G.R. Scott $^{a}$, P.R. Renne $^{a,b}$ and J.M. Feinberg $^{b}$
$^{a}$ Berkeley Geochronology Center, Berkeley, California, USA. $^{b}$ Department of Earth and Planetary Science, University of California, Berkeley, California, USA.
The past performance of the geomagnetic field potentially provides constraints on the thermal and physical evolution of Earth’s inner and outer core. Relevant paleomagnetic information must be deciphered from records preserved in natural materials (rocks), which typically contain mixtures of different minerals, with different grain-sizes and magnetic stability. Rocks produce a compound signal whose fidelity is additionally compromised by lithification processes, burial temperatures and long exposures to later ambient magnetic fields. The intent of laboratory demagnetization procedures is to unravel some of these complexities. Another technique is to examine only certain crystals within a rock so that mineral species can then be chosen that have optimal characteristics of magnetic, thermal and chemical stability.
Discrete iron-oxide crystals are not necessarily stable in a magnetic, thermal and chemical sense. Records from iron-oxides (and sulfides) are degraded by variations in grain-size (and shape) and a tendency to oxidize or hydrate under mild metamorphism and exposure to near surface conditions (as well as during many laboratory experiments). Another recording source is silicate-hosted magnetites. These iron-oxides are subsolidus exsolution products formed as microcrystals within the silicate lattice. Crystallographically oriented magnetic inclusions (COMIS) are arrays of these magnetite microcrystals. The usual form of the microcrystals is as long thin rods or laths, with a narrow range of sizes and low Ti contents. This type of paleomagnetic recording media has the advantages of: 1] extreme shape anisotropy (producing single domain behavior); 2] restricted grain-size (limiting the coercivity and unblocking spectra); 3] enclosure within a refractory silicate (isolating the iron-oxides from later alteration); 4] chemical and thermal stability with its host silicate (even moderate metamorphism may have no effects). Examining silicates with these advantages should significantly expand the geologic range of the paleomagnetic record. Among the common rock-forming minerals: feldspar, clinopyroxene, orthopyroxene, and amphibole can exsolve COMIS. Our initial research has been on clinopyroxene, a common constituent of basic igneous rocks. Two arrays of abundant needle-shaped magnetites occur in clinopyroxenes from slowly cooled crystalline rocks. These arrays are absent from volcanic pyroxenes with the same chemical composition. Constrained to the (010) clinopyroxene plane they are labeled the ‘X’ and ‘Z’ subarrays. Individual magnetite laths have an extremely high ratio of length to thickness (70:1). All rock magnetic and paleomagnetic features indicate single-domain behavior from these COMIS. The intensity of magnetization measured from single crystals (average: 100 A/m) is sufficient to account for both the strong remanence and large magnetic anomalies produced by these igneous rocks. Silicate-hosted remanent magnetism is probably the source of many crustal anomalies, including marine and planetary anomalies. The protracted duration of exsolution and cooling will result in an integrated recording of the ambient magnetic field. Therefore, average field strengths will be recorded that cover an extended period of time (1,000-100,000 yrs). This is in contrast to typical paleointensity measurements from lavas, which cool rapidly to produce an instantaneous (>1yr) recording of the field. By measuring the COMIS, a general, time-averaged record of the geomagnetic field can be obtained. A disadvantage is the self-canceling intensity that would accrue during periods of rapid reversals. Our initial work suggests the feasibility of using silicates with exsolved magnetite to examine the geomagnetic field strength in the Precambrian. Precambrian rock types are biased towards lithologies known to possess COMIS, and largely devoid of the pristine volcanic rocks conventionally required for paleointensity purposes. For the development of silicate-hosted paleomagnetism some technological difficulties must be confronted, including single crystal manipulation and orientation, narrow unblocking temperature Thellier-Thellier experiments, and the effects of extreme shape anisotropy on direction and intensity acquisition. Additional research must also expand our understanding of the thermodynamics and crystal chemistry of oxide exsolution within silicate minerals.