If magma cools quickly, crystals don't have much time to form, so they're very small. If magma cools slowly, the crystals have enough time to grow and enlarge. For crystals to have a long time to grow, the evaporation rate must be slow (see below). The evaporation rate of beakers, crystallization plates, Petri dishes and watch crystals can be controlled in part by covering the plate with aluminum foil and drilling holes in the foil.
The foil delays evaporation compared to an open container, while the holes allow some solvent vapor to escape, preventing a completely closed system. The number of holes in the sheet can be increased for less volatile solvents or fewer holes can be made to slow the evaporation of a highly volatile solvent. To crystallize a solid, impure compound, add enough hot solvent to completely dissolve. The flask then contains a hot solution, in which the solute molecules, both the desired compound and the impurities, move freely between the hot solvent molecules.
As the solution cools, the solvent can no longer contain all the solute molecules and they begin to exit the solution and form solid crystals. During this cooling, each solute molecule, in turn, approaches a growing crystal and rests on the surface of the crystal. If the geometry of the molecule conforms to that of the crystal, it is more likely to remain in the crystal than to return to solution. Therefore, each growing crystal consists of only one type of molecule, the solute.
Once the solution has reached room temperature, it is carefully placed in an ice bath to complete the crystallization process. The cooled solution is then filtered to isolate the pure crystals and the crystals are rinsed with cooled solvent. This will encourage crystals to grow on the side of the vial, as there is more solvent in contact with the side and the angle prevents newly formed crystals from falling directly to the bottom of the vial. Recrystallizations for purification purposes are a well-known and widely applied technique, but the cultivation of crystals suitable for single-crystal X-ray diffraction (XRD) is less well known and is more of an art than a science.
Small crystals have a large surface area-to-volume ratio and the impurities are found on the surface of the crystals and are trapped within the matrix. Under the old Garbage In %3D Garbage Out rule, a crystal structure is only as good as the glass used for data collection. If after two weeks no crystals have formed in the sample, it may be time to reconsider your culture technique with solvents or crystals and try another method. The impurities of the yellow triangle are trapped inside the crystals that are formed by the orange hexagons, so the isolated crystals are impure.
Very small holes reduce the maximum resolution at which the crystal is diffracted, larger holes destroy the glass. If you find that the evaporation or slow cooling technique provides crystals, but the crystals are not the right size or shape for XRD, both techniques can be extended to include a binary or tertiary solvent system. Single-crystal X-ray diffraction is an excellent way to characterize a compound, but cultivating crystals that provide good data is a skill and an art.