Skip to main content
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2003 Mar 3;100(6):3047–3050. doi: 10.1073/pnas.0437778100

Nuclear georeactor origin of oceanic basalt 3He/4He, evidence, and implications

J Marvin Herndon 1,*
PMCID: PMC152243  PMID: 12615991

Abstract

Nuclear georeactor numerical simulation results yield substantial 3He and 4He production and 3He/4He ratios relative to air (RA) that encompass the entire 2-SD (2σ) confidence level range of tabulated measured 3He/4He ratios of basalts from along the global spreading ridge system. Georeactor-produced 3He/4He ratios are related to the extent of actinide fuel consumption at time of production and are high near the end of the georeactor lifetime. Georeactor numerical simulation results and the observed high 3He/4He ratios measured in Icelandic and Hawaiian oceanic basalts indicate that the demise of the georeactor is approaching. Within the present level of uncertainty, one cannot say precisely when georeactor demise will occur, whether in the next century, in a million years, or in a billion years from now.

Keywords: helium|mantle|nuclear reactor|Earth core


Early in 1939, Hahn and Strassmann (1) published their discovery of nuclear fission. Later in the same year, Flügge (2) speculated on the possibility that self-sustaining nuclear fission chain reactions might have taken place under natural conditions within uranium ore deposits. Applying Fermi's nuclear reactor theory (3), in 1956 Kuroda (4) demonstrated the feasibility that thick seams of uranium ore might have undergone sustained nuclear fission 2,000 million years ago or earlier when the relative proportion of 235U was greater. In 1972, French scientists (5) discovered the intact remains of a natural nuclear fission reactor that had operated 1,800 million years ago in a 0.5-m-thick seam of uranium ore at Oklo, in the Republic of Gabon. Later other reactor zones were discovered in the region (6). In 1992, Herndon (7), applying Fermi's nuclear reactor theory, demonstrated the feasibility of planetary-scale nuclear fission reactors as energy sources for the giant outer planets, three of which radiate approximately twice as much energy as they each receive from the Sun. Beginning in 1993, Herndon (810) demonstrated the feasibility of a planetary-scale nuclear fission reactor at the center of the Earth as the principal energy source for the geomagnetic field and as a contributive energy source for other geodynamic processes, such as plate movement. In 2001, Hollenbach and Herndon (11) published results of numerical simulations of a deep-Earth nuclear fission reactor, conducted at the Oak Ridge National Laboratory in Oak Ridge, TN, which confirmed the previous considerations of Herndon (810) and demonstrated that 3He and 4He would be produced by the georeactor.

Clarke et al. (12) discovered that 3He and 4He are venting from the Earth's interior. The 3He/4He ratio of helium released to the oceans at midoceanic ridges is about eight times greater than in the atmosphere (R/RA = 8 ± 1, where R is the measured value of 3He/4He and RA is the same ratio measured in air = 1.4 × 10−6), and, therefore, cannot be ascribed to atmospheric contamination. Iceland plume 3He/4He values have been found (13) as high as ≈37 RA. Natural radioactive decay of uranium and thorium will lead to 4He production; but for three decades geophysicists have been unaware of any mechanism deep within the Earth that can account for substantial 3He production. Lacking knowledge of a deep-source production mechanism, deep-Earth 3He has been assumed to be of primordial origin (12, 13), trapped within the mantle at the time that the Earth formed. In the belief that deep-Earth 3He is primordial, various implications have been drawn concerning mantle structure and dynamics (14, 15). But the ratio of primordial 3He/4He is thought to be ≈10−4, a value inferred from gas-rich meteorites (16), which is ≈1 order of magnitude greater than helium released from the mantle. In ascribing a primordial origin to the observed deep-Earth 3He/4He, the assumption implicitly made is that the primordial component is diluted by a factor of ≈10 with 4He produced by the natural radioactive decay of U and Th in the mantle and/or in the crust. The alternative suggestion (17), that the 3He/4He arises instead from cosmic dust, subducted into the mantle, necessitates the assumption that the influx of interplanetary dust particles was considerably greater in ancient times than at present and also necessitates the assumption of a 10-fold dilution by 4He. Based on nuclear reactor numerical simulation results, Hollenbach and Herndon (11) have suggested instead that the observed deep-source helium is in fact the product of and evidence for a deep-Earth nuclear fission reactor (810).

Previous georeactor numerical simulations by Hollenbach and Herndon (11) were conducted at a single power level with the SAS2 analysis sequence contained in the SCALE Code Package from the Oak Ridge National Laboratory (18). Because these codes were developed for use with government and commercial nuclear reactors, cumulative fission yields are reported over time. The 3He/4He values published by Hollenbach and Herndon (11) were likewise cumulative. But instantaneous values are more geophysically representative and more revealing. One purpose of the present article is to present instantaneous helium fission yields ratios through steps in time at multiple power levels, thus facilitating comparison with 3He/4He ratios measured in deep-source lavas. Another purpose of the present article is to show that the nuclear reactor fission yield helium isotope ratios are not necessarily constant, but rather appear to be related to the extent of actinide fuel consumption at time of production. Still another purpose of the present article is to address the question of the georeactor lifetime and demise.

Methodology

The background as to why a large portion of the Earth's reservoir of uranium is expected to exist in the core, precipitate, and ultimately collect at the center of the Earth has been set forth in refs. 811 and stems from the deep interior of the Earth having a state of oxidation similar to the Abee enstatite chondrite (10). The numerical simulations presented in this article were conducted at the Oak Ridge National Laboratory by using the same computer codes and input parameters as described in Hollenbach and Herndon (11), the source to refer to for details.

Calculations were made with the SAS2 analysis sequence contained in the SCALE Code Package from the Oak Ridge National Laboratory (18) that has been developed over 30 years and has been extensively validated against isotopic analyses of commercial reactor fuels (1923). The SAS2 sequence invokes the ORIGEN-S isotopic generation and depletion code to calculate concentrations of actinides, fission products, and activation products simultaneously generated through fission, neutron absorption, and radioactive decay. The SAS2 sequence performs the 1D transport analyses at selected time intervals, calculating an energy flux spectrum, updating the time-dependent weighted cross sections for the depletion analysis, and calculating the neutron multiplication of the system.

With the exception of power levels, the values used as input to the SAS2 are the same used by Hollenbach and Herndon (11) and are as follows: initial volume of uranium = 5.6807 × 1017 cm3; initial atom ratio 235U/238U = 0.3038; uranium density = 36.84 g/cm3; steady-state fission power = 3.0 Terra-watts (TW) (3.0 × 1019 ergs/s), 5.0 TW, or 6.0 TW. Time steps of 2 × 106 years were used throughout. Reactor operation was assumed to have commenced 4.5 × 109 years ago and ceased when the effective neutron multiplication constant Keff < 1 (3). In each case, fission products were removed on formation; all 3H is assumed to have escaped the high neutron flux of the subcore reactor region before decaying to 3He.

Results and Discussion

From the Oak Ridge National Laboratory numerical simulations, values of the 3He/4He ratio, relative to the same ratio in air, RA, at each 2 × 106 year time step for each power level are shown in Fig. 1. For comparison, the range of values of the same ratio, measured in oceanic basalts, is shown in Table 1 at a 2σ confidence level. The entire range of values from oceanic basalts, shown in Table 1, are produced by self-sustaining nuclear fission chain reactions as demonstrated by the georeactor numerical simulations results presented in Fig. 1. The agreement is extremely strong evidence for a deep-Earth nuclear reactor and the solution of the three-decade-long mantle helium controversy and is unlike the alternative view, which rests on assumptions.

Figure 1.

Figure 1

Nuclear reactor numerical simulation results for three power levels showing the 3He/4He RAs produced during 2 × 106-year increments over the lifetime of the georeactor. Each data point represents the ratio of the 3He and 4He fission yields for a single time step. The pronounced upward trend of the data results from the continuing reduction of 238U, the principle source of 4He, by decay and breeding.

Table 1.

Statistics of 3He/4He relative to air (RA) of basalts from along the global spreading ridge system at a 2-SD (2σ) confidence level

Propagating lithospheric tears 11.75  ± 5.13 RA
Manus Basin 10.67  ± 3.36 RA
New rifts 10.01  ± 4.67 RA
Continental rifts or narrow oceans 9.93  ± 5.18 RA
South Atlantic seamounts 9.77  ± 1.40 RA
Mid-Ocean Ridge Basalt 8.58  ± 1.81 RA
EM Islands 7.89  ± 3.63 RA
North Chile Rise 7.78  ± 0.24 RA
Ridge abandoned islands 7.10  ± 2.44 RA
South Chile Rise 6.88  ± 1.72 RA
Central Atlantic Islands 6.65  ± 1.28 RA
HIMU Islands 6.38  ± 0.94 RA
Abandoned ridges 6.08  ± 1.80 RA

Adapted from ref. 24

In Fig. 1, the upward trend over time of the ratio data for each power level is principally the consequence of the gradual removal of 238U, the major source of 4He, by way of its natural decay and by its conversion to transuranic actinide fuels (a process of neutron absorption and β-decay termed fuel-breeding). For a particular power level, the highest 3He/4He values represent the most recent production, especially near the end of the nuclear fission lifetime of the georeactor.

The limitation on the upper limits for 3He/4He depends on the georeactor being critical, i.e., able to sustain chain reactions (3), as its actinide fuel approaches depletion. The main factors affecting that circumstance are the amount and nature of the initial actinide subcore and the operating history of the georeactor.

For the present investigation, no special efforts were made to extend the range of 3He/4He values, for example by assuming variable power levels over time or by including 232Th. One may reasonably expect, therefore, that the high values for 3He/4He, shown in Fig. 1, may not be true upper limits. As with the range of isotope ratios, the number of atoms of 3He and 4He produced by the georeactor numerical simulations over the lifetime of its criticality, as shown in Table 2, may likewise not be true upper limits. The initial uranium content used for the nuclear reactor numerical simulations is close to the maximum one might reasonably expect. Thorium, however, was not included because of uncertainties in its abundance in the core (11) and may provide additional fissile material by transmuting to 233U by neutron capture and double β-decay. But at the present time no one knows georeactor power level history, and, hence, fuel consumption in the past. Ultimately, one may hope to narrow the uncertainty by improved understanding of oceanic basalt helium data and a deeper knowledge of nuclear georeactor boundary conditions and dynamics.

Table 2.

For each power level, over-lifetime-of-georeactor production of 3He and 4He, in atoms, time of reactor demise, and over-lifetime-of-georeactor ratios of 3He/4He

3He atoms 4He atoms Demise in years 3He/4He RA
3 TW 1.73  × 1036 2.59  × 1041 5.6  × 109 4.77 RA
5 TW 2.21  × 1036 2.26  × 1041 4.4  × 109 6.99 RA
6 TW 2.39  × 1036 2.14  × 1041 4.0  × 109 7.98 RA

Previously, in the absence of knowledge of a deep-Earth production mechanism for 3He, the assumed primordial origin of 3He was essentially taken as fact with little justification. In light of the evidence presented for a deep-Earth nuclear reactor origin of the 3He/4He of oceanic basalts, the burden of proof now falls on those who would still argue for a primordial or cosmic origin to show in detail the specific geophysical circumstances whereby their individually assumed separate helium reservoirs, differing in space and time and differing by nearly an order of magnitude, mix to yield the relatively narrow range of 3He/4He values shown in Table 1.

Conclusions previously drawn relating to the geophysical implications of oceanic basalt helium data, for example, mantle degassing, should now be reassessed. Such reassessment is beyond the intent and scope of the present paper. Nevertheless, the subject of high 3He/4He values in certain measurements of so-called plumes, specifically Icelandic and Hawaiian, deserves comment.

For years efforts have been made to find unambiguously high 3He/4He values in plume-derived oceanic basalts (25, 26). A main motivation of those investigations, based on the assumed primordial origin of the 3He, was to find helium least diluted by 4He. Those investigations should be continued and encouraged, not for the original motivation, but because the high 3He/4He values may very well reflect the beginnings of the demise of the georeactor and should be investigated.

One shortcoming of oceanic basalt helium isotopic measurements is that the time of formation of the helium is unknown. But from Fig. 1, one can see that helium time of formation is important for assessing the time of demise of the georeactor. Efforts should be made to address that shortcoming, such as described below.

At the pressures that prevail within the Earth's core, density is a function almost exclusively of atomic number and atomic mass. Only very light elements might be able to escape from the core and find transport to the surface through some volcanic system. Helium is one example. When an actinide nucleus fissions, it typically splits into two heavy fragments. But once in approximately every 104 binary fission events, the actinide nucleus splits into three pieces, two heavy fragments and one very light fragment. Tritium (3H), which decays into 3He, is a light fragment from ternary fission. Other ternary fission products, which should be sought and which might be found in deep-source oceanic basalts, are shown in Table 3.

Table 3.

Potential in-core nuclear fission signatures in oceanic basalts

Isotopes Nuclear data Deep-earth data
3He, 4He Have Have
6Li, 7Li Have Need
9Be, 10Be Have Need
10B, 11B Need Need
20Ne, 21Ne, 22Ne Need Have

All of the isotopes shown in Table 3, with the exception of 10Be, are stable. Generally, light-element, ternary fission products, if radioactive, have very short half-lives. A notable exception, however, is 10Be, with a half-life of 1.5 × 106 years. Both 10Be and 9Be are produced by the georeactor with an initial ratio 10Be/9Be = 6. Although a major technological challenge, serious efforts should be made to find evidence of nuclear fission produced beryllium in high 3He/4He oceanic basalt samples and then to devise a means for using 10Be to obtain helium time-of-formation data.

In Fig. 1, the 3-TW, 5-TW, and 6-TW nuclear reactors cease to maintain criticality at 5.6, 4.4, and 4.0 gigayears, respectively. That these times are very close to the present epoch in the lifetime of the Earth may well be cause for concern. The long-standing idea that the Earth will continue much as it has for at least another 4.5 gigayears stems from the 1940 reasoning of Birch, who could not have known of the implications (27) resulting from the 1960's discovery of nickel silicide and silicon-containing metal in enstatite chondrite meteorites. The data presented in Fig. 1 show that terminal failure of the georeactor is approaching, but that time frame is not well defined, considering the uncertainties, and might be as short as 102 years or as long as 109 years.

Conclusions

The helium observed for the past three decades in oceanic basalts has been demonstrated to have been produced by a nuclear reactor at the center of the Earth. The nuclear georeactor numerical simulation results, even for the simple, preliminary cases shown, yield a narrow range of 3He/4He RAs that encompass the entire 2-SD (2σ) confidence level range of tabulated (24) measured 3He/4He ratios of basalts from along the global spreading ridge system and lead to substantial 3He and 4He production.

Nuclear georeactor produced 3He/4He ratios are not necessarily constant, but rather appear to be related to the extent of actinide fuel consumption at time of production. High 3He/4He ratios are produced near the end of the georeactor lifetime.

Nuclear georeactor numerical simulation results and the observed high 3He/4He ratios measured in Icelandic and Hawaiian oceanic basalts indicate that the demise of the georeactor is approaching, but the time is not yet precisely determined. As the georeactor dies, the geomagnetic field that it presumably powers after a time will begin to collapse. But unlike previous geomagnetic collapses, that have restarted and re-energized the field, a time will come when the actinide fuel of the georeactor is too diminished to initiate self-sustaining neutron-induced chain reactions; the georeactor will die and sometime thereafter the geomagnetic field will die and will not restart. At some point in time after the georeactor dies, there will be no geomagnetic field and life on Earth will never be the same. The challenge now is to determine precisely the time of georeactor demise. Within the present level of uncertainty, one cannot say whether that time will come in the next century, in the next millennium, in a million years, or in a billion years. But one thing is certain: georeactor demise will occur.

Acknowledgments

High praise and deep appreciation are extended to the Oak Ridge National Laboratory and particularly to Drs. D. J. Hill, D. F. Hollenbach, and C. V. Parks for graciously assisting a small business conducting unfunded, not-for-profit, but important, basic research.

Abbreviations

RA

ratio relative to air

TW

Terra-watt

References

  • 1.Hahn O, Strassmann F. Naturwissenschaften. 1939;27:11. [Google Scholar]
  • 2.Flügge F. Naturwissenschaften. 1939;27:402. [Google Scholar]
  • 3.Fermi E. Science. 1947;105:27–32. doi: 10.1126/science.105.2715.27. [DOI] [PubMed] [Google Scholar]
  • 4.Kuroda P K. J Chem Phys. 1956;25:781–782. [Google Scholar]
  • 5.Neuilly M, Bussac J, Fréjacques C, Nief G, Vendryes G, Yvon J. C R Acad Sci Paris. 1972;275:1847–1849. [Google Scholar]
  • 6.Gauthier-Lafaye F, Holliger P, Blanc P L. Geochim Cosmochim Acta. 1969;60:4831–4852. [Google Scholar]
  • 7.Herndon J M. Naturwissenschaften. 1992;79:7–14. [Google Scholar]
  • 8.Herndon J M. J Geomagn Geoelectr. 1993;45:423–437. [Google Scholar]
  • 9.Herndon J M. Proc R Soc London Ser A. 1994;445:453–461. [Google Scholar]
  • 10.Herndon J M. Proc Natl Acad Sci USA. 1996;93:646–648. doi: 10.1073/pnas.93.2.646. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Hollenbach D F, Herndon J M. Proc Natl Acad Sci USA. 2001;98:11085–11090. doi: 10.1073/pnas.201393998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Clarke W B, Beg M A, Craig H. Earth Planet Sci Lett. 1969;6:213–220. [Google Scholar]
  • 13.Hilton D R, Grönvold K, Macpherson C G, Castillo P R. Earth Planet Sci Lett. 1999;173:53–60. [Google Scholar]
  • 14.O'Nions P K. J Geol Soc London. 1987;144:259–274. [Google Scholar]
  • 15.McDougall I, Honda M. In: The Earth's Mantle. Jackson I, editor. Cambridge, U.K.: Cambridge Univ. Press; 1998. pp. 159–190. [Google Scholar]
  • 16.Pepin R O, Singer P. Science. 1965;149:253–265. doi: 10.1126/science.149.3681.253. [DOI] [PubMed] [Google Scholar]
  • 17.Anderson D L. Science. 1993;261:170–176. doi: 10.1126/science.261.5118.170. [DOI] [PubMed] [Google Scholar]
  • 18. Computational Physics and Engineering Division (April 1995) SCALE: A Modular Code System for Performing Standardized Analyses for Licensing Evaluations, NUREG/CR-0200, Rev. 4, ORNL/NUREG/CSD-2/R4, Vols. I, II, and III, CCC-545 (Radiation Safety Information Computational Center, Oak Ridge National Laboratory, Oak Ridge, TN).
  • 19.Hermann O W, Bowman S M, Brady M C, Parks C V. Validation of the SCALE System for PWR Spent Fuel Isotopic Composition Analyses, ORNL/TM-12667. Oak Ridge National Laboratory, Oak Ridge, TN: Martin Marietta Energy Systems; 1995. [Google Scholar]
  • 20.DeHart M D, Hermann O W. An Extension of the Validation of SCALE (SAS2H) Isotopic Predictions for PWR Spent Fuel, ORNL/TM-13317. Oak Ridge National Laboratory, Oak Ridge, TN: Lockheed Martin Energy Research; 1996. [Google Scholar]
  • 21.Hermann O W, DeHart M D. Validation of SCALE (SAS2H) Isotopic Predictions for BWR Spent Fuel, ORNL/TM-13315. Oak Ridge National Laboratory, Oak Ridge, TN: Lockheed Martin Energy Research; 1998. [Google Scholar]
  • 22.Hermann O W. San Onofre PWR Data for Code Validation of MOX Fuel Depletion Analyses, ORNL/TM-1999/018, R1. Oak Ridge National Laboratory, Oak Ridge, TN: Lockheed Martin Energy Research; 2000. [Google Scholar]
  • 23.England T R, Wilson R E, Schenter R E, Mann F M. Summary of ENDF/B-V Data for Fission Products and Actinides, EPRI NP-3787 (LA-UR 83–1285) (ENDF-322) Palo Alto, CA: Electric Power Research Institute; 1984. [Google Scholar]
  • 24.Anderson D L. Geophys Res Lett. 2000;27:2401–2404. [Google Scholar]
  • 25.Kurz M D, Jenkins W J, Hart S R, Clague D. Earth Planet Sci Lett. 1983;66:388–406. [Google Scholar]
  • 26.Honda M, McDougall I, Patterson D B, Doulgeris A, Clague D A. Geochim Cosmochim Acta. 1993;57:859–874. [Google Scholar]
  • 27.Herndon J M. Phys Earth Planet Inter. 1998;105:1–4. [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES