Abstract
Nuclear Magnetic Resonance imaging in the medical and biological fields has been well established since the early 1970's, and high quality, well resolved images are now routinely obtained. Only recently has the enormous potential of Nuclear Magnetic Resonance as a means of non-invasive imaging of solid samples been recognised. The broad natural linewidths associated with solid samples has been the cause of the paucity of published solid imaging papers, and the techniques put forward in order to overcome the broad lines and resolve the spatially encoded contributions to the lineshapes range from the simple, single radiofrequency pulse methods, to the more complex multiple-pulse sequences. This thesis describes an attempt to return to the simple radiofrequency pulse methods, employing as few pulses as possible, by way of the "Solid Echo". A theory of solid echo imaging has been put forward, and a general result governing the form of the images in terms of the Fourier transformed, modulated, quadrature solid echo peak is postulated. Validation of the theory is provided by a series of one and two dimensional images of various nuclear spin configurations and spin species. Rapid gradient field switching was not a viable option during this period of research and, because of this, the gradient fields had to be left on during the radiofrequency pulses. The resulting nuclear spin motions about the two effective fields of the solid echo pulse sequence due to the combination of the gradient and radiofrequency fields has been modelled by a first order approximation. An evaluation of solid imaging using the MREV-8 multiple-pulse sequence is also given, and profiles obtained using this method are demonstrated.