This report will explore the subject of naturally-occurring nuclear fission reactions, focusing on the reactors of Gabon, Africa, which were operational approximately two billion years ago. This is intended to complement a past report, with this report having an emphasis on the following topics: the analysis of Xe isotopes from the Gabon sites to elucidate the operation of the reactors, the natural reactors as models for the management of modern nuclear fission waste, and the implication of the reactors' operation on the possible variance of the fine-structure constant (α) over time. 
In 1972, a researcher analyzing uranium from the Oklo deposit in Gabon, Africa discovered that the samples contained a U-235 content of only 0.717 percent, significantly less than the 0.720 percent typical of modern uranium ore. Further, researchers found that at least one portion of the Oklo deposit was depleted, with approximately 200 kg of U-235 missing.  In addition, the depleted region of the deposit was found to be rich in nuclear fission products. This suggested that the uranium deposit had undergone a self-sustaining fission reaction that consumed the missing U-235 and generated fission products. [2,3]
The idea of a naturally-occurring fission reaction had been postulated in 1953 by Wetherwill and Inghram and further studied in 1956 by Kuroda. [2,4,5] As discussed elsewhere, the Gabon sites (16 in total, between the Oklo mine and the neighboring Okelobondo uranium mine) satisfied the conditions (e.g., appropriate size, presence of a moderator, and absence of neutron poisons) that had been proposed for natural fission reactors.  One criterion is particularly of note - while a modern, natural uranium deposit could not become a reactor due to its decreased U-235 content (0.720 percent), at the time that the Gabon reactors were operational, natural uranium contained approximately 3 percent U-235. This higher concentration enabled the self-sustaining fission reactors; in fact, it is comparable to the level of U-235 enrichment of the uranium fuel used in most modern nuclear power stations. 
Meshik and co-workers studied (via mass spectrometry) the isotopic composition of xenon from Oklo rock, in order to probe the nature of the Gabon reactors. The researchers initially made two surprising discoveries. First, they observed that the xenon was not largely located in uranium-rich mineral grains as they had anticipated but instead was found in aluminum phosphate minerals. Second, they observed that the isotopic distribution differed from that of modern nuclear reactors. Of the nine stable isotopes of xenon, the material from Oklo was depleted in Xe-136 and Xe-134.  Meshik and co-workers recognized that none of the xenon isotopes were produced directly by the fission of uranium itself but, instead, were formed by the decay of other fission products (specifically, radioactive iodine and tellurium). Thus, they recognized that the formation of the various xenon isotopes would be dependent upon the lifespans of their precursors. This would result in Xe-136 forming within one minute of the initiation of the self-sustained fission reaction, with Xe-134 forming after one hour, with Xe-132 and Xe-131 forming within days, and with Xe-129 only forming after millions of years. 
These key insights led Meshik and co-workers to propose that the Gabon reactors likely operated through a series of "on/off" cycles. [2,6] Essential to this conclusion was the proposal (by Meshik and others) that the reactors were moderated by water. [2,3] In the presence of water, neutrons are slowed to thermal energies, and a fission chain reaction can occur. However, the heat generated by the reaction would eventually boil the water away, halting fission until groundwater returned. [2,3] This cycle explains the xenon isotope distribution: when the reactor was operating, the rapidly-formed Xe-134 and Xe-136 gases were driven off, but the precursors that would ultimately yield Xe-132, Xe-131, and Xe-129 were incorporated into aluminum phosphate minerals as the reactor cooled, during an "off" cycle. This explains the absence of Xe-134 and Xe-136 from these aluminum phosphate minerals. Moreover, this dependence on a water moderator could also explain the absence of xenon from the uranium-rich mineral grains: the water would have washed away the water-soluble tellurium and iodine isotopes that would have led to xenon. 
From the Xe-131 / Xe-134 and Xe-132 / Xe-134 ratios of the aluminum phosphate from the Oklo site, Meshik and co-workers calculated the operating schedule for the Oklo reactors. Their calculations indicated a 30-minute "on" period of fission (with concomitant boiling of the water moderator) followed by a 2-hour, 30-minute "off" cooling period before water returned and the reactor again became self-sustaining. [2,6]
As the Gabon reactors produced large amounts of radioactive fission waste products, they have been studied as a natural model for fission waste management. Meshik has highlighted the ability of aluminum phosphate minerals to store fission products, including gaseous products, for billions of years. Such storage could be an alternative to the venting of radioactive gases from nuclear plants into the atmosphere.  In addition, Gauthier-Lafaye has examined the topic in great detail. Specifically, Gauthier-Lafaye has noted that many fission products from the Oklo reactors (e.g. Pu, Th, Zr, Ru, Rh, Pd, and rare-earth elements [REE]) have been retained, by virtue of their high solubility in uraninite and low solubility in groundwater. Furthermore, other matrices such as apatites, chlorites, and clay minerals have been implicated in the storage of actinides and fission products.  The rate of dissolution of the Oklo uraninite and the subsequent release of its stored fission products has also been discussed. Gauthier-Lafaye notes that although uraninite dissolution has been observed, the present existence of uraninite (two billion years after the reactors were operational) suggests that the dissolution rate is very low. This high stability of uraninite has been explained by the presence of a reducing environment (mediated by Fe(II) minerals) and by natural protection of the material by clay minerals and organic material. 
Thus, the Oklo reactors constitute a useful long-term study on the storage of nuclear fission products, and the fact that the products have been stored so effectively by these various means suggests that similar storage techniques could possibly be employed for the storage of fission waste from modern reactors.
The fine-structure constant (α) is a fundamental physical constant which affects a multitude of physical phenomena. Its relation to other physical values and its approximate value are as follows: 
As Barrow and Webb note, the precise value of this constant has a significant impact on physical events: the value affects the density of solid matter, the temperatures of chemical bond dissociations, and the stability of nuclei. If the value of α were to become greater than 0.1, nuclear fusion would be impossible. Furthermore, a shift of just four percent in the value of α would yield a change in the energy levels of carbon nuclei so dramatic that its production in stars would not occur.  Given that α has such in importance to fundamental physical processes, there has been significant interest in studying how the value of α may have changed over time. With nuclear processes having such a strong dependence on even minute variations in α, the Gabon reactors play an important role in determining the degree to which α may have changed over a very long period of time.
In 1976, Shlyakhter recognized that the operation of the Oklo reactors was dependent upon the ability of Sm-149 to undergo neutron capture. He also he recognized that, in order for this process to have occurred at Oklo, it was necessary for a value of a particular resonant energy level of the Oklo Sm-149 to be very similar to its present value. This resonant energy level is dependent upon α and, thus, is sensitive to changes in α. [8-11] From the isotopic composition of samples from Oklo, the capture cross-section value (at the time that the reactors were operational), as well as the possible range of its variation, could be determined. From this, the upper boundaries for the range of variation in the resonance energy level and, therefore, in α, could be determined. . Shlyakhter thus determined that the rate of change of α (specifically, [rate of change of α]/[α]) must be < 10-17 / year. [11,12] In 2004, Lamoreaux and Torgerson analyzed the Oklo reactor data and its implications on the time variation of α.  A key aspect of their analysis was the recognition that the uranium present in the Oklo reactor could act as a 1/v neutron absorber, a factor which alters the neutron energy spectrum of the reactor. Taking this into account, Lamoreaux and Torgerson calculated that the value of α has decreased over the past two billion years. Specifically, they calculated the |[rate of change of α]/[α]| to be < 3.8 x 10-17 / year (95% confidence).  In 2006, Petrov and co-workers published a comprehensive analysis in which a complete computer model of one of the Oklo reactors (RZ2) was constructed.  Using this model, the researchers determined the averaged cross section for Sm-149, which was then used to determine the limits on the variation of α. Petrov and co-workers determined these limits to be -3.7 x 10-17 / year < [rate of change of α]/[α] < 3.1 x 10 -17 / year.  Thus, their analysis indicates that a distinct increase or decrease in α over the past two billion years is not a certainty.
Petrov and co-workers commented that "the discovery of the Oklo natural nuclear reactor in Gabon (West Africa) in 1972 was possibly one of the most momentous events in reactor physics since 1942, when Enrico Fermi and his team achieved an artificial self-sustained fission chain reaction."  As discussed herein, the implications of this event have been significant and far-reaching. The discovery, geological distribution, and isotopic pattern of fission products (namely, xenon) led to significant insight into the operation of the Gabon reactors, which may find additional utility should evidence of other natural reactors be found in the future. The discovery that the reactor fission products were well-contained and were resistant to dissolution in groundwater may suggest methods of managing modern nuclear waste products. Finally, the very fact that the reactors were operational at all has stimulated discussion regarding the time variation of a fundamental physical constant, leading to comprehensive analyses of the reactor data in an effort to answer this important question. Given the manifold consequences of these natural fission reactors, the statement by Petrov and co-workers rings true, with the story of natural nuclear fission being one of special interest within the realm of nuclear physics.
© James T. Masters. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 T. Blake, "Fission in Gabon," Stanford University, Physics 240, Fall 2011.
 A. P. Meshik, "The Workings of an Ancient Nuclear Reactor," Scientific American 293, No. 5, 82 (November 2005).
 G. A. Cowan, "A Natural Fission Reactor," Scientific American 235, No. 1, 36 (July 1976).
 G. W. Wetherill, "Spontaneous Fission Yields from Uranium and Thorium," Phys. Rev. 92, 907 (1953).
 P. K. Kuroda, "On the Nuclear Physical Stability of the Uranium Minerals," J. Chem. Phys. 25, 781 (1956).
 A. P. Meshik, C. M. Hohenberg, O. V. Pravdivtseva, "Record of Cycling Operation of the Natural Nuclear Reactor in the Oklo/Okelobondo Area in Gabon," Phys. Rev. Lett. 93, 182302 (2004).
 F. Gauthier-Lafaye, "2 Billion Year Old Natural Analogs for Nuclear Waste Disposal: The Natural Nuclear Fission Reactors in Gabon (Africa)," Comptes Rendus Physique 3, 839 (2002).
 J. D. Barrow and J. K. Webb, "Inconstant Constants," Scientific American 292, No. 6, 56 (June 2005).
 J. D. Barrow, "Cosmology and Immutability," in Science and Ultimate Reality: Quantum Theory, Cosmology, and Complexity, ed. by J. D. Barrow, P. C. W. Davies, and C. L. Harper, Jr. (Cambridge U. Press, 2004).
 J. D. Barrow, Constants of Nature (Pantheon, 2003).
 A. I. Shlyakhter, "Direct Test of the Constancy of Fundamental Nuclear Constants," Nature, 264, 340 (1976).
 Y. V. Petrov et al., "Natural Nuclear Reactor Oklo and Variation of Fundamental Constants: Computation of Neutronics of Fresh Core," Phys. Rev. C. 74, 064610 (2006).
 S. K. Lamoreaux and J. R. Torgerson, "Neutron Moderation in the Oklo Natural Reactor and the Time Variation of α," Phys. Rev. D. 69, 121701 (2004).