This honestly sounds like it's basically impossible, but here we go anyway.

We've picked up a strange reading from a spot orbiting a star in a system elsewhere in the Milky Way galaxy. We've come into orbit nearby, and decided to investigate in person. We suit up, head out with our self-propelled space backpacks and, oddly enough, our systems indicate that we can safely remove our helmets. Here? In open space? With no planet or even a large asteroid anywhere remotely nearby? How could this be possible?

Well, I suppose that means that, despite near-impossibly slim probability, this star...

  • Has some sort of gas cloud orbiting it right where we happen to be, and among the gasses are just the right gasses in the right proportions and pressures for humans to breathe and also nothing fatal. Maybe a circumstellar ring? Permissibly, there can be lots of space junk also in proximity.
  • Is of an appropriate size, temperature, radiation level, etc to have a region that is the right temperature, pressure, etc for us to survive.
  • Probably many other factors we aren't even considering.

Could such an environment really exist? We don't need to survive long-term; we can go back to the ship for water, food, etc. We would just like to be able to remove our helmets in this unusual place. What would have to happen for this to exist?

Bonus points if we can safely look at the star with the naked eye, and super bonus points if it can actually occupy a significant arc of the "sky."

  • 3
    $\begingroup$ The Integral Trees, by Larry Niven, explores something like this--a neutron star with a "gas torus" in orbit around it. I'm not sure how feasible/realistic the physics for it are. $\endgroup$
    – Qami
    Apr 5, 2021 at 21:21
  • 1
    $\begingroup$ A link to a page explaining some of the related physics: kaiserscience.wordpress.com/physics/gravity/… $\endgroup$
    – Qami
    Apr 5, 2021 at 21:26
  • $\begingroup$ "Short" is not a specific amount of time. You could definitely survive for a "short" time on lot of diferent kind of environments. Somes are just really really "short" time though. $\endgroup$
    – Kepotx
    Apr 7, 2021 at 10:05

3 Answers 3


Niven's The Integral Trees has exactly this premise. It has a somewhat implausibly benign gas ring around a star, pulled out of the atmosphere of a convenient gas giant that's too close to its parent star to fully retain its atmosphere.

Could such an environment really exist?

Almost certainly not. A major part of the problem is that when you're close enough to a star to benefit from things like liquid water, you're also close enough for UV and other ionising radiation to split the chemicals in the atmosphere into radicals and for the solar wind and radiation pressure to either blow them away or de-orbit them into the sun. A dense enough ring just isn't going to be able to form.

Bonus points if we can safely look at the star with the naked eye, and super bonus points if it can actually occupy a significant arc of the "sky."

If the star is weak enough to look at with the naked eye, it might not actually be able to do useful things like keep water liquid, or drive photosynthesis.

  • $\begingroup$ @ StarishPrime: Could it be realistic-ified by having a star with slightly higher energy output of lower spectrum...(ie. leaning more into infrared than UV) ? $\endgroup$
    – Qami
    Apr 5, 2021 at 21:43
  • $\begingroup$ @Qami - solar wind is still an issue. The energy output of the sun required to provide things like liquid water prevent it from hanging around unaccompanied in space. If you've got the one, you can't have the other, and if you don't have the one, you also can't have the other. Kind of a lose-lose. $\endgroup$
    – jdunlop
    Apr 5, 2021 at 21:54
  • $\begingroup$ "either blow them away or de-orbit them into the sun" - so there is a sweet spot between the two? "UV and other ionising radiation" / "If the star is weak enough to look at with the naked eye, it might not actually be able to do useful things" - again, perhaps a sweet spot between the two is possible? $\endgroup$ Apr 6, 2021 at 0:08
  • 1
    $\begingroup$ @PcMan radiation pressure from the star is gonna blow that ring away. The page you've linked doesn't consider it at all, which is a major oversight. The integral trees was a clever idea for an unusual setting, but it is alas not practical in real life. $\endgroup$ Apr 7, 2021 at 13:29
  • 1
    $\begingroup$ @PcMan - at 1.7Au, you can still use a solar sail. Goodbye free gases. $\endgroup$
    – jdunlop
    Apr 7, 2021 at 17:41

The closest thing to this in real life would probably be a brown dwarf. They're not technically stars (they're classified as substellar objects, something between a large planet and a small star) but early in their lives they do fuse hydrogen (deuterium mostly) and some of the largest fuse Lithium.


  • They emit light mostly in the infrared spectrum, though some do emit x-rays.
  • They cool over time, and will eventually reach temperatures that a human could survive.
  • You could look at one safely, and get quite close without being incinerated (assuming it's no longer undergoing nuclear fusion, and not emitting x-rays)
  • They are suspected to be fully convective, so if the brown dwarf contains breathable quantities of oxygen it won't be trapped in the core due to gravity


  • "Surface" gravity in excess of 20Gs
  • As with any planet, any orbiting cloud of gas dense enough to breathe would disperse, and if you descended far enough into the brown dwarf for a breathable atmosphere, the gravity would kill you.
  • $\begingroup$ To clarify, the star can all be perfectly natural. It was the device emitting the signal that appears to be manmade. Ultimately, this device doesn't matter, so I'll edit it out shortly. $\endgroup$
    – Devsman
    Apr 7, 2021 at 17:49

Let's look at the ways this would be possible:

Kordylewski Clouds at the L4 or L5 Points

enter image description here

There are natural saddles in empty space between any two massive objects in orbit around one another.

These saddle regions can trap dust.

But can they hold a 101.1 kiloPascal atmosphere at 300 degrees Kelvin?

Mean Speed of Gas -

The mean velocity of $O_2$ at $300^{o} K$ is $v = [\sqrt{{3 k_b T} \over {m}}]$ where T = 300K, $k_b$ $\approx 1.38 \times 10^{-23}$ $J \over K$ and $m = 9.64 \times 10^{-25}$ kg.

Plugging in the values, about 110 $m \over s$

Velocity at an Arbitrary L4/L5 Point -

The equation here is ${v \over R} = {V \over c}(1 + {m \over M})$. v is the velocity of the gas (110 $m \over s$), $R \approx c (sin (60))$, c is the distance between the star and the gas giant, V is the orbital velocity of the gas giant around the star. $V = \sqrt{{GM} \over c}$ for circular orbits. And $m \over M$ is the ratio of masses between the gas giant and the star.

If you manipulate the equation ${v \over R} {c \over V} $ must be $ \ge 1$ for any $m \over M$ value to be valid.

I tried several M solar masses and c radii. I found one match where a Jupiter-mass primary (M) orbiting a $1 \over 3$ Jupiter mass (m) companion at 10,000 astrononical units.

I think this would be a pretty huge gas cloud between the two that was in the process of coalescing. Maybe the system is very new, and the second planet is a recent capture.

Radiation Pressure

enter image description here

It's possible that the star emits enough radiation pressure to push a cloud of 101 kPa pressure.

The equation for radiation pressure is $P = {{4 \sigma} \over {3c}}T^4$, where $\sigma \approx 5.67 \times 10^{-8}$, c is the speed of light, P is 101.1 kPa, and T is temperature (not yet known).

Plugging all of that in, I get a temperature of 141,000 Kelvin for a star to generate 101.1 kPa of radiation pressure. Is that a reasonable value? According to this list, it is.

However, you'd need some as-yet-unknown-to-science buffer gas that's 141 thousand degrees on the hot side, and a cool 300 degrees K (26 C) on the cool side.


The hot zone would need to be deep enough down the gravity well of the star that gas is trapped. It would fall in, if not for the radiation pressure keeping it out.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .