About 5 billion years from now, the Sun will begin to swell into a red giant. This will cause some problems, because life on a planet orbiting a red giant is hard. Even if Earth isn't engulfed by the expanding Sun, it's going to be scorched. We're looking at surface temperatures anywhere from 500$^\circ$F to about 3200$^\circ$ F, which is highly unpleasant. 3 billion years later, the Sun will have settled down to become a white dwarf, surrounded by a protoplanetary nebula, and things will become relatively quiet and boring for what remains of the Solar System.

I'm working on a story set in the far future. Specifically, it involves a couple of intrepid time travelers intent on seeing the wonders of the galaxy - pretty standard time travel stuff. Most of their journey is irrelevant for now, but the key event is that they decide to travel 8 billion years into the future to see what the protoplanetary nebula around the Solar System looks like, out of sheer morbid curiosity. I'd like them to find, to their surprise, that the Sun is no longer a white dwarf, but a normal (-ish?) star, insofar as it's undergoing significant nuclear fusion again. Life might even be possible again, at some part of the Solar System.

Needless to say, our explorers are more than a little shocked. The thing is, this whole premise rests on the idea that after the Sun becomes a white dwarf, it could somehow revert to a normal star, and when I started writing, I wasn't sure if this would be possible. Can the Sun somehow, 8 billion years in the future, leave the white dwarf track and start fusing hydrogen, helium, or heavy elements again? Here are my criteria, which are pretty strict:

  • Fusion must be stable on a timescale of at least a few thousand years. That's admittedly small on stellar timescales, but I'm not pushing my luck. A few million years would be nice, but my expectations are low.
  • The Sun must have a luminosity at least half as large as its current luminosity - preferably a couple times greater.
  • I'd rather there not have been any catastrophic astronomical events, like a collision (or interaction) with another star, because then there's the possibility that the Solar System could be thrown into disarray.
  • The event should be natural, not prompted by an advanced civilization or anything far-out like that. For instance, I would disallow star lifting, or the creation of some sort of megastructure.
  • $\begingroup$ I admit the you idea is definitely an interesting one, but you if you wanted you could have the Sun have been turned into a Neutron Star but Jupiter and Saturn collided and absorbed some more material from the sun, and now are a small sun themselves orbiting the now dead sun... might be cool to look at. $\endgroup$ Commented Jun 7, 2018 at 13:46
  • 1
    $\begingroup$ @BladeWraith That's a possibility (although the Sun's nowhere near massive enough to become a neutron star, and even the combined mass of Saturn and Jupiter isn't high enough to begin nuclear fusion). $\endgroup$
    – HDE 226868
    Commented Jun 7, 2018 at 13:54
  • $\begingroup$ I completely agree, but there's been 8 billion years for something to have happened that could have increased the chances of it happening maybe even absorbing some of the matter that the sun would have thrown off... just let me dream... sob sob sob $\endgroup$ Commented Jun 7, 2018 at 14:01
  • $\begingroup$ What if, instead of the sun reversing to a "normal" state, the mass of its expansion as a red giant drove the earth away, in a new Goldilock zone, thus the temperatures are close to normal? $\endgroup$
    – kikirex
    Commented Jun 7, 2018 at 14:43
  • $\begingroup$ @kikirex Earth's orbit will expand, yes, although it might not be enough; surface temperatures will still be high. Plus, I'm really looking for events that might happen after the Sun enters the white dwarf stage, or thereabouts. $\endgroup$
    – HDE 226868
    Commented Jun 7, 2018 at 14:44

3 Answers 3


I actually figured this out shortly after beginning the story, and I'm going to therefore post a self-answer, because it might be useful to others. Obviously, I'm still open to more ideas if this one is flawed or incomplete.

TL;DR: Yes, it could work, on short enough timescales.

So, the Sun will transition from a red giant to a white dwarf as it traverses the asymptotic giant branch of stellar evolution, becoming an AGB star. It will remain in this phase for a few million years at the most, losing dramatic amounts of mass via stellar pulsations. After this extreme mass loss - on the order of half a solar mass - it will transition into a post-AGB star, and head towards the white dwarf track.

Now, the pulsations in the star continue throughout the AGB phase, and include shell helium flashes, where helium fusion suddenly begins in a shell of hydrogen in the star. This happens several times for an AGB star, and may continue as the Sun transitions through the post-AGB phase and forms a planetary nebula.

Under the right conditions, something called a very late thermal pulse may occur as the star enters the white dwarf branch. This rapidly depletes the hydrogen and can initiate the fusion of heavy elements inside the star. The now former white dwarf travels back to the AGB phase extremely quickly - within decades or a century - and begins life as an AGB star again, staying on the branch for hundreds of years, depending on its mass loss. At the end of this era, the star will be forced to become a white dwarf for good, as it will have lost almost all of its hydrogen.

So, does this happen? Very likely! There are several notable candidate cases:

  • Sakurai's Object, which began to display this behavior in 1996. The very late thermal pulse model was applied by Duerbeck & Benetti (1996) later that year.
  • V605 Aquilae, which brightened in 1918. It may also have undergone a nova instead (or merged with another star), but the born-again hypothesis remains plausible.
  • FG Sagittae, which may have begun to transition back to an AGB star about a century ago. It has continued cooling, which makes sense, as white dwarfs are usually hotter than AGB stars or supergiants.

FG Sagittae may have stopped cooling; the first two stars will likely stop cooling in a few decades (Lawlor & MacDonald (2003)). After that, they will enter the AGB phase again. Now, Sakurai's Object formed a planetary nebula $\sim$17,000 years ago, and it's still visible. Protoplanetary nebulae don't last long, on astronomical timescales, but this one isn't going anywhere any time soon.

Now, let's say the Sun also undergoes a very late thermal pulse. What sort of conditions are we looking at? The three objects above give us a pretty good idea:

  • It will have a mass about half that of its current mass, as per models of AGB evolution.
  • Its luminosity will be $\sim$10,000 times its current luminosity. This means that the effective temperature of a planet $\sim$100 AU from the star will have the same effective temperature as Earth does right now.
  • The cooling phase should last about a century, and the second AGB phase will last at least several hundred years (models predict different timescales; see Herwig (2001)). $\sim$1,000 years is likely the upper limit.

Earth likely won't be habitable, although it's possible that Planet Nine (or any moons it might have), at its closest approach to the Sun, would be (its orbital elements aren't terribly well-constrained). The Kuiper Belt would also be warmer, which would be nice. Perhaps even the outer ice giants could provide refuge for any life-forms daring enough. Of course, 1,000 years (or several hundred years!) isn't much time at all for life to arise, but perhaps life could arrive via panspermia. It's still possible.

Some less well-developed ideas

I also had some other, less promising ideas that could still work, so I'll outline them here. I like them a bit less because of a lack of observational support for their evolutionary paths.

So, if fusion begins again, the Sun is going to be a hydrogen-deficient star, because it will have lost all of its hydrogen during the AGB and post-AGB phases. Now, we know of some classes of stars that are hydrogen-deficient, besides AGB stars:

The Sun will never become a Wolf-Rayet star; it's not massive enough. But these particular subdwarfs show promise. Subdwarf B stars may form when red giants prematurely lose their outer layers of hydrogen (possibly on the horizontal branch), and may evolve in subdwarf O stars (which could also form from the merger of white dwarfs). The only problem, of course, is that this early hydrogen loss is not yet well-understood; a binary system may be necessary.

  • $\begingroup$ That is cool. Well. Not cool. Hot. But... you get what I mean. $\endgroup$
    – Joe Bloggs
    Commented Jun 7, 2018 at 13:43
  • $\begingroup$ "late thermal pulse" stage is estimated to be "lasting only about 200 years", and it's not clear if there will be 10,000s years of stable cooldown after that. $\endgroup$
    – Alexander
    Commented Jun 7, 2018 at 17:25
  • $\begingroup$ @Alexander Looks like I'd mixed up some exponents; I'd used a number of the characteristic timescale of thermal pulses prior to leaving the asymptotic giant branch. Models differ; depending on the mass loss rate and the very late thermal pulse, we could be looking at anywhere from perhaps 200 to 700 years. Thanks for the correction! $\endgroup$
    – HDE 226868
    Commented Jun 7, 2018 at 17:43
  • $\begingroup$ your time travellers are extremely lucky to be able to observe such a phenomenon. they would have missed it if they were off by even a few millennia. $\endgroup$
    – fraglord
    Commented Dec 21, 2018 at 11:12

Here's another thought... Some time after the sun becomes a white dwarf, another less massive young star passes very nearby, and some - or a lot - of its hydrogen is drawn off by the sun. This would shortly have the effect of reigniting hydrogen fusion in the transferred mass.

By the time the time travellers arrive, the sun is calmly fusing its stolen hydrogen, and the victim of the theft has either left the area minus a lot of mass, or has been entirely absorbed. The former is far more likely.

Of course, careful examination of the stellar thief would reveal an awful lot iron and other end-stage elements, and it wouldn't be too hard to deduce what had happened.

  • $\begingroup$ I think this may have the adverse effect of distorting the orbits of the planets. $\endgroup$
    – user44399
    Commented Jan 5, 2019 at 14:52

(please note that while this answer explicitly violates the terms laid out by the OP, I am not deleting it for future reference purposes to avoid a DenverCoder9 situation.)

Human Intervention

There is a technique called Starlifting which lifts mass from stars using magnetic fields, allowing civilizations to gather immense amounts of matter.



The matter would be used for megastructures such as Dyson Swarms and would also allow extension of the sun's lifespan by reduction in mass, so your time traveler would see a smaller G-type yellow dwarf (or said star gone red-giant), or potentially a red dwarf with a lifespan of trillions of years.

But what if you want humans gone?

using 10% of the Sun's total power output would allow 5.9 × 10^21 kilograms of matter to be lifted per year (0.0000003% of the Sun's total mass)


An expression for the main sequence lifetime can be obtained as a function of stellar mass and is usually written in relation to solar units.

$ \frac{t_{MS}}{t_\odot} \sim (\frac{M}{M_\odot})^{-2.5} $

$t_\odot$ = Sun MS lifetime

$M$ = mass of star

$M_\odot$ = solar mass

-Swineburn University

Let's say a gamma-ray burst or something wiped out the humans (which realistically probably would have gone interstellar by than) because we don't want our future solar system that is visited by our protagonists to see other humans. How much would they have to starlift to say, double the solar lifespan? With the above assumptions, we can calculate the change in mass needed to achieve. I thought the above equation was kind of confusion so I simplifed it to

$f(x) = \left(\frac{x}{1}\right)^{-2.6}$

where $x$ is mass in terms of solar masses and $f(x)$ is time in terms of the Sun's lifespan (am I rehashing the same thing over and over again?). We see that it would take a reduction of 25% in solar mass to double its lifespan, which would take $8.3 \cdot 10^7 $ years, or 83 million years. Using all of solar energy output would still take 8.3 million years, though energy probably won't be a problem since most of the output would be hydrogen, reused for nuclear fusion. By this time, humans probably can achieve Kardashev-3 status, so this option probably isn't viable.

  • $\begingroup$ Please note that in my question I asked for a natural phenomenon, and specifically ruled out star lifting or similar methods. $\endgroup$
    – HDE 226868
    Commented Jan 5, 2019 at 19:25
  • $\begingroup$ @HDE226868 oh shoot sorry about that. $\endgroup$ Commented Jan 5, 2019 at 19:40

You must log in to answer this question.

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