Distribution and Concentration Changes of Fe-Ti Oxides Due to Heating 500-700C of a Basaltic Rock


Risa Madoff
University of Rochester
Department of Earth and Environmental Sciences



Through the use of SEM, x-ray spectral analysis and x-ray dot maps, distributional and concentration changes in Fe-Ti oxides in grains from rocks sampled from basaltic dikes, were tracked through heating from 500-700 C. Regions of Ti-rich and Ti-poor, Fe-rich, were detected, imaged, and analyzed, at each heating interval. The dot maps displayed increases in concentration of the original element detected in the regions distinguished as temperature increased. X-ray specra analysis gave a quantification of the concentration changes by showing changes in the abundances of Fe and Ti in the two regions analyzed.


Ferromagnetic minerals in mafic igneous rocks, formed during the cooling of magma, record direction and intensity of the earths magnetic field. In paleomagnetism, the study of the earths magnetic field in geologic time, the characteristics both of the magnetic acquisition in rocks and of the magnetic field, can be understood by measuring magnetic properties of Fe-Ti oxides titanomagnetites and titanohematites in rock samples. Laboratory heating procedures are used to remove overprinting of magnetic fields recorded subsequently to the original in order to determine the original characteristic remanent magnetization (ChRM) of a sample of a certain age. However, during the original cooling and possibly during laboratory heatings, Fe-Ti oxides can undergo oxidation and exsolution, alteration processes which can affect magnetic properties recorded in the rocks. Because changes in Fe-Ti oxide concentration and abundances within a grain can affect grain size and composition, two important factors affecting the ability of a grain to acquire and retain a recording of the magnetic field, measuring these properties can help determine the suitability of a sample for use in a paleomagnetic study. Scanning electron microscopy and x-ray dot maps were used to investigate changes in the abundance and concentration of Fe-Ti oxides in individual grains resulting from heating 500-700C


Sample Prep.

A coating applied by gold sputtering on a polished basaltic "rock chip", 1 cm in diameter and 3 mm thick, taken from a hand-drilled core, prepared the sample for the SEM. To track the same grains through repeated trials, a diamond etcher was used to mark the region on the chip for investigation.

Secondary Electron Imaging

Secondary electron images, taken at different scales during each trial, were used to compare changes in grain size and intergrowth spacing.

Back Scatter Imaging

Back scatter images were used at the highest magnification needed to differentiate Ti-poor and Ti-rich regions.

X-ray Detection

Through the use of EDAX, x-ray spectra of Ti and Fe were selected. The atomic weight percent of each element was used to determine the x parameter of titanomagnetite's solid solution line between magnetite and ulvospinel.

X-ray Dot Maps

X-ray dot maps, colored bit maps of each of the selected elements identified in the x-ray spectra, were generated and used to track the regions and concentration changes of Fe and Ti in the grains.





The same grain at the same low magnification. Narrow bands are Ti-rich regions, while the rest is Ti-poor or Fe-rich. Temperature differences - heating to 550C on the left and heating to 700C on the right - do not show significant grain size changes or structural differences at this magnification (except for some fracturing in the right image). But the x-ray spectra and dot map tell a different story....


This triangular portion, now with magnification x3000, is able to show the concentrations of the Ti and Fe regions by means of the proximity of dots in the bit map. Also, given the atomic weight percents, it is shown that the dark narrow regions, colored red in the dot map, are primary titanohematite (x = 0.8) or illmenite, while the broad regions, colored green in the dot map, consist of primary titanomagnetite (x = 0.2). This image is from the 575 C heating. Compare with the more diffuse spacing of dots in the image below.


A grain of similar size and appearance at only 500C showed more diffuse spacing of bit map dots for each of the elements detected by the x-ray spectra. Also, when calculated, the broad green areas consisted of primary titanomagnetite (x = 0.3), while the dark bands consisted of paramagnetic titanohematite (x = 1.43). For paramagnetic grains, magnetization equals 0 when the magnetizing field is removed.


The back-scatter detector was used in conjunction with the maximum magnification possible to isolate narrow regions so that the element involved could be detected with as little overlap as possible with the neighboring element. The lighter regions, of the atomically heavier element Fe, in the back-scatter image on the left, are more clearly distinguishable from the darker regions consisting of the atomically lighter element Ti, than when compared with the secondary electron image on the right.

Data Analysis

Calculating the parameter, x = (3*Ti atomic weight %)/(atomic weight % Fe + atomic weight % Ti), gives a value 0-1, showing where on a solid solution line between magnetite and ulvospinel, the chemical formula for titanomagnetite falls.

Fe (3-x)Ti (x)O 4

0 < x < 0.8 primary titanomagnetite

0.8 < x < 0.95 primary titanohematite

x >1 paramagnetic ( magnetization = 0)







Fresh sample






















From the atomic weight percents given by the x-ray spectra analysis, changes in the Fe:Ti were calculated. The x values together with the dot maps show that the Ti-poor, Fe-rich, regions become more concentrated with increased temperatures, but for the most part remain titanomagnetite. The narrow bands of Ti-rich intergrowths also become more concentrated through heating, but are paramagnetic (magnetization = 0) nearly throughout.


Narrow bands of paramagnetic Ti-rich intergrowths in grains of Fe-Ti oxides were found to occur from room temperature to 700 C in a single rock sample. The broad Ti-poor, Fe-rich regions, probably remain as stable carriers of the ChRM up until 600 C. At 700 C there is a large drop in Ti. The degree to which the presence of paramagnetic regions would affect the overall magnetic recording of the particular sample would be difficult to determine, as measurements of magnetic data are averaged for a whole sample in laboratory procedures. The usefulness of this kind of information would be more evident in comparative studies between samples from rocks at different locations or that underwent a different magmatic process. Also, an inventory of the composition of the grains generally occurring in each sample would contribute to understanding Fe-Ti oxide variations between genetically different events.


The ability of SEM to detect and image Fe-Ti oxides in regions smaller than 10 μm can contribute to advancing a distributional and compositional anlaysis of ferromagnetic grains in igneous rocks used for paleomagnetic studies. Through the use of various detection and imaging techniques, x-ray spectra analysis, and x-ray dot maps, such analysis can track the distribution and concentration of Fe-Ti oxides in ferromagnetic grains of a single sample and thereby increase the ability to make comparisons between samples within a study region. Also, the temperatures at which significant changes occur in such grains can be determined, aiding in the process of selecting samples that are the most stable recorders of a magnetic field.


Background Reading

Butler, R. F., 1992, Paleomagnetism: Magnetic Domains to Geological Terranes, Blackwell: Boston.

Postek, M. T., Howard, K. S., Johnson, A. H., McMichael, 1980, Scanning Electron Microscopy, Ladd Research Industries: Vermont.

Slayter, E. M. and Slayter, H.S., 1992, Light and Electron Microscopy, Cambridge University Press: Cabridge.

Smirnov, A. V. and Tarduno, J. A., in submission, Thermochemical remanent magnetization in Precambrian rocks: Are we sure the geomagnetic field was weak?, Journal of Geophysical Research.



Please enter any comments, criticisms, questions, etc. below.

Your name:

Email address: