CS-Symmetric Pyridine(diimine) Iron Methyl Complexes for Catalytic [2+2] Cycloaddition and Hydrovinylation: Metallacycle Geometry Determines Selectivity

A series of CS-symmetric (aryl,alkyl)-substituted pyridine(dimine) iron methyl (CyARPDI)FeCH3 complexes have been prepared, characterized, and evaluated as precatalysts for the [2+2]-cycloaddition of butadiene and ethylene. Mixtures of vinylcyclobutane and (Z)-hexa-1,4-diene were observed in each case. By comparison, C2v-symmetric, arylated (PDI) iron catalysts are exclusively selective for reversible [2+2]-cycloaddition to yield vinylcyclobutane. The alteration in the chemoselectivity of the catalytic reaction was investigated through a combination of precatalyst stability studies, identification of catalytic resting state(s), and 2H and 13C isotopic labeling experiments. While replacement of an aryl-imine substituent with an N-alkyl group decreases the stability of the formally iron(0) dinitrogen and butadiene complexes, two diamagnetic metallacycles were identified as catalyst resting states. Deuterium labeling and NOESY/EXSY NMR experiments support 1,4-hexadiene arising from catalytic hydrovinylation involving reversible oxidative cyclization leading to accessible cis-metallacycle. Cyclobutane formation proceeds by irreversible C(sp3)–C(sp3) bond-forming reductive elimination from a trans-metallacycle. These studies provide key mechanistic understanding into the high selectivity of bis(arylated) pyridine(diimine) iron catalysts for [2+2]-cycloaddition, unique, thus far, to this class of iron catalysts.


I. General Considerations
All air-and moisture-sensitive manipulations were carried out using vacuum line, Schlenk and cannula techniques or in an MBraun inert atmosphere (nitrogen) dry box unless otherwise noted. All glassware was stored in a pre-heated oven prior to use. The solvents used for air-and moisture-sensitive manipulations were dried and deoxygenated using literature procedures. i Butadiene, butadiene-d6, ethylene, ethylene-d6 and 13 C-labeled ethylene were purchased in reagent grade from either Matheson or Aldrich. The butadiene was dry and deoxygenated by stirring vigorously with n-BuLi below 0 °C for 10 minutes, then vacuum transferred into a thick-walled glass vessel containing 4 Å molecular sieves.
Ethylene was stored over activated 4 Å molecular sieves for at least 24 hours before use.
Labelled gases were stored in thick-walled glass vessels and used as received. The following compounds were prepared according to literature procedures: vinylcyclobutane, ii Cy A Me PDIFeCl2, iii Cy A iPr PDIFeCl2, iii [( Me PDI)Fe(N2)]2(-N2). iv 1

Special details
Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.          ,7.93;N,9.72. Found: C,69.36;H,7.67;N,9.67.       The crude product 4-CH3 (0.232 g, 0.520 mmol, 97%) was obtained as a dark green solid.

Special details
Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Figure S25. Truncated 1 H NMR (400 MHz, C6D6) spectrum of the reaction of 5 with CD3OD.

Special details
Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.
The tube was sealed and heated to 60 °C for 24h after which time full conversion of 5 to ( Cy A Me PDI)2Fe was observed by 1 H NMR spectrum of the sample.

Special details
Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.

v. Preparation and analytic procedure for 3-(N2)2.
In a nitrogen-filled glovebox, 3-CH3 (50 mg, 0.116 mmol) was dissolved into hexanes (2 ml) and transferred to a 20 ml scintillation vial. The solution was maintained at -35 °C for two weeks after which the green solution turned deep red. The cold solution was transferred to a cold IR cell using a chilled glass pipette. The cell was brought out of the glovebox for IR measurement within 5 minutes. The temperature of the sample was estimated to not

III. Iron-catalyzed [2+2]-cycloaddition/hydrovinylation of ethylene and butadiene
i. Extended Optimization ii. Time-course of the Reaction Figure S34. Time course of the cycloaddition/hydrovinylation of butadiene and ethylene with 1-CH3 (5 mol%) as catalyst.  Table S10.   Characterization of the metallacycle fragment: In a nitrogen-filled glovebox, a solution of 1-CH3 (0.009 mg, 0.022 mmol) in 500mg of benzene-d6 was transferred to a J. Young tube.
The tube was sealed, removed from the glovebox and frozen in liquid dinitrogen. The headspace was evacuated and butadiene (0.44 mmol) followed by 13 C-enriched ethylene (0.44 mmol) were added by vacuum transfer via a calibrated bulb. The tube was sealed under static vacuum, thawed and mixed by inversion for 10 minutes at room temperature. 1H NMR spectrum was recorded at room temperature and reveals the formation of the title compounds in >99% overall yield and ethane as a byproduct. 13C{1H} and HSQC NMR spectrum of the was recorded at room temperature showing parallel conversion of the organic substrates to 13 C enriched vinylcyclobutane and (Z)-hexa-1,4-diene. Ethylene fragment proton and carbon chemical shifts were determined by analysis of the NMR spectra obtained and are reported and Table S11.  In a nitrogen-filled glovebox, a solution of 1-CH3 (0.009 mg, 0.022 mmol) in 500mg of benzene-d6 was transferred to a J. Young tube. The tube was sealed, removed from the glovebox and frozen in liquid dinitrogen. The head-space was evacuated and butadiene or butadiene-d6 (0.44 mmol) followed by ethylene (0.44 mmol) were added by vacuum transfer via a calibrated bulb. The tube was sealed under static vacuum, thawed, mixed by inversion for 10 minutes at room temperature and cooled down to 0 °C. 1H NMR spectrum was recorded at 0 °C and reveals the formation of the title compounds in >99% overall yield and ethane as a byproduct. 13C{1H} spectra were recorded at 0 °C. Butadiene fragment proton and carbon chemical shifts were determined by subtraction of the of the NMR spectra obtained with butadiene and butadiene-d6 and are reported and Table 11.  In a nitrogen-filled glovebox, a solution of 1-CH3 (0.009 mg, 0.022 mmol) in 500mg of toluene-d8 was transferred to a J. Young tube. The tube was sealed, removed from the glovebox and frozen in liquid dinitrogen. The head-space was evacuated and butadiene (0.44 mmol) followed by 13C enriched ethylene (0.44 mmol) were added by vacuum transfer via a calibrated bulb. The tube was sealed under static vacuum, thawed, mixed by inversion for 10 minutes at room temperature and cooled down to 0 °C. 1H NMR spectrum was recorded at 0 °C and reveals the formation of the title compounds in >99% overall yield and ethane as a byproduct. 13C{1H} and HBMC (constant13 = 15, 3 scans) spectra were recorded at 0 °C. Correlation signals between alpha and gamma-position of the metallacycle are found for 7a and 7b. Figure S43. 1 H-13 C HMBC (101 MHz, toluene-d8) spectrum of 7a and 7b generated with 13 C enriched ethylene. Correlation signals indicated in blue for 7a and red for 7b.

ii. Freeze-quenched Mössbauer Spectroscopy
In a nitrogen-filled glovebox, a solution of 1-CH3 (0.046 mg, 0.11 mmol) in 500mg of benzene-d6 was transferred to a J. Young tube. The tube was sealed, removed from the glovebox and frozen in liquid dinitrogen. The head-space was evacuated and butadiene (0.44 mmol) followed by ethylene (0.44 mmol) were added by vacuum transfer via a calibrated bulb. The tube was sealed under static vacuum, thawed and mixed by inversion for 10 minutes at room temperature. 1 H NMR spectrum was recorded to confirmed full conversion of 1-CH3 to 7a and 7b. The tube was entered in the glovebox and the content transferred to a chilled Mössbauer cell. Figure S44. Zero-field 57Fe Mössbauer (freeze-quench, 80 K) spectrum of 7a and 7b after 10 minutes of reaction.
The same sample was prepared and the reaction allowed to run for one hour at room temperature. The tube was entered in the glovebox and the content transferred to a chilled Mössbauer cell. Figure S45. Zero-field 57Fe Mössbauer (freeze-quench, 80 K) spectrum of 7a and 7b after 60 minutes of reaction.
iii. X-ray diffraction analysis for 7a.

iv. Variable Temperature 1 H NMR Spectroscopy
In a nitrogen-filled glovebox, a solution of 1-CH3 (0.009 mg, 0.022 mmol) in 500mg of toluene-d8 was transferred to a J. Young tube. The tube was sealed, removed from the glovebox and frozen in liquid dinitrogen. The head-space was evacuated and butadiene (0.44 mmol) followed by 13 C enriched ethylene (0.44 mmol) were added by vacuum transfer via a calibrated bulb. The tube was sealed under static vacuum, thawed, mixed by inversion for 10 minutes at room temperature and cooled down to -78 °C. The sample was analyzed by 1H NMR spectroscopy at -80 °C then by 10 °C increasing increments to 30 °C. 5-minute intervals were allowed between scans to equilibrate the solution temperature. The compiled data are reported in Figure S47. isotopologs of 1,4-HD, which were spectroscopically consistent with 1,4-HD-d0 and 1,4-HD-d4, were observed. Figure S48. 1 H NMR (400 MHz, benzene-d6) spectrum of (Z)-hexa-1,4-diene-d4. Assignment of the signals corresponding to (Z)-hexa-1,4-diene are given. Figure S49. 13 C NMR (101 MHz, benzene-d6) spectrum of (Z)-hexa-1,4-diene-d4. Assignment of the signals corresponding to (Z)-hexa-1,4-diene are given.   In order to record a GC-HRMS chromatogram, the same protocol was run in mesitylene as the solvent of the reaction. After 14h, the reaction was filtered over a plug of silica to remove traces of catalyst. Both reaction solutions, run with ethylene-d4 and an equimolar mixture of ethylene and ethylene-d4, were analyzed by GC-HRMS using mesitylene as solvent, with a solvent delay of 1 minute and switching off the detector at t = 5 minutes. pentane, dichloromethane, benzene, toluene or dodecane could not be used as reaction solvent or GC sample solvent as it would interfere with the retention time of C6 volatile fraction or its mass spectrum.
A distribution of the d1, d2, d3 isotopologues was observed while only the d4 isotopologues was expected when using only ethylene-d4. Analysis of the HRMS trace of the recovered PDI ligand following the completion of the reaction with ethylene-d4 gave a mixture of isotopologues, consistent with a background H/D scrambling process (see below). The distribution pattern experimentally observed is therefore a result of either: (i) a competitive (PDI) to olefin substrates ligand-to-ligand hydride transfer; (ii) a low resolution between fragmentation and isotopologues mass peaks at this mass range. The comparison between the GC-HRMS spectra of the two experiments rules out 1,4-hexadiene-d1, -d2 and -d3 isotologues formation due to a Cossee-Arlman mechanism. For better visualization, subtraction of HRMS spectrum (experimental x=0) to spectrum (experimental x=1) was provided in the main text of the paper. Figure S53. Crossover experiment isotopolog distributions expected for a catalyst system operating by either of two mechanistic possibilities, compared with experimental HRMS(EI) results. x corresponds to the ratio of ethylene-d0 to ethylene-d4.

ii. H/D Kinetic Isotopic Effect Measurement
In a nitrogen-filled glovebox, a solution of 1-CH3 (0.009 mg, 0.022 mmol) in 500 mg of benzene-d6 transferred to a J. Young tube. The tube was sealed, removed from the glovebox and frozen in liquid dinitrogen. The head-space was evacuated and butadiene (0.44 mmol) followed by ethylene (0.44 mmol) or ethylene-d4 (0.44 mmol) were added by vacuum transfer via a calibrated bulb. The tube was sealed under static vacuum, thawed and mixed by inversion at room temperature while 1H NMR spectra were recorded throughout the course of the reaction. Relative concentration of each products were determined by comparing 1H NMR integration of the products to butadiene. Each run was performed in duplicate.

iii. (PDI) to Substrates H/D Scrambling Evidence
At the end of the time course both reaction tubes (with ethylene-d0 and ethylene-d4) were opened and the solutions were exposed to air for 24h.

iii. 13 C-labeled ethylene experiment to study the C-C reductive elimination step
In a nitrogen-filled glovebox, a solution of [Fe] precatalyst (0.022 mmol) in 500mg of benzene-d6 was transferred to a J. Young tube. The tube was sealed, removed from the glovebox and frozen in liquid dinitrogen. The head-space was evacuated and butadiene