A protocol for the high-throughput analysis of polymerization catalyst, chain transfer polymerizations, polyethylene characterization, and reaction kinetic analysis is presented.
We demonstrate a method for high-throughput catalyst screening using a parallel pressure reactor starting from the initial synthesis of a nickel α-diimine ethylene polymerization catalyst. Initial polymerizations with the catalyst lead to optimized reaction conditions, including catalyst concentration, ethylene pressure and reaction time. Using gas-uptake data for these reactions, a procedure to calculate the initial rate of propagation (kp) is presented. Using the optimized conditions, the ability of the nickel α-diimine polymerization catalyst to undergo chain transfer with diethylzinc (ZnEt2) during ethylene polymerization was investigated. A procedure to assess the ability of the catalyst to undergo chain transfer (from molecular weight and 13C NMR data), calculate the degree of chain transfer, and calculate chain transfer rates (ke) is presented.
Polyolefins are an important class of industrial polymers with uses in thermoplastics and elastomers. Significant advances in the design of single-site catalysts for the production of polyolefins has led to the ability to tune molecular weight, polydispersity, and polymer microstructure, which leads to a wide range of potential applications.1-3 More recently, chain transfer and chain shuttling polymerizations have been developed to give an additional route to modify the properties of the polymer without having to modify the catalyst.4-6 This system employs a single-site transition metal catalyst and a chain transfer reagent (CTR), which is typically a main group metal alkyl. During this polymerization, the growing polymer chain is able to transfer from the catalyst to the CTR, where the polymer chain remains dormant until it is transferred back to the catalyst. Meanwhile, the alkyl group that was transferred to the catalyst can initiate another polymer chain. In a chain transfer polymerization, one catalyst can initiate a greater number of chains compared to a standard catalytic polymerization. The polymer chains are terminated with the chain transfer metal; therefore further end-group functionalization is possible. This system can be used to change the molecular weight and molecular weight distribution of polyolefins,7 to catalyze Aufbau-like alkyl chain growth on main group metals,8 and for the synthesis of specialty polymers involving multicatalyst systems, such as block copolymers.9,10
Chain transfer polymerizations have been observed most commonly with early transition metals (Hf, Zr) and alkylzinc or alkylaluminum reagents, although examples exist across the transition metal series.5,7,8,11-16 In typical early transition metal catalyst systems, chain transfer is fast, efficient and reversible leading to narrow molecular weight distributions. Chain transfer/shuttling has been observed in mid-to-late transition metals (e.g. Cr, Fe, Co and Ni) with group 2 and 12 metal alkyls, although the rates of transfer are highly variable compared to early metals.4,7,17-19 Two main factors are apparently necessary for efficient chain transfer: a good match of metal-carbon bond dissociation energies for the polymerization catalyst and chain transfer reagent, and an appropriate steric environment to promote bimolecular formation/breakage of alkyl-bridged bimetallic intermediates.20 In the case of late transition metals, if the catalyst does not contain enough steric bulk, beta-hydride (β-H) elimination will be the dominant termination pathway and will generally out-compete chain transfer.
Herein we report on a study of bimetallic chain transfer from nickel to zinc in a bis(2,6-dimethylphenyl)-2,3-butanediimine-based catalyst system with diethylzinc (ZnEt2) through small-scale high-throughput reactions. Chain transfer will be identified by examining changes in the molecular weight (Mw) and dispersity index of the resulting polyethylene through gel-permeation chromatography analysis. Chain transfer will also be identified through 13C NMR analysis of the ratio of vinyl to saturated chain ends as a function of chain transfer agent concentration. An in-depth kinetic analysis of the rates of propagation and chain transfer will also be presented.
Caution: Please consult all relevant material safety data sheets (MSDS) before use. Several of the chemicals used in these syntheses are acutely toxic and carcinogenic, while several are pyrophoric and ignite in air. Please use all appropriate safety practices when performing these reactions including the use of engineering controls (fume hood, glovebox) and personal protective equipment (safety glasses, gloves, lab coat, full length pants, closed-toe shoes). Portions of the following procedures involve standard air-free handling techniques.
1. Preparation of [bis(2,6-dimethylphenyl)-2,3-butanediimine]NiBr2,21-25
2. Preparation of Catalytic Stock Solutions
3. Catalytic Polymerizations Using a Parallel Pressure Reactor
Reaction Vessel | Pressure (psi) | Catalyst Vol. (ml) | ZnEt2 Vol. (ml) | Toluene Vol. (ml) |
1 | 15 | 0 | 0 | 3 |
2 | 15 | 0.1 | 0 | 2.9 |
3 | 30 | 0 | 0 | 3 |
4 | 30 | 0.1 | 0 | 2.9 |
5 | 60 | 0 | 0 | 3 |
6 | 60 | 0.1 | 0 | 2.9 |
7 | 150 | 0 | 0 | 3 |
8 | 150 | 0.1 | 0 | 2.9 |
Reaction Vessel | Pressure (psi) | Catalyst Vol. (ml) | ZnEt2 Vol. (ml) | Toluene Vol. (ml) |
1 | 60 | 0.1 | 0 | 2.9 |
2 | 60 | 0.1 | 0.005 | 2.9 |
3 | 60 | 0.1 | 0.01 | 2.89 |
4 | 60 | 0.1 | 0.015 | 2.89 |
5 | 60 | 0.1 | 0.025 | 2.88 |
6 | 60 | 0.1 | 0.042 | 2.86 |
7 | 60 | 0.1 | 0.06 | 2.84 |
8 | 60 | 0.1 | 0.085 | 2.82 |
4. Kinetic Analysis of Polymerizations: Rates of Chain Transfer and Propagation
The ethylene gas consumption versus time is presented in Figure 1 for the different ethylene pressures tested. This data is used to determine optimized reaction conditions. The ethylene gas consumption versus time is presented in Figure 2A for the catalyst alone samples, which is used to calculate the rate of propagation (kp). Figure 2B shows gel permeation chromatography (GPC) traces for chain transfer polymerizations with 0-1,000 equivalents of diethylzinc. The GPC is used to calculate the molecular weight (Mn) and dispersity (Đ) of the polymer samples, which is presented in Table 1. Figure 3 shows 13C NMR of the polyethylene samples, with 3A showing the spectra of the full series and 3B showing a zoomed in spectrum with the peaks labeled. The molecular weight data is used to calculate the number of chains initiated (Figure 4A and Table 1) and the Mayo plot (Figure 4B). The fit of the Mayo plot is used to calculate the ratio of the rate of chain transfer to the rate of propagation (ke/kp), which is used to calculate the rate of chain transfer (ke).
Figure 1: Ethylene consumption versus time at selected pressures.
Figure 2: (A) Ethylene consumption versus time for catalyst [α-diimine]NiBr2 at 60 psi ethylene. The slope of the linear region was used to calculate kp. (B) GPC traces of polymer obtained from catalyst [α-diimine]NiBr2 activated with MAO in the presence of ZnEt2 (0-1,020 eq.) in 1,2,4-trichlorobenzene at 135 °C. Figure adapted from reference 20.
Figure 3: (A) 13C NMR in C2D2Cl4 at 130 °C of polyethylene from [α-diimine]NiBr2 activated with MAO. The concentration of ZnEt2 increases from bottom to top. (B) 13C NMR in C2D2Cl4 at 130 °C of polyethylene from [α-diimine]NiBr2 activated with MAO with 1,020 eq. ZnEt2. Polyethylene peak assignments showing saturated end groups labeled Sx.28 Figure adapted from reference 20. Please click here to view a larger version of this figure.
Figure 4: (A) Average polymer chains initiated per nickel catalyst versus the amount of ZnEt2 for [α-diimine]NiBr2 over 3 runs. The error bars represent the standard deviation. (B) Mayo plot of the catalyst ligand steric effects on ethylene polymerization with [α-diimine]NiBr2 and ZnEt2, and calculations of ke/kp and ke. Figure adapted from reference 20.
Entry | Equiv. ZnEt2a | Yield (g) | Activity (g * mol-1 * hr-1 x 10-5) | Mn (x 10-5)b |
Đ | Mol (Zn-R)ext (x 107)c |
Chains/Nid |
1 | 0 | 0.199 | 19.9 | 1.52 | 2.37 | – | 13.1 |
2 | 60 | 0.18 | 18 | 1.31 | 2.56 | 13.8 | 13.8 |
3 | 120 | 0.299 | 29.9 | 1.12 | 2.41 | 26.7 | 26.7 |
4 | 180 | 0.216 | 21.6 | 0.953 | 2.46 | 22.7 | 22.7 |
5 | 300 | 0.178 | 17.8 | 0.689 | 2.39 | 25.8 | 25.8 |
6 | 500 | 0.189 | 18.9 | 0.506 | 2.17 | 37.2 | 37.2 |
7 | 720 | 0.179 | 17.9 | 0.406 | 2.08 | 44.1 | 44.1 |
8 | 1,020 | 0.268 | 26.8 | 0.278 | 2.16 | 96.4 | 96.4 |
Table 3: Data for ethylene polymerizations with the [α-diimine]NiBr2 catalyst and ZnEt2. All values are the average of at least 3 runs. Conditions: 1 x 10-7 moles catalyst, 500 equivalents of MAO, 60 psi ethylene, room temperature, 1 hr, toluene solvent (3.0 ml). aEquivalents of ZnEt2 based on the amount of catalyst. bDetermined by GPC. cDefined as the number of ethyl groups that has been extended with ethylene, determined by GPC. dThe number of chains initiated per total molar amount of polymerization catalyst.
A methyl-substituted cationic [α-diimine]NiBr2 ethylene polymerization catalyst activated with MAO was examined for its competency for ethylene chain transfer polymerizations. The reactions were monitored via gas uptake measurements to determine the rate and extent of polymerization and catalyst lifetime, and the molecular weight of the resultant polymers were determined via gel permeation chromatography (GPC). Initially, the nickel catalyst was tested over a range of ethylene pressures (from 15-225 psi) to determine optimal conditions for this system in the absence of ZnEt2. Using a 3.33 x 10-5 M reaction solution of catalyst (0.1 ml of the catalyst stock solution and 3.0 total reaction volume), a pressure of 60 psi of ethylene was found to produce an optimal combination of catalyst lifetime and ethylene consumption. It is important that the catalyst is active for the entire polymerization while also producing significant amounts of polyethylene for analysis. Figure 1 shows the ethylene consumption versus time at 4 selected pressures of 15, 30, 60 and 150 psi. At pressures of 15 and 30 psi, the ethylene consumption is linear over the time period tested, indicating an active catalyst, however small amounts of polymer are produced, which could complicate characterization. At 150 psi, the initial gas consumption is linear, and a significant amount of polymer is produced. However, it reaches a saturation point 25 min into the run, indicating an inactive catalyst. At 60 psi, the gas consumption is linear and a significant amount of polymer is produced allowing for easy isolation and characterization of the sample. Using the optimal conditions at 60 psi, multiple polymerizations were attempted to probe the reproducibility of the system in addition to the initial rate of propagation (kp). Figure 2A shows the ethylene uptake for four runs under identical conditions. From the initial slope (from 500 to 2,000 sec), kp was calculated to be 0.00319 M-1 sec-1.
Chain transfer polymerizations were tested in the presence of excess molar equivalents (0 to 1,020 equiv.) of ZnEt2 holding all other conditions constant. In the cases where polymerization was successful, catalyst systems that underwent chain transfer were identified by a reduction of Mn as the concentration of main group ZnEt2 increased. Figure 2B shows the GPC traces moving to longer retention times as a function of increasing [ZnEt2], where the longer retention times indicate a polymer with a lower molecular weight. Table 3 also lists the Mn of the produced polyethylene indicating successful chain transfer from Ni to Zn. The polymerization activity and molecular weight dispersity did not systematically vary across all [ZnEt2] (Table 3). In addition, the polymer microstructure determined from the 13C NMR in Figure 3A remains constant over the range [ZnEt2]. Figure 3B shows only saturated chain ends are present (marked Sx) indicating chains terminated on the Zn. Vinyl chain ends (at 111 and 136 ppm) would be expected from chain termination on the Ni due to β-H elimination, however these peaks are not present in this sample.20, 28 This data proves successful chain transfer from Ni to Zn and disproves competing side reactions involving the catalyst and chain transfer reagent that could alter the resulting polymer. Analysis of the polymer microstructure is also important because a change in the polymer microstructure could be indicative of a change in the nature of the catalyst species, which would make kinetic comparison impossible.
The amount of chain transfer can be quantified in two ways: first, as the amount (mol) of main group metal alkyl groups extended (equation 1),7 and second, by examining the number of chains initiated per total molar amount of polymerization catalyst (equation 2). Figure 4A shows a plot of the number of chains initiated versus the concentration of ZnEt2. This catalyst systems shows a linear dependence on the number of polymer chains generated per Ni and the [ZnEt2]. This type of plot can be used to compare the degree of chain transfer across different catalyst systems to determine ideal matches; for example multiple catalysts and the same chain transfer agent or multiple chain transfer agents and a single catalyst.
Mayo plot analysis of the catalytic system can reveal the relative rates of chain transfer with Zn (ke) to propagation (kp) based on the Mayo equation (equation 3). Fitting Figure 4B reveals ke/kp = 0.00355 for [α-diimine]NiBr2. The absolute rates of chain transfer (ke) in these systems, calculated from kp in the absence of ZnEt2, yields ke = 1.14 x 10-5 M-1 sec-1 ± 4.42 x 10-7. Using this type of kinetic analysis, it is straightforward to determine how the system is affected by addition of chain transfer agent, whether it is the rate of propagation, chain transfer or both. However, Mayo plot kinetic analysis is limited to systems that undergo chain transfer polymerizations.
We have demonstrated the ability of a [α-diimine]NiBr2 polymerization catalyst to undergo chain transfer with metal alkyls during ethylene polymerization. Although chain transfer from Ni to Zn is slower than many previously reported systems, using high-throughput screening and straightforward kinetic analysis it will be possible to study a large number of late transition metals, ligand frameworks and metal combinations in a short amount of time. With the data obtained from high-throughput chain transfer polymerizations and prudent catalyst system design, it will be possible to finely tune and exploit chain transfer with late transition metal complexes to produce new polymers, efficient routes to common polymers and specialty block copolymers.
The authors have nothing to disclose.
Financial support was provided by the University of Minnesota (start up funds) and the ACS Petroleum Research Fund (54225-DNI3). Equipment purchases for the Chemistry Department NMR facility were supported through a grant from the NIH (S10OD011952) with matching funds from the University of Minnesota. We acknowledge the Minnesota NMR Center for high-temperature NMR. Funding for NMR instrumentation was provided by the Office of the Vice President for Research, the Medical School, the College of Biological Science, NIH, NSF, and the Minnesota Medical Foundation. We thank John Walzer (ExxonMobil) for a gift of PEEK high-throughput stirring paddles.
Endeavor Pressure Reactor | Biotage | EDV-1N-L | |
Blade Impellers | Biotage | 900543 | |
Glass Liners | Biotage | 900676 | |
2,3-butanedione, 99% | Alfa Aesar | A14217 | |
2,6-dimethylaniline, 99% | Sigma Aldrich | D146005 | |
formic acid, 95% | Sigma Aldrich | F0507 | |
methanol, 99.8% | Sigma Aldrich | 179337 | ACS Reagent |
nickel (II) bromide, 99% | Strem | 28-1140 | anhydrous, hygroscopic |
triethylorthoformate, 98% | Sigma Aldrich | 304050 | dried with K2CO3 and distilled |
1,2-dimethoxyethane, 99.5% | Sigma Aldrich | 259527 | dried with Na/Benzophenone and distilled |
pentane, 99% | Fisher | P399 | HPLC Grade * |
dichloromethane, 99.5% | Fisher | D37 | ACS Reagent * |
toluene, 99.8% | Fisher | T290 | HPLC Grade * |
methylaluminoxane | Albemarle | MAO | pyrophoric, 30% in toluene |
diethylzinc, 95% | Strem | 93-3030 | pyrophoric |
1,2,4-trichlorobenzene, 99% | Sigma Aldrich | 296104 | |
1,1,2,2-tetrachloroethane-D2, 99.6% | Cambridge Isotopes | DLM-35 |