IntroductionI n 1988, a report by Amundson on behalf of the U.S. National Research Council (Amundson, 1988) identified nuclear magnetic resonance (NMR) spectroscopy as a potentially useful tool for studying catalyst structure and surface chemistry, noting the existence of magnetic resonance imaging (MRI). Fourteen years later, there is a substantial body of evidence to say that magnetic resonance (MR) techniques, far more wide ranging in scope than was originally envisaged, have become integrated into chemical engineering research and are becoming increasingly more popular in this role. This said, the diversity of measurements that can be made and the systems that can be studied are still relatively poorly understood. The opportunities for market penetration are still quite high.From a chemical engineer's perspective, the most important attributes of MR lie in its ability to record 1-D, 2-D and 3-D images with the resolution as high as 5-10 µm, along with spatially resolved maps of transport characteristics (diffusion, dispersion, and flow vector), chemical composition, and variations in the state of matter being probed-all this information being accessed noninvasively and without the need for a tracer. It is the ability to combine different MR measurements within a single experiment (e.g., combining spectroscopy and transport measurements) that makes it so useful in chemical engineering research. To date, most effort has been focused on transferring the imaging and flow visualization measurements originally developed for medical applications across the nonmedical arena. In itself this is not a trivial exercise, since the MR characteristics of nonmedical systems and their requirements on the detail of implementing the MR experiment are quite different from those of medical subjects.
Ongoing MR Research in Chemical EngineeringMR is established, but is still relatively underused in three distinct areas of application: characterizing fluid flows; visualizing structure-hydrodynamics relationships in process units; and the integrated use of MR with theoretical modeling and numerical simulation studies. The way in which MR is used in each of these applications can be quite different.
Characterizing fluid flowsUntil recently, applications of MR to study flow phenomena have almost entirely been restricted to probing the liquid, as opposed to gas, phase-a direct result of the relatively poor signalto-noise inherent in MR techniques. Nuclear spin densities in the liquid phase are ~10 3 higher than those in the gas, thereby making signal acquisition faster and easier. The potential use of MR in studying fluid flows is huge, because methods can be used to probe molecular displacements over length scales of 10 -6 m up to the physical dimension of the sample. The time scales over which these displacements can be studied in a single measurement is sample-dependent (and depends on the characteristic relaxation time properties of the system), but can lie in the range of milliseconds to seconds. For those who are brave enough to r...