How did we know the lowest possible temperature is minus 273°C?
― Figured out from how air shrinks, long before anyone could cool anything that far
Temperature has no ceiling, but it does have a floor: minus 273.15°C, "absolute zero." What's striking is that this number was pinned down long before humans could actually cool anything that cold. The clue was the same thing that makes a balloon shrink in the cold.
A beach ball, blown up tight in the summer sun, looks a bit saggy in a chilly room at night. A bike tyre feels softer on a winter morning. In neither case did anyone let the air out.
Air shrinks when it cools. So what happens if you keep cooling it? If it keeps shrinking the colder it gets, does its volume eventually hit zero?
Scientists about 300 years ago wondered exactly the same thing. And they worked out that "zero-volume temperature" with calculation alone.
Just two clues were enough
Cool air at 0°C by 1°C, and its volume drops by about 1/273. Cool it by 2°C, and it drops by 2/273. This fraction was nearly the same whether the gas was air, oxygen, or hydrogen.
Plot temperature against volume, and the points fall neatly on a straight line. A straight line can be extended beyond what you actually measured. Follow it far enough, and you hit the temperature where volume reaches zero.
In other words, nobody actually cooled anything that far. They read off the dead end from the "slope" measured over a range they could reach. Let's follow the story in order.
The people who measured how air shrinks
The first person to get close to this idea was Amontons in France. Around 1702, using a thermometer with trapped air, he studied the relationship between temperature and the air's pushing force. From that slope, he is said to have estimated that the pushing force would vanish at around minus 240°C.
About 80 years later, Charles, also French, studied how the volume of various gases changed when heated. Charles is also known as the first person to fly in a hydrogen balloon. He never published his results as a paper, though.
In 1802, Gay-Lussac published more careful measurements. The surprise was that the rate of expansion was almost the same regardless of the type of gas. Heated from 0°C to 100°C, every gas expanded to about 1.37 times its original volume.
Take a look at Figure 1. The solid line in the upper right is the range actually measured, from 0°C to 100°C. Extend that line down to the lower left as a dotted line, and it hits zero volume at minus 273°C. Try the dial: change the temperature and see how the air inside the cylinder at upper left shrinks.
The person who grasped what "zero volume" really meant
But there was a snag. Real air turns to liquid well before it reaches minus 273°C. Zero volume never actually happens. So was that dead end at the end of the dotted line just a number on paper?
The answer came from William Thomson of Britain, later Lord Kelvin. In 1848, he rethought what temperature itself actually means. At the time, the idea that heat is really "the motion of tiny particles inside matter" was gaining ground.
Look at Figure 2. Particles of gas are always flying around and bouncing off the walls of their container. The warmer they are, the faster they move, and the harder they push the walls outward. Cool them down, and their motion slows, so the push weakens and the gas shrinks. In other words, that dead end on the dotted line is the temperature where the particles' motion is as small as it can possibly be. Nothing can move slower than not moving at all. That's why there's no temperature below it.
Thomson proposed a new temperature scale starting from this dead end as its origin: the "kelvin," the unit we still use today. It sets minus 273.15°C as 0 kelvin, with each degree the same size as 1°C. An ordinary room temperature works out to about 300 kelvin.
Checking the answer took a century
Once calculation had located the dead end, a race began to see how close anyone could actually get. Oxygen was liquefied in 1877, hydrogen in 1898. In 1908, the Dutch physicist Kamerlingh Onnes liquefied helium, reaching minus 269°C.
Even cooled that far, the gas shrank by exactly the predicted fraction — the dead end hadn't moved. The dotted line on paper turned out to be real. Three years later, Onnes also discovered "superconductivity": the sudden disappearance of electrical resistance in chilled mercury.
Today, laboratories can get within about a billionth of a degree of that floor. Even so, it can never be reached exactly: the closer you get, the less effective the very tools used to remove heat become.
The background of space sits at about minus 270°C. But a cloud of gas called the Boomerang Nebula is thought to be even colder, around minus 272°C. That's because gas rushing outward cools as it expands. It's the same mechanism as a chilled spray can, playing out on the scale of a star.
Summary
Absolute zero wasn't found by cooling something all the way down to it. It was found by precisely measuring the "slope" at which a gas shrinks between 0°C and 100°C, then extending that line. Its true meaning — the point where particle motion is as small as possible — was understood later, and experiment caught up a century after that.
Humans measured their way to a place nobody could ever reach.
The straight line traced by shrinking air pointed the way there.
You can read about what makes space cold in "Is space really cold?", and how carrying heat away cools things in "Why does a fridge warm up the room?". For another story about cleverly weighing the unreachable, see "How did we weigh the Earth?"
- Blow up a balloon and tie it off, then use a piece of string to measure around its widest point.
- Put the balloon in the freezer for about 30 minutes, then take it out and immediately measure the same spot again.
- Warm it with your hands for a while, and check that it returns to its original size.
Moving from a 25°C room to a minus 18°C freezer should shrink the volume by about 14% by calculation. That works out to roughly a 5% drop in circumference — about 1.5cm narrower for a 30cm balloon. Because rubber has its own elasticity, the actual result won't match this exactly.
Want to go deeper? ― Terms, formulas, and where this fits in the curriculumEach item is labeled by level, from middle-school science to university specialist courses
- MSCovered in middle-school science
- HSCovered in high-school physics or chemistry
- HS+Advanced high-school material, or a textbook sidebar topic
- UnivNot covered in high school — university-level thermodynamics/statistical mechanics
- ResearchNot yet settled even at university level — an active research question
MSTerms: this phenomenon has names
- Absolute zero: the lower limit below which temperature cannot go. Minus 273.15°C.
- Charles's law: at constant pressure, a gas's volume increases with temperature at a fixed rate. Also called Gay-Lussac's law.
- Absolute temperature (kelvin): a temperature scale with absolute zero as 0. Symbol K. 0°C is 273.15 K.
MSHSCheck it with a formula: finding the dead-end temperature from a measured "slope"
Let's calculate the zero-volume temperature using only the same kind of data Gay-Lussac had: "volume at 0°C and at 100°C."
| In symbols | V = V0 × ( 1 + t ÷ 273 ) → V = 0 when t = −273 |
| In words | Current volume = volume at 0°C × (1 + temperature ÷ 273). The zero-volume temperature is minus (the reciprocal of the fraction gained per 1°C) |
| Where it comes from | The straight-line relationship between volume and temperature at constant pressure (Charles's law). At its root is the link between temperature and how vigorously particles move |
| Symbol | Meaning and unit |
| V | Gas volume at that moment (litres) |
| V0 | Volume at 0°C (litres) |
| t | Celsius temperature (°C) |
| Air's volume at 0°C | 1 litre |
| Volume after heating to 100°C (measured) | 1.366 litres |
| Temperature range heated | 100 °C |
| Volume gained over 100°C | 1.366 − 1 = 0.366 litres |
| Volume gained per 1°C | 0.366 ÷ 100 = 0.00366 litres |
| Temperature drop needed to lose 1 litre entirely | 1 ÷ 0.00366 ≒ 273 °C |
| Dead-end temperature | 0 − 273 = −273 °C |
Using only measurements between 0°C and 100°C, the dead end comes out to minus 273°C. Gay-Lussac's own measurements gave a fraction closer to 1/267, putting the dead end at around minus 267°C. As measurement precision improved, the figure converged toward 273.
| Room temperature (27°C) in absolute temperature | 27 + 273 = 300 K |
| Liquid nitrogen temperature (−196°C) in absolute temperature | −196 + 273 = 77 K |
| Resulting volume fraction | 300 ÷ 77 ≒ 3.9 |
Cool a balloon with liquid nitrogen, and its volume should shrink to roughly a quarter, by calculation. That's the dramatic shrinkage often shown in science-museum demonstrations.
HSHS+Putting volume, pressure, and temperature together
HSHigh school covers "Boyle-Charles' law," which combines Charles's law with Boyle's law (at constant temperature, volume and pressure are inversely proportional). Add in the amount of substance, and you get the ideal gas equation, PV = nRT. Here T is absolute temperature — the equation doesn't come out as cleanly using Celsius.
HS+In the kinetic theory of gases, the average kinetic energy of a single particle is proportional to absolute temperature: (3/2)kT (where k is the Boltzmann constant). This equation shows that absolute temperature is, in effect, "a scale for the energy of particle motion."
UnivA temperature scale independent of any substance, and the third law of thermodynamics
Thomson's 1848 proposal was a "thermodynamic temperature," based on Carnot's theory of heat engines. Its breakthrough was defining temperature without relying on any particular substance's properties; it was later shown to agree with the scale read off a gas thermometer. The fact that absolute zero cannot be reached by any finite process is also captured in the third law of thermodynamics, which originated with Nernst's heat theorem. According to quantum mechanics, particle motion never fully stops even at absolute zero — a minimal residual jitter called zero-point motion remains.
📖 For the derivation and further reading: Absolute Zero (Wikipedia, Japanese) ／ Third Law of Thermodynamics (Wikipedia, Japanese)
ResearchWhat's still not fully understood
- How close can we actually get? Experiments have reportedly cooled groups of atoms to within tens of trillionths of a degree. Where the limits of cooling technology actually lie is still an open race.
- "Negative absolute temperature." In specially prepared groups of atoms, states have been created where, by the statistical definition, temperature comes out as negative kelvin. This isn't colder than absolute zero — it's generally interpreted as hotter than "infinitely hot," and how to interpret it is still debated.
- Behaviour of matter seen only at extreme cold. Phenomena that appear only near absolute zero — such as superconductivity and Bose-Einstein condensation, where atoms behave as a single wave — still have mechanisms that aren't fully worked out.
In other words, this article too reflects "what's understood so far." The location of the dead end is certain, but the world just this side of it is still being explored.
Curriculum links (by level)
| Level | Subject/Unit | Where in this article |
|---|---|---|
| MS | Science: states of matter and particle models | Shrinking on cooling, particle motion and temperature |
| HS | Physics/Chemistry: "gas laws" | Charles's law, Boyle-Charles' law, absolute temperature |
| HS+ | Physics: "kinetic theory of gases" | Average kinetic energy and absolute temperature |
| Univ | Thermodynamics/statistical mechanics | Thermodynamic temperature, third law of thermodynamics, zero-point motion |
| Research | Low-temperature physics/cold atoms | Extreme-cold records, negative absolute temperature |
| ― | Everyday connections | Tyres and balls sagging in winter, liquid nitrogen demonstrations |
- Wikipedia, "Absolute Zero" (絶対零度)
- Wikipedia, "Charles's Law" (シャルルの法則)
- Wikipedia, "Third Law of Thermodynamics" (熱力学第三法則)
- Yoshitaka Yamamoto, Netsugaku Shisō no Shiteki Tenkai (熱学思想の史的展開, "The Historical Development of the Concept of Heat"), Chikuma Shobo (ちくま学芸文庫)
- National Astronomical Observatory of Japan (国立天文台), ed., Rika Nenpyō (理科年表, "Chronological Scientific Tables"), Maruzen Publishing (丸善出版)
※This article is a general-audience science explainer. The figures given are approximations meant to aid understanding. Historical dates and measured values vary somewhat between sources. Extremely cold substances such as liquid nitrogen should only be handled with proper equipment and under expert supervision.