Fluoro- & trifluoromethoxy-modified bis(imino)pyridyliron chloride enhancing the activity and stability of vinyl polyethylenes

Abstract

The bis(imino)pyridylferrous chloride complexes bearing para-trifluoromethoxy and ortho-difluorobenzhydryl substituents, namely, 2-(1-{2,6-bis[bis(4-fluorophenyl)methyl]-4-trifluoromethoxyphenyl-imino}ethyl)-6-[1-(2,6-di(R¹)-4-R²-phenylimino) ethyl]pyridyl iron(II) chloride (where R¹ = Me, R² = H, Fe1; R¹ = Et, R² = H, Fe2; R¹ = i-Pr, R² = H, Fe3; R¹ = Me, R² = Me, Fe4; R¹ = Et, R² = Me, Fe5; and R¹ = (4-FC₆H₄)₂CH, R² = OCF₃, Fe6) were successfully synthesized and thoroughly characterized. The X-ray crystal structures of Fe4 and Fe6 were also determined. The catalytic performance of all the new iron complexes was systematically evaluated for ethylene polymerization, with a detailed comparative analysis conducted against previously published studies. Notably, iron complexes Fe1–Fe5 demonstrated exceptional catalytic activity and thermal stability upon activation with MAO, while Fe6 showed inferior performance due to its significant steric hindrance. Among them, Fe1 demonstrated the highest activity, achieving a peak of 17.28 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ at 70 °C in n-hexane, which markedly surpassed the activity levels of other structurally related iron complexes, indicating the beneficial impact of incorporating multi-fluoro groups. Remarkably, the current iron pre-catalysts generally revealed superior performance in hexane compared to toluene, even when using a reduced quantity of co-catalyst, thereby underscoring the pivotal role of solvent polarity in governing the polymerization process. Furthermore, the analysis of the generated polyethylenes (PEs) revealed a high degree of linearity in the polymers, featuring vinyl end groups and exhibiting a broad spectrum of molecular weights (M_w: 4.5–567.9 kg mol⁻¹).

---

Introduction

Polyethylene plays a central role in daily life and industrial applications due to its widespread use in the production of span films, insulated cables, piping systems, and engineering plastics.¹ Among the myriad carbon–carbon bond-forming reactions,² transition metal-catalyzed ethylene polymerization of ethylene holds a pivotal position in the chemical industry.³ The advent of late transition metal catalysts in the mid-1990s marked a transformative breakthrough in polyolefin catalytic technology, building on the foundational developments with Ziegler–Natta catalysts in the 1950s and metallocene catalysts in the 1980s. In particular, tridentate bis(imino)pyridyl iron complexes (A, Chart 1) have emerged as highly efficient pre-catalysts, exhibiting remarkable performance in producing highly linear α-olefins/polyethylenes,⁴ symbolizing a significant achievement in the field of late transition metal catalysts for polyolefins. Subsequently, extensive research efforts have been directed toward modifying the bis(imino)pyridine backbone and N-aryl groups through the incorporation of diverse electronic and steric features, aiming to significantly enhance the catalytic proficiency of these systems.⁵

Chart 1 The classic bis(imino)pyridine-iron chloride A, along with its benzhydryl-substituted iron derivatives, B–D.

Previous studies have demonstrated that the presence of bulky substituents at the ortho-positions of the N-aryl groups significantly enhances the catalytic performance and thermal stability of the resulting complexes. In other words, the presence of bulky groups such as benzhydryl (CHPh₂) in iron complexes (B, Chart 1) not only remarkably generates higher molecular weight PEs by reducing the rate of chain transfer but also increases thermal stability and catalytic performance by shielding the apical positions of the metal centers.⁶ Meanwhile, the electronic properties of para-substituents in the N-aryl groups can also influence the catalytic performance of catalysts by modulating the ligand structure and electron density around the central metal of the complex. Specifically, compared with complexes containing electron-donating groups, generally both the catalytic activity and the molecular weight of polyethylene produced by iron complexes with electron-withdrawing groups at the para-position of the N-aryl groups can be enhanced.⁶ Nevertheless, excessive steric hindrance from the vertical side could significantly hinder ethylene coordination and insertion reactions, leading to drastically reduced activity.

Given the significant influence of the electronic effects of the ligand skeleton on the catalytic behavior of iron complexes, various remote substituents with distinct electronic properties (specifically, F, Cl, and OCH₃) have been introduced into the diphenylmethyl moiety to investigate how electronic modifications can modulate and enhance catalytic performance.⁷ When the diphenylmethyl group was modified using electron-withdrawing para-fluoro or chloro substituents, the resulting iron complexes C (R = F and Cl, Chart 1) demonstrated enhanced catalytic activity, but led to the production of polyethylenes with lower molecular weights.^7a,b Additionally, the thermal stability of another di(arylimine)pyridine iron(II) complex that incorporated a remote fluorine group within the N-2,6-di{di(4-fluorophenyl)methyl}-4-nitrophenyl group was significantly improved,^7d compared to its fluorine-free iron analogues.^6e For cobalt complexes, which typically exhibit lower performance in ethylene polymerization, significant improvements in both polymerization activity and thermal stability can be achieved by structural refinement through the incorporation of remote electron-withdrawing substituents, such as para-fluoride (F) or chloride (Cl) substituents.⁵ Conversely, the introduction of an electron-donating para-methoxy group at the benzhydryl substituent into the iron complexes C (R = OCH₃, Chart 1) resulted in a notable increase in the polyethylene's molecular weight and a broadening of its molecular weight distribution, albeit at the expense of significantly diminished catalytic activity.^7c In general, the electronic properties of the remote substituents on the benzhydryl groups significantly influenced the catalytic efficiency and thermal stability of bis(imino)pyridine-iron complexes, as well as the properties of the resulting polyethylenes, regardless of their distance from the iron center.

Inspired by a recently published series of ligands,⁸ we synthesized a series of unsymmetrical bis(imino)pyridine-iron pre-catalysts (D, Chart 1) featuring one N-aryl group substituted with a para-trifluoromethoxy group (OCF₃) and two ortho-difluorobenzhydryl groups to gain a deeper understanding of the effects of electron-withdrawing groups on iron-catalyzed ethylene polymerization. This design allowed us to further investigate the beneficial role of remote fluoro groups and to assess the combined effects of the strong σ-inductive effect and minimal π-donating influence of the OCF₃ group.⁹ Hexane, a widely used industrial solvent, was chosen as the polymerization medium to investigate its influence on the formation of active species and the polymerization process, with comparisons drawn to toluene.^6b In order to investigate their catalytic behaviour thoroughly, polymerization reactions were conducted under various conditions, including ethylene pressure, polymerization time, polymerization temperature, co-catalyst type, and co-catalyst dosage, aimed at identifying optimal polymerization conditions and exploring how these factors influence the catalytic performance and properties of the resulting polyethylenes.

Results and discussion

Synthesis and identification of the ligands and iron complexes

A distinct set of unsymmetrically substituted ligands, 2-[CMeN{2,6-{(4-FC₆H₄)₂CH}₂-4-OCF₃C₆H₂}]-6-(CMeNAr)C₅H₃N (Ar = 2,6-Me₂C₆H₃, L1; 2,6-Et₂C₆H₃, L2; 2,6-i-Pr₂C₆H₃, L3; 2,4,6-Me₃C₆H₂, L4; 2,6-Et₂-4-MeC₆H₂, L5; 2,6-[(4-F-C₆H₄)₂]-4-OCF₃-C₆H₂, L6), has been synthesized through two consecutive Schiff base condensation reactions in our previous work (Scheme 1).⁸ Subsequently, the iron complexes, [2-[CMeN{2,6-{(FC₆H₄)₂CH}₂-4-OCF₃C₆H₂}]-6-(CMeNAr)C₅H₃N]FeCl₂ (Ar = 2,6-Me₂C₆H₃, Fe1; 2,6-Et₂C₆H₃, Fe2; 2,6-i-Pr₂C₆H₃, Fe3; 2,4,6-Me₃C₆H₂, Fe4; 2,6-Et₂-4-MeC₆H₂, Fe5; 2,6-[(4-F-C₆H₄)₂]-4-OCF₃-C₆H₂ Fe6), were prepared by the stoichiometric reactions of iron dichloride with the respective ligands (L1–L6) in a blend of ethanol and dichloromethane under an inert nitrogen atmosphere at ambient temperature, achieving high yields ranging from 73% to 84% (Scheme 1). All newly synthesized iron complexes were identified by FT-IR, ¹⁹F NMR spectroscopy and elemental analysis. Notably, the molecular structures of Fe4 and Fe6 were verified by single-crystal X-ray diffraction.

Scheme 1 The synthesis pathway for six iron(II) chloride complexes, Fe1–Fe6.

When comparing the ¹⁹F NMR spectra of the ligands with those of their corresponding iron complexes, it was noted that the fluorine signals of the free ligands underwent a downfield shift, mirroring the trend observed in cobalt analogues.⁸ Specifically, the CH(4-FC₆H₄)_a(4-FC₆H₄)_b groups, which resided in distinct chemical environments, exhibited two non-equivalent fluorine groups within the range of −114.21 to −118.12 ppm. Conversely, the OCF₃ functional group comprised equivalent fluorine atoms, leading to a single peak within the range of −57.69 to −59.63 ppm (Fig. S1–S6†). Additionally, the IR spectra of the iron complexes exhibited a noticeable shift in the C=N stretching vibrations towards lower wavenumbers (1580–1605 cm⁻¹) compared to those observed in the free ligands (1600–1646 cm⁻¹), signifying the existence of a robust binding interaction between the imine-nitrogen atoms and the iron center.¹⁰

Two single crystals of iron complexes (Fe4 and Fe6) were obtained by the slow diffusion of diethyl ether into their respective dichloromethane solutions under a nitrogen-protected atmosphere. The molecular structures of Fe4 and Fe6 are shown in Fig. 1 and 2, respectively, while the selected bond lengths and angles are shown in Table 1. Both Fe4 and Fe6 exhibit similar coordination structures that can be described as a distorted tetrahedral geometry, featuring a five-coordinate backbone where the iron center is coordinated by two chloride atoms and three nitrogen atoms from the 2,6-(bisarylimino)pyridine ligand.

Fig. 1 Molecular structure of Fe4 with the thermal ellipsoids shown at the 30% probability level; one molecule of dichloromethane and hydrogen atoms have been omitted for clarity.

Fig. 2 Molecular structure of Fe6 with the thermal ellipsoids shown at the 30% probability level; hydrogen atoms have been omitted for clarity.

Table 1 Selected bond lengths (Å) and angles (°) for Fe4 and Fe6

	Fe4	Fe6
Bond lengths (Å)
Fe(1)–N(1)	2.260(7)	2.2557(19)
Fe(1)–N(2)	2.114(7)	2.0748(19)
Fe(1)–N(3)	2.323(6)	2.2364(19)
Fe(1)–Cl(1)	2.289(3)	2.2380(7)
Fe(1)–Cl(2)	2.279(3)	2.3045(8)
N(1)–C(2)	1.282(12)	1.289(3)
N(1)–C(10)	1.431(10)	1.442(3)
N(2)–C(3)	1.362(10)	1.345(3)
N(2)–C(7)	1.337(10)	1.346(3)
N(3)–C(8)	1.296(10)	1.295(3)
N(3)–C(19)/C(43)	1.445(9)	1.436(3)
O(1)–C(38)/O(2)–C(76)	1.337(10)	1.324(4)
Bond angles (°)
Cl(1)–Fe(1)–Cl(2)	108.52(12)	113.35(3)
N(1)–Fe(1)–N(2)	74.1(2)	73.19(7)
N(1)–Fe(1)–N(3)	146.1(2)	140.19(7)
N(1)–Fe(1)–Cl(1)	96.8(2)	99.45(5)
N(1)–Fe(1)–Cl(2)	105.0(2)	103.31(5)
N(2)–Fe(1)–N(3)	72.3(2)	73.35(7)
N(2)–Fe(1)–Cl(1)	118.4(2)	151.38(6)
N(2)–Fe(1)–Cl(2)	133.0(2)	95.26(6)
N(3)–Fe(1)–Cl(1)	102.76(18)	99.63(5)
N(3)–Fe(1)–Cl(2)	94.71(18)	100.62(5)

---

In the asymmetric structure of Fe4, the dihedral angles associated with the rings bonded to N(1) and N(3) are comparable (79.93° and 88.43°), indicating that the phenyl rings appended to the imine groups are nearly perpendicular to the coordination plane (Fig. 1). Regarding Fe6, it exhibits a configuration that closely approximates C_s symmetry, yet it is characterized by differing dihedral angles: specifically, 75.94° for the phenyl ring attached to N(1) and 83.84° for the phenyl ring attached to N(3). These discrepancies underscore the significant steric hindrance imparted by the benzhydryl substituents positioned on the N1-aryl group (Fig. 2). The iron atom exhibits distances of 0.125 Å and 0.570 Å to the plane formed by the coordinated nitrogen atoms (N1, N2, and N3) in Fe4 and Fe6, respectively. Moreover, the central Fe–N2(Py) bond in Fe4, measuring 2.114(7) Å, is marginally elongated compared to the Fe–N2(Py) bond in Fe6, which spans 2.0748(19) Å. However, in both cases, the Fe–N2(Py) bonds are notably shorter than the Fe–N3(imine) bonds, which are 2.323(6) Å and 2.2364(19) Å, respectively, illustrating the robust coordination bond between the N_pyridine atom and the central iron metal, as well as the pronounced axial influence exerted by large hindrance groups on the coordination center.^6e–f,7d The bite angles N1–Fe–N2 and N2–Fe–N3 in Fe6 are virtually identical, measuring 73.19(7)° and 73.35(7)° respectively. In contrast, the bite angles in Fe4 exhibit a notable discrepancy, with values of 74.1(2)° and 72.3(2)°. This indicates that Fe6 possesses a more symmetrical structure, whereas Fe4 exhibits an unequal chemical environment around its iron center, attributed to the presence of two different N-aryl groups. However, the ortho-di(4-fluorophenyl)methyl substituents present in both N-aryl groups effectively obstruct the axial sites of the iron center in the symmetrical Fe6 complex, thereby impeding the process of ethylene polymerization. The distance between the carbon atom and its neighboring oxygen atom in OCF₃, which measures 1.337(10) Å in Fe4 and 1.324(4) Å in Fe6, reveals characteristics indicative of a double bond. This observation emphasizes the significant contribution of the ionic limiting resonance form ArO⁺=CF₂F⁻, akin to our previously reported findings.^6c,8

Catalytic potential toward ethylene polymerization

To uncover the catalytic potential of the synthesized iron pre-catalysts Fe1–Fe6, polymerization experiments were conducted utilizing two distinct co-catalysts: methyl aluminoxane (MAO) and its modified counterpart, MMAO. Subsequently, the influence of the solvent was explored using the more potent co-catalyst. Prior to this, a systematic optimization process was undertaken, encompassing key polymerization parameters such as temperature, Al : Fe molar ratio, duration, and ethylene feed pressure. Upon establishing the optimal conditions, the remaining pre-catalysts (Fe2–Fe6) were evaluated for their efficacy in ethylene polymerization. The physical attributes of the resultant polyethylene (PE), including molecular weight (M_w) and melting point (T_m), were ascertained through gel permeation chromatography (GPC) and differential scanning calorimetry (DSC), respectively. Additionally, high-temperature ¹H/¹³C NMR spectroscopy was employed to elucidate the microstructural characteristics of the obtained PE samples.

Catalytic evaluation of Fe1–Fe6/MAO in toluene

To refine the polymerization conditions, Fe1 was selected as the test pre-catalyst and paired with MAO as the co-catalyst, and the polymerization results are detailed in Table 2. Initially, the influence of temperature on the catalytic performance was examined by varying the temperature between 40 and 100 °C with the Al : Fe molar ratio fixed at 2000 and a reaction run time of 30 min (runs 1–7, Table 2). These temperature adjustments revealed a significant impact on the catalytic properties and the resultant polyethylene (PE) properties. It was observed that Fe1/MAO exhibited lower catalytic activity at 40–50 °C but produced higher molecular weight PEs (198.7 to 567.9 kg mol⁻¹) (runs 1–2, Table 2), indicating a greater probability of chain propagation within this temperature range. As the temperature increased from 60 to 80 °C, the catalytic performance gradually improved, peaking at 14.23 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ at 80 °C (runs 3–5, Table 2). When compared with the structurally related N-2,6-dibenzhydryl-4-trifluoromethoxyphenyl-containing B_OCF3 (Chart 1),^6b it is evident that the presence of the difluorobenzhydryl groups leads to an improvement in both the catalytic activity and thermal stability (optimal temp. 60 °C, 5.81 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ for B_OCF3 vs. optimal temp. 80 °C, 14.23 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ for Fe1).

Table 2 Ethylene polymerization by Fe1−Fe6/MAO using toluene as the polymerization solvent^a

Run	Pre-cat.	T (°C)	t (min)	Al : Fe	Act.^b	Mw ^c	Mw/Mn ^c	Tm ^d (°C)
a Conditions: 2.0 μmol of Fe pre-catalyst, 100 mL toluene, 10 atm C2H4.b 10^6 g (PE) mol^−1 (Fe) h^−1.c Mw (kg mol^−1), Mw and Mw/Mn determined by GPC.d Determined by DSC.e PC2H4 = 5 atm.f PC2H4 = 1 atm.
1	Fe1	40	30	2000	4.54	567.9	77.24	133.7
2	Fe1	50	30	2000	5.10	198.7	28.4	131.8
3	Fe1	60	30	2000	6.62	64.6	9.8	131.2
4	Fe1	70	30	2000	10.95	59.0	5.4	131.3
5	Fe1	80	30	2000	14.23	35.7	4.5	129.0
6	Fe1	90	30	2000	10.68	13.5	3.1	130.0
7	Fe1	100	30	2000	2.50	4.5	3.0	124.3
8	Fe1	80	30	1000	0.28	195.5	7.4	133.8
9	Fe1	80	30	1500	6.98	40.7	4.2	131.7
10	Fe1	80	30	2500	9.95	37.0	4.3	130.8
11	Fe1	80	30	2750	8.09	16.2	3.5	129.7
12	Fe1	80	30	3000	6.65	10.9	3.5	128.1
13	Fe1	80	05	2000	39.51	4.1	2.6	125.5
14	Fe1	80	15	2000	15.14	14.7	4.0	128.7
15	Fe1	80	45	2000	9.83	41.3	3.7	130.4
16	Fe1	80	60	2000	7.66	47.3	7.8	128.8
17^e	Fe1	80	30	2000	6.12	9.4	4.5	126.1
18^f	Fe1	80	30	2000	1.50	2.9	3.5	122.8
19	Fe2	80	30	2000	15.11	12.7	3.2	128.5
20	Fe3	80	30	2000	8.71	15.2	5.4	128.2
21	Fe4	80	30	2000	14.5	10.9	4.1	127.3
22	Fe5	80	30	2000	15.71	11.4	2.6	128.8
23	Fe6	80	30	2000	0.58	13.0	4.6	128.2

---

Above 80 °C, the catalytic performance declined slightly to 10.68 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ at 90 °C (run 6, Table 2), potentially due to reduced monomer solubility or partial deactivation of the active species.^11,17 Despite the slight decrease at higher temperatures (above 90 °C), the catalytic performance remained robust at 2.5 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹, outperforming related pre-catalysts,^6e,7d underscoring the high thermal stability of the current system. In contrast to the catalytic performance, the molecular weights of the produced PEs consistently decreased with the increase of reaction temperature (M_w: 567.9 to 4.5 kg mol⁻¹), likely due to a higher propensity for chain transfer compared to chain propagation.^6,11,12 Fig. S7† shows that the PEs exhibited broad molecular weight distributions at lower temperatures (40–70 °C), shifting to narrower distributions at higher temperatures (80–100 °C), indicating a transition towards single-site active species. Furthermore, the high melting temperatures, ranging from 125.5 to 133.8 °C, affirm the linear nature of the resulting PEs, which will undergo further analysis via high-temperature ¹H/¹³C NMR in the sections that follow.

Secondly, the impact of varying the Al : Fe molar ratio within the range of 100 to 3000 was investigated while maintaining a constant temperature of 80 °C and a reaction time of 30 minutes (runs 5, 8–12, Table 2). Notably, at an Al : Fe molar ratio of 1000, the molecular weight of the resulting PEs reached a peak of 195.5 kg mol⁻¹, subsequently decreasing to 10.9 kg mol⁻¹ as the ratio was increased to 3000. This observation could likely be attributed to the elevated concentrations of co-catalyst, which facilitated chain transfer from the iron active species to the aluminum center.¹³ The optimal catalytic performance, achieving a rate of 14.23 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹, was achieved at a 2000 Al : Fe molar ratio (run 5, Table 2). Consistent with previous observations, all polymers demonstrated a wide range of molecular weight distributions, exhibiting a general tendency towards narrowing as the quantity of co-catalyst utilized increased (Fig. S8†).

Third, to assess the longevity of Fe1/MAO, polymerization reactions were conducted over varying durations of 5, 15, 30, 45 and 60 minutes, with a reaction temperature of 80 °C and an Al : Fe molar ratio of 2000 (runs 5, 13–16, Table 2). Analysis of the results revealed a peak in catalytic performance at 5 min, achieving an activity level of 39.5 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹. This surge in performance can be attributed to the rapid generation of numerous active species following the injection of MAO into the reaction mixture. Subsequently, the catalytic performance decreased steadily due to the partial deactivation of these active sites.¹⁴ Despite this decline, a considerable performance of 7.66 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ was maintained after 1 h of reaction, highlighting the impressive longevity of this pre-catalyst. With regard to the molecular weight of the resulting PEs, a gradual increase from 4.1 to 47.3 kg mol⁻¹ was observed as the reaction time prolonged. This trend can be attributed to the persistence of sufficient active species over extended reaction periods (Fig. S9†).¹⁵ The influence of ethylene pressure was also investigated under the optimized reaction conditions. As the ethylene pressure was reduced from 10 atm (run 5, Table 2) to 5 atm (run 17, Table 2) and then to 1 atm (run 18, Table 2), the catalytic performance decreased accordingly, from 14.23 to 6.12 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ and then to 1.5 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹. Similarly, the molecular weights of PEs decreased from 35.7 kg mol⁻¹ to 9.4 kg mol⁻¹ and then to 2.9 kg mol⁻¹, indicating a close correlation between the coordination and insertion rate of ethylene and the concentration of ethylene supplied.⁷

Finally, to explore the electronic and steric effects on catalytic performance, the remaining pre-catalysts (Fe2–Fe6) were evaluated under the optimized conditions [Al : Fe = 2000, 80 °C, 30 minutes, 10 atm] (runs 19–23, Table 2). All iron pre-catalysts except Fe6, which possessed the largest steric hindrance, exhibited high catalytic performance ranging from 8.71 to 15.71 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹, and the activity trend was listed as follows: Fe5 > Fe2 > Fe4 > Fe1 > Fe3 ≫ Fe6. In comparison with previously reported results obtained with the iron precatalyst B_OCF3 (Chart 1),^6b the current iron systems gave improved activity even at higher polymerization temperatures [(3.61–9.66) × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ for B_OCF3 at 60 °C vs. (8.71–15.71) × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ for D at 80 °C], highlighting the better thermal stability and catalytic efficacy of the current catalysts containing remote fluoro-substituents. The presence of substantial steric hindrance at the ortho-position in Fe3 and Fe6 could impede the seamless incorporation of the ethylene monomer and retard the transfer of iron active species along the chain, ultimately resulting in diminished catalytic efficiency and the generation of polyethylenes with increased molecular weights.^6a,d,e,16 Moreover, Fe4 and Fe5 bearing electron-donating methyl groups at the para-position of the N-aryl ring exhibited superior activities compared to their counterparts Fe1 and Fe2, respectively, consistent with previous observations for iron pre-catalysts.^6d,e In general, much lower molecular weight PEs were obtained by Fe1–Fe6 (M_w: 10.9–35.7 kg mol⁻¹) (Fig. S10†) in comparison with those obtained from structurally related iron pre-catalysts (B and C, Chart 1), indicating that the introduction of remote electron-withdrawing substituents had a significant impact on chain transfer from the active centers.^6,7 Notably, all the resulting PEs exhibited unimodal molecular weight distributions, coupled with high melting points, indicating their linear nature as well as the uniform active species during the polymerization process. Within the current reaction setup, the deactivation of the active species tended to occur as the temperature rose or the reaction proceeded for a longer period. The root causes are that the iron active species undergoes easier redox deactivation upon binding with impurities (e.g., aluminum alkyls) and the inactive species may form due to structural changes (e.g., ligand dissociation and dimerization) and β-H elimination, which effectively removes it from the polymerization cycle.

Catalytic evaluation of Fe1–Fe6/MAO in hexane

To investigate the influence of the polymerization solvent on catalytic performance, experiments were conducted using hexane as the solvent. In a manner similar to that described for Fe/MAO in toluene, the polymerization conditions in hexane were again optimized by employing Fe1 as the test pre-catalyst. Meanwhile, the catalytic activities of Fe1–Fe6 and the properties of the resulting polyethylenes (PEs) were evaluated, with the results summarized in Table 3.

Table 3 Ethylene polymerization by Fe1–Fe6/MAO using n-hexane as the solvent^a

Run	Pre-cat.	T (°C)	t (min)	Al : Fe	Act.^b	Mw ^c	Mw/Mn ^c	Tm ^d (°C)
a Conditions: 2.0 μmol of Fe pre-catalyst, 100 mL of hexane, 10 atm C2H4.b 10^6 g (PE) mol^−1 (Fe) h^−1.c Mw (kg mol^−1), Mw and Mw/Mn determined by GPC.d Determined by DSC.
1	Fe1	50	30	1500	6.23	411.3	39.2	133.0
2	Fe1	60	30	1500	9.31	354.6	59.8	131.9
3	Fe1	70	30	1500	12.52	56.4	11.0	129.6
4	Fe1	80	30	1500	10.20	7.8	2.5	126.8
5	Fe1	70	30	500	8.36	267.4	19.9	133.4
6	Fe1	70	30	750	13.84	148.9	12.0	132.9
7	Fe1	70	30	1000	17.28	134.4	12.5	132.4
8	Fe1	70	30	1250	12.97	88.7	10.1	132.0
9	Fe2	70	30	1000	13.80	177.3	24.0	132.3
10	Fe3	70	30	1000	10.45	290.2	34.9	132.5
11	Fe4	70	30	1000	16.12	143.0	18.2	132.7
12	Fe5	70	30	1000	13.58	227.9	34.3	131.6
13	Fe6	70	30	1000	0.50	178.2	61.7	131.3

---

Initially, the thermal stability of the Fe1/MAO system was assessed between 50 °C and 80 °C with an Al : Fe molar ratio of 1500 : 1 in hexane under 10 atm of C₂H₄ (runs 1–4, Table 3). The best activity of 12.52 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ was again achieved at 70 °C, exceeding the performance observed in toluene at the same temperature (run 4, Table 2), indicating that the Fe1/MAO system exhibited better heat resistance r in n-hexane. GPC analysis (Fig. S11†) showed a trend similar to that in toluene, with higher molecular weight PEs (354.6–411.3 kg mol⁻¹) produced at lower temperatures. The bimodal polyethylenes obtained at low temperatures (50–60 °C) could be attributed to the multiple active species generated by residual alkyl aluminum in MAO, while the polyethylenes with the lowest molecular weight of 7.8 kg mol⁻¹ produced at 80 °C exhibited the narrowest distribution, indicating the failure of unstable active species at high temperatures (run 4, Table 3).¹¹ In most cases, the resultant polyethylenes obtained in n-hexane had much wider polydispersities than those obtained in toluene, partly due to the low solubility of iron complexes in n-hexane.^4e

Next, with the temperature held at 70 °C, the Al : Fe molar ratio was varied from 500 : 1 to 1500 : 1 (runs 3, 5–8, Table 3). Remarkably, an Al : Fe molar ratio of 1000 yielded the peak performance value of 17.28 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ (run 7, Table 3). The optimal dosage of MAO for the catalytic system in hexane was significantly lower compared to the amounts typically utilized in toluene, emphasizing the economic viability of the current system and its applicability for industrial applications. However, no comparable enhancement was noted in the analogous system employing B_OCF3 (Chart 1), despite the use of an even greater quantity of cocatalyst to activate the corresponding iron complexes.^7b

Under the above optimized conditions in n-hexane, the remaining five iron pre-catalysts (Fe2–Fe6) were tested toward ethylene polymerization (runs 7, 9–13, Table 3). All six iron complexes revealed moderate or good catalytic activity ((0.50–17.28) × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹), yielding high molecular weight polyethylenes (134.4–290.2 kg mol⁻¹) with broad molecular weight distributions (M_w/M_n: 12.5–61.7). The overall catalytic performances followed the order: Fe1 > Fe4 > Fe2 > Fe5 > Fe3 ≫ Fe6. In terms of steric hindrance, Fe6 and Fe3, which had the most significant steric hindrance, were the least active systems, whereas Fe1 and Fe4, with the least steric hindrance, were the most active. Unlike the Fe/MAO system in toluene, the electronic properties of a para-methyl group on the N-aryl had no obvious positive effect on the catalytic performance.^13c,16 As illustrated in Fig. S13,† the molecular weight distributions of the resulting PEs exhibited a notable range, spanning from distinctly bimodal (Fe2, Fe3, Fe5, Fe6) to apparently unimodal (Fe1, Fe4). This variation underscores the differing proportions of two active species, which arise from distinct termination kinetics. The relative abundance of these species is, to some extent, influenced by the varying configurations of the active species, a consequence of the steric hindrance imposed by ortho-substituents.^4e Generally, these iron pre-catalysts demonstrated superior activities and delivered polyethylenes with higher molecular weights in hexane compared to toluene.

Catalytic evaluation of Fe1–Fe6/MMAO in toluene

To investigate the impact of co-catalyst type on catalytic performance, MMAO was employed as an additional primary co-catalyst to activate these iron complexes for the purpose of ethylene polymerization in toluene; the polymerization data are summarized in Table 4.

Table 4 Ethylene polymerization by Fe1–Fe6/MMAO using toluene as the polymerization solvent^a

Run	Pre-cat.	T (°C)	t (min)	Al/Fe	Act.^b	Mw ^c	Mw/Mn ^c	Tm^d (°C)
a Conditions: 2.0 μmol Fe, 100 mL toluene, 10 atm of C2H4.b 10^6 g of (PE) mol^−1 (Fe) h^−1.c Mw: kg mol^−1, determined by GPC.d Determined by DSC.e 5 atm of C2H4.f 1 atm of C2H4.
1	Fe1	50	30	2000	6.50	25.5	10.8	128.1
2	Fe1	60	30	2000	12.99	20.8	5.1	127.9
3	Fe1	70	30	2000	13.31	17.7	4.0	129.5
4	Fe1	80	30	2000	8.90	9.2	2.8	128.0
5	Fe1	90	30	2000	2.63	3.7	2.5	123.3
6	Fe1	70	30	1000	4.90	48.2	4.9	131.3
7	Fe1	70	30	1500	7.80	18.2	3.2	130.5
8	Fe1	70	30	2500	12.11	9.2	3.3	127.9
9	Fe1	70	30	3000	8.66	8.6	3.2	126.5
10	Fe1	70	05	2000	24.18	3.3	2.2	123.0
11	Fe1	70	15	2000	9.08	4.5	2.5	124.5
12	Fe1	70	45	2000	9.46	20.3	14.3	124.6
13	Fe1	70	60	2000	7.16	22.9	10.7	125.6
14^e	Fe1	70	30	2000	2.40	1.5	1.5	119.8
15^f	Fe1	70	30	2000	0.70	1.1	1.6	115.6
16	Fe2	70	30	2000	6.88	3.8	2.7	123.3
17	Fe3	70	30	2000	5.39	4.2	3.0	123.8
18	Fe4	70	30	2000	7.38	3.6	2.4	123.2
19	Fe5	70	30	2000	6.22	4.2	2.5	123.8
20	Fe6	70	30	2000	0.30	2.0	2.1	123.8

---

When adjusting the temperature within the range of 50 to 90 °C (runs 1–5, Table 4) and modifying the Al : Fe molar ratio from 1000 : 1 to 3000 : 1 (runs 3, 6–9, Table 4), a peak activity of 13.31 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ was attained at 70 °C with an Al : Fe molar ratio of 2000 over a duration of 30 minutes. Analysis of the polymerization temperature-related data revealed that Fe1/MMAO exhibited inferior thermal stability compared to Fe1/MAO, producing PEs with decreased molecular weights and narrower molecular weight distributions (Fig. S14†).^6e,11 Furthermore, the Al : Fe molar ratio played a pivotal role in regulating the catalytic performance of iron complexes. As the Al : Fe molar ratio increased to 3000, the catalytic activity gradually declined from 13.31 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ to 8.66 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹, accompanied by a reduction in molecular weight from 48.2 kg mol⁻¹ to 8.6 kg mol⁻¹, suggesting an elevated rate of chain transfer from active iron species to aluminum (Fig. S15†).⁷ As the reaction time extended from 5 to 60 min, the catalytic activity decreased from 24.18 × 10⁶ to 7.16 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹, whereas the molecular weight of polyethylenes increased from 3.3 to 22.9 kg mol⁻¹ (Fig. S16†), indicating a rapid generation of active species upon the addition of MMAO and a long lifetime of these pre-catalysts despite partial deactivation (runs 3, 10–13, Table 4). Additionally, the ethylene pressure was found to be directly proportional to both the activity and molecular weight of the resulting polymer, showing that a specific monomer concentration is a necessary condition for overcoming mass transport limitations (runs 3, 14 and 15, Table 4). When compared to the Fe/MAO system, Fe1–Fe5/MMAO exhibited considerably lower activities and generated polyethylenes with decreased molecular weights and narrower molecular weight distributions, with the exception of Fe6/MMAO (Fig. S17†). As previously observed, reducing steric hindrance at the ortho-aryl position and replacing the para-aryl proton with a methyl group can enhance catalytic performance.^6f,15,17a

A comparative view of the previously reported and current iron pre-catalysts

To gain insights into how the structural enhancements achieved through the incorporation of ortho-difluorobenzhydryl and p-trifluoromethoxy groups influence the catalytic performance of iron pre-catalysts, a meticulous comparison of the performance attributes of the current precatalyst family (D, Chart 1) with previously documented B and C (Chart 1)^{6a,b,f,7b–d} was conducted and is summarized in Fig. 3. The activity levels and optimal operating temperatures of Fe4, in particular, are presented alongside corresponding data of the mesityl-containing instances of B and C. All catalytic reactions were executed under conditions of 10 atm of C₂H₄, utilizing MAO as the activator.

Fig. 3 Comparative catalytic performance of D (Fe4 in this work) with mesityl-containing B_Me, C_F, C_Cl, C_OMe and B_OCF3; all runs were conducted with MAO as the activator under optimum conditions.

In terms of catalytic activity, Fe4 (D, Fig. 3) and C_Cl^7b emerged as notably more potent than other iron complexes, with Fe4 achieving a peak activity of 15.71 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ at 80 °C. Among all iron analogues, Fe4 exhibited the highest optimal temperature (80 °C), highlighting the beneficial effects of incorporating multiple-fluoro-substituents into the precatalyst structure on both catalytic activity and thermal stability. Regarding the molecular weight of the resulting polyethylenes, the inclusion of a remote electron-donating group within the benzhydryl framework had a favorable effect, with the methoxy-containing C_OMe^7c yielding polymers of the highest molecular weight. This finding illustrates the capacity of the remote methoxy group to positively influence molecular weight within this set. Conversely, Fe4 (D, Fig. 3) and C_Cl,^7b which contain electron-withdrawing substituents in their benzhydryl groups, exhibited the lowest molecular weight value (around 10.9 kg mol⁻¹), indicating the detrimental effects of electron-withdrawing groups on polymer chain growth. Due to the combined electronic effects of the remote substituents on benzhydryl and the para-substituents, Fe4 (D, Fig. 3) exhibited a greater similarity to its chloride counterpart, C_Cl,^7b than to C_F.^7a Notably, the introduction of a remote fluoro group significantly boosted activity and led to the formation of a distinct type of polyethylene when compared to other trimethoxy-containing iron complexes, B_OCF3.^6b Furthermore, a comparative analysis between the previously documented cobalt analogues⁸ and the present iron complexes reveals that the iron complexes demonstrate markedly superior thermal stability and yield polyethylenes characterized by broader molecular weight distributions. These findings strongly highlight the predominant advantages that iron complexes hold in facilitating efficient ethylene polymerization.

Microstructural properties of the resultant polyethylenes

All polyethylenes produced in this study exhibited high melting temperatures (≥126 °C), irrespective of the solvent or co-catalyst variations, indicating their high linearity.¹⁸ Notably, only the samples produced at low pressure (i.e. 1 atm C₂H₄, run 18, Table 2) displayed melting temperatures below 126 °C. To further confirm the linearity and elucidate the end group configuration of these polyethylenes, high-temperature ¹H and ¹³C NMR spectroscopy was performed on several selected samples: PE_{Fe1/MAO/toluene} (run 5, Table 2, M_w = 35.7 kg mol⁻¹), PE_{Fe1/MAO/hexane} (run 7, Table 3, M_w = 134.4 kg mol⁻¹), and PE_{Fe1/MMAO/toluene} (run 3, Table 4, M_w = 17.7 kg mol⁻¹). For adequate solubility, the polyethylenes were pre-dissolved in 1,1,2,2-tetrachloroethane-d₂ at 100 °C, and the spectra were recorded at the same temperature.

For the typical sample PE_{Fe1/MAO/toluene} (run 4, Table 2, M_w = 59.0 kg mol⁻¹), the presence of an intense signal at δ 1.30 in the ¹H NMR spectrum and δ 29.41 in the ¹³C NMR spectrum served as evidence of the polyethylenes’ linearity (Fig. 4). Moreover, the ¹H NMR spectrum revealed proton signals attributable to a vinyl group at δ 5.89 (H_b), δ 4.93 (H_a) and δ 4.99 (H_a′) in a 1 : 1 : 1 ratio. The existence of two vinylic carbon signals at δ 115.87 (a) and δ 141.69 (b) in the ¹³C NMR spectrum further supported this finding. Additionally, the ratio of the vinylic H_b proton to the methyl H_g protons at the opposite end of the PE chain was approximately 1 : 3, providing strong evidence for the presence of an unsaturated chain end. The vinylic H_b proton, indicative of a terminal double bond, suggested the presence of vinyl functionality at one end of the polyethylenes, while the methyl H_g protons confirmed a saturated structure at the other end. The ¹³C NMR spectrum also provided clear evidence for the existence of a saturated chain end in the structure of polyethylenes, with distinct signals at δ 22.44 (f), corresponding to the methylene carbon and δ 13.85 (g) associated with the methyl end-group. These observations were consistent with the expected polymerization mechanism, where chain termination occurs through β-hydride elimination, leaving a vinyl group at the terminal position.⁷ Importantly, identical signal integration and chemical shift assignments were also observed in other polyethylenes, namely, PE_{Fe1/MMAO/toluene} and PE_{Fe1/MAO/hexane}, produced using different co-catalysts or polymerization solvents (Fig. S18–S20†), providing further evidence for the mechanistic reliability of the current catalytic systems. Interestingly, an analysis of the NMR spectra of polyethylene (PE) samples produced in hexane, with MAO serving as the co-catalyst, revealed the presence of an analogous vinyl chain-terminated group. This observation suggests that the polymerization solvent exerts a relatively minor influence on the chain termination mechanisms during the polymerization process. Meanwhile, the ¹³C NMR spectrum of PE_{Fe1/MMAO/toluene} exhibits several weak signals at chemical shifts, δ 39.12 (c), 27.93 (b), 27.30 (d) and 22.56 (a), which can be ascribed to an isobutyl end group, implying that both chain transfer to aluminum and β-H elimination served as viable chain termination pathways with β-H elimination being the predominant mechanism.

Fig. 4 The high temperature ¹³C NMR spectrum of PE_{Fe1/MAO/toluene} (run 5, Table 2) along with an inset of its high temperature ¹H NMR spectrum recorded in 1,1,2,2-tetrachloroethane-d₂ at 100 °C.

Experimental

General methods

To handle chemicals sensitive to air or moisture, standard Schlenk techniques were employed under an atmosphere of dry nitrogen or within a nitrogen-filled glove box. Prior to use, the solvents hexane and toluene were dried over sodium and distilled under nitrogen. Methylaluminoxane (1.46 M solution in toluene) and modified methylaluminoxane (1.93 M solution in n-heptane) were sourced from Anhui Botai Electronic Materials Co., while high-purity ethylene, obtained from Beijing Yanshan Petrochemical Corp., was used without further purification. Additional reagents were supplied by Aldrich, Acros, or local providers. The aniline, 2,6-bis(bis(4-fluorophenyl)methyl)-4-(trifluoromethoxy)aniline,¹⁹ 2-acetyl-6-[1-(2,6-bis(bis(4-fluorophenyl)methyl)-4-trifluoromethoxyphenylimino)ethyl] pyridine (A) and the organic ligands (L1–L6)⁸ were synthesized based on our previously published methods. FT-IR spectra were recorded using a PerkinElmer System 2000 FT-IR spectrometer, while elemental analyses were conducted on a Flash EA 1112 microanalyzer. ¹H and ¹³C NMR spectra of all novel organic and inorganic compounds were obtained on a Bruker Avance 400 MHz spectrometer at room temperature, employing CDCl₃ as the solvent and TMS as the internal standard. ¹⁹F NMR spectra were measured on a Bruker AVANCE 700 MHz instrument at ambient temperature. The melting temperatures (T_m) of the polyethylene (PE) samples were determined using a PerkinElmer TA-Q2000 DSC analyzer under a nitrogen atmosphere. The method involved heating a weighed sample (4.0–6.0 mg) to 160 °C at a rate of 20 °C min⁻¹, holding at 160 °C for 5 minutes to eliminate thermal history, and then cooling to −20 °C at a rate of 20 °C min⁻¹. Finally, the sample was heated to 160 °C at a rate of 10 °C min⁻¹. Gel permeation chromatography (GPC) of the PEs was performed using an Agilent PLGPC 220 instrument equipped with a refractive index (RI) detector operating at 160 °C, with 1,2,4-trichlorobenzene as the eluent. For the ¹H and ¹³C NMR spectroscopy of the PEs, a weighed amount of PE (20–40 mg) was first dissolved in 1,1,2,2-tetrachloroethane-d₂ (1 mL) at elevated temperature, using TMS as the internal standard. The spectra were then recorded on a Bruker AVANCE III 500 MHz instrument at 100 °C. The operating conditions for the ¹H NMR spectra included a spectral width of 15.000 kHz, an acquisition time of 2.1845 s, a relaxation delay of 2.0 s, and 64 scans. For the ¹³C NMR spectra, the conditions were a spectral width of 31.250 kHz, an acquisition time of 0.52 s, a relaxation delay of 1.5 s, and 1024 scans.

Synthesis and characterization

Synthesis of [2-[CMeN{2,6-{(4-FC₆H₄)₂CH}₂-4-OCF₃}]-6-(CMeNAr)C₅H₃N]FeCl₂. Ar = 2,6-Me₂C₆H₃ (Fe1): Under a nitrogen atmosphere, a mixture comprising L1 (0.27 mmol, 0.21 g) and FeCl₂ (0.27 mmol, 0.03 g) in a solvent blend of dichloromethane (10 mL) and ethanol (5 mL) was stirred at room temperature overnight. Subsequently, all volatiles were then removed under reduced pressure to give a concentrated solution. To precipitate the product, an excess of diethylether was introduced, and the resulting precipitate was collected via filtration and washed with diethylether (3 × 10 mL), yielding Fe1 as a blue powder (0.21 g, 84%). FT-IR (cm⁻¹): 2970 (w), 2917 (W), 1628 (w), 1595 (C=N, m), 1505 (s), 1478 (w), 1440 (m), 1374 (w), 1245 (s), 1213 (s), 1198 (w), 1168 (w), 1157 (s), 1102 (w), 1015 (w), 837 (s), 815 (w), 839 (s), 734 (w). ¹⁹F NMR (672 MHz, CDCl₃): δ −57.06, −116.83, −118.12. Anal. calc. for C₅₀H₃₈Cl₂FeF₇N₃O (956.61): C, 62.78; H, 4.00; N, 4.39. Found: C, 62.54; H, 4.28; N, 4.63.

Ar = 2,6-Et₂C₆H₃ (Fe2): By employing a methodology analogous to that detailed for Fe1, with molar ratios similarly adjusted, but utilizing L2 in lieu of L1, Fe2 was synthesized and isolated as a blue powder (0.19 g, 73%). FT-IR (cm⁻¹): 3067 (w), 2968 (w), 2931 (w), 1600 (C=N, m), 1506 (s), 1447 (m), 1372 (w), 1253 (w), 1221 (m), 1159 (s), 1198 (w), 1099 (w), 1015 (w), 1102 (w), 836 (s), 656 (w). ¹⁹F NMR (672 MHz, CDCl₃): δ −56.95, −116.37, −117.02. Anal. calc. for C₅₂H₄₂Cl₂FeF₇N₃O (984.66): C, 63.43; H, 4.30; N, 4.27. Found: C, 63.51; H, 4.53; N, 4.55.

Ar = 2,6-i-Pr₂C₆H₃ (Fe3): By employing a methodology analogous to that detailed for Fe1, with molar ratios similarly adjusted, but utilizing L3 in lieu of L1, Fe3 was synthesized and isolated as a blue powder (0.23 g, 85%). FT-IR (cm⁻¹): 3064 (w), 2965 (w), 2871 (w), 2161 (w), 1600 (C=N, w), 1581 (w), 1505 (vs), 1445 (w), 1370 (w), 1321 (w), 1250 (w), 1227 (w), 1205 (w), 1167 (w), 1158 (m), 1098 (w), 1014 (w), 875 (w), 836 (m), 820 (w), 789 (w), 770 (w), 733 (w), 712 (w). ¹⁹F NMR (672 MHz, CDCl₃): δ −54.64, −114.66, −115.90. Anal. calc. for C₅₁H₄₆Cl₂FeF₇N₃O (1012.72): C, 64.05; H, 4.58; N, 4.15. Found: C, 63.82; H, 4.61; N, 4.39.

Ar = 2,4,6-Me₃C₆H₂ (Fe4): By employing a methodology analogous to that detailed for Fe1, with molar ratios similarly adjusted, but utilizing L4 in lieu of L1, Fe4 was synthesized and isolated as a blue powder (0.21 g, 80%). FT-IR (cm⁻¹): 2930 (w), 2962 (w), 1642 (w), 1602 (C=N, m), 1505 (s), 1439 (w), 1374 (w), 1248 (m), 1216 (m), 1178 (w), 1158 (w), 1094 (w), 1014 (w), 867 (w), 837 (w), 818 (w), 786 (w), 655 (w). ¹⁹F NMR (672 MHz, CDCl₃): δ −57.06, −116.33, −117.06. Anal. calc. for C₅₁H₄₀Cl₂FeF₇N₃O (970.63): C, 63.11; H, 4.15; N, 4.33. Found: C, 63.00; H, 4.20; N, 4.35.

Ar = 2,6-Et₂-4-MeC₆H₂ (Fe5): By employing a methodology analogous to that detailed for Fe1, with molar ratios similarly adjusted, but utilizing L5 in lieu of L1, Fe5 was synthesized and isolated as a blue powder (0.19 g, 73%). FT-IR (cm⁻¹): 2930 (w), 2962 (w), 1642 (w), 1602 (C=N, m), 1505 (s), 1439 (w), 1374 (w), 1248 (m), 1216 (m), 1178 (w), 1158 (w), 1094 (w), 1014 (w), 867 (w), 837 (w), 818 (w), 786 (w), 655 (w). ¹⁹F NMR (672 MHz, CDCl₃): δ −57.06, −116.33, −117.06. Anal. calc. for C₅₃H₄₄Cl₂FeF₇N₃O (998.69): C, 63.74; H, 4.44; N, 4.21. Found: C, 63.58; H, 4.43; N, 4.37.

Ar = 2,6-(4-FPh)₂-4-OCF₃C₆H₂ (Fe6): By employing a methodology analogous to that detailed for Fe1, with molar ratios similarly adjusted, but utilizing L6 in lieu of L1, Fe6 was synthesized and isolated as a blue powder (0.31 g, 81%). FT-IR (cm⁻¹): 2930 (w), 2962 (w), 1642 (w), 1602 (C=N, m), 1505 (s), 1439 (w), 1374 (w), 1248 (m), 1216 (m), 1178 (w), 1158 (w), 1094 (w), 1014 (w), 867 (w), 837 (w), 818 (w), 786 (w), 655 (w). ¹⁹F NMR (672 MHz, CDCl₃): δ −56.28, −114.21, −114.94. Anal. calc. for C₇₅H₄₉Cl₂FeF₁₄N₃O₂ (1416.96): C, 63.57; H, 3.49; N, 2.97. Found: C, 63.51; H, 3.53; N, 3.05.

X-ray diffraction studies

Single crystals of Fe4 and Fe6, suitable for X-ray diffraction analysis, were grown through the slow diffusion of diethyl ether into their respective dichloromethane solutions at room temperature, all under a nitrogen atmosphere to ensure an inert environment. A crystal of optimal quality from each compound was carefully selected and mounted onto an XtaLAB Synergy R, HyPix diffractometer, which featured a graphite-monochromated Cu-Kα radiation source (λ = 1.54184 Å) and a nitrogen cold stream for temperature control. Throughout the data collection process, the crystals were maintained at a precise temperature of 169.98(10) K to minimize thermal effects on the diffraction patterns. Utilizing the Olex2 software suite,²⁰ the crystal structures were solved by employing the ShelXT²¹ structure solution program, which leveraged intrinsic phasing techniques. Following the initial solution, the structures were refined using the ShelXL²² refinement package, which applied least-squares minimization to optimize the structural parameters. During the structural refinement, the disordered solvent in the cell structure was squeezed (Fe6) with PLATON software.²³ The comprehensive details of the X-ray structure determinations and refinement processes are documented in Table S1.† In the case of Fe4, twinned crystals were obtained, and resolving the twinning proved challenging. Nevertheless, the data obtained fully support the proposed structure.

General methods for ethylene polymerization

Ethylene polymerization at ethylene pressures of 5 or 10 atm. The polymerization reactions conducted at ethylene pressures (P_C2H4) of either 5 or 10 atm were executed within a 250 mL stainless-steel autoclave, which was furnished with a mechanical stirrer, along with sophisticated systems for pressure and temperature control. Prior to the experiment, the autoclave was meticulously dried and subsequently evacuated under vacuum. It was then purged twice with nitrogen and once with ethylene to ensure an inert atmosphere. Upon attainment of the desired temperature, a solution comprising the iron pre-catalyst dissolved in toluene or hexane (50 mL) was injected into the autoclave, immediately followed by an additional 25 mL of toluene or hexane. Subsequently, the requisite quantity of aluminum activator (either MAO or MMAO) was introduced, and a further 25 mL of toluene or hexane was injected using a syringe to finalize the addition process. The autoclave was promptly pressurized to the specified ethylene pressure and vigorously agitated at a speed of 400 revolutions per minute. Following the completion of the predetermined reaction duration, the ethylene pressure was safely released, and the reaction mixture was quenched using a solution of 10% hydrochloric acid in ethanol. Ethanol was employed to precipitate and collect the resulting polymer. Subsequently, the obtained polymer was thoroughly washed with ethanol, dried under reduced pressure at 100 °C, and then weighed for further analysis.

Ethylene polymerization at an ethylene pressure of 1 atm

For the polymerization reactions conducted at an ethylene pressure (P_C2H4) of 1 atm, the co-catalyst was first introduced into a dry Schlenk vessel, which was equipped with a stir bar. The vessel was then evacuated to create a vacuum and subsequently backfilled with ethylene. Freshly distilled toluene (30 mL) was introduced into the vessel, followed by the necessary amount of activator (either MAO or MMAO). The reaction mixture was stirred under an ethylene pressure of 1 atm for a duration of 30 minutes. Once the stirring period was complete, the ethylene supply was terminated, and the pressure was vented from the vessel. The reaction mixture was then quenched using a solution of 10% hydrochloric acid in ethanol. The resulting polymer was collected by precipitation, washed with ethanol, and finally dried under reduced pressure at 100 °C for further use or analysis.

Conclusions

A series of bis(imino)pyridine iron(II) chloride complexes, featuring trifluoromethoxy and difluorobenzhydryl moieties, was synthesized, thoroughly characterized and systematically evaluated as precatalysts for ethylene polymerization. Upon activation with methylaluminoxane (MAO) or modified methylaluminoxane (MMAO), these iron complexes exhibit exceptionally high catalytic proficiency and remarkable thermal stability during ethylene polymerization, producing highly linear polyethylenes with predominantly vinyl-terminated groups, suitable for applications in the production of functional polymers, coatings and lubricant additives. Notably, when MAO-activated, Fe1 demonstrated peak catalytic activity, achieving a level of 17.28 × 10⁶ g (PE) mol⁻¹ (Fe) h⁻¹ at 70 °C in hexane. The molecular weight of the resultant PE ranged from 2.0 to 567.7 kg mol⁻¹, adjustable through modifications in the ligand structure (Fe1–Fe6) or alterations in the polymerization conditions. Intriguingly, sterically hindered complexes, such as Fe6 and Fe3, generated PEs with elevated molecular weights, underscoring the beneficial influence of bulkier substituents on chain propagation processes. Furthermore, the incorporation of remote fluoro groups and trifluoromethoxy substituents was found to positively impact both catalytic performance and thermal stability, aligning with our prior research findings. Notably, superior catalytic performance was achieved in hexane, a solvent of significant industrial relevance, highlighting the potential of these iron complexes for practical, industrial applications.

Author contributions

Hosseinzadeh Z., Zhang Q., and Sun W.-H. made substantial contributions to the conception and design of the study and performed data analysis and interpretation; Wang Y., Wang Q., Ren G., Liang T., and Ma Y performed data acquisition and provided administrative, technical, and material support.

Conflicts of interest

All authors declared that there are no conflicts of interest.

Data availability

The data supporting this article have been included as part of the ESI.†

Acknowledgements

Z. H. is grateful to the Chinese Academy of Sciences President's International Fellowship Initiative (no. 2020PM0055).

References

1. (a) J. Zheng, Y. Zhang, D. Hou, Y. Qin, W. Guo, C. Zhang and J. Shi, Front. Mech. Eng., 2018, 13, 535–545; (b) M. Jacobs and J. Van Dingenen, J. Mater. Sci., 2001, 36, 3137–3142.
2. E. I. Negishi, Angew. Chem., Int. Ed., 2011, 50, 6738–6764.
3. (a) G. Natta, P. Pino, P. Corradini, F. Danusso, E. Mantica, G. Mazzanti and G. Moraglio, J. Am. Chem. Soc., 1955, 77, 1708–1710; (b) M. Stürzel, S. Mihan and R. Mülhaupt, Chem. Rev., 2016, 116, 1398–1433; (c) M. Delferro and T. J. Marks, Chem. Rev., 2011, 111, 2450–2485; (d) W. Kaminsky, Polyolefins: 50 Years After Ziegler and Natta II, Springer, Berlin, 2013; (e) P. E. Figgins and D. H. Busch, J. Am. Chem. Soc., 1960, 82, 820–824; (f) G. J. Britovsek, V. C. Gibson and D. F. Wass, Angew. Chem., Int. Ed., 1999, 38, 428–447.
4. (a) B. L. Small, M. Brookhart and A. M. Bennett, J. Am. Chem. Soc., 1998, 120, 4049–4050; (b) G. P. Britovsek, V. Gibson, S. McTavish, G. Solan, A. P. White, D. Williams, B. Kimberley and P. Maddox, Chem. Commun., 1998, 7, 849–850; (c) G. J. Britovsek, M. Bruce, V. C. Gibson, B. S. Kimberley, P. J. Maddox, S. Mastroianni, S. J. McTavish, C. Redshaw, G. Solan and S. Strömberg, J. Am. Chem. Soc., 1999, 121, 8728–8740; (d) W.-H. Sun, S. Jie, S. Zhang, W. Zhang, Y. Song, H. Ma, J. Chen, K. Wedeking and R. Fröhlich, Organometallics, 2006, 25, 666–677; (e) D. Takeuchi, S. Takano, Y. Takeuchi and K. Osakada, Organometallics, 2014, 33, 5316–5323.
5. (a) Z. Wang, G. A. Solan, W. Zhang and W.-H. Sun, Coord. Chem. Rev., 2018, 363, 92–108; (b) S.-F. Yuan, Y. Yan, G. A. Solan, Y. Ma and W.-H. Sun, Coord. Chem. Rev., 2020, 411, 213254; (c) A. M. F. Phillips, H. Suo, F. C. Maria, A. J. Pombeiro and W.-H. Sun, Coord. Chem. Rev., 2020, 416, 213332; (d) H. Suo, G. A. Solan, Y. Ma and W.-H. Sun, Coord. Chem. Rev., 2018, 372, 101–116; (e) Y. Wang, W. Zhang, X. Wang, Y. Ma and W.-H. Sun, Acta Polym. Sin., 2023, 54, 564–583.
6. (a) J. Yu, H. Liu, W. Zhang, X. Hao and W.-H. Sun, Chem. Commun., 2011, 47, 3257–3259; (b) M. Liu, Z. Ning, Y. Ma, G. A. Solan, T. Liang and W.-H. Sun, J. Organomet. Chem., 2023, 994, 122740; (c) Q. Mahmood, J. Guo, W. Zhang, Y. Ma, T. Liang and W.-H. Sun, Organometallics, 2018, 37, 957–970; (d) X. Cao, F. He, W. Zhao, Z. Cai, X. Hao, T. Shiono, C. Redshaw and W.-H. Sun, Polymer, 2012, 53, 1870–1880; (e) Q. Mahmood, E. Yue, J. Guo, W. Zhang, Y. Ma, X. Hao and W.-H. Sun, Polymer, 2018, 159, 124–137; (f) T. Liu, Y. Ma, G. A. Solan, T. Liang and W.-H. Sun, Appl. Organomet. Chem., 2021, 35, e6259.
7. (a) W.-H. Sun, W. Zhao, J. Yu, W. Zhang, X. Hao and C. Redshaw, Macromol. Chem. Phys., 2012, 213, 1266–1273; (b) T. Liu, Y. Ma, A. S. Solan, Y. Sun and W.-H. Sun, New J. Chem., 2023, 47, 5786–5795; (c) T. Liu, Y. Ma, G. Solan, T. Liang and W.-H. Sun, Appl. Organomet. Chem., 2021, 35, e6259; (d) R. Zhang, M. Han, Y. Ma, G. A. Solan, T. Liang and W.-H. Sun, Dalton Trans., 2019, 48, 17488–17498.
8. Z. Hosseinzadeh, Q. Zhang, Q. Wang, A. M. Ashfaq, T. Liang, Y. Ma and W.-H. Sun, ChemPlusChem, 2025, 90, e202500007.
9. (a) E. Castagnetti and M. Schlosser, Chem. – Eur. J., 2002, 8, 799–804; (b) F. R. Leroux, B. Manteau, J.-P. Vors and S. Pazenok, Beilstein J. Org. Chem., 2008, 4, 13.
10. J. Lai, W. Zhao, W. Yang, C. Redshaw, T. Liang, Y. Liu and W.-H. Sun, Polym. Chem., 2012, 3, 787–793.
11. (a) T. M. Smit, A. K. Tomov, G. J. Britovsek, V. C. Gibson, A. J. White and D. J. Williams, Catal. Sci. Technol., 2012, 2, 643–655; (b) S. Wang, B. Li, T. Liang, C. Redshaw, Y. Li and W.-H. Sun, Dalton Trans., 2013, 42, 9188–9197; (c) W. Zhao, E. Yue, X. Wang, W. Yang, Y. Chen, X. Hao, X. Cao and W.-H. Sun, J. Polym. Sci., Part A: Polym. Chem., 2017, 55, 988–996; (d) N. V. Semikolenova, W.-H. Sun, I. E. Soshnikov, M. A. Matsko, O. V. Kolesova, V. A. Zakharov and K. P. Bryliakov, ACS Catal., 2017, 7, 2868–2877; (e) D. P. Gates, S. A. Svejda, E. Oñate, C. M. Killian, L. K. Johnson, P. S. White and M. Brookhart, Macromolecules, 2000, 33, 2320–2334; (f) C. S. Popeney, A. L. Rheingold and Z. Guan, Organometallics, 2009, 28, 4452–4463.
12. (a) S. Zhang, I. Vystorop, Z. Tang and W.-H. Sun, Organometallics, 2007, 26, 2456–2460; (b) S. Zhang, W.-H. Sun, X. Kuang, I. Vystorop and J. Yi, J. Organomet. Chem., 2007, 692, 5307–5316.
13. (a) S. A. Cantalupo, H. E. Ferreira, E. Bataineh, A. J. King, M. V. Petersen, T. Wojtasiewicz, A. G. DiPasquale, A. L. Rheingold and L. H. Doerrer, Inorg. Chem., 2011, 50, 6584–6596; (b) J. Yuan, W.-B. Shi and H.-Z. Kou, Transition Met. Chem., 2015, 40, 807–811; (c) C. Huang, Y. Zhang, G. A. Solan, Y. Ma, X. Hu, Y. Sun and W.-H. Sun, Eur. J. Inorg. Chem., 2017, 36, 4158–4166.
14. (a) T. Xiao, P. Hao, G. Kehr, X. Hao, G. Erker and W.-H. Sun, Organometallics, 2011, 30, 4847–4853; (b) Y. Zhang, C. Huang, X. Hao, X. Hu and W.-H. Sun, RSC Adv., 2016, 6, 91401–91408; (c) V. K. Appukuttan, Y. Liu, B. C. Son, C.-S. Ha, H. Suh and I. Kim, Organometallics, 2011, 30, 2285–2294; (d) Q. Mahmood, Y. Zeng, X. Wang, Y. Sun and W.-H. Sun, Dalton Trans., 2017, 46, 6934–6947.
15. Z. Wang, R. Zhang, W. Zhang, G. A. Solan, Q. Liu, T. Liang and W.-H. Sun, Catal. Sci. Technol., 2019, 9, 1933–1943.
16. N. E. Mitchell, J. W. C. Anderson and B. K. Long, J. Polym. Sci., Part A: Polym. Chem., 2017, 55, 3990–3995.
17. (a) W.-H. Sun, X. Tang, T. Gao, B. Wu, W. Zhang and H. Ma, Organometallics, 2004, 23, 5037–5047; (b) V. C. Gibson, C. Redshaw, G. A. Solan, A. J. White and D. J. Williams, Organometallics, 2007, 26, 5119–5123.
18. W. Zhang, W.-H. Sun and C. Redshaw, Dalton Trans., 2013, 42, 8988–8997.
19. Z. Hosseinzadeh, M. Liu, Q. Zhang, T. Liang, G. A. Solan, Y. Ma and W.-H. Sun, Catalysts, 2022, 12, 1520.
20. O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. Howard and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
21. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Adv., 2015, 71, 3–8.
22. G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71, 3–8.
23. L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2009, 65, 148–155.

---

Footnote

† Electronic supplementary information (ESI) available: GPC and DSC curves are collected as well as representative ¹³C NMR of the resultant polyethylenes. Crystal data and structure refinement are provided for Fe4 and Fe6. CCDC 2266904 and 2266905. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5dt00837a

---