Dark energy

• 1. Introduction

  ›

  Dark energy is the name scientists gave to whatever is pushing the universe apart at an ever-faster rate, and it makes up 68% of everything that exists. That single number hides a profound embarrassment: nobody knows what dark energy actually is. What physicists do know is that ordinary matter, the atoms that make up planets, stars, and human beings, accounts for only 5% of the universe's total mass-energy. Dark matter contributes another 27%. Dark energy fills the rest. Its density is extraordinarily low, roughly 7 times 10 to the negative 30 grams per cubic centimetre, far thinner than any ordinary substance. Yet because it spreads uniformly across all of space, it outweighs everything else combined. The questions the following chapters will explore are deceptively simple: how did we discover something so thin it cannot be detected in a laboratory, what might it actually be, and what will it do to the universe in the end?

• 2. Einstein's Blunder

  ›

  Albert Einstein first added a term called the cosmological constant to his equations of general relativity not because he thought the universe was expanding, but because he thought it was not. He wanted a mathematical mechanism to prevent gravity from pulling everything into a single point, and the constant, which he labelled with the Greek capital letter Lambda, gave empty space its own gravitational property. Einstein described the concept by stating that empty space takes the role of gravitating negative masses distributed all over interstellar space. The arrangement was inherently precarious. If the universe expanded even slightly, the resulting vacuum energy would drive still more expansion. A slight contraction would snowball into collapse. Neither outcome is stable. When Edwin Hubble's observations in 1929 showed that the universe is in fact expanding, Einstein reportedly called his failure to predict a dynamic universe his greatest blunder. What he could not have anticipated was that decades later a real cosmological constant, or something very much like it, would be precisely what the data demanded.

• 3. The 1998 Discovery

  ›

  In 1998, the High-Z Supernova Search Team published observations of Type Ia supernovae, and in 1999 the Supernova Cosmology Project followed with further evidence that the universe's expansion is speeding up rather than slowing down. The finding caught scientists off guard because the prevailing expectation was that gravity would gradually brake the expansion begun by the Big Bang. Type Ia supernovae made that test possible because they all explode with the same intrinsic brightness. By comparing how bright a supernova looks against how fast it appears to be receding, astronomers can measure the expansion history of the universe with unusual precision. The 2011 Nobel Prize in Physics was awarded to Saul Perlmutter, Brian P. Schmidt, and Adam G. Riess for their leadership in the discovery. The term dark energy itself was coined by cosmologist Michael S. Turner in 1998 for a paper written with Saul Perlmutter and Martin White. Independent confirmation came quickly: in 2000, the BOOMERanG and Maxima experiments measured the first acoustic peak in the cosmic microwave background and found that the total mass-energy density is close to 100% of the critical density, while the 2dF Galaxy Redshift Survey in 2001 showed that matter accounts for only about 30% of that total, leaving a large gap filled by something smooth and pressureless.

• 4. Weighing an Invisible Force

  ›

  Three independent lines of evidence now underpin dark energy's existence. The first comes from supernovae, which show the universe expanded more rapidly in the latter half of its life than in the first half. The second comes from the geometry of space: measurements of cosmic microwave background anisotropies show the universe is close to flat, but all the matter counted adds up to only about 30% of the density needed to produce that flat geometry, implying something else supplies the rest. The third comes from the large-scale clustering of galaxies. A 2011 survey called WiggleZ examined more than 200,000 galaxies from the Australian Astronomical Observatory. By measuring the characteristic spacing of baryon acoustic oscillations, roughly 150 megaparsecs in diameter, researchers used those voids as standard rulers to estimate distances as far as 2,000 megaparsecs, equivalent to a redshift of 0.6. The data confirmed cosmic acceleration reaching back roughly 7 billion years, while constraining the universe's large-scale inhomogeneity to 1 part in 10. Much more precise measurements from the WMAP spacecraft between 2003 and 2010 continued to support the same picture, and work based on the Planck spacecraft in 2013 yielded a refined estimate of 68.3% dark energy, 26.8% dark matter, and 4.9% ordinary matter.

• 5. What Dark Energy Might Be

  ›

  The simplest candidate for dark energy is a cosmological constant: a fixed energy density of empty space that never changes. Quantum field theory predicts that the vacuum should carry energy through the constant creation and annihilation of particle-antiparticle pairs. The trouble is that theory predicts a vacuum energy roughly 120 orders of magnitude larger than what is observed, a disagreement sometimes called the cosmological constant problem, and it remains unresolved. A second family of candidates falls under the name quintessence, which replaces a fixed constant with a dynamical field that can vary across time and space. One particular quintessence scenario, called Quintom, was proposed by Xinmin Zhang's group in 2004 after data hinted that the equation of state had crossed the cosmological constant boundary from above to below, a behaviour that requires at least two types of scalar fields. Phantom dark energy, a special case, is unusual because its energy density would actually increase with time, potentially culminating in a Big Rip. Interacting dark energy models take a different approach entirely, treating dark energy and dark matter as different aspects of a single unknown substance or proposing that cold dark matter decays into dark energy. In the early 2020s, researchers briefly theorized that excess signal observed in the XENON1T detector in Italy might reflect a chameleon model of dark energy, but further experiments ruled that out.

• 6. New Evidence of Change

  ›

  In March 2025, the Dark Energy Spectroscopic Instrument collaboration, known as DESI, announced that evidence for evolving dark energy had emerged from analysis combining DESI data on baryon acoustic oscillations with measurements of the cosmic microwave background, weak lensing, and supernovae. The statistical significance ranged from 2.8 to 4.2 sigma, and the results suggest that the density of dark energy is slowly decreasing with time. Earlier data from DESI had already indicated that the amount of dark energy may be about 10% lower now than it was 4.5 billion years ago. Neither result yet rules out the cosmological constant, but together they give quintessence and variable dark energy models their first serious observational footing. The Hubble Space Telescope's Higher-Z Team had previously provided evidence that dark energy has been present for at least 9 billion years, reaching back well before the onset of cosmic acceleration.

• 7. Fates of the Universe

  ›

  Cosmologists estimate that the accelerated expansion began roughly 5 billion years ago, after dark energy's influence overtook matter's gravitational pull. What happens next depends almost entirely on what dark energy turns out to be. If the cosmological constant holds, galaxies beyond the Local Group will recede ever faster. The current distance to what amounts to a cosmological event horizon stands at about 16 billion light years: any event happening today beyond that threshold will never send a signal that reaches Earth. Galaxies approaching that boundary will appear increasingly redshifted until they vanish entirely. The Local Group, including the Milky Way, would remain bound together while the rest of the universe fades from view, ultimately reaching heat death. If phantom dark energy governs instead, the growing force would eventually tear apart galaxies, solar systems, and ultimately individual atoms in a scenario called the Big Rip. A third possibility is that dark energy weakens or reverses, allowing gravity to eventually win. That would end in a Big Crunch. Some models even propose a cyclic universe in which each iteration, from Big Bang to Big Crunch, spans roughly a trillion years. None of these alternatives is supported by current observations, but none has been eliminated either, which means the fate of everything depends on understanding a substance that fills all of space yet remains beyond laboratory detection.