Most likely the universe ends cold and dark in a heat death lasting past 10100 years. The timeline, the rival endings, and what new dark energy data say.
By The Orbital Record editorial teamPublished Last reviewed
The future of the universe, according to current physics, is most likely a slow heat death: space keeps expanding, the last stars fade within about 100 trillion (1014) years, and black holes evaporate by roughly 10100 years. The details depend on dark energy, which DESI data since 2024 hint may be changing.
Measured against that future, the universe has barely begun. It is 13.8 billion years old, and we live in the Stelliferous Era, the age of stars, right at its start. This guide answers with physics, not prophecy: what has been measured, what is extrapolated, and what is still open.
The most likely ending is a heat death, often called the Big Freeze. The universe keeps expanding, galaxies run out of gas, stars burn out, stellar remnants cool and drift apart, and black holes, the last concentrations of mass, slowly evaporate. Nothing explodes at the end. The universe simply runs out of ways to do anything.
How confident is that? It follows from two things measured well: the expansion of the universe has been speeding up since the 1998 discovery of cosmic acceleration, confirmed independently by the Supernova Cosmology Project, and the physics of stars, which is among the best tested in astronomy. The far end of the timeline, beyond about 1030 years, rests on particle physics nobody has yet observed, such as proton decay.
The deciding factor is dark energy, the unknown component that makes up about 68% of the universe’s energy (68.5% in the Planck 2018 results). If it is a true constant, heat death follows. If it grows, space tears apart in a Big Rip. If it fades and turns negative, the expansion reverses into a Big Crunch.
On current physics, the timeline of the far future runs from today’s 13.8 billion years to beyond 10100 years. Times below are measured from the Big Bang unless marked “from now”. You can scroll through the same timeline, one power of ten at a time, in our Deep Time explorer.
| When | What happens | Confidence |
|---|---|---|
| 13.8 billion years (now) | Universe aged 13.787 ± 0.020 billion years (Planck 2018) | Measured |
| About 2 billion years from now | Milky Way swallows the Large Magellanic Cloud, almost certainly (Sawala et al. 2025) | High |
| About 5 to 7.6 billion years from now | Sun becomes a red giant; its tip is reached about 7.6 billion years from now and Earth is engulfed (Schröder and Smith 2008) | High |
| Within 10 billion years from now | About a 50% chance the Milky Way and Andromeda merge | Coin flip |
| About 100 billion years | Galaxies beyond our Local Group slip out of view (Krauss and Scherrer 2007) | High, if dark energy stays constant |
| 1 to 10 trillion years | Galaxies exhaust their gas for new stars | Moderate |
| About 12 trillion years | The smallest red dwarfs run out of hydrogen | Moderate |
| 1014 years | Normal star formation ends; the Stelliferous Era closes | Moderate |
| About 1019 to 1020 years | Galaxies evaporate: dead stars are flung out or fall into the central black hole | Extrapolated |
| More than 1034 years | Protons may decay, dissolving white dwarfs and neutron stars (never observed) | Speculative |
| About 1067 years | A black hole of one solar mass evaporates | Extrapolated |
| About 10100 years | The largest black holes evaporate; the Dark Era begins | Extrapolated |
Near-term forecasts keep changing as data improve. In 2025, Gaia and Hubble measurements turned the “certain” collision with Andromeda into a coin flip, and a 2026 study found that the Sun’s lumpy mass loss as it dies could break up the outer Solar System within a few billion years of it becoming a white dwarf.
Fred Adams and Gregory Laughlin divided cosmic time into five eras in their 1997 review “A dying universe” and their 1999 book The Five Ages of the Universe. They count time in “cosmological decades”: η is the power of ten of the age in years, so today η is about 10.
Yes, by any measure that uses years. Its 13.8 billion years are about 0.014% of the 1014-year Stelliferous Era alone. Squeeze the age of stars into a 100-year human life and the universe is about five days old. We show the arithmetic, and the honest counterarguments, in is the universe young?
Physicists take five endings seriously. Which one is ours depends mainly on how dark energy behaves over time.
The heat death of the universe is the default in the standard cosmological model, ΛCDM. Expansion continues, everything cools toward a floor set by the cosmic horizon (around 10-30 kelvin), and once the last black hole is gone, no source of usable energy remains.
If dark energy grows denser as space expands (“phantom” energy, with an equation of state w below −1), expansion eventually tears apart galaxy clusters, galaxies, planets and atoms. In the 2003 Big Rip paper by Robert Caldwell, Marc Kamionkowski and Nevin Weinberg, an example with w = −1.5 ends the universe 22 billion years from now.
If dark energy weakens and turns negative, gravity wins and the universe recollapses. A model by Hoang Nhan Luu, Yu-Cheng Qiu and S.-H. Henry Tye, fitted to DES and DESI data and published in JCAP in 2025, gives a crunch about 20 billion years from now. That figure rests on their specific model and its best-fit parameters.
Some theories replace an ending with a new beginning: a crunch that rebounds into a new expansion, as in the cyclic model of Paul Steinhardt and Neil Turok (2002). These ideas are mathematically developed but have no observational support yet.
Vacuum decay does not depend on expansion at all. If the Higgs field is not in its lowest possible energy state, a bubble of “true vacuum” could appear anywhere and expand at nearly the speed of light, rewriting the laws of physics inside it.
| Scenario | What dark energy must do | When | How likely, as of October 2026 |
|---|---|---|---|
| Heat death (Big Freeze) | Stay constant (w = −1) or fade but stay positive | Stars end around 1014 years, black holes around 10100 | Leading scenario |
| Big Rip | Grow stronger forever (w below −1) | 22 billion years from now for w = −1.5 | Disfavored: DESI prefers w above −1 today |
| Big Crunch | Fade and turn negative | About 20 billion years from now in one 2025 model | Possible, model-dependent |
| Big Bounce or cycles | Recollapse, then rebound | Model-dependent | Speculative |
| Vacuum decay | Independent of dark energy | Expected wait far beyond 10100 years | Possible in principle, very unlikely soon |
For a side-by-side look at the three classic endings, see Big Rip vs Big Crunch vs Big Freeze.
Dark energy controls the future of the universe because it already dominates its energy budget and, unlike matter, does not thin out (or thins out slowly) as space expands. Physicists describe it with an equation of state, w: the ratio of its pressure to its energy density. A cosmological constant, Einstein’s Λ, has w = −1 exactly, and its density never changes. Combining Planck with supernovae gave w = −1.03 ± 0.03, consistent with a constant.
That consensus has wobbled since 2024. The Dark Energy Spectroscopic Instrument (DESI), which maps millions of galaxies and quasars, reported in its first-year results a preference for dark energy that changes with time, at 2.5 to 3.9 sigma depending on the supernova sample. DESI’s second data release in March 2025 strengthened the hint to 3.1 sigma with the cosmic microwave background, and 2.8 to 4.2 sigma with supernovae. A July 2026 Lyman-alpha analysis still prefers evolving dark energy, at 2.7 to 3.2 sigma.
None of this reaches the 5-sigma standard for a discovery, and the strongest numbers lean on supernova calibrations that are still being revised. We track the evidence in is dark energy weakening? and in our dark energy guide. DESI’s full five-year map, finished in April 2026, should give a sharper answer in 2027.
Probably yes. Under current data, the universe will expand forever: dark energy has dominated for billions of years and nothing measured so far suggests it will switch sign. Even if it is weakening, as DESI hints, a slow fade keeps the expansion going, just less steeply.
The expansion would stop and reverse only if dark energy became negative, as in the 2025 crunch model. Matter alone cannot do it: there is far too little of it, about 31.5% of the critical density according to Planck. So the answer to “will the universe stop expanding?” is: not unless dark energy holds a surprise.
A side effect of eternal expansion is cosmic isolation. In about 100 billion years every galaxy beyond our Local Group will have moved out of view, and astronomers then would see one merged galaxy in an apparently static universe.
Once the stars are gone, matter keeps changing, only much more slowly.
Proton decay. Grand unified theories predict that protons are unstable. If so, white dwarfs and neutron stars slowly evaporate into radiation and lighter particles; Adams and Laughlin adopted a representative lifetime of 1037 years. No proton decay has ever been seen. Super-Kamiokande, a 50,000-ton water detector in Japan, set a lower limit of 2.4 × 1034 years for the main predicted decay mode (2020). If protons are stable, remnants last far longer.
Hawking radiation. In 1974 Stephen Hawking showed that black holes are not eternal: quantum effects make them glow faintly and lose mass. Don Page’s 1976 calculations turned that into lifetimes, which grow as the cube of the mass. A solar-mass black hole lasts about 1067 years, and one of 100 billion solar masses, the mass of a large galaxy, about 10100 years. Hawking radiation has never been detected from a real black hole.
A contested shortcut. In 2023 and 2025, Heino Falcke, Michael Wondrak and Walter van Suijlekom proposed that spacetime curvature alone, without an event horizon, makes all compact objects evaporate. Their 2025 JCAP paper puts the last white dwarfs at roughly 1078 years. A published comment argued their method conflicts with exact results in simpler settings; the authors disagree. Treat it as a proposal.
Nobody knows, and the answer has changed with the data. In 1979 Freeman Dyson argued in “Time without end” that life could last forever in an open universe by thinking ever more slowly and hibernating to save energy. Accelerating expansion undercut that hope: in 2000, Lawrence Krauss and Glenn Starkman concluded that in an accelerating universe any civilization’s total conscious lifetime is finite, though it could be astronomically long.
Later ideas, such as computing near the cold horizon or harvesting black holes, remain speculative. What physics does say is that the Stelliferous Era is by far the most energy-rich time the universe will ever have. Life, if it is going to flourish anywhere, has its best chance now and over the next few trillion years around red dwarfs.
The study of the far future has a formal name: physical eschatology. It became a quantitative science with Dyson’s 1979 paper and Adams and Laughlin’s 1997 review, which follows stars, galaxies and particles over more than 150 orders of magnitude in time.
The method is simple to state: take physics measured today, such as stellar evolution, orbital dynamics, quantum field theory and the expansion rate, and run it forward. Some steps rest on direct measurements, like the ages of stars and the cosmic expansion. Others extrapolate tested theory far beyond any test, like Hawking radiation. A few depend on untested physics, like proton decay and the Higgs vacuum.
The biggest lever is observational: precise maps of cosmic expansion from DESI, Euclid, the Vera C. Rubin Observatory and the Nancy Grace Roman Space Telescope, launched on 30 August 2026. If you prefer to watch, melodysheep’s 29-minute “Timelapse of the Future” and PBS Space Time’s “How Will the Universe End?” are in our video library.
Not with a single bang, on current evidence. The leading forecast is a gradual heat death: stars stop forming within about 100 trillion years, black holes evaporate by roughly 10100 years, and space keeps expanding into a cold, near-empty state. Alternative endings depend on dark energy, which is still being measured.
There is no date. In the standard model the universe never stops existing, it just runs down. Its last stars fade around 1014 years after the Big Bang and its last black holes evaporate around 10100 years. One model fitted to recent dark energy data instead predicts a collapse about 20 billion years from now.
Heat death is a process rather than a moment. Normal star formation ends around 1014 years, stellar remnants dominate until roughly 1039 years, and the biggest black holes finish evaporating near 10100 years. After that the Dark Era begins, with almost nothing left to happen.
Probably. Since 1998, supernova data have shown the expansion is speeding up, driven by dark energy. If dark energy stays constant or fades slowly, expansion continues forever. Only if dark energy eventually turns negative, which some models fitted to DESI data allow, would expansion stop and reverse.
Slowly, in one sense. The universe forms stars at a far lower rate than at its peak about 10 billion years ago. But most stars that will ever shine are small red dwarfs that last trillions of years, so on the scale of its full future the universe is early in its life.
Red dwarfs, the smallest true stars. A red dwarf of 0.08 solar masses burns hydrogen for about 12 trillion years, heats up into a blue dwarf and fades as a white dwarf. With gas running out over trillions of years, normal star formation should end around 100 trillion years.
In principle, yes, through vacuum decay: if the Higgs field sits in a metastable state, a bubble of lower energy could form and spread at nearly light speed. Central estimates from Standard Model calculations range from about 10139 to 10790 years, and unknown physics could change the answer either way.
No. The 2036 date is an internet meme with no scientific basis. No published physics predicts any end of the universe in this century. The heat death, if it happens, unfolds over trillions of trillions of years, and even the fastest proposed collapse is about 20 billion years away.
Page, Physical Review D, 13, 198-206, 1976
Calculated in 1976 what an evaporating black hole emits and how long it lives, a time growing as its mass cubed. Don Page's Hawking radiation numbers.
Dyson, Reviews of Modern Physics, 51, 447-460, 1979
Asked in 1979 what happens in an ever-expanding universe over unimaginable times, and whether life could last forever. Freeman Dyson's founding paper.
Adams and Laughlin, Reviews of Modern Physics, 69, 337-372, 1997
Mapped the universe's future into eras, from the age of stars to a dark era beyond 10100 years. Fred Adams and Gregory Laughlin's 1997 review, explained.
Caldwell et al., Physical Review Letters, 91, 071301, 2003
Showed in 2003 that dark energy with w below -1 would tear apart galaxies, planets and atoms in a finite time. Caldwell et al. on the Big Rip.
Krauss and Scherrer, General Relativity and Gravitation, 39, 1545-1550, 2007
Argued in 2007 that accelerating expansion will erase the evidence of the Big Bang, leaving a static-looking universe. Krauss and Scherrer's essay.
Schröder and Smith, Monthly Notices of the Royal Astronomical Society, 386, 155-163, 2008
Calculated in 2008 that the red giant Sun will engulf Earth about 7.6 billion years from now, despite losing a third of its mass. Schröder and Smith.
Falcke et al., Journal of Cosmology and Astroparticle Physics, 2025
Proposed in 2025 that white dwarfs and neutron stars evaporate by Hawking-like radiation, the last within 1078 years. Falcke, Wondrak, van Suijlekom.
Sawala et al., Nature Astronomy, 2025
Found in 2025 only about a 50% chance that the Milky Way and Andromeda merge in the next 10 billion years, using Hubble and Gaia data. Till Sawala's team.
A cinematic timelapse of the future of the universe, jumping forward in ever larger steps from the present to the death of the Sun, the end of all stars and beyond. Pure visuals and music, no prior knowledge needed. melodysheep
An animated look at the far future of the universe that asks what the very last event will be once every star has burned out. A gentle, non-technical introduction to the idea of heat death. Kurzgesagt - In a Nutshell
Explains heat death, the slow fading of stars, galaxies and black holes into a cold, empty universe, then explores whether some loophole could let life continue. Animated and aimed at newcomers. Kurzgesagt - In a Nutshell
Covers the long-term future of the universe after the Stelliferous Era of shining stars ends, about 100 trillion years from now, and what can still happen in the dark that follows. Suits viewers comfortable with some physics. PBS Space Time
Imagines what an observer would see in the far future of our galaxy as stars die, and how long life might hold on. A thoughtful companion to the eras of deep time, pitched at curious non-specialists. PBS Space Time
Explains white dwarfs, the dead cores of stars that will be the last lights in the universe, and how they will cool into black dwarfs. Short, animated and easy to follow. Kurzgesagt - In a Nutshell
Starts from the heat death of the universe and walks through calculations predicting one final kind of supernova that can only occur near the end of time. A detailed episode for viewers who like the theory. PBS Space Time
Explains Roger Penrose's Conformal Cyclic Cosmology, the idea that the end of our universe could become the Big Bang of the next one. Physics-heavy and best for viewers who already know the basics of cosmology. PBS Space Time
Theoretical astrophysicist Katie Mack's public lecture on how the universe could end, from heat death to more abrupt scenarios, and what each would look like. Clear and engaging for a general audience over an hour. Perimeter Institute for Theoretical Physics
Astrophysicist Becky Smethurst explains how the fate of the universe depends on the contest between gravity and cosmic expansion, and why recent results question whether heat death is certain. Good for following current research. Dr. Becky