Years ago I came across this quote

...nothing would be more tragic in the American exploration of space than to encounter alien life and fail to recognize it...".

Now one of my guiding research principles, this quote reminds me to resist “terracentricity”—the notion that habitable exoplanet environments must be closely related to Earth’s.

We were motivated to study life in molecular hydrogen-dominated environments simply because H₂-dominated  planet atmospheres are larger in extent and will therefore be easier to observationally study than their heavier atmosphere (i.e. N₂- or CO₂-dominated) counterparts. Atmosphere densities fall exponentially with increasing vertical altitude. The e-folding distance is the pressure scale height, which depends on surface gravity, temperature, and atmosphere mean molecular mass. The lightness of H₂, therefore, increases the atmospheric scale height, i.e., makes atmospheres  “puffy”. 

A caveat is we don’t yet know if rocky planets with hydrogen atmospheres exist, we can only say that theory supports them in some scenarios.  We also haven't yet discovered a rocky planet that fits the bill—a planet more massive and receiving less stellar energy than Earth, yet not obviously swamped by an H₂ or H₂-He gas envelope—partly because observational selection effects disfavor small, cold planets.

With two biologists and a chemistry grad student on my team, we borrowed some lab space and set up a custom "bioreactor" to study a simple (prokaryote) microbe E. coli, and a more complex (eukaryote) microbe, yeast, in different atmosphere environments. We used pure hydrogen as a proxy for a hydrogen-dominated atmospheres. We also used pure helium for comparison. Air and a 20% CO₂ and 80% N₂ mixture rounded out the controls. 

We found that the microbes we studied could survive and grow in a pure hydrogen-dominated atmosphere, although at a slower rate than in air. The main take away for our astronomy peers is that lots of hydrogen is not detrimental to the microbes we studied, or to life in general.

Biologists are not surprised by our results, as hydrogen is not known to be toxic to life. The type of life we studied could survive with any atmosphere because it is gaining energy from the glucose broth and not from the hydrogen atmosphere itself. 

We aimed to provide clear and concise experimental validation of inferred knowledge (of survival strategies of microbes)  presented to the astronomy community in an accessible way.  Rocky exoplanet atmospheres with hydrogen will be relatively easy (though still challenging) to study, hence we want astronomers to accept them in the menu of options when it comes to the search for signs of life. 

What is surprising from an astronomer's point of view is that E. coli can produce many different types of gases—meaning that E. coli has a diverse metabolic machinery, surprising for such a simple life form. This gives us scientific hope that a range of interesting gases might also be produced by simple hypothetical microbial-type life on exoplanets.

When near future, next-generation ground- and space-based telescopes are online,  we want to be ready to recognize signs of life on any of the few precious planets we will have available for observation.