Synthesis, APPI Mass-Spectrometric Characterization, and Polymerization Studies of Group 4 Dinuclear Bis(ansa-metallocene) Complexes

: New ligand platforms of the type p - or m -Ph{-CR(3,6- t Bu 2 Flu)(Cp)} 2 ( para -, R = Me ( 2a ), H ( 2b ); meta -, R = Me ( 2c )) were synthesized via nucleophilic addition of the 3,6- t Bu 2 -ﬂuorenyl-anion onto the parent phenylene-bridged difulvenes ( 1a – c ). The corresponding discrete homodinuclear zirconium and hafnium bis(dichloro ansa -metallocene) complexes, Ph[{-CR(3,6- t Bu 2 Flu)(Cp)}MCl 2 ] 2 ( p -, R = Me ( 3a-Zr 2 , 3a-Hf 2 ), R = H ( 3b-Zr 2 ); m -, R = Me ( 3c-Zr 2 ), were prepared by salt metathesis reactions. An attempt to generate in situ a heterodinuclear complex 3a-Zr - Hf was also undertaken. For the ﬁrst time, Atmospheric Pressure PhotoIonization (APPI) mass-spectrometric data were obtained for all dinuclear compounds and found to be in excellent agreement with the simulated ones. Preliminary studies on the catalytic performances of these dinuclear complexes, upon activation with MAO, in ethylene homopolymerization and ethylene/1-hexene copolymerization revealed a few differences as compared to those of the monometallic analogues. In particular, slightly lower molecular weights and a greater formation of short methyl and ethyl branches were obtained with the dinuclear systems.


Synthesis of Group 4 Dinuclear Bis(dichloro ansa-metallocene) Complexes.
In order to prepare the corresponding group 4 bis(dichloro ansa-metallocene) complexes, standard salt-metathesis reactions between the ligand tetraanions, generated in situ in Et2O, and MCl4 salts (2 equiv.), were used (Scheme 4). Thus, the homodinuclear bis(dichloro ansa-zirconocenes) 3a-c-Zr2 and bis(dichloro ansa-hafnocene) 3a-Hf2 were isolated in good yields as red and yellow solids, respectively. As 3a-c-Zr2 and 3a-Hf2 were derived from diastereomeric mixtures of proligands, two diastereomers for each of these compounds were anticipated, featuring Cs-/Ci-symmetries for the para-phenylene-bridged complexes 3a,b-M2 and Cs-/C1-symmetries for the meta-phenylene-bridged complex 3c-Zr2 (Scheme 5). Accordingly, the 1 H and 13 C NMR spectra of the crude 3a-Zr2 ( Figures  S24-S26), 3a-Hf2 ( Figures S28 and S29, respectively), and 3c-Zr2 ( Figures S35 and S36, respectively) complexes displayed two sets of resonances corresponding to the two diastereomers. Unexpectedly, only one set of resonances assigned to a single diastereoisomer of either Cs-or Ci-symmetry was observed in the 1 H NMR spectrum of 3b-Zr2 ( Figure S31). As this compound was isolated in a lower yield than the other ones, one cannot discard that only one diastereoisomer was recovered in the workup. As 3a-c-Zr 2 and 3a-Hf 2 were derived from diastereomeric mixtures of proligands, two diastereomers for each of these compounds were anticipated, featuring C s -/C i -symmetries for the para-phenylene-bridged complexes 3a,b-M 2 and C s -/C 1 -symmetries for the meta-phenylene-bridged complex 3c-Zr 2 (Scheme 5). Accordingly, the 1 H and 13 C NMR spectra of the crude 3a-Zr 2 ( Figures S24-S26), 3a-Hf 2 ( Figures S28 and S29, respectively), and 3c-Zr 2 ( Figures S35 and S36, respectively) complexes displayed two sets of resonances corresponding to the two diastereomers. Unexpectedly, only one set of resonances assigned to a single diastereoisomer of either C s -or C i -symmetry was observed in the 1 H NMR spectrum of 3b-Zr 2 ( Figure S31). As this compound was isolated in a lower yield than the other ones, one cannot discard that only one diastereoisomer was recovered in the workup. Unfortunately, all attempts to grow single-crystals of these complexes suitable for X-ray diffraction studies have failed so far. However, the identity of these bis(dichloro ansa-metallocene) compounds was confirmed unambiguously by mass spectrometry (vide infra).
In order to obtain a better clue about the possible structures of the dinuclear bis(metallocenes), the corresponding geometries of the two Cs-and Ci-symmetric isomers of 3a-Zr2 ( Figure S68; see the Experimental Section for details) and the two Cs-and C1-symmetric isomers of 3c-Zr2 ( Figure S69) were modeled by DFT computations. It is noteworthy that the optimized geometries of the isomers belonging to both dinuclear systems 3a-Zr2 and 3c-Zr2 featured relatively long Zr…Zr intermetallic distances of 10.5-10.8 Å and 9.2-9.8 Å, respectively. Also, the respective orientations of the metallocenic fragments in these structures resulted in the coordination sites, represented by the chlorine ligands, pointing in opposite directions. Such an orientation of the metallocenic moieties in both para-and meta-phenylene-bridged systems may not be favorable to the mutual approach of the two metal centers in dinuclear active species derived thereof during polymerization (vide infra). Note, however, that the above observations were made on the most stable neutral isomers as determined by DFT, and they do not necessarily reflect the proximity that can be reached from dynamic conformations in those species. Also, the behavior of the active cationic species associated with counterionic moieties may be quite different.
In an attempt to synthesize a hetero-bis(metallocene) incorporating both zirconium and hafnium metals, a similar salt-metathesis protocol as that utilized for the synthesis of the homo-bis(dichloro ansa-metallocenes) 3a-Zr2 and 3a-Hf2 was probed using 1 equiv. of each of the metal precursors ZrCl4 and HfCl4 (Scheme 6). In this case, as anticipated, a statistical 1:1:2 mixture of the homodinuclear 3a-Zr2 and 3a-Hf2 complexes and the heterodinuclear 3a-Zr/Hf complex was obtained, as revealed by 1 H NMR spectroscopy of the crude sample. No single-crystal suitable for X-ray diffraction studies has been grown thus far. Due to the complexity of the mixture and the obvious difficulties associated with regular elemental and spectroscopic analyses, only its mass-spectrometric characterization was performed (vide infra). Unfortunately, all attempts to grow single-crystals of these complexes suitable for X-ray diffraction studies have failed so far. However, the identity of these bis(dichloro ansa-metallocene) compounds was confirmed unambiguously by mass spectrometry (vide infra).
In order to obtain a better clue about the possible structures of the dinuclear bis(metallocenes), the corresponding geometries of the two C s -and C i -symmetric isomers of 3a-Zr 2 ( Figure S68; see the Experimental Section for details) and the two C s -and C 1 -symmetric isomers of 3c-Zr 2 ( Figure S69) were modeled by DFT computations. It is noteworthy that the optimized geometries of the isomers belonging to both dinuclear systems 3a-Zr 2 and 3c-Zr 2 featured relatively long Zr . . . Zr intermetallic distances of 10.5-10.8 Å and 9.2-9.8 Å, respectively. Also, the respective orientations of the metallocenic fragments in these structures resulted in the coordination sites, represented by the chlorine ligands, pointing in opposite directions. Such an orientation of the metallocenic moieties in both paraand meta-phenylene-bridged systems may not be favorable to the mutual approach of the two metal centers in dinuclear active species derived thereof during polymerization (vide infra). Note, however, that the above observations were made on the most stable neutral isomers as determined by DFT, and they do not necessarily reflect the proximity that can be reached from dynamic conformations in those species. Also, the behavior of the active cationic species associated with counterionic moieties may be quite different.
In an attempt to synthesize a hetero-bis(metallocene) incorporating both zirconium and hafnium metals, a similar salt-metathesis protocol as that utilized for the synthesis of the homo-bis(dichloro ansa-metallocenes) 3a-Zr 2 and 3a-Hf 2 was probed using 1 equiv. of each of the metal precursors ZrCl 4 and HfCl 4 (Scheme 6). In this case, as anticipated, a statistical 1:1:2 mixture of the homodinuclear 3a-Zr 2 and 3a-Hf 2 complexes and the heterodinuclear 3a-Zr/Hf complex was obtained, as revealed by 1 H NMR spectroscopy of the crude sample. No single-crystal suitable for X-ray diffraction studies has been grown thus far. Due to the complexity of the mixture and the obvious difficulties associated with regular elemental and spectroscopic analyses, only its mass-spectrometric characterization was performed (vide infra). Unfortunately, all attempts to grow single-crystals of these complexes suitable for X-ray diffraction studies have failed so far. However, the identity of these bis(dichloro ansa-metallocene) compounds was confirmed unambiguously by mass spectrometry (vide infra).
In order to obtain a better clue about the possible structures of the dinuclear bis(metallocenes), the corresponding geometries of the two Cs-and Ci-symmetric isomers of 3a-Zr2 ( Figure S68; see the Experimental Section for details) and the two Cs-and C1-symmetric isomers of 3c-Zr2 ( Figure S69) were modeled by DFT computations. It is noteworthy that the optimized geometries of the isomers belonging to both dinuclear systems 3a-Zr2 and 3c-Zr2 featured relatively long Zr…Zr intermetallic distances of 10.5-10.8 Å and 9.2-9.8 Å, respectively. Also, the respective orientations of the metallocenic fragments in these structures resulted in the coordination sites, represented by the chlorine ligands, pointing in opposite directions. Such an orientation of the metallocenic moieties in both para-and meta-phenylene-bridged systems may not be favorable to the mutual approach of the two metal centers in dinuclear active species derived thereof during polymerization (vide infra). Note, however, that the above observations were made on the most stable neutral isomers as determined by DFT, and they do not necessarily reflect the proximity that can be reached from dynamic conformations in those species. Also, the behavior of the active cationic species associated with counterionic moieties may be quite different.
In an attempt to synthesize a hetero-bis(metallocene) incorporating both zirconium and hafnium metals, a similar salt-metathesis protocol as that utilized for the synthesis of the homo-bis(dichloro ansa-metallocenes) 3a-Zr2 and 3a-Hf2 was probed using 1 equiv. of each of the metal precursors ZrCl4 and HfCl4 (Scheme 6). In this case, as anticipated, a statistical 1:1:2 mixture of the homodinuclear 3a-Zr2 and 3a-Hf2 complexes and the heterodinuclear 3a-Zr/Hf complex was obtained, as revealed by 1 H NMR spectroscopy of the crude sample. No single-crystal suitable for X-ray diffraction studies has been grown thus far. Due to the complexity of the mixture and the obvious difficulties associated with regular elemental and spectroscopic analyses, only its mass-spectrometric characterization was performed (vide infra). Scheme 6. Attempted synthesis of the heterodinuclear bis(metallocene) 3a-Zr/Hf. Scheme 6. Attempted synthesis of the heterodinuclear bis(metallocene) 3a-Zr/Hf.

Synthesis of Mononuclear Ansa-metallocene Analogues
For comparative studies of mass-spectrometric analyses and of catalytic properties of the bis(metallocene)s in α-olefin polymerization, their mononuclear analogues were also synthesized (Scheme 7). The complexes 3a',b'-Zr and 3b'-Hf were isolated in good yields and characterized by 1 H and 13 C NMR spectroscopic studies, X-ray diffraction (for 3a'-Zr; Figure S67), and APPI mass-spectrometry (vide infra).

Synthesis of Mononuclear Ansa-metallocene Analogues
For comparative studies of mass-spectrometric analyses and of catalytic properties of the bis(metallocene)s in α-olefin polymerization, their mononuclear analogues were also synthesized (Scheme 7). The complexes 3a',b'-Zr and 3b'-Hf were isolated in good yields and characterized by 1 H and 13 C NMR spectroscopic studies, X-ray diffraction (for 3a'-Zr; Figure S67), and APPI massspectrometry (vide infra). Scheme 7. Synthesis of the mononuclear metallocene analogues.

Mass Spectrometric Studies of Mononuclear and Dinuclear Bis(dichloro ansa-metallocene) Complexes
Atmospheric Pressure PhotoIonization (APPI) was chosen instead of the more common electrospray (ESI), as APPI is very efficient for the ionization of aromatic molecules that do not contain polar groups; also, it allows for the use of dry toluene as a solvent to preserve the rather sensitive metallocene complexes [39,40]. The APPI mass spectra of the mononuclear complexes 3a'-Zr, 3b'-Zr, and 3b'-Hf are summarized in Figure 1. In each case, the corresponding intact species was detected as a M +• molecular ion. A free ligand was also observed in the spectra at m/z 444.3 and m/z 430.3, as well as the C21H26 moiety at m/z 278.

Mass Spectrometric Studies of Mononuclear and Dinuclear Bis(dichloro ansa-metallocene) Complexes
Atmospheric Pressure PhotoIonization (APPI) was chosen instead of the more common electrospray (ESI), as APPI is very efficient for the ionization of aromatic molecules that do not contain polar groups; also, it allows for the use of dry toluene as a solvent to preserve the rather sensitive metallocene complexes [39,40]. The APPI mass spectra of the mononuclear complexes 3a'-Zr, 3b'-Zr, and 3b'-Hf are summarized in Figure 1. In each case, the corresponding intact species was detected as a M +• molecular ion. A free ligand was also observed in the spectra at m/z 444.3 and m/z 430.3, as well as the C 21 H 26 moiety at m/z 278.

Synthesis of Mononuclear Ansa-metallocene Analogues
For comparative studies of mass-spectrometric analyses and of catalytic properties of the bis(metallocene)s in α-olefin polymerization, their mononuclear analogues were also synthesized (Scheme 7). The complexes 3a',b'-Zr and 3b'-Hf were isolated in good yields and characterized by 1 H and 13 C NMR spectroscopic studies, X-ray diffraction (for 3a'-Zr; Figure S67), and APPI massspectrometry (vide infra). Scheme 7. Synthesis of the mononuclear metallocene analogues.

Mass Spectrometric Studies of Mononuclear and Dinuclear Bis(dichloro ansa-metallocene) Complexes
Atmospheric Pressure PhotoIonization (APPI) was chosen instead of the more common electrospray (ESI), as APPI is very efficient for the ionization of aromatic molecules that do not contain polar groups; also, it allows for the use of dry toluene as a solvent to preserve the rather sensitive metallocene complexes [39,40]. The APPI mass spectra of the mononuclear complexes 3a'-Zr, 3b'-Zr, and 3b'-Hf are summarized in Figure 1. In each case, the corresponding intact species was detected as a M +• molecular ion. A free ligand was also observed in the spectra at m/z 444.3 and m/z 430.3, as well as the C21H26 moiety at m/z 278. The zoomed areas showcase the theoretical and experimental isotopic clusters. For each isotopic distribution, only the most intense isotope peak is labelled.
By analogy, the M +• molecular ions were identified for the dinuclear bis(zirconocene) 3a-b-Zr 2 compound and the dinuclear bis(hafnocene) 3a-Hf 2 compound ( Figure 2). The accurate masses and isotopic distributions (m/z 1130.2074, 1102.1754, and 1310.2865, respectively) are in very good agreement (<5 ppm) with those expected theoretically based on the corresponding ions' molecular formula (m/z 1130.2013, 1102.1700, and 1310.2850, respectively). In addition to the dinuclear species 3a,b-Zr 2 and 3a-Hf 2 , molecular ions derived from the monometallated fragments, i.e., 3a,b-Zr and 3a-Hf, were detected in each case at m/z 972.3776, 944.3398, and 1062.4139, respectively. These ions were most likely not generated by gas-phase fragmentation, as they were not produced by collision-induced dissociation of the dinuclear molecular ions. They, however, may have been produced by partial degradation (e.g., hydrolysis) during the sample handling or in the atmosphere source that may contain traces of water. . In addition to the dinuclear species 3a,b-Zr2 and 3a-Hf2, molecular ions derived from the monometallated fragments, i.e., 3a,b-Zr and 3a-Hf, were detected in each case at m/z 972.3776, 944.3398, and 1062.4139, respectively. These ions were most likely not generated by gas-phase fragmentation, as they were not produced by collisioninduced dissociation of the dinuclear molecular ions. They, however, may have been produced by partial degradation (e.g., hydrolysis) during the sample handling or in the atmosphere source that may contain traces of water. For the mixture containing the heterodinuclear bis(dichloro ansa-metallocene) compound 3a-Zr/Hf and its homodinuclear counterparts 3a-Zr2 and 3a-Hf2, a set of five isotopic distributions was observed in the APPI mass-spectrum ( Figure 3). Besides the distributions corresponding to the homodinuclear 3a-Zr2 and 3a-Hf2 and their monometallated versions, i.e., 3a-Zr and 3a-Hf, respectively, an isotopic distribution at m/z 1222.2463 was identified and unequivocally assigned to 3a-Zr/Hf. APPI should present a low ionization discrimination for these species, so their relative abundance should be representative of their actual amount in the sample, although the most airsensitive molecules may present a lower abundance because of higher degradation. This possibly accounts for the observed lower intensity of peaks arising from 3a-Hf2. For the mixture containing the heterodinuclear bis(dichloro ansa-metallocene) compound 3a-Zr/Hf and its homodinuclear counterparts 3a-Zr 2 and 3a-Hf 2 , a set of five isotopic distributions was observed in the APPI mass-spectrum ( Figure 3). Besides the distributions corresponding to the homodinuclear 3a-Zr 2 and 3a-Hf 2 and their monometallated versions, i.e., 3a-Zr and 3a-Hf, respectively, an isotopic distribution at m/z 1222.2463 was identified and unequivocally assigned to 3a-Zr/Hf. APPI should present a low ionization discrimination for these species, so their relative abundance should be representative of their actual amount in the sample, although the most air-sensitive molecules may present a lower abundance because of higher degradation. This possibly accounts for the observed lower intensity of peaks arising from 3a-Hf 2 . The zoomed areas showcase the theoretical and experimental isotopic clusters of the dinuclear bis(dichloro ansa-metallocene) complexes 3a-Zr-Hf. For each isotopic distribution, only the most intense isotope peak is labelled.

Polymerization Catalysis
The dinuclear bis(dichloro ansa-metallocene) complexes 3a-M 2 , 3b-Zr 2 , and 3c-Zr 2 , and their mononuclear analogues 3a,b'-Zr and 3b'-Hf, in combination with methylalumoxane (MAO), were evaluated in homogeneous ethylene polymerization (Table 1) and ethylene/1-hexene copolymerization ( Table 2). Each polymerization experiment was repeated independently twice under the same conditions (toluene, 5.5 bar of ethylene, 60 • C), which revealed good reproducibility in terms of productivity (polymer yield) and physicochemical properties (T m , M w , polydispersity index (PDI)) of the isolated polymer.    In general, no significant or limited difference was observed for the experiments involving dinuclear bis(metallocene) with respect to those using the corresponding mononuclear analogues (Table 1; compare entries 1-2/3-4, 5-6/7-8, and 10-11/12, respectively). Indeed, all of the systems that were produced with similar productivities used polyethylene samples exhibiting quite similar molecular weight distributions and T m values. A slight drop in productivity was observed for the meta-bridged system 3c-Zr 2 (entry 9), also yielding a lower molecular weight polyethylene (PE) as compared to the para-bridged congeners 3a,b-Zr 2 . Interestingly, as established by 13 C NMR spectroscopy, 3b-Zr 2 induced the formation of short branches (methyl and, to a lesser extent, ethyl) to a significantly greater extent than its mononuclear analogue and any other dinuclear system; the same observation was made in ethylene/1-hexene copolymerization (vide infra). This may presumably arise from a chain-walking mechanism.
The Hf-based systems 3a-Hf 2 and 3b'-Hf appeared to be ca. 3-fold less productive (entries 10-12) than their Zr-based counterparts while affording much higher molecular weight PEs. The addition of BHT (entry 11) to scavenge the excess "free" AlMe 3 [41] present in MAO did not affect the productivity of this system, and also resulted in an insoluble polymer (likely due to the formation of high molecular weight polyethylene (HMWPE) because of the absence of transfer to AlMe 3 ).
Also, no strikingly different results were obtained upon using the different mono-and dinuclear compounds in ethylene/1-hexene copolymerization ( Table 2). The Zr-based systems, both di-and mononuclear, afforded a very narrow range of productivities. It is of note that, in the para-bridged series 3a,b-Zr 2 and 3a'b'-Zr, the dinuclear bis(metallocenes) gave slightly but significantly lower molecular weight copolymers and narrower dispersities (compare entries 1-2/3-4 and 5-6/7-8, respectively) than their respective mononuclear analogues. Conversely, the meta-bridged system 3c-Zr 2 afforded a higher molecular weight copolymer than its mononuclear analogue (compare entries 3/4 and 9). Again, higher molecular weight copolymers were obtained with Hf-based catalysts, but with lower productivities than their Zr counterparts (entries 10-12). Also, a somewhat higher incorporation of 1-hexene in copolymers was achieved with the Hf-based catalysts.

Instruments and Measurements
The NMR spectra of air-and moisture-sensitive compounds were recorded on Bruker AM-400 and AM-500 spectrometers in Teflon-valved NMR tubes at room temperature. 1 H and 13 C chemical shifts are reported in ppm versus SiMe4 and were determined using residual solvent signal. Coupling constants are given in Hertz. Assignments of signals were carried out using 1D ( 1 H, 13 C{ 1 H}, JMOD) and 2D (COSY, HMBC, HMQC) NMR experiments. Elemental analyses were performed on a Carlo Erba 1108 Elemental Analyzer instrument at the London Metropolitan University by Stephen Boyer or on a Flash EA1112 CHNS Thermo Electron apparatus at CRMPO, Rennes, and were the average of a minimum of two independent measurements. The 13 C{ 1 H} NMR and GPC analyses of polymer samples were performed in the research center of Total Research and Technologies in Feluy (Feluy, Belgium). The 13 C{ 1 H} NMR analyses were run on a 500 MHz Bruker Avance III with a 10 mm cryoprobe HTDUL in trichlorobenzene/C6D6 (2 mL/0.5 mL). The GPC analyses were performed in 1,2,4-trichlorobenzene at 135 °C using PS standards for calibration. Differential scanning calorimetry (DSC) analyses were performed on a Setaram DSC 131 apparatus under a continuous flow of helium and using aluminum capsules. Glass transition and melting temperatures were measured during the second heating (10 °C·min −1 ).
The ESI (ElectroSpray Ionization) mass spectra of organic compounds, including proligands, were recorded at CRMPO-Scanmat (Rennes, France) on an Orbitrap Thermo Fisher Scientific Q-Exactive instrument with an ESI source in positive mode by direct introduction at 5-10 µg·mL −1 . Samples were prepared in CH2Cl2 at 10 µg·mL −1 .
The ASAP (Atmospheric Solids Analysis Probe) mass spectra of proligands were recorded at the CRMPO-Scanmat (Rennes, France) on a Q-TOF Bruker Maxis 4G instrument with an APCI source in positive mode at desorption temperatures of 255 and 300 °C.

APPI Mass Spectrometric Characterization of Metal Complexes
The mass spectra of metal complexes were recorded on a hybrid quadrupole time-of-flight instrument (Waters, Synapt G2, Manchester, England) equipped with an APPI source. The instrument was operated in the positive ion mode. The ionization experimental conditions were set as follows: desolvation gas flow, 700 L h −1 ; source temperature, 120 °C; probe temperature, 400 °C; sampling cone, 20 V; extraction cone, 3 V. The time-of-flight was operated in the 'resolution mode' yielding a resolving power of about 20,000. The samples were prepared in a glove box using dried toluene in 1.5 mL glass vials with a final concentration of 20 µM. For analysis, the sample was taken with a dry syringe stored in an oven. The solution was directly infused into the source using a syringe pump at a flow rate of 200 µL h −1 . Data were acquired over the m/z 50-2000 range for 2-5 min. Note that the given accurate masses are given with Water Mass Lynx 4.1 that do not take into account the mass of the electron removed during ionization. All given masses are monoisotopic values. In the mass spectra, only the highest abundant isotope is labelled.

Instruments and Measurements
The NMR spectra of air-and moisture-sensitive compounds were recorded on Bruk and AM-500 spectrometers in Teflon-valved NMR tubes at room temperature. 1 H and 1 shifts are reported in ppm versus SiMe4 and were determined using residual solvent signa constants are given in Hertz. Assignments of signals were carried out using 1D ( 1 H, 13 C{ and 2D (COSY, HMBC, HMQC) NMR experiments. Elemental analyses were performed Erba 1108 Elemental Analyzer instrument at the London Metropolitan University by Ste or on a Flash EA1112 CHNS Thermo Electron apparatus at CRMPO, Rennes, and were of a minimum of two independent measurements.
The 13 C{ 1 H} NMR and GPC analyses of polymer samples were performed in the rese of Total Research and Technologies in Feluy (Feluy, Belgium). The 13 C{ 1 H} NMR analys on a 500 MHz Bruker Avance III with a 10 mm cryoprobe HTDUL in trichlorobenzene/C6D mL). The GPC analyses were performed in 1,2,4-trichlorobenzene at 135 °C using PS st calibration. Differential scanning calorimetry (DSC) analyses were performed on a Setara apparatus under a continuous flow of helium and using aluminum capsules. Glass tra melting temperatures were measured during the second heating (10 °C·min −1 ). The ESI (ElectroSpray Ionization) mass spectra of organic compounds, including were recorded at CRMPO-Scanmat (Rennes, France) on an Orbitrap Thermo Fisher S Exactive instrument with an ESI source in positive mode by direct introduction at 5-Samples were prepared in CH2Cl2 at 10 µg·mL −1 . The ASAP (Atmospheric Solids Analysis Probe) mass spectra of proligands were reco CRMPO-Scanmat (Rennes, France) on a Q-TOF Bruker Maxis 4G instrument with an APC positive mode at desorption temperatures of 255 and 300 °C.

APPI Mass Spectrometric Characterization of Metal Complexes
The mass spectra of metal complexes were recorded on a hybrid quadrupole tim instrument (Waters, Synapt G2, Manchester, England) equipped with an APPI s instrument was operated in the positive ion mode. The ionization experimental conditio as follows: desolvation gas flow, 700 L h −1 ; source temperature, 120 °C; probe temperat sampling cone, 20 V; extraction cone, 3 V. The time-of-flight was operated in the 'resolu yielding a resolving power of about 20,000. The samples were prepared in a glove box toluene in 1.5 mL glass vials with a final concentration of 20 µM. For analysis, the sampl with a dry syringe stored in an oven. The solution was directly infused into the source usin pump at a flow rate of 200 µL h −1 . Data were acquired over the m/z 50-2000 range for 2that the given accurate masses are given with Water Mass Lynx 4.1 that do not take into mass of the electron removed during ionization. All given masses are monoisotopic va mass spectra, only the highest abundant isotope is labelled.

Instruments and Measurements
The NMR spectra of air-and moisture-sensitive compounds were recorded on Bruker AM-400 and AM-500 spectrometers in Teflon-valved NMR tubes at room temperature. 1 H and 13 C chemical shifts are reported in ppm versus SiMe 4 and were determined using residual solvent signal. Coupling constants are given in Hertz. Assignments of signals were carried out using 1D ( 1 H, 13 C{ 1 H}, JMOD) and 2D (COSY, HMBC, HMQC) NMR experiments. Elemental analyses were performed on a Carlo Erba 1108 Elemental Analyzer instrument at the London Metropolitan University by Stephen Boyer or on a Flash EA1112 CHNS Thermo Electron apparatus at CRMPO, Rennes, and were the average of a minimum of two independent measurements. The 13 C{ 1 H} NMR and GPC analyses of polymer samples were performed in the research center of Total Research and Technologies in Feluy (Feluy, Belgium). The 13 C{ 1 H} NMR analyses were run on a 500 MHz Bruker Avance III with a 10 mm cryoprobe HTDUL in trichlorobenzene/C 6 D 6 (2 mL/0.5 mL). The GPC analyses were performed in 1,2,4-trichlorobenzene at 135 • C using PS standards for calibration. Differential scanning calorimetry (DSC) analyses were performed on a Setaram DSC 131 apparatus under a continuous flow of helium and using aluminum capsules. Glass transition and melting temperatures were measured during the second heating (10 • C·min −1 ).
The ESI (ElectroSpray Ionization) mass spectra of organic compounds, including proligands, were recorded at CRMPO-Scanmat (Rennes, France) on an Orbitrap Thermo Fisher Scientific Q-Exactive instrument with an ESI source in positive mode by direct introduction at 5-10 µg·mL −1 . Samples were prepared in CH 2 Cl 2 at 10 µg·mL −1 .
The ASAP (Atmospheric Solids Analysis Probe) mass spectra of proligands were recorded at the CRMPO-Scanmat (Rennes, France) on a Q-TOF Bruker Maxis 4G instrument with an APCI source in positive mode at desorption temperatures of 255 and 300 • C.

APPI Mass Spectrometric Characterization of Metal Complexes
The mass spectra of metal complexes were recorded on a hybrid quadrupole time-of-flight instrument (Waters, Synapt G2, Manchester, England) equipped with an APPI source. The instrument was operated in the positive ion mode. The ionization experimental conditions were set as follows: desolvation gas flow, 700 L h −1 ; source temperature, 120 • C; probe temperature, 400 • C; sampling cone, 20 V; extraction cone, 3 V. The time-of-flight was operated in the 'resolution mode' yielding a resolving power of about 20,000. The samples were prepared in a glove box using dried toluene in 1.5 mL glass vials with a final concentration of 20 µM. For analysis, the sample was taken with a dry syringe stored in an oven. The solution was directly infused into the source using a syringe pump at a flow rate of 200 µL h −1 . Data were acquired over the m/z 50-2000 range for 2-5 min. Note that the given accurate masses are given with Water Mass Lynx 4.1 that do not take into account the mass of the electron removed during ionization. All given masses are monoisotopic values. In the mass spectra, only the highest abundant isotope is labelled. were dissolved in methanol (200 mL). To this solution, pyrrolidine (7.5 mL, 89.0 mmol) was added at 0 • C. The reaction mixture was stirred at room temperature for 7 days. After neutralization with glacial acetic acid (7.5 mL) and separation of the organic phase, volatiles were evaporated under vacuum to give a yellow powder (5.51 g, 21.3 mmol, 72%). 1 H NMR (CDCl 3 , 400 MHz, 25 • C): δ 7.35 (s, 4H, CH-Ar), 6.59 (dt, 3 J = 5.2, 4 J = 1.6, 2H, CH-Cp), 6.51 (dt, 3 J = 5.2, 4 J = 1.6, 2H, CH-Cp), 6.43 (dt, 3 J = 5.2, 4 J = 1.6, 2H, CH-Cp), 6.16 (dt, 3 J = 5.2, 4 J = 1.6, 2H, CH-Cp), 2.50 (s, 6H, CH 3 ). 13  In a Schlenk flask, to a solution of 3,6-di-tert-butyl-fluorene (2.17 g, 7.8 mmol) in THF (100 mL), was added n-BuLi (3.13 mL of a 2.5 M solution in hexane, 7.8 mmol). This solution was added dropwise to a solution of 1a (1.00 g, 3.9 mmol) in THF (100 mL) at −10 • C over 10 min. After completion of the addition, the reaction mixture was stirred for 2 days at room temperature. The mixture was hydrolyzed with 10% aqueous hydrochloric acid (20 mL), the organic phase was dried over sodium sulfate, and volatiles were evaporated in vacuo. The solid residues was washed with pentane (200 mL) and dried under reduced in vacuo to afford a white powder (1.  13  3.9. 1,3-Ph(MeC-(3,6-tBu 2 FluH)(CpH)) 2 (2c) Using a protocol similar to that described above for 2a, compound 2c was prepared from 3,6-di-tert-butyl-fluorene (2.17 g, 7.8 mmol), n-BuLi  13

Instruments and Measurements
The NMR spectra of air-and moisture-sensitive compounds were recorded on Bruker AM-400 and AM-500 spectrometers in Teflon-valved NMR tubes at room temperature. 1 H and 13 C chemical shifts are reported in ppm versus SiMe4 and were determined using residual solvent signal. Coupling constants are given in Hertz. Assignments of signals were carried out using 1D ( 1 H, 13 C{ 1 H}, JMOD) and 2D (COSY, HMBC, HMQC) NMR experiments. Elemental analyses were performed on a Carlo Erba 1108 Elemental Analyzer instrument at the London Metropolitan University by Stephen Boyer or on a Flash EA1112 CHNS Thermo Electron apparatus at CRMPO, Rennes, and were the average of a minimum of two independent measurements. The 13 C{ 1 H} NMR and GPC analyses of polymer samples were performed in the research center of Total Research and Technologies in Feluy (Feluy, Belgium). The 13 C{ 1 H} NMR analyses were run on a 500 MHz Bruker Avance III with a 10 mm cryoprobe HTDUL in trichlorobenzene/C6D6 (2 mL/0.5 mL). The GPC analyses were performed in 1,2,4-trichlorobenzene at 135 °C using PS standards for ). A combination of ωand ϕ-scans was carried out to obtain at least a unique data set. The crystal structures were solved by direct methods, and the remaining atoms were located from a difference Fourier synthesis followed by full-matrix least-squares refinement based on F2 (programs SIR97 [43] and SHELXL-97 [44]). Hydrogen atoms were placed at calculated positions and forced to ride on the attached atom. All non-hydrogen atoms were refined with anisotropic displacement parameters. The locations of the largest peaks in the final difference Fourier map calculation as well as the magnitude of the residual electron densities were of no chemical significance. The main crystal and refinement data are summarized in Table S1. Crystal data, details of data collection, and structure refinement for compound 3a'-Zr (CCDC 1863222) can be obtained from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Computational Details
All calculations were performed with the TURBOMOLE program package using density functional theory (DFT) [45][46][47][48]. The gradient-corrected density functional BP86 in combination with the resolution identity approximation (RI) [49,50] was applied for the geometry optimizations of a stationary point. The triple-ζ zeta valence quality basis set def-TZVP was used for all atoms [51].
The stationary points were characterized as energy minima (no negative Hessian eigenvalues) by vibrational frequency calculations at the same level of theory.

Conclusions
The synthesis of original bis(Cp/Flu) ligand systems linked at the C1-bridge through a phenylene group was developed starting from difulvene precursors. These ligand platforms were utilized for the preparation of homodinuclear zirconium and hafnium bis(dichloro ansa-metallocene) complexes via a regular salt-metathesis metallation protocol. The synthesis of a heterodinuclear zirconium/hafnium bis(dichloro ansa-metallocene) was also performed, although the desired product was generated as a statistical mixture with the corresponding homodinuclear complexes. For the first time, an advanced APPI mass-spectrometric method was applied to the characterization of dinuclear bis(ansa-metallocene) complexes and their mononuclear ansa-metallocene analogues, and relevant data were obtained.
Ethylene homopolymerization as well as ethylene/1-hexene copolymerization were conducted using the homodinuclear dichloro catalyst precursors, as well as with their mononuclear analogues, in combination with MAO. Limited cooperativity evidence has been observed with dinuclear systems so far, with, in some cases, slightly different molecular weights or a greater formation of short methyl and ethyl branches as compared to the mononuclear reference systems. The apparent lack of significant cooperative behavior observed for the dinuclear systems was substantiated by a computational analysis. Thus, the computed paraand meta-phenylene-bridged neutral dinuclear structures suggest that the two metallocenic fragments may orientate their coordination spheres in opposite directions, hence resulting in distant (>9 Å) isolated metal centers. Further investigations in our laboratories are focused on the elaboration of other polynuclear precatalysts with improved performances and identification of the nature of possible intermetallic cooperative effects.