Use this URL to cite or link to this record in EThOS:
Title: Experiments on ordered spin systems
Author: Grant, Margaret F.
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 1967
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
In this thesis the results obtained from several experiments in which radioactive nuclei were statically oriented in various hosts are described and discussed. The work is divided into eight chapters, the first four giving the theory and experimental details, and the remaining four being concerned with the results. We commence with a brief introduction to the theory of magnetic and hyperfine interactions and describe the difficulties involved in the calculation of their strengths in metals. A discussion follows of the interpretation of measurements of the hyperfine field strength in metals and this short review concludes with an evaluation of the various experimental techniques which may be exploited to measure the hyperfine interaction. A more detailed account follows of the particular technique used to obtain our results, namely that of static nuclear orientation. A phenomenological description of the hyperfine interaction is given, together with the theory of the angular distribution of gamma radiation emitted by oriented nuclei, and a discussion of the scope of gamma anisotropy measurements. The cooling apparatus required to statically orient nuclei is described, together with the gamma radiation detection equipment and the data collection process. Also discussed are the corrections applied to all raw data and the procedure followed in analyzing the measurements in terms of a hyperfine magnetic field acting at the nucleus. Measurement of the sense of any circularly polarised component in the gamma radiation emitted by oriented nuclei yields the sign of the hyperfine interaction, or hyperfine field in our case. The equations describing the gamma radiation polarization are presented, together with a discussion of the theory of, and the apparatus required for, detection of the circularly polarised component via the dependence of the scattering cross-section for polarised electromagnetic radiation on the orientation of the spin of the scattering electron. Gamma radiation anisotropy measurements were taken from four systems, namely 109mAg in iron, 54Mn and 60Co in palladium, and 60Co in thermal neutron-irradiated cobalt. Circular polarization measurements were taken from the palladium host samples, and from a calibrating 60Co in iron sample. An alloy of 1% Pd in Fe was irradiated to form the 14hr 109Pd isotope; this nuclide decays via beta emission (undetected) to a 40sec isomeric state, 88keV above the 109Ag ground state. Anisotropy measurements of the electric octupole 88keV transition yielded an estimate of 340±20kG for the hyperfine field at silver nuclei in iron, assuming complete re-orientation among the nuclear spin levels of the isomeric state. Hyperfine field measurements at diamagnetic nuclei dissolved in ferromagnets can only be interpreted collectively and we describe such an analysis, due to Shirley et al. Two differently prepared samples of 54Mn in nominally pure palladium wore examined and results from them were found to agree to to within the experimental errors. Up to 45% reduction was observed in the intensity of the axially emitted gamma radiation, measured in applied fields up to 5kG and at temperatures between 0.05 and 0.011°K. When analyzed assuming orientation parallel to the applied magnetic field, values for the 'apparent' hyperfine field were obtained which reduced in magnitude with reducing temperature and with reducing applied field at applied fields below ~4kG. This data was reanalyzed assuming the Mn ions to interact, describing this interaction by a randomly oriented magnetic field. The theory for data analysis is given for a constant magnitude interaction and for ''square-wave' distribution in magnitudes' interaction, and the results re-analyzed assuming the true hyperfine field to be a constant. The cut-off value of this 'square-wave form' interaction field was deduced for the two samples to be 0.4±0.1kG and 1.1±0.1kG and the hyperfine field to be 361±15kG. The different interaction field estimates were probably due to a different impurity content of the two samples. Both values correspond to interaction temperatures an order of magnitude above the temperature range of measurement and a later spectroscopic analysis of the initial palladium showed an impurity content high enough to cause magnetic ordering at about the deduced interaction temperatures. From circular polarisation measurements, where a 1% effect was observed, the sign of the hyperfine field at Mn nuclei in palladium was determined to be negative; the hyperfine field value is discussed in terms of the possible contribution from the polarization of the host's conduction electrons. Four samples of 60Co in palladium were examined. The two samples in nominally pure palladium, prepared in different ways and examined in applied fields up to 5kG and at temperatures between 0.05 and 0.012°K, were found to substantially agree to within the experimental errors. The 'apparent' hyperfine field results showed the same temperature- and applied field dependence as the MnPd data but with less accuracy, the axial gamma radiation reduction being always less than 10%. A Co-Co 'interaction field' analysis, as for the MnPd data, yielded estimates of the 'square-wave' form interaction of 1.4±0.2kG and > 1.5kG, and the value for the hyperfine field at Co nuclei in Pd 110±8kG. Any comments concerning the interaction field estimates are similar to those for the MnPd data. Anisotropy measurements from 60Co in two palladium samples containing respectively 0.2% and 0.05% of stable cobalt, sufficient to achieve ferromagnetic ordering below 1°K, were taken in applied fields up to 16kG and at temperatures between 0.03 and 0.011°K. The 0.002CoPd sample yielded 'apparent' He values larger than the dilute samples, but was far from being magnetically saturated. The 0.0005CoPd sample yielded even larger 'apparent' He values, which indicated the approach to a saturation value of either +200±20kG or −240±20kG in applied fields ~20kG. Circular polarisation measurements yielded, from 1% 'effects', the surprising result that the hyperfine fields at 60Co in one of the 'pure' samples and in the 0.002CoPd sample were positive. Thus the result from the 0.0005CoPd alloy is in agreement with n.m.r. data (He = 213±5kG.). Anisotropy measurements from one 'pure' sample were taken in applied fields up to 16kG and the results yielded a value of +116±10kG for the hyperfine field, and were consistent with saturation in applied fields ~5kG. It is concluded that the hyperfine field at cobalt nuclei in palladium metal may well be a function of the cobalt concentration in the very dilute alloy range, possibly due to a variation in the degree of conduction electron polarization. The last experiment reported herein involved the irradiation of a single crystal of cobalt metal at 4°K in a thermal neutron beam. The 60Co nuclei so formed were cooled to ~0.01°K and gamma radiation a anisotropy measurements taken. Some 58Co diffused into the cobalt crystal acted as the thermometer. These measurements were repeated after the sample had been annealed at 20°K, then at 77°K and 300°K. Little change in the data with annealing temperature was observed, in contrast with resistivity measurements from similarly treated cobalt; particularly, the hyperfine field value measured after annealing at 4°K was only 6±4% larger than that measured after annealing at 300°K. From our result it is argued, that even after absorption of a thermal neutron and emission of ~6MeV energy, the 60Co nuclei occupy lattice sites, with any vacancy or interstitial more than one lattice constant distant.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available