Note: If this question does not belong on this Stack Exchange, please direct me to a more appropriate Stack Exchange.

This is a "creative chemistry" question as much or more-so than a "give me the facts" type of question. Should be of interest to anyone who finds the topic of plausible alternative biochemistries intriguing.

I recently came across several papers outlining a hypothetical type of photosynthesis that would be highly efficient in a Hydrogen ($\ce{H2}$) dominant atmosphere containing Methane ($\ce{CH4}$). The photosynthetic reaction is:

$$\ce{CH4 + H2O + $y$ -> CH2O + 2H2}$$

"$y$" represents energy imparted by photons. "$\ce{CH2O}$" is the simple organic compound formaldehyde, which could be used to build biomass.

This made me wonder how "animals" in such an environment would obtain the energy needed to live. In an "$\ce{O2}$" heavy atmosphere the answer is simple: use "$\ce{O2}$" to oxidize the available organic compounds, getting a ton of energy. This is aerobic respiration. What if there is essentially zero available "$\ce{O2}$"? Which brings me to...

My question in detail: What is a plausible chemical reaction (or series of reactions) that could take the place of $\ce{O2}$ respiration in a "Hydrogen Dominant" world with the following molecules and compounds to work with:


  • $\ce{H2}$ (10-90%)
  • $\ce{N2}$ (10-90%)
  • $\ce{H2O}$ Vapor (≈1%)
  • $\ce{CH4}$ (0.01 - 5%)
  • $\ce{NH3}$ Vapor (0.01 - 5%)
  • Other trace to minimally-present compounds may include $\ce{CO2}$, $\ce{Ar}$, etc.


  • Built off of $\ce{CH2O}$ (the organic product of hydrogenetic photosynthesis)


  • Bodies of water containing a mix of $\ce{H2O}$ + $\ce{NH3}$, and plausibly available compounds in the crust (like iron).

By "take the place of" I mean provide enough energy to run complex life-processes. I do not mean superficially resemble aerobic respiration (i.e. The process need not require the "breathing in" of atmospheric gases. It could, but need not).

The answer should try to meet the spirit of the following criteria:

  • The reaction(s) could make use of compounds in the air (breathing) or could be obtained through other methods (eating, drinking) in sufficient quantity to run biological processes.
  • Impart significant amounts of "free" energy, more so than other plausible reactions.
  • Make use of available compounds (see details above).
  • Plausibly occur at temperatures and pressures not inimical to carbon life. $\pu{500\\K}$ temperatures or pressures of $\pu{100\\bar}$ are no-goes.
  • Be "plausibly likely" to evolve. This requires a measure of judgement. If the chemical reaction or series of reactions is wildly complicated, far-less efficient than other alternative reactions, or needs to make use of rare materials, then it's probably not plausible.

I'd love the form of the answer to contain the equation, a brief discussion on why it's plausible, and a discussion of the amount of energy released, and/or other "good to know" things.



Photosynthesis in Hydrogen-Dominated Atmospheres – https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4284464/

BIOSIGNATURE GASES IN H2-DOMINATED ATMOSPHERES ON ROCKY EXOPLANETS – https://iopscience.iop.org/article/10.1088/0004-637X/777/2/95/meta

A BIOMASS-BASED MODEL TO ESTIMATE THE PLAUSIBILITY OF EXOPLANET BIOSIGNATURE GASES – https://iopscience.iop.org/article/10.1088/0004-637X/775/2/104#apj480437s4

  • 1
    $\begingroup$ Note: The reason that there is far more methane than carbon dioxide is that life-forms could derive easy energy from methanogenisis (hydrogen + carbon dioxide = energy + methane + water). The reason there is ammonia is that life could derive easy energy from a "cold haber" process (hydrogen + nitrogen = energy + ammonia). These processes are therefore likely to evolve, causing the build-up of methane and ammonia in the atmosphere. $\endgroup$
    – n_bandit
    Commented Jul 20, 2019 at 19:24
  • 1
    $\begingroup$ In terms of formatting your question, be invited to have a look at mhchem's functionality -- available in questions, answers and comments of ChemSE -- to render chemistry-related content easier to read: chemistry.meta.stackexchange.com/questions/86/… $\endgroup$
    – Buttonwood
    Commented Jul 21, 2019 at 16:29
  • $\begingroup$ Can you put a quantitative value on "Impart significant amounts of "free" energy, more so than other plausible reactions."? Significant can mean a lot of things. $\endgroup$
    – Andrew
    Commented Jul 22, 2019 at 11:59
  • $\begingroup$ @Andrew "Significant amount of free energy" means more than other plausible reactions. That's it. More energy for the work than other competing reactions. There is no absolute measure, it $\endgroup$
    – n_bandit
    Commented Jul 22, 2019 at 13:22
  • $\begingroup$ [got cut off] @Andrew ...it all depends on what other reactions are out there that might provide more energy per the amount of effort (energy invested, resources gathered, etc.). In this case the other relevant reactions are Methanogenesis (yielding ≈193kJ/ mole?) and Fermentation (yielding 74kJ/mole?). So more energetic than those. It'd be nice if it got up to aerobic respiration levels of energy (2,820kJ/mol) but not necessary. $\endgroup$
    – n_bandit
    Commented Jul 22, 2019 at 13:29

2 Answers 2


The obvious approach to is to just do the direct analog of what Earthling metabolism does: exactly reverse the photosynthetic reaction (on net, anyway; the intermediate processes may be rather different, as they are on Earth).

In this case, that means reacting hydrogen with carbohydrates to get back methane and water. Based on this paper on photosynthesis in hydrogen-dominated atmospheres, building carbohydrates from methane and water requires considerably less energy than building them from carbon dioxide and water, so we'll equally get less energy out by decomposing them again--I'd estimate somewhere around a quarter to a third as much power density as Earthling animals relying on aerobic respiration, if we assume organisms in this environment will be building biomolecules optimized for this environment, rather than just copying ours. Given how much time I spend every day not eating, and not hyperventilating, I'm pretty sure that would still be enough to power complex animals; they just might need to eat more / breathe more / be more lazy than equivalent Earth animals. Even if you only get one fifth the power density (based on the Bains paper's estimate that hydrogenic photosynthesis can proceed with 20% of the energy required by oxygenic photosynthesis), that's still 7-8 alien-equivalent-of-ATPs per glucose, as opposed to the 2 ATPs per glucose provided by anaerobic respiration (cf. 38 ATPs-per-glucose for aerobic respiration).

As Andrew noted, however, land animals will have an extra energy supply available in the form of reducing N2 into ammonia--i.e., they can supplement their energy supply simply by breathing, and only really need to eat for the materials. Incidentally, autotrophs could also make use of excess ammonia production to power chemosynthesis in the absence of light.

Despite the relatively low abundance of ammonia in the atmosphere, however, it has much higher solubility in water than hydrogen does, and thus may be more easily available to aquatic organisms than free hydrogen is. That gives you a slightly modified "respiration" reaction that takes 3 formaldehyde units and 4 ammonia molecules to produce methane, water, and free nitrogen. Decomposing ammonia into hydrogen and nitrogen is endothermic, however, so this would be less energetic than hydrogen respiration. Not sure how well that would be balanced by the reduced energy needs for pumping water over the gills compared to processing enough water to get the equivalent quantity of free hydrogen. This is a Good Thing for the long-term stability of the atmosphere, as otherwise you'd expect life to just keep producing ammonia until either nitrogen or hydrogen is exhausted.

Now, that is not the most energetic set of reactions imaginable. Obviously, if they had access to oxygen, that would be way better. But, it's the only option that is consistently renewable. Other potential oxidizers, which are more oxidized than the plant and animal biomass that serves as food in this scenario, won't be produced by plant life (by definition--anything that is produced by plantlife is biomass available as food, and thus not more oxidized than itself!) and thus can only be regenerated in small quantities by geochemical processes, and will have very short environmental half-lives. E.g., if creature could eat lithium metal, they could react that with nitrogen, ammonia, water, or complex organic biomolecules and get energy out of any of those reactions--but you just aren't going to find large quantities of free lithium to power your biosphere!

For some back-of-the-envelope estimated specifics, let's initially suppose that plants are hydrogenically producing glucose and animals are hydrogenating glucose. That leads to the following metabolic reaction:

${C_6H_{12}O_6}^{174 kJ} + 12{H_2}^{0 kJ} \rightarrow 6{CH_4}^{-74.85 kJ} + 6H_2O^{-285.83 kJ}$

Which provides approximately 2338 kJ/mol of glucose, based on the heats of formation for each reaction component as listed in the NIST Chemistry WebBook; the net energetics for aerobic respiration are found just by replacing methane with ${CO_2}^{-393.51kJ}$ in that equation (diatomic oxygen, like diatomic hydrogen, conventionally has a heat of formation of 0, so that doesn't make any difference to the calculation), which gives us a theoretical output of 4250kJ/mol. (Looking up the actual empirical energy yield of aerobic respiration of glucose, we get 2880 kJ/mol--so, assume that all other energy calculations here are only good to within about a power of two!) So, if hydrogen-breathing animals can just manage to evolve a slightly more efficient energy system than we have (which seems plausible, especially if they are living at lower temperatures which permit better thermodynamic efficiency), they may even get up to a full half the power density of oxygen-breathing Earthlings. Given the seriously speculative and highly imprecise nature of these calculations, though, I'm more comfortable sticking with my earlier 1/3 estimate as a plausible ceiling.

Plants in a hydrogen-dominated atmosphere aren't necessarily going to be producing as much glucose as our plants do in our highly oxidized environment, so let's look at some other options for biomolecules that might be available for animals to eat. Suppose, for example, that plants produce acetylene as an energy storage molecule; hydrogenating that provides

$C_2H_2^{226.73 kJ} + 3H_2^{0 kJ} \rightarrow 2CH_4^{-74.85 kJ}$

376 kJ/mol of acetylene -- or about 14.5 kJ per gram. About half as much per gram as glucose.

And if we look at the above-mentioned ammonosynthesis reaction, using hydrogen and nitrogen directly from the atmosphere, we get

$3H_2 + N_2 \rightarrow 2{NH_3}^{-45.9 kJ}$

about 91 kJ/mol of nitrogen, or 5.3 kJ/gram of inhaled hydrogen/nitrogen mix. That's not a ton compared to glucose (or acetylene), but there's just so much more of it available, for cheap! After all, it takes quite a lot energy just to consume and digest food--and practically none to just breathe it in.

Also consider that, largely because of the above reaction, nitrogen is far more easily bioavailable in this world than on Earth, so maybe they use a lot more cyanide groups or even azides. Hydrogenating hydrogen cyanide gets you

${HCN}^{135.14 kJ} + 3{H_2}^{0 kJ} \rightarrow {CH_4}^{-74.85 kJ} + {NH_3}^{-45.9 kJ}$

255.75 kJ/mol, or about 9.5 kJ / gram.

Simply decomposing hydrogen azide, even before you consider the energy gains from hydrogenation, gets you

$2{HN_3}^{264 kJ} \rightarrow {H_2}^{0 kJ} + 3{N_2}^{0 kJ}$

about 6 kJ/gram.

Those last few reactions (somewhat surprisingly, given the notorious explosivity of azides) would tend to drive the average energy density of food down a smidge from the optimistic third-to-half estimates, so once again about a quarter of the total energy density of aerobically-metabolized Earthling food seems reasonable.

  • $\begingroup$ Thanks Logan! Hmm...would you be game to putting some numbers to the energy outputs? It's not clear to me how you arrive at a "quarter to a third" for reacting hydrogen and carbs, and that's something I'd really like to understand. Same with the ammonia reduction. $\endgroup$
    – n_bandit
    Commented Jul 23, 2019 at 1:16
  • $\begingroup$ @n_bandit I arrive at a "quarter to a third" just be reversing the synthesis reactions. It takes a certain amount of energy to build biomass hydrogenically, so that's about how much you'll get out by taking it apart again. I can look up some back-of-the-envelope energetics numbers In A Little Bit, though, and edit those in to the answer. $\endgroup$ Commented Jul 23, 2019 at 14:51
  • $\begingroup$ @n_bandit Note the energetics calculations that have now been edited in. $\endgroup$ Commented Jul 23, 2019 at 17:05

This is only a partial answer, but maybe will stimulate some thinking by others.

For a good sense of plausible reactions, one need only look at the existing biology of anaerobic organisms. Many organisms produce energy in the absence of oxygen, either by fermentation or anaerobic respiration (the distinction being the absence or presence of an energy-harvesting electron transport chain). The methanogens you mentioned are one example, and that would be one viable source of energy in a hydrogen-rich atmosphere, as long as the supply of CO2 was sufficient.

In general, fermentation and anaerobic respiration are problematic for larger organisms for a few related reasons. Because the ATP yield per reaction is low, a large organism with high energy demand needs to achieve a very high throughput of the reaction in question. That means not only having a large metabolic capacity in terms of the required enzymes and cell structures, but also the ability to take in large amounts of substrate and remove large amounts of end-product at a high rate.

The oxidized form of the electron acceptor thus must be something that is stable enough to be fairly abundant as well as easily taken in, and the waste products (reduced forms) need to be removed (or at least sequestered) quickly in order to maintain energetic favorability of the process.

For an animal living on land, gas exchange through lung-type organs seems to be the only really plausible mechanism for disposing of waste, since sequestration internally would have an energy cost. Based on existing biological products, that suggests H2S and NH3 as leading candidates for end-products. For H2S, there is no plausible gaseous precursor, so you would have to imagine the organism consuming large amounts of polysulfur or sulfur oxides. Possible, but challenging. That leaves NH3. With high N2/NH3 ratio in the atmosphere, conversion of N2 to NH3 as a respiratory pathway is hypothetically viable and seems the only plausible option. With a high H2 atmosphere, it might be conceivable to live mostly on hydrogen and not need to eat carbon compounds for energy, but just for biomass.

For an animal living in water, there is a wider range of possibilities, since the substrates can be dissolved ions such as sulfates or nitrate or metals, and the waste disposal could be via liquid-liquid exchange. The main challenge would be to find substrates that are sufficiently abundant in oxidized form despite the hydrogen-rich atmosphere and to identify a viable parallel process for reoxidation of the endproducts.


Not the answer you're looking for? Browse other questions tagged or ask your own question.