I.M. Thomas

Magnetic susceptibility tomography for three-dimensional imaging of diamagnetic and paramagnetic objects

A tomographic technique for reconstructing the three-dimensional distribution of magnetic susceptibility in an object is described, A SQUID magnetometer may be used to measure the perturbations imposed by the object on an applied magnetic field and these data contain information about the susceptibility distribution. To assess the technique, a model object was defined, simulated magnetic field data were generated, and a matrix inversion was carried out with singular value decomposition to yield a least-squares solution for the susceptibility distribution. Various relative geometries of the three interacting physical systems (the applied field, the object and the measurement space) were used and the algorithms performance was investigated for each of the cases in which one of the systems was moved while keeping the other two fixed. With either strategy involving relative motion between the object and the measurement space, accurate, convergent solutions were obtained, but the algorithm failed when only the direction of the uniform applied field was varied. A suitable nonuniform applied field may make the algorithm robust. Applications for a tomographic imaging susceptometer in biomedical imaging, nondestructive evaluation, and geophysics are envisaged. >

Magnetic susceptibility imaging for nondestructive evaluation (using SQUID magnetometer)

High-resolution superconducting magnetometers such as MicroSQUID (superconducting quantum interference device) have been shown to be effective for nondestructive evaluation. MicroSQUID can also be used with a room-temperature magnet to image the magnetic susceptibility of materials. A diamagnetic or paramagnetic sample is scanned in the applied field, and the local perturbations are measured. For thin samples, such as plates, sheets, or thin sections of rock, the data are deconvolved to generate two-dimensional susceptibility images. Three-dimensional structures can be imaged with magnetic susceptibility tomography: deconvolution of a large data set obtained by applying the field and scanning in multiple orientations. Extremely small surface defects on nonmagnetic or weakly magnetic samples are imaged by decorating the sample with paramagnetic microspheres prior to scanning. Magnetic susceptibility imaging demonstrates the feasibility of SQUID nondestructive evaluation on materials that could previously be examined only with X-rays or ultrasound.<<ETX>>

Spatial resolution and sensitivity of magnetic susceptibility imaging

The use of a scanning susceptometer to image the 2-D distribution of magnetic susceptibility in thin samples is discussed. High-resolution magnetic field data recorded above complex shapes of plexiglass, scanned in a uniform applied field, are presented. Deconvolution of these data yielded magnetization distributions that were accurate images of the samples. It was found that the susceptibility of plexiglass is -9.0*10/sup -6/ (SI) and it was demonstrated that the present system is sensitive to susceptibility contrasts as small as 5*10/sup -7/ (SI), with a spatial resolution better than 1 mm. This performance is limited by the strength of the applied field, by the size and distance from the sample of the sensing coils, and by superconducting quantum interference device (SQUID) noise. It is estimated that one could increase spatial resolution by a factor of two, and sensitivity to susceptibility contrast by at least one order of magnitude by using an imaging susceptometer whose SQUIDs are integrally mounted within a superconducting magnet.<<ETX>>

High resolution magnetic susceptibility imaging of geological thin sections: Pilot study of a pyroclastic sample from the Bishop Tuff, California, U.S.A.

High resolution magnetic susceptibility imaging is a new technique for studying the magnetic properties of geological thin sections. The two-dimensional distribution of both remanent and induced magnetization can be determined with a spatial resolution (< 1.0 mm) that is similar to the size of phenocrysts in the sample. Amongst many problems in rock magnetism to which it could be applied, the technique holds great potential for understanding the origin of the anisotropy of magnetic susceptibility in pyroclastic flows. Preliminary tests on a single sample of ignimbrite indicate that secondary iron-titanium oxide particles deposited within vesicle walls prior to their collapse are responsible for the bulk susceptibility.

Reconstruction of two-dimensional magnetization and susceptibility distributions from the magnetic field of soft magnetic materials

We apply a spatial filtering technique to solve the two-dimensional magnetization imaging problem. This allows us to reconstruct the magnetization distribution from the measured magnetic field above a planar soft magnetic sample exposed to an applied magnetic field. Knowledge of the magnetizing field distribution can then be used to determine the susceptibility distribution and hence to obtain information about the material characteristics of the sample. By simulation of an actual experiment, we examine the spatial resolution of the resulting image and discuss the uniqueness of solutions to the inverse problem.

A high resolution imaging susceptometer

A high-resolution, MicroSQUID (superconducting quantum interference device)-based susceptometer has been constructed for imaging the susceptibility distribution in diamagnetic and paramagnetic objects. The maximum available applied field, produced by the Helmholtz pair, is 300 mu T, which provides an adequate signal-to-noise ratio for common susceptible materials. The calculated variation in the applied field within 20 mm of the center is less than 0.1% of the field at the center. In a 100-Hz bandwidth, the minimum detectable susceptibility-induced field change, (limited by SQUID noise) is 3*10/sup -9/ of the applied field. The sensitivity of the system may be improved by increasing the applied field, for instance by incorporating a superconducting magnet into the magnetometer dewar, and by using lower-noise SQUIDs.<<ETX>>

High resolution SQUID imaging of current and magnetization distributions

Interpretation of high-resolution SQUID (superconducting quantum interference device) magnetometer data is facilitated by inverting them into current or magnetization images. Different deconvolution algorithms utilizing the finite element method (FEM), spatial filtering (SF) and lead field analysis (LFA) are compared. The authors applied an appropriate algorithm to magnetic field data produced by current distributions in high-T/sub c/ superconducting thin films (TlBaCaCuO), by the diamagnetic magnetization of plexiglass and by planar and cylindrical current-carrying conductors with flaws. They demonstrate and compare the performances of the algorithms and conclude that, with all of the approaches, the deconvolved images are easier to interpret than the original field maps. The FEM and SF can both solve for discontinuous sources. The FEM allows the solution to be constrained with boundary conditions but it is much slower than SF, which cannot constrain the solution. LFA is numerically simple for complex geometries but it is slow and cannot deal with discontinuous sources.<<ETX>>

Superconducting magnetometry: a possible technique for aircraft NDE

Superconducting quantum interference device (SQUID) magnetometers offer promise as multi- mode instruments capable of obtaining high resolution images of extremely low frequency injected currents or eddy currents, and they can be configured to image the magnetic susceptibility of titanium, aluminum, and nonmetallic composites. While high resolution SQUID magnetometers will generally be noisier than conventional SQUIDs, the small coils and reduced coil-to-source spacing more than compensate to provide low-noise, high- resolution images. To explore SQUID NDE, we have developed research facilities that include the high-resolution MicroSQUID magnetometer, a magnetic shield, a scanning stage, and a computer-based control and data acquisition system. Using this instrumentation, we have imaged magnetic fields produced by varied sources. In support of the experimental studies, we have developed analytical and numerical models for the simulation of flaws with several geometries inside thick and thin current-carrying plates and thin-walled tubes, and have demonstrated that two-dimensional magnetic images can be deconvolved into images of current or magnetization by filtering techniques, finite element models, lead field analyses, and maximum entropy methods.

A distributed quasi-static ionic current source in the 3-4 day old chicken embryo

We report measurements of slowly varying magnetic field patterns close to fertilized eggs of the chicken Gallus domesticus during the first few days of incubation. These fields are generated by ionic currents within the egg that are associated with the development of the embryo. Since they are very weak (no greater than tens of pT) and vary over distances of a few millimetres, it has been necessary to develop specialized instrumentation and analysis techniques. We describe the use of high-spatial-resolution SQUID magnetometers to measure the field patterns and appropriate imaging algorithms to model the current sources responsible for producing the fields. Our results provide strong evidence for a distributed source in the extra-embryonic membranes. There is also indication of a more localized source within the embryo itself.

SQUID NDE: detection of surface flaws by magnetic decoration

A nondestructive evaluation (NDE) surface-flaw detection technique involving decoration of the samples with a ferromagnetic or paramagnetic tracer and subsequent scanning with a high-resolution magnetometer or susceptometer is considered. The authors present the results of a preliminary study demonstrating that this technique is sensitive enough to detect and image defects at least as small as 254 mu m*51 mu m*76 mu m deep in a nickel-alloy test block. Applied field strength and superconducting quantum interference device (SQUID) noise limit sensitivity, and sensing coil characteristics limit spatial resolution.<<ETX>>