J. Verhoogen
1977 Journal of geomagnetism and geoelectricity  
Introduction The recognition that many rocks carry a remanent magnetization that reliably reflects the direction, and in fewer rocks also the intensity, of the magnetic field prevailing at the time and place of their formation, has lead in the past 25 years to spectacular developments in several branches of geology and geophysics. The discovery of temporal variations of the earth's magnetic field on a time scale of 103-107 years, and in particular of frequent reversals of its polarity, has lead
more » ... to new ideas regarding the origin of the field and the behavior of the earth's core. Paleomagnetism fi nally provided, after years of debate, the clinching arguments for continental drift, and lead to the revolutionary concepts of sea-floor spreading and plate tectonics. Yet, paradoxically, the very mechanism by which rocks acquire their tell-tale remanence has remained somewhat obscure, particularly so in the case of the thermal remanent magnetization (TRM) acquired by igneous rocks as they cool in the earth's weak field. The two features of TRM that have made it so useful in paleomagnetism are its intensity and an extraordinary stability that enables it to survive over billions of years. The intensity of TRM in rocks is usually greater by several orders of magnitude than the remanence that can be imparted by exposure, at room temperature, to the earth's field; and destruction of TRM by a-c demagnetization may require an a-c field of several hundred gauss or more. The theory of the acquisition of TRM has been satisfactorily developed, and experimentally tested, in the two extreme cases of 1) magnetic grains small enough to consist of single domains (SD), and 2) grains large enough to have a multi-domain (MD) structure. (The threshold for single-domain behavior in magnetite appears to be close to 0.05 pm for equidimensional grains.) It turns out that a field as weak as the earth's field may induce in an assembly of non-interacting single-domain grains a TRM comparable in magnitude to the saturation remanence induced at room temperature by fields of several thousand gauss, and of high stability. By contrast, weakfi eld TRM in multidomain grains is usually much weaker and less stable. A curious feature is that grains of size between 0.05 and about 15 pm (for magnetite), and which are clearly much too large to be SD, nevertheless exhibit an intensity of remanence and a resistance to demagnetization that are typical of SD grains. Such grains are said to show pseudo single-domain behavior (PSD). The TRM of PSD grains also shows an inverse dependence on grain size that is not typical of MD remanence, but the dependence of TRM on the magnitude of the inducing field is not that predicted for SD behavior. PSD behavior is now generally attributed to the presence of residual moments that cannot be removed by demagnetization. Attempts to explain these residual moments by invoking internal strains (dislocations), inhomogeneity, or Barkhausen discreteness in domain-wall position, have generally failed to account for one or more characteristics of PSD, e.g., the inverse dependence of remanence on grain size. In the past 3 or 4 years, the search has focussed on the internal structure of domain walls ('walls
doi:10.5636/jgg.29.231 fatcat:gtxgfzqn6bg2teqcc5n2uhimjq