The closest anyone has come to absolute zero is roughly 38 picokelvins, or 38 trillionths of a degree above zero on the Kelvin scale. German physicists achieved this in 2021 by dropping a cloud of ultracold rubidium atoms down a tower in microgravity-like conditions, then allowing the gas to expand freely until its effective temperature plunged to a level almost indistinguishable from the theoretical floor of zero kelvin. That record sits at the current frontier of a centuries-long push toward colder and colder matter, a pursuit that has repeatedly rewritten physics along the way.
Why Absolute Zero Remains Out of Reach
Absolute zero, defined as 0 kelvin (about −273.15 °C), is the point at which a system would have no thermal energy left to give up. It is not simply very cold; it represents a hard boundary that thermodynamics forbids you from crossing or even touching. The principle at the heart of this prohibition, sometimes called the unattainability principle, states that no finite sequence of cooling steps can bring a system all the way to zero kelvin. Theoretical work has confirmed this constraint even in models that initially seemed to violate it; once realistic energy-level spacings are accounted for, the cooling rate drops to zero as the temperature approaches absolute zero.1PubMed Central. Cooling by heating: Restoration of the third law of thermodynamics You can always get colder, in other words, but each step takes more effort and removes less heat than the one before it, and the final step to exactly zero would require infinite work.
This is not just a practical obstacle like running out of coolant. It is baked into the mathematics of how energy states work at the quantum level. As a sample’s temperature drops, the atoms settle into lower and lower energy states, and extracting the last traces of energy becomes exponentially harder. The result is an asymptotic approach: researchers keep shaving off more decimal places, but they never land on zero.
From Liquid Helium to Bose-Einstein Condensates
The modern era of extreme cold began in 1908, when the Dutch physicist Heike Kamerlingh Onnes liquefied helium for the first time after more than a decade of preparation, building on earlier work by James Dewar, who had liquefied hydrogen in 1898.2Physica B: Condensed Matter. Cryogenics at the end of the 19th and the first half of the 20th century (1880-1940) Liquid helium boils at about 4.2 kelvin, and for decades that neighborhood represented the bottom of what laboratories could reliably reach. The liquefaction breakthrough also led almost immediately to the discovery of superconductivity, which Onnes stumbled onto just three years later when he found that mercury’s electrical resistance vanished below about 4 kelvin.
Through the 20th century, a cascade of improved techniques pushed temperatures lower. Dilution refrigerators, which exploit a mixture of two helium isotopes, brought bulk samples down to the millikelvin range. Adiabatic demagnetization, which removes energy by cycling a magnetic field applied to a paramagnetic material, pushed nuclear spin systems even further. Modern integrated cryogenic systems combine multiple stages to reach the 0.1 kelvin range and below.3IOP Conference Series: Materials Science and Engineering. Conceptual design of key components for a single-shot dilution refrigerator (DR) with small helium-3 Inventory assisted by adiabatic demagnetization refrigerator (ADR) Nuclear demagnetization techniques applied to metals like rhodium eventually brought spin temperatures into the picokelvin regime.4Physica B: Condensed Matter. Nuclear cooling and spin properties of rhodium down to picokelvin temperatures
The real revolution, though, came from a completely different direction. In 1995, researchers at JILA in Colorado created a Bose-Einstein condensate (BEC) in a vapor of rubidium-87 atoms. By confining the atoms in magnetic fields and evaporatively cooling them, they watched the condensate form near a temperature of about 170 nanokelvin, or 170 billionths of a degree above absolute zero.5PubMed. Observation of bose-einstein condensation in a dilute atomic vapor A BEC is a state of matter in which a group of atoms loses individual identity and behaves as a single quantum object, and producing one requires extraordinary cold. That 1995 achievement won the Nobel Prize and opened the door to the picokelvin experiments that would follow.
The Current Record and How It Was Set
The current lowest temperature on record was reported in 2021 by a team at the University of Bremen in Germany, who cooled a cloud of rubidium atoms to an effective temperature of about 38 picokelvins. To put that in perspective, 38 picokelvins is 38 trillionths of a kelvin, roughly two billion times colder than the depths of outer space. The team used the Bremen Drop Tower, a 120-meter-tall facility designed for microgravity experiments. After pre-cooling the atoms with lasers and magnetic traps, they released the cloud into free fall and turned off the trap, allowing the gas to expand in every direction. That expansion further cooled the cloud as it fell.6Physical Review Letters. New record set for lowest temperature—38 picokelvins
An earlier experiment along similar lines used a matter-wave lens, a brief pulse of laser light that acts like a lens for atoms, to collimate an expanding rubidium cloud and cool it in two dimensions to an effective temperature below about 50 picokelvins.7Physical Review Letters. Matter wave lensing to picokelvin temperatures That technique showed the basic principle: if you let a very cold gas expand and then refocus it, you can squeeze out even more thermal energy. The Bremen experiment refined and extended the approach by using the longer free-fall time available in their drop tower.
One subtlety worth flagging: at these extremes, “temperature” gets tricky. What researchers measure is the spread of velocities in the atom cloud. A narrower spread corresponds to a lower effective temperature. This is a legitimate and standard way to define temperature for a dilute gas, but it is not the same as cooling a solid block of metal to 38 picokelvins. The atoms in these experiments number in the thousands to millions, and they exist in a carefully engineered quantum state, not in a chunk of material you could hold in your hand.
How Researchers Push Temperatures This Low
Getting to picokelvin territory requires stacking multiple cooling techniques on top of one another, each one picking up where the previous one becomes ineffective.
- Laser cooling: A set of carefully tuned laser beams slows atoms by hitting them with photons whenever they move toward the light source. Each photon delivers a tiny momentum kick opposing the atom’s motion, gradually sapping kinetic energy. This alone can bring atoms down to the microkelvin range. Researchers have demonstrated sub-Doppler cooling of molecules like yttrium monoxide (YO) to about 4 microkelvin.8PubMed Central. Sub-Doppler Cooling and Compressed Trapping of YO Molecules at μK Temperatures Newer approaches like Sisyphus cooling schemes that exploit long-lived atomic clock states can push below even the recoil limit, the temperature associated with a single photon’s kick.9PubMed. Clock-Line-Mediated Sisyphus Cooling
- Evaporative cooling: Once atoms are trapped, the hottest ones are selectively allowed to escape. The remaining atoms re-equilibrate at a lower average energy, much like a cup of coffee cooling faster when steam carries away the most energetic water molecules. This is the technique that produced the first BEC.
- Magnetic and optical traps: Specially shaped magnetic fields or intersecting laser beams confine the cold atoms in a tiny region. The trap geometry matters enormously, because releasing the trap in the right way lets the cloud expand and cool further.
- Matter-wave lensing: A brief pulse of light or a magnetic field gradient acts as a lens on the expanding atom cloud, refocusing the atoms’ momentum distribution and reducing the effective temperature still further. This is the final step in the picokelvin experiments.
Each stage only works in the temperature range where it is efficient. Laser cooling is useless below a few microkelvin; evaporative cooling takes over from there down to nanokelvin; magnetic lensing and free expansion push into the picokelvin zone. The engineering challenge is stitching all of these stages together without losing atoms or introducing heat at each handoff.
Cooling Atoms in Space
Gravity is a nuisance when you are trying to observe an ultracold gas. On Earth, even inside a drop tower, free-fall time is limited to a few seconds. In orbit, free fall is essentially unlimited, which gives atoms much more time to expand and cool. NASA’s Cold Atom Lab (CAL), installed on the International Space Station in 2018, was built to exploit this advantage. Pathfinder experiments on CAL have demonstrated atom interferometry with free-expansion times exceeding 150 milliseconds, far longer than what is practical on the ground.10PubMed Central. Pathfinder experiments with atom interferometry in the Cold Atom Lab onboard the International Space Station
The primary goal of space-based cold-atom experiments is not simply to set temperature records but to build exquisitely sensitive measurement tools. Atom interferometers, which exploit the wave nature of ultracold atoms, can detect tiny variations in gravitational fields, test whether gravity behaves exactly as general relativity predicts, and search for forces that the Standard Model of particle physics does not account for. Persistent microgravity lets these instruments reach sensitivities that would be impossible in a ground-based lab, where the experiment ends the moment the atoms hit the floor of the vacuum chamber.
The Coldest Natural Place in the Universe
While labs manufacture extreme cold in tiny volumes, nature has produced at least one environment colder than the cosmic microwave background radiation that fills all of space. The Boomerang Nebula, a dying star roughly 5,000 light-years from Earth, has been confirmed as the coldest known natural object in the universe. ALMA telescope observations showed that its massive, high-speed outflow has cooled well below the background temperature of about 2.7 kelvin.11The Astrophysical Journal. The Coldest Place in the Universe: Probing the Ultra-cold Outflow and Dusty Disk in the Boomerang Nebula The nebula reaches roughly 1 kelvin in its coldest regions.
The cooling mechanism is analogous to how a can of compressed gas chills when you release the valve. The nebula’s central star is ejecting gas at very high speed, and as that gas expands rapidly into space, it drops in temperature, overshooting the ambient background. In most of the universe, any object left alone will eventually warm up to the 2.7 kelvin background by absorbing microwave photons, so the Boomerang Nebula’s extreme cold is a transient phenomenon, sustained only as long as the outflow keeps expanding faster than the background can reheat it.
Still, 1 kelvin is balmy compared to what labs achieve. The gap between the coldest natural environment and the coldest laboratory temperature spans about 11 orders of magnitude, a reminder of how far human engineering has outrun nature in this particular race.
When Temperature Goes Negative
There is a genuinely strange corner of thermodynamics where temperature can be defined as negative, and it is worth understanding why this does not mean “colder than absolute zero.” In 2013, a team in Munich created a quantum gas of potassium atoms in an optical lattice where the atoms occupied high-energy states more than low-energy states. By the formal thermodynamic definition of temperature, that population inversion corresponds to a negative absolute temperature.12PubMed. Negative absolute temperature for motional degrees of freedom
The confusing part is that a system at negative temperature is not colder than zero. It is, in a well-defined sense, hotter than any positive temperature. Temperature in physics describes how a system’s energy changes when you add entropy, and at negative temperature, the system has so much energy packed into its highest states that adding more entropy actually means lowering its energy. If you brought a negative-temperature gas into contact with a positive-temperature one, energy would flow from the negative-temperature system to the positive-temperature one, the same direction heat flows from a hot object to a cold one. The negative-temperature system is the hot one.
These states can only exist in systems with a finite upper limit on energy, such as atoms confined in a lattice where they cannot move faster than a certain speed. They cannot occur in normal gases or solids, where there is no cap on how much kinetic energy a particle can have. So negative temperatures are not a loophole to reaching or surpassing absolute zero; they are a separate phenomenon that lives on the other side of infinity on the temperature scale, accessible only in carefully engineered quantum systems.
Laser Cooling Antimatter
One of the more striking recent achievements in ultracold physics is the laser cooling of antihydrogen, the antimatter counterpart of ordinary hydrogen. In 2021, the ALPHA collaboration at CERN demonstrated that antihydrogen atoms trapped in a magnetic bottle could be slowed and cooled using precisely tuned laser light, the same basic principle that has been cooling ordinary atoms for four decades.13PubMed Central. Laser cooling of antihydrogen atoms
The motivation is not to set temperature records with antimatter. It is to make antimatter cold enough to study with extreme precision. One of the deepest open questions in physics is why the universe contains overwhelmingly more matter than antimatter, when the Big Bang should have produced them in equal amounts. Comparing the spectral properties and gravitational behavior of antihydrogen to those of ordinary hydrogen could reveal tiny asymmetries between matter and antimatter that current theories do not predict. Colder antihydrogen stays trapped longer, moves less, and produces sharper spectral lines, all of which make these measurements more sensitive. The ALPHA team noted that the achievement opens the door to future experiments like anti-atom fountains and antimatter interferometry, precision tools that would have been unthinkable without laser cooling.
What Gets Easier and What Gets Harder Near the Bottom
An underappreciated aspect of the push toward absolute zero is how the challenges shift as you go lower. In the millikelvin range, the main enemy is external heat leaking in through vibrations, radiation, or poorly insulated wiring. Shielding and isolation are paramount, and the engineering is largely mechanical: better vacuum chambers, cleaner materials, more stages of thermal insulation. By the nanokelvin range, those problems are largely solved, and the challenge becomes controlling the atoms themselves. You need magnetic fields stable to parts per billion, laser frequencies locked to extraordinary precision, and vacuum pressures so low that a stray air molecule hitting the cloud would destroy the experiment.
At the picokelvin frontier, even the measurement becomes an issue. You cannot stick a thermometer into a cloud of a few thousand atoms. Instead, you infer the temperature from the cloud’s expansion rate after it is released from a trap: you take an image of the cloud at two different times, measure how fast it spread, and work backward to a temperature. Any imperfection in the imaging, any residual magnetic field gradient nudging the atoms, or any interaction between atoms that your model does not account for introduces uncertainty. The 38-picokelvin result, for instance, is an effective kinetic temperature inferred from the center-of-mass motion of a BEC released in microgravity. Claiming a temperature that low requires painstaking control over every variable, and the error bars at this frontier are a significant fraction of the measurement itself.
Reaching still lower temperatures will likely depend on longer free-fall times in space, more sophisticated matter-wave optics, and perhaps new trap geometries that have not yet been invented. Whether someone will report a femtokelvin measurement in the coming decades is hard to predict, but the unattainability principle guarantees that there will always be another decimal place to chase.