Dilution refrigerator

The dilution refrigerator emerged because cryogenics had reached a temperature wall—evaporating helium-3 could only cool to about 0.3 Kelvin—and Heinz London recognized that mixing helium isotopes could break through to millikelvin realms.

In 1948, Landau and Pomeranchuk proposed that extremely dilute helium-3 dissolved in superfluid helium-4 would behave like a gas of quasi-particles. Heinz London built on this in 1951, proposing that the mixing of helium-3 into helium-4 could provide continuous refrigeration below what simple evaporation could achieve.

The physics is counterintuitive: below 0.87 Kelvin, mixtures of helium-3 and helium-4 separate into two phases. The 'concentrated' phase floats on top (nearly pure helium-3); the 'dilute' phase settles below (helium-4 with about 6.6% helium-3). When helium-3 crosses from the concentrated to the dilute phase, it absorbs heat—analogous to evaporative cooling but occurring entirely in the liquid state.

The idea remained theoretical for over a decade. Helium-3 was scarce and expensive. Properties of helium-3/helium-4 mixtures had to be systematically studied: vapor pressure, mixing enthalpy, phase separation, and dilute solution behavior down to millikelvins.

In 1962, London, Clarke, and Mendoza published sketches of two possible refrigerator designs. The first experimental realization came in 1964 at Leiden University's Kamerlingh Onnes Laboratory—appropriately, where Heike Kamerlingh Onnes had first liquefied helium in 1908. P. Das, R. de Bruyn Ouboter, and K.W. Taconis constructed a prototype reaching approximately 220 millikelvin.

Rapid improvements followed. Hall, Ford, and Thompson reached 80 mK in 1966. Neganov, Borisov, and Liburg achieved 25 mK the same year. At the University of Illinois, improved heat exchangers reached 20 mK, and a single-cycle refrigerator precooled by a continuous system reached 4.4 mK—temperatures once considered unattainable.

The dilution refrigerator proved essential for studying quantum phenomena that only manifest at millikelvin temperatures. Today, every superconducting quantum computer operates inside a dilution refrigerator, cooled to temperatures colder than interstellar space. London's 1951 insight enabled technologies he could never have imagined.