Liquid helium is made by cooling purified helium gas to about −269 °C (4.2 K), just a few degrees above absolute zero. Reaching that temperature requires repeated cycles of compression, heat removal, and controlled expansion, typically using variations of a thermodynamic process called the Claude cycle. The raw helium itself almost always starts as a trace component of natural gas, separated out and purified before it ever enters a liquefaction plant. Getting from underground gas well to a dewar of clear, ultra-cold liquid involves an unusual number of steps, each of which has to work at extremes that push engineering to its limits.
Where the Helium Comes From
Helium is the second most abundant element in the universe, but on Earth it is surprisingly scarce. Nearly all commercial helium is found trapped in underground natural gas deposits, where it accumulates over geological timescales from the radioactive decay of uranium and thorium in rock. When those elements shed alpha particles, the particles pick up electrons and become helium-4 atoms. Over millions of years the gas migrates upward and collects beneath the same cap-rock formations that trap natural gas.
The concentrations are tiny. Among roughly 1,500 samples analyzed by the U.S. Bureau of Mines, only about one percent showed helium concentrations above one percent, with most falling between 0.5 and 2 percent of the total gas mixture. Over 90 percent of the world’s known helium reserves sit in just four countries: roughly 34 percent in Russia, 25 percent in Qatar, 18 percent in the United States, and 17 percent in Algeria, according to U.S. Geological Survey data from 2024. Global estimated reserves stand at about 39.9 billion cubic meters.1ScienceDirect (Natural Gas Industry B). A review of helium resources and development That geographic concentration is one reason helium supply has been so volatile in recent decades, with periodic shortages rippling through hospitals, research labs, and semiconductor fabs worldwide.
Extracting Helium from Natural Gas
Before helium can be liquefied, it has to be pulled out of the natural gas stream and purified to very high levels, typically 99.995 percent or better. Several technologies handle this separation, and modern plants often combine more than one approach. The main methods are cryogenic distillation, membrane separation, and adsorption-based techniques.2Industrial & Engineering Chemistry Research. Review of Technologies for Helium Extraction from Natural Gas
Cryogenic distillation is the workhorse for large-scale production. It exploits the fact that helium has the lowest boiling point of any element. When natural gas is chilled progressively, methane, ethane, nitrogen, and other components condense or freeze out long before helium does. What remains as a gas after nearly everything else has been removed is a crude helium stream, typically around 50 to 70 percent pure, still mixed with some nitrogen and traces of hydrogen and neon.
That crude stream then goes through further purification. Pressure swing adsorption, in which gas passes through beds of material that selectively grabs nitrogen and other impurities, can push purity well above 99 percent. A final cleanup stage, often involving a catalytic oxidizer to convert trace hydrogen to water and a cold trap to remove the water, brings the helium up to the ultra-high purity needed for liquefaction. Any remaining contaminant would freeze solid inside a liquefier’s heat exchangers and eventually block them, so purity is not optional.
How the Liquefaction Process Works
Once you have a stream of very pure helium gas at roughly room temperature, the challenge is removing enough thermal energy to bring it down to 4.2 K. No single cooling step can get there. Instead, the gas passes through a cascade of stages that progressively strip heat away.
The basic pattern looks like this: compress the helium, remove the heat of compression, then let the gas expand so it cools further. Repeat. Each cycle shaves off more thermal energy. In a large plant, pre-cooling with liquid nitrogen (which boils at 77 K) handles the first big temperature drop, getting the helium from ambient temperature down to roughly 80 K in one efficient step. From there, the helium enters a series of expansion stages.
Most industrial liquefiers use some version of the Claude cycle, named after the French engineer Georges Claude. The key feature of a Claude-type system is that part of the compressed helium is diverted through a turbine expander, where it does mechanical work and cools dramatically in the process. That cold gas is then used to pre-cool the remaining helium stream through a heat exchanger. The process is repeated at successively lower temperatures, with two or more turbine expanders operating at different temperature levels.
At the final stage, the coldest helium passes through a throttling valve, a constriction where it expands rapidly without doing external work. This is a Joule-Thomson expansion, and for helium it only produces net cooling below about 40 K, which is why all the earlier turbine stages are necessary to get the gas cold enough for the final throttle to work. A fraction of the gas liquefies on the other side of the valve; the rest circulates back through the heat exchangers to help cool the incoming stream. Even in a well-designed plant, only a portion of the helium liquefies on each pass; the unliquefied gas recycles continuously.
The Machinery That Makes It Happen
A helium liquefier is not one machine but an integrated system of compressors, heat exchangers, expanders, and control valves, all connected by piping that has to handle pressures up to about 20 bar and temperatures spanning from 300 K down to 4 K.
Compression is where the process begins, and it consumes most of the electrical power. Oil-flooded rotary screw compressors are the standard choice in helium cryogenics. These machines inject oil into the compression chamber, which seals internal clearances, absorbs heat, and lubricates the rotors. A study modeling the performance of a helium twin rotary screw compressor found that isothermal efficiency and discharge temperature can vary by up to 25 percent depending on the oil-to-gas mass ratio, highlighting how sensitive the process is to operating conditions.3Volume 6: Energy. A One-Dimensional Model of an Oil-Injected Helium Screw Compressor for Cryogenic Refrigeration Systems After compression, the helium passes through coalescing filters and charcoal adsorbers to strip out every trace of oil before entering the cold box; even a few parts per million of oil would freeze and foul the heat exchangers.
The heat exchangers themselves are typically brazed aluminum plate-fin units, chosen for their enormous surface area packed into a compact volume. At the coldest stages, copper or stainless steel replaces aluminum because aluminum becomes too brittle. Turbine expanders in modern plants are gas-bearing turboexpanders, spinning at tens of thousands of revolutions per minute on a thin film of process gas rather than conventional bearings. They are remarkably efficient at extracting energy from the helium stream and converting it to cooling.
All of this sits inside a vacuum-insulated enclosure, the “cold box,” wrapped in dozens of layers of reflective foil separated by spacer material. That multi-layer insulation minimizes radiative heat leak from the warm surroundings. Even a small heat leak at 4 K can burn through a significant fraction of the plant’s cooling capacity, so thermal isolation is taken extremely seriously.
Small-Scale Liquefiers for Research Labs
Not all liquid helium comes from large industrial plants. Over the past couple of decades, small cryocooler-based liquefiers have become increasingly common in university labs and research facilities. These machines allow a single lab or a small cluster of instruments to produce their own liquid helium on-site, reducing dependence on bulk deliveries that may be unreliable or expensive.
One approach uses pulse tube cryocoolers, which have no moving parts at the cold end and therefore generate very little vibration. A prototype pulse tube helium liquefier demonstrated continuous production of 4.8 liters per day while consuming 4.6 kilowatts of electrical power.4ScienceDirect (Elsevier). Helium liquefaction with a 4 K pulse tube cryocooler That is a tiny output compared to a commercial plant, which might produce thousands of liters per day, but it is enough to keep a single scanning tunneling microscope or a small magnet system running without outside supply.
The trade-off is efficiency. Small-scale liquefiers use far more energy per liter of liquid helium produced than large plants do, because the thermodynamic overhead of maintaining a 4 K cold stage does not scale down proportionally. But for labs spending tens of thousands of dollars a year on delivered liquid helium, or for researchers working in locations where supply is unreliable, the economics can still work out favorably.
Storing Liquid Helium and Recovering Boil-Off
Liquid helium is among the most difficult cryogenic fluids to store. Its extremely low temperature and tiny latent heat of vaporization mean that even modest heat leaks cause rapid boil-off. Storage vessels, called dewars, use double-walled vacuum-insulated construction with multi-layer insulation in the vacuum space, and larger dewars often incorporate a nitrogen-temperature radiation shield to intercept heat before it reaches the helium.
Despite all this, some boil-off is inevitable. In a typical research dewar, losses of one to two percent of the stored volume per day are common. For a hospital MRI system containing over a thousand liters of liquid helium, that adds up to significant ongoing cost. The helium that boils off is still perfectly pure gas, but if it escapes to the atmosphere it is lost for good, since helium is too light to be recaptured from air at any practical scale.
Recovery systems capture the boil-off gas and either store it in high-pressure cylinders for later reliquefaction or feed it directly back into a small on-site liquefier. One system designed for vibration-sensitive instruments demonstrated a recovery rate above 94 percent, capturing gas from both routine boil-off and the refilling process, which is when losses are typically highest.5PubMed Central. An Economical and Efficient Helium Recovery System for Vibration-Sensitive Applications Recovery has become a priority at many institutions because the cost of helium has roughly tripled over the past fifteen years, and supply shortages in 2012-2013, 2019, and again in 2022 left some labs rationing or even shutting down experiments.
Why Liquid Helium Matters
The effort to make liquid helium is not academic curiosity. Several technologies that most people interact with, directly or indirectly, depend on it. The most widespread is magnetic resonance imaging. An MRI scanner’s superconducting magnet must be bathed in liquid helium to keep the niobium-titanium wire cold enough to carry current without resistance. Without that helium bath, the magnet quenches, meaning it abruptly loses superconductivity, dumps its stored energy as heat, and becomes a very expensive paperweight until it can be recooled and recharged.
Particle accelerators are another major consumer. The Large Hadron Collider at CERN, for instance, is the single largest cryogenic installation on Earth, requiring tens of thousands of liters of liquid helium to cool its superconducting magnets to 1.9 K, even colder than ordinary liquid helium. Semiconductor manufacturing uses helium (mostly gaseous, but the supply chain depends on liquid-phase logistics) as a carrier and cooling gas. Quantum computing research groups, meanwhile, use dilution refrigerators that cool experimental chips to millikelvins, and those refrigerators rely on liquid helium as a first-stage coolant.
Rocket engine testing, fiber-optic manufacturing, deep-sea breathing mixtures, leak detection, and nuclear fusion research round out a list that is surprisingly broad for a substance most people never see. The common thread is that helium’s unique combination of properties, an inert gas with the lowest boiling point of any element, makes it irreplaceable in each application. No cheap substitute exists for cooling a superconducting magnet to 4 K.
The Race to Reach 4 Kelvin
Helium was the last of the so-called permanent gases to be liquefied, and the story of how it happened illustrates just how extreme the challenge was. By the late 1800s, physicists had managed to liquefy oxygen, nitrogen, and even hydrogen. Helium held out. Its critical temperature is only 5.2 K, meaning no amount of pressure alone can liquefy it above that temperature. You have to get the gas below 5.2 K first, and in 1900 nobody had a reliable way to do that.
Heike Kamerlingh Onnes, a Dutch physicist at the University of Leiden, spent years building up a cascaded cooling infrastructure: liquid air to pre-cool hydrogen, liquid hydrogen to pre-cool helium, and then a Joule-Thomson expansion to push the helium below its critical point. On July 10, 1908, he succeeded in producing the first liquid helium, reaching temperatures the Earth had never seen.6PubMed. Cryogenics at the end of the 19th and the first half of the 20th century (1880-1940) The accomplishment earned him the Nobel Prize in Physics in 1913, though the prize was officially for his broader investigations into matter at low temperatures.
Having liquid helium in hand almost immediately led to a cascade of discoveries. In 1911, Kamerlingh Onnes found that the electrical resistance of mercury dropped to zero below about 4.2 K, the first observation of superconductivity.7The Harvest of a Century. Kamerlingh Onnes – Liquid Helium and Superconductivity That discovery would eventually make liquid helium commercially important, because superconducting magnets became the backbone of MRI machines and particle accelerators. For decades after 1908, Leiden was the only laboratory in the world with the ability to produce liquid helium. It was not until the 1920s and 1930s that other labs, notably in Toronto and Oxford, managed to replicate the feat, and industrial-scale helium liquefaction did not emerge until after World War II.
Helium-3 and the Deeper Cold
Everything discussed so far concerns helium-4, the common isotope that makes up about 99.9999 percent of terrestrial helium. But a vanishingly rare lighter isotope, helium-3, plays a critical role in reaching temperatures far below what helium-4 alone can achieve.
Helium-3 liquefies at 3.2 K, slightly lower than helium-4’s 4.2 K. More importantly, mixtures of liquid helium-3 and helium-4 exhibit a peculiar phase separation below about 0.87 K: the mixture spontaneously splits into a helium-3-rich layer floating on a helium-4-rich layer. Dilution refrigerators exploit this separation. By continuously pumping helium-3 from the dilute (helium-4-rich) phase and cycling it back into the concentrated phase, the device extracts heat and can reach sustained temperatures in the range of 10 millikelvins or lower. This is the technology behind essentially all quantum computing experiments that use superconducting qubits.
Helium-3 is phenomenally expensive compared to helium-4 because it is not produced by the same radioactive decay process. Most commercial helium-3 comes from the radioactive decay of tritium in nuclear weapons stockpiles, and supply is measured in tens of thousands of liters per year worldwide rather than millions. Making liquid helium-3 follows the same general cryogenic principles as helium-4 liquefaction, but it requires even more careful handling and typically uses a helium-4-based cryocooler as the first cooling stage.
Energy Cost and the Efficiency Problem
Producing liquid helium is staggeringly energy-intensive relative to the amount of liquid you get. The theoretical minimum work required to liquefy helium starting from room temperature is already large because of the enormous temperature ratio involved, but real plants operate well above that minimum. Turbine expanders, heat exchangers, and compressors all have irreversibilities. A modern large-scale liquefier might consume on the order of 500 to 700 watts of electrical power for every watt of cooling delivered at 4.2 K. That ratio, sometimes called the Carnot penalty, is a reflection of the second law of thermodynamics: the colder the target, the more work you have to do per unit of heat removed.
At the small-cryocooler end, the numbers are even more unfavorable. The pulse tube liquefier that produces 4.8 liters per day while drawing 4.6 kilowatts illustrates this: the cooling power at 4 K is on the order of one watt, but the wall-plug power is thousands of times larger.4ScienceDirect (Elsevier). Helium liquefaction with a 4 K pulse tube cryocooler Industrial plants do better by scaling up, using turboexpander efficiency gains and nitrogen pre-cooling, but they still consume enormous amounts of electricity. This is a major reason why helium recovery is so important. Every liter of helium gas that escapes to the atmosphere represents not just the cost of the helium itself but the substantial energy investment that went into cooling it.
Researchers continue to explore ways to improve efficiency, including magnetic refrigeration at cryogenic temperatures and advanced expander designs. But the fundamental physics sets a hard floor: getting anything down to 4 K requires moving a lot of entropy uphill, and there is no shortcut around that.