergo a phosphine-acetate reductive elimination by Amatore and coworkers to eventually reduce to a [Pd⁰(PPh₃)_n] species and an oxidized phosphine product. In our case, the cis configuration of all compounds shown in Scheme 2 is enforced from the start by the cis-chelating acac ligand. The proposed PPh₃-OAc reductive elimination in the trans isomer to generate the Pd⁰ species relies on the formation of a chelate ring between PPh₃ and the bound acetate.⁷ In contrast, the cis configuration of the acetate ligands may be primed for hydrogen bonding with protic solvents or water, as seen in the solid-state structure of 3 (Fig. S43†). This may prevent the formation of a chelate with PPh₃ and hence prevent reductive elimination with PPh₃ at room temperature, which is consistent with the lack of [Pd⁰(PPh₃)_n] or Pd black formation in this reaction system.

The solid-state structures shown in Fig. 7 merit some discussion. All complexes demonstrate square planar geometries, with geometry index values of 0.046, 0.053, and 0.070, respectively, as expected for 4-coordinate Pd^II complexes. Complexes 1-PPh^2-OMePh₂, 2, and 3 show Pd–P bond lengths of 2.260, 2.247, and a Pd–P_ave bond length of 2.275 Å (values ranging from 2.255 to 2.289 Å, two molecules of 3 per asymmetric unit), respectively. Complex 1-PPh^2-OMePh₂ also shows longer average Pd-acac bond lengths as compared to 2 (2.059 vs. 2.022 Å) due to a stronger trans influence from the κ¹-C_γ binding mode of the acac ligand and more electron-donating phosphine ligand. Furthermore, the asymmetry of the Pd–O bond lengths in the Pd-acac moiety is significantly larger in complex 2 compared to 1-PPh^2-OMePh₂ due to disparity in the trans influence going from a C-bound acac to an acetate ligand (1-PPh^2-OMePh₂: 2.060 & 2.058 Å; 2: 1.992 & 2.053 Å). It is also worth noting that 3 has shorter Pd–P_ave distances (cis: 2.27 Å, trans: 2.33 Å) and longer Pd–O bond lengths (cis: 2.06 Å, trans: 2.03 Å) than the trans isomer,⁷ due to the stronger trans influence of PPh₃.

Conversion of Cat-1 to Pd⁰ species

As alluded to in the previous sections, the mechanism of Pd^II reduction in the Pd(OAc)₂/PPh₃ catalytic system has garnered significant interest, yet little work has been done in investigating the mechanism of reduction of the Pd^II(acac)₂/PPh₃ catalytic system. During the course of our studies, we found that reduction of the Pd^II to Pd⁰ did not happen at room temperature, yet it was facile at higher temperatures in the absence of acetic acid. This section deals with the mechanistic insights we obtained in this regard, with the key findings being summarized in Scheme 3.

Scheme 3 Key reactions involved in the generation of Cat-1 and Pd⁰ species from Pd(acac)₂/PPh₃, both at room temperature and elevated temperature in methanol.

As noted previously, when a methanolic solution of Pd(acac)₂ is treated with 2 equiv. PPh₃ in the absence of acetic acid, formation of Cat-1 was observed in the ³¹P NMR and 1 crashed out of the solution (Fig. S29†). However, on heating the same reaction mixture up to 50 °C, significant formation of Pd⁰ black was observed. ESI-MS analysis (Fig. S27 and Table S1†) of the organic fraction revealed mass fragments for [(acac)-PPh₃]⁺ and O=PPh₃, providing strong evidence of reductive elimination of (acac) and PPh₃ from Pd^II to generate Pd⁰ species (Scheme 4). This would presumably proceed similarly to the mechanism proposed by Amatore and Jutand^6,34 for the reductive elimination of Pd^II bound acetate and triphenylphosphine. The generation of O=PPh₃ can be rationalized by the hydrolysis of [(acac)-PPh₃]⁺, which can be generated from trace water in methanol or in the solution used for mass spectrometry analysis. The ³¹P NMR of this solution showed the peak corresponding to Cat-1, along with the formation of a bisphosphine species. This was evident from the appearance of two equal intensity doublets (J = 32 Hz). In the absence of other evidence, we cannot confirm the exact identity of this species (Fig. S30†). We hypothesize that this is an intermediate that decomposes by reductive elimination. GC-MS analysis of the organic fraction of this reaction mixture shows stoichiometric biphenyl formation (integrated relative to PPh₃), which suggests P–C_aryl bond cleavage of PPh₃ (Fig. S32†). This process has precedence from the work of Hursthouse and coworkers,³⁵ who reported the synthesis and characterization of a Pd^II dimer formed by the cleavage of the P–C_aryl bond – [Pd^II₂(μ-OAc)₂Ph₂(PPh₃)₂] in 41% yield when Pd(OAc)₂ is refluxed with excess PPh₃ in methanol. Interestingly, formation of biphenyl was observed by Fairlamb and coworkers⁷ in a crude reaction mixture containing the Pd^I dimer shown in Fig. 1, indicating that such intermediates may be accessed in the Pd(acac)₂/2 PPh₃ system as well. Based on this evidence, we propose that Pd^II(acac)-derived precursors may undergo the same types of reductive transformations as the more ubiquitous Pd(OAc)₂/PPh₃ systems.⁷

Scheme 4 Proposed mechanism of the generation of Pd⁰ species from Pd(acac)₂/PPh₃ at elevated temperature.

In the presence of acetic acid, the solution chemistry of Pd(acac)₂/PPh₃ is remarkably different. Addition of 1 equivalent AcOH to this reaction at 50 °C does not result in any new reactivity, only Cat-1 being formed, as observed by ³¹P NMR and ¹H NMR (Fig. S33–S35†). The main differentiating factor, however, is that Pd black is not observed in any reaction when AcOH is added, even at elevated temperatures (80 °C) or longer reaction times (16 hours). Critical evidence was obtained from ESI-MS studies (Fig. S28†), which notably showed only trace triphenylphosphine oxide and no evidence of [(acac)-PPh₃]⁺. This indicates an absence of reductive elimination pathways. Instead, new Pd-based mass fragments were seen, which could be assigned to Cat-1 and other Pd^II species. Possible explanation for the lack of Pd^II reduction could be the reoxidation of any Pd⁰ species being formed by acetic acid, which can react further with free acac and PPh₃ to form Cat-1. It is interesting to note that this pattern of reactivity is completely divergent from the facile, room-temperature reduction of the trans-(PPh₃)₂Pd^II(OAc)₂ generated when Pd(OAc)₂ reacts with 2 equivalents of PPh₃.⁵

Conversion of Cat-1 to Cat-2 and Cat-2′ in the presence of MOD-1 via Pd⁰ intermediates

Having established the reductive chemistry of the Pd(acac)₂/2 PPh₃ system without MOD-1, we then investigated the steps to generate the Pd^II-octadienyl intermediates Cat-2 or Cat-2′ in the presence of MOD-1. It must be mentioned that significant prior work by Jolly and Beller has focused on the investigation of the solution-state speciation of the intermediates involved in Pd-catalyzed reactions of 1,3-dienes.^36–39 Yet there has been no investigation into the generation of these Pd^II-octadienyl species when there is no butadiene in the reaction mixture, and the sole source of the octadiene is in fact MOD-1, the product of the butadiene telomerization reaction with methanol. In this section, we provide evidence for the C–O bond cleavage of the allyl ether and propose feasible routes for the generation of Cat-2 and Cat-2′ from Pd(acac)₂, PPh₃, AcOH, and MOD-1 (Scheme 5).

Scheme 5 Key reactions involved in the attempted generation of Pd^II-octadienyl species from Pd(acac)₂/PPh₃ at room temperature and elevated temperature in the presence of MOD-1 in methanol.

At temperatures below 50 °C for the reaction between Pd(acac)₂, 2 equivalents PPh₃, and 1 equivalent AcOH in the presence of excess MOD-1, no formation of Cat-2 is observed by ³¹P NMR spectroscopy (Fig. S34†). Since the formation of Cat-2 from MOD-1 would involve oxidative addition to an allyl ether, it is likely that a low-valent Pd species must be generated via a C–P reductive elimination as highlighted in the previous section. This is not observed at lower temperatures, which can be attributed to slow oxidative addition of the octadienyl ether or slow acac dissociation from Cat-1. However, when the same reaction mixture is heated at 70 °C, a complicated mixture of species is observed (Fig. 8, bottom panel). Most of the characteristic ³¹P signals were identified by comparison with independently synthesized compounds or literature precedents, which helped us assign most of the major species in solution. The most characteristic set of peaks was a set of two doublets at 25.6–26.3 ppm (J = 41.1 Hz) which are assigned to the bisphosphine species Cat-2′ (confirmed by independent synthesis, Fig. 8, middle panel). In addition, a singlet could be identified at 26.05 ppm which was confirmed to be Cat-2 (Fig. 8, top panel) by independent synthesis following the protocol reported by Weckhuysen et al.⁴⁰ By fitting the peaks an approximate ratio of 1 : 3 can be identified for Cat-2 : Cat-2′ in the reaction mixture (Fig. S36†). ESI-MS analysis of this reaction mixture shows two major organic products at m/z = 279.0936 and m/z = 371.1921 (Fig. S38 and Table S4†), which can be assigned to triphenylphosphine oxide and an octadienyl-phosphonium (Oct-PPh₃) salt, respectively. The latter has been reported²⁴ to be a decomposition product in butadiene telomerization reactions, and the former is known to be a hydrolysis product of Oct-PPh₃. Moreover, we could also identify the octadienylphosphonium salt in the ³¹P NMR spectrum that reveals a singlet at 21.6 ppm.^41,42 Other key ³¹P NMR signals were attributed to 3 (28.6 ppm) and the bridging Pd^II species arising from C–P bond cleavage (30.5 ppm), as observed by Hursthouse when Pd(OAc)₂ reacts with PPh₃ in methanol (vide supra).³⁵ Control experiments mimicking the generation of a MOD-1 bound Pd⁰ PPh₃ compound (Fig. S37†) suggest that the peak at 32.5 ppm may be (PPh₃)Pd⁰(MOD-1), which we also see in the ESI-MS of the reaction mixture.

Fig. 8 ³¹P NMR (202 MHz) spectra related to the generation of Cat-2 and Cat-2′. (A) ³¹P NMR spectrum of Cat-2 synthesized by the method reported by Weckhuysen et al.; (B) ³¹P NMR spectrum obtained on adding one equivalent PPh₃ to Cat-2 to generate Cat-2′; (C) crude ³¹P NMR spectrum of Pd(acac)₂, PPh₃, AcOH heated to 70 °C in 3 : 1 MeOH/MOD-1.

With the NMR and ESI-MS evidence in hand and based on our knowledge from the works of Jolly, Beller, and others, we propose the simplified mechanism outlined in Scheme 6. Importantly, this work focuses on how Cat-2 or Cat-2′ can be generated starting from the precatalyst solution used in the commercial process of the Dow Chemical Company, in order to eliminate induction period in the Pd-catalyzed telomerization reactions.²⁸ In summary, we know that heating Cat-1 in the absence of acetic acid produces Pd⁰ species, which are presumably reoxidized in the presence of acetic acid. However, in the presence of a large excess of MOD-1 this putative phosphine-ligated Pd⁰ species can bind to a molecule of MOD-1 to generate [LPd⁰(MOD-1)], whose protonated adduct was detected by ESI-MS. This can then oxidatively add to the allyl ether bond where the methoxy group acts as a leaving group to generate the monophosphine Pd^II octadienyl species Cat-2, detected in both ESI-MS and ³¹P NMR. Key evidence for this step was obtained from the reaction of Pd^II(acac)₂ with 2 equiv. PPh₃ and 10 equiv. MOD-1 in CD₃OH in the absence of acetic acid. GC-MS analysis of the reaction mixture shows the incorporation of 3 deuterium atoms into MOD-1 with no observed formation of octadienes, which indicates that C–O bond cleavage is reversible under these conditions and the exchange of the methoxy group with the deuterated methanol solvent is occurring during this process (Fig. S39†).

Scheme 6 Proposed mechanism of the generation of Cat-2 and Cat-2′ species from Pd(acac)₂/PPh₃ at elevated temperature in the presence of MOD-1. The ³¹P NMR shifts of the relevant species in methanol and the observed m/z values from ESI-MS are given as well.⁴¹

While the above discussion implies that the Pd center is cleaving the C–O bond that it forms during the catalysis, it is important to note that the catalytic reaction is initiated with a large excess of butadiene relative to the palladium catalyst. All experimental data presented in this study consider reaction conditions in which butadiene is not involved. In addition, typical reaction times tend to be quite short – usually two hours.²⁴ Thus, it is expected that the oxidative dimerization of two butadiene molecules at Pd⁰, followed by nucleophilic attack of methoxide should still be the predominant reaction pathway under normal telomerization reaction conditions.

While Cat-2 is the critical selectivity-determining intermediate for the formation of MOD-1, we also see NMR and ESI-MS evidence of a bisphosphine species. This can be readily explained by a reversible addition of a further equivalent of PPh₃ to Cat-2, to generate Cat-2′. In fact, this is the major species we see in solution and also has precedent. It has been shown that less sterically demanding phosphines, without an ortho substituent leads to the formation of bisphosphine species. Finally, to account for the formation of octadienyl-phosphonium species detected in the ³¹P NMR and ESI-MS, we propose two possible pathways. One possibility is the reductive elimination of a molecule of PPh₃ with the octadienyl species which can regenerate the Pd⁰ species [LPd⁰(MOD-1)]. Another route for generating the octadienyl-phosphonium salt is the nucleophilic attack of a free phosphine on the π-allyl system of the octadienyl ligand bound to Cat-2, which would also generate a putative monophosphine-Pd⁰ species that can participate in downstream reactivity; a step that has been investigated by Weckhuysen and coworkers.^24,30,41

Conclusions

In this study, we have undertaken a systematic investigation of the chemical steps involved in the generation of the [(PPh₃)₂Pd^II(acac)]⁺ species starting from Pd(acac)₂. The summary of the key Pd species and the conditions under which they are generated has been shown in Scheme 7. We have also uncovered a solvent-dependent speciation for this process, whereby a polar protic solvent (methanol) can stabilize the cationic species better than a polar aprotic solvent (dichloromethane). In addition, we provided the first structural evidence for one of the intermediates (2) involved in the generation of [(PPh₃)₂Pd^II(acac)]⁺ and isolated a rare cis-[(PPh₃)₂Pd^II(OAc)₂] (3) during these investigations. Then we sought to understand the mechanism of reduction of the Pd^II species to generate Pd⁰. We found that in the absence of acetic acid, this process is facilitated by an acac-PPh₃ reductive elimination, akin to the OAc-PPh₃ reductive elimination seen in the Pd(OAc)₂/PPh₃ system. Importantly, this study provides strong support for the potential use of Pd(acac)₂/triarylphosphine systems as a precatalyst combination for traditional Pd-catalyzed reactions. This is especially relevant as the Pd(OAc)₂/PPh₃ system tends to form more complicated mixtures of Pd species. One surprising departure from the Pd(OAc)₂/PPh₃ system was the finding that adding AcOH to the Pd(acac)₂/PPh₃ system prevents the formation of Pd⁰ species, even at higher temperatures.

Scheme 7 Summary of overall palladium reactivity for the Pd(acac)₂/2 PPh₃ system in the context of precatalyst speciation for butadiene telomerization. Reaction conditions are shown in green.

Finally, in the context of butadiene telomerization, we have provided a mechanistic explanation for how this precatalyst system can react with MOD-1 to generate on-cycle intermediates such as Cat-2 and Cat-2′, or “naked” [Pd⁰(PPh₃)_n] species that skip the induction period typically observed during catalysis with Cat-1. This is somewhat counterintuitive, as MOD-1 is the major product of the telomerization reaction. However, the precatalyst solution is generated separately, without any butadiene present, which aids the formation of on-cycle species like Cat-2 and Cat-2′. The key new finding is the direct observation of the reversible C–O bond cleavage of the methoxy-octadiene fragment to generate a π-allyl species, a step that mimics the formation of Pd^II π-allyl compounds in Tsuji–Trost reactions. We anticipate that these insights will facilitate the design of more active catalytic systems for this important reaction. Overall, these results map an underused catalytic system – the combination of Pd(acac)₂ with triarylphosphines, into the broader context of Pd catalysis by examining the mechanism and conditions for the generation of catalytically active Pd⁰ and Pd^II-octadienyl compounds in protic solvents.³⁰

Data availability

Experimental details of the synthesis of new compounds (3 and 1-PPh^2-OMePh₂), NMR and ESI-MS of reaction mixtures, additional characterization data including ¹H, ¹³C, and ³¹P NMR spectra, and crystallographic information are available in the ESI.† The deposition numbers CCDC 2376744–2376746 at the Cambridge Crystallographic Data Centre (CCDC) contain the crystallographic data for compounds 1-PPh^2-OMePh₂, 2, and 3. These data are provided free of charge by the CCDC (https://www.ccdc.cam.ac.uk/).

Author contributions

BSB, DYB, SC, MR, RDK, and LMM designed the experiments. Experimental work and data analysis were performed by SC, DYB and BSB. The manuscript was written through the contributions of all authors.

Conflicts of interest

The authors have no conflicts to declare.

Acknowledgements

We thank the Dow Chemical Company (UPI 226772AR) for financial support. We also thank the research facilities and staff scientists at the University of Illinois at Urbana-Champaign for their help. We also thank Drs Toby Woods and Danielle Gray for their assistance in collecting the X-ray diffraction data. We thank Drs Alex Nett, Peter Groenendijk, and Jerzy Klosin for valuable discussions.

Notes and references

1. X. Chen, K. M. Engle, D. H. Wang and J. Q. Yu, Palladium(II)-Catalyzed C-H Activation/C-C Cross-Coupling Reactions: Versatility and Practicality, Angew. Chem., Int. Ed., 2009, 48, 5094–5115.
2. N. Miyaura and A. Suzuki, Palladium-Catalyzed Cross-Coupling Reactions of Organoboron Compounds, Chem. Rev., 1995, 95, 2457–2483.
3. C. C. Johansson Seechurn, M. O. Kitching, T. J. Colacot and V. Snieckus, Palladium-catalyzed cross-coupling: a historical contextual perspective to the 2010 Nobel Prize, Angew. Chem., Int. Ed., 2012, 51, 5062–5085.
4. D. Astruc, The 2010 Chemistry Nobel Prize to R.F. Heck, E. Negishi, and A. Suzuki for palladium-catalyzed cross-coupling reactions, Anal. Bioanal. Chem., 2011, 399, 1811–1814.
5. C. Amatore, A. Jutand and M. A. M'Barki, Evidence of the formation of zerovalent palladium from Pd(OAc)₂ and triphenylphosphine, Organometallics, 1992, 11, 3009–3013.
6. C. Amatore and A. Jutand, Anionic Pd(0) and Pd(II) Intermediates in Palladium-Catalyzed Heck and Cross-Coupling Reactions, Acc. Chem. Res., 2000, 33, 314–321.
7. N. W. J. Scott, M. J. Ford, C. Schotes, R. R. Parker, A. C. Whitwood and I. J. S. Fairlamb, The ubiquitous cross-coupling catalyst system ‘Pd(OAc)₂’/2PPh₃ forms a unique dinuclear Pd^I complex: an important entry point into catalytically competent cyclic Pd₃ clusters, Chem. Sci., 2019, 10, 7898–7906.
8. T. Mandai, T. Matsumoto, M. Kawada and J. Tsuji, Stereocontrolled formation of cis and trans ring junctions in hydrindane, decalin, and steroid systems by palladium-catalyzed regioselective and stereospecific hydrogenolysis of allylic formates, Tetrahedron, 1993, 49, 5483–5493.
9. T. Mandai, T. Matsumoto, M. Kawada and J. Tsuji, A novel method for stereospecific generation of either C-20 epimer in steroid side chains by palladium-catalyzed hydrogenolysis of C-20 allylic carbonates, J. Org. Chem., 1992, 57, 6090–6092.
10. J. L. Fabre and M. Julia, Organic synthesis with sulfones no. XXIX stereospecific hydrogenolysis of vinylic sulfones with grignards and transition metal catalysts, Tetrahedron Lett., 1983, 24, 4311–4314.
11. J. Yang, P. Wang, H. Neumann, R. Jackstell and M. Beller, Industrially applied and relevant transformations of 1,3-butadiene using homogeneous catalysts, Ind. Chem. Mater., 2023, 1, 155–174.
12. M. Jaganmohan, Global demand for linear low-density polyethylene (LLDPE) from 1999 to 2019 actual, and forecast for 2020 to 2040, by region, https://www.statista.com/statistics/1248064/worldwide-demand-for-lldpe/, accessed 9 February 2024.
13. E. J. Smutny, Oligomerization and dimerization of butadiene under homogeneous catalysis. Reaction with nucleophiles and the synthesis of 1,3,7-octatriene, J. Am. Chem. Soc., 1967, 89, 6793–6794.
14. S. Takahashi, T. Shibano and N. Hagihara, Dimerization of Butadiene by Palladium Complex Catalysts, Tetrahedron Lett., 1967, 2451–2452.
15. P. F. Britt, G. W. Coates, K. I. Winey, J. Byers, E. Chen, B. Coughlin, C. Ellison, J. Garcia, A. Goldman and J. Guzman, Report of the Basic Energy Sciences Roundtable on Chemical Upcycling of Polymers, USDOE Office of Science, Office of Science and Technical Information, 2019,  DOI:10.2172/1616517.
16. F. Vollmüller, J. Krause, S. Klein, W. Mägerlein and M. Beller, Control of Chemo- and Regioselectivity in the Palladium-Catalyzed Telomerization of Butadiene with Methanol – Catalysis and Mechanism, Eur. J. Inorg. Chem., 2000, 2000, 1825–1832.
17. C.-F. Huo, R. Jackstell, M. Beller and H. Jiao, Mechanistic study of palladium-catalyzed telomerization of 1,3-butadiene with methanol, J. Mol. Model., 2010, 16, 431–436.
18. R. Jackstell, S. Harkal, H. Jiao, A. Spannenberg, C. Borgmann, D. Röttger, F. Nierlich, M. Elliot, S. Niven, K. Cavell, O. Navarro, M. S. Viciu, S. P. Nolan and M. Beller, An Industrially Viable Catalyst System for Palladium-Catalyzed Telomerizations of 1,3-Butadiene with Alcohols, Chem.–Eur. J., 2004, 10, 3891–3900.
19. F. Benvenuti, C. Carlini, M. Lami, M. Marchionna, R. Patrini, A. M. R. Galletti and G. Sbrana, Telomerization of 1,3-butadiene with alcohols catalyzed by homogeneous palladium(0) complexes in the presence of mono- and diphosphine ligands, J. Mol. Catal. A: Chem., 1999, 144, 27–40.
20. R. Patrini, M. Lami, M. Marchionna, F. Benvenuti, A. M. R. Galletti and G. Sbrana, Selective synthesis of octadienyl and butenyl ethers via reaction of 1,3-butadiene with alcohols catalyzed by homogeneous palladium complexes, J. Mol. Catal. A: Chem., 1998, 129, 179–189.
21. R. Palkovits, A. N. Parvulescu, P. J. C. Hausoul, C. A. Kruithof, R. J. M. Klein Gebbink and B. M. Weckhuysen, Telomerization of 1,3-butadiene with various alcohols by Pd/TOMPP catalysts: new opportunities for catalytic biomass valorization, Green Chem., 2009, 11, 1155–1160.
22. M. Camargo, P. Dani, J. Dupont, R. F. de Souza, M. Pfeffer and I. Tkatchenko, Cationic cyclopalladated complexes: new catalyst precursors for the telomerization of butadiene with alcohols, J. Mol. Catal. A: Chem., 1996, 109, 127–131.
23. R. Jackstell, A. Grotevendt, M. G. Andreu and M. Beller, A Practical Palladium-Catalyzed Telomerization for the Synthesis of Functionalized Alcohols, Org. Process Res. Dev., 2009, 13, 349–353.
24. J. R. Briggs, J. T. Patton, S. Vermaire-Louw, P. M. Margl, H. Hagen and D. Beigzadeh, US Pat., US8558030B2, 2009.
25. J. R. Briggs, H. Hagen, S. Julka and J. T. Patton, Palladium-catalyzed 1,3-butadiene telomerization with methanol. Improved catalyst performance using bis-o-methoxy substituted triarylphosphines, J. Organomet. Chem., 2011, 696, 1677–1686.
26. Z. Kanda, Y. Nakamura and S. Kawaguchi, Terminal-Carbon-Bonded Palladium(II) Complexes of Acetylacetone, Chem. Lett., 1976, 199–200,  DOI:10.1246/cl.1976.199.
27. D. S. Suslov, M. V. Bykov, Z. D. Abramov, I. A. Ushakov, T. N. Borodina, V. I. Smirnov, G. V. Ratovskii and V. S. Tkach, Cationic palladium(II)–acetylacetonate complexes containing phosphine and aminophosphine ligands and their catalytic activities in telomerization of 1,3-butadiene with methanol, J. Organomet. Chem., 2020, 923, 121413.
28. J. L. Klinkenberg and K. P. Lawry, Sterically Encumbered and Poorly Electron-Donating Oxaphosphaadamantane Ligands for the Pd-Catalyzed Telomerization of Butadiene with Methanol, Org. Process Res. Dev., 2019, 23, 1654–1658.
29. R. Bigler, D. Spiess, J. Wellauer, M. Binder, V. Carré and S. Fantasia, Synthesis of Biaryl Phosphine Palladium(0) Precatalysts, Organometallics, 2021, 40, 2384–2388.
30. P. R. Melvin, D. Balcells, N. Hazari and A. Nova, Understanding Precatalyst Activation in Cross-Coupling Reactions: Alcohol Facilitated Reduction from Pd(II) to Pd(0) in Precatalysts of the Type (η³-allyl)Pd(L)(Cl) and (η³-indenyl)Pd(L)(Cl), ACS Catal., 2015, 5, 5596–5606.
31. N. Hazari and D. P. Hruszkewycz, Dinuclear Pd(I) complexes with bridging allyl and related ligands, Chem. Soc. Rev., 2016, 45, 2871–2899.
32. S. Kawaguchi, Variety in the coordination modes of β-dicarbonyl compounds in metal complexes, Coord. Chem. Rev., 1986, 70, 51–84.
33. S. Baba, T. Ogura and S. Kawaguchi, Reactions of Bis(acetylacetonato)palladium(II) with Triphenylphosphine and Nitrogen Bases, Bull. Chem. Soc. Jpn., 2006, 47, 665–668.
34. C. Amatore, E. Carre, A. Jutand, M. A. M'Barki and G. Meyer, Evidence for the Ligation of Palladium(0) Complexes by Acetate Ions: Consequences on the Mechanism of Their Oxidative Addition with Phenyl Iodide and PhPd(OAc)(PPh3)2 as Intermediate in the Heck Reaction, Organometallics, 1995, 14, 5605–5614.
35. M. B. Hursthouse, O. D. Sloan, P. Thornton and N. P. C. Walker, Preparation and x-ray crystal structure of [Pd₂(μ-O₂CMe)₂(C₆H₅)₂(PPh₃)₂], a binuclear acetate-bridged palladium(II) phenyl complex formed by P-C bond cleavage of triphenylphosphine, Polyhedron, 1986, 5, 1475–1478.
36. R. Benn, P. W. Jolly, R. Mynott and G. Schenker, Intermediates in the palladium-catalyzed reactions of 1,3-dienes. 1. (η³,η³-dodecatrienediyl)palladium [Pd(η³,η³-C₁₂H₁₈)], Organometallics, 1985, 4, 1136–1138.
37. R. Benn, P. W. Jolly, R. Mynott, B. Raspel, G. Schenker, K. P. Schick and G. Schroth, Intermediates in the palladium-catalyzed reactions of 1,3-dienes. 2. Preparation and structure of (η¹,η³-octadienediyl)palladium complexes, Organometallics, 1985, 4, 1945–1953.
38. P. W. Jolly, R. Mynott, B. Raspel and K. P. Schick, Intermediates in the palladium-catalyzed reactions of 1,3-dienes. The reaction of (η¹,η³-octadiendiyl)palladium complexes with acidic substrates, Organometallics, 1986, 5, 473–481.
39. R. Benn, G. Gabor, P. W. Jolly, R. Mynott and B. Raspel, Intermediates in the palladium-catalysed reactions of 1,3- dienes: IV. The reactions of η¹,η³-octadienediyl-palladium complexes with alkynes and activated alkenes, J. Organomet. Chem., 1985, 296, 443–447.
40. P. J. C. Hausoul, A. N. Parvulescu, M. Lutz, A. L. Spek, P. C. A. Bruijnincx, B. M. Weckhuysen and R. J. M. Klein Gebbink, Facile Access to Key Reactive Intermediates in the Pd/PR₃-Catalyzed Telomerization of 1,3-Butadiene, Angew. Chem., Int. Ed., 2010, 49, 7972–7975.
41. P. J. C. Hausoul, M. Lutz, J. T. B. H. Jastrzebski, P. C. A. Bruijnincx, B. M. Weckhuysen and R. J. M. Klein Gebbink, Mechanistic Study of the Pd/TOMPP-Catalyzed Telomerization of 1,3-Butadiene: Influence of Aromatic Solvents on Bis-Phosphine Complex Formation and Regioselectivity, Organometallics, 2013, 32, 5047–5057.
42. K. Yada, K. Shimoyamada, M. Muranaka, J. Fuji and S. Suzuki, Eur. Pat., EP1249455A1, 2002.

---

Footnotes

† Electronic supplementary information (ESI) available. CCDC 2376744–2376746. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5sc02171e

‡ These authors have contributed equally.

---