We
sought the crystal packing preferences of the chalcone scaffold
by analyzing 232 single-component crystal structures of chalcones
with a small (six or fewer non-hydrogen atoms) substituent on either
or both rings, including the unsubstituted molecule. This covers 216
molecules, as some are polymorphic, and 277 independent molecular
conformations, as 16% of the crystal structures have more than one
symmetry independent molecule. Quantum mechanical conformational profiles
of the unsubstituted molecule and the almost 5000 crystal structures
within 20 kJ mol
–1
of the global minimum generated
in a crystal structure prediction (CSP) study have been used to complement
this analysis. Although π conjugation would be expected to favor
a planar molecule, there are a significant number of crystal structures
containing nonplanar molecules with an approximately 50° angle
between the aromatic rings. The relative orientations of the molecules
in the inversion-related dimers and translation-related dimers in
the experimental crystal structures show the same trends as in the
CSP-generated structures for the unsubstituted molecule, allowing
for the substituent making the side-to-side distances larger. There
is no type of dimer geometry associated with particularly favorable
lattice energies for the chalcone core. Less than a third of the experimental
structures show a face-to-face contact associated with π···π
stacking. Analysis of the experimental crystal structures with XPac
and Mercury finds various pairs of isostructural crystals, but the
largest isostructural set had only 15 structures, with all substituents
(mainly halogens) in the para position. The most common one-dimensional
motif, found in half of the experimental crystal structures, is a
translation-related side-to-side packing, which can be adopted by
all the observed conformations. This close-packed motif can be adopted
by chalcones with a particularly wide variety of substituents as the
substituents are at the periphery. Thus, although the crystal structures
of the substituted chalcones show thermodynamically plausible packings
of the chalcone scaffold, there is little evidence for any crystal
engineering principle of preferred chalcone scaffold packing beyond
close packing of the specific molecule.