Novel Route to Cationic Palladium(II)–Cyclopentadienyl Complexes Containing Phosphine Ligands and Their Catalytic Activities

The Pd(II) complexes [Pd(Cp)(L)n]m[BF4]m were synthesized via the reaction of cationic acetylacetonate complexes with cyclopentadiene in the presence of BF3∙OEt2 (n = 2, m = 1: L = PPh3 (1), P(p-Tol)3, tris(ortho-methoxyphenyl)phosphine (TOMPP), tri-2-furylphosphine, tri-2-thienylphosphine; n = 1, m = 1: L = dppf, dppp (2), dppb (3), 1,5-bis(diphenylphosphino)pentane; n = 1, m = 2 or 3: 1,6-bis(diphenylphosphino)hexane). Complexes 1–3 were characterized using X-ray diffractometry. The inspection of the crystal structures of the complexes enabled the recognition of (Cp–)⋯(Ph-group) and (Cp–)⋯(CH2-group) interactions, which are of C–H…π nature. The presence of these interactions was confirmed theoretically via DFT calculations using QTAIM analysis. The intermolecular interactions in the X-ray structures are non-covalent in origin with an estimated energy of 0.3–1.6 kcal/mol. The cationic palladium catalyst precursors with monophosphines were found to be active catalysts for the telomerization of 1,3-butadiene with methanol (TON up to 2.4∙104 mol 1,3-butadiene per mol Pd with chemoselectivity of 82%). Complex [Pd(Cp)(TOMPP)2]BF4 was found to be an efficient catalyst for the polymerization of phenylacetylene (PA) (catalyst activities up to 8.9 × 103 gPA·(molPd·h)−1 were observed)


Introduction
The elucidation of ferrocene's structure in 1952 by Wilkinson, Woodward, and Fischer [1,2] inspired chemists to extensively study transition metal complexes with a cyclopentadienyl (Cp) ligand. This ligand exhibits unique electronic properties and often behaves as a stable three-coordinate spectator ligand [3]. Furthermore, Cp proved to be a good supporting ligand for catalyst design [4]. Although in most cases, industrial application requires the use of heterogeneous, supported catalysts, homogeneous and metal nanoparticle catalysts [5][6][7], which may present an intermediate option between homogeneous and heterogeneous catalysts, continue to evolve [8]. A famous example of this is metallocene catalysts (with titanium-family metals) that have been applied in the industrial production of α-olefin polymers owing to their single reaction active center and high catalytic activity [9][10][11][12].
metallocene catalysts (with titanium-family metals) that have been applied in the industrial production of α-olefin polymers owing to their single reaction active center and high catalytic activity [9][10][11][12].
In this paper, we present a novel route for the syntheses of cationic η 5 -cyclopentadienyl palladium(II) complexes, featuring mono-and bidentate tertiary phosphines ligands. This chemistry is summarized in Scheme 2. Ten new compounds have been fully characterized, and the crystal structures of [Pd(Cp)(PPh3)2]BF4, [Pd(Cp)(dppp)]BF4, and [Pd(Cp)(dppb)]BF4 have been determined via X-ray diffraction. NMR spectroscopy features of the prepared cationic complexes are also discussed. Additionally, we disclose our findings on the catalytic activity of the novel complexes in the polymerization of phenylacetylene (PA) and telomerization of butadiene (BD) with methanol.
In this paper, we present a novel route for the syntheses of cationic η 5 -cyclopentadienyl palladium(II) complexes, featuring mono-and bidentate tertiary phosphines ligands. This chemistry is summarized in Scheme 2. Ten new compounds have been fully characterized, and the crystal structures of [Pd(Cp)(PPh 3 ) 2 ]BF 4 , [Pd(Cp)(dppp)]BF 4 , and [Pd(Cp)(dppb)]BF 4 have been determined via X-ray diffraction. NMR spectroscopy features of the prepared cationic complexes are also discussed. Additionally, we disclose our findings on the catalytic activity of the novel complexes in the polymerization of phenylacetylene (PA) and telomerization of butadiene (BD) with methanol.

Synthesis of Palladium(II) Complexes
We have found that complexes 1-10 (Scheme 2) can be prepared through reacting cationic acetylacetonate complexes with cyclopentadiene in the presence of BF 3 ·OEt 2 . The starting palladium complexes are easily prepared in high yields from readily available [Pd(acac)(MeCN) 2 ]BF 4 [38,39]. Use of methanol as a solvent was found to be more convenient than non-coordinating dichloromethane or 1,2-dichloroethane. For example, the reaction of [Pd(acac)(PPh 3 ) 2 ]BF 4 with 1 eq. of BF 3 ·OEt 2 and 1 eq. of CpH in CH 2 Cl 2 led to the formation of desired complex 1 within 1 h. However, in this case, the reaction was complicated by a side reaction. In the IR and 1 H NMR spectra of the precipitate containing complex 1, signals that can be attributed to cyclopentadiene/dicyclopentadiene (DCPD) oligomers were observed. As a consequence, the preparation of complex 1 required additional recrystallization of the reaction product from a MeCN/Et 2 O mixture, which reduced the yield of the complex. The review [40] indicated that DCPD can be oligomerized using initiators of carbocationic polymerization. We carried out the reaction between cyclopentadiene (CpH) and BF 3 ·OEt 2 in the absence of a palladium compound and obtained a white powder. According to IR and 1 H NMR data, this can be attributed to a mixture of exo-2,3and exo-2,7-polyDCPD [41,42]. The side process of oligomerization does not occur in methanol, but a longer reaction time and the use of an excess of CpH and BF 3 ·OEt 2 are required to increase the yield of complex 1. On the other hand, the formation of complex 1 was not observed in more coordinating solvents such as MeCN or DMSO.
The new compounds (1-10) were characterized using multinuclear and two-dimensional (COSY, NOESY, HMBC, HSQC) NMR, IR, and UV spectroscopy; ESI-MS; and elemental analysis (see Supporting Information File (SI) for details). The obtained complexes appear to be stable solids even in air at ambient temperature, although they decompose slowly in solution. For example, solutions of complexes 1-10 in acetonitrile or MeCN/toluene mixtures were stable for days at room temperature, while those in MeOH, CH 2 Cl 2 , or CHCl 3 decomposed within 10 h, forming an unidentified black precipitate.
In all cases, coordination of the phosphines caused a downfield shift ∆(δ complex − δ ligand ) of the 31 P{ 1 H} NMR signals in the range of 34.4 to 55.4 ppm (Table 1). It should be noted that the 31 P{ 1 H} and 1 H NMR spectra of complex 3 are severely broadened at room temperature (e.g., ∆ν P (FWHM) = 365 Hz at δ P = 8.2 ppm), indicating that the complex shows dynamic exchange, most likely between "TOMPP-Pd"-rotamers in solution.
The NMR signal broadening observed for complex 3 is comparable to that observed for cis-[Pd((1−3η)-but-2-en-1-yl)(TOMPP) 2 ]BF 4 [43]. The δ values obtained from 1 H and 13 C NMR spectra of compounds 1-10 are consistent with NMR data for known palladium complexes with the same phosphine ligands [38,[50][51][52][53][54]. In each case, the 1 H NMR signal for the cyclopentadienyl group appears as a triplet (J HP = 2.1-2.3 Hz), due to coupling to the two phosphorus atoms (Table 2). This was additionally confirmed in the phosphorus-decoupled 1 H{ 31 P} NMR spectra, where these signals appeared as singlets. The 1 H NMR chemical shifts of the C 5 H 5 group were dependent on the nature of the phosphine ligand. The C 5 H 5 resonance in complex 3, for example, appears at 5.12 ppm (shielded by an aromatic ring current) when using two bulky tris(orthomethoxyphenyl)phosphine ligands (cone angle θ =176 • [55]), while the same group signal for complex 4 with non-bulky tris(2-furyl)phosphine was observed at 6.04 ppm. In addition, the 13 C{ 1 H} NMR spectra of compounds 1-9 revealed diagnostic carbon peaks of Cp ligands as triplets at 100-105 ppm (for complex 10, a broadened asymmetric singlet from C Cp was observed at 101.5 ppm). In the aromatic region of the 13 C{ 1 H} NMR spectra, the signals corresponding to the ipso-C, ortho-C, and meta-C of the PAr-moieties were observed as the expected virtual triplets [56], whereas signals for para-C were still observed as singlets.
The NMR spectra of compounds 9 and 10 suggest that they exist in different isomeric forms in solution. In particular, the 1 H and 13 C NMR spectra of compound 10 display three distinct sets of resonances in C 5 H 5 group region. The major isomer gave resonances at δ 5.78 (t, J = 2.0 Hz, C 5 H 5 ) and 101.51 (s, br., C 5 H 5 ), while the minor isomers were assigned to resonances at δ 5.76-5.69 (m), 5.61 (t, J = 2.0 Hz, C 5 H 5 ), and 102.18 (s, br., C 5 H 5 ). The 31 P{ 1 H} NMR spectrum of complex 10 displayed a singlet at δ 28.61 for the major isomer, while singlets at δ 28. 39-28.22, 27.11 were observed for the minor isomers in a 6:4 intensity ratio. We assume that the isomers observed in solution are due to presence of µ-P-type coordination dimers and oligomers. The ESI-MS isotope distribution data ( Figure S76, SI) for compound 10 do not contradict this assumption. The distribution can be interpreted in favor of the presence of bi-and trinuclear palladium complexes in solution. A quantitative calculation of the corresponding intensoids in the enviPat program [57] gave the ratio of dimers and trimers as 4 to 6. From ESI-MS and NMR data for complex 9, mononuclear compounds were mainly found in solution, and the suitable isotope distribution ratio was modeled as [M 1 ] + :[M 2 ] 2+ :[M 3 ] 3+ = 9.5:0.1:0.4. In the case of compounds 1-8, ESI-MS data confirmed the formation of the expected mononuclear complex cations.
Although the mechanism of synthesis of cationic cyclopentadienyl palladium complexes 1-10 from acetylacetonate precursors is not known, it is assumed that the process occurs stepwise, with an early step being the reaction of boron trifluoride molecules with the acetylacetonate ligand to produce a κ 1 -C-acac-Pd species, which then reacts with cyclopentadiene. Evidence supporting this hypothesis has been obtained via examining  4 . The presence of carbon-ligated acac is indicated by the doublet peaks (J = 3.9 Hz) of the methine carbon of the acac ligand at 101.96 ppm in the 13 C{ 1 H} NMR spectrum. The 1 H NMR spectrum of the residue in CDCl 3 shows signals at 2.30 and 6.00 ppm from the methyl and methine protons of the κ 1 -C-acac ligand and characteristic resonance at 3.81 ppm from the 2methoxy-group of the TOMPP ligand. The 31 P{ 1 H} NMR spectrum shows a singlet from the major product at 32.7 ppm shifted downfield when compared to [Pd(acac)(TOMPP) 2 ]BF 4 (17.5 ppm [50]). The obtained precipitate also displays characteristic resonances in the 13 C{ 1 H} NMR spectrum for CO and CH 3 groups of the κ 1 -C-acac ligand (at 192.49 and 24.22 ppm). The 19 F NMR spectrum shows two major sets of signals with an intensity ratio of approximately 3:4, corresponding to the BF 3 and [BF 4 ] − structural fragments in the intermediate. Furthermore, upon addition of cyclopentadiene to the intermediate in CH 2 Cl 2 , green complex 3 is formed along with BF 2 (acac), which was detected in the GC-MS spectra of the solution. Single crystals of compounds 1, 7, and 8 suitable for X-ray crystallography were obtained via slow diffusion of toluene into acetonitrile solutions of the complexes. Figure 1 shows the molecular structures of 1, 7, and 8. Selected interatomic distances and angles are given in the figure captions. In all cases, the asymmetric unit of the crystals contains the complex cation, the [BF 4 ] − anion, and a solvent molecule. The molecular structures of these complexes reveal that the planes containing the metal and the two-ligand phosphorus atoms are almost perpendicular (88.05 • , 89.33 • , 88.05 • for 1, 7, and 8, respectively) to the Cp ring planes. The alignments of the ML 2 planes relative to the Cp rings are typically described as "eclipsed" or "staggered" conformations [27,34], which differ from each other due to a rotation of the cyclopentadienyl ring around the metal-ring centroid vector (Scheme 4). Our study of the [PdCp(PPh 3 ) 2 ] + cation 1 (Figure 1) reveals that it adopts eclipsed conformation. In the molecular structure of complex 1, the cyclopentadienyl C-C bond lengths radiating from C2 (1.447(3) and 1.417(3) Å) are long, and the bond lengths of the C3-C4-C5-C1 sequence show characteristic [27,34] butadiene-like bond length alternation (1.399(3), 1.454(3), and 1.382(4) Å). The variations in the Pd-C distances are also typical of the pattern expected for an eclipse conformation; thus, the Pd-C2 distance of 2.268(2) Å is much shorter than the Pd1-C1 and Pd1-C3 distances (2.372(2) and 2.329(2) Å, respectively). Structural results for compounds 7 and 8 reveal that cations [PdCp(dppp)] + and [PdCp(dppb)] + adopt a geometry that is closer to staggered than eclipsed. For instance, in complex 8, the structural features include three shorter-than-average bond distances (C26-C25 = 1.407(8), C24-C23 = 1.370 (8), and C22-C26 = 1.407(7) Å) and two longer bond lengths (C25-C24 = 1.436(7) and C23-C22 = 1.428(7) Å). In addition, Pd1-C22 and Pd1-C25 are shorter than the other metal-carbon bonds (Figure 1). The Pd-P distances and P-Pd-P bond angles in complex 7 (Pd1-P1(a) = 2.244 Å, ∠P1-Pd-P1a = 94.99 • ) are in good agreement with the corresponding values in complex 8 (Pd1-P1 = 2.263(1) Å, Pd1-P2 = 2.258(1) Å, ∠P1-Pd-P2 = 96.12(4) • ), as well as with those previously reported for cationic cyclopentadienyl palladium-diphosphine complexes [26,27,34]. In the crystal packing of all presented complexes C-H…π and π-π contacts play one of the main structure-directed roles. Cpand Ph-groups give a lot of possibilities to realize a whole potential of this kind interactions. While all presented structures are different due to the difference in the contact types provided by the geometry of coordinated phosphines. This aspect plays the most important role in the crystal packing formation for the presented complexes. In the case of PPh3 (1) intramolecular π-π interactions bound Phgroups of the adjacent phosphine ligands. Intermolecular contacts are of C-H…π nature and involved Ph-groups and Cpligands of adjacent molecules ( Figure 2). In the crystal structure of complex 7, the situation changes dramatically due to the presence of the (CH2)3 linker in the diphosphine ligand. We found strong CH2…π intermolecular contact-    In the crystal packing of all presented complexes C-H . . . π and π-π contacts play one of the main structure-directed roles. Cpand Ph-groups give a lot of possibilities to realize a whole potential of this kind interactions. While all presented structures are different due to the difference in the contact types provided by the geometry of coordinated phosphines. This aspect plays the most important role in the crystal packing formation for the presented complexes. In the case of PPh 3 (1) intramolecular π-π interactions bound Ph-groups of the adjacent phosphine ligands. Intermolecular contacts are of C-H . . . π nature and involved Ph-groups and Cpligands of adjacent molecules ( Figure 2). In the crystal structure of complex 7, the situation changes dramatically due to the presence of the (CH 2 ) 3 linker in the diphosphine ligand. We found strong CH 2 . . . π intermolecular contact-directed formation of 1D supramolecular structures ( Figure 3a) with suppression of π-π interactions. On the other hand, complex 8 has practically the same (CH 2 ) 4 linker (one CH 2 -group longer) which is practically inactive in the formation of intermolecular interactions. Otherwise π-π and C-H . . . π contacts strongly direct orientation of Ph-groups of the diphosphine ligands ( Figure 3b). This fact can be of electronic nature caused by the linker elongation.   QTAIM analysis of model structures demonstrates the presence of bond critical points (3, −1) for intermolecular interactions C-H•••π in the X-ray structures 1, 7, and 8 ( Table 3). The low magnitude of the electron density (0.002-0.009 a.u.), positive values of the Laplacian of electron density (0.008-0.031 a.u.), and zero or very close to zero positive energy density (0.000-0.002 a.u.) in these bond critical points (3, −1) and estimated strength of appropriate short contacts (0.3-1.6 kcal/mol) are typical for very weak noncovalent interactions involving π-systems [59][60][61][62][63][64][65][66][67]. The balance between the Lagrangian kinetic energy G(r) and potential energy density V(r) at the bond critical points (3, −1) (the ratio -G(r)/V(r) ≥ 1) reveals that a covalent contribution in all intermolecular interactions C-H•••π in the X-ray structures 1, 7, and 8 is absent [68] ( Table 3). The Laplacian of electron density is typically decomposed into the sum of contributions along the three principal axes of maximal variation, giving the three eigenvalues of the Hessian matrix (λ1, λ2 and λ3), and the sign of λ2 can be utilized to distinguish bonding (attractive, λ2 < 0) weak In order to confirm or disprove the hypothesis on the existence of intermolecular interactions C-H· · · π in the X-ray structures 1, 7, and 8 and approximately quantify the strength of these supramolecular contacts from a theoretical viewpoint, DFT calculations followed by a topological analysis of the electron density distribution using the QTAIM approach [58] were carried out (see Computational Details and Table S2 in Supporting Information). Results of QTAIM analysis are summarized in Table 3. The contour line diagram of the Laplacian of electron density distribution ∇ 2 ρ(r), bond paths, and selected zero-flux surfaces as well as a visualization of electron localization function (ELF) and reduced density gradient (RDG) analyses for intermolecular interactions C-H· · · π in the X-ray structure 7 are shown in Figure 4. interactions from non-bonding ones (repulsive, λ2 > 0) [69,70]. Thus, intermolecular interactions C-H•••π in the X-ray structures 1, 7, and 8 are attractive (Table 3). Table 3. Values of the density of all electrons-(r), Laplacian of electron density- 2 (r) and appropriate λ2 eigenvalues, energy density-Hb, potential energy density-V(r), and Lagrangian kinetic energy-G(r) (a.u.) at the bond critical points (3, −1), corresponding to intermolecular interactions C-H•••π in the X-ray structures 1, 7, and 8, and estimated strength for these contacts Eint (kcal/mol).  [71] van der Waals radii for H and C atoms are 1.20 and 1.70 Å, respectively, and "modern" values of van der Waals radii suggested by Alvarez [72] H and C atoms are 1.20 and 1.77 Å, respectively. ** Eint ≈ −V(r)/2 [73].  QTAIM analysis of model structures demonstrates the presence of bond critical points (3, −1) for intermolecular interactions C-H· · · π in the X-ray structures 1, 7, and 8 (Table 3). The low magnitude of the electron density (0.002-0.009 a.u.), positive values of the Laplacian of electron density (0.008-0.031 a.u.), and zero or very close to zero positive energy density (0.000-0.002 a.u.) in these bond critical points (3, −1) and estimated strength of appropriate short contacts (0.3-1.6 kcal/mol) are typical for very weak noncovalent interactions involving π-systems [59][60][61][62][63][64][65][66][67]. The balance between the Lagrangian kinetic energy G(r) and potential energy density V(r) at the bond critical points (3, −1) (the ratio -G(r)/V(r) ≥1) reveals that a covalent contribution in all intermolecular interactions C-H· · · π in the X-ray structures 1, 7, and 8 is absent [68] ( Table 3). The Laplacian of electron density is typically decomposed into the sum of contributions along the three principal axes of maximal variation, giving the three eigenvalues of the Hessian matrix (λ 1 , λ 2 and λ 3 ), and the sign of λ 2 can be utilized to distinguish bonding (attractive, λ 2 < 0) weak interactions from non-bonding ones (repulsive, λ 2 > 0) [69,70]. Thus, intermolecular interactions C-H· · · π in the X-ray structures 1, 7, and 8 are attractive (Table 3). Table 3. Values of the density of all electrons-ρ(r), Laplacian of electron density-∇ 2 ρ(r) and appropriate λ 2 eigenvalues, energy density-H b , potential energy density-V(r), and Lagrangian kinetic energy-G(r) (a.u.) at the bond critical points (3, −1), corresponding to intermolecular interactions C-H· · · π in the X-ray structures 1, 7, and 8, and estimated strength for these contacts E int (kcal/mol).
As shown in Table 4, the selectivity and BD conversion largely depended on the ture of the catalyst. The reaction was performed in the presence of an excess of nuc philes relative to the diene ([MeOH]0:[BD]0 = 1), so turnover number (TON) is based butadiene conversion. As one can see from Table 4, for phosphine-ligated complexes, T decreased in the following order: 1, 2 > 3 > 5 > 4 ≫ 6-10. For triarylphosphines as liga this correlates with the decreasing basicity of the phosphine ligand (cf. [90]), ex TOMPP, which is more basic than PPh3 [91]. However, the ease of oxidation of the p phine ligand also increases with its basicity, and therefore, the loss of productivity of catalyst could be explained by a larger loss of the phosphine ligand through oxida during the catalysis as proposed by van Leeuwen et al. [76]. Complexes 6-10 with dip phines were not suitable for this telomerization reaction. It is known that diphosph often perform worse than monophosphine catalysts as ligands for the telomerizatio dienes with alcohols [76,86]. Next, the most interesting pre-catalyst 1 was studied w lower palladium concentration. When only 0.0021 mol% of Pd loading is used, hig turnovers are obtained (23,700) with practically the same selectivity as in entry 1. It sho be noted that at 90 °C, the conversion of butadiene and the selectivity decrease (62% 82% 1-MOD selectivity at 90 °C and 70 °C, respectively), showing that the catalyst form from compound 1 is unstable at high temperatures. Finally, with a [BD]0/[Pd]0 rati 120,000:1 the conversion of BD as well as selectivity substantially dropped, and s amounts of vinylcyclohexene as a result of the Diels-Alder reaction were detected. compound 1, the catalyst selectivities obtained are similar to those reported by van L wen et al. [84], who used a somewhat similar catalyst system Pd 0 (dvd)2/2PPh3/5NaO (dvd-tetramethyldivinyldisiloxane), or by Beller et al. [80], who utilized conventi Scheme 5. Telomerization of 1,3-butadiene with methanol.
As shown in Table 4, the selectivity and BD conversion largely depended on the nature of the catalyst. The reaction was performed in the presence of an excess of nucleophiles relative to the diene ([MeOH] 0 :[BD] 0 = 1), so turnover number (TON) is based on butadiene conversion. As one can see from Table 4, for phosphine-ligated complexes, TON decreased in the following order: 1, 2 > 3 > 5 > 4 6-10. For triarylphosphines as ligands, this correlates with the decreasing basicity of the phosphine ligand (cf. [90]), except TOMPP, which is more basic than PPh 3 [91]. However, the ease of oxidation of the phosphine ligand also increases with its basicity, and therefore, the loss of productivity of the catalyst could be explained by a larger loss of the phosphine ligand through oxidation during the catalysis as proposed by van Leeuwen et al. [76]. Complexes 6-10 with diphosphines were not suitable for this telomerization reaction. It is known that diphosphines often perform worse than monophosphine catalysts as ligands for the telomerization of dienes with alcohols [76,86]. Next, the most interesting pre-catalyst 1 was studied with lower palladium concentration. When only 0.0021 mol% of Pd loading is used, higher turnovers are obtained (23,700) with practically the same selectivity as in entry 1. It should be noted that at 90 • C, the conversion of butadiene and the selectivity decrease (62% vs. 82% 1-MOD selectivity at 90 • C and 70 • C, respectively), showing that the catalyst formed from compound 1 is unstable at high temperatures. Finally, with a [BD] 0 /[Pd] 0 ratio of 120,000:1 the conversion of BD as well as selectivity substantially dropped, and some amounts of vinylcyclohexene as a result of the Diels-Alder reaction were detected. For compound 1, the catalyst selectivities obtained are similar to those reported by van Leeuwen et al. [84], who used a somewhat similar catalyst system Pd 0 (dvd) 2 /2PPh 3 /5NaOMe (dvd-tetramethyldivinyldisiloxane), or by Beller et al. [80], who utilized conventional Pd(OAc) 2 /3PPh 3 /100NEt 3 , albeit with TON being 30% lower in our case (23,500 vs. 30,000-34,000 reported in [80,84]). Notably, these conversions and selectivities were achieved under solventless conditions and without the use of any added base.   For compound 1, the obtained catalyst TONs and selectivity are as might be expected from results by Beller et al. using catalyst system Pd(OAc) 2 /3PPh 3 .
At this point, we decided to test the catalytic activity of compounds 1-10 towards another type of substrate, and the addition polymerization of phenylacetylene (PA) was examined (Scheme 6, Table 5). Polyacetylenes are used in applications such as photonics, light-emitting diodes, conductors or semi-conductors, sensors, gas separation membranes, and chiral materials [92,93]. Poly(phenylacetylene) (PPA) is mainly produced from catalyzed reactions of phenylacetylene with early-and late-transition metal complexes, driven either via metathesis or insertion polymerization mechanisms [92]. Rhodium-based latetransition metal catalysts are commonly used, but they are associated with a high-cost disadvantage [94][95][96]. Therefore, there is a need to develop other catalytic systems, with Pdbased catalysts attracting most attention [92,93,[97][98][99][100]. It has been established [92,93,97] that efficient palladium catalysts for the polymerization of phenylacetylene also fall under the category of cationic organometallic compounds bearing bulky phosphine ligands.  [93]. Therefore, we expected that compounds 1-10 would exhibit interesting catalytic properties during screening.
les 2023, 28, x FOR PEER REVIEW selectivity decreased significantly. Increasing the temperature ab cantly decreased productivity resulting from catalyst deactivatio of compound 3 in the polymerization of phenylacetylene, therefor meation chromatography curves ( Figures S95-S100, SI) of the obt displayed bimodal distribution, which is characteristic of Pd initia The effect of reaction time on conversion, activity, and molecular in the entries in Table 5. The polymerization was quenched at th acidified methanol. As shown, conversion increased with time, same molecular weight was observed independent of conversion larger than expected for living polymerization based on monome dicate that catalyst 3 exhibits slow initiation. Polymerization pr tested (see entries 22-27, Table 5). However, polymerization was acetonitrile, presumably due to its strong coordination to the pall hibits coordination with the monomer. The reaction in THF affo terms of molecular weight (Mw = 23.2 kDa), isolated yield (87%), a PPA samples obtained with catalyst 3 showed a sharp absorpti spectra at 740 cm −1 and a broad peak at 890 cm −1 (Figure S101, SI) [102]. The 1 H NMR spectra of the polymers also show a sharp sing to the vinylic protons in the polymer, and a set of broad peaks at δ (m, 3H) ppm, which are associated with a head-to-tail structure ( Figure S102, SI) [94,101]. Regarding the formation of the active lysts, we suggest that the reaction involves the transformation of ligand to the η 1 -Cp-isomer. In the next step, insertion of the coord C bond occurs. Therefore, the 1 H NMR spectra of PPA show the p at 3.6 ppm, indicative of the Cp fragment from the cyclopentad obtained polyphenylacetylene samples ( Figure S102, SI).

General Procedures and Materials
All air-and/or moisture-sensitive compounds were manip Scheme 6. Polymerization of phenylacetylene with Pd catalysts 1-10.  As shown in Table 5, the polymer yields and molecular weights largely depended on the nature of the catalyst. Typically, low-molecular-weight powdery products were obtained. However, cationic pre-catalyst 3 with a bulky TOMPP ligand produced highmolecular-weight polyphenylacetylene with significant product yield. This is consistent with the known hypothesis that sterically bulky phosphine [ In order to achieve more stable reaction conditions using catalyst 3, we investigated the polymerization of PA while varying temperature, ratio of [PA] 0 :[Pd] 0 , reaction time, and solvent. Polymerization temperature had a significant effect on catalytic activity and the polydispersity index (PDI) of the resulting polymers. As temperature (entries 11-14, Table 5) increased, yield and activity were enhanced; however, molecular weight and selectivity decreased significantly. Increasing the temperature above 80 • C led to significantly decreased productivity resulting from catalyst deactivation. The thermal stability of compound 3 in the polymerization of phenylacetylene, therefore, was modest. Gel permeation chromatography curves (Figures S95-S100, SI) of the obtained PPAs at 25-50 • C displayed bimodal distribution, which is characteristic of Pd initiators [93,99,101,107,108]. The effect of reaction time on conversion, activity, and molecular weight is summarized in the entries in Table 5. The polymerization was quenched at the respective time using acidified methanol. As shown, conversion increased with time, and approximately the same molecular weight was observed independent of conversion. The PDI and M n were larger than expected for living polymerization based on monomer weight, which may indicate that catalyst 3 exhibits slow initiation. Polymerization proceeded in all solvents tested (see entries 22-27, Table 5). However, polymerization was substantially slower in acetonitrile, presumably due to its strong coordination to the palladium center, which inhibits coordination with the monomer. The reaction in THF afforded optimal results in terms of molecular weight (M w = 23.2 kDa), isolated yield (87%), and high TON (435). The PPA samples obtained with catalyst 3 showed a sharp absorption peak in the infrared spectra at 740 cm −1 and a broad peak at 890 cm −1 (Figure S101, SI) characteristic of cis-PPA [102]. The 1 H NMR spectra of the polymers also show a sharp singlet at δ 5.86 (s, 1H) due to the vinylic protons in the polymer, and a set of broad peaks at δ 6.72 (m, 2H) and δ 6.98 (m, 3H) ppm, which are associated with a head-to-tail structure of a cis-transoidal PPA ( Figure S102, SI) [94,101]. Regarding the formation of the active species with such catalysts, we suggest that the reaction involves the transformation of the η 5 -cyclopentadienyl ligand to the η 1 -Cp-isomer. In the next step, insertion of the coordinated PA into the Pd-C bond occurs. Therefore, the 1 H NMR spectra of PPA show the presence of a weak peak at 3.6 ppm, indicative of the Cp fragment from the cyclopentadienyl end group in the obtained polyphenylacetylene samples ( Figure S102, SI).

General Procedure for 1,3-Butadiene Telomerization
All the reactions were carried out in a custom-made stainless-steel autoclave with a volume of 20 mL. The 1,3-butadiene (39 mmol) was first condensed into the reactor and its volume was controlled. The catalyst was then added directly into the reactor pot as a solution in CH 2 Cl 2 . After that, methanol (39 mmol) was added to the reactor via a syringe. The reactor was then closed and placed in an oil bath at a temperature of 70 • C. The reaction was stirred magnetically at 700 rpm for 2 h. The reaction was stopped using an ice bath. The catalytic reactions were analyzed using GC-FID, and the response factor for the telomers was determined using the pure fraction of telomers obtained (identified via GC-MS) after vacuum distillation of the reaction mixture.

General Procedure for Phenylacetylene Polymerization
All polymerizations were conducted in a 5 mL glass reactor equipped with a magnetic stirrer under an Ar atmosphere in an oil bath. The Pd catalyst was stirred in 1 mL of solvent, after which 1 mL of PA (9.1 mmol) was added. The reaction mixture was then stirred for the designated time, and subsequently poured into a large amount of acidified methanol to precipitate the polymer as a yellow powder. The product was isolated via filtration and dried under vacuum to reach a constant weight. If necessary, the polymer was purified through dissolving it in THF and precipitating it with methanol to obtain a fine yellow powder. The polymer samples were characterized using 1 H NMR, FT-IR, and GPC analysis. PPA with high cis-content: 1 H NMR (CDCl 3 ): δ 6.97 (m, 3H), 6.73 (m, 2H), 5.86 (s, 1H).

Computational Details
The single-point calculations based on the experimental X-ray geometries of compounds 1, 7, and 8 have been carried out at the DFT level of theory (ωB97XD/CEP-121G) with the help of the Gaussian-09 [114] program package. The topological analysis of the electron density distribution has been performed using the Multiwfn program (version 3.7) [115]. The Cartesian atomic coordinates for model structures are presented in Table S2.