Minestrone and magnetic resonance

Researchers in the US and France have solved a decade old mystery in magnetic resonance studies by spotting a discrepancy in the way nuclear spins behave. Their new mathematical model of the process improves our understanding of atomic behaviour and could lead to better NMR spectra, sharper magnetic resonance images, and perhaps one day a fully portable MRI machine.

Writing in the Journal of Chemical Physics, chemist Philip Grandinetti of Ohio State University and colleagues there and at the Centre National de la Recherche Scientifique, the Université d'Orléans, and the Université de Lyon, France, explain how difficult it is to control completely the behaviour of atomic nuclei in some nuclear magnetic resonance (NMR) experiments. An adiabatic experiment applies a firmer grip in the quest to boost NMR resolution but the nuclei do not always behave as they should.

Even then, although NMR is a wonderful tool for determining molecular structure, picturing more complex objects, such as human brains with the same level of detail is not quite as simple. Specifically, he says, these images often lack detail because whenever atoms happen to "broadcast" in opposite directions, they cancel each other out of the final image.

The team's steps towards improving this situation was somewhat serendipitous. They had been examining an earlier University of Alberta experiment, which had been aimed at enhancing NMR signal strength, but discovered anomalous behaviour that led to the new insights. "We originally wanted to work out a rigorous theoretical description of the Alberta experiment," says Grandinetti, "but the more we tried to understand even the simplest adiabatic process in magnetic resonance, the more we realized that there was a disturbing discrepancy between theory and experiment."

The researchers then uncovered the same contradiction in many other studies that have been carried out during several decades of NMR research. Even though all these experiments seemed to run properly and yield good results, the nuclear spins were not behaving in as controlled a manner as previous NMR theory would have use believe. There was scant mention of the effect in many years of scientific papers, but the observation had not been pursued to its logical conclusion.

The researchers found that atoms were not spinning out of control, they were moving in a "predictable" way, following the rules of super-adiabaticity, a quantum mechanical concept first proposed in the late 1980s by Sir Michael Berry at the University Of Bristol. A super-adiabatic process is still adiabatic. This explains why there were so many seemingly contradictory experiments that nevertheless worked. The atoms were behaving in an adiabatic way, just not following the commonly accepted adiabatic rules.

Grandinetti and colleagues have now derived a mathematical explanation for this nuclear deviancy and suggest it might be applied in sharpening up MRI images. The advance may one day help scientists to look inside people and objects without needing the giant magnetic cylinder of current MRI scanners.

"To be fair, though, it wasn't clear that this discrepancy posed a real problem, and most people thought the conventional theoretical approach was doing a fine job," Grandinetti adds, "It was only after we fully understood the reason for the discrepancy that we realized the conventional theoretical approach contained a flaw that might prevent better adiabatic processes from being discovered."

The mathematics boils down to the difference between adiabatic and superadiabatic. In the former, the atoms being studied move metaphorically speaking from one point on a globe to another, slowly, and following a very carefully designed path. In super-adiabaticity, the atoms follow a different sometimes, wildly different, path, but still end up travelling from A to B, just as do their adiabatic counterparts.

Grandinetti has a nice analogy for the difference: "Imagine you have to carry a bowl of soup from the kitchen to the dining room and you don't want to spill it. You want the surface of the soup to remain flat during the entire process. In adiabatic "delivery" you move infinitely slowly so that the soup surface not only stays perfectly flat but also remains perpendicular to the force of gravity (locking force). Of course, the problem with an infinitely slow process is that the soup always gets cold! In the superadiabatic "delivery" you accelerate and decelerate smoothly so that the soup surface still stays perfectly flat but will now be perpendicular to new locking force which combines the original locking force (gravity) with the inertial forces of the moving frame."

Grandinetti now hopes to incorporate the algorithm into software for controlling NMR and MRI measurements so that it can be used to boost image resolution. One day, it might even help these instruments record the signals from objects that are not inside the magnet.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.