On this page
By 1960, physicists were finding new particles by studying photographs of tiny bubbles. The instrument behind them was the bubble chamber, and the person who worked out that it could exist was Donald Glaser. This post follows the reasoning in his Nobel Lecture, explains the physics of a superheated liquid, and includes a short probability example from his own detective work.
Who was Donald Glaser?
The Nobel Prize facts page records that Glaser was born on 21 September 1926 in Cleveland, Ohio, and died on 28 February 2013 in Berkeley, California. He received the 1960 Nobel Prize in Physics "for the invention of the bubble chamber". His Nobel biography says he finished his PhD at Caltech in 1949, moved to the University of Michigan that year, began developing the bubble chamber in 1952, and moved to Berkeley in 1959. In 1962 he left physics for molecular biology, and later worked on human vision.
What problem was he trying to solve?
In his Nobel Lecture, Glaser explains that in 1950 the new "strange particles" were being found in cosmic rays, mostly in cloud chamber photographs. A cloud chamber is filled with gas, and gas has low density, so a particle rarely collides with anything inside it. Production events "almost never occurred in the gas itself". High-pressure cloud chambers existed, but needed 15 to 30 minutes between expansions. The accelerators coming in the early 1950s would deliver pulses every few seconds.
So he needed a detector that was dense, large, and able to reset in seconds. Our post on C. T. R. Wilson and the cloud chamber explains the gas detector he was trying to improve on.
Why does a superheated liquid make tracks?
Glaser listed every unstable state he could think of that a single particle might trigger. He chose a superheated liquid: a liquid heated above its boiling point but kept from boiling, then held at low pressure for a moment. It is unstable. Any small disturbance can start a bubble, which then grows fast.
A fast charged particle leaves a trail of deposited energy. If enough energy lands in a tiny volume, it forms a vapour bubble big enough to keep growing. Glaser's estimate showed this needs low surface tension and high vapour pressure, which means running the liquid "about one-half to two-thirds of the way from the boiling point to the critical point".
How did a 1924 chemistry paper help him?
Glaser chose diethyl ether, which boils at 34.6 °C at one atmosphere. One of his calculations predicted that radiation would trigger boiling at about 140 °C. That seemed absurdly hot, so he searched the chemistry literature and found a 1924 paper by Kenrick, Gilbert and Wismer. They had kept ether superheated for hours at 130 °C, but at 140 °C it erupted at erratic times, and they listed 30 waiting times to show how "capricious" it was.
Glaser saw that the waiting times fitted a Poisson distribution with an average of about 60 seconds. He then estimated how often cosmic rays and background radiation would hit their small sample. The answer was also about once every 60 seconds.
Here is what the Poisson fit means, as a worked example. For random events with an average waiting time τ, the chance of waiting longer than t is e^(−t/τ).
Take τ = 60 s.
Chance of waiting more than 60 s: e^(−1) ≈ 0.37.
Chance of waiting more than 120 s: e^(−2) ≈ 0.14.
Chance of an eruption within 10 s: 1 − e^(−1/6) ≈ 0.15.
Common mistake: assuming that because 60 seconds is the average, an eruption is "due" after a long wait. Random events have no memory. Our post on radioactive decay and half-life uses the same exponential.
How did he test it?
He sealed ether in a thick glass tube with one bulb held at 140 °C, then cooled another part of the tube so the pressure fell to about one atmosphere. When he brought a cobalt-60 gamma source near, the hot ether burst into violent boiling, "instantly as far as the eye can tell".
He admits one detour first. He took bottles of beer, ginger ale and soda water into his lab, warmed them, and opened them with and without a radioactive source nearby. They foamed the same way. Water, he notes, is the wrong liquid: its surface tension and critical pressure are too high.
Next he built small glass chambers holding a few cubic centimetres of ether and filmed them at 3,000 pictures per second. Bubbles grew to more than a millimetre in 300 microseconds, along clean tracks. His first report, Some Effects of Ionizing Radiation on the Formation of Bubbles in Liquids, appeared in Physical Review in 1952.
What happened to the idea?
Liquid hydrogen was "the physicists' dream" because collisions with single protons are simple to interpret. Glaser estimated it would work near 27 K. By the time of his lecture, the largest chamber was Luis Alvarez's 180-cm, 500-litre hydrogen chamber, and Glaser noted that no liquid tested seriously had failed to work.
What can a student take from how Glaser worked?
List every option before choosing. He wrote down all the unstable states he could imagine.
Search old literature. A 1924 chemistry paper held the key data.
Test the cheap idea first. The beer test was quick, and it ruled out a shortcut.
If exponential waiting times feel slippery, the Fermi Problem Library has a step-by-step solver that explains the reasoning rather than printing an answer.
Next step: with τ = 60 s, work out the waiting time t for which there is an even chance of an eruption, then check it against ln 2.
