Sterically and Electronically Modified Aryliminopyridyl-Nickel Bromide Precatalysts for an Access to Branched Polyethylene with Vinyl/Vinylene End Groups.

A series of 2-((arylimino)ethyl)pyridine derivatives (L1-L5), each containing N-2,4-bis(dibenzocycloheptyl) groups with variations in the steric/electronic properties of the ortho-substituent in the aryl ring, and the corresponding nickel bromide precatalysts [2-N{2,4-(C15H13)-6-R-C6H2}C7H7N]NiBr2 (R = Me (Ni1), Et (Ni2), i-Pr (Ni3), Cl (Ni4), or F (Ni5)), have been prepared in high yield. All the precatalysts are air-stable and characterized by Fourier transform infrared spectroscopy and elemental analysis. The molecular structures of Ni2 and Ni5 were proved through single-crystal X-ray diffraction analysis. The steric/electronic impact of the catalysts on ethylene polymerization and the resulting polymer properties were studied. Upon activation with either MAO or EASC, all the complexes displayed higher activities (up to 7.93 × 106 g of PE (mol of Ni)-1 h-1 with MAO) in ethylene polymerization and produced moderate to highly branched unsaturated polyethylene with a molecular weight of up to 16.55 kg/mol with narrow dispersities (1.6-2.4). Significantly, the generated polyethylenes are branched and unsaturated with a major class of internal double bond (-CH=CH-) as compared to the terminal double bond (-CH=CH2) (vinylene/vinyl = 9.8:1 to 1.8:1). Notably, their catalytic activities, types of unsaturation, and branches are highly affected by the nature of the ortho-substituent and reaction temperature. Moreover, the precatalysts Ni4 and Ni5 (with N-ortho = Cl and F) exhibited lower catalytic activities, produced low-molecular-weight polyethylene with a high melt temperature and the least number of branches with an increased level of terminal double bonds.

Introduction

Since the 1990s, the effective design of a nickel precatalyst for ethylene polymerization has been the subject of continued investigations.[1] Especially, broad branching contents of the obtained polyethylenes, because of the propensity of the catalysts with facile “chain walking”, led to producing promising materials with a wide range of applications.[2] Structural modifications of catalysts have been explored, which include the diazabutadiene-class of complexes[3] and other nickel complexes in which the steric and electronic effects of the ligand backbone, as well as the N-aryl substituents, played a vital role in the catalytic performance and polymer properties.[4] An early example of a nickel precatalyst bearing pyridine-based ligand framework (A, Chart ),[5] in which the steric hindrance at the ortho-position of the N-aryl ring played a significant role, leading to produce mono and dinuclear complexes and showed high activity for ethylene polymerization. The catalytic activity has been influenced by variations in the pyridine moiety using methyl or mesityl in nickel precatalysts for ethylene polymerization (B, Chart ),[6] and generated high molecular weight polyethylene with narrow dispersities and a high degree of branches. Over the past several years, our group has developed several effective cycloalkyl-fused pyridinyl–nickel complexes for ethylene polymerization.[7] It has been observed that variations in the N-aryl ring particularly using the electron-withdrawing groups at the para position showed an enhanced effect on the catalytic activity for ethylene polymerization, producing low-molecular-weight-branched polyethylene with narrow dispersities.[8] Furthermore, the incorporation of dibenzhydryl substituents into the N-naphthyl ring (CChart )[4l,4m] led to enhanced catalytic activity and generated low-molecular-weight polyethylene with narrow dispersities. Meanwhile, highly active and thermally stable nickel precatalysts have been investigated for ethylene polymerization and produced materials of semicrystalline nature in the range of ultrahigh molecular weight.[9] It has been recognized that the ortho-substituents in the 2-(arylimino)pyridine-nickel and palladium precatalysts could also be important for catalytic performance, thermal stability, and polymer properties.[10] With this view, the nickel complexes (D and E, Chart )[4k] bearing dibenzhydryl substituents at the 2,4-positions of the N-aryl ring have been investigated for ethylene polymerization. These complexes exhibited high catalytic activity and produced highly branched polyethylene with narrow molecular weight distributions. Similarly, the 2,4-fluorenyl substituents in the N-aryl ring proved substantial (F, Chart )[11] and produced high molecular weight polyethylene with moderate to broad polydispersities. However, the incorporation of cycloalkyl groups into the N-aryl group showed slightly lower activity and produced low-molecular-weight polyethylene.[7b] More recently, we have developed 8-(arylimino)-5,6,7-trihydroquinoline-nickel precatalysts for ethylene polymerization and generated branched polyethylene of vinyl/vinylene functionalities with narrow dispersities; such unsaturated polymer materials highly demanded further functionalization.[12]

Chart 1

Development of 2-(Arylimino)pyridine-nickel(II) Halide Precatalysts Bearing Modified N-Aryl Substituted via Benzhydryl, Fluorenyl, and Dibenzocycloheptyl Groups (A–G).

In the present work, we are developing dibenzocycloheptyl groups as the counterpart of dibenzhydryl groups in D and E or fluorenyl groups in F (Chart ); we assumed that the steric effect, as well as acidity of methine protons in 2,4-bis(dibenzhydryl)-6-R-imine and 2,4-bis(fluorenyl)-6-methyl-imine along with variable steric/electronic properties of the ortho-position near the central metal may have some influence on the catalytic performance and polymer properties.[13] In this regard, we report 2-(2,4-bis(dibenzocycloheptyl-6-R-phenylimino)ethyl)pyridine-nickel precatalysts (G, Chart ), where R = Me, Et, i-Pr, Cl, and F. The influence of the steric and electronic properties of the ortho-substituent was investigated and then catalytic evaluation of G was investigated in depth that is how the co-catalyst as well as reaction parameters can impact the catalytic performance and properties of the resulting polyethylene. Additionally, nearly related D, E, and F of the reported and current G precatalysts are also compared.

Results and Discussion

Synthesis and Characterization

Using the Friedel–Crafts alkylation reaction, five examples of anilines (A1–A5) such as 2,4-(C15H13)-6-R-C6H2NH2 {R = Me (A1), Et (A2), i-Pr (A3), Cl (A4), and F(A5)} were prepared and reasonable yield was obtained (Scheme S1). The condensation reactions of the respective aniline (A1–A5) with 1 M equivalent of acetylpyridine offered a series of ligands 2-N{2,4-(C15H13)-6-RC6H2}C7H7N {R = Me (L1), Et (L2), i-Pr (L3), Cl (L4), or F (L5)} as a yellow powder in 56–86% yield (Scheme ). All the ligands were characterized by Fourier transform infrared (FT-IR), 1H, and 13C NMR spectroscopy as well as by elemental analysis. Treatment of ligands (L1–L5) with (DME)NiBr2 (DME = 1,2-dimethoxyethane) in a mixture of ethanol and dichloromethane at room temperature afforded the corresponding nickel(II) bromide complexes (Ni1–Ni5) as a green powder in good yield (82–95%). All these complexes were air-stable and characterized by FT-IR spectroscopy and elemental analysis, as well as single-crystal X-ray diffraction analysis for Ni2 and Ni5.

Scheme 1

Synthesis of Iminopyridine Ligands (L1–L5) and Their Corresponding Nickel(II) Bromide Complexes (Ni1–Ni5)

In the FT-IR spectra of the complexes, the absorption bands for ν(C=N) shifted to lower wavenumbers of 1627–1621 cm–1 compared to their corresponding free ligands (L1–L5) in the range of 1645–1634 cm–1, such shifts toward lower wavenumbers reveal the effective coordination between the donor nitrogen atom of the ligand framework and the metal center.[4k,4m,5,8a,9,11,12] The elemental analysis results were consistent with the results of the prepared compounds. Moreover, single-crystals of Ni2 and Ni5 suitable for X-ray determination were grown by slow diffusion of heptane into a solution of the respective complex in dichloromethane at room temperature. During our attempt to growing the crystals, we obtained the bis-ligated Ni2(a) and 2Ni5·HO as the dinuclear complex. The images of the corresponding structures are shown in Figures and 2, respectively; their selected bond lengths and angles are collected in Table . As shown in Figure , the structure of the biligated Ni2(a) complex can be best described as distorted square-pyramidal geometry around the nickel center comprising four chelating nitrogen atoms (N(1), N(2), N(1a), and N(2a)) of the two ligands, one bromide (Br(1)), the other bromide (Br(2)) acting as a free-anion, similar findings have been reported using bis(2-((2,4-dibenzohydryl-6-ethylphenylimino)pyridyl nickel bromide complexes.[4k] In addition, each ligand forms five-membered heterocyclic rings around the nickel center as Ni(1), N(1), C(5), C(6), and N(2) and Ni(1), N(1a), C(5)a, C(6a), and N(2a), in which the atoms C(5) and C(6) deviate (0.139–0.44 Å) versus C(5a) and C(6a) (0.169–0.242 Å) from the coplane of atoms Ni(1), N(1), and N(2) versus Ni(1), N(1a), and N(2a), respectively. The Ni(1) atom deviates by 0.187 or 0.275 Å from the coplane of atoms N(1), C(5), C(6), and N(2) or N(1a), C(5a), C(6a), and N(2a), respectively. Moreover, the plane of bidentate nitrogen atoms was alongside the nickel center and the N-aryl plane is nearly perpendicular with a dihedral angle of 85.51°, similar observations have been reported in the literature for 2-aryliminopyridylnickel analogs.[2d,4d,4k−4m,5,7c] The bond length of Ni(1)–N(1)pyridine (2.039(4) Å) is slightly shorter than the bond length of Ni(1)–N(2)imine (2.059(4) Å), highlighting more effective coordination of the pyridine nitrogen atom compared to the imine nitrogen atom, which is consistent with the reported literature.[4k−4m,5,8b,9−12]

Figure 1

ORTEP diagram of bis-ligated Ni2(a) with thermal ellipsoids shown at a probability level of 30%. All the hydrogen atoms have been omitted for clarity.

Figure 2

ORTEP diagram of 2Ni5·HO with thermal ellipsoids shown at a probability level of 30%. All the hydrogen atoms have been omitted for clarity.

Table 1

Selected Bond Lengths (Å) and Angles (deg) of Ni2(a) and 2Ni5·HO

Ni2(a)		2Ni5·H2O
Bond Lengths (Å)
Ni(1)–N(1)	2.039(4)	Ni(1)–N(1)	2.042(3)
Ni(1)–N(2)	2.059(4)	Ni(1)–N(2)	2.102(3)
Ni(1)–N(1a)	2.051(4)	Ni(1)–Br(1)	2.4975(8)
Ni(1)–N(2a)	2.057(4)	Ni(1)–Br(2)	2.5384(6)
Ni(1)–Br(1)	2.4014(8)	Ni(1)–O(1)	2.161(4)
N(2)–C(6)	1.273(6)	N(2)–C(6)	1.275(4)
N(2)–C(8)	1.437(6)	N(2)–C(8)	1.437(4)
Bond Angles (deg)
N(1)–Ni(1)–N(2)	78.98(15)	N(1)–Ni(1)–N(2)	78.58(11)
N(1a)–Ni(1)–N(2a)	78.56(15)	N(1)–Ni(1)–Br(1)	92.98(11)
N(1)–Ni(1)–Br(1)	95.46(10)	N(1)–Ni(1)–Br(2)	170.27(10)
N(1a)–Ni(1)–Br(1)	98.32(11)	N(2)–Ni(1)–Br(1)	100.72(9
N(2)–Ni(1)–Br(1)	119.82(10	N(2)–Ni(1)–Br(2)	92.98(7)
N(2a)–Ni(1)–Br(1)	132.52(10)	O(1)–Ni(1)–Br(1)	173.09(8)
N(1)–Ni(1)–N(2a)	92.84(14)	N(1)–Ni(1)–O(1)	87.25(14)
N(1a)–Ni(1)–N(2)	93.22(15)	N(2)–Ni(1)–O(1)	86.10(12)
N(1)–Ni(1)–N(1a)	166.18(15)	N(1)–Ni(1)–Br(2i)	95.19(9)
N(2)–Ni(1)–N(2a)	107.66(14)	N(2)–Ni(1)–Br(2i)	166.36(9)

The molecular structure of 2Ni5·HO as shown in Figure revealed that the centrosymmetric dimer in which one bromide (Br(2) or Br(2)) per nickel center was bridged between the two nickel centers while the remaining bromides (Br(1) or Br(1)) occupied the apical position, similar structural observations have been reported in the literature.[2d,4k,7a,7c,8a] In addition, each nickel center is bounded by two neutral chelating nitrogen atoms of the ligand and a molecule of water per nickel center, which acts as a monodentate ligand to Ni(1) or Ni(1), this led to a distorted octahedral geometry conformed to the nickel center {O(1)–Ni(1)–Br(1) 173.09(8)°, N(1)–Ni(1)–Br(2) 170.27(10)°, and N(2)–Ni(1)–Br(2) 166.36(9)°}.[4k−4m,12] The planes between the N(1), Ni(1), and N(2) chelating rings and the aryl group are nearly perpendicular with a dihedral angle of 87.01°, similar observations have been reported for 2-aryliminopyridylnickel analogs.[4d,7a,14] Besides, each nickel forms five-membered rings consisting of Ni(1), N(1), C(5), and C(6), in which C(5) and C(6) deviate from the coplane of atoms Ni(1), N(1), and N(2) with a distance of 0.082 or 0.144 Å, respectively. Ni(1) deviates at 0.152 Å from the coplane consisting N(1), C(5), C(6), and N(2). There is no direct bonding between the nickel atoms; however, the intermolecular distance of Ni1···Ni1 was observed to be 3.544 Å.[4d,4k,9,11−14] Furthermore, the Ni(1)–N(1)pyridine bond length (2.042(3) Å) is shorter than the Ni(1)–N(2)imine bond length (2.102(3) Å), indicating the better donor capability of the pyridine nitrogen atoms.[4k−4m,5,8b,9,11,12]

Ethylene Polymerization

To identify the most suitable co-catalyst for ethylene polymerization, five different alkyl-aluminum reagents such as EASC (ethylaluminum sesquichloride), Et2AlCl (diethylaluminum chloride), Me2AlCl (dimethylaluminum chloride), MAO (methylaluminoxane), and MMAO (modified methylaluminoxane) were assessed with Ni1 selected as the test precatalyst. A typical polymerization run as conducted under 10 atm of C2H4 over 30 min in toluene (100 mL) at a temperature of 30 °C; and the results obtained are summarized in Table . Inspection of the results shows that the catalytic performance of the co-catalysts decreased in the following order as MAO > EASC > MMAO > Me2AlCl > Et2AlCl. Therefore, considering the level of activity as well as the different chemical nature of the aluminum activators, MAO and EASC were selected for in-depth ethylene polymerization reactions.

Table 2

Evaluation of Cocatalysts by Using Precatalyst Ni1a

run	co-cat.	Al/Ni	T (°C)	t (min)	PE (g)	activityb	Mwc	Mw/Mnc	Tm (°C)d
1	EASC	400	30	30	6.38	6.38	6.93	2.3	77.9
2	Et2AlCl	400	30	30	4.93	4.93	6.92	2.3	85.7
3	Me2AlCl	400	30	30	5.84	5.84	5.19	2.1	71.4
4	MAO	2000	30	30	6.87	6.87	9.21	2.0	94.7
5	MMAO	2000	30	30	5.96	5.96	7.55	2.3	86.4

Conditions: 2.0 μmol Ni, 100 mL toluene, 10 atm of C2H4, 30 min, and 30 °C.

106 g of PE (mol of Ni)−1 h–1.

Mw: kg mol–1, determined by GPC.

Determined by DSC.

Ethylene Polymerization Using Ni1–Ni5/MAO

To establish an optimum ethylene polymerization condition, precatalyst Ni1 was employed with MAO as a co-catalyst under 10 atm pressure of ethylene. At a fixed temperature of 30 °C and a run time of 30 min, the Al/Ni ratio was gradually increased from 1500 to 3000 (runs 1–5, Table ). Inspection of the results shows that in a relatively low ratio of 2500, the catalytic activity reached the maximum of 7.93 × 106 g PE mol–1(Ni) h–1 (run 3, Table ). Above 2500, the activity gradually decreased and at 3500 the lowest activity of 5.08 × 106 g PE mol–1(Ni) h–1 was reached (run 5, Table ). The molecular weight of the obtained polyethylene was increased in a similar way reaching 9.35 kg mol–1 with a molar ratio of 2500 (run 3, Table ) and then steadily decreased to 7.93 kg mol–1 (run 5, Table ). This latter decrease in the molecular weight can be ascribed to higher rates of chain transfer as compared to chain propagation at a higher molar ratio of the co-catalyst.[3b−3d,5,11,15] In addition, the molecular weight distribution (Mw/Mn = 2.0–2.3) remained narrow and unimodal as shown in the GPC curves (Figure ).

Table 3

Optimization of the Polymerization Conditions Using Ni1/MAOa

run	T (°C)	t (min)	Al/Ni	PE (g)	activityb	Mwc	Mw/Mnc	Tm (°C)d
1	30	30	1500	5.64	5.64	7.13	2.0	85.4
2	30	30	2000	6.87	6.87	9.21	2.0	94.7
3	30	30	2500	7.93	7.93	9.35	2.3	92.7
4	30	30	3000	6.81	6.81	8.59	2.3	89.7
5	30	30	3500	5.78	5.08	7.93	2.2	86.5
6	20	30	2500	4.48	4.48	15.19	2.4	104.8
7	40	30	2500	5.31	5.31	5.35	2.1	69.8
8	50	30	2500	2.06	2.06	4.30	2.0	62.6
9	30	05	2500	1.21	7.25	7.44	2.2	85.7
10	30	15	2500	3.76	7.52	7.89	2.3	96.1
11	30	45	2500	10.04	6.69	10.63	2.5	97.0
12	30	60	2500	11.15	5.58	12.43	2.5	101.0
13e	30	30	2500	2.80	2.80	5.73	2.2	68.8
14f	30	30	2500	trace	trace

Conditions: 2.0 μmol Ni1, 100 mL toluene, and 10 atm of C2H4.

106 g of PE (mol of Ni)−1 h–1.

Mw: kg mol–1, determined by GPC.

Determined by DSC.

5 atm of C2H4.

1 atm of C2H4.

Figure 3

GPC curves of the obtained polyethylene using Ni1/MAO at different Al/Ni molar ratios (runs 1–5, Table ).

Using the fixed Al/Ni molar ratio of 2500 with a reaction time of 30 min, the temperature effect was studied from 20 to 50 °C and the highest activity was achieved at 30 °C (runs 3, 6–8, Table ). Upon further increasing the temperature, the catalytic activity rapidly decreased to 2.06 × 106 g PE mol–1(Ni) h–1 (run 8, Table ) at 50 °C, which can be attributed to the deactivation of active species and lower solubility of ethylene at elevated temperature.[3b−3d,5,7c,8,9,11,14c] The molecular weight of the obtained polymer decreased from 15.19 to 4.30 kg mol–1 (run 3, 6–8 Table ) as a function of elevation in temperature, suggesting a higher chain termination rate compared to chain propagation at an elevated reaction temperature.[3b−3d,5,7,15,16] Also, the molecular weight distribution remained narrow (Mw/Mn = 2.0–2.4) and unimodal at various temperatures indicating the single-site precatalyst, the corresponding GPC curves are shown in Figure .

Figure 4

GPC curves of the obtained polyethylene using Ni1/MAO at different reaction temperatures (runs 3, 6–8, Table ).

The lifetime of the precatalyst was explored with the Al/Ni molar ratio fixed at 2500 and temperature at 30 °C by conducting the polymerization runs at 5, 15, 30, 45, and 60 min intervals using Ni1/MAO (runs 3, 9–12, Table ). We observed that the catalytic activity was gradually increased in the first 30 min resulting in the maximum activity of 7.93 × 106 g PE mol–1(Ni) h–1 (run 3, Table ). However, the activity slightly decreased in the next 30 min, but still remains relatively high as 5.58 × 106 g PE mol–1(Ni) h–1 even after 60 min (run 12, Table ); this highlights the longer lifetime of the active species.[3c,3d,6,7a,14,15a,15b] The molecular weight of the PE was gradually increased reaching a maximum of 12.43 kg mol–1 at 60 min. Significantly, all the run time of the catalytic reaction and the molecular weight distribution remained narrow (Mw/Mn = 2.2–2.5) and unimodal as shown in the GPC curves (Figure ). The pressure of ethylene also greatly affected the catalytic activity, as a much lower activity at 5 atm and only a trace amount of polymer was observed at 1 atm (runs 13 and 14, Table ). Additionally, the molecular weight of the obtained polymer at 5 atm of ethylene was nearly half to that obtained with 10 atm, which can be attributed to a lower propagation rate at lower ethylene pressure (runs 3 and 13, Table ).[3b−3d,4k,12]

Figure 5

GPC curves of the obtained polyethylene using Ni1/MAO over different reaction times (runs 3, 9–12, Table ).

Using the optimal conditions established for Ni1/MAO (Al/Ni molar ratio = 2500, reaction temp. = 30 °C, and run time = 30 min), the catalytic potential of all the remaining precatalysts (Ni2–Ni5) was evaluated toward ethylene polymerization with MAO (runs 1–5, Table ). The overall relative catalytic activities fell in the following order, Ni1 > Ni2 > Ni3 > Ni4 > Ni5. Evidently, the catalytic activities were greatly influenced by the steric and electronic effects of the N-aryl substituents. In terms of steric properties, the least bulky precatalyst Ni1 displayed the highest performance of 7.93 × 106 g PE mol–1(Ni) h–1 followed by Ni2 and Ni3. In terms of electronic effects, Ni4 and Ni5 displayed low activities because of the combination of electron-withdrawing properties of F or Cl and the steric effect of the dibenzocycloheptyl group. The molecular weight of the obtained polymer was significantly affected by the N-aryl substituents with most sterically bulkier precatalyst Ni3 generating the highest molecular weight of up to 16.55 kg mol–1 followed by Ni2 with 12.19 kg mol–1 and dropped to 9.35 kg mol–1 with Ni1 (runs 1–3 Table ). The lower molecular weight polymers (2.22–5.81 kg mol–1) were observed with ortho-halide counterpart precatalysts. Besides, the molecular weight distribution remained narrow (Mw/Mn = 2.1–2.4) and unimodal; suggesting a single-site active center as shown in the GPC curves (Figure ). In contrast, the melting temperature (Tm) of polyethylene displayed by the precatalysts bearing ortho-halide Ni4 or Ni5 (Tm = 106.8 or 109.0 °C) is higher than that generated by the precatalysts bearing alkyl counterparts as for Ni1–Ni3 (Tm = 89.1–96.1 °C). These observations suggest that electron-withdrawing ortho-substituents (Cl or F) suppressed “chain walking” and favored β-H elimination. It is uncertain to explain this finding but it might be because of weak hydrogen bonding or dipole interaction of ortho-fluorine or chlorine with coordinated ethylene during the insertion transition state; these findings are consistent with the literature.[4k,12,17]

Table 4

Ethylene Polymerization by Using Ni1–Ni5/MAO under Optimized Conditionsa

run	precat.	mass of PE (g)	activityb	Mwc	Mw/Mnc	Tm (°C)d
1	Ni1	7.93	7.93	9.35	2.3	92.7
2	Ni2	5.62	5.62	12.19	2.4	96.1
3	Ni3	2.53	2.53	16.55	2.3	89.1
4	Ni4	2.24	2.24	5.81	2.4	106.8
5	Ni5	2.12	2.12	2.22	2.1	109.0

Conditions: 2.0 μmol Ni, 100 mL toluene, 10 atm of C2H4, 30 min, 30 °C, and the Al/Ni ratio of 2500.

106 g of PE (mol of Ni)−1 h–1.

Mw: kg mol–1,determined by GPC.

Determined by DSC.

Figure 6

GPC curves of the obtained polyethylene using Ni1–Ni5/MAO under optimized conditions (runs 1–5, Table ).

Ethylene Polymerization Using Ni1–Ni5/EASC

With EASC as the co-catalyst, a parallel investigation was performed as that described for the MAO system. Once again Ni1 was selected as the test precatalyst for optimization of polymerization parameters; the results are summarized in Table . Upon varying the Al/Ni molar ratio from 300 to 700, the highest catalytic performance of 6.75 × 106 g PE mol–1(Ni) h–1 was observed at a ratio of 500 at 30 °C (runs, 1–5, Table ). While the highest molecular weight of polyethylene was 7.27 kg mol–1 achieved using an Al/Ni ratio of 400 and then gradually decreased to 6.66 kg mol–1, this reduction in the molecular weight can be attributed to a higher chain transfer rate versus chain propagation rate at a higher molar ratio of the co-catalyst;[3b−3d,5,11,15] the corresponding GPC curves are shown in Supporting Information (S2, Figure S1). The molecular weight distribution remained in the narrow range (Mw/Mn = 2.0–2.3) and was unimodal; suggesting a single-site active species, displaying the same results as those obtained with the Ni1/MAO system.

Table 5

Optimization of the Polymerization Conditions Using Ni1/EASCa

run	T (°C)	t (min)	Al/Ni	mass of PE (g)	activityb	Mwc	Mw/Mnc	Tm (°C)d
1	30	30	300	2.14	2.14	6.93	2.3	77.9
2	30	30	400	6.38	6.38	7.27	2.0	84.8
3	30	30	500	6.75	6.75	6.81	2.0	87.4
4	30	30	600	5.82	5.82	6.75	2.0	84.1
5	30	30	700	5.03	5.03	6.66	2.0	83.9
6	20	30	500	5.04	5.04	12.67	2.3	101.1
7	40	30	500	4.50	4.50	5.01	2.0	70.3
8	50	30	500	1.97	1.97	3.96	1.8	64.0
9	30	05	500	1.03	6.18	6.00	2.0	88.8
10	30	15	500	3.16	6.32	6.26	2.0	81.9
11	30	45	500	7.93	5.29	7.58	2.2	85.1
12	30	60	500	9.31	4.66	7.81	2.1	87.1
13e	30	30	500	2.34	2.34	4.77	1.8	65.3
14f	30	30	500	trace	trace

Conditions: 2.0 μmol Ni1, 100 mL toluene, and 10 atm of C2H4.

106 g of PE (mol of Ni)−1 h–1.

Mw: kg mol–1, determined by GPC.

Determined by DSC.

5 atm of C2H4.

1 atm of C2H4.

Monitoring the activity of Ni1/EASC, the temperature was varied from 20 to 50 °C (runs 3, 6–9), again exhibiting similar trends to that of the Ni1/MAO system and the peak performance of 6.75 × 106 g PE mol–1(Ni) h–1 was observed at 30 °C (run 3, Table ). Furthermore, upon increasing the temperature, a dramatic drop in the activity (1.97 × 106 g PE mol–1(Ni) h–1) was observed at 50 °C (run 8, Table ). This decrease in the activity at elevated temperature can be attributed to the lower solubility of ethylene and the deactivation of the active species. Furthermore, a decrease trend from 12.67 to 3.96 kg mol–1 was observed in the molecular weight of the obtained polyethylene (run 3, 6–8 Table ) as a result of increasing the temperature, suggesting a higher chain termination rate versus chain propagation rate at elevated temperature;[3b−3d,7,15,16] the temperature effects on the molecular weight of the obtained polyethylene are further shown in the GPC curves (Supporting Information S2, Figure S2). Additionally, the molecular weight distribution remained in the narrow range (Mw/Mn = 1.8–2.3) and unimodal at various reaction temperatures.

With a fixed Al/Ni molar ratio of 500 and a reaction temperature of 30 °C, polymerization runs were performed between intervals of 5 and 60 min (runs 3, 9–12, Table ). The activity was gradually increased reaching its peak performance of 6.75 × 106 g PE mol–1(Ni) h–1 (run 3, Table ) at 30 min and then progressively dropped over longer time with the lowest value of 4.66 × 106 g PE mol–1(Ni) h–1 at 60 min (run 12, Table ). The molecular weight of the obtained polyethylene steadily increased as the time was prolonged reaching 7.81 kg mol–1 at 60 min, nevertheless lower than that obtained using the Ni1/MAO system; further illustrated in the GPC curves (Supporting Information S2, Figure S3). Moreover, the unimodal molecular weight distribution in the narrow range (2.0–2.2) was the characteristic feature of the obtained polyethylene. Furthermore, similar observations are made for the effect of pressure on the activity and molecular weight to those observed with Ni1/MAO.

Under the optimized conditions determined for Ni1/EASC such as the Al/Ni molar ratio = 500, reaction temperature = 30 °C, and run time = 30 min, all the remaining nickel precatalysts Ni2–Ni5 were also screened and the results are shown in Table . The activities followed a similar decreasing order to those observed earlier with the MAO-activated precatalysts. Moreover, these EASC-activated precatalysts displayed slightly lower activities than those of Ni1–Ni5/MAO indicating the importance of the alkyl-aluminoxane activator. Almost similar steric and electronic effects can be used for explanation of activities with least steric precatalyst Ni1 again exhibiting the highest activity of 6.75 × 106 g PE mol–1(Ni) h–1, while the highest molecular weight of 14.42 kg mol–1 of polyethylene was achieved with the bulkier precatalyst Ni3 (runs 1 and 3, Table ). Interestingly, similar trends for the molecular weight were observed to those obtained with Ni/MAO system. Moreover, the molecular weight distribution was in the narrow range (Mw/Mn = 1.6–2.2) and was unimodal; suggesting again that the current precatalyst generates a single-site active species as shown in the GPC curves (Supporting Information S2, Figure S4). The melting temperatures (Tm) of the obtained polyethylene again show similar trends to those observed with the Ni/MAO system; similar observations have been reported in the literature.[4k,4l,12,17]

Table 6

Ethylene Polymerization Using Ni1–Ni5/EASC under Optimized Conditionsa

run	precat.	mass of PE (g)	activityb	Mwc	Mw/Mnc	Tm (°C)d
1	Ni1	6.75	6.75	6.81	2.0	87.4
2	Ni2	4.58	4.58	9.76	2.2	85.4
3	Ni3	2.07	2.07	14.42	2.1	83.3
4	Ni4	2.13	2.13	4.01	2.1	99.8
5	Ni5	2.01	2.01	1.41	1.6	101.3

Conditions: 2.0 μmol Ni, 100 mL toluene, 10 atm of C2H4, 30 min, 30 °C, and the Al/Ni ratio of 500.

106 g of PE (mol of Ni)−1 h–1.

Mw: kg mol–1,determined by GPC.

Determined by DSC.

Microstructural Studies of Polyethylenes

The melting point (Tm) of the obtained polymers using Ni1–Ni5 with either MAO or EASC fell in the range 83.3–109.0 °C with a slight difference seen between the cocatalysts (89.1–109.0 °C with MAO versus 83.3–101.3 °C with EASC) as shown in Tables and 6, respectively. This range of melting points would suggest the branching in the polyethylenes. To study the influence of steric/electronic variations in the ligand framework as well as the cocatalyst and temperature effect on the microstructure of polyethylene, high-temperature 1H and 13C NMR spectroscopy was conducted for representative samples of polyethylene. The NMR spectral data of the obtained samples using Ni4/MAO and Ni5 bearing an electron-withdrawing ortho-substituent (Cl or F) were compared to the data obtained using Ni1–Ni3/MAO with an ortho-alkyl group (Me, Et, or i-Pr). All the calculations for the data and peak assignments were performed according to the literature reports;[18] the NMR data and the calculations of the branching levels for the selected polyethylene samples are provided in Supporting Information S3.

In general, the 1H NMR spectra of the obtained polyethylenes using Ni1–Ni5 with MAO (runs 1–5, Table ) disclose the existence of vinyl and vinylene functional groups with characteristics signals at δ 5.91 (Hb) and 5.06 (Ha) with an integration ratio of 1:2 (−CH=CH2) and a signal at δ 5.50 (Hc/Hc′) for the −CH=CH– functional group. Inspection of the spectral results reveals that the vinylene to a vinyl ratio ((−CH=CH−)/(−CH=CH2)) was notably affected by the ortho-substituents in the precatalysts and was in the range of 6.9:1 to 1.8:1 (Figure ) and also see Supporting Information S4, Figure S5. The vinyl end group as a function of precatalysts followed the order as Ni5 ≈ Ni4 > Ni1 > Ni2 > Ni3, suggesting that the electron-withdrawing fluorine or chlorine group at the ortho-position favored generating a polymer with vinyl end groups.

Figure 7

1H NMR spectra of the obtained polyethylene samples with Ni1, Ni2, and Ni5 at 30 °C using MAO (runs 1, 2, 5, Table ); recorded at 100 °C in d-C2D2Cl4.

This vinylene/vinylene unsaturation was further explained using the 13C NMR spectrum of the polyethylene samples obtained using Ni1/MAO (run 1, Table ), indicating the characteristics peaks for the corresponding alkene-carbon atoms such as Ca, Cb, Cc, and Cc′ (Figure ). Analysis of the sample spectrum reveals highly branched contents (68 branches/1000 Cs) including methyl (73.6%), ethyl (2.2%), propyl (1.4%), butyl (3.9%), amyl (1.9%), 1,4-paired methyl (4.6%), 1,6-paired methyl (3.7%), and longer-chain branches (8.7%). Almost similar observations were made for the obtained sample of polyethylene using Ni2/MAO (run 2, Table ), which displayed almost similar types of branching with the corresponding peaks for vinylene and vinyl groups along with upfield saturated peaks and showed total branches of 69/1000 Cs (Supporting Information S4, Figure S6).

Figure 8

13C NMR spectrum of the obtained polyethylene sample with Ni1/MAO at 30 °C (run 1, Table ); recorded at 100 °C in d-C2D2Cl4.

In stark contrast, the 13C NMR spectrum of the obtained polyethylene using Ni5/MAO (run 5, Table ) displayed much lower branches of 17/1000 Cs (Figure ), consisting of methyl (58.4%), butyl (6.1%), 1,4-paired methyl (3.2%), and longer-chain branches (32.3%). Moreover, the signal intensity of the vinyl end groups (Ca and Cb) is higher than that seen for the vinylene group (Hc or Hc′), which is in accordance with the corresponding 1H NMR data (vinylene to vinyl ratio = 1.8:1). These observations suggest that substitution of fluorine at the ortho-position favored β-H elimination and reduced “chain walking”. It is uncertain to explain this finding but it might be because of weak hydrogen bonding between the coordinated ethylene and ortho-fluorine during the insertion transition state. This lower branching level was also supported by the high melt temperature (Tm) of the corresponding polyethylene sample; similar findings have been found in the literature.[2d,8a,12,17c,17d,18]

Figure 9

13C NMR spectrum of the obtained polyethylene sample with Ni5/MAO at 30 °C (run 5, Table ); recorded at 100 °C in d-C2D2Cl4.

Additionally, the high-temperature 13C NMR spectrum of polyethylene samples obtained with Ni4/MAO (run 4, Table ) was recorded and also showed lower branches of 34/1000 Cs (Supporting Information S4, Figure S7). The explanation for this lower degree of branching is again ambiguous but may be because of the high electronegative nature of chlorine, which may interact with the coordinated ethylene that suppresses chain-walking and favors β-H elimination. Furthermore, the temperature greatly influences the catalytic performance and its effect was investigated on the microstructure of the polyethylene generated by Ni1/MAO at 30, 40, and 50 °C (runs 3, 7, 8, Table ). The 1H NMR spectra displayed that upon increasing the temperature from 30 to 50 °C, the vinylene/vinyl ratio increases from 5.81:1 to 9.82:1 (Figure ); suggesting that higher temperature had increased the rate of termination and hence within the polymer chain increased the vinylene portions.

Figure 10

1H NMR spectra of the obtained polyethylene samples with Ni1/MAO at 30, 40, and 50 °C (runs 3, 7, 8, Table ); recorded at 100 °C in d-C2D2Cl4.

In contrast to 30 °C, the 13C NMR spectrum of a sample obtained at 50 °C (run 8, Table ) showed a higher number of branches of 121/1000 Cs (Supporting Information S4, Figure S8); suggesting that higher temperature increases β-H elimination. Additionally, the effect of the co-catalyst on the microstructure of polyethylene was studied and the 1H and 13C NMR spectra of the obtained samples using Ni1 (run 1, Table ) and Ni5 (run 5, Table ) with EASC were recorded. Similar variations in the melting points were seen to those observed using Ni/MAO, which suggests similar effects on the microstructure. Likewise, again both types of the unsaturated peaks were observed with a downfield chemical shift in the 1H NMR spectra with the vinylene/vinyl ratio decreasing from 7.6:1 to 2.5:1, which can be attributed to the fact that the electron-withdrawing substituent (fluorine) at the ortho-position compared to the ortho-alkyl substituent favored to produce polyethylene of vinyl end groups (Figure ).

Figure 11

1H NMR spectra of the obtained polyethylene samples with Ni1 and Ni5 using EASC at 30 °C (runs 1 and 5, Table ); recorded at 100 °C in d-C2D2Cl4.

These vinylene/vinyl units were also confirmed by the 13C NMR spectrum obtained using EASC/Ni1 (run 1, Table ), which showed high branches of 83 per 1000 Cs (Supporting Information S4, Figure S9); including methyl (69.8%), ethyl (5.4%), propyl (1.8%), butyl (3.1%), amyl (1.8%), 1,4-paired methyl (4.2%), 1.6-paired methyl (2.7%), and longer-chain branches (11.2%). In addition, much contrast results of the 13C NMR spectrum of the resulting polyethylene using Ni5/EASC were obtained (run 5, Table ), again displaying much lower branches of 16/1000 Cs; mainly comprising methyl (54.6%), butyl (4.2%), amyl (7.1%), 1,4-paired methyl (1.9%), 1,6-paired methyl (3.4%), and longer-chain branches (28.8%) (Figure ). These lower-branching contents suggest that electron-withdrawing fluorine at the ortho-position reduced “chain walking” and preferred β-H elimination.[2d,8a,12,17a,17d,18,19] Furthermore, these findings also suggest that the co-catalysts (MAO and EASC) have a slight impact on the microstructure for the desired unsaturated (vinyl/vinylene double bonds) branched polyethylene.

Figure 12

13C NMR spectrum of the obtained polyethylene sample with Ni5/EASC at 30 °C (run 5, Table ); recorded at 100 °C in d-C2D2Cl4.

Comparison between the Reported and Current Precatalysts

To explore the overall influence of the 2,4-bis(dibenzocycloheptyl) groups along with the variations in the steric and electronic properties of the ortho-substituent, the catalytic performance, molecular weight, and dispersity of the current nickel precatalyst system (G) are compared with those of the previously reported systems D, E, and F under optimized conditions at 10 atm of ethylene (Figure ).[4k,11] In terms of activity, the precatalyst D (13.8 × 106 g PE mol–1(Ni) h–1) was proved to be the most active followed by the current system G. The relative catalytic activities were found to be in the order D > G > E > F. However, the molecular weight of the obtained polyethylene with current system G is the second highest with a narrow dispersity (Mw/Mn = 2.4), which is comparable to those obtained with D and E; and much narrow than that observed with F. It shows that the catalytic performance of the current system bears some resemblance to those of D, E, and F. More significantly, the current catalytic system has the propensity of generating highly branched unsaturated (vinyl/vinylene) polyethylene with a major class of the internal double bond (vinylene).

Figure 13

Comparison of the previously reported systems bearing N-2,4-bis(dibenzhydryl) or N-2,4-bis(fluorenyl) and the current system with N-2,4-bis(dibenzocycloheptyl) groups under optimal conditions at 10 atm of C2H4 (D–G).

Conclusions

In summary, we have successfully synthesized a series of 2-(2,4-dibenzocycloheptyl-6-R-phenylimino)ethyl)pyridine, R = Me (L1), Et (L2), i-Pr (L3), Cl (L4), or F (L5)) and the corresponding nickel(II) bromide complexes (Ni1–Ni5) in good yield. All the organic compounds and the precatalysts were stable in the air and were characterized by FT-IR, 1H, and 13C NMR spectroscopy as well as by elemental analysis. The molecular structures of complexes Ni2 and Ni5 were the subject for the single-crystal X-ray diffraction analysis. The structure of complex Ni2 can be best described as a distorted-square pyramidal and Ni5 with a bromide bridged dimeric form revealed distorted-octahedral geometry. When activated with either MAO or EASC, all the precatalysts displayed high activity in the range of 2.12 to 7.93 × 106 g of PE (mol of Ni)−1 h–1 (MAO) and 2.01 to 6.75 × 106 g of PE (mol of Ni)−1 h–1 (EASC) producing highly branched polyethylene of low-molecular-weights in the range of 1.41–16.55 kg/mol with unsaturated end groups and with the internal double bond as the major class (vinylene/vinyl = 9.8:1 to 1.8:1). The catalytic activity, the type of unsaturation, and branching contents are greatly influenced by the nature of the ortho-substituent and temperature. Significantly, the precatalysts Ni4 and Ni5 (R = Cl or F) displayed lower catalytic activities and the obtained polyethylene showed high melting temperature with the least number of branches as compared to that with Ni–Ni3 (R = Me, Et, or i-Pr); these observations were also consistent with the high-temperature 1H and 13C NMR spectral analyses.

Experimental Section

General Considerations

All the moisture and/or air-sensitive compounds were handled under an inert nitrogen atmosphere using standard Schlenk techniques. All the solvents were heated to reflux prior to use, distilled, and used under the nitrogen atmosphere. The cocatalysts such as MAO (methylaluminoxane, 1.46 M in toluene) and MMAO (modified methylaluminoxane, 1.93 M in heptane) were purchased from Akzo Nobel Corp; whereas Et2AlCl (diethylaluminum chloride, 0.79 M in toluene) and (EASC (Et3Al2Cl3, 0.87 M in toluene)) were purchased from Acros Chemicals. Ethylene of high purity was purchased from Beijing Yanshan Petrochemical Co., used as received, and all other reagents were purchased from Aldrich, Acros, or local suppliers. The corresponding anilines (2,4-bis(dibenzocycloheptyl)-R-phenylamine, R = Me, Et, i-Pr, Cl, or F) were prepared according to the procedure reported in the literature.[3b,4i,12,20] The NMR spectra were recorded on a Bruker DMX (400 MHz instrument) at ambient temperature using TMS as an internal standard. A PerkinElmer System 2000 FT-IR spectrometer was used for the analysis of FT-IR spectra and a Flash EA 1112 microanalyzer was used for elemental analysis. Molecular weight (Mw) and molecular weight distribution (Mw/Mn) of the obtained polyethylene were determined by PL-GPC220 at 150 °C using 1,2,4-trichlorobenzene as a solvent. The melting temperatures were determined by using differential scanning calorimetry (DSC, TA2000) under a nitrogen atmosphere. A typical polyethylene sample in the range of 4.5–5.5 mg was heated up to 130 °C at a heating rate of 20 °C per min for 5 min at the same temperature to remove its thermal history and then cooled to −50 °C at the same heating rate. For recording high-temperature 1H and 13C NMR spectra, a weighed amount of polyethylene (90–100 mg) in 1,1,2,2-tetrachloroethane-d2 (2 mL) was used with TMS as an internal standard. Inverse gated 13C spectra were recorded on a Bruker DMX 300 spectrometer at 75.47 MHz in 5 mm standard glass tubes at 100 °C with the number of scans between 3642 and 4982. The operational conditions: spectral width 1882.4 kHz; acquisition time 0.870–1.8 s; relaxation delay 2.0 s, and with a pulse width of 15.5 μs. The branching contents were calculated from the integration of the corresponding peaks in the 13C NMR spectra according to the literature.[18] The procedure for the synthesis of a series of sterically and electronically modified bulky anilines are given in the Supporting Information.

Synthesis of Ligands (L1–L5)

2-{N-2,4-(C15H13)-6-MeC6H2}C7H7N (L1)

A mixture of 2-actylpyridine (0.25 g, 2.07 mmol) and 2,4-bis(dibenzocycloheptyl)-6-methylaniline (1.01 g, 2.07 mmol) in toluene (50 mL) was refluxed for 20 min using Dean–Stark trap, followed by the slow addition of p-toluenesulfonic acid (cat. 15 mol %) and the reaction mixture was further refluxed for 8 h. Upon cooling to room temperature, all the volatiles were removed under reduced pressure and the residue was purified through alumina (basic) column chromatography using petroleum ether/ethyl acetate (25:2) as an eluent to offer L1 as a yellow powder (0.78 g 63%). mp 130–132 °C. FT-IR (cm–1): 3055 (w), 3014 (w), 2925 (w), 2871 (w), 2831 (w), 1644 (ν(C=N), m), 1588 (w), 1566 (w), 1491 (m), 1465 (m), 1434 (m), 1362 (m), 1302 (w), 1241 (w), 1215 (w), 1163 (w), 1133 (w), 1103 (m), 1044 (w), 1024 (w), 993 (w), 967 (w), 942 (w), 917 (w), 882 (w), 842 (w), 808 (w), 754 (s), 707 (m), 677 (w). 1H NMR (400 MHz, CDCl3, TMS): δ 8.64 (d, J = 4.80 Hz, 1H, Py-H), 8.40 (d, J = 8.00 Hz, 1H, Py-H), 7.83 (t, J = 7.80 Hz, 1H, Py-H), 7.40–7.37 (m, 1H, Py-H), 7.20–6.82 (m, 15H, Ar-H), 6.59 (s, 1H, Ar-Hm), 6.51 (t, J = 7.40 Hz, 1H, Ar-H), 6.42 (s, 1H, Ar-Hm), 5.09 (s, 1H, −CH−), 4.96 (s, 1H, −CH−), 3.12–2.89 (m, 3H, −CH2−), 2.75–2.55 (m, 4H, −CH2−), 2.32–2.26 (m, 1H, −CH2−), 1.79 (s, 3H, −CH3), 1.41 (s, 3H, −CH3). 13C NMR (100 MHz, CDCl3, TMS): δ 168.9, 156.0, 148.4, 145.7, 141.4, 141.2, 140.2, 140.0, 139.7, 139.6, 139.3, 138.6, 138.4, 136.1, 131.5, 131.4, 131.2, 131.1, 131.0, 130.9, 130.5, 130.2, 129.3, 128.6, 127.1, 127.0, 126.8, 126.6, 126.4, 126.0, 125.9, 125.6, 125.4, 124.7, 124.6, 121.4, 57.8, 56.4, 32.6, 32.5, 31.9, 30.4, 17.9, 16.1. Anal. calcd for C44H38N2 (594.80): C, 88.85; H, 6.44; N, 4.71. Found: C, 88.47; H, 6.84; N, 4.59.

2-{N-2,4-(C15H13)-6-EtC6H2}C7H7N (L2)

Using the same procedure and molar ratios as described for the preparation of L1, L2 was prepared as a yellow powder (1.08 g, 86%). mp 182–184 °C. FT-IR (cm–1): 3059 (w), 3014 (w), 2965 (w), 2930 (w), 2894 (w), 2869 (w), 2831 (w), 1634 (ν(C=N), m), 1588 (w), 1566 (w), 1491 (m), 1468 (m), 1447 (m), 1363 (m), 1343 (w), 1300 (w), 1267 (w), 1241 (w), 1213 (w), 1163 (w), 1134 (w), 1103 (m), 1066 (w), 1045 (w), 1025 (w), 996 (w), 970 (w), 941 (w), 896 (w), 875 (w), 844 (w), 813 (w), 762 (s), 680 (w). 1H NMR (400 MHz, CDCl3, TMS): δ 8.64 (d, J = 4.80 Hz, 1H, Py-H), 8.40 (d, J = 8.00 Hz, 1H, Py-H), 7.83 (t, J = 7.80 Hz, 1H, Py-H), 7.39–7.36 (m, 1H, Py-H), 7.26–6.79 (m, 15H, Ar-H), 6.62 (s, 1H, Ar-Hm), 6.50 (t, J = 7.40 Hz, 1H, Ar-H), 6.43 (s, 1H, Ar-Hm), 5.10 (s, 1H, −CH−), 4.95 (s, 1H, −CH−), 3.14–2.90 (m, 3H, −CH2−), 2.75–2.56 (m, 4H, −CH2−), 2.32–2.22 (m, 1H, −CH2−), 2.20–2.02 (m, 2H, −CH2−), 1.55 (s, 3H, −CH3), 0.91 (t, J = 7.60 Hz, 3H, −CH3). 13C NMR (100 MHz, CDCl3, TMS): δ 168.6, 156.0, 148.4, 145.3, 141.4, 141.2, 140.2, 140.0, 139.9, 139.7, 139.3, 138.7, 138.5, 136.1, 131.5, 131.4, 131.1, 131.0, 130.9, 130.5, 130.2, 130.1, 129.3, 128.5, 127.0, 126.8, 126.6, 126.4, 125.9, 125.8, 125.6, 125.4, 125.2, 124.6, 121.3, 57.9, 56.5, 32.6, 32.4, 31.9, 24.2, 16.3, 13.5. Anal. calcd for C45H40N2 (608.83): C, 88.78; H, 6.62; N, 4.60. Found: C, 88.48; H, 6.67; N, 4.61.

2-{N-2,4-(C15H13)-6-i-PrC6H2}C7H7N (L3)

Using the same procedure and molar ratios as described for the preparation of L1, L3 was prepared as a yellow powder (1.00 g 78%). mp 208–210 °C. FT-IR (cm–1): 3059 (w), 3014 (w), 2961 (w), 2930 (w), 2894 (w), 2869 (w), 2828 (w), 1638 (ν(C=N), m), 1588 (w), 1566 (w), 1491 (m), 1467 (m), 1446 (s), 1363 (s), 1338 (w), 1299 (w), 1269 (w), 1241 (w), 1214 (w), 1164 (w), 1127 (w), 1103 (m), 1080 (w), 1044 (w), 1026 (w), 996 (w), 966 (w), 943 (w), 898 (w), 875 (w), 836 (w), 812 (w), 783 (m), 762 (s), 680 (w). 1H NMR (400 MHz, CDCl3, TMS): δ 8.65 (d, J = 4.80 Hz, 1H,, Py-H), 8.40 (d, J = 8.00 Hz, 1H, Py-H), 7.84 (t, J = 7.80 Hz, 1H, Py-H), 7.38 (t, J = 6.20 Hz, 1H, Py-H), 7.21–6.78 (m, 15H, Ar-H), 6.70 (s, 1H, Ar-Hm), 6.50 (t, J = 7.20 Hz, 1H, Ar-H), 6.44 (s, 1H, Ar-Hm), 5.11 (s, 1H, −CH−), 4.94 (s, 1H, −CH−), 3.10–2.88 (m, 3H, −CH2−), 2.76–2.68 (m, 1H, −CH2−), 2.62–2.58 (m, 3H, −CH2−), 2.57–2.47 (m, 1H, −CH2−), 2.30–2.25 (m, 1H–CH−), 1.40 (s, 3H, −CH3), 0.91 (d, J = 6.80 Hz, 3H, −CH3), 0.87 (d, J = 6.80 Hz, 3H, −CH3). 13C NMR (100 MHz, CDCl3, TMS): δ 168.7, 156.0, 148.4, 144.6, 141.5, 141.3, 140.2, 139.9, 139.8, 139.7, 139.3, 138.8, 138.7, 136.1, 135.1, 131.5, 131.4, 131.1, 130.9, 130.8, 130.2, 129.8, 129.3, 128.4, 126.9, 126.8, 126.6, 126.4, 125.9, 125.8, 125.6, 125.4, 124.6, 122.5, 121.4, 58.0, 56.5, 32.6, 32.4, 31.9, 30.4, 27.4, 23.4, 22.4, 16.4. Anal. calcd for C46H42N2 (622.86): C, 88.71; H, 6.80; N, 4.50. Found: C, 88.35; H, 6.85; N, 4.45.

2-{N-2,4-(C15H13)-6-ClC6H2}C7H7N (L4)

Using the same procedure and molar ratios as described for the preparation of L1, L4 was prepared as a yellow powder (0.71 g 56%). mp 195–197 °C. FT-IR (cm–1): 3058 (w), 3015 (w), 2930 (w), 2879 (w), 2829 (w), 1645 (ν(C=N), m), 1588 (w), 1568 (w), 1491 (m), 1467 (m), 1446 (m), 1361 (m), 1303 (m), 1241 (w), 1217 (w), 1163 (w), 1136 (w), 1104 (m), 1046 (w), 1024 (w), 996 (w), 972 (w), 945 (w), 924 (w), 903 (w), 877 (w), 837 (w), 808 (w), 756 (s), 696 (w), 648 (w). 1H NMR (400 MHz, CDCl3, TMS): δ 8.65 (d, J = 4.80 Hz, 1H, Py-H), 8.41 (d, J = 7.60 Hz, 1H, Py-H), 7.85 (t, J = 7.80 Hz, 1H, Py-H), 7.41–7.38 (m, 1H, Py-H), 7.20–6.80 (m, 16H, Ar-H), 6.51 (t, J = 7.20 Hz, 2H, Ar-H), 5.07 (s, 1H, −CH−), 5.01 (s, 1H, −CH−), 3.04–2.90 (m, 3H, −CH2−), 2.78–2.57 (m, 4H, −CH2−), 2.32–2.26 (m, 1H, −CH2−), 1.46 (s, 3H, −CH3). 13C NMR (100 MHz, CDCl3, TMS): δ 171.3, 155.6, 148.5, 143.7, 140.7, 140.6, 140.1, 139.9, 139.5, 139.2, 138.8, 137.6, 136.2, 133.3, 131.7, 131.5, 131.4, 131.2, 131.1, 130.9, 130.5, 130.3, 129.3, 129.1, 127.3, 127.2, 126.9, 126.7, 126.2, 126.1, 126.0, 125.8, 125.5, 125.9, 122.5, 121.7, 57.5, 56.5, 32.6, 32.5, 31.9, 30.3, 16.7. Anal. calcd for C43H35ClN2 (615.22): C, 83.95; H, 5.73; N, 4.55. Found: C, 83.61; H, 5.80; N, 4.52.

2-{N-2,4-(C15H13)-6-FC6H2}C7H7N (L5)

Using the same procedure and molar ratios as described for the preparation of L1, L5 was isolated as a yellow powder (0.86 g, 69%). mp 202–204 °C. FT-IR (cm–1): 3058 (w), 3014 (w), 2966 (w), 2930 (w), 2888 (w), 2868 (w), 2829 (w), 1643 (ν(C=N), m), 1588 (w), 1567 (w), 1491 (m), 1470 (m), 1447 (s), 1361 (m), 1302 (m), 1241 (w), 1216 (w), 1163 (w), 1136 (w), 1104 (s), 1046 (w), 1026 (w), 995 (w), 970 (w), 945 (w), 906 (w), 876 (w), 838 (w), 808 (w), 756 (s), 702 (w), 679 (w), 650 (w). 1H NMR (400 MHz, CDCl3, TMS): δ 8.65 (d, J = 4.50 Hz, 1H, Py-H), 8.38 (d, J = 8.00 Hz, 1H, Py-H), 7.84 (t, J = 7.70 Hz, 1H, Py-H), 7.38 (t, J = 6.10 Hz, 1H, Py-H), 7.22–6.86 (m, 16H, Ar-H), 6.70 (d, J = 11.00 Hz, 1H, Ar-H), 6.35 (s, 1H, Py-H), 5.12 (s, 1H, −CH−), 5.08 (s, 1H, −CH−), 3.01–2.95 (m, 3H, −CH2−), 2.69–2.33 (m, 5H, −CH2−), 1.58 (s, 3H, −CH3). 13C NMR (100 MHz, CDCl3, TMS): δ 171.0, 155.7, 151.7, 149.3, 148.5, 140.5, 140.4, 139.8, 136.1, 134.9, 134.3, 134.2, 131.4, 127.2, 126.7, 126.1, 125.7, 124.8, 121.7, 112.5, 112.3, 57.6, 56.2, 56.1, 32.3, 16.5, 16.4. Anal. calcd for C43H35FN2 (598.77): C, 86.26; H, 5.89; N, 4.68. Found: C, 85.87; H, 5.88; N, 4.66.

Synthesis of Nickel Complexes (Ni1–Ni5)

[2-{N-2,4-(C15H13)-6-MeC6H2}C7H7N]NiBr2 (Ni1)

Under the nitrogen atmosphere, 2-(2,4-bis(dibenzocycloheptyl)-6-methylphenyl)iminoethyl)pyridine (0.31 g, 0.52 mmol) and NiBr2(DME) (0.08 g, 0.26 mmol) were loaded into dichloromethane (10 mL) and ethanol (10 mL) and stirred for 14 h at room temperature. After which, all the volatiles were evaporated under reduced pressure, followed by the addition of diethyl ether into the residue to induce precipitation. The precipitate was filtered, washed with an excess of diethyl ether (3 × 10 mL), and dried to afford Ni1 as a green powder (0.40 g, 95%). FT-IR (cm–1): 3058 (w), 3016 (w), 2931 (w), 2884 (w), 2832 (w), 1624 (ν(C=N), m), 1595 (m), 1572 (w), 1492 (m), 1438 (m), 1370 (m), 1319 (m), 1259 (w), 1210 (w), 1162 (w), 1102 (w), 1050 (w), 1023 (w), 986 (w), 945 (w), 880 (w), 837 (w), 765 (s), 704 (m), 676 (m). Anal. calcd for C44H38Br2N2Ni·2H2O (849.33): C, 62.22; H, 4.98; N, 3.30. Found: C, 62.18; H, 4.86; N, 3.25.

[2-{N-2,4-(C15H13)-6-EtC6H2}C7H7N]NiBr2 (Ni2)

Using a similar procedure and molar ratios to those described for the synthesis of Ni1, Ni2 was isolated as a green powder (0.41 g, 95%). FT-IR (cm–1): 3059 (w), 3016 (w), 2970 (w), 2933 (w), 2877 (w), 2833 (w), 1623 (ν(C=N), m), 1595 (m), 1571 (w), 1492 (m), 1447 (m), 1370 (m), 1317 (m), 1259 (w), 1206 (w), 1161 (w), 1102 (w), 1052 (w), 1022 (w), 982 (w), 946 (w), 880 (w), 877 (w), 833 (w), 764 (s), 704 (m), 676 (m). Anal. calcd for C45H40Br2N2Ni·2H2O (863.36): C, 62.60; H, 5.14; N, 3.24. Found: C, 62.86; H, 4.95; N, 3.22.

[2-{N-2,4-(C15H13)-6-i-PrC6H2}C7H7N]NiBr2 (Ni3)

Using a similar procedure and molar ratios to those described for the synthesis of Ni1, Ni3 was isolated as a green powder (0.36 g, 82%). FT-IR (cm–1): 3058 (w), 3014 (w), 2965 (w), 2932 (w), 2875 (w), 2830 (w), 1621 (ν(C=N), w), 1595 (m), 1571 (w), 1493 (m), 1445 (m), 1371 (m), 1317 (m), 1259 (w), 1204 (w), 1163 (w), 1136 (w), 1103 (w), 1051 (w), 1025 (w), 982 (w), 947 (w), 877 (w), 830 (w), 766 (s), 704 (m), 678 (m). Anal. calcd for C46H42Br2N2Ni·H2O (859.37): C, 64.29; H, 5.16; N, 3.26. Found: C, 64.30; H, 5.12; N, 3.25.

[2-{N-2,4-(C15H13)-6-ClC6H2}C7H7N]NiBr2 (Ni4)

Using a similar procedure and molar ratios to those described for the synthesis of Ni1, Ni4 was isolated as a green powder (0.40 g, 93%). FT-IR (cm–1): 3059 (w), 3013 (w), 2933 (w), 2878 (w), 2831 (w), 1627 (ν(C=N), m), 1595 (m), 1595 (w), 1571 (w), 1492 (m), 1447 (m), 1405 (w), 1369 (m), 1318 (m), 1259 (w), 1220 (w), 1162 (w), 1132 (w), 1101 (w), 1051 (w), 1022 (w), 983 (w), 948 (w), 878 (w), 827 (w), 765 (s), 695 (m), 673 (m), 652 (m). Anal. calcd for C43H35Br2ClN2Ni·H2O (851.73): C, 60.64; H, 4.38; N, 3.29. Found: C, 60.67; H, 4.30; N, 3.22.

[2-{N-2,4-(C15H13)-6-FC6H2}C7H7N]NiBr2 (Ni5)

Using a similar procedure and molar ratios to those described for the synthesis of Ni1, Ni5 was isolated as a green powder (0.39 g, 95%). FT-IR (cm–1): 3060 (w), 3014 (w), 2933 (w), 2876 (w), 2832 (w), 1627 (ν(C=N), m), 1597 (m), 1570 (w), 1490 (m), 1448 (w), 1423 (m), 1370 (m), 1320 (w), 1282 (w), 1260 (w), 1204 (w), 11621 (w), 1133 (w), 1104 (w), 1050 (w), 1023 (w), 995 (w), 945 (w), 915 (w), 875 (w), 830 (w), 766 (s), 738 (s), 704 (m), 675 (m). Anal. calcd for C43H35Br2FN2Ni·H2O (835.28): C, 61.83; H, 4.47; N, 3.35. Found: C, 61.71; H, 4.41; N, 3.34.

Typical Procedure for Ethylene Polymerization

High pressure (10 atm) ethylene polymerization was conducted in a 250 mL stainless steel autoclave, equipped with pressure and temperature control systems and a mechanical stirrer. The autoclave was evacuated and back-filled three times with nitrogen and once with ethylene. The precatalyst (2.0 μmol) was dissolved in toluene (25 mL) and when the required temperature was reached, the complex solution was injected into the autoclave containing ethylene (ca. 1 atm) followed by the addition of more toluene (25 mL). The required amounts of co-catalyst (Me2AlCl, Et2AlCl, EASC, MAO, or MMAO), followed by the addition of toluene (50 mL), were successively added using a syringe making the total volume to 100 mL. The autoclave was immediately pressurized to the predetermined ethylene pressure and stirred (400 rpm). After reaching the required reaction time, the reactor was cooled in a water bath and stopped the flow of ethylene and then slowly released the ethylene pressure. The reaction was quenched with 10% hydrochloric acid solution in ethanol. The polymer was collected, filtered, washed with ethanol, and dried under reduced pressure at 40 °C and then weighed. Using a similar procedure to that described for 10 atm of ethylene, 5 atm ethylene polymerization was also conducted. While the Schlenk tube was used instead of an autoclave for 1 atm ethylene polymerization.

X-ray Crystallographic Studies

X-ray diffraction studies were conducted for the suitable single-crystal of Ni2 and Ni5. Crystals were grown by the layering of heptane into the solution of the corresponding complex in dichloromethane at ambient temperature. The X-ray structural determination was carried out on a Rigaku Saturn 724+ CCD diffractometer provided with graphite monochromatic Mo Kα radiation (λ = 0.71073 Å) at 173(2) K. The cell parameters were obtained by a global refinement of the positions of all the collected reflections. The intensities for Lorentz and polarization effects were corrected out and an empirical absorption was applied. The structures were resolved by the direct methods and refined by full-matrix least-squares on F2. All hydrogen atoms were placed in the calculated positions. Structure solution and refinement were conducted by using SHELXT-97.[21] The free solvent molecules were squeezed with PLATON software.[22] The details of the X-ray structure determination data and structural refinements for Ni2 and Ni5 are provided in Table .

Table 7

Crystal Data and Structure Refinement for Ni2(a) and 2Ni5·HO

	Ni2(a)	2Ni5·H2O
CCDC no.	1976741	1976742
empirical formula	C90H80Br2N4Ni	C43H37Br2FN2NiO
formula weight	1436.11	835.27
temperature/K	170.01(10)	170.01(10)
wavelength/Å	0.71073	0.71073
crystal system	monoclinic	monoclinic
space group	P21/n	P21/c
a/Å	11.4100(4)	15.6936(4)
b/Å	38.9819(12)	18.8357(5)
c/Å	18.2057(7)	14.8346(7)
α/deg	90	90
β/deg	97.329(3)	92.195(3)
γ/deg	90	90
volume/Å3	8031.4(5)	4381.9(3)
Z	4	4
Dcalcd (g/cm3)	1.188	1.266
μ/mm–1	1.282	2.301
F(000)	2984.0	1696.0
crystal size/mm3	0.194 × 0.101 × 0.085	0.232 × 0.168 × 0.082
radiation	Mo Kα (λ = 0.71073)	Mo Kα (λ = 0.71073)
2θ range for data collection/deg	6.832–54.998	6.99–61.25
index ranges	–14 ≤ h ≤ 14, –48 ≤ k ≤ 50, –21 ≤ l ≤ 23	–22 ≤ h ≤ 22, –23 ≤ k ≤ 26, –21 ≤ l ≤ 19
reflections collected	77553	46334
independent reflections	18415 [Rint = 0.0807, Rsigma = 0.0915]	12143 [Rint = 0.0562, Rsigma = 0.0687]
data/restraints/parameters	18415/0/878	12143/0/453
completeness to θ	0.998	0.898
goodness-of-fit on F2	1.063	1.040
final R indexes [I ≥ 2σ(I)]	R1 = 0.0859, wR2 = 0.1759	R1 = 0.0615, wR2 = 0.1333
final R indexes [all data]	R1 = 0.1336, wR2 = 0.1933	R1 = 0.1089, wR2 = 0.1496
largest diff. peak/hole/e Å–3	1.31/–0.95	0.90/–1.11