Focus: Tracking Nuclei on the Move

Published October 25, 2006  |  Phys. Rev. Focus 18, 13 (2006)  |  DOI: 10.1103/PhysRevFocus.18.13
+Enlarge image Figure 1
Phys. Rev. Lett. 97, 175502 (2006)

Full of holes. An x-ray computed tomography (CT) scan of sandstone shows a range of sizes of pores that can hold oil. A new technique probes the motion of molecules between such pores to learn how readily oil could be extracted.

Geologists routinely probe subterranean atomic nuclei using nuclear magnetic resonance (NMR) to learn about the porous structure of rocks that hold petroleum. Now, in the 27 October PRL, researchers show how to extend the technique to track the motion of liquid between different pores. The method could help petroleum engineers extract oil more efficiently. It could also allow researchers to monitor the motion of molecules between different microscopic regions in liquid crystals or biological cells.

To learn about a rock’s structure using NMR, researchers first expose a water-saturated rock to a strong magnetic field. The nuclei of hydrogen atoms in the water molecules act like tiny bar magnets because of their internal spins, which tend to line up with the applied field. Then an oscillating field is briefly added to the initial field to jostle these aligned spins, causing them to wobble in synch like off-balance tops. Monitoring the wobbles provides information about the rock structure.

For more than a decade, petroleum engineers have been sending NMR rigs down candidate oil wells to learn about the surrounding rock. For example, they can measure a time called T2, the time it takes the wobbling spins to lose synchronization with the original oscillating field. Collisions with the walls of a pore hasten this so-called dephasing, so molecules in a small pore, which hit the walls more frequently, have a shorter T2. “It’s a marker for the size of a pore,” says Paul Callaghan of Victoria University of Wellington, New Zealand. But knowing the range of pore sizes doesn’t tell you whether the pores are isolated or are connected well enough to allow trapped oil to move through the rock and out of the well. For that, you need to know if molecules move from pore to pore, and if so, how rapidly. This information comes from measuring changes in T2 over time.

Two research teams have previously measured a change in T2 values in the lab, but only over a fixed amount of time [1, 2]. They measured T2, waited a “mixing time” of tens of milliseconds, and then measured again. The technique allowed them to track populations of molecules and learn the fraction that had moved to a new T2, and therefore a different-sized pore. But the teams didn’t vary the mixing time to see how rapidly the entire population of water molecules moved to new pores, although they did measure many other properties.

Callaghan and his Victoria graduate student Kate Washburn have now varied the mixing time for a chunk of sandstone and “followed the actual dynamics of the exchange process,” says Callaghan. Using the right computational techniques was critical to their success. Extracting T2s from the raw data involves a difficult mathematical maneuver called a two-dimensional inverse Laplace transform, and the team benefited from recent advances in the computer algorithms, Callaghan says. He believes similar techniques should be useful for measuring molecular motions between distinct regions of other samples, such as domains of liquid crystals or the interior and exterior of biological cells.

Yi-Qiao Song, of Schlumberger-Doll Research in Ridgefield, Connecticut, says the new work, including the other recent papers, is “very exciting.” He says the technique will allow petroleum researchers to go beyond mere structural measurements “to something that will probe new physics.”

–Don Monroe


References

  1. P. J. McDonald et al., “Surface Relaxation and Chemical Exchange in Hydrating Cement Pastes: A Two-Dimensional NMR Relaxation Study,” Phys. Rev. E 72, 011409 (2005).
  2. J. H. Lee et al., “Two-Dimensional Inverse Laplace Transform NMR: Altered Relaxation Times Allow Detection of Exchange Correlation,” J. Am. Chem. Soc. 115, 7761 (1993).