Although we’ve been focusing lately on photosynthesis, radiolysis — the dissociation of molecules by ionizing radiation — can produce food and energy for life below the surface and in deep oceans. Our interest in surface conditions thus needs to be complemented by the investigation of what may lie within, as Alex Tolley explains in today’s essay. Indeed, biospheres in a planet’s crust could withstand even the destruction of all surface life. The possible range of microorganisms well beyond the conventional habitable zone defined by liquid water is wide, and while detecting it will be challenging, we may be able to investigate the possibilities in our own system with landers, looking to a day when interstellar probes are possible to explore exoplanet interiors.

by Alex Tolley

“There may be only one garden of Eden here for large life forms such as ourselves. But living beings small enough to populate tiny pore spaces may well exist within several – and perhaps many-other planetary bodies.”

– Thomas Gold, The Deep Hot Biosphere, 1999 [1]

Thomas Gold was probably wrong about subsurface microbes being the source of fossil fuels using fossil methane (CH4), but he was the first to suggest that the newly discovered microbes in the Earth’s crust might be common in other planetary bodies. This essay will explore whether the molecules and energy available from the radiolysis of water (H2O) might support similar biospheres in other worlds in space.

Follow the water, but don’t forget the energy

NASA’s mantra of “follow the water” is important when searching for life, because liquid water is required for carbon-based terrestrial life. Life can still exist if the water freezes, but it will be in a non-metabolizing state and dormant. [Even in frozen water, such as the snow on mountains, a speck of dark material can melt a tiny volume of water around it, allowing microbes to live in these microscopic habitats.]

But while liquid water is necessary, it is insufficient to support life. Inoculate microbes in a dark, sealed flask of distilled water and they will die or go dormant, unable to acquire the energy needed for metabolism. [This is why you can keep containers of distilled water for a long time, even if bacteria contaminate the contents before sealing.]

The rich surface biosphere on Earth is powered by the sun. Photosynthesis fixes the sun’s energy from carbon dioxide (CO2) and water. Before photosynthesis evolved energy was anaerobically harvested from molecules that could liberate energy when respired. Bacteria living in the ocean’s dark, hot smoker vents metabolise the molecules erupting from the mantle and in turn provide the food and energy for the complex life living near these vents.

By the mid-1990s it was accepted that microorganisms discovered kilometers down in the crust were active in the interstices between the mineral grains. Water percolating in these rocks was responsible for keeping these microorganisms actively metabolizing rather than being in a dormant state. But what were they using for food and energy where it was lightless? Carbon was available as CO2 and CH4. The archaea kingdom of anaerobic organisms include all the methanogens that can convert CO2 to CH4 extracting energy and the using the carbon for metabolism. These were the dominant forms of life on and in the early Earth, and possibly the source of the traces of seasonal CH4 detected on Mars.

However, there is another more energetic molecule, molecular hydrogen (H2) that can be used for metabolism. As an electron donor, it can be coupled with an electron acceptor to be part of an energy harvesting metabolism. If H2 is a metabolic energy source, what is its source in the crustal biosphere?

The standard explanation is that some forms of the serpentinization reactions can produce H2 as well as the better known production of CH4.

However there is another source of H2, created by radiolysis of H2O by decaying radioactive elements such as unstable isotopes of potassium (K), thorium (Th), and uranium (U).

Figure 1. Radiolysis of water in the interstices of rock when water is present.

Besides creating H2, radiolysis also produces oxidants which in turn react with the rocks, most notably with sulfides, producing sulfates.

Li et al

“We have demonstrated that the S-MIF-bearing dissolved sulfate in the saline fracture waters at Kidd Creek originates from sulfides in the Archaean host rocks. The most likely mechanism for sulfate production in these anoxic fracture water systems is the indirect oxidation of sulfide minerals by oxidants from radiolytic decomposition of water” [3]

Radiolysis vs Serpentinization

Experiments on the radiolysis of water suggested that radiolysis was not an important source of H2 compared to serpentinization. Serpentinization occurs wherever high iron (Fe) igneous rocks from the mantle, water and heat interact. The ocean ridges between the plates are important zones where this takes place.

However, later experiments with oceanic sediments showed that radiolysis production of H2 was catalyzed by the minerals increasing production of H2 many fold. In the sediments on the ocean floor it was found that radiolysis was the main source of H2 as an energy source for microbes.

Sauvage et al: [9]

“Radiolytic H2 has been identified as the primary electron donor (food) for microorganisms in continental aquifers kilometers below Earth’s surface. […] all common marine sediment types catalyse radiolytic H2 production, amplifying yields by up to 27X relative to pure water. […] Comparison of radiolytic H2 consumption rates to organic oxidation rates suggests that water radiolysis is the principal source of biologically accessible energy for microbial communities in marine sediment older than a few million years.”

Moreover, radiolytic H2 is as dominant a source of food and energy as marine photosynthesis powered by the sun.

Sauvage et al: [10]

“[…] radiolytic H2 production in marine sediment locally produces as much electron donor (food) as photosynthetic carbon fixation in the ocean.”

In summary globally, radiolysis can provide both food and energy comparable to that of the marine photosynthetic organisms.

Methanogens & Sulfur-reducing bacteria

In anoxic environments where both archaeal methanogens and sulfur-reducing bacteria coexist, the biomass and types of the latter are greater than the former. One reason may be that the available energy from the reduction of sulphur from sulfates is greater than the reduction of carbon to CH4 from CO2.

CH4 can be created by the reduction of carbon dioxide.

Serpentinization is a geologic source. However, archaean methanogens are believed the dominant source of methane in the atmosphere on the early Earth reducing carbon dioxide anaerobically to methane [12]. It is the uncertainty of the source of the methane detected on Mars that intrigues astrobiologists.

Sulfate-reducing bacteria such as Desulfovibrio and Desulfobacter use the H2 to reduce sulfates created by the radiolysis oxidants on mineral sulfides to again reduce the sulfur to silfides.

As table 1 indicates, this is a more energetic reaction than methanogenesis and may account for the very many different bacteria utilizing H2 and sulphate as an energy source.The radiolytic oxidants also react with CH4 to form simple organic molecules such as formate and acetate which can be used as food sources by bacteria, further indicating the value of radiolysis in maintaining a subsurface habitat.


As I have noted in previous posts, the search for life has typically been focused on surface-living, complex, aerobic life, as the low hanging fruit of detectability. This restricts the search to planets in the habitable zone (HZ). However, as unicellular life dominated Earth’s history and anaerobic respiration was dominant until the evolution of photosynthesis, and the Great Oxidation Event increased the partial pressure of oxygen in the atmosphere, such worlds may give rise to false negatives when analyzing the atmosphere by spectroscopy. Furthermore, as professor Tyrrell has suggested, surface life on Earth-like worlds may have a low probability of being sustained over 3 billion years due to events perturbing the surface temperature into runaway conditions.

Unlike the variable conditions on the surface, subject to wide ranges of conditions and vulnerable to cosmic and geologic disruption that saw 5 major extinctions on Earth, as well as an ongoing 6th extinction in the Anthropocene, conditions in the crust are far more stable, and less vulnerable to the disruptions on the surface. Such crustal biospheres once established may survive even after surface life has been extinguished, especially once the star’s luminosity renders the surface uninhabitable.

Such a biosphere may even allow for an evolutionary reset starting with microorganisms should surface conditions become uninhabitable for a temporary period and subsequently returning to habitability.

So we have evidence that life is in the subsurface crustal rocks and that radiolysis may be an important, if not the most important, source of food and energy for this life. But what about bodies elsewhere?

Other Celestial Bodies

1. Mars

Mars has all the same ingredients as Earth for subsurface microorganisms to live. At some depth below the surface the temperature should be sufficient to create liquid water [13]. While serpenitization can occur, especially to generate CH4, CO2 should be available for methanogens to respire and release CH4. Residual radioactive elements should be able to produce the needed H2 and SO4.for sulfur reducing bacteria. The race is on to determine whether there is an extant microbial biosphere on Mars. Looking for frozen microbes in ejecta from large meteor impacts that have penetrated to the needed depths might be the easiest approach for robotic vehicle discovery. If there are any near surface hot spots from residual volcanism or local concentrations of radioactive elements, these might also be good places to look. Whether earth, Venus, or Mars is the original world where abiogenesis occurred, panspermia between these worlds due to ejecta and microbes propelled by solar radiation, early Mars may have been a home for life.

Tarnas et al: [14]

“We have demonstrated that radiolysis alone produced sufficient quantities of reductants to have sustained a subsurface biosphere during the Noachian for hundreds of millions of years. Given sufficient oxidant availability, this habitat could have fostered chemolithotrophic microbial communities that would have imprinted organic, morphological, and isotopic biosignatures on their habitat’s host rock.”

2. Icy moons

Jupiter’s Europa and Saturn’s Enceladus are 2 icy moons that have subsurface oceans and tidally induced warming. Radiolysis of subsurface water in the core below the ocean sediments should provide the conditions necessary for a microbial biosphere. Analysis of the plumes by a flyby or orbiting probe is one method to search for life, although this is more likely to detect life in the oceans, rather than below the ocean-crust interface. This later will prove a much more difficult target.

3. Titan

Saturn’s moon Titan is believed to have a rocky core, overlaid with a liquid ocean, and topped with hydrocarbons with a dense nitrogen atmosphere. As with the icy moons, below the crust-ocean interface is a possible biosphere.

4. Ceres

Like the icy moons, Ceres has a rocky core overlaid with brines. Evidence of cryovolcanism suggests that these brines must be partially liquid. Unlike the icy moons, there is no tidal heating. This suggests that any biosphere in the core must be powered by radiolysis from any residual radioactive element decay. Catillo-Rogers recently suggested Ceres has the potential to host life as it has the radioactive elements for both heating and radiolysis. [5]

5. Eris

The Trans-Neptunian dwarf planet Eris has a density of 2.52 g/cm^3, indicating that it must be composed of rocky material and ices. If, like Pluto, there is evidence of cryovolcanism then it is possible that a core with radioactive elements and liquid water provides a habitat for a microbial biosphere. If so, then other dwarf planets extending out into the Kuiper belt could also have similar subsurface habitats.

6. Comets and Kuiper belt Objects

Holm focused on serpentinization for the production of CH4 and H2 on celestial bodies [7] He notes that radioactivity could also warm these bodies, making serpentinization possible. However, he did not consider radiolysis that might have been an important contribution to the production of energy rich molecules that could be used for metabolism.

Comets and their parent bodies, such as Transneptunian Objects (Kuiper Belt Objects—KBOs), accreted from a mixture of volatile ices, carbonaceous matter, and rocks in the coldest regions of the protosolar nebula. […] However, the rocky material contained in comets includes radioactive isotopes, whose decay can provide an important source of heat, possibly significantly altering the internal structure of these icy objects after their formation. There is a general agreement that short-lived radioactive isotopes like 26Al and 60Fe could have played a major role during the early evolution of both comets and their parent bodies, possibly leading to the melting of water ice and to the triggering of serpentinization and FTT reactions.

A more recent paper by Bouquet emphasized the importance of radiolysis in icy bodies which not only produced H2, but sulfates to support metabolism [2].

We found that radiolysis can produce H2 quantities equivalent to a few percent of what is estimated from serpentinization. Higher porosity, which is unlikely at the scale of a body’s entire core but possible just under the seafloor, can increase radiolytic production by almost an order of magnitude. The products of water radiolysis also include several oxidants, allowing for production of life-sustaining sulfates. Though previously unrecognized in this capacity, radiolysis in an ocean world’s outer core could be a fundamental agent in generating the chemical energy that could support life.

7. Rogue/Free Floating Planets

Rogue planets ejected from their systems would include bodies similar to those in the solar system. Given the prevalence of conditions needed for a subsurface biosphere, especially in bodies at the edge of our system, there seems every reason to believe that these rogue planets should also host subsurface conditions suitable for a microbial biosphere.

Habitable yes, but inhabited?

The above suggests that if radioactive elements can also heat the surrounding material so that water is kept liquid, then almost any celestial body with a rocky material and water could potentially be a microbial habitat in space, irrespective of whether it is in the HZ or not. As suggested earlier, planets in the HZ that have lost surface habitability could retain refugia for life in the crust.

For other non-Earth-like bodies which may have the conditions for a subsurface biosphere the question becomes whether they are living or sterile. Is abiogenesis possible on these worlds, or must they be inoculated by life from living worlds? We don’t yet know the answers to such questions, but it does suggest that astrobiologists take seriously the possibility that any body with a suitable subsurface environment could be inhabited and therefore instruments to detect such life should be included with exploratory probes. As we increase the exploration of our system, landers and rovers should include technologies to detect life, especially within the habitable zone below the surface.

Could we detect subsurface biospheres on exoplanets?

Detection of subsurface biospheres on exoplanets is going to be very difficult. Seager [6] produced a catalog of possible biosignature molecules, of which hydrogen sulphide (H2S) is primarily of biologic origin and therefore its presence is likely an unambiguous biosignature.

Although H2S is likely to have a very low concentration in the atmosphere it has a distinctive IR signal which could be detectable in principle,

As we can currently only analyze exoplanets spectroscopically, if it becomes possible to detect the very small amounts H2S in an otherwise unpromising atmosphere with possibly unsuitable surface conditions for life, then we should attempt to devise the technology to detect the presence of this gas as an unambiguous biosignature.

Based on the terrestrial history of life, it seems likely that on living exoplanets, extant life will be mostly unicellular, possibly even just prokaryotes. On Earth-like worlds geologic processes will ensure that such life will also inhabit a deep, crustal biosphere. Detecting such life will be very difficult, but perhaps not impossible with suitable technology. In the distant future, interstellar probes with landers should be able to detect such life as we explore our stellar neighborhood and catalog and map the forms of life we find.


Gold, T. (2021). The Deep Hot Biosphere: The Myth of Fossil Fuels (November 6, 1998) Hardcover. Springer.

Bouquet, A., Glein, C. R., Wyrick, D., & Waite, J. H. (2017). Alternative Energy: Production of H 2 by Radiolysis of Water in the Rocky Cores of Icy Bodies. The Astrophysical Journal, 840(1), L8.

Li, L., Wing, B. A., Bui, T. H., McDermott, J. M., Slater, G. F., Wei, S., Lacrampe-Couloume, G., & Lollar, B. S. (2016). Sulfur mass-independent fractionation in subsurface fracture waters indicates a long-standing sulfur cycle in Precambrian rocks. Nature Communications, 7(1).

Lin, L. H., Wang, P. L., Rumble, D., Lippmann-Pipke, J., Boice, E., Pratt, L. M., Lollar, B. S., Brodie, E. L., Hazen, T. C., Andersen, G. L., DeSantis, T. Z., Moser, D. P., Kershaw, D., & Onstott, T. C. (2006). Long-Term Sustainability of a High-Energy, Low-Diversity Crustal Biome. Science, 314(5798), 479-482.

Castillo-Rogez, J. C., Neveu, M., Scully, J. E., House, C. H., Quick, L. C., Bouquet, A., Miller, K., Bland, M., De Sanctis, M. C., Ermakov, A., Hendrix, A. R., Prettyman, T. H., Raymond, C. A., Russell, C. T., Sherwood, B. E., & Young, E. (2020). Ceres: Astrobiological Target and Possible Ocean World. Astrobiology, 20(2), 269-291.

Seager, S., Bains, W., & Petkowski, J. (2016). Toward a List of Molecules as Potential Biosignature Gases for the Search for Life on Exoplanets and Applications to Terrestrial Biochemistry. Astrobiology, 16(6), 465-485.

Holm, N., Oze, C., Mousis, O., Waite, J., & Guilbert-Lepoutre, A. (2015). Serpentinization and the Formation of H2 and CH4 on Celestial Bodies (Planets, Moons, Comets). Astrobiology, 15(7), 587-600.

Lollar, B. S., Onstott, T. C., Lacrampe-Couloume, G., & Ballentine, C. J. (2014). The contribution of the Precambrian continental lithosphere to global H2 production. Nature, 516(7531), 379-382.

Sauvage, J. F., Flinders, A., Spivack, A. J., Pockalny, R., Dunlea, A. G., Anderson, C. H., Smith, D. C., Murray, R. W., & D’Hondt, S. (2021). The contribution of water radiolysis to marine sedimentary life. Nature Communications, 12(1).

Sauvage, J. F., (2018) Sedimentary Catalysis of Radiolytic Hydrogen Production, Open Access Dissertations. Paper 704.

Cepelewicz, J, (2021, May 24). Radioactivity May Fuel Life Deep Underground and Inside Other Worlds, Quanta Magazine, Accessed online June 2021

Tolley, A., Detecting Early Life on Exoplanets. February 23, 2018

Tolley, A., Is Most Life in the Universe Lithophilic? – AMT prior article JANUARY 11, 2019

Tarnas, J., Mustard, J., Sherwood Lollar, B., Bramble, M., Cannon, K., Palumbo, A., & Plesa, A. C. (2018). Radiolytic H2 production on Noachian Mars: Implications for habitability and atmospheric warming. Earth and Planetary Science Letters, 502, 133-145.