KGS Home Geology Home

Kansas Geological Survey, Bulletin 199, pt. 1, originally published in 1970

Dating Conodonts Using Electron Spin Resonance: A Possible Technique

by Maurice Morency, Pierre L. Emond, Peter H. von Bitter

Originally published in 1970 as part of Kansas Geological Survey Bulletin 199, pt. 1, p. 17-19. This is, in general, the original text as published. The information has not been updated. An Acrobat PDF version of the complete bulletin (5 MB) is also available.


Electron spin resonance is a means of detecting, with microwaves, the presence of electrons trapped in the structurc of crystalline material. Prior to being trapped, the electrons are released from their parent atoms by radiation from radioisotopes present in the conodonts and the host rock.

The feasibility of using electron spin resonance as a dating technique was tested on conodonts (fossil tooth-like structures) from the Holts Summit Formation (upper Devonian), Missouri, the Bushberg Formation (upper Devonian to lower Mississippian), Missouri, and the Heebner Shale Member of the Oread Limestone (upper Pennsyvanian). All of the conodonts yielded an electron spin resonance spectrum arising principally from trapped electrons. It was found that this spectrum would be suitable for the age determination calculations described herein.


Interest in the use of electron spin resonance as a dating technique arises from two principal considerations: The electron spin resonance signal produced by phosphate minerals is relatively simple; it is proportional to the total radiation received. The results presented are only preliminary. A more detailed investigation is under way using a series of samples that range in age from lower to upper Pennsylvanian.

Electron spin resonance refers to the magnetic resonance of a permanent dipole moment. [Note: A rough macroscopic analogy might be that of a spinning top precessing about the direction of a gravitational field.] Pictorially, an electron with moment μ, when placed in a constant magnetic field H0 precesses about the direction of H0 at a constant angle θ with a natural frequency given by:

ω = γH0 (Larmor precession),

where γ is a constant.

If we apply microwave radiation, which is a small rotating magnetic field 'h' in a direction perpendicular to H0 and having the same frequency as that of the precessing electron, θ will increase as the system absorbs power from the incident microwaves. Under these conditions the system is said to be "on resonance." Experimentally, we fix the frequency of the incident microwaves and vary H0 such that as we go through resonance the power absorbed by the sample is recorded by observing the variation of the power level of the microwaves. As we have assumed a paramagnetic system (one in which the electrons do not interact), any measure of the power absorbed (e.g., peak height, width at half-power points) will be a relative measure of the number of trapped electrons in the sample. Background radiation produces free electrons which wander through the crystalline structure and are eventually trapped in defects called electron traps. Traps are various types of crystalline imperfections introduced during the crystallization process. They are vacant lattice points, impurity atoms, interstitial atoms, and the like. Alpha and beta radiation can also cause defects, but according to Zeller, Levy, and Mattern (1967) the formation of additional defects will be negligible if the total dose is low.

The procedure employed in dating the conodonts consists of first determining the electron spin resonance signal induced by the natural background radiation and then subjecting the sample to small additional doses of artificial gamma radiation in order to establish the dependence of signal growth on radiation dose. If the signal growth is linear, we can, by assuming a constant natural dose rate, make a linear extrapolation to determine the total natural radiation dose received by the sample (Fig. 1).

Figure 1--Hypothetical age (determination diagram (after Zeller, Levy, and Mattern, 1967).

Natural signal is extended by radiation dosage; linear trend can be extended backward to estimate prior dose.

The background radiation, and hence the dose rate, is measured with a solid-state detector. Thus:

Total natural dose (Rads) / Dose rate (Rads/year) = Age (years)

The composition of conodonts places them in the dahlite-francolite isomorphous series (Deer, Howie, and Zussman, 1962). A prior investigation of minerals suitable for electron spin resonance dating (Zeller, 1968) yielded an age for an apatite sample whose composition approximated that of the aforementioned series. Since conodonts show essentially the same x-ray diffraction pattern as members of the dahlite-francolite series (Lindström, 1964), similar results to those obtained by Zeller could be expected.

The initial work was done on conodonts from the Holts Summit Formation (sandstone) of upper Devonian age, and a subsequent study on conodonts of the Bushberg Formation (sandstone) (Fig. 2).

Figure 2--Absorption peaks for the Holts Summit and Bushberg formations as a function of 60Co gamma-ray dose. The signal has been extrapolated to determine the prior doses. Insert shows typical differential power absorption signal obtained for the Holts Summit and Bushberg formations.

Holts Summit has lower maximum absorption than Bushberg; Holts Summit extrapolates back to a prior dose of 1; Bushberg extrapolates back to a prior dose of 4.5.

The natural signal intensity was measured and plotted on the vertical axis at dose = 0. The sample was then subjected to successive doses of gamma radiation to establish the signal growth-rate. The signal response to radiation is linear until saturation of the traps is achieved (Fig 2). An interesting feature of the results obtained from the Bushberg sample is the marked decrease in signal intensity once initial saturation has been attained. Only unpaired electrons will contribute to the electron spin resonance signal. Since pairing of electrons will be more likely to occur once trap saturation is achieved (Vaz and Zeller, 1966), increased pairing will thus tend to diminish the signal intensity.

Conodonts from the Heebner Shale Member (Oread Limestone) were studied primarily because of the presence of high background radioactivity associated with black shales. Because of the presence of high amounts of uranium and thorium in their environment, the radiation damage, caused mainly by alpha particles, is detectable. This is reflected in the electron spin resonance signal emitted by the conodonts from the Heebner shale. The differential power absorption signal obtained from the Heebner conodonts (Fig. 3) shows several peaks occurring at different field strengths. The Holts Summit and Bushberg conodonts yielded broad single absorption peaks. These characteristics reflect the nature and distribution of types of traps within the samples. Thermal annealing experiments quantitatively support this explanation.

Figure 3--Electron spin resonance absorption curves for the conodonts of the Heebner after irradiation in 60Co gamma-ray source and heat treatment.

Absorption is higher for Heebner samples irradiated for 30 minutes than 45; heated samples are lowest; sample irradiated 30 minutes has two even spikes, 45-minute sample has one high spike and one lower spike.


From the results presented above we can say:

  1. Conodonts yield measurable electron spin resonance signals suitable for use in the age determination technique described above.
  2. From irradiation and thermal annealing experiments we know that the signals obtained arise from trapped electrons.
  3. The signal response to radiation is linear until trap saturation is achieved.

One must keep in mind that after the conodonts are deposited in a particular envronment, heat, pressure, and other geological processes can alter both trap population and number of traps.


We wish to express our gratitude to E. J. Zeller and A. J. Rowell, of the University of Kansas, for their interestiiig and stimulating discussions and the State Geological Survey of Kansas for the use of their facilities. We gratefully acknowledge the financial assistance from the U.S. Atomic Energy Commission under contract number AT(11-1)1057-7.


Deer, W. A., Howie, R. A., and Zussman, J., 1962, Rock-Forming Minerals: v. 5, Non-silicates: Longmans, London, p. 323-338.

Lindström, Maurits, 1964, Conodonts: Elsevier Publishing Co., Amsterdam, 196 p.

Vaz, J. E., and Zeller, E. J., 1966, Thermoluminescence of calcite from high gamma radiation doses: Am. Mineral. v. 51, p. 1156-1166.

Zeller, E. J., 1968, chap. 5.1, Use of electron spin resonance for measurement of natural radiation damage; in, Thermoluminescence of Geological Materials, D.J. McDougall, ed.: Academic Press, London, p. 271-279.

Zeller, E. J., Levy, P. W., and Mattern, P. L., 1967, Geologic dating by electron spin resonance; in, Radioactive Dating and Methods of Low-Level Counting: Proceedings of symposium, International Atomic Energy Agency in Cooperation with joint Commission on Applied Radioactivity (ICSU), Monaco, March 2-10, 1967, International Atomic Energy Agency, Vienna, p. 531-540.

For additional information relating to this subject see:

Levy, P. W., 1968, chap. 2.2, A brief survey of radiation effects applicable to geology problems; in, Thermoluminescence of Geological Materials, D.J. McDougall, ed.: Academic Press, London, p. 25-38.

Slichter, C. T., 1963, Principles of Magnetic Resonance: Harper & Rowe Publishers, New York, 246 p.

Kansas Geological Survey, Dating Conodonts Using Electron Spin Resonance: A Possible Technique
Placed on web May 11, 2009; originally published in March 1970.
Comments to
The URL for this page is