Interplay of Supramolecular Chemistry and Photochemistry with Palladium-Catalyzed Ethylene Polymerization

Open AccessCCS ChemistryRESEARCH ARTICLE1 Jul 2021Interplay of Supramolecular Chemistry and Photochemistry with Palladium-Catalyzed Ethylene Polymerization Guohong Wang†, Dan Peng†, Yao Sun and Changle Chen Guohong Wang† CAS Key Laboratory of Soft Matter Chemistry, Hefei National Laboratory for Physical Sciences at the Microscale, Department of Polymer Science and Engineering, University of Science and Technology of China, Hefei 230026 , Dan Peng† CAS Key Laboratory of Soft Matter Chemistry, Hefei National Laboratory for Physical Sciences at the Microscale, Department of Polymer Science and Engineering, University of Science and Technology of China, Hefei 230026 , Yao Sun CAS Key Laboratory of Soft Matter Chemistry, Hefei National Laboratory for Physical Sciences at the Microscale, Department of Polymer Science and Engineering, University of Science and Technology of China, Hefei 230026 and Changle Chen *Corresponding author: E-mail Address: [email protected] CAS Key Laboratory of Soft Matter Chemistry, Hefei National Laboratory for Physical Sciences at the Microscale, Department of Polymer Science and Engineering, University of Science and Technology of China, Hefei 230026 https://doi.org/10.31635/ccschem.020.202000414 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Metal–metal cooperativity effects have been extensively explored in olefin polymerization, along with the design and preparation of many binuclear transition metal catalysts. However, their synthesis and the tuning of metal–metal distance are complicated and time-consuming. In this contribution, a supramolecular chemistry strategy was used to construct multinuclear olefin polymerization catalysts. Urea functional groups were installed into several α-diimine palladium catalysts to enable hydrogen-bonding-induced self-assembly. Compared with methylated counterparts (devoid of supramolecular interactions), the assembled structures and their catalytic properties were affected by the concentration, ligand sterics, temperature, and solvents, which ultimately changed the polymerization properties. Furthermore, the introduction of azobenzene units into the catalyst framework gave rise to photoresponsive behaviors in the assembled catalysts. During ethylene polymerization and copolymerization with methyl acrylate, important parameters were easily modulated, including the activity, comonomer incorporation, polymer branching density, molecular weight, and molecular weight distribution. The introduction of supramolecular chemistry and photochemistry strategies to transition-metal-catalyzed olefin polymerization opens up new possibilities for the design of new polyolefin materials. Download figure Download PowerPoint Introduction The annual production of polyolefins exceeds 180 million tons, which accounts for more than half of the global plastics market.1 In this field, transition-metal catalysts based on various ligands have played critical roles in the development of new technologies and materials.2–5 The majority of research has involved the development of catalysts, followed by extensive ligand electronic/steric modifications. Recently, some design strategies beyond the simple electronic/steric modulation of catalysts have emerged, which can be applied to different catalyst systems.6 For example, chain shuttling polymerization strategies can produce random multiblock copolymers of polyolefins with alternating amorphous and semicrystalline regions.7 The installation of a secondary coordinating ligand that can interact with the metal center or growing polymer chain can efficiently modulate olefin polymerization and copolymerization processes (Scheme 1, I and II).8–13 The use of specially designed comonomers with functional groups that can interact with metal centers has been demonstrated for (co)polymerization reactions (Scheme 1, III).14–17 Redox control strategies (Scheme 1, IV) can be used to generate two or more different catalyst systems using only a single catalyst precursor for olefin polymerization and copolymerization.18–20 Scheme 1 | Previously reported strategies that can modulate the properties of olefin polymerization catalysts (I–V); dinuclear and multinuclear metal complexes assembled via supramolecular chemistry (VI and VII). Download figure Download PowerPoint In addition to the aforementioned strategies, metal–metal cooperativity effects have been extensively explored for polymerization and have proven effective at modulating polymer microstructures, as well as comonomer incorporation for copolymerization (Scheme 1, V).21–25 However, this technique requires the synthesis of binuclear or multinuclear transition-metal catalysts, which can be complicated and time-consuming. Moreover, the tuning of metal–metal cooperativity effects (e.g., metal–metal distances) usually demands the design/synthesis of totally new ligand motifs. Recently, supramolecular systems held together by noncovalent bonds have found wide application in many domains of chemistry.26 Specifically, multinuclear metal complexes can be assembled through diverse noncovalent interactions such as metal–ligand coordination, hydrogen bonding, van der Waals interactions, aromatic stacking, and so forth.27 This provides a simple and versatile strategy that enables the rapid construction and modulation of dinuclear/multinuclear catalysts.28 For example, Gianneschi et al.29 assembled two salen-based Cr(III) centers by coordinating Rh(I) with phosphine and thioether moieties (Scheme 1, VI) and studied their properties during the asymmetric ring opening of cyclohexene oxide. In another example, Park et al.30 designed a chiral bimetallic Co(II)-salen catalyst (Scheme 1, VII) that self-assembled through hydrogen bonding and significantly accelerated the reaction rate and also displayed excellent enantioselectivity in the Henry reaction. The dynamic and reversible features of noncovalent interactions in supramolecular assemblies enable their structures to be easily tuned. For example, the addition of CO to system VI cleaved the Ru(I)-thioether coordination bonds, and this allosteric control enabled easy control of the catalytic properties of this system. Various stimuli (such as concentration, solvents, light, and postmodification reactions) have been used to trigger transformations of supramolecular architectures, which have formed new structures with new properties.31 The utilization of supramolecular strategies to construct multinuclear olefin polymerization catalysts remains largely unexplored. In this contribution, urea functional groups were introduced into α-diimine palladium (Pd) catalysts to take advantage of hydrogen-bonding self-assembly. Compared with methylated counterparts that lack supramolecular interactions, the assembled structures and their catalytic properties were affected by the concentration, ligand sterics, temperature, and polymerization solvents, which translated into changes in the catalytic properties. Furthermore, the introduction of azobenzene units into the catalyst framework gave rise to photoresponsive behaviors in the assembled catalysts. Interestingly, some of these Pd complexes produced polymers with bimodal gel permeation chromatography (GPC) distributions, whose molecular weight distributions could be modulated by changing the experimental conditions. The experimental results supported a mechanism in which the assembled Pd species produced high-molecular-weight fractions, while the monomeric complexes produced low-molecular-weight fractions. The molecular weight distribution is an important parameter that determines many polymer properties.32,33 For polyolefin materials, bimodal polyethylene has many unique properties, such as the ability to limit shear forces during extrusion,34,35 but its preparation requires relatively complicated procedures that involve combinations of catalysts and/or reactors. This work provides an alternative strategy to control the polyolefin molecular weight distribution using supramolecular chemistry and photochemistry. Experimental Section All manipulations were carried out using standard Schlenk technique or in glovebox unless otherwise mentioned. Toluene, CH2Cl2, CHCl3, CDCl3, Et2O, and n-hexane were purified over 4 Å molecular sieves. All the other reagents were used as received from commercial sources. Nuclear magnetic resonance (1H NMR and 13C NMR) spectra were recorded on a Bruker 400 MHz instrument at room temperature unless otherwise stated. Molecular weight and molecular weight distribution of the polymer were determined by GPC equipped with two linear Styragel columns (HR2 and HR4) at 40 °C using tetrahydrofuran (THF) as a solvent and calibrated with polystyrene standards. THF was employed as the eluent at a flow rate of 1.0 mL·min−1. Single crystals of the metal complexes were obtained by slow diffusion of MeOH into CH2Cl2 solutions. X-ray diffraction data were collected at 298(2) K on a Bruker Smart charge-coupled device (CCD) area detector with graphite-monochromated Mo Kα (λ = 0.71073 Å) radiation at room temperature. The size of nanoparticles was characterized by NanoBrook 90Plus PALS in 25 °C. The diameter of the nylon syringe filter we used to filter the solution before dynamic light scattering (DLS) test was 13 mm, and the pore size of it was 0.45 μm. Elemental analysis was performed by the Analytical Center of the University of Science and Technology of China (Anhui, China). Mass spectra were recorded on a P-SIMS-Gly of Bruker Daltonics Inc. using ESI-TOF (electrospray ionization-time of flight). High-resolution mass spectrometry (HRMS) spectra were measured on a GCT premier CAB048 mass spectrometer operating in matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mode. See Supporting Information for details of preparation and characterization of ligands as well as the Pd complexes. Results and Discussion 3-(Aromatic imino)butan-2-one was prepared from the starting anilines following a literature procedure.36 Subsequently, condensation with diamino durene led to the formation of the aniline-functionalized α-diimines that were mixed with phenyl isocyanate to provide the desired α-diimine ligands ( L1– L4) in moderate yields (Scheme 2). The corresponding methylated ligands ( L1 Me– L4 Me) were synthesized in high yields through reactions between NaH and CH3I.37 Pd complexes ( Pd1– Pd4 and Pd1 Me– Pd4 Me) were obtained in 61–89% yields from the reaction of these ligands with one equivalent of (COD)PdMeCl (COD = 1,5-cyclooctadiene). The ligands ( L1– L4 and L1 Me– L4 Me) were characterized by 1H NMR, 13C NMR, and electrospray ionization mass spectrometry (ESI-MS), and the Pd catalysts were characterized using 1H NMR, 13C NMR, MALDI-MS, elemental analysis, and infrared (IR) spectroscopy. Scheme 2 | Synthesis of ligands and Pd complexes. Download figure Download PowerPoint Single crystals of the metal complexes were obtained by slow diffusion of MeOH into CH2Cl2 solutions. The molecular structures of Pd1 and Pd1 Me were determined by X-ray diffraction (Figures 1a and 1b). The Pd center adopted a spare-planar geometry. The distances and angles around the Pd center were typical of conventional α-diimine Pd complexes. Self-assembly through hydrogen bonding was observed from the crystal packing (Figure 1c). The self-assembled structure adopted a crossed "shoulder-to-shoulder" conformation with short NH–O distances (2.15 and 2.215 Å). The distance between two Pd centers was relatively long, but it may be smaller in solution due to bond rotation after the abstraction of chloride during polymerization. Fourier transform infrared (FTIR) spectroscopy provided evidence for the existence of urea–urea hydrogen bonding, since free N–H groups and hydrogen-bonded N–H groups have different frequencies. The FTIR spectra of solids Pd1– Pd4 at 25 °C revealed strong hydrogen-bonded NH stretching vibrations (3313 cm−1 for Pd1, 3296 cm−1 for Pd2, 3348 cm−1 for Pd3, and 3328 cm−1 for Pd4) compared with free N–H stretching vibrations (3500 cm−1) (see Supporting Information Figure S3).38 Figure 1 | X-ray crystallographic structures of (a) Pd1 and (b) Pd1 Me.a Ellipsoids were set at 30% probability. Selected bond lengths (Å) and angles (°) for Pd1 Pd1–N1 2.068 (5), Pd1–N2 2.099 (5), Pd1–C34 2.13 (5), Pd1–Cl1 2.257 (10), N1–Pd1–N2 77.3 (2), C34–Pd1–Cl1 88.1(15). Selected bond lengths (Å) and angles (°) for Pd1 Me: Pd1–N1 2.049 (3), Pd1–N2 2.146 (3), Pd1–C36 2.065 (4), Pd1–Cl1 2.2919(11), N1–Pd1–N2 77.75 (12), C36–Pd1–Cl1 88.73 (11). (c) Crystal packing to show the self-assembly of dimeric Pd complex Pd1. Download figure Download PowerPoint Catalyst activation An in situ activation procedure was used to study ethylene polymerization and ethylene/methyl acrylate (MA) copolymerization. The (co)polymerization was initiated with the addition of 1.2 equiv sodium tetrakis(3,5-bis(trifluoromethyl)phenyl)borate (NaBAF) to complexes Pd1– Pd4 and Pd1 Me– Pd4 Me, which abstracted a chloride to generate cationic Pd active species. These catalysts were all active in ethylene polymerization (Table 1), with parameters (activity, polymer molecular weight, etc.) comparable with previously reported α-diimine Pd catalysts. It is expected that the weakly coordinating and charge-dispersing perfluoroaryl counteranions39 will not interfere with the hydrogen bonding or assembly process. In the following sections, the factors that may affect the self-assembly and catalytic properties of the Pd complexes will be investigated (Scheme 3). Scheme 3 | Influence of different factors on the self-assembly process. Download figure Download PowerPoint Table 1 | Ethylene Polymerization Under Different Conditionsa Entry Catalyst Solvent Temperature (°C) Yieldb (g) Activityb Mnc, 104 PDIc Bd 1 Pd1 CH2Cl2/3 mL 25 1.20 0.40 12.6/0.41 1.16/1.26 111 2 Pd1 CH2Cl2/8 mL 25 1.72 0.57 10.7/0.37 1.88/1.16 117 3 Pd1 CH2Cl2/18 mL 25 1.75 0.58 9.60/0.26 1.36/1.01 121 4 Pd1 CH2Cl2/38 mL 25 1.93 0.64 3.48/0.27 1.16/1.12 130 5 Pd2 CH2Cl2/3 mL 25 0.72 0.24 15.5/2.37 1.20/1.12 99 6 Pd2 CH2Cl2/8 mL 25 0.74 0.25 14.3/2.35 1.22/1.08 100 7 Pd2 CH2Cl2/18 mL 25 0.67 0.22 14.0/2.11 1.19/1.13 104 8 Pd2 CH2Cl2/38 mL 25 0.57 0.19 12.2/1.71 1.11/1.26 106 9 Pd3 CH2Cl2/3 mL 25 1.44 0.48 16.2 1.26 94 10 Pd3 CH2Cl2/8 mL 25 1.67 0.56 17.6 1.27 96 11 Pd3 CH2Cl2/18 mL 25 2.17 0.72 20.9 1.30 98 12 Pd3 CH2Cl2/38 mL 25 3.12 1.04 22.8 1.34 100 13 Pd2 CH2Cl2/18 mL 0 0.72 0.24 9.10/2.77 1.07/1.08 88 14 Pd2 CH2Cl2/18 mL 20 1.19 0.40 10.6/2.47 1.10/1.11 94 15 Pd2 CH2Cl2/18 mL 40 2.12 0.71 13.9/1.80 1.15/1.27 97 16 Pd2 CH2Cl2/18 mL 55 2.55 0.85 21.4/1.73 1.13/1.26 97 17 Pd2 CH2Cl2/18 mL 60 3.11 1.04 0.89 1.67 109 18 Pd2 CHCl3/18 mL 25 0.084 0.028 2.28/0.47 1.16/1.11 105 19 Pd2 Et2O/18 mL 25 0.15 0.05 1.21/0.34 1.03/1.21 107 20 Pd2 Toluene/18 mL 25 0.93 0.31 7.07 1.38 98 21 Pd2 Hexane/18 mL 25 0.82 0.27 16.0 1.20 85 22 Pd1 Hexane/18 mL 25 4.92 1.64 17.9 1.64 100 aReaction conditions: precatalyst (10 μmol) in 2 mL CH2Cl2, NaBAF (1.2 equiv), 8 atm, 3 h. bThe overall activity (105 g mol−1 h−1) was determined from the mass of the polymer product and was the average of at least two runs. cMolecular weight and PDI were determined by GPC using polystyrene standards. dBranches per 1000 carbon atoms, determined by 1H NMR analysis. Influence of catalyst concentration Higher catalyst concentrations usually promote self-assembly, as supported by DOSY (diffusion ordered spectroscopy) NMR measurements40,41 of complexes Pd1 and Pd1 Me. Using the Einstein–Stokes equation,42 the hydrodynamic volume of Pd1 was calculated to be ca. 6.1 times that of Pd1 Me ( Supporting Information Figure S6). DLS measurements were used to study the assembly of the activated cationic species of Pd2 in solution ( Supporting Information Figure S2). At 2.0 mM, the average hydrodynamic diameter was 307 nm, which increased at higher concentrations, indicating the formation of larger aggregates.43 Polymerization was conducted using catalysts Pd1 and Pd2 at four different concentrations (Table 1, Entries 1–8; 2, 1, 0.5, and 0.25 mM) in CH2Cl2. The polyethylenes displayed bimodal GPC curves under all conditions. Interestingly, the ratio of the high-molecular-weight fraction to the low-molecular-weight fraction increased with increasing catalyst concentration ( Supporting Information Figure S1 and Figures 2a and 2b). At higher catalyst concentrations, more Pd complexes assembled/aggregated, which increased the steric environment around the Pd center. It is also possible that metal–metal cooperativity effects may be involved at high concentrations. Figure 2 | GPC curves of the polyethylene products generated under different conditions. (a) Polyethylenes generated by Pd2 at different catalyst concentrations. (b) Polyethylenes generated by Pd3 at different catalyst concentrations. (c) Polyethylenes generated by Pd2 at different polymerization temperatures. (d) Polyethylenes generated by Pd2 in different solvents. Download figure Download PowerPoint In the literature, numerous dinuclear olefin polymerization catalysts have been reported.21 Generally, dinuclear catalysts produce higher-molecular-weight polyolefins than their mononuclear counterparts due to ligand steric effects or metal–metal cooperativity effects. For α-diimine-type Pd catalysts, branched polyethylenes are generated during ethylene polymerization due to chain walking.44–48 Interestingly, dinuclear α-diimine Pd catalysts typically produce polyethylenes with lower branching densities than their mononuclear counterparts. This has been attributed to the inhibition of rotation about the N-aryl moieties in the ligand framework49 or to metal–metal cooperativity effects that slow β-hydride elimination and the corresponding chain-walking process.50–52 In this system, the molecular weight of high-molecular-weight fraction decreased as the catalyst concentration decreased from 12.6 × 104 to 3.48 × 104 for Pd1 and from 15.5 × 104 to 12.2 × 104 for Pd2 (Table 1, Entries 1–8). After increasing the catalyst concentration, the polymer branching densities decreased from 130 to 111/1000 C for Pd1 and from 106 to 99/1000 C for Pd2. These results may originate from metal–metal cooperativity effects or steric effects induced by self-assembly. Notably, the differences in both the molecular weight and polymer branching were much more dramatic for Pd1 than Pd2. It is possible that hydrogen bonding was stronger in Pd2 (as evidenced by the FTIR analysis in Supporting Information Figure S3: 3296 cm−1 for Pd2 vs 3313 cm−1 for Pd1), thereby making its assembly–disassembly process less sensitive to concentration than complex Pd1. For the methylated counterparts ( Pd1 Me and Pd2 Me), the polymer molecular weights and branching densities did not depend on the catalyst concentration ( Supporting Information Table S1, Entries 1–6). The use of Pd1 Me formed polyethylene with a molecular weight twice as high as that produced by Pd2 Me due to the greater steric bulk of Pd1 Me than Pd2 Me. However, Pd2 led to the formation of polyethylene with a much higher molecular weight than Pd1, likely due to the presence of two sets of urea moieties and correspondingly stronger self-assembly. These results strongly support the aforementioned mechanism of hydrogen-bonding-induced catalyst self-assembly. Despite their self-assembly feature, Pd1 and Pd2 generated polymers with higher branching density than their methylated counterparts. This is probably due to the presence of a significant amount of low-molecular-weight fraction and high branching fraction produced from the nonassembled section of Pd1 and Pd2. Influence of ligand steric effects It is expected that the assembly of these Pd complexes will be sensitive toward ligand steric effects; therefore, catalyst Pd3 with sterically bulky diphenyl-methyl substituents was prepared and studied. FTIR analysis showed that Pd3 possessed a much higher N–H stretching frequency ( Supporting Information Figure S3, σ = 3348 cm−1) than Pd1 and Pd2, indicating weaker hydrogen bonding. The bulky substituents inhibited the assembly/aggregation of Pd3. At high catalyst concentrations (Table 1, Entries 9 and 10; 2 and 1 mM), a very small amount of high-molecular-weight fraction was formed using Pd3. At low concentrations (Table 1, Entries 11 and 12; 0.5 and 0.25 mM), complex Pd3 produced polyethylenes with monomodal molecular weight distributions. These results are consistent with the hypothesis that sterically bulky substituents are detrimental to the self-assembly process. Similar to Pd1 Me and Pd2 Me, the methylated counterpart Pd3 Me led to the formation of monomodal polyethylenes at all concentrations ( Supporting Information Table S1, Entries 7–10). Influence of polymerization temperature Temperature can influence the hydrogen-bonding-based self-assembly, and high temperatures can weaken hydrogen bonding, which translates into changes in the catalyst properties (molecular weight distributions in this case). As expected, the proportion of the low-molecular-weight fraction increased when the temperature was increased from 0 to 60 °C. The high-molecular-weight fraction disappeared at 60 °C (Figure 2c and Table 1, Entries 13–17), indicating complete dissociation of the catalyst assembly. The C=O absorbance peak (ca. 280 nm) in the UV–Vis spectra of the cationic form of complex Pd2 in dimethyl sulfoxide (DMSO) weakened when the temperature increased from 10 to 70 °C ( Supporting Information Figure S4). This indicates that the hydrogen bonds were weakened, which is consistent with the polymerization results, showing that this provides a simple strategy to modulate the molecular weight distributions of polyethylenes. Influence of polymerization solvent The solvent polarity affects the self-assembly/aggregation state of these Pd species; therefore, ethylene polymerization was conducted in different solvents. Interestingly, bimodal polyethylenes were generated using Pd2 in polar solvents, including CH2Cl2, CHCl3, and Et2O (Figure 2d and Table 1, Entries 7, 18, and 19). In contrast, only high-molecular-weight monomodal polyethylenes were produced in nonpolar solvents such as toluene and hexanes (Table 1, Entries 20 and 21). The hydrogen bonding was stronger in nonpolar solvents, leading to catalyst self-assembly (the equilibrium shifted to the left in Scheme 3). Similar behavior was observed for ethylene polymerization catalyzed by Pd1 in hexanes (Table 1, Entry 22). In contrast, the methylated counterparts did not show solvent-dependent behavior during ethylene polymerization ( Supporting Information Table S1, Entries 11–15). Influence of photochemistry To further demonstrate the effect of supramolecular-induced self-assembly in this system and to explore other stimuli that influence self-assembly, we installed a light-sensitive moiety in the catalyst framework. Light can be used to noninvasively modulate polymerization processes.53–55 With the introduction of an azobenzene group, the catalyst structure and self-assembly can be modulated using irradiation, which will affect the catalyst properties and polymer properties. The UV–Vis absorption spectrum of Pd4 Me contained a maximum at ca. 345 nm ( Supporting Information Figure S5b) due to the π–π* transition of the azobenzene unit. After irradiation, the intensity of this peak decreased without affecting the shape of the overall curve, indicating the formation of the cis isomer. In contrast, the entire UV–Vis spectrum of Pd4 shifted after light irradiation ( Supporting Information Figure S5a), suggesting that trans–cis isomerization also affected the self-assembly of Pd4. UV irradiation completely changed the ethylene polymerization behavior using Pd4. Monomodal polyethylenes were generated by Pd4 in the dark, while bimodal polymers were obtained by Pd4 under UV irradiation (Table 2, Entries 1 and 2). Similar light-dependent properties were observed when the polymerization was carried out at 0 °C instead of 25 °C (Table 2, Entries 3 and 4). The light-induced trans–cis isomerization likely changed the self-assembled state of complex Pd4 and thereby affected the polymerization properties. In contrast, Pd4 Me showed almost the same properties (activity, polymer molecular weight, and branching density) during ethylene polymerization in the dark or under UV irradiation (Table 2, Entries 5–8). The trans–cis isomerization of Pd4 Me only slightly affected the ligand electronics without influencing the ligand sterics. This is consistent with previous studies that have shown that ligand electronics only slightly influence the properties of α-diimine Pd catalysts.56,57 Table 2 | Ethylene Polymerization Using the Azobenzene Functionalized Complex Pd4/Pd4Me and Ethylene-MA Copolymerization Studiesa Entry Catalyst MA (M) Yieldb (g) Activityb X (%) Mnc, 104 PDIc Bd Light 1 Pd4 — 1.12 3.73 — 6.99 1.64 99 Dark 2 Pd4 — 0.24 0.80 — 9.95/1.81 1.05/1.18 100 UV 3e Pd4 — 0.62 2.07 — 5.87 1.43 94 Dark 4e Pd4 — 0.44 1.47 — 11.0/2.36 1.09/1.10 95 UV 5 Pd4 Me — 1.38 4.60 — 11.7 1.56 94 Dark 6 Pd4 Me — 0.82 2.73 — 10.7 1.48 96 UV 7e Pd4 Me — 0.89 2.97 — 6.26 1.67 96 Dark 8e Pd4 Me — 0.78 2.60 — 6.42 1.67 95 UV 9 Pd1 1 0.63 2.10 3.2 2.17 1.40 97 — 10 Pd1 Me 1 1.13 3.77 0.4 3.31 1.55 102 — 11 Pd3 1 0.91 3.03 1.6 5.66 1.19 98 — 12 Pd3 Me 1 1.53 5.10 0.9 8.10 1.32 93 — aReaction conditions: precatalyst (10 μmol), NaBAF (1.2 equiv), CH2Cl2 (20 mL), 25 °C, 8 atm, 3 h. bThe overall activity (104 g mol−1 h−1) was determined from the mass of the polymer product and was the average of at least two runs. cMolecular weight and PDI were determined by GPC using polystyrene standards. dBranches per 1000 carbon atoms, determined by 1H NMR analysis. ePolymerization temperature: 0 °C. Copolymerization studies The incorporation of polar comonomers during olefin polymerization represents a simple and effective strategy to modulate important properties of polyolefin materials.58–62 α-Diimine Pd catalysts have been extensively studied for this purpose.63,64 The presence of large amounts of polar comonomers (MA for this study) would break the hydrogen-bonding-induced self-assembly. The utilization of other supramolecular strategies (metal–ligand coordination, for example) might be able to address this issue. Interestingly, Pd1/ Pd3 led to the formation of copolymers with much higher amounts of incorporated MA than their methylated counterparts Pd1 Me/ Pd3 Me(Table 2, Entries 9–12). This phenomenon is not fully understood and may originate from the high local MA concentration near the Pd center due to hydrogen-bonding interactions between the urea moiety and the MA comonomer. Conclusion α-Diimine Pd complexes bearing urea functional units were designed and synthesized, and their corresponding methylated counterparts were prepared for comparison. The experimental results suggested that these Pd complexes self-assembled through urea-based hydrogen-bonding interactions. During ethylene polymerization and copolymerization with MA, various important parameters (activity, polymer molecular weight, molecular weight distribution, branching density, comonomer incorporation, etc.) were modulated by changing the catalyst concentration, ligand steric effects, temperature, and polymerization solvent. To further demonstrate the effect of supramolecular-induced self-assembly in this system and to explore other stimuli that influence self-assembly, a light-sensitive azobenzene unit was installed in the catalyst framework. Light irradiation tuned the self-assembly state and the polymerization properties, demonstrating the ability to nonintrusively tune the molecular weight distributions through the temperature or light. This work provides a new avenue for designing high-performance olefin polymerization catalysts and provides new possibilities for tuning polymerization processes and polymer properties. Footnotes a CCDC 2000115 ( Pd1) and CCDC 2000114 ( Pd1Me) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre. Supporting Information Supporting Information is available. Conflict of Interest There is no conflict of interest to report. Acknowledgments This work was supported by the National