If you ever took a chemistry lab class in college, chances are you once stared desperately at a flask of liquid, crossing your fingers for tiny crystals to appear. Your lab instructor may have offered advice that sounded like voodoo: "Scratch the inside of the flask to make the crystal grow." But the trick worked--and now scientists have uncovered new details behind it.
Compared with the fast wiggling and whizzing of molecules, crystallization takes forever. A crystal has a specific, ordered pattern, and it’s quite unlikely that a disorganized soup of molecules will suddenly reach that state. But once the molecules start to organize themselves, a process called nucleation, they act as a template for others to get in line--and crystallization takes off. Grooves and pits in glass surfaces help crystals grow faster, because they act as nucleation hot spots. Chemists have known why for decades: The interface between crystal and liquid is unstable, and grooves minimize this interface.
To learn more about how these grooves aid nucleation, physical chemists Amanda Page and Richard Sear of the University of Surrey in the United Kingdom ran computer simulations of droplets of simple molecules such as argon or methane inside v-shaped grooves. The researchers recorded how long nucleation took as they changed the groove's angle. The optimal angle was about 70 degrees, the scientists report online this month in the Journal of the American Chemical Society. Nucleation occurs 48 orders of magnitude faster at this angle than it does on a flat surface.
The reason why one angle is optimal, the researchers found, has to do with the three-dimensional repeating pattern the molecules make in the crystal. The simulated argon and methane molecules in this study, for example, like to assemble into a type of pattern called a face-centered cubic lattice, which fits comfortably into the 70-degree wedges. Other groove angles, such as 45 degrees, force the molecules to disrupt their preferred pattern (see picture), slowing crystallization. "The crystal says, 'I want to be 70 degrees,' and the wedge says, 'No, you have to be 45 degrees,' " Sear says. "So there's frustration."
A surprising result from these simulations is that crystals of simple molecules need only a simple-shaped template to help them nucleate, says theoretical chemist Peter Harrowell of the University of Sydney in Australia: "It will be an eye opener for people."
Still, notes Sear, a 70-degree groove won't work for all molecules. Complex molecules such as drugs can crystallize into more than one pattern, for example. That has practical consequences; because some crystal lattices are more soluble than others, the right crystalline shape can influence how much of a drug enters the bloodstream. In 1998, the drug company Abbott discovered that a less-soluble crystalline form of the HIV drug ritonavir was rampant in their production lines, forcing them to rework how they made the drug. Designing apparatuses that have nano-sized grooves with specific shapes might help favor the crystals drugmakers want over those they don't, Sear says.