Aqueous Keto-Polyethylene Dispersions from Catalytic Copolymerization of Ethylene and Carbon Monoxide in Water

Water-soluble [P,O]Ni(II) catalysts enable the direct catalytic nonalternating copolymerization of fundamental comonomers ethylene and carbon monoxide (CO) in water as an environmentally friendly reaction medium. This yields stable aqueous dispersions of high molecular weight polyethylene containing ∼1 mol % of largely isolated in-chain keto groups in the form of particles with sizes between 100 nm and 1 μm. The intermediate species of chain growth resulting from incorporation of polar comonomers are amenable to specific chain termination pathways in conjunction with water.


Materials and general methods
Unless noted otherwise, all manipulations of air and moisture sensitive materials were carried out under an inert gas atmosphere using standard glovebox and Schlenk techniques.Solvents were dried and degassed using standard laboratory techniques.Benzene was distilled from sodium, THF from sodium benzophenone ketyl.Pentane and toluene were dried over molecular sieves (3 Å) and degassed by passing through a MB-SPS-800 solvent purification system by MBRAUN.Deionized water for aqueous polymerizations was distilled under a constant stream of N2 to remove dissolved oxygen.Nickel precursor [(tmeda)NiMe2] was prepared according to literature procedures. 1 2tert-butyl-4-fluoro phenol 2 , 2-(4-fluoro phenoxy)tetrahydro-2H-pyran 3 and 2-(2-(tert-butyl)-4fluorophenoxy)tetrahydro-2H-pyran were prepared according to reported procedures. 3α-amino-ωmethoxy-polyethylene glycol (Mn 5516 g mol -1 ) was purchased from Iris Biotech.4-Fluoro phenol and sodium dodecyl sulfate (SDS) were purchased from Sigma Aldrich, and CsOH was supplied by ABCR.Ethylene of grade 3.5 and carbon monoxide of grade 4.7 were purchased from Air Liquide and used as received.Deuterated solvents were bought from Eurisotop.All other commercially available reagents were supplied by Sigma Aldrich, Acros, ABCR, or Activate Scientific.

Instruments and characterization
NMR-spectra were recorded on a Bruker Avance III 400, a Bruker Avance III HD or a JEOL ECZ 500R spectrometer. 1H chemical shifts were referenced to the solvent residual proton signals (C6D6: 7.16 ppm, CD3OD: 3.31 ppm, C2D2Cl4: 6.00 ppm) and 13 C chemical shifts were referenced to the carbon signals of the deuterated solvent (C6D6: 128.1 ppm, CD3OD: 49.0 ppm, C2D2Cl4: 73.8 ppm). 19F chemical shifts were referenced to external BF3•OEt2.Data evaluation was performed with MestreNova software by MestreLab SL.NMR spectra of polymers were acquired at 110 °C in 1,1,2,2-tertachlorethane-d2.Size exclusion chromatography (SEC) was performed on a PolymerChar GPC-IR instrument equipped with an integrated four-capillary viscometer and an IR5 dual wavelength infrared detector (selective for methylene and methyl groups) on PSS Polefin Linear XL columns (3 × 30 cm) and with an additional guard column at 160 °C and 0.5 mL min -1 in 1,2-dichlorobenzene.Universal calibration using narrow polystyrene standards was employed.The raw data was evaluated with PSS WinGPC UniChrom software.
Differential scanning calorimetry (DSC) was performed on a Netzsch DSC 204 F1 with a bicyclic temperature program and heating/cooling rates of 10 K min -1 .For the measurements, the polymers were weighed into sealed 40 μL aluminum pans.ATR-IR spectra of polymers were acquired on a Perkin Elmer Spectrum 100 instrument.Quantitative analysis of keto-contents was performed according to previously reported procedures. 4For calculation of the carbon monoxide incorporation, the ratio of the intensity of the C=O signal (peak ~ 1714 cm -1 ) to the intensity of the C-H signal of the PE at 2915 cm -1 was calculated and referenced with linear polyketone samples with known C=O content synthesized via ADMET copolymerization and subsequent hydrogenation (Figure S1). 5 Dynamic Light Scattering (DLS) measurement was performed on a diluted polyethylene dispersion using a Malvern Zetasizer Nano-ZS ZEN 3600 instrument (633 nm) in backscattering mode (173°) at 25 °C.Particle size distribution was analyzed using the Malvern Zetasizer Software, version 7.12.Transmission electron microscopy (TEM) images were acquired on a Zeiss Libra 120 EF-TEM instrument (120 kV).The respective samples were diluted to a solids content of ca.0.03 wt.% and dialyzed in a Spectrum Laboratories Spectra/Por Dialysis Membrane 1, MWCO 6000-8000 with deionized water for 14 days to remove free sodium dodecyl sulfate.The resulting dispersions were dropped onto a TEM copper grid and dried for 2 h.Keto-PE films were generated by dropcasting of copolymer dispersions (ca.0.05 to 0.5 wt.% polymer content) on a cleaned glass surface.Particle dispersions were first dialyzed to remove excessive SDS, which could disturb film formation.The samples were synthesized via ADMET copolymerization of docosa-1,21-dien-11-one and undeca-1,10-diene followed by hydrogenation. 5 The signal intensity used for calculation of ratios for carbonyl stretching vibrations (~ 1714 cm -1 ) vs. the polyethylene C-H vibration (2915 cm -1 ) are depicted in red.b: The intensity ratio is directly proportional to the concentration of C=O groups in the polymer χ.That is, χ ≈ nCO/nC2H4 which is a valid approximation if nC2H4 >> nCO.

General procedure for aqueous polymerizations
All polymerization experiments were carried out in a high-pressure polymerization set-up consisting of a 300 mL MiniClave by BüchiGlasUster equipped with a Cyclone 075 magnetically coupled mechanical pitched blade stirrer, in-and outlet valves, a temperature sensor, a digital pressure sensor, a Julabo CF41 thermostat and continuous (co-)monomer gas feeds.Gas flows of ethylene and carbon monoxide were individually monitored and regulated by El-Flow mass flow controllers by Bronkhorst.Gas feeds and temperature were monitored and controlled via LabVision software automatization by Hitec Zang.The reactor was evacuated and purged with nitrogen three times prior to the reaction while the internal temperature was > 70 °C.Sodium dodecyl sulfate (SDS) and CsOH were dissolved in 100 mL of deionized and deoxygenized water.For polymerizations which required low amounts of CsOH (pH 9.8 and 10.8) a respective stock solution was used instead of solid CsOH.For polymerizations under acidic conditions, KHSO4 was used to adjust the desired pH value instead of CsOH.90 mL of the obtained SDS/CsOH solution were transferred to the reactor vessel by cannula while 10 mL were used to dissolve the respective precatalysts.After heating to the desired reaction temperature, the precatalyst solution was added to the reactor by cannula transfer.The stirring rate was adjusted to 1000 rpm and the reactor was pressurized to 30 bar with the desired mixture of carbon monoxide and ethylene.This mixture was further fed to the reactor by the automated mass flow controllers to maintain a stable pressure (thus replenishing consumed gaseous monomer) over the course of the entire reaction.After the desired polymerization time, the reactor was vented and cooled.The obtained polymer dispersion was weighed and separated in two parts: 10-50 wt.% were filtered over cotton wool and used for dispersion and particle analysis.The remaining portion was precipitated in methanol (1000 mL).The precipitated polymers were filtered off, washed thoroughly with water and methanol and vacuum dried at 60 °C for > 24 h.Reference homopolymerizations of ethylene were carried out following the same protocol, without addition of CO to the gas feed.

Phosphinophenols [P,O]-1, [P,O]-2, [P,O] F -3c, [P,O] F -3d, [P,O] F -4c and [P,O] F -4d
9][10] General procedure for the synthesis of hydrophilic catalyst precursors The α-amino-PEG complexes were synthesized according to a procedure reported for salicylaldiminato complexes. 11A solution of the respective phosphinophenol (1.05 equiv.) in benzene (4 mL) was added to [(tmeda)NiMe2] (1.10 equiv.).During the addition, methane evolution was observed and the resulting orange solution was stirred at r.t. for 1 h.A solution of NH2-PEG-OMe (M= 5516 g mol -1 , 1.00 equiv.) in 1 mL of benzene was added and the reaction was stirred for further 3 h at r.t..The resulting orange solution was filtered via syringe filter to remove nickel black.The clear solution was vitrified by cooling the flask in liquid nitrogen and volatile compounds were removed by sublimation in vacuum.The obtained solid was washed with pentane (3 x 5 mL) and dried in vacuum to give the respective complex as yellow solid.Complexes 3 SO3 -pyr and 4 SO3 -pyr with pyridine as labile ligand were synthesized according to a reported procedure. 10

Figure S24
. 1 H NMR spectrum (400 MHz, 383 K, C2D2Cl4) of a keto-PE obtained from aqueous copolymerization at pH 12.8 employing complex 1-NH2PEG (Table S2, entry 7).S3, entry 3).Note the occurrence of small extents of backbone oxidation and chlorination 6 by the solvent due to the long high-temperature measurement.
Thermal properties from differential scanning calorimetry (DSC) . Representative differential scanning calorimetry (DSC) traces (second heating, 10 K min -1 ) of selected keto-PE polymers obtained from aqueous copolymerization with different catalysts and under different conditions (cf.Table S2).

Figure S1 .
Figure S1.Referencing of IR spectra.a: Polyketones with known carbonyl contents were analyzed by ATR-IR.The samples were synthesized via ADMET copolymerization of docosa-1,21-dien-11-one and undeca-1,10-diene followed by hydrogenation.5The signal intensity used for calculation of ratios for carbonyl stretching vibrations (~ 1714 cm -1 ) vs. the polyethylene C-H vibration (2915 cm -1 ) are depicted in red.b: The intensity ratio is directly proportional to the concentration of C=O groups in the polymer χ.That is, χ ≈ nCO/nC2H4 which is a valid approximation if nC2H4 >> nCO.

Figure S10. a :
Figure S10.a: Inhibitory effect of the presence of CO in the monomer feed on copolymerization yield (TableS1).b: Comparison of C=O incorporation ratios (blue) and yields (black) in dependence of pH.Copolymers obtained from polymerization employing complex 1-NH2PEG (TableS2, entries 2 -7).

Figure S13 .
Figure S13.ATR-IR spectra (left) with details of carbonyl region (right, 1675 -1775 cm -1 ) of keto-PEs obtained from aqueous nonalternating copolymerization of ethylene and CO with catalyst precursor 3-NH2PEG at different pH values (TableS2, entries 12 -16).Note that different ratios of isolated C=O and more alternating C=O motifs are indicated by broadening or a shoulder of the C=O absorption band.

Figure S14 .
Figure S14.ATR-IR spectra (left) with details of carbonyl region (right, 1675 -1775 cm -1 ) of keto-PEs obtained from aqueous nonalternating copolymerization of ethylene and CO with catalyst precursor 3 SO3 -pyr at different pH values (TableS2, entries 26 -30).Note that different ratios of isolated C=O and more alternating C=O motifs are indicated by broadening or a shoulder of the C=O absorption band.

Figure S19 .
Figure S19.ATR-IR spectra (left) with details of carbonyl region (right, 1675 -1775 cm -1 ) of keto-PEs obtained from aqueous nonalternating copolymerization of ethylene and CO with catalyst precursor 4 SO3 -pyr at high pH values (TableS2, entries 29 and 30).Note that different ratios of isolated C=O and more alternating C=O motifs are indicated by broadening or a shoulder of the C=O absorption band.

Figure S32. 1 H
Figure S32.1 H NMR spectrum (400 MHz, 383 K, C2D2Cl4) of a PE homopolymer obtained from aqueous polymerization employing complex 1-NH2PEG (TableS3, entry 3).Note the occurrence of small extents of backbone oxidation and chlorination6 by the solvent due to the long high-temperature measurement.

Table S1 .
Results of catalytic ethylene/CO copolymerizations in aqueous reaction media at different ethylene-CO feed ratio.

Table S2 .
Results of aqueous ethylene/CO copolymerizations at different pH values of the reaction medium.Note that entries 1 -7 are identical to Table1in the manuscript.
* Determined by SEC in 1,2-dichlorobenzene at 160 °C, 0.5 mL min -1 via universal calibration versus narrow polystyrene standards.** Determined by DSC (10 K min -1 ), second heating cycle.‡ ‡ Determined by dynamic light scattering.X Approximate CO content obtained from ATR-IR spectroscopy.Note that the employed calibration method is not accurate at very high CO incorporation ratios.

Table S4 .
Ratio of α,β-unsaturated carbonyl endgroups formed by deprotonation of carbonyl intermediates vs. unfunctionalized olefinic endgroups from chain transfer by β-H elimination in dependence of the pH/OH -concentration.