The crystal structure ofN,N′-methylenebisacrylamide was determined through the geometry optimization of the molecular unit with density functional theory and conformational analysis, and then through the calculation of the packingviaa crystal structure prediction protocol, based on lattice energy minimization. All the calculated structures were ranked, comparing their powder pattern with the laboratory low-quality X-ray diffraction data. Rietveld refinement of the best three proposed structures allowed the most probable crystal arrangement of the molecules to be obtained. This approach was essential for disentangling the twinning problems affecting the single-crystal X-ray diffraction data, collected on samples obtainedviarecrystallization of powder, which definitely confirmed the predicted model. It was found thatN,N′-methylenebisacrylamide shows a monoclinic structure in the space groupC2/c, with lattice parametersa= 17.822 (12),b= 4.850 (3),c= 19.783 (14) Å, β = 102.370 (9)°,V= 1670 (2) Å3. Two strong interactions between the amide protons and the carbonyl groups of neighbouring molecules were found along thebaxis, determining the crystal growth in the form of wires in this direction. This work provides a further example of how computational methods may help to investigate low-quality molecular crystals with standard diffraction techniques.
Machine Learning-based Crystal Structure Prediction for X-Ray Microdiffraction
