Thermostable α-Diimine Nickel Complexes with Substituents on Acenaphthequinone-backbone for Ethylene Polymerization

In order to promote the thermostability of α-diimine nickel complex by ligand backbone structure, a series of α-diimine nickel complexes with substituents on acenaphthequinone backbone were synthesized and used as catalysts for ethylene polymerization. When the hydroxyethyl phenoxyl group was introduced to the acenaphthequinone-backbone, the thermal stability and activity of the catalyst could be significantly improved. The catalytic activity of complex C2 [5-(4-(2-hydroxyethyl)phenoxyl)-N,N-bis(2,6-diisopropyl)acenaphthylene-1,2-diimine]nickel(II) dibromide with isopropyl substituents on N-aryl reached 8.2 × 106 g/(molNi·h) at 70 °C and 2 MPa. The activity of [5-(4-(2-hydroxyethyl)phenoxyl)-N,N-bis(2,6-dibenzhydryl-4-menthylphenyl)acenaphthylene-1,2-diimine]nickel(II) dibromide (C3) still maintained at 6.7 × 105 g/(molNi·h) at 120 °C. Compared with C3 containing bulky dibenzhydryl substituents, the activity of C2 was sensitive to the change of the polymerization pressure. However, the polyethylenes obtained from complex C3 had lower branching density. Meanwhile, the molecular weight could reach 971 kg/mol, which is almost 5 times as much as that of the polyethylene obtained from complex C2.


INTRODUCTION
Since Brookhart's seminal discovery on nickel and palladium complexes bearing α-diimine ligands, [1,2] there has been tremendous interest in exploring new α-diimine ligands. These catalysts are capable of producing polymers with various types of branches, and have good tolerance to many polar comonomers. [3−9] Despite their unique properties, Brookharttype α-diimine catalysts suffer from poor thermal stability, which has greatly limited their potential industrial application. Driven by the desire for enhanced thermal stability, higher activity and polymer molecular weight, tremendous efforts have been made on typical α-diimine nickel/palladium complexes, which involve modifying N-aryl substituents and ligand backbone structures. [10−19] For example, Guan and co-workers developed α-diimine ligands bearing different substituents, and designed novel cyclic α-diimine complexes, which showed high activity and thermal stability for ethylene polymerization. [20,21] It is worth mentioning that the bulky substituents on N-aryl can significantly enhance the catalytic properties of the corresponding precatalysts. More recently, the investigation of benzhydryl-derived ligand frameworks aroused high discussion interests. [22−26] Sun's group developed a series of unsymmetrical diimine complexes bearing the dibenzhydryl substituent, which are better for inducing high thermal stability. [22,23] Long's group reported the synthesis of a series of sterically demanding α-diimine nickel(II) complexes by the use of 2,6-bis(diphenyl-methyl)-4-methylaniline. [24,25] Moreover, Chen's group reported a series of dibenzhydryl-based αdiimine complexes bearing a range of electron-donating and electron-withdrawing substituents and investigated the electronic effects on the ethylene polymerization. [26] All of the complexes show highly catalytic activity in ethylene polymerization and the excellent thermal stability of these catalysts makes them suitable for use in industrial polymerization temperature (70−100 °C).
For the modification of ligand backbone structure, some groups provided advanced strategies by increasing the steric bulk of the α-diimine ligand backbone for enhancing the thermal stability of catalysts. [21,27−30] Gao's group reported catalysts containing a camphyl backbone and dibenzobarrelene-derived α-diimine nickel catalysts for ethylene polymerization to promote the thermostability of nickel catalyst by the ligand backbone. [27,28] Sun's group explored the ethylene polymerization performance of the nickel dihalides (X = Cl, Br), bearing 4,5-bis(arylimino)pyrenylidene N^N-ligands. [29] Coates' group has investigated the ethylene polymerization behavior by a dibenzobarrelene-derived α-diimine nickel complex. [30] Comparing with modifications of the N-aryl substituents, there have been few works related to the modifications of the acenaphthequinone backbone structures for improving thermal stability of the catalyst. So, three α-diimine nickel catalysts with substituents on acenaphthequinone backbone (C1−C3) were synthesized in this work. In order to investigate the effect of substituents on the acenaphthequinone backbone, C1−C3 were compard with Brookhart-type catalyst C concerning the effect of the nickel α-diimine complex structure on the thermal stability and catalytic performances in ethylene polymerization. Meanwhile, the effects of the polymerization pressure and temperature on catalyst activity, branching degree, and molecular weight and distribution of polymers were also investigated.

EXPERIMENTAL General Considerations
All manipulations of the air-and/or moisture-sensitive materials were carried out under a dry argon atmosphere using standard Schlenk techniques. Toluene and hexane were dried through an activated molecular sieve (4Å) and then refluxed and distilled over sodium diphenylketyl complex prior to use. Dichloromethane was distilled from calcium hydride. Methylaluminoxane (MAO, 1.5 mol/L solution in toluene) and diethylaluminium chloride (AlEt 2 Cl, 1.0 mol/L in toluene) were purchased from Infinity Scientific (Beijing) Co., Ltd. 5-Bromoacenphthylene-1,2dione and 2,6-(diphenylmethyl)-4-methylaniline were purchased from Zhengzhou Te Li Kai Chemical Technology Co., Ltd. High-purity ethylene was purchased from Air Products&Chemicals (Tianjin) Co., Ltd. and used as received. All other chemicals were obtained commercially and used as received.
Fourier transform infrared (FTIR) spectra were recorded on a Bruker Tensor-27 spectrometer using the pressed KBr pellets. Elemental analysis was carried out using a Flash EA 1112 microanalyzer. NMR spectra of ligands and complexes were recorded on a Bruker DMX 400 MHz instrument at ambient temperature in deuterated chloroform using TMS as an internal standard; δ values were given in ppm and J values in Hz. MS elemental analyses were performed on a Vario EL microanalyzer. Molecular weights and molecular weight distribution of polyethylene were determined by a PL-GPC 220 at 150 °C, with 1,2,4-trichlorobenzene as the solvent. 13 C-NMR spectra of polyethylene were also recorded on a Bruker DMX 400 MHz instrument at 120 °C in deuterated 1,2-dichlorobenzene with TMS as an internal standard. The melting temperatures (T m ) and the fusion enthalpy (ΔH f ) of the polymers were determined by differential scanning calorimetry (DSC) with a DSC-Diamond (Perkin-Elmer Co.) operating at a heating rate of 10 °C/min from 0 °C to 150 °C under nitrogen atmosphere, and T m was determined in the second scan. The crystallinity (χ c , %) was calculated from the heat of fusion, (ΔH f /ΔH f 0 ) × 100%, where ΔH f 0 is the heat of fusion of folded-chain polyethylene (289.0 J/g). [31]

Ethylene Polymerization
Different polyethylene samples were synthesized by using different α-diimine nickel (II) complexes (Scheme 1). The polymerization was carried out in a 100 mL reactor with stirring bar and temperature and pressure controller. It should be noted that the reactor was dried under vacuum at 100 °C for 30 min and cooled to room temperature under an argon atmosphere, then purged with dry argon two times and ethylene once. Then, toluene and a prescribed amount of catalyst solution and cocatalyst solution were injected and the mixture was maintained at the desired temperature and ethylene pressure. The polymerization mixture was quenched by the addition of 10 vol% HCl/ethanol solution after 60 min. The obtained polymers were washed with water and ethanol, and then dried at 60 °C in vacuum oven to constant weight.

Synthesis of α-Diimine Ligands and Complexes C1−C3
The complex C1 was synthesized according the literature. [32] The synthetic procedures of α-diimine nickel(II) complexes C2 and C3 are shown in Scheme 2.

Ethylene Polymerization with α-Diimine Nickel Complexes
It is well known that the α-diimine nickel complex C (shown in Scheme 1) showed a poor thermal stability. As shown in Table 1, the activity decreased rapidly as the polymerization temperature increased from 30 °C to 70 °C. In order to study the modification in backbone group for improving thermal stability of catalyst, we synthesized complex C1, which has a phenoxy substituent attached to the acenaphthequinone-backbone and isopropyl substituents on N-aryl structures.
However, contrary to our original wish, the activity of complex C1 was lower than that of complex C at 30 and 50 °C. This indicates that the benzyloxy group is not good for the activity of catalyst. The same phenomenon was also found by Kim et al. in the ethylene polymerization catalyzed by pyridine-bis-imine iron catalyst with benzyloxy substituent. [33] The authors suggested that the electron-donating characteristic of benzyloxy group resulted in a decreased coordination power with respect to ethylene monomer, thus causing a lower propagation rate. Furthermore, the decreased activity was also caused by complicated complexation between cationic active species and benzyloxy group. This complexation stabilized the active species and retarded the coordination of incoming monomer, resulting in decreased catalytic activity. Similarly, for the α-diimine nickel catalyst C1 with benzyloxy group, the decrease of catalyst activity was related to the electron-donating character, as well as the complexation between cationic active species and benzyloxy group. Despite all of these, the activity of C1 was higher than that of C at 70 °C, which shows the steric hindrance of phenoxy substituent can improve the thermal stability of α-diimine complex to a certain extent.
Furthermore, we synthesized complex C2, which has an additional hydroxyethyl group on the benzyloxy group compared to complex C1. However, the hydroxyethyl group made C2 show a higher catalytic activity. At 70 °C, the activity of complex C2 was 1.76 × 10 5 g/(mol Ni ·h), which is higher not only than that of C1, but also than that of C. It is suggested that the presence of hydroxy group can make up for the negative effects of benzyloxy group, and improve the catalytic activity, especially at the higher polymerization temperature. One possible reason is that hydroxy group may interact with the cocatalyst anion in the active center ion pair, which increased the distance between cation species and anion. So, the insertion of ethylene monomer was easier, and catalytic activity and thermal stability were enhanced. Another possible reason is that after the hydroxy group reacted with the cocatalyst, the formed bulkier substituent blocked the forma-tion of complexation between cationic active species and benzyloxy group. Anyhow, the hydroxy group plays an important role in improving the catalytic behaviors. Meanwhile, it can be seen that complex C2 with AlEt 2 Cl showed higher activity than with MAO in ethylene polymerization. So, the cheaper AlEt 2 Cl was selected as the co-catalyst for further investigations.
For investigating the stability of complex C2 at elevated temperature and pressure, polymerizations were conducted at different temperatures and ethylene pressures ( Table 2). At 70 °C and 0.5 MPa ethylene pressure, the catalytic activity of C2 reached 2.36 × 10 6 g/(mol Ni ·h) (entry 2, Table 2), which suggested that C2 had good thermal stability. As the temperature increased to 90 °C, the activity reduced by an order of magnitude, which should result from the smaller steric hindrance of isopropyl on the N-aryl. Nevertheless, although the ortho-site substituents of N-aryl structure are small steric hindrance isopropyl, the hydroxyethyl phenoxyl group on acenaphthequinone-backbone made complex C2 show higher catalytic activity and thermal stability than reported αdiimine catalysts with isopropyl substituents on N-aryl.
To further improve the thermal stability of the catalyst, two dibenzhydryl groups were introduced into the ortho-N-aryl substituents, and a new complex C3 was obtained. At 70 °C, the activity of C3 was lower than that of C2, which indicates that the dibenzhydryl groups could inhibit the insertion of monomer and lead to the decrease of activity. However, when the temperature was improved to 90 °C, as compared to complex C2, complex C3 demonstrated much greater thermal stability by showing higher activity by up to an order of magnitude (2.0 × 10 6 g/(mol Ni ·h), entry 6, Table 2). It is worth mentioning that the activity of C3 still maintained at 6.7 × 10 5 g/(mol Ni ·h) (entry 10, Table 2) at 120 °C. Few studies on catalytic polymerization of α-diimine nickel catalysts at such high temperatures have been reported in the literatures.
Furthermore, it is found that C3 had higher catalytic activity than C' under the same polymerization condition (entries 7 and 14, Table 2). It indicates that the hydroxy group makes the dibenzhydryl-based α-diimine complex have better activity. This is consistent with the comparison of catalytic activity between C2 and C. These data clearly indicate that under the dual actions of large steric hindrance substituent dibenzyl and hydroxyl on the backbone, the α-diimine complex C3 has high thermal stability and high catalytic activity at high temperature.
Employing different ethylene pressures, higher activities of complexes C2 and C3 were gradually observed along with increasing the ethylene pressure from 0.5 MPa to 2 MPa (entries 2, 4, and 5 for C2, entries 7, 12, and 13 for C3, Table 2). By comparison, it is found that the increase of C2 activity was more notable. At 2 MPa and 70 °C, complex C2 showed the better activity up to 8.2 × 10 6 g/(mol Ni ·h). The significant in- crease of activity suggests that the small steric hindrance isopropyl substituents on N-aryl can promote the insertion reaction of ethylene, and the polymerization activity is remarkably improved under high pressure and high ethylene concentration conditions. However, the steric hindrance of the diphenylmethyl substituents on the C3 N-aryl is significantly greater than that of the isopropyl substituents on C2, which is not conductive to the insertion reactions for a large amount of ethylene and results in a lower polymerization rate and relatively slow increase of activity for complex C3 when the ethylene pressure is increased. The catalyst structure also has great effect on the molecular weight of the resulting polymer. The polyethylene obtained from complex C3 showed the high molecular weight of 971 kg/mol (entry 12, Table 2), which is almost 5 times as much as that of the polyethylene obtained from complex C2 (185 kg/mol, entry 4, Table 2) under the same conditions. For complex C3, the large steric hindrance of dibenzhydryl groups can result in a lower polymerization rate, but they also effectively inhibit the chain transfer reaction, which promotes the chain growth reaction.
When the ethylene pressure was increased from 0.1 MPa to 0.5 MPa, the molecular weight of the polymer produced by complex C3 increased from 212 kg/mol to 696 kg/mol. However, as the pressure further increased to 1 MPa, the molecular weight increased slowly as indicated by the GPC curves in Fig. 1. These observations suggest that the higher ethylene concentration contributes to ethylene coordination and insertion; meanwhile, it also suppresses the chain transfer reaction, leading to the higher molecular weights.
Temperature also influences the molecular weight of the resultant polyethylene obviously. The molecular weight was decreased at elevated temperatures. For example, the molecular weight dropped from 696 kg/mol at 70 °C to 360 kg/mol at 110 °C for complex C3 (entries 7 and 9, Table 2). This phenomenon should be related to the accelerated chain transfer at high temperatures.
The resultant polyethylenes were characterized by differential scanning calorimetry (DSC). Their melting temperatures (T m ) and crystallinity degrees are shown in Table 3. Comparison of melting temperatures clearly demonstrates that complex C2 with 2,6-diisopropyl substituents afforded polyethylenes with much lower melting temperatures than C3 with dibenzhydryl substituents. Following the same trend as other reported α-diimine Ni(II) catalysts, the melting point of the obtained polymer decreased or disappeared with the increasing polymerization temperatures. As the temperature increased, the chain growth and chain walking accelerate and form more branches. The high number of branches hinders the crystallization of the polymer. The different melting temperatures should be due to the different branching degrees. So, the precise microstructures of representative polyethylenes were further studied by 13 C-NMR spectroscopy, and the results are summarized in Table 3. As shown in Table 3, when Table 3 The melting behavior and branched chain distributions of polyethylenes obtained with C2 and C3.

Entry
Cat.   Table 2). the temperature increased from 50 °C to 70 °C, the branching density of polyethylene produced using complex C2 increased from 96/1000C to 114/1000C, and the melting point changed from 47 °C to no melting point (entries 1 and 2, Table 3).
As shown in Fig. 2 and Table 3, being interpreted according to the literature, [34] the polyethylene obtained by C2 at 70 °C and 0.5 MPa possessed 114/1000C branches, including methyl (63.7%), ethyl (10.1%), propyl (5.1%), butyl (6.6%), amyl (5.1%), and LCB (9.4%), while the polyethylene obtained by C3 under the same polymerization condition had 61 branches/1000C, containing 66.0% methyl, 2.7% ethyl, 3.7% propyl, 2.2% butyl, 2.4% amyl, and 23.0% LCB. The reducing branching density could be reasonably attributed to the slow chain walking suppressed by ortho bulky substituents. [23,25,26] However, beside the methyl chains, the polyethylenes containing high content of LCB were also obtained by C3. Similar observations have been found by Gao et al., when they used α-diimine nickel catalysts with bulky aryl on aniline moieties to synthesize polyethylenes. [35] They proposed a different "chain walking" mechanism to interpret the phenomenon. As described in the mechanistic model, the long branching chains were formed by ethylene insertion into the primary Nialkyl species originating from nickel migration to methyl terminal of the growing chain because of restricted ethylene insertion into secondary Ni-alkyl species with an α-ethyl or a bulkier alkyl group. For branched chain distributions of polyethylenes obtained from the complex C3, another interesting phenomenon is that the proportion of short branch chains decreased and meanwhile the proportion of LCB increased with the increase of reaction temperature, implying that the nickel migration to methyl terminal of the growing chain, as the different "chain walking" mechanism proposed by Gao, became very easy. For the polyethylenes obtained from the complex C3, the increases in pressure and monomer concentration were favorable for the linear growth reaction of the chain, which reduced the degree of polymer branching. If the pressure was too high, e.g., 2 MPa, there was no significant change in the branching density. Although the polymers in entries 8 and 9 had similar branching densities, the polymer in entry 8 had a slightly higher melting point than that in entry 9.

CONCLUSIONS
To conclude, we synthesized a series of α-diimine nickel complexes with substituents on acenaphthequinone backbone. Complex C1 with phenoxyl group on acenaphthequinone backbone had lower activity than that with no substituents. Howerver, when the phenoxyl group was changed to hydroxyethyl phenoxyl group, the thermal stability and activity could be improved obviously. Complex C2 showed higher catalytic activity than reported α-diimine catalysts with isopropyl substituents on N-aryl, and the activity of C3 still maintained at 6.7 × 10 5 g/(mol Ni ·h) at 120 °C. For the complexes with hydroxyethyl phenoxyl group (C2, C3), the substituents on N-aryl also had a lot of influence on catalyst activity, branch degree, and molecular weight of resultant polymers. At 70 °C and below, complex C2 containing 2,6-diisopropyl substituents had a higher activity and was more sensitive to pressure variations than that containing bulky dibenzhydryl substituents. In comparison, complex C3 containing larger substitution group on N-aryl had better activity over 90 °C, and the resultant polyethylene from C3 had higher molecular weight and lower branching degree. They are more conducive to industrial application. In addition, this kind of complexes could be used to prepare the supported catalysts by the reaction of their hydroxyl group with the active groups of supports. The further exploration of the supported catalyst is in progress.

Electronic Supplementary Information
Electronic supplementary information (ESI) is available free of charge in the online version of this article at https://doi.org/10.1007/s10118-020-2430-x.    Table 2).