Raising the N-aryl Fluoride Content in Unsymmetrical Diaryliminoacenaphthylenes as a Route to Highly Active Nickel ( II ) Catalysts in Ethylene Polymerization

Please do not adjust margins a. School of Chemistry and Chemical Engineering, University of Chinese Academy of Sciences, Beijing 100049, China. E-mail: cyguo@ucas.ac.cn. b. Key Laboratory of Engineering Plastics and Beijing National Laboratory for Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China. E-mail: whsun@iccas.ac.cn. c. Department of Chemistry, University of Leicester, University Road, Leicester LE1 7RH, UK. E-mail: gas8@leicester.ac.uk. †Electronic Supplementary Information (ESI) available: NMR spectra for the ligands L1 – L5 and complexes Ni1 – Ni10; crystallographic data in CIF format. CCDC 1504611 (Ni2) 1504612 (Ni4). See DOI: 10.1039/x0xx00000x Received 00th January 20xx, Accepted 00th January 20xx


Introduction
The discovery of α-diiminonickel catalysts for ethylene polymerization by Brookhart and co-workers 1 in the mid-1990's has been instrumental in the renaissance of late transition metal mediated ethylene oligomerization and polymerization. 2Among the many types of α-diimine systems to be explored, the 1,2-diiminoacenaphthylyl-nickel halides (A, Scheme 1) 1 have attracted widespread attention and produce polyethylenes with some unique properties. 3Recently the addition of bulky benzhydryl groups to the 2,6-positions of one of the two N-aryl groups of the ligand frame has resulted in nickel systems that display improved thermostabilities and / or catalytic activities toward ethylene polymerization (B, Scheme 1), 4 while affording polyethylenes with high branching contents and narrow polydispersities.From our viewpoint, these catalytic systems show considerable opportunities for potential commercialization as they address two of the major drawbacks of this type of catalyst relating to their thermostability and catalytic activity. 2With a view to improving these parameters further we have found that the introduction of halide substituents to the N-2,6-substituted phenyl group (C, Scheme 1) 6,7 or the aryl groups belonging to the benzhydryl substituents (D, Scheme 1), 8 has a positive effect on activity.Some justification for these results comes from computational work that points towards the electronegativity of these substituents (F or Cl) and the effect on the net charge of the active catalyst as influential. 5Hence, Scheme 1. Symmetrical and unsymmetrical diiminoacenaphthylene-nickel(II) halide precatalysts (A -E) This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins additional targeted halogenation of the N,N-ligand offers the potential for further improvements in catalytic performance.
In this work, we report a series of novel unsymmetrical 1-[2,6-bis(bis-(4-fluorophenyl) methyl)-4-fluoro]-2aryliminoacenaphthylenes in which the N-2,6benzhydrylphenyl group is fluorinated at all five para-aryl positions, while for the other N'-aryl group, the 2-, 4-and 6positions are systematically appended with alkyl groups.A detailed investigation of the effects of these substitution patterns on the performance of the resultant nickel(II) bromide and chloride complexes (E, Scheme 1) in ethylene polymerization is disclosed; the effects on polymer properties are also highlighted.In addition, full synthetic and characterization details for both the ligands and complexes are reported including a 19 F NMR study of the ligands, precatalysts and active species.
Crystals of Ni2 and Ni4 suitable for single crystal X-ray diffraction studies were grown by the slow diffusion of diethyl ether into their respective dichloromethane solutions.Their molecular structures are shown in Figures 1 and 2; selected bond lengths and angles are listed in Table 1.Both structures are closely related and will be discussed together.Each consists of a single nickel center bound by two bromide ligands and an N,N-chelating diaryliminoacenaphthylene to complete a 4-coordinate geometry that can be best described as distorted tetrahedral. 4,9The N-aryl groups within each N,N-ligand are inequivalent with one pair based on 2
The 1 H NMR spectra of the complexes, recorded in deuterated dichloromethane, exhibit broad paramagnetically shifted peaks in the range δ +34 to -16 that display some common features (Table 2).Assignment of the peaks was made through inspection of their peak integrations, consideration of spin delocalization effects, 10 by comparison of the other spectra within this series and by analysis of spectra recorded for simpler symmetrical diiminoacenaphthylenenickel dihalide species such as [1,2-(ArN)2C2C10H6]NiBr2 (Ar = 2,6-Me2C6H3, 11 2,6-Et2C6H3, 12 2,6-iPr2C6H3, 11,13 2,4,6-Me3C6H2 11b ) (Figures S27-S30).In general, the spectra for the nickel bromide complexes were more amenable to assignment while those for the nickel chlorides showed significantly broader and overlapping peaks (Ni8 being the exception); it unclear as to the origin in this difference.In Ni1 -Ni5, the acenaphthalene protons can be identified in the range δ +26 to +5 as six independent signals that integrate to one proton reflecting the unsymmetrical nature of the ligand backbone.The CH(4-FC6H4)2 protons for the fluorinated N-aryl substituents are particularly broad and are seen as 2H resonances at ca. δ 11.5.On the nonfluorinated N-aryl group, the aryl para-protons in Ni1, Ni2 and Ni3 can be seen upfield at δ -15.5.Support for this assignment comes from the observation that replacing this para-proton for a methyl group in Ni4 and Ni5 results in loss of this upfield resonance and the formation of a new downfield 3H resonance at δ 34 corresponding to the para-methyl group.We have also explored the use of 19 F{ 1 H} NMR spectroscopy to characterize Ni1 -Ni10.As a representative series, the 19 F NMR spectra for Ni2, Ni7 and the corresponding free ligand L2 are shown in the top three images in Figure 3.In general, the halide complexes display five fluoride resonances four of which are closely positioned or slightly overlapping; in comparison the free ligand shows only three.It would seem likely that in the free ligand the two 4-FC6H4 groups on the same 2-substituted CH group are inequivalent due to restricted rotation (and π-π interactions) about the Ar-CH(4-FC6H4)a(4-FC6H4)b bond that is mirrored at the 6-position leading two resonances in total; the independent N-4-F-aryl resonance is seen more upfield at ca. δ -123.By contrast in the complexes, in which a distorted tetrahedral geometry is adopted by the nickel(II) center, all four of the fluorides belonging to the 4-fluorophenylmethyl groups become inequivalent; the separate N-4-F-aryl resonance is again upfield shifted.The exchange of a bromide for a chloride has a only minor effect on the four more downfield fluoride chemical shifts.On inspection of the molecular structures of Ni2 and Ni4, it is not immediately obvious as to the reason for the inequivalency of the 4-fluorophenylmethyl substituents, but could be due to the observed inclination of the nonfluorinated N-aryl group away from the perpendicular.a The 1 H NMR spectra of chloride-containing Ni6, Ni7, Ni9, Ni10 were not as well defined with many broad overlapping peaks which has limited their full assignment.b An-H = protons on the acenaphthylene unit.c The peak for the Ar-CHMe2 protons in Ni3 was not identified as was also the case in With a view to probing the active catalytic species for the polymerization (vide infra), we also used 19 F NMR spectroscopy to monitor the effect of addition of the co-catalyst ethylaluminum sesquichloride (Et3Al2Cl3, EASC) to a sample of Ni2.Typically a toluene solution of Ni2 and EASC were transferred to a NMR tube in the glovebox containing a CD2Cl2 glass insert and the NMR spectrum recorded immediately.The spectra were recorded at Al/Ni2 ratios of 10, 50 and 100.At 10 equivalents full consumption of Ni2 was observed with the formation of a mixture of new products as evidenced by multiple signals in the δ -115 to -118 range.At 50 equivalents some sharpening of the resonances was apparent with at least two species visible (Figure 3, bottom image).At higher ratios of EASC (Al/Ni2 = 100), at levels approaching that used to activate the catalyst in the polymerization studies, the signal to noise ratio was poor but still a mixture of species was evident.We are uncertain as to the identity of the species but it seems reasonable to assume that square planar ethyl complexes of the type [(L2)NiEt(BrAlEtCl2)] and [(L2)NiEt(BrAlEt2Cl)] are involved.

Catalytic evaluation for ethylene polymerization
Co-catalyst screen.In order to determine the most compatible co-catalyst, the polymerization study was first conducted using Ni4 in conjunction with various alkylaluminum reagents including methylaluminoxane (MAO), modified methylaluminoxane (MMAO), diethylaluminum chloride (Et2AlCl) and ethylaluminum sesquichloride (Et3Al2Cl3, EASC).Typically the tests were performed at 30 ºC in toluene under ten atmospheres of ethylene pressure with a run time of 30 minutes.In all cases, high activities were achieved with Ni4/EASC the highest (Table 3).Given the good performance of the latter along with the fact that relative low amounts of EASC were needed (i.e., 500 eq.), subsequent studies focused on the use of EASC as the co-catalyst.Ethylene polymerization with Ni1 -Ni5/EASC.With a view to optimizing the catalytic conditions, namely the Al/Ni molar ratio, reaction temperature and run time, Ni4 was selected as the test precatalyst using EASC as the co-catalyst; the results are compiled in Table 4. Firstly, on increasing the Al/Ni molar ratio from 300 to 600 (entries 1-4, Table 4) the activities of Ni4/EASC at 30 ºC steadily increased to a maximum of 21.95 ×

Please do not adjust margins
Please do not adjust margins 10 6 g mol -1 h -1 .However on further increasing the Al/Ni molar ratio a decrease in activity was observed (entry 5, Table 4).This would suggest that with higher molar ratios of Al/Ni that the rate of chain termination (e.g., chain transfer to aluminum) exceeds the rate of chain propagation forming polyethylenes with lower molecular weights (Figure 4).Secondly, with the Al/Ni ratio fixed at 600, the reaction temperature was increased from 20 o C to 80 o C (entries 4, 6-11, Table 4).On inspection of the data the best activity was observed at 30 o C (entry 4); higher temperatures lead to a marked decrease in activity down to 0.24 × 10 6 g of PE (mol of Ni) 1 h 1 at 80 o C (entry 11).This drop in catalyst performance with increasing temperature can be attributed to partial deactivation of the active species at elevated temperatures. 8imilar trends have been previously reported for pre-catalysts bearing related dibenzhydryl-substituted unsymmetrical 1,2diiminoacenaphthylenes. 4,8,9 With regard to the polyethylene properties, higher molecular weights were achieved at lower reaction temperatures (entries 4, 6-11, Table 3 and Figure 5).Again it is assumed that the higher chain transfer and termination takes place more readily at elevated temperatures.In comparison with previous studies it would appear the introduction of para-fluorides to all five of the phenyl groups of the N-2,6-benzydrylphenyl unit described herein is having a positive effect on the catalytic activity.For example, for precatalysts bearing one 2,6-bis(benzhydryl)-4-fluorophenyl group (C, Scheme 1), 7 the highest activity reported, under comparable condition, was 12.68 × 10 6 g of PE (mol of Ni) 1 h 1 which compares with 21.95 × 10 6 g of PE (mol of Ni) 1 h 1 in this work.Likewise, a similar conclusion is drawn when comparing 2,6-bis(di(4-fluorophenyl)methyl)-4-methylphenylcontaining counterparts (D, Scheme 1) 8 in which the activity is again lower at 12.13 × 10 6 g of PE (mol of Ni) 1 h 1 .While parafluorides can exhibit a positive electron-donating mesomeric effect (via p−π F−Ar bonding), 14 they can also display powerful inductive effects and it is this property that is viewed as responsible for the performance characteristics identified here.Indeed our computational work has highlighted the importance of electron withdrawing groups on the net charge of the active catalyst and resultant activity. 5It is noteworthy that at even 70 o C, Ni4/EASC maintains good activity at 4.22 × 10 6 g of PE (mol of Ni) 1 h 1 .4).
Thirdly, the ethylene polymerization study of Ni4/EASC was conducted over different run times, namely 15, 30, 45 and 60 minutes (entries 4 and 12-14, Table 4).The results reveal the highest activity of 21.95 × 10 6 g of PE (mol of Ni) 1 h 1 is observed over 30 minutes (entry 4, Table 4).It is apparent that a long induction period is required to form the active species following the addition of EASC.Beyond 30 minutes the catalytic activity decreases 15 and after 60 minutes the activity drops to 13.54 × 10 6 g of PE (mol of Ni) 1 h 1 as the active species starts to deactivate.9b The polyethylenes obtained show higher molecular weights with increased reaction times This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins (Figure 6), illustrating that despite partial deactivation there are still some species that remain active. 16ith the optimal conditions for Ni4/EASC established with a Al/Ni molar ratio of 600 and a temperature of 30 °C, all the other nickel bromide complexes were investigated (entries 15-18 in Table 4).All four precatalysts displayed good activities    5).
Ethylene polymerization with Ni6 -Ni10/EASC.The nickel chlorides (Ni6 -Ni10) also showed good catalytic activities towards ethylene polymerization upon treatment with EASC; the data are collected in Table 5.The catalytic performances of these nickel chlorides complexes showed some similar features when compared with their nickel bromide analogues.For example for Ni6/EASC, the catalytic activity at 30 o C increased on raising the Al/Ni molar ratios from 300 but, in this case the highest activity was reached with 500 equivalents of EASC (entry 3 in Table 5); the molecular weights of the corresponding polyethylenes also increased (entries 1-3 in Table 5).On further increasing the Al/Ni ratios beyond 500, both the activity and the polyethylene molecular weight decreased (Figure 7).Unlike for the bromide-containing Ni1 -Ni5, the range in catalytic activities for Ni6 -Ni10 [14.69 -11.63 x 10 6 g PE (mol of Ni) 1 h 1 ] is less pronounced.Nevertheless Ni9, the nickel chloride analogue of Ni4 (the most active system) was at the upper end of the range.It is unclear as to the origin of the variations in catalytic performances between the chlorides and bromides but is likely due to subtle variations occurring during the activation process and to differences in counterion type.Polyethylene microstructures.As a representative sample the polyethylene obtained using Ni4/EASC under optimal conditions was characterized by high-temperature 13 C NMR spectroscopy.Based on assignments listed in the literature, 17 the polyethylene produced possessed 85 branches per 1000 carbons including methyl (56.6%), ethyl (23.3%) and longer chains (20.1%) (Figure 8).By contrast, the polyethylene obtained using Ni6/EASC (Figure 9) contained 116 branches per 1000 carbons, including methyl (52.6%), ethyl (6.4%) and longer chains (41.0%).9a Figure 8. 13 C NMR spectrum of the polyethylene obtained using Ni4/EASC (entry 4 in Table 4) Figure 9. 13 C NMR spectrum of the polyethylene obtained using Ni6/EASC (entry 3 in Table 5) In support of this NMR-based branching analysis, the Tm value for the polyethylene obtained using bromide Ni4/EASC was higher (49.4 o C) than that with chloride Ni6/EASC (43.5 o C) consistent with the lower branching content in the former. 18In Please do not adjust margins Please do not adjust margins addition, the monotonic tensile stress-strain testing data was obtained for a polyethylene sample produced using Ni4/EASC (entry 4 in Table 4).Each mechanical test was performed with five samples in order to obtain statistically reliable results.The ultimate tensile stress and elongation at break of these samples were 2.667 MPa and 256%, respectively. 19

Conclusions
Ten examples of nickel(II) halide (bromide and chloride) complexes bearing unsymmetrical 1,2-diarylimino acenaphthylenes, in which one N-2,6-benzhydrylphenyl group has been para-fluorinated and the other N-aryl group systematically decorated with alkyl groups, have been prepared and characterized; both 1 H and 19 F NMR studies have been informative as to their structure of these paramagnetic species in solution.On activation with EASC at relatively low Al/Ni ratios (ca.600 equiv.),these complexes exhibited high activities up to 2.20 × 10 7 g of PE (mol of Ni) 1 h 1 toward ethylene polymerization at 30 o C. In comparison with previously reported unsymmetrical nickel catalysts, these nickel systems exhibit higher activities toward ethylene polymerization which has been ascribed to the electron withdrawing properties of the para-fluorides and its effect on the net charge of the active catalyst.The bromide precatalysts showed higher activities than their chloride analogues, while the chloride precatalysts required less co-catalyst.Branching analysis using a combination of 13 C NMR spectroscopy and DSC (Tm values below 75 o C) revealed that the polyethylenes possessed high levels of branching.This work further illustrates how fine tuning of the ligand frame can influence catalytic performance and polymer microstructure.

Experimental General procedures
All manipulations of air and/or moisture sensitive compounds were carried out under a nitrogen atmosphere using standard Schlenk techniques.Solvents were distilled under nitrogen from appropriate drying agents prior to use.Methylaluminoxane (MAO) (1.46 M in toluene) and methylaluminoxane (MMAO) (1.93 M in heptane) were purchased from Akzo Nobel Corporation.Diethylaluminum chloride (Et2AlCl) (1.17 M in toluene) and ethylaluminum sesquichloride (Et3Al2Cl3, EASC, 0.87 M in toluene) were purchased from Acros Chemical.High-purity ethylene was purchased from Beijing Yanshan Petrochemical Company and used as received.Other reagents were purchased from Aldrich, Acros or local suppliers. 1H, 13 C NMR and 19 F NMR spectra were recorded on a Bruker AVANCE 600 MHz instrument at ambient temperature.Chemical shifts (ppm) for the 1 H and 13 C NMR spectra are referenced using TMS as an internal standard; 19 F NMR spectra were referenced to external CF3COOH.Coupling constants (J values) are given in Hz.FT-IR spectra were recorded on a Perkin-Elmer System 2000 FT-IR spectrometer.Elemental analyses were carried out using a Flash EA 1112 microanalyzer.Molecular weights (Mw) and molecular weight distributions (MWD) of the polyethylenes were determined using a PL-GPC220 at 150 o C, with 1,2,4trichlorobenzene as the solvent.The melting points of the polyethylenes were measured from the second scanning run on Perkin-Elmer TA-Q2000 DSC analyzer under a nitrogen atmosphere.In the procedure, a sample of about 2.0 -4.0 mg was heated to 150 o C at a heating rate of 20 o C min -1 , and maintained for 5 min at 150 o C to remove the thermal history and then cooled at a rate of 20 o C min -1 to -20 o C. The 13 C NMR spectra of the polyethylenes were recorded on a Bruker DMX 300 MHz instrument at 135 o C in deuterated 1,2-dichlorobenzene with TMS as an internal standard.

X-ray Crystallographic Studies
Single crystals of the nickel complexes Ni2 and Ni4 were obtained by layering diethyl ether onto their dichloromethane solutions at room temperature.X-ray determinations were carried out on a Rigaku Saturn 724 + CCD with graphite-monochromatic Mo-Kα radiation (λ = 0.71073 Å) at 173(2) K, the cell parameters were obtained by global refinement of the positions of all collected reflections.Intensities were corrected for Lorentz and polarization effects and empirical absorption.The structure was solved by direct methods and refined by full-matrix least squares on F 2 .All hydrogen atoms were placed in calculated positions.Structure solution and refinement were performed by using the Olex2 1.2 package. 20etails of the crystal data and structure refinements for Ni2 and Ni4 are shown in Table 5.

Polymerization studies
Ethylene polymerization at 1 atm ethylene pressure.The polymerization at 1 atm ethylene pressure was carried out in a Schlenk tube.Complex Ni4 was added followed by toluene (30 ml) and then the required amount of co-catalyst (EASC) introduced by syringe.The solution was then stirred at 30 o C under 1 atm of ethylene pressure.After 30 min, the solution was quenched with This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins 10% hydrochloric acid in ethanol.The polymer was washed with ethanol, dried under reduced pressure at 40 o C and then weighed.
Ethylene polymerization at 5 / 10 atm ethylene pressure.The polymerization at high ethylene pressure was carried out in stainless steel autoclave (0.25 L) equipped with an ethylene pressure control system, a mechanical stirrer and a temperature controller.At the required reaction temperature, freshly distilled toluene (30 ml) was injected into the autoclave, followed by the complex (2.0 μmol) dissolved in toluene (50 ml).The required amount of co-catalyst (MAO, MMAO, Et2AlCl, EASC) and more toluene (20 ml) were then injected successively to complete the addition.The autoclave was immediately pressurized to high ethylene pressure and the stirring commenced.After the required reaction time, the ethylene pressure was released and the polymer collected and washed with ethanol.Following drying under reduced pressure at 40 ºC and the polymer sample was weighed.

Figure 1 .
Figure 1.ORTEP drawing of Ni2 with thermal ellipsoids at a 30% probability level.Hydrogen atoms and molecule of diethyl ether have been omitted for clarity.

Figure 2 .
Figure 2. ORTEP drawing of Ni4 with thermal ellipsoids at a 30% probability level.Hydrogen atoms have been omitted for clarity.

Table 5 .
Crystal data and structure refinements for Ni2 and Ni4