NMR noninvasively yet powerfully probes for a variety of data, including chemical reaction progress, the structure and dynamics of biochemicals, and a variety of indicators for both industrial quality control and security screening. NMR's spatially enhanced offspring, magnetic resonance imaging (MRI) typically generates crucial graphical representations of fluid flow, rates of diffusion, and the interiors of materials and living organisms. Therefore, the general technique of magnetic resonance (MR) plays an indispensible role in fields ranging from molecular biology, genetics, and diagnostic radiology to geotechnical research, pharmaceuticals, and materials science. However, MR experiments typically demand the placement of the sample or subject inside the relatively small bore of a large, solenoidal, superconducting magnet, which provides a fairly homogeneous magnetic field. This requirement inherently constrains the sample size and makes high-field magnets bulky and expensive. Typical MR techniques and instrumentation justify these inconveniences because the spins require a spatially uniform field in order to resonate at uniform frequencies and give a strong, clear signal.
MR magnet users also typically tweak the static field into uniformity by means of a complicated and expensive array of "shim coils." Shim coils generate obvious obstacles to the portability and simplicity of MR systems. The amount of inhomogeneity correction a given shim coil can supply remains strictly limited by hardware constraints; it can prove difficult to construct shim coils capable of correcting for more significant inhomogeneities (such as those present in portable and single-sided systems.) Fortunately, the Pines lab and other researchers have demonstrated that shim coils and homogeneous, solenoidal magnets are not strictly necessary.
A researcher can, in fact, manipulate the most fundamental of MR hardware components to refocus even fairly large inhomogeneities -- this is the keystone accomplishment and driving philosophy of the toolbox of methods, developed primarily in the Pines lab, called ex-situ MR. These manipulations temporarily refocus the inhomogeneities at a single point in time (i.e. the "homogeneity echo",) thus facilitating the stroboscopic or two-dimensional acquisition of effectively homogeneous signal. The homogeneity echo draws its versatility from two sources:
- The researcher exploits control over the time dependences of the RF pulse and imaging gradient hamiltonians. Full exploration of this area has included ongoing investigation into areas of pulse sequence design, numerical optimization, and quantum control.
- In the highly inhomogeneous case of one-sided systems, a researcher or engineer can fine-tune a pre-existing natural correlation between the RF and static magnetic fields. This poses challenges in the optimization (both intuitive and numerical) of RF coil designs, within the constraints of inexpensive machining.
In addition to these more mature ex-situ methodologies, our explorations into the general areas of pulse and hardware design have outlined entirely new paths for improvement of MR in inhomogeneous fields. While some of these remain nascent areas of development, we would like to survey our shift from experiments inside commercial high-field magnets (with artificially imposed inhomogeneities,) towards the testing and utilization of custom-built portable and single-sided systems, and on to the ongoing, in-house construction of novel hardware designs.
Pulse Design
We are developing a unique toolbox of pulse sequences and hardware-based methods that can overcome to a large extent the limitations when attempting to use MR in inhomogeneous fields. These new methods use pre-image manipulation of the phase of nuclear spins to provide spatial and spectral resolution. The computational demands of this approach, namely the determination of the appropriate spin trajectory to provide sharp images, are expected to be far less than the computational demands associated with post-image processing algorithms. In general, the magnet design (since not all open magnets will be tractable) will dictate the ideal pulse scheme, which should incorporate RF efficiency, short length, and broadband behavior,together with tunability. We focus on the mutual optimization of the magnet and RF coil geometry, however, ex-situ pulses significantly decrease the precision required of magnet design, though hardware design and pulse design do still remain interdependent.
[ Back to top ]
Magnet Design
The magnet is one of the most critical elements of any MR device, and despite several advances and techniques for field optimization and corrections, the design and construction remains a crucial aspect of the system. The magnet’s size, cost, field strength, homogeneity and sensitive volume will in a large part determine the performance of an ex-situ system. For traditional MR magnets, the designs aim for stringent requirements on the field strength, homogeneity, and general field profile (e.g. flat slices for imaging) whereas in ex-situ these requirements are comparatively relaxed to allow for considerably less expensive and smaller magnets, and more open configurations. What the optimal magnet designs are for these smaller and cheaper MR systems remains an open question and an active area of research. We currently have several small one-sided magnets to demonstrate our ex-situ methodologies. Our newest magnet arises from a design methodology novel in its application to permanent magnet design and incorporates features that allow for an easy adjustment of the magnetic field.
[ Back to top ]
Coil Design
One of the requirements for the realization of an ex-situ matching is to have the RF field matched with the static field, meaning that both fields should have a similar spatial dependence. Due to several pulse sequence advances this matching requirement can be relaxed. One should start by designing a coil that to a first degree matches the static field profile. In addition coil design becomes more critical in specific applications for sensitivity reasons, hence some of our efforts are aimed towards rf field design. In addition we are designing and building customized single sided gradient coils and shims. Results suggest the possibility of substantially pushing the current limits of high resolution NMR spectroscopy in weak and inhomogeneous fields.
[ Back to top ]
Electronics Miniaturization
The sustainability of portable MR systems for chemical analysis and/or imaging of samples in the field depends in further miniaturizing of the system electronics, decrease in the power requirements sensitivity and resolution enhancement. RF transmitter and receiver components in single boards have been developed from several companies. We have worked with Quantum Magnetics (now GE Security) for a custom-made system that allows for continuous RF amplitude and phase modulation,gradient modulation, and fast data transfer. The system also contains a class D amplifier. We are currently building custom portable spectrometers, with miniature electronics.
[ Back to top ]