A multi-substrate screening approach for the identification of a broadly applicable Diels–Alder catalyst

When developing a synthetic methodology, chemists generally optimize a single substrate and then explore the substrate scope of their method. This approach has led to innumerable and widely-used chemical reactions. However, it frequently provides methods that only work on model substrate-like compounds. Perhaps worse, reaction conditions that would enable the conversion of other substrates may be missed. We now show that a different approach, originally proposed by Kagan, in which a collection of structurally distinct substrates are evaluated in a single reaction vessel, can not only provide information on the substrate scope at a much earlier stage in methodology development, but even lead to a broadly applicable synthetic methodology. Using this multi-substrate screening approach, we have identified an efficient and stereoselective imidodiphosphorimidate organocatalyst for scalable Diels–Alder reactions of cyclopentadiene with different classes of α,β-unsaturated aldehydes.


Procedure for aldehydes 1h-1j
A screw-cap vial was charged with aldehyde (1.0 equiv), dimethylamine hydrochloride (1.2 equiv) and 37 % aq formaldehyde solution (1.2 equiv). Upon heating the heterogeneous mixture under air at 70 °C for 1 h, the formation of a clear, homogenous layer was observed.
Prolonged heating at 70°C for 12 to 15 h again resulted in the separation in a biphasic mixture.
After allowing the mixture to cool to ambient temperature, water was added and the phases were separated. The aqueous layer was extracted with CH2Cl2 (× 3), and the combined organic layers were washed with brine (× 1), dried (Na2SO4) and concentrated under reduced pressure. The crude residue was purified via column chromatography on silica gel (n-pentane : Et2O = 90 : 10) to furnish the desired α-methylenated aldehyde.

[Method A]
To a solution of aldehyde 1 (1.0 mmol) in anhydrous toluene (1.0 mL) was added diene (5.0 mmol). The resulting mixture was stirred at 100 °C for 12 h. Purification was performed by column chromatography or preparative thin layer chromatography on silica gel using 5-10% Et2O/n-pentane as the eluent.
Purification was performed by column chromatography or preparative thin layer chromatography on silica gel using 5-10% Et2O/n-pentane as the eluent.

Procedure of the Multi-Substrate Screening
To a vial with a catalyst, immersed in a dry ice-acetone bath at -78 °C , were added a solution of aldehydes (1a-1f or 1a-1d) and cyclopentadiene 2. The resulting mixture was stirred at the described temperature for described reaction times. After addition of NEt3 (30 µL), the reaction mixture was warmed to ambient temperature. Conversions were determined by 1 H NMR analysis by comparison to p-nitrobenzaldehyde as an internal standard. The diastereomeric and enantiomeric ratios were measured by GC analysis on a chiral stationary phase.

Supplementary Table 1.
Effects of catalysts: acidity. a : a. Unless otherwise indicated, reactions were performed with aldehydes (0.02 mmol each, total 0.12 mmol), diene 2, and the corresponding catalyst (1.2 μmol) in CH2Cl2 (120 μL) for 24 h. b. Conversions were determined by 1 H NMR analysis by comparison to p-nitrobenzaldehyde as an internal standard after addition of TEA. c. The diastereomeric and enantiomeric ratios were measured by GC analysis on a chiral stationary phase.

Supplementary Table 2.
Effects of catalysts: substituent. a : a. Unless otherwise indicated, reactions were performed with aldehydes (1a1f, 0.02 mmol each, total 0.12 mmol), diene 2, and catalyst 4 (1.2 μmol) in CH2Cl2 (120 μL) at 78 °C for 24 h. b. Conversions were determined by 1 H NMR analysis by comparison to p-nitrobenzaldehyde as an internal standard after addition of TEA. c. The diastereomeric and enantiomeric ratios were measured by GC analysis on a chiral stationary phase.

Supplementary Table 3.
Effects of solvents. a : a. Unless otherwise indicated, reactions were performed with aldehydes (0.02 mmol each, total 0.12 mmol), diene 2, and catalyst 4h (1.2 μmol) in a solvent (120μL) at 78 °C for 24 h. b. Conversions were determined by 1 H NMR analysis by comparison to p-nitrobenzaldehyde as an internal standard after addition of TEA. c. The diastereomeric and enantiomeric ratios were measured by GC analysis on a chiral stationary phase.

Supplementary Table 4.
Effects of catalyst loadings, diene equivalents, concentrations. a : a. Unless otherwise indicated, reactions were performed with aldehydes (0.02 mmol each, total 0.12 mmol), diene 2, and catalyst 4h in CH2Cl2at 78 °C for 24 h. b. Conversions were determined by 1 H NMR analysis by comparison to p-nitrobenzaldehyde as an internal standard after addition of TEA. c. The diastereomeric and enantiomeric ratios were measured by GC analysis on a chiral stationary phase.

Supplementary Table 5.
Effects of temperatures. a : a. Unless otherwise indicated, reactions were performed with aldehydes (0.02 mmol each, total 0.12 mmol), diene 2, and catalyst 4h or 4i (1.2 μmol) in CH2Cl2 (120μL). b. Conversions were determined by 1 H NMR analysis by comparison to p-nitrobenzaldehyde as an internal standard after addition of TEA. c. The diastereomeric and enantiomeric ratios were measured by GC analysis on a chiral stationary phase.
To a solution of the catalyst in anhydrous CH2Cl2, immersed in a liquid nitrogen-ethanol slush at -116 °C, were added aldehyde 1 (1.0 equiv) and diene 2 (5.0 equiv). The resulting mixture was stirred at the described temperature. Upon completion of the reaction, NEt3 (50 µL) was added and the reaction mixture was warmed to ambient temperature. After removal of the solvent, the crude mixture was dry loaded onto Celite ® and purified by column chromatography on silica gel to afford product 3.

Gram Scale Reactions
To a flame-dried Schlenk tube, charged with aldehyde 1b ( [15][16][17] The topology of the molecule was generated from the starting structure (CCDC code SEFWAE 2 ) using the CGenFF ParaChem server (https://cgenff.paramchem.org/, interface v. 1.0.0, force field v. 3.0.1), and then modified manually to ensure the structure remained consistent with crystallography data (CCDC code SEFWAE). Specifically, the central P-N-P angle was set to 170° (vs. 135° in the original Parachem topology), the N-P-N-P and P-N-P-N dihedral angles were set to 180° (by fixing the improper angle defined by N-P(-N)-P-N to 180°), and the corresponding P-N and N-P bond distances were set to 1.50 Å (vs. 1.79 Å ). Furthermore, the core of the molecule was stabilized by specifically defining an H-bond between H0 and O52 and setting the corresponding S-N-H-P improper dihedral angle to 0°.
The molecule was placed in a box of 396 trichloromethane molecules, 18 ensuring a minimum distance of 10 Å between of any IDPi atom and the edge of the box. All bonds were constrained using the LINCS method 19 and the timestep was 2 fs. The Verlet cutoff scheme 20 was used for the neighbor lists (10 Å short-range cutoffs for both Coulomb and van der Waals forces) and the particle mesh Ewald method 21,22 was employed for electrostatics. The contact information from the measured nuclear Overhauser effects (NOEs) was included as simple bond-type restraints (i.e. time averaging was not considered).
The system was first energy-minimized by steepest descent to a maximum force of 10 kJ•mol −1 •nm −1 , then equilibrated, first in the isothermal ensemble (NVT) at 300 K for 100 ps, then in the isothermal isobaric ensemble (NPT) for 100 ps at 300 K and 1 bar, then (only for the NOE-restrained production run) for an additional 100 ps under the same conditions with the NOE restraints applied. The two production simulations (one with and one without NOE restraints) were then performed for 10 ns in the same NPT ensemble. The temperature and pressure of the system were controlled using a Berendsen thermostat 23 and a Parrinello-Rahman barostat. 24 Supplementary Table 13 shows that 16/20 of the NOE contacts are already satisfied in the unrestrained simulation. Adding them simply stabilizes the structure-the standard deviations and ranges of the distances are reduced-and ensures that the remaining four restraints are satisfied.