Unprecedentedly High Activity and/or High Regio-/Stereoselectivity of Fluorenyl-Based CGC Allyl-Type η3:η1-tert-Butyl(dimethylfluorenylsilyl)amido Ligated Rare Earth Metal Monoalkyl Complexes in Olefin Polymerization

A series of fluorenyl-based constrained-geometry-configuration (CGC) allyl-type rare earth metal monoalkyl complexes bearing the divalent anionic η3:η1-tert-butyl(dimethylfluorenylsilyl)amido (η3:η1-FluSiMe2NtBu) ligand (η3:η1-FluSiMe2NtBu)Ln(CH2SiMe3)(THF)2 (1–3) have been synthesized via the alkane elimination reaction between the FluHSiMe2NHtBu ligand and rare earth metal tri(trimethylsilylmethyl) complexes Ln(CH2SiMe3)3(THF)n. Their structures are characterized by means of NMR spectrum, elemental analyses, and X-ray diffraction. These complexes 1–3 are isostructural and isomorphous, and each of them adopts a distorted-trigonal-bipyramidal configuration containing one η3:η1-FluSiMe2NtBu ligand, one CH2SiMe3 ligand, and two THF molecules. Unlike traditional CGC allyl-type rare earth metal complexes showing no or low activity and regio-/stereoselectivity in styrene or MMA polymerization, these complexes 1–3 exhibit high catalytic activities and/or high regio-/stereoselectivities in the cis-1,4-polymerization of isoprene and myrcene or in the syndiotactic polymerization of styrene under the aid of different activators (borate or borane) and AlR3. The in situ 1H NMR spectra suggest that the exchanges of chelating ligands such as alkyl groups and divalent anionic η3:η1-FluSiMe2NtBu ligands between rare earth metal centers and Al centers result in the formation of a heterobimetallic tetraalkylaluminate complex R2Al(μ-R)2Ln(R)(μ-R)2AlR2, which is activated by activators to form a divalent cationic species [Ln(μ-R)2AlR2]2+ as a catalytically active species in the coordination–insertion polymerization of olefins.


Introduction
The development of highly efficient and highly regio-/stereoselective rare earth metal catalysts has become a hot topic in the coordination-insertion polymerization of olefin over the past two Recently, we have paid much attention to the synthesis of the half-sandwich Flu-ligated rare earth metal dialkyl complexes Flu'Ln(CH 2 SiMe 3 ) 2 (THF) n and their applications in the coordination-insertion (co)polymerization of olefins such as ST or conjugated dienes. In 2013, these complexes displayed high activities up to 3.4 × 10 7 (g of polymer)/(mol Ln h) and syndiotacticities up to >99% in ST polymerization when activated by an activator with or without a small amount of Al i Bu 3 [30]. Moreover, such complexes also showed very high activities up to 1.9 × 10 7 (g of polymer)/(mol Ln h) and high cis-1,4-selectivities, 93% in the polymerization of isoprene (IP) in the presence of activator and AlR 3 [31]. In addition, such catalysts were also active in the regioselective polymerization of 1,3-cyclohexadiene and copolymerization with ST and IP [32]. These results demonstrate that the effective adjustment of the skeleton of the Flu ligand of these complexes has an important impact on their catalytical performance in the olefin polymerization, which arouses our interests to explore more Flu-based rare earth metal complexes and detect their catalytic performance in olefin polymerization. Herein, we report the synthesis and structural characterization of three Flu-based CGC allyl-type rare earth metal monoalkyl complexes (η 3 :η 1 -FluSiMe 2 N t Bu)Ln(CH 2 SiMe 3 )(THF) 2 1-3 (1: Ln = Sc; 2: Ln = Lu; 3: Ln = Y) via the alkane elimination reaction between the tert-butyl(dimethylfluorenylsilyl)amido (FluHSiMe 2 NH t Bu) ligand and the rare earth metal trialkyl complexes Ln(CH 2 SiMe 3 ) 3 (THF) n . Activated by different cocatalysts, these complexes 1-3 unprecedentedly exhibit high activities and high regio-/stereoselectivities in the polymerization of IP, myrcene (MY), or ST, affording the cis-1,4-poly(conjugated diene)s or syndiotactic polystyrenes with high molecular weights and moderate molecular weight distributions. The possible coordination-insertion polymerization mechanism is investigated by means of the in situ 1 H NMR spectrum.

Method
By using J. Young valve NMR tubes, the samples of rare earth metal catalysts were prepared for NMR spectroscopic measurements in the glove box. 1 H, 13 C NMR spectra of ligand and catalysts were tested on a Bruker AVANCE 400 spectrometer in C 6 D 6 or C 7 D 8 at room temperature. 1 H, 13 C NMR spectra of polyisoprene (PIP), polymyrcene (PMY) and polystyrene (PST) samples were recorded on a Bruker AVANCE 400 spectrometer in CDCl 3 at room temperature or at 60 • C. The molecular weights and the molecular weight distributions (PDI) of the poly(conjugated dienes)s were performed at 25 • C by gel permeation chromatography (GPC) on a WATERS 1515 apparatus. THF was selected as the eluent at a flow rate of 1 mL/min. For SPSTs, GPC data were performed in 1,2,4-trichlorobenzene at 150 • C using IR detection and calibration against polystyrene. Differential scanning calorimetry (DSC) measurements were carried out on a TA 60 (TA Co.) at a rate of 10 • C/min. Any thermal history difference in the poly(conjugated diene)s was eliminated by first heating the specimen to 100 • C, cooling at 10 • C/min to -100 • C, and then recording the second DSC scan. For SPSTs, DSC parameter was set to 10 • C/min to speed up to 300 • C, then cooled at 10 • C/min to room temperature, before recording the second DSC scan. Elemental analyses were performed on an Elementary Vario MICRO CUBE (Germany).

X-ray Crystallographic Analysis
The crystals of complexes 1-3 were oil sealed under a microscope in the glove box. For data collection at -100 • C, a CCD area detector using graphite-monochromated Mo Kα radiation (λ = 0.71073 Å) was chosen on a Bruker Smart-Apex CCD diffractometer. The SMART program package was used to determine the crystal class and unit cell parameters. SAINT and SADABS were adopted to process the original frame data and generated the reflection data file. Shelxtl-97 program was applied to solve the structure. F2 anisotropic non-hydrogen atoms were refined by using the full matrix least square method. All the non-hydrogen atoms were anisotropy refined, and all hydrogen atoms were introduced in the calculated positions and were included in the structure calculation without further refinement of the parameters. Crystallographic data (excluding structure factors) have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication nos. CCDC-1904924 (1), CCDC-1904923 (2), and CCDC-1904925 (3) containing the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif from The Cambridge Crystallographic Data Centre.

A Typical Procedure for MY Polymerization in
In a glovebox at 25 • C, toluene solution (3.5 mL), Al i Bu 3 (100 µL, 1.0 M, 100 µmol), complex 3 (0.0063 g, 10 µmol), a toluene solution (1.5 mL) of B(C 6 F 5 ) 3 (0.0052 g, 10 µmol), and MY (0.68 g, 5 mmol) were added into a 50 mL round bottom flask in succession. The reaction system became sticky rapidly. After 1 h, the flask was taken outside and then quenched by addition of ethanol (50 mL, containing 5% butylhydroxytoluene (BHT) as stabilizing agent). The mixture was washed with ethanol and then dried under vacuum at 45 • C to a constant weight (0.49 g, yield = 72%). The resulting polymer was soluble in THF and chloroform at room temperature. The isomer contents of the PMY products were calculated from the 1 H NMR spectra (Formulas (6)-(8) and 13 C NMR spectra (Formulas (9)- (12 Table 4 Entry 9 In a glovebox at 25 • C, toluene solution (3.5 mL), Al i Bu 3 (100 µL, 1.0 M, 100 µmol), complex 1 (0.0057 g, 10 µmol), a toluene solution (1.5 mL) of [PhNHMe 2 ][B(C 6 F 5 ) 4 ] (0.0081 g, 10 µmol), and ST (0.52 g, 5 mmol) were added into a 50 mL round bottom flask in succession. Some solids were gradually precipitated from the reaction system. After 20 h, the flask was taken outside and then quenched by addition of ethanol (50 mL, containing 5% butylhydroxytoluene (BHT) as stabilizing agent). The mixture was washed with ethanol and then dried under vacuum at 45 • C to a constant weight (0.22 g, yield = 43%). The resulting polymer is soluble in CHCl 3 and 1,2,4-trichlorobenzene at high temperature. The isomer contents of the polystyrene products were calculated from the 1 H and 13 C NMR spectra according to the following Formulas (13) and (14): in which I C1 is the integration of the resonance at 146.8 ppm (mmmm) and I C2 is the integration of the resonance at 145.4 ppm (rrrr) in the 13 C NMR spectrum.

Structural Characterization of Flu-Based CGC Allyl-Type Rare Earth Metal Monoalkyl Complexes 1-3
These complexes 1-3 have good solubilities in common organic solvents such as hexane, toluene and THF. In the 1 H NMR spectra of the complexes 1-3 in C7D8 and C6D6, the disappearance of the proton signals attributed to the Flu−H and N−H group of the FluHSiMe2NH t Bu ligand suggests the generation of a dianionic chelating ligand in these complexes. Moreover, the eight Flu protons are divided into four peaks, indicating an asymmetric coordination mode of the Flu ligand around the metal center [29]. In each case, the molar ratio of the integral areas of the signals for the FluSiMe2N t Bu ligand, the CH2SiMe3 ligand, and THF molecules is 1:1:2. Similar to the flexible [(3,6t Bu2C13H6)SiR2N t Bu]Y(CH2SiMe3)(THF)2 at room temperature [27], these complexes also have a flexible structure and the CH2SiMe3 group in these complexes can not be fixed at the NMR time scale at room temperature since the two methylene protons of the Ln-CH2SiMe3 groups show only a singlet at high field for 1 at -1.09 ppm, for 2 at -0.82 ppm, and for 3 at -1.06 ppm, respectively.
In a mixed toluene/hexane solution at -30 °C, single crystals of the complexes 1-3 were cultivated for an X-ray determination. The ORTEP (Oak Ridge Thermal-Ellipsoid Plot Program) drawings of the complexes 1-3 are shown in Figure 1 and the representative bond distances and angles are summarized in Table 1. The X-ray diffraction study reveals that these complexes 1-3 are isomorphous and isostructural. Similar to the previous Flu-based CGC allyl-type complex [η 3 :η 1 -(3,6t Bu2C13H6)SiR2N t Bu]Y(CH2SiMe3)(THF)2 [27], each of them contains one dianionic FluSiMe2N t Bu ligand, one CH2SiMe3 group, and two coordinated THF molecules and adopts a distorted-trigonalbipyramidal configuration. Moreover, the 9-position carbon atom (C1) and the two adjacent carbon atoms (C2, C3) of one phenyl (Ph) ring of the Flu ligand are bound to the metal center in an asymmetric η 3 -allyl mode. The distance of Ln-C1

Structural Characterization of Flu-Based CGC Allyl-Type Rare Earth Metal Monoalkyl Complexes 1-3
These complexes 1-3 have good solubilities in common organic solvents such as hexane, toluene and THF. In the 1 H NMR spectra of the complexes 1-3 in C 7 D 8 and C 6 D 6 , the disappearance of the proton signals attributed to the Flu−H and N−H group of the FluHSiMe 2 NH t Bu ligand suggests the generation of a dianionic chelating ligand in these complexes. Moreover, the eight Flu protons are divided into four peaks, indicating an asymmetric coordination mode of the Flu ligand around the metal center [29]. In each case, the molar ratio of the integral areas of the signals for the FluSiMe 2 N t Bu ligand, the CH 2 SiMe 3 ligand, and THF molecules is 1:1:2. Similar to the flexible [(3,6-t Bu 2 C 13 H 6 )SiR 2 N t Bu]Y(CH 2 SiMe 3 )(THF) 2 at room temperature [27], these complexes also have a flexible structure and the CH 2 SiMe 3 group in these complexes can not be fixed at the NMR time scale at room temperature since the two methylene protons of the Ln-CH 2 SiMe 3 groups show only a singlet at high field for 1 at -1.09 ppm, for 2 at -0.82 ppm, and for 3 at -1.06 ppm, respectively.

Cis-1,4-Polymerization of Ip by the Complexes 1-3/activator/AlR3 Ternary Systems
The complexes 1-3 alone, the complexes 1-3/AlR3 binary systems, and the complexes 1-3/activator binary systems were inactive in IP polymerization. In the presence of both activator and AlR3, however, these complexes 1-3 unprecedentedly exhibited high activities and regio-/stereoselectivities in IP polymerization under mild conditions as shown by 1 H and 13 C NMR analysis ( Table 2 and supporting information). At the very beginning, the best catalytic system was investigated for IP polymerization. At first, the Y complex 3 and 2.  (Table 2, entry 2). By contrast, neutral borane B(C6F5)3 (C) was inert for the polymerization of IP even in a long polymerization time ( Table 2, entry 3). In order to prepare PIPs with high cis-1,4-selectivities, the borate A was chosen as an optimum activator in the following IP polymerization. In the presence of borate A and 10 equivalent units of Al i Bu3, high activity

Cis-1,4-Polymerization of Ip by the Complexes 1-3/activator/AlR 3 Ternary Systems
The complexes 1-3 alone, the complexes 1-3/AlR 3 binary systems, and the complexes 1-3/activator binary systems were inactive in IP polymerization. In the presence of both activator and AlR 3 , however, these complexes 1-3 unprecedentedly exhibited high activities and regio-/stereoselectivities in IP polymerization under mild conditions as shown by 1 H and 13 C NMR analysis (Table 2 and supporting information). At the very beginning, the best catalytic system was investigated for IP polymerization. At first, the Y complex 3 and 2.5 equivalent units of Al i Bu 3 were fixed for screening of activators. As an activator, trityl borate [Ph 3 C][B(C 6 F 5 ) 4 ] (A) showed low activity, approximately 4 × 10 3 (g of polymer)/(mol Ln h) and moderate cis-1,4-selectivity of approximately 83% in 30 min. (Table 2, entry 1), while anilinum borate [PhNHMe 2 ][B(C 6 F 5 ) 4 ] (B) displayed moderate activity, approximately 18 × 10 3 (g of polymer)/(mol Ln h) and low cis-1,4-selectivity of approximately 65% under the same condition ( Table 2, entry 2). By contrast, neutral borane B(C 6 F 5 ) 3 (C) was inert for the polymerization of IP even in a long polymerization time ( Table 2, entry 3). In order to prepare PIPs with high cis-1,4-selectivities, the borate A was chosen as an optimum activator in the following IP polymerization. In the presence of borate A and 10 equivalent units of Al i Bu 3 , high activity approximately 1.1 × 10 6 (g of polymer)/(mol Ln h) and high cis-1,4-selectivity of approximately 90% were obtained in IP polymerization catalyzed by the Sc complex 1 only in 2 min, affording main cis-1,4-PIP with high molecular weight and moderate molecular weight distribution (M n = 700 kg/mol, M w /M n = 2.26) ( Table 2, entry 4). In comparison, the Lu complex 2 and the Y complex 3 had moderate activities (1.6 × 10 4 -1.7 × 10 4 (g of polymer)/(mol Ln h)) and similar or slightly lower cis-1,4-selectivities (86%-90%) under the same conditions to prepare main cis-1,4-PIPs with lower molecular weights and broader molecular weight distributions (M n : 100-300 kg/mol, M w /M n : 3.17-3.26) ( Table 2, entries 5-6). Therefore, the Sc complex 1 served as an optimized catalyst in the following polymerization of IP.
All of the resulting PIPs have good solubilities in THF and CHCl 3 . The 1 H NMR spectra of these PIPs in CDCl 3 indicate the presence of main 1,4-PIP unit and a trace amount of 3,4-PIP unit. The 13 C NMR spectra of these PIPs give diagnostic signals assigned as main cis-1,4 configuration (δ = 23.6, 26.

Polymerization Mechanism Study
In general, the catalytically active species in the coordination-insertion polymerization of olefin is usually generated from the rare earth metal dialkyl/dihalides complex and activator with or without AlR 3 (Chart 2a,b) [8]. In such catalytic systems, an activator usually eliminates one alkyl group from the metal center to afford cationic species containing a metal-alkyl (Ln-R) bond for the coordination and insertion of olefin monomer to finally give polyolefin with different regio-/stereoselectivity and microstructure. Unlike an activator, AlR 3 can perform a lot of functions in olefin polymerization, such as scavenging impurities, transforming (alkylating and reducing) the cationic species, and/or acting as a chain transfer agent [34][35][36]. Later, AlR 3 , especially AlMe 3 , was found to react with the rare earth metal trialkyl/dialkyl complex to form a heterobimetallic tetraalkylaluminate complex as a catalyst precursor. Activated by an activator, the corresponding cationic heterobimetallic tetraalkylaluminate complex was formed as a truly active species in the olefin polymerization (Chart 2b) [37][38][39]. Recently, AlR 3 was found to remove coordinated solvent molecules such as THF or pyridine from the metal center or transfer the anionic chelating ligand from the rare earth metal center (Chart 2b) [40][41][42][43]. But in comparison, the catalytically active species in the coordination-insertion polymerization of olefins by use of rare earth metal monoalkyl complex/activator/AlR 3 ternary system is difficult to calculate and understand (Chart 2c). According to the conventional synthesis method a, the resulting cationic species does not have the alkyl group, which inhibits coordination and insertion of olefin monomer into the Ln-R bond. As a result, the high regio-/stereoselective polymerization of olefins can't occur. Therefore, some special reaction must be happened during the formation of cationic active species. More recently, we found that in the cis-1,4-polymerization of isoprene catalyzed by the dipyrromethene ligated scandium monoalkyl complex/activator/AlR 3 ternary system, both of two anionic dipyrromethene chelating ligands transferred from the Sc center to the Al center. This was observable by using naked eyes, UV irradiation, fluorescence spectrum, and in situ 1 H NMR spectrum, and affords a catalyst precursor heterobimetallic tetraalkylaluminate complex (Chart 2d) [40]. Then, one alkyl group of such catalyst precursor was removed by an activator to form the cationic heterobimetallic tetraalkylaluminate complex as a truly active species in cis-1,4-polymerization of IP (Chart 2d). In this paper, these Flu-based CGC allyl-type rare earth metal monoalkyl complexes/activator/AlR 3 ternary system also exhibited high regio-/stereoselectivities and/or high activities in the polymerization of olefins such as IP, MY, and ST. The catalytically active species of such ternary systems in the coordination-insertion polymerization of olefins also aroused our interest. Therefore, the polymerization initiation processes by the Y complex 3/activator/AlR 3  The analysis of the in situ 1 H NMR spectra of the active species generated from the Y complex 3/[Ph3C][B(C6F5)4]/AlMe3 ternary system in d-toluene at 25 °C was taken as an example ( Figure 2). Firstly, the reaction between the Y complex 3 and 5 equivalent units of AlMe3 was carried out in a J. Young valve NMR tube for 5 min consistent with polymerization procedure (Figure 2, E). The in situ 1 H NMR spectrum demonstrated that the peaks assigned to the dianionic FluSiMe2N t Bu ligand became weak and moved to the high field. Moreover, the position of peaks assigned to the coordinated THF molecules and AlMe3 had obvious changes. It was very interesting that this 1 H NMR spectrum very much looked like the in situ 1 H NMR spectrum of the reaction of Y(CH2SiMe3)3(THF)2 and 5 equivalent units of AlMe3 after 5 min in d-toluene at 25 °C, in which the heterobimetallic tetramethylaluminate complex Y(Me)[(μ-Me)2Al(Me)2]2 was formed with a broad signal at -0.3 ppm for all of the methyl groups (slightly different with AlMe3) in addition with the byproduct Al(CH2SiMe3)3 (Figure 2, F). These results implied that the Y complex 3 had decomposed during this Chart 2. Coordination-insertion polymerization of olefin by using the different rare earth metal catalysts.
The analysis of the in situ 1 H NMR spectra of the active species generated from the Y complex 3/[Ph 3 C][B(C 6 F 5 ) 4 ]/AlMe 3 ternary system in d-toluene at 25 • C was taken as an example ( Figure 2). Firstly, the reaction between the Y complex 3 and 5 equivalent units of AlMe 3 was carried out in a J. Young valve NMR tube for 5 min consistent with polymerization procedure ( Figure 2E). The in situ 1 H NMR spectrum demonstrated that the peaks assigned to the dianionic FluSiMe 2 N t Bu ligand became weak and moved to the high field. Moreover, the position of peaks assigned to the coordinated THF molecules and AlMe 3 had obvious changes. It was very interesting that this 1 H NMR spectrum very much looked like the in situ 1 H NMR spectrum of the reaction of Y(CH 2 SiMe 3 ) 3 (THF) 2 and 5 equivalent units of AlMe 3 after 5 min in d-toluene at 25 • C, in which the heterobimetallic tetramethylaluminate complex Y(Me)[(µ-Me) 2 Al(Me) 2 ] 2 was formed with a broad signal at -0.3 ppm for all of the methyl groups (slightly different with AlMe 3 ) in addition with the byproduct Al(CH 2 SiMe 3 ) 3 ( Figure 2F). These results implied that the Y complex 3 had decomposed during this reaction. In view that no free FluHSiMe 2 NH t Bu ligand was observed from the above in situ 1 H NMR spectrum, the dianionic FluSiMe 2 N t Bu ligand should precipitate from the polymerization solvent. Then 1 equivalent unit of Taking above results into account, a plausible coordination-insertion mechanism is proposed for the regio-/stereoselective polymerization of olefins catalyzed by the Flu-based CGC allyl-type rare earth monoalkyl complexes (η 3 :η 1 -FluSiMe2N t Bu)Ln(CH2SiMe3)(THF)2 (1-3)/activator/AlR3 ternary systems in Scheme 1. At first, 2 equivalent units of AlR3 abstracts two coordinated THF molecules from the rare earth metal center of these complexes 1-3 to produce intermediate (η 3 :η 1 - Taking above results into account, a plausible coordination-insertion mechanism is proposed for the regio-/stereoselective polymerization of olefins catalyzed by the Flu-based CGC allyl-type rare earth monoalkyl complexes (η 3 :η 1 -FluSiMe 2 N t Bu)Ln(CH 2 SiMe 3 )(THF) 2 (1-3)/activator/AlR 3 ternary systems in Scheme 2. At first, 2 equivalent units of AlR 3 abstracts two coordinated THF molecules from the rare earth metal center of these complexes 1-3 to produce intermediate (η 3 :η 1 -FluSiMe 2 N t Bu)Ln(CH 2 SiMe 3 ) (a). Then the exchange of the CH 2 SiMe 3 group of a and alkyl group of AlR 3 gives a new rare earth metal alkyl complex (η 3 :η 1 -FluSiMe 2 N t Bu)LnR (c) with the release of AlR 2 (CH 2 SiMe 3 ). The continued ligand exchange between metal center of 0.5 equivalent units of the above intermediate c and Al center of AlR 3 forms 0.5 equivalent units of a heterobimetallic tetraalkylaluminate complex (R) 2 Al(µ-R) 2 Ln(R)(µ-R) 2 Al(R) 2 (h) as catalyst precursor and 0.5 equivalent units of (η 3 :η 1 -FluSiMe 2 N t Bu)AlR (g). Then the 0.5 equivalent units of g continues to react with the other 0.5 equivalent units of c to finally give an insoluble Y-Aluminum salt solid [(R) 2 Al(µ-R) 2 Ln(µ-R) 2 Al(R) 2 ] + [Al(η 3 :η 1 -FluSiMe 2 N t Bu) 2 ] − (k) as a byproduct. Then two alkyl groups of 0.5 equivalent units of the catalyst precursor h are removed from metal center by 1 equivalent unit of activator to finally generate a divalent cationic species {Ln[(µ-R) 2 AlR 2 ] 2+ } (l) as catalytically active species in the regio-/stereoselective polymerization of olefins. Based on the actual circumstance of syndiotactic polymerization of ST by the Y complex 3/B/Al i Bu 3 ternary system and the Sc(CH 2 SiMe 3 ) 3 (THF) 2 /B/Al i Bu 3 ternary system with different molar ratios as 1:1:20 and 1:1:10 in Table 4, such a divalent cationic active species is quite reasonable. The less steric hindrance around the metal center of this cationic species permits the coordination and insertion of IP/MY monomers in cis-1,4 mode to form the anti-allyl form intermediate, which finally affords the cis-1,4-PIPs/PMYs with cis-1,4-selectivities up to 96% or 100%. Such high cis-1,4-selectivity in the coordination-insertion polymerization of conjugated dienes is in agreement with the high cis-1,4-selectivity obtained by the heterobimetallic tetraalkylaluminate active species {LLn[(µ-Me) 2 AlMe 2 ] 2+ } in the isoprene polymerization [40,41]. Similarly, such a cationic species also promotes the backbiting of the last phenyl group in PST chains with a metal center. As a result, the SPSTs are obtained by such Flu-based CGC allyl-type rare earth monoalkyl complexes 1-3/activator/AlR 3 ternary systems. polymer)/(mol Ln h) to 1.1 × 10 6 (g of polymer)/(mol Ln h) and high cis-1,4-selectivities up to 96% in IP polymerization in PhCl 2 , yielding the main cis-1,4-PIPs with high molecular weights (M n up to 1000 kg/mol) and bimodal molecular weight distributions (M w /M n = 1.78-3.71). Activated by the cocatalyst borane B(C 6 F 5 ) 3 (C) and Al i Bu 3 , these complexes 1-3 display high activities up to 1.2 × 10 5 (g of polymer)/(mol Ln h) and high cis-1,4-selectivities up to 100% in MY polymerization in Tol, producing the cis-1,4-PMYs with high molecular weights (M n up to 900 kg/mol) and bimodal molecular weight distributions (M w /M n = 1.66-3.47). Moreover, these complexes 1-3/[PhNHMe 2 ][B(C 6 F 5 ) 4 ] (B)/Al i Bu 3 ternary systems also promote the syndiotactic polymerization of ST with moderate activities up to 2.8 × 10 3 (g of polymer)/(mol Ln h) in PhCl 2 to give SPSTs with high molecular weights (M n up to 1300 kg/mol) and bimodal molecular weight distributions (M w /M n = 1.49-18.78). In comparison with no, or very low activity and regio-/stereoselectivity of the previous CGC allyl type rare earth metal complexes, such results demonstrate that the transfer of the chelating ligand from the rare earth metal center to the Al center to form the heterobimetallic tetraalkylaluminate complex plays a key role on the excellent catalytic performance of these CGC allyl-type rare earth metal monoalkyl complexes in olefin polymerization. These findings will benefit the design of highly efficient and regio-/stereoselective rare earth metal catalysts as well as the precise synthesis of natural rubber. Further studies will be focused on the research of excellent rare earth metal catalysts for the polymerization of olefins. Author Contributions: X.L. conceived and designed the experiments; G.G. carried out the experiment and analyzed the data; X.W., X.Y., L.Y., S.Z. and N.Q. contributed reagents/materials/analysis tools; G.G. and X.L. wrote the paper.