Kinetics, Mechanism and Theoretical Studies of Norbornene-Ethylene Alternating Copolymerization Catalyzed by Organopalladium(II) Complexes Bearing Hemilabile α-Amino–pyridine

Cationic methylpalladium complexes bearing hemilabile bidentate α-amino–pyridines can serve as effective precursors for catalytic alternating copolymerization of norbornene (N) and ethylene (E), under mild conditions. The norbornyl palladium complexes in the formula of {[RHNCH2(o-C6H4N)]Pd(C7H10Me)(NCMe)}(BF4) (R = iPr (2a), tBu (2b), Ph (2c), 2,6-Me2C6H3 (2d), 2,6-iPr2C6H3 (2e)) were synthesized via single insertion of norbornene into the corresponding methylpalladium complexes 1a–1e, respectively. Both square planar methyl and norbornyl palladium complexes exhibit facile equilibria of geometrical isomerization, via sterically-controlled amino decoordination–recoordination of amino–pyridine. Kinetic studies of E-insertion, N-insertion of complexes 1 and 2, and the geometric isomerization reactions have been examined by means of VT-NMR, and found in excellent agreement with the results estimated by DFT calculations. The more facile N-insertion in the cis-isomers, and ready geometric isomerization, cooperatively lead to a new mechanism that accounts for the novel catalytic formation of alternating COC.


General procedure for copolymerization of ethylene-norbornene
Into a 600 mL Parr autoclave equipped with a magnetic stirring bar was placed norbornene (0.5-10 g) in dried CH 2 Cl 2 (50 mL). The autoclave was sealed. Upon flush with ethylene gas several times, the ethylene gas was pressurized. The solution was stirred for 20 min in order to be saturated with ethylene gas. After release of ethylene pressure, the palladium complexes (0.06 mmol) were added, and then refilled with ethylene gas up to 21 bar. The mixture was stirred for 30 min, and the ethylene pressure was kept constant during the copolymerization runs. The reaction was quenched with venting the autoclave followed by adding 100 mL MeOH-HCl in 4:1 v/v ratio. The precipitated polymers were filtered from solution, washed with methanol and dried in vacuum oven at 80 o C overnight.

General procedures for the kinetic study determined by variable-temperature 1 H NMR spectroscopy VT-NMR Insertion Kinetics for Ethylene
An NMR tube was charged with 0.7 mL of [(N^N)Pd(Me) (NCMe)]BF4] solution (2.5 × 10 -2 ~ 1.1 × 10 -3 M) in CDCl3 and tetramethylsilane (TMS) as internal standard under nitrogen atmosphere, and then frozen with dry ice/acetone bath. After evacuation, desired amount of ethylene (4-8 mL) was injected by gastight syringe. The tube was warmed briefly around the melting point of CDCl 3 and shaken vigorously in order to adequately dissolve the ethylene. The sample was placed in a precooled NMR probe at desired temperature, and the decrease in intensity of Pd-Me was measured by using 1H NMR spectroscopy. Spectra were taken at intervals of 91-301 sec. The slope of the line defined by the Ln(integral of Pd-Me) signal vs time afforded the first-order rate constant, and the reaction was monitored for 2-4 half-lives.

Insertion Kinetics for Norbornene
An NMR tube was charged with 0.7 mL of [(N^N)Pd(Me) (NCMe)]BF 4 solution (2.5 × 10 -2 -1.1 × 10 -3 M) in CDCl 3 and tetramethylsilane (TMS) as internal standard, and then frozen with dry ice/acetone bath. Desired amount of norbornene (1.2 -104.0 mg) was added directly. The sample was warmed briefly around the melting point of CDCl 3 and mixed by an iron stirrer. The tube was placed in a precooled NMR probe at desired temperature, and the decrease in intensity of Pd-Me for cis-isomers and Py H-4 or Py H-6 for trans-isomers were measured by using 1 H NMR spectroscopy. Spectra were taken at intervals of 46 -301 sec. The slope of the line defined by the Ln(integral of Pd-Me, Py H-4 or Py H-6) signal vs time afforded the first-order rate constant, and the reaction was monitored for 2 -4 half-lives.

Isomerization Kinetics
In a typical run, a single crystal of C-2e (3 mg, 4.9 μmol) was ground into powder, and then placed into an NMR tube with dry ice/acetone bath. CDCl 3 (0.7 mL) was added slowly at -78 o C in dry ice/acetone bath. The sample was warmed briefly and mixed by an iron stirrer around the melting point of CDCl 3 . The tube was placed in a precooled NMR probe at desired temperature, and the decrease in intensity of Py H-6 for C-isomer was measured by using 1 H NMR spectroscopy. Spectra were taken at intervals of 46-301 sec. The slope of the line defined by the Ln(integral of Py H-6) signal vs time afforded the first-order rate constant, and the reaction was monitored for 2-4 half-lives.

X-ray crystallographic analysis
Diffraction data were measured on a Nonius CAD-4, SmartCCD, or Nonius KappaCCD diffractometer with graphitemonochromatized Mo K α radiation (λ = 0.7103 Å). No significant decay was observed during the data collection. The data were processed on a PC using the SHELXTL refinement software package. 1 The structures were solved using the direct method and refined by full-matrix least-squares on the F 2 value.
All the non-hydrogen atoms were refined anisotropically. Hydrogen atoms were identified by calculation and refined using a riding mode, and their contributions to structure factors were included. Atomic scattering factors were taken from the International Tables of Crystallographic Data, Vol IV. Computing programs are from the NRC VAX package. 2

Computational Details
All geometries of the reactants, intermediates, transition states and products of reactions were fully optimized by using the Gaussian 09 program, 3 and the B3LYP method with the LanL2DZ basis sets. No symmetry constrains were used for transition state optimization. 4-5 Saddle points were determined by relaxed potential energy surface (PES) scan for bond distance or dihedral angle. Harmonic vibration frequency calculations were performed on all stationary points to identify the local minimum (without imaginary frequency) or transition state (with one imaginary frequency), and determinate zero-point energies (ZPE). The relative energies (E + ZPE), obtained from vibrational frequency analyses on optimized structures by single point calculations, were corrected for basis set superposition error (BSSE) 6 using the counterpoise method. Intrinsic reaction coordinate (IRC) calculations 7 were also performed to confirm the connectivity between the transition state and the designed intermediates. The enantiomers are omitted for discussion, because they generally show nearly identical energies in the preliminary calculations.. The chirality of the coordinated amino group is fixed in R-form for the comparing purpose.

Fineman-Ross Plot
The copolymer compositions are determined according to the literature. 9 Where Pd-N-polymer and Pd-E-polymer are the growing chain with norbornene and ethylene as the last inserted monomer, and the reactivity ratios are defined as r1 = kNN/kNE and r2 = kEE/kEN. The r1 represents the ratio of rate constants between the successive norbornene insertions and norbornene insertion followed by ethylene, and the r 2 represents the ratio of rate constants between the successive ethylene insertions and ethylene insertion followed by norbornene. By using quasi-steadystate assumption for propagation, the copolymer composition equation can be derived as:  (3)(4)(5)(6)(7)(8)(9)(10)(11) According to the N/E feeding ratios and norbornene calculated from 13 C NMR spectra in Table S1, the values of F and f in Eq. (3)(4)(5)(6)(7)(8)(9)(10)(11) are collected in Table S2. The Fineman-Ross data of E-N copolymerization catalyzed by 1e give linear relationships between F 2 /f and F(f-1)/f as shown in Figure S1. Such data give the r 1 = k EE /k EN = 0.013, r 2 = k NN /k NE = 0.31 and r1×r2 = 0.004. The product of r1×r2 is significantly smaller than 1.0, indicating the consecutive hetero-olefin insertions are faster than the consecutive homo-olefin insertions. Figure S3. Fineman-Ross relationships for E-N copolymerization catalyzed by 1e.

Determination of norbornene content and alternating percentage for copolymers
The norbornene content,  NB was evluated with use of 13 C NMR integrations as designated in the following table according to Eq. (1) used by Kaminsky. 10 The alternating percentage in a copolymer may be calculated according to Eq. (1-2).

Van't Hoff Plot
Theoretical Calculation

C-T isomerization
The first possible C-T isomerization is the ligand association-substitution reaction, involving square pyramidal intermediates, trigonal bipyramidal and square pyramidal transition states (Scheme S1). 11 Decrease of the distance for an external ethylene and palladium denoted as 6 and 6', respectively, affords the transition states TS(6-6) and TS(6'-6') in trigonal bipyramidal geometries, which the energies lie 14.1 and 7.0 kcal/mol above the 6 and 6', respectively (Scheme S2). Subsequent decrease of the ethylene-palladium distance resulted in the dissociation of original ethylene. Our trials to locate a ground state for 6 and 6' with external ethylene in square pyramidal geometry resulted in a simple dissociation of ethylene. Relaxation of TS(6-6) and TS(6'-6') to the product sides give the 6 and 6', indicative of the ethylene exchange process. Such low ethylene exchange energies would result in the broadening signals in 1 H NMR spectra at low temperature, which are corresponded to the experimental observation. Therefore, the C-T isomerization via associative pathway has been ruled out.