Will the Universe End in Heat Death? What Physics Can Say

“Heat death” sounds like the universe bursting into flames. It means the opposite: a universe that has run out of usable differences — no hot next to cold, no fuel next to empty space — so nothing can do work any more. Our Disordra mug keeps the score: entropy only goes up. This post separates what is solid physics, what is a forecast, and what is open. It builds on our explainer Why Does Time Only Go Forward?.

Where did the idea come from?

In 1852 William Thomson (later Lord Kelvin) published a short paper titled “On a Universal Tendency in Nature to the Dissipation of Mechanical Energy”. Its conclusions include that there is at present in the material world a universal tendency to the dissipation of mechanical energy, and that restoring it without more than an equivalent dissipation is impossible in inanimate processes. His third conclusion is the seed of the heat-death idea: “Within a finite period of time past, the earth must have been, and within a finite period of time to come the earth must again be, unfit for the habitation of man as at present constituted…” (Thomson 1852^[1]). Kelvin was talking about Earth and its energy sources. Later writers extended the reasoning to the whole universe; Adams and Laughlin’s review traces the concept to the mid-nineteenth century work of Helmholtz and Clausius (Adams & Laughlin 1997^[2]).

What does entropy have to do with it?

The second law of thermodynamics says that the entropy S of an isolated system does not decrease:

ΔS ≥ 0

Entropy (in joules per kelvin) measures, roughly, how many microscopic arrangements are compatible with what we see at large scale; Boltzmann’s S = k_B ln W makes that precise, where k_B is the Boltzmann constant and W the number of microstates (see What Exactly Is Entropy?). Engines, stars and living things all run on gradients — differences in temperature, pressure or chemical potential — and every time they use one, total entropy goes up. A system at maximum entropy has no gradients left. That is the classical heat death: the universe in thermodynamic equilibrium, at one temperature everywhere, unable to run any heat engine (Adams & Laughlin, §VI.D^[2]).

Why the classical picture needs updating

The universe is expanding, and that changes the question. Adams and Laughlin point out that a continually expanding universe never reaches true thermodynamic equilibrium, so classical heat death in the strict sense is avoided. What can still happen is what they call cosmological heat death: the expansion becomes effectively adiabatic, the entropy in a comoving volume approaches a constant, and the universe becomes “a dull and lifeless place with no ability to do physical work” (Adams & Laughlin 1997^[2]). Their paper works through the eras ahead in powers of ten of years: the present star-forming era; an era dominated by stellar remnants; an era in which black holes dominate; and a final dark era. They estimate that energy and entropy production continue until roughly 10¹⁰⁰ years, postponing heat death at least that long.

Two caveats are essential. First, that review was published in 1997, a year before the accelerating expansion was announced, so it surveys several possible futures rather than one. Second, some steps in that timeline rely on processes that have never been observed. Proton decay is predicted by some unified theories but has not been seen; Super-Kamiokande’s search sets a lower limit on the partial lifetime for p → e⁺π⁰ of 2.4 × 10³⁴ years (Super-Kamiokande 2020^[3]). Black-hole evaporation by Hawking radiation is a firm theoretical prediction but is also unobserved for real black holes (our post Do Black Holes Evaporate? covers why).

Dark energy tips the balance toward “ice”

In 1998 and 1999 two teams measuring distant Type Ia supernovae found them fainter than expected, implying that the expansion of the universe is speeding up (Riess et al. 1998^[4]; Perlmutter et al. 1999^[5]). Saul Perlmutter, Brian Schmidt and Adam Riess received the 2011 Nobel Prize in Physics “for the discovery of the accelerating expansion of the Universe through observations of distant supernovae” (NobelPrize.org 2011^[6]). The Nobel press release draws the obvious consequence: if the expansion continues to speed up, the universe “will end in ice” (Nobel 2011 press release^[7]).

The acceleration is attributed to dark energy, which NASA describes as making up roughly 68.3 to 70% of the universe and whose nature, in NASA’s words, is a short answer of “We don’t know” (NASA Science^[8]). Planck’s final cosmic-microwave-background analysis is well fitted by the standard ΛCDM model, in which dark energy is a cosmological constant Λ (Planck 2018^[9]). If that is right, galaxies outside our gravitationally bound neighbourhood keep receding faster, stars eventually stop forming, and the universe trends toward cold, dark dilution: a heat death by expansion, not by everything reaching one warm temperature.

The equation of state that decides it

Cosmologists summarise dark energy by its equation-of-state parameter

w = p / (ρc²)

where p is its pressure, ρ its mass density and c the speed of light. A cosmological constant has w = −1 exactly, forever. If instead w were below −1 (“phantom energy”), Caldwell, Kamionkowski and Weinberg showed the dark-energy density would become infinite in a finite time, ripping apart galaxies, planets and finally atoms — the “Big Rip” (Caldwell et al. 2003^[10]). That is a hypothetical case, not a prediction.

What is open right now

Whether w is exactly −1 is an active measurement question. In 2025 the DESI collaboration reported that its baryon-acoustic-oscillation data, combined with the CMB and supernovae, prefer a time-evolving dark energy over a pure cosmological constant at between 2.8σ and 4.2σ, depending on which supernova sample is used (DESI DR2, 2025^[11]). That is a hint, not a discovery, and it does not by itself say what the far future holds. If dark energy evolves, the long-term fate could differ from the simple ΛCDM forecast — in either direction. We label the ultimate fate of the universe open.

So, will the universe end in heat death?

• Solid: the second law holds; usable energy differences are being used up; the expansion is accelerating.
• Best current forecast: a cold, ever more dilute universe in which star formation ends and structure slowly decays — a heat death by expansion.
• Open: what dark energy is and whether it changes; whether protons decay; how black-hole evaporation finishes.

In the shop

The Disordra mug is entropy keeping score, and Void Whisperer is dark energy pushing the galaxies apart. Browse more in Thermodynamics and Relativity & Cosmos.

FAQ

Does heat death mean the universe gets hot?

No. It means no usable temperature differences remain. In an accelerating universe the expected end state is very cold.

When will it happen?

Not on any human timescale. Adams and Laughlin expect conventionally evolving stars to stay central for several thousand Hubble times (a Hubble time is comparable to the universe’s current age), and estimate energy production continuing to about 10¹⁰⁰ years, under assumptions that include unobserved processes.

Could the universe end in a Big Crunch instead?

Present measurements favour continued, accelerating expansion, which is why the 2011 Nobel press release spoke of “ice”. Because the nature of dark energy is unknown, the far future is a forecast from today’s data, not a certainty.

References

1. W. Thomson (Lord Kelvin) (1852), “On a Universal Tendency in Nature to the Dissipation of Mechanical Energy” (transcription; Phil. Mag. 4, 304, DOI 10.1080/14786445208647126). https://zapatopi.net/kelvin/papers/on_a_universal_tendency.html
2. F. C. Adams & G. Laughlin (1997), “A dying universe: the long-term fate and evolution of astrophysical objects,” Rev. Mod. Phys. 69, 337 (arXiv:astro-ph/9701131). https://arxiv.org/pdf/astro-ph/9701131
3. Super-Kamiokande Collaboration, A. Takenaka et al. (2020), “Search for proton decay via p→e⁺π⁰ and p→μ⁺π⁰ with an enlarged fiducial volume in Super-Kamiokande I–IV,” Phys. Rev. D 102, 112011 (arXiv:2010.16098). https://arxiv.org/abs/2010.16098
4. A. G. Riess et al. (1998), “Observational Evidence from Supernovae for an Accelerating Universe and a Cosmological Constant,” Astron. J. 116, 1009 (arXiv:astro-ph/9805201). https://arxiv.org/abs/astro-ph/9805201
5. S. Perlmutter et al. (1999), “Measurements of Ω and Λ from 42 High-Redshift Supernovae,” Astrophys. J. 517, 565 (arXiv:astro-ph/9812133). https://arxiv.org/abs/astro-ph/9812133
6. NobelPrize.org, The Nobel Prize in Physics 2011 (Perlmutter, Schmidt, Riess). https://www.nobelprize.org/prizes/physics/2011/summary/
7. NobelPrize.org, Press release: The Nobel Prize in Physics 2011. https://www.nobelprize.org/prizes/physics/2011/press-release/
8. NASA Science, “Dark Energy”. https://science.nasa.gov/dark-energy/
9. Planck Collaboration (2020), “Planck 2018 results. VI. Cosmological parameters,” A&A 641, A6 (arXiv:1807.06209). https://arxiv.org/abs/1807.06209
10. R. R. Caldwell, M. Kamionkowski & N. N. Weinberg (2003), “Phantom Energy: Dark Energy with w < −1 Causes a Cosmic Doomsday,” Phys. Rev. Lett. 91, 071301 (arXiv:astro-ph/0302506). https://arxiv.org/abs/astro-ph/0302506
11. DESI Collaboration, M. Abdul-Karim et al. (2025), “DESI DR2 Results II: Measurements of Baryon Acoustic Oscillations and Cosmological Constraints” (arXiv:2503.14738). https://arxiv.org/abs/2503.14738

Written by Pixelated Physics. Every factual claim is linked to the source we checked; points that are still debated are labelled open.