Borata-Alkene Derived Syntheses of (F5C6)2B-Substituted Bis(indenyl

---

This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article pubs.acs.org/Organometallics

Borata-Alkene Derived Syntheses of (F5C6)2B‑Substituted Bis(indenyl) Group 4 Metal Complexes Sonja Kohrt,† Gerald Kehr,† Constantin G. Daniliuc,† René S. Rojas,‡ Bernhard Rieger,§ Carsten Troll,§ and Gerhard Erker*,† †

Organisch-Chemisches Institut, Westfälische Wilhelms-Universität, Corrensstraße 40, 48149 Münster, Germany Departamento de Química Inorgánica, Facultad de Química, Pontificia Universidad Católica de Chile, Casilla 306, Santiago 22, Chile § WACKER-Lehrstuhl für Makromolekulare Chemie, Technische Universität München, Lichtenbergstraße 4, 85748 Garching bei München, Germany ‡

S Supporting Information *

ABSTRACT: An indene based borata-alkene was used as a reagent for the synthesis of B(C6F5)2-functionalized group 4 bent-metallocene complexes. The respective borata-alkene was prepared by HB(C6F5)2 hydroboration of dimethylbenzofulvene followed by deprotonation with lithium tetramethylpiperidide. Treatment of CpTiCl3 or CpZrCl3 with the borataalkene gave the respective [(C6F5)2B-indenyl]CpMCl2 bentmetallocene complexes. Both were characterized by X-ray diffraction. The boryl-functionalized metallocenes gave active ethene or propene polymerization catalysts, respectively, upon activation with triethylaluminum.

■

INTRODUCTION Group 4 bent metallocenes bearing pendant boryl functionalitites show an interesting organometallic chemistry.1 Some have found use in catalysis, especially in homogeneous group 4 metallocene based Ziegler−Natta olefin polymerization.2,3 Mostly, the BR2 functional groups are found attached through connecting hydrocarbyl chains, ranging from −CH2−1a,f to ethylene-4 and oligomethylene-bridged systems1b−d,g−i,3a and sometimes also involving unsaturated organic linkers.1f Most frequently those systems were obtained by means of hydroboration reactions of the respective bent metallocenes featuring the respective alkenyl or alkynyl side chains attached at their Cp rings. This route was especially favored for the attachment of the terminal B(C6F5)2 Lewis acid function using Piers’ borane (HB(C6F5)2)5 as the hydroboration reagent. There were also systems reported that had the BR2 functionality bonded directly at the η5-Cp, η5-indenyl, or η5fluorenyl ligands of the group 4 bent metallocenes or constrained-geometry complexes.6 This was in most cases achieved by synthetic routes involving e.g. nucleophilic attack at suitable XBR2 or X2BR reagents.7 This is straightforward in the case of, for example, the respective dialkylamino boranes (e.g., R = NMe2), since selective functionalization is greatly facilitated by the mesomeric stabilization of the Lewis acid borane site by the NR2 substituent. Subsequent deprotonation followed by reaction with a group 4 metal halide reagent or simply treatment with a metal amido complex followed by exchange of the residual metal amido groups for halide by treatment with a chlorosilane then may selectively lead to the respective −[B]− NMe2 substituted bent-metallocene complexes.8,9 © XXXX American Chemical Society

It seems to be more difficult to prepare group 4 bent metallocenes with the strongly Lewis acidic B(C 6 F 5 ) 2 functionalities directly attached at the Cp or indenyl ligands. Bochmann et al. developed a synthesis of the B(C6F5)2substituted indene 1. Its reaction with Zr(NMe2)4, for example, resulted in attachment of a borylindenyl ligand at zirconium, but during this process a C6F5 group at boron was exchanged for the strongly stabilizing −NMe2 group, leading to complex 2 (see Scheme 1).10 Reetz et al. prepared a rare example of a group 4 metal complex containing the Cp−B(C6F5)2 ligand.2 This synthesis followed a reaction scheme devised by Jutzi and Seufert:11 treatment of the silyl-/stannyl-substituted cyclopentadiene 4 with ClB(C6F5)2 gave the reagent 5, which was used in a chlorodesilylation reaction with ZrCl4 to form the doubly Scheme 1

Received: May 27, 2016

A

DOI: 10.1021/acs.organomet.6b00427 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics

were obtained from toluene/pentane by the diffusion method. This analysis shows the central bent-metallocene structure with a Cp ring and the substituted indenyl ring both η5 coordinated to the TiCl2 unit. The (η5-Cp)−Ti bonding is quite uniform with the C(Cp)−Ti bond lengths found in a narrow range between 2.335(5) and 2.382(5) Å, whereas the coordination of the indenyl five-membered ring is somewhat distorted. We note short bonds in the usual range between Ti and carbon atoms C3, C2, and C3A, whereas the Ti1−C1 and especially the Ti1− C7A bonds are longer (see Table 1). This is probably caused by

B(C6F5)2 functionalized zirconocene dichloride complex 6 (Scheme 1). Bochmann et al. used a variation of this route to successfully prepare the complex [Cp-B(C6F5)2]TiCl3 (3).10 Complexes 2 and 3 were both shown to generate “selfactivating” homogeneous Ziegler−Natta ethene polymerization catalysts upon treatment with excess triethylaluminum. Lancester et al. used compound 3 as the starting material for the synthesis of a series of [Cp-B(C6F5)2]metal complexes, usually isolated as stabilized derivatives (with pyridine, dimethyl sulfide, or internal chloride coordination).12 These synthetic entries have in common that they need to make use of the (F5C6)2B−halide reagents, which are difficult to make and handle.13 We thought that it would be much nicer if we had a route to such systems which would only involve the [HB(C6F5)2] reagent (such as in the −(CH2)n−B(C6F5)2 examples discussed above), which is conveniently obtained by Piers’ method and is easy to handle. We here describe such a synthetic entry to group 4 bis(indenyl−B(C6F5)2) metal complex examples.

Table 1. Selected Structural Parameters of the Group 4 BentMetallocene Complexes 10−12a MX2 M−Cl1 M−Cl2 M−C1 M−C2 M−C3 M−C3A M−C7A C1−C2 C2−C3 C3−C3A C3A−C7A C1−C7A C3−B1

■

RESULTS AND DISCUSSION We had previously shown that hydroboration of the dimethylbenzofulvene 7 with Piers’ borane (HB(C6F5)2) under thermodynamic control (several hours at room temperature) followed by deprotonation with the bulky amide base lithium tetramethylpiperidide (LiTMP) gave the borata-alkene system 8 in good yield (Scheme 2).14,15 The X-ray crystal Scheme 2

a

10

12

11

TiCl2 2.321(1) 2.472(1) 2.402(4) 2.332(4) 2.282(4) 2.382(4) 2.514(4) 1.417(5) 1.411(5) 1.463(5) 1.428(5) 1.437(5) 1.596(6)

ZrCl(OH) 2.443(1) 2.138(3)b 2.547(5) 2.463(5) 2.398(4) 2.477(5) 2.575(5) 1.423(7) 1.422(7) 1.436(7) 1.441(7) 1.423(7) 1.608(7)

ZrCl2 2.433(1) 2.399(1) 2.623(3) 2.509(2) 2.496(2) 2.625(3) 2.673(3) 1.385(4) 1.449(4) 1.469(4) 1.424(3) 1.453(4) 1.505(4)

Bond lengths in Å. bZr1−O1.

the unsymmetrical bonding situation at the front of the bentmetallocene wedge. There we found a terminal Ti1−Cl1 unit and a Ti1−Cl2−B1 bridging chloride (B1−Cl2 2.034(5) Å, angle Ti1−Cl2−B1 85.9(1)°). The Cl1−Ti1−Cl2 angle in complex 10 amounts to 91.9(5)°. The Cl2−B1 contact has caused a distorted-tetrahedral coordination geometry at the boron atom (angles Cl2−B1−C3 94.4(3)°, C11−B1−C21 109.6(3)°, ∑B1CCC 343.2°) (see Figure 1). The 11B NMR signal at δ 6.2 confirms the presence of a tetracoordinated boron atom in complex 10. Due to the element of planar chirality of the substituted indenyl−Ti moiety, we have monitored the 1H/13C NMR signals of

structure analysis and the NMR spectroscopic characterization indicated that the anion 8 has to be regarded as a conjugatively stabilized borata-alkene6b,16,17 rather than a boryl-substituted indenyl anion. Our present study now showed that nevertheless the reagent 8 reacted cleanly with the group 4 metal halide reagents CpMCl3 (M = Ti (9a), M = Zr (9b)) under mild reaction conditions to give the respective B(C6F5)2-substituted bent-metallocene complexes. The borata-alkene 8 (with Li[HTMP]+ countercation) was treated with 1 molar equiv of CpTiCl3 (9a) suspended in toluene at ambient temperature. Workup involving removal of the lithium chloride by filtration, evaporation of the solvent, and washing with pentane eventually gave complex 10 as an olive green solid in 54% yield. The complex was characterized by C, H elemental analysis, by spectroscopy, and by X-ray diffraction. Single crystals for the X-ray crystal structure analysis

Figure 1. Projection of the molecular structure of complex 10 (thermal ellipsoids are shown with 15% probability). B

DOI: 10.1021/acs.organomet.6b00427 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics diastereotopic methyl groups of the isopropyl substituent at the indenyl core. The C6F5 groups at the tetracoordinated boron atom are likewise diastereotopic (for details see the Supporting Information). We then reacted the borata-alkene 8 with CpZrCl3 (9b).18 The reaction was performed analogously as in the Ti case. Workup eventually gave the zirconocene complex 11 as a yellow solid in 68% yield. The compound was also characterized by C, H elemental analysis, by spectroscopy, and by X-ray diffraction. Our attempts to get single crystals suitable for an X-ray crystal structure analysis initially furnished a few crystals that turned out to be the hydrolysis product 12, where apparently the reaction with 1 equiv of water had taken place with elimination of 1 molar equiv of HCl. Since we had only a few crystals of this serendipitously formed material, it was only characterized by X-ray diffraction (see Figure 2). The

Figure 3. Molecular structure of the zirconium complex 11 (thermal ellipsoids are shown with 15% probability).

In solution (d6-benzene) complex 11 shows a 1H/13C NMR Cp resonance at δ 5.88/117.4. It shows a 11B NMR signal at δ 55.7, which is typical of a planar tricoordinate R−B(C6F5)2 species. At 299 K we have monitored three broad 19F NMR resonances with Δδ(19F) = 10.1. Lowering the monitoring temperature resulted in decoalescence of the p-C6F5 signals (193 K: δ −147.1, −150.5), as expected for hindered rotation about the B−indenyl vector. In addition, hindered rotation around the B−C6F5 vectors also sets in at low temperature, which eventually leads to the observation of a total of 10 19F NMR signals of the B(C6F5)2 moiety in complex 11. Complex 11 has a planar-chiral (indenyl)Zr moiety, and consequently we have monitored the 1H/13C NMR signals of a pair of diastereotopic isopropyl methyl groups (for further details see the Supporting Information). Olefin Polymerization Reactions. The B(C6F5)2-functionalized metallocenes 10 and 11 were used as transition-metal components in the homogeneous metallocene Ziegler−Natta polymerization of ethene and propene, respectively (Scheme 3).19 Both of these complexes serve as “self-activating” systems

Figure 2. Molecular structure of the hydrolysis product 12 (thermal ellipsoids are shown with 15% probability).

compound has a structure similar to that of the titanocene complex 10, only in this case we find a OH group bridging between boron and the group 4 metal atom (here Zr). For further structural details see Table 1 and the Supporting Information. We were able to obtain single crystals of the intact complex 11 by crystallization under rigorously dry conditions. The X-ray crystal structure analysis showed a bent-metallocene complex with a Cp ligand (Zr1−C(Cp)) featuring bond lengths between 2.473(9) and 2.513(7) Å and the substituted indenyl ligand (Zr−C bond lengths to the five-membered ring between 2.496(2) and 2.673(3) Å; see Table 1). The structure of the zirconocene complex 11 differs markedly from that of its Ti congener 10: it lacks any direct Cl···B contact (B1···Cl2 separation 4.68 Å) (Figure 3). The ZrCl2 unit in complex 11 shows metal−halogen bond lengths of 2.399(1) Å (Zr1−Cl2) and 2.433(1) Å (Zr1−Cl1), respectively (Cl1−Zr1−Cl2 angle: 96.0(1)°). The boron atom at the indenyl ring shows a trigonalplanar coordination geometry (∑B1CCC 359.6°). The B1−C3 bond to the indenyl five-membered ring is slightly shorter than that of the B1−C11/C21 bonds to the C6F5 substituents (1.592(4) Å/ 1.588(4) Å), and we note that the C1−C2 and B1−C3 bonds in compound 11 are rather short. Taken together, this may probably indicate some slight residual boratafulvene character in the structure of this bent-metallocene complex.

Scheme 3

in the presence of excess AlEt3.2,10,12 Ethene polymerization with the titanium catalyst system 10/AlEt3 gave only low activities, whereas the zirconium catalyst system 11/AlEt3 gave polyethylene with high activities under suitable reaction conditions (see Table 2). The system 11/AlEt3 was also an active catalyst for propene polymerization (see Table 2). 13C NMR pentad analysis (see Figure 4) revealed isotactic polypropylene formation by enantiomorphic site control with a ca. 77:23 m/r diad ratio (α = 85.6; for details see the Supporting Information).20

■

CONCLUSIONS We had shown that a sequence of hydroboration/deprotonation starting from benzofulvene 7 with the bulky Li+[TMP]− base resulted in the formation of borata-alkene salt 8 (with C

DOI: 10.1021/acs.organomet.6b00427 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics Table 2. Olefin Polymerization with the Catalyst Systems 10/AlEt3 and 11/AlEt3a entry

precatalyst

monomer

pressure (bar)

temp (°C)

time (h)

amt (mg)

activityb

mp (°C)

1 2 3 4 5 6 7 8

Cp2ZrCl2 10 10 11 11 11 11 11

ethene ethene ethene ethene ethene ethene ethene propene

2 30 30 2 30 30 30 8

20 20 50 20 20 20 50 20

72 2 2 18