Neon illuminates the mantle

The outer Earth grew largely from material added by impacts from planetesimals, rather than by capture of dust grains from the solar nebula — or at least that's the inference from the latest geochemical analyses.

A record of Earth's formation and its evolutionary history during the past 4,500 million years is preserved within the chemical and isotopic composition of the mantle. Fluids and the magmas expelled at the Earth's surface as basalt rocks provide samples for deciphering this record. In particular, isotopes of the noble gases contain unique clues to the structure of the mantle, the formation of the hydrosphere and atmosphere, and the history of the building blocks used during our planet's accretion. On page 33 of this issue, Ballentine et al.1 provide high-precision measurements of neon and helium isotopes in carbon-dioxide-rich well gases from New Mexico. Their results illuminate all of these issues, and have profound implications for our understanding of Earth's accretion history.

The initial (primordial) noble gases in the Earth were either trapped directly from a gas-rich solar nebula, or implanted as ions during intense irradiation by a young Sun2. Terrestrial noble gases differ in their isotopic make-up from primordial values because they have been modified by the radioactive decay of uranium (U), thorium (Th) and potassium (K), the major heat-producing elements. The ratio of primordial to radiogenic noble gases in Earth's mantle therefore reflects the time-integrated ratio of primordial noble gas to U, Th and K. For example, the relatively high ratios of helium isotopes (3He/4He) observed in ocean island basalts (OIBs) from localities such as Hawaii and Iceland indicate a mantle source that is characterized by high 3He/(U+Th). This OIB source has a higher 3He/4He than that of mid-ocean-ridge basalts (MORBs), and is therefore less degassed and generally considered to lie somewhere below the upper mantle3.

Support for this model is found by comparing the neon-isotope compositions of OIBs and MORBs4,5,6,7,8,9. Elevated 21Ne/22Ne is a result of 21Ne production by nuclear processes involving the collision of energetic α-particles (4He atoms produced by U and Th radioactive decay) with 18O in mantle silicates—the silicon- and oxygen-rich rocks that make up most of the mantle. Hence, the trend in OIBs from Hawaii and Iceland6,7,8, towards high 20Ne/22Ne and low 21Ne/22Ne when compared with MORBs4,5,9 (Fig. 1), is consistent with a deep, relatively undegassed ‘mantle plume’ source beneath those ocean islands. Elevated 20Ne/22Ne cannot be explained by nucleogenic processes, and is attributed to the presence of a solar neon component in the Earth4,5,6,7,8,9,10. A major goal is therefore to identify the upper limit for 20Ne/22Ne in various parts of the mantle, because this potentially distinguishes between different accretion scenarios for the Earth7.

Figure 1: A three-component isotope mix.

The diagram illustrates the neon-isotope compositions for air, solar energetic particles, Ne-B (the neon component in meteorites that underwent ion implantation by solar energetic particles) and the solar wind. Basalt rocks from mid-ocean ridges (MORBs) and ocean islands (OIBs) have air as one end-member because they contain atmospheric contamination released during mass spectrometric analysis. The OIBs from Hawaii and Iceland define a mixing line between air and a deep-mantle component similar to that of the solar wind, and MORBs define a mixing line between air and a primordial neon-isotope component that has been modified by addition of nucleogenic 21Ne from α-particle collisions with 18O in mantle silicates (arrows). Ballentine and colleagues' data1 for well gases from New Mexico define a wedge-shaped field, because air- and crustal-derived neon are pre-mixed before the addition of gases from the upper mantle. This provides an estimated upper limit for 20Ne/22Ne in the upper mantle, which implicates Ne-B as the primordial composition for most of the mantle.

Ballentine and colleagues' results1 establish an upper limit of 12.2 to 12.5 for 20Ne/22Ne in Earth's upper mantle. In contrast, 20Ne/22Ne ratios for the deep mantle, estimated from analyses of basalts at Hawaii and Iceland6,7,8, and rocks from the mantle-plume province of Russia's Kola Peninsula10, extend to 13.0 or higher. These higher 20Ne/22Ne values approach the value for the solar wind (13.8), a present-day proxy for the early solar nebula. The shallow- and deep-mantle sources are systematically different in 21Ne/22Ne as well (upper mantle, 0.056; deep mantle, <0.04). The primordial neon-isotope composition for the upper mantle strongly resembles the neon component (Ne-B) observed in meteorites that underwent significant ion implantation by solar energetic particles (SEPs). Therefore, the primordial Ne-isotope composition of the deep mantle (OIB source) resembles that produced by direct trapping from a gas-rich solar nebula, whereas the primordial Ne-isotope composition of the upper mantle (MORB source) resembles that produced by a mixture of solar wind and SEPs (Fig. 1).

These measurements suggest that deep-mantle sources, such as those beneath Hawaii and Iceland, do not contribute much to the inventory of noble gases in the convecting upper mantle. Evidently, steady-state models for upper-mantle noble gases11 that invoke a flux from these deep-mantle sources need to be re-evaluated.

More remarkably, however, the results indicate that accretion of the outer portions of the Earth was dominated by aggregated solids (planetesimals) that had been heavily irradiated by solar ions. This is remarkable because such intense irradiation is likely to have occurred during an active phase of the early Sun, and only after the rotating disk of nebular gas had been swept clear. The effects of this process have recently been imaged around the main-sequence star β Pictoris, where sub-micrometre dust has been swept out of this extrasolar planetary system by radiation pressure12. With respect to Earth, only the deep-mantle regions feeding ocean islands such as Hawaii and Iceland seem to retain a considerable remnant of gases from the early solar nebula, captured from a dense atmosphere during the earliest parts of planetary formation.

One outstanding problem in this research is achieving a self-consistent model that incorporates the noble-gas constraints together with trace-element and isotope ratios of lithophile elements (those elements that tend to be concentrated in silicates, such as the alkaline earths and rare earths). The new neon-isotope results suggest that there is little or no exchange between the deep-mantle regions feeding ocean islands and the upper mantle. Yet there is currently no evidence in the lithophile tracers for any vestiges of primitive, undifferentiated mantle13. Evidence emerging from tungsten isotopes in oceanic basalts also seems to exclude significant interaction between the core and deep mantle14, making it unlikely that the core is the ultimate source of the solar neon-isotope signature observed in mantle plumes.

Consequently, the ultimate source seems to be remnants of the very earliest silicates involved in terrestrial accretion, and these remnants have remained effectively isolated from overlying mantle convection throughout Earth's history. If this source is associated with the seismically anomalous (D″) layer at the base of the mantle, the neon-isotope results indicate that this layer may have formed during Earth's accretion15.


  1. 1

    Ballentine, C. J., Marty, B., Sherwood Lollar, B. & Cassidy, M. Nature 433, 33–38 (2005).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Pepin, R. O. & Porcelli, D. Rev. Mineral Geochem. 47, 191–246 (2002).

    CAS  Article  Google Scholar 

  3. 3

    Kurz, M. D., Jenkins, W. J. & Hart, S. R. Nature 297, 43–47 (1982).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Sarda, P., Staudacher, T. & Allègre, C. J. Earth Planet. Sci. Lett. 91, 73–88 (1988).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Marty, B. Earth Planet. Sci. Lett. 94, 45–56 (1989).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Honda, M., McDougall, I., Patterson, D. B., Doulgeris, A. & Clague, D. A. Nature 349, 149–151 (1991).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Trieloff, M., Kunz, J., Clague, D. A., Harrison, D. & Allègre, C. J. Science 288, 1036–1038 (2000).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Moreira, M., Breddam, K., Curtice, J. & Kurz, M. D. Earth Planet. Sci. Lett. 185, 15–23 (2001).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Moreira, M., Kunz, J. & Allègre, C. J. Science 279, 1178–1181 (1998).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Yokochi, R. & Marty, B. Earth Planet. Sci. Lett. 225, 77–88 (2004).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Porcelli, D. & Wasserburg, G. J. Geochim. Cosmochim. Acta 59, 4921–4937 (1995).

    ADS  CAS  Article  Google Scholar 

  12. 12

    Okamoto, Y. K. et al. Nature 431, 660–663 (2004).

    ADS  CAS  Article  Google Scholar 

  13. 13

    Hofmann, A. W. Nature 385, 219–228 (1997).

    ADS  CAS  Article  Google Scholar 

  14. 14

    Scherstén, A., Elliott, T., Hawkesworth, C. & Norman, M. Nature 427, 234–236 (2004).

    ADS  Article  Google Scholar 

  15. 15

    Tolstikhin, I. & Hofmann, A. W. Phys. Earth Planet. Inter. (in the press).

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Graham, D. Neon illuminates the mantle. Nature 433, 25–26 (2005).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.