NMR Enters The Crystallography Realm

For most chemists, the term crystallography is interchangeable with X-ray diffraction. But really crystallography is just the study of the forms and structures of crystals, and there are plenty of crystalline materials for which diffraction fails to reveal a structure. In recent years, advances in solid-state nuclear magnetic resonance techniques have allowed NMR to emerge as an important addition to crystallographers’ toolboxes.

In some cases, researchers have used NMR to independently reproduce structures determined by X-ray diffraction. “There is deep symmetry between X-ray diffraction and NMR,” which makes the two methods complementary, says Manish A. Mehta, a professor of chemistry at Oberlin College. X-ray diffraction yields electron densities that allow scientists to then infer a molecule’s nuclear coordinates. “In solid-state NMR, it works the other way around,” Mehta says. “What you have are the nuclei that tell you something about what’s going on with the electrons around them.”

But the real power of NMR crystallography lies in getting structural information when X-ray diffraction cannot, such as when crystals are too small to yield good diffraction data, or when they incorporate elements that can’t be distinguished by diffraction, contain defects or lack periodic order.

Additionally, NMR allows scientists to study the motion of atoms in crystals, says Francis Taulelle, a research group director at the Lavoisier Institute of Versailles and a professor at the Centre for Surface Chemistry & Catalysis at KU Leuven. “Very often diffraction gives the impression that a crystal is a static object—that a crystal is solid and nothing moves. This is absolutely wrong.”

The NMR crystallography field grew up over the past couple of decades, as higher-field magnets enabled higher sensitivity and spectral resolution and as techniques advanced for data analysis and quantum mechanical modeling. The latter helped the field develop in part because, although researchers can solve a crystal structure by NMR crystallography alone, the technique benefits from X-ray diffraction as well as computational modeling of possible structures to aid interpretation of experimental data (J. Am. Chem. Soc. 2013, DOI: 10.1021/ja311649m). “Diffraction and NMR provide different complementary constraints” to help understand crystalline structures, says Bradley F. Chmelka, a professor of chemical engineering at the University of California, Santa Barbara.

Taulelle and Chmelka separately use NMR crystallography to study microporous silicate zeolites that may be used for catalysis or gas separation. Determining zeolite structures is challenging because they can be difficult to crystallize into large enough crystals to generate good diffraction data. Additionally, X-ray diffraction can’t distinguish between silicon and aluminum or boron atoms found in the zeolites.

As a result, X-ray data limited the conclusions researchers could make about some zeolites’ structures. “For a long, long period, we all believed that the elements were homogeneously distributed, and this is absolutely wrong,” Taulelle says. “Every zeolite has a different element distribution, and the properties come from that distribution. We have to go beyond the crystal network to find why the aluminum is distributed in a particular way and correlate that distribution to a zeolite’s properties.”

Chmelka sees similar characteristics in phosphors for solid-state lighting (Chem. Mater. 2013, DOI: 10.1021/cm401598n). “You take a white powder such as yttrium aluminum garnet or calcium scandate and put in a tiny amount of a rare-earth element, and it becomes a very bright phosphor,” Chmelka says. The rare-earth elements incorporate into the material seemingly at random, without long-range order, and with phosphor performance depending strongly on rare-earth element amount and distribution. Understanding why they have such a strong effect on the photoluminescence requires more structural information than diffraction can provide.

Studying disorder in crystalline materials is also the specialty of Sharon Ashbrook, a chemistry professor at the University of St. Andrews. “Sometimes people skip over the idea of disorder and say it doesn’t matter,” Ashbrook says. “But disorder really changes the physical and chemical properties of the material.”

Her research includes investigating hydration of anhydrous silicates such as wadsleyite, a form of Mg₂SiO₄ that exists at high pressures in Earth’s mantle (Chem. Sci. 2013, DOI: 10.1039/c3sc21892a). “The amount of protons in that material has implications for plate tectonics, but no one knows where the protons sit,” Ashbrook says. A better understanding of water in wadsleyite could also inform use of similar minerals for water and hydrogen storage applications.

Researchers also use NMR crystallography to study composite and organic materials. Bone, for example, is largely a composite of stacks of crystalline platelets of calcium phosphate that are separated by water layers and housed within proteinaceous collagen fibrils. NMR was critical to identifying the presence and role of water and, more recently, citrate in bone mineral.

Studying an octacalcium phosphate model of bone with powder X-ray diffraction can determine only that the mineral’s unit cell size expands with the addition of citrate, says Melinda J. Duer, a chemistry professor at the University of Cambridge. She and colleagues used NMR to reveal that citrate binds in multiple ways to bridge the water between the crystalline layers, perhaps helping to create a viscoelastic layer that dissipates energy (J. Magn. Reson. 2015, DOI: 10.1016/j.jmr.2014.12.011). Duer is also investigating other organic acids that affect bone structure. “We’re hoping to map those changed mineral structures onto changes that arise in osteoporosis,” she says.

Oberlin’s Mehta and Lyndon Emsley, head of the Laboratory of Magnetic Resonance at Swiss Federal Institute of Technology, Lausanne, each also study mixtures of materials with an eye toward pharmaceutical applications.

Mehta’s research focuses on the formation and structures of cocrystals, which are crystals that contain two or more molecular entities held together by noncovalent interactions. Pharmaceutical makers add so-called coformers such as malonic acid to the active drug in a formulation to generate a different crystal form. Tweaking these structures can affect the drug’s properties, such as its ability to be absorbed by the body.

Mehta wants to understand the mechanisms behind how ingredients combine and cocrystallize. He’s been using NMR to study the combination of caffeine and malonic acid, which spontaneously cocrystallize when mixed (J. Phys. Chem. Lett. 2014, DOI: 10.1021/jz501699h). He hopes to identify intermediates and determine the energetics of the process.

Emsley, for his part, would like to be able to “take a pill, subject it to solid-state NMR, and determine de novo what the structure of the active pharmaceutical ingredient is in the formulation and how it interacts with excipients,” such as binders or preservatives, he says. To that end, he is experimenting with combining solid-state NMR with dynamic nuclear polarization, which involves adding a stable radical compound to a sample to enhance NMR signals (J. Am. Chem. Soc. 2014, DOI: 10.1021/ja4092038).

Fifteen years ago, the term NMR crystallography meant nothing to people, Emsley notes. Now, the International Union of Crystallography has created a commission for it in recognition of its usefulness, and conferences are being organized that aim to promote structure elucidation using multiple techniques. Says Emsley, “We’re seeing something emerging about using methods in convergent ways to solve structures.”