Nancy Washton, who looks after 28 Pacific Northwest National Laboratory instruments that use superconducting magnets, was forced to shut down five NMR spectrometers last year when her supplier began rationing liquid helium, cutting deliveries from 2400 L to 962 L per month. Although all five NMRs are back in operation, the lab now pays $39/L, double the cost from two years ago. Washton says that at one point she was forced to pay $55/L to cool a magnet back to its superconducting state and keep it there.

William Halperin, a physicist at Northwestern University, says the university pays on average $30/L for helium today, in comparison with $7/L a decade ago. Christopher Nicholson, a chemist at Marian University in Indianapolis, Indiana, says he’s paid $45/L for the 9 L per month he needs to keep the campus’s sole NMR spectrometer cold. But since his supplier’s rigid delivery schedule doesn’t align with what it takes to keep the instrument operational, his department will pay even more by signing up for a service contract from the NMR manufacturer. “They do charge a little more for the helium, but not so much more that it’s not worth it,” he says. Many small liberal arts colleges with one or two instruments face similar helium predicaments, he adds.

The helium shortage that last year forced many scientific users to get by with less than half of their pre-2022 usage levels has eased a bit, according to a wide variety of users. (See “Helium is again in short supply,” Physics Today online, 4 April 2022.) Yet prices continue to soar above the already high levels of a year ago, they say.

Helium is critical to low-temperature physics, chemistry, and life-sciences experiments, yet laboratory usage accounts for just 10% of helium consumption worldwide, well below medical MRI and semiconductor manufacturing.


Helium is critical to low-temperature physics, chemistry, and life-sciences experiments, yet laboratory usage accounts for just 10% of helium consumption worldwide, well below medical MRI and semiconductor manufacturing.


Close modal

The situation is no better outside the US. Donald Thomas, a chemist at the University of New South Wales in Sydney, Australia, says he’s heard that researchers in Western Australia and New Zealand are paying suppliers up to Aus$100/L ($68/L), “if they even bother to promise to deliver.” He says other Australian universities have had NMR magnets come close to warming up and quenching their superconductivity. “I know of three magnets that were in trouble and at least as many where conversations were had about shutting down magnets.”

At the University of Crete, chemist Apostolos Spyros says he paid €44/L ($49/L) for his most recent shipment of liquid helium, up from the €24/L charged last year. That doesn’t include a 23% value-added tax. “We’ve gone from something like €7000 to €8000 per year to €17 000 or €18 000. It’s very difficult for the department to get that money,” he says. The price is only a couple euros lower in mainland Greece, he notes.

A recent survey conducted by the Canadian Helium Users Group, an organization of NMR spectroscopists, found that 72% of facilities had difficulty procuring liquid helium within the last nine months. It also found many labs have been subjected to unscheduled price increases, ranging from 25% to as much as 400%.

Still, there are always exceptions to the rule with helium: Gregory Wylie, the NMR facility manager at Texas A&M University, says he pays $19/L, up from $16/L two years ago. Like other institutions, the university has had to deal with helium rationing by suppliers, Wylie says, but he’s never had a problem obtaining enough to get by. “We are a big user, so we have pretty good contracts.”

The immediate cause of last year’s helium supply pinch, from which the market has yet to fully recover, was the five-month-long shutdown in early 2022 of the Cliffside crude helium enrichment plant in Amarillo, Texas. That followed an extended outage in 2021. Cliffside feeds helium from the US helium reserve and other privately owned helium sources into the 684 km pipeline that has long been the source for about half of US helium supply. The pipeline is tapped at various points along its length by four privately owned helium refineries. Cliffside has operated without further interruptions since resuming operation in June 2022 under new private-sector management.

A cryoplant newly commissioned at SLAC will feed the superconducting RF cavities of the Linac Coherent Light Source-II with helium cooled to 2 K. The closed system circulates 4 tons of helium between the linac and the cryoplant, with very low losses.


A cryoplant newly commissioned at SLAC will feed the superconducting RF cavities of the Linac Coherent Light Source-II with helium cooled to 2 K. The closed system circulates 4 tons of helium between the linac and the cryoplant, with very low losses.


Close modal

The General Services Administration is now scheduled to auction off both the pipeline and associated assets, plus the 51 million cubic meters (51 billion L) of helium left in the reserve, in mid-November. The sale, first ordered by Congress in 1996, has been delayed repeatedly, most recently last year.

The American Physical Society has petitioned lawmakers in recent years, to no avail, to require that buyers of reserve assets pledge to continue fulfilling scientists’ ongoing needs on a preferential basis. At press time, those efforts were continuing with staff on Capitol Hill.

Despite helium’s indispensability in low-temperature physics, chemistry, and life sciences, research accounts for only 10% of overall helium consumption (see the chart on page 18).

No matter who buys the reserve, the flow of helium is unlikely to change immediately, says helium market consultant Phil Kornbluth. But the flow will naturally diminish each year as the remaining helium, and the corresponding pressure in the formation, continually decline. He estimates that it will take 10 years or more to empty the reserve.

In the short term, a monthlong maintenance shutdown at Exxon Mobil’s natural gas processing plant in Shute Creek, Wyoming, will exacerbate the shortage for a couple of months, says Kornbluth. After that, the market should return to its previous condition of relatively mild shortage. Shute Creek is the largest US helium source and accounts for about 20% of global supply. Helium occurs in varying proportions in natural gas. It is separated cryogenically when its fraction is economically viable, generally over 0.3%.

The wild card for the near-term global supply picture is the status of Gazprom’s Amur natural gas processing complex in eastern Russia. The complex hasn’t operated since an October 2021 fire and explosion, a month after its commissioning. Amur could add 21–42 million cubic meters per year of helium to the global supply. But that depends on whether Gazprom can operate the plant and can get the helium to the international market as the war in Ukraine continues. Gazprom was planning to begin commissioning Amur last month, Kornbluth says, but the company had also promised to open last year. In any case, there is no scenario in which Amur will reach full production immediately, he adds.

“If and when Amur puts a substantial amount of gas into the market, the shortage should end,” says Kornbluth. To date, Western nations have not imposed sanctions on Russian exports of helium. Should they do so, helium from Amur could be shipped to China, India, or other countries that haven’t imposed sanctions, he says.

A complicating factor for Amur, however, is a US export restriction on domestically manufactured containers that are used to ship large quantities of liquid helium. The US manufacturer Gardner Cryogenics produces the bulk of the global supply of those superinsulated cylinders, and there are limited spares available around the world that could transport helium from a new source.

The opening of another liquefied natural gas plant in Qatar in 2027 could add as much as 42 million cubic meters to the world’s helium supply, Kornbluth says. Depending on what comes out of Russia, helium could be in oversupply by the end of the decade. In that case, he adds, prices are likely to moderate.

Development of new helium sources around the globe has picked up in response to rising prices. One major new source could be Blue Spruce Minerals, a gas-processing plant that’s being planned to commence operations on land adjacent to Exxon Mobil’s Wyoming plant by 2028. Robert Ferguson, Blue Spruce’s managing partner, says the hope is to produce 22 million cubic meters of helium annually, about half the output of Exxon Mobil’s plant.

The prolonged supply and price squeeze has driven more large consumers of helium—and increasingly many smaller ones—to acquire recovery and liquefaction systems. The efficiencies of those systems vary from 75% to 95%, according to users.

Physicists require helium for conducting experiments at low temperatures and for cooling quantum computers. They also need it for some scientific instruments, including some superconducting magnets, superconducting quantum interference devices, and accelerators—the last of which uses copious quantities. The newly commissioned helium recovery system at SLAC’s Linac Coherent Light Source-II is among the largest cryoplants in the US. (See the photo on page 19.) The closed-loop system will circulate 4 tons of helium (equivalent to 32 000 L liquid at 4.5 K) to cool the LCLS’s 23 superconducting RF cavities to 2 K, says Eric Fauve, SLAC’s cryogenic division director. He notes that the Large Hadron Collider at CERN circulates a helium inventory of 96 tons (768 000 L liquid equivalent). The circulating helium in both systems is part liquid and part gaseous.

With the purchase of an Aus$5 million helium liquefier from Linde five years ago, the University of New South Wales now can provide much of the 20 000 L used by its quantum computing group and the 2000 L needed for its 11 NMRs annually. Because nearly all the helium from the NMRs is recovered and liquefied, Thomas says his cost is an “embarrassingly low”Aus$5/L. Before it began liquefying helium, the university had been capturing helium and selling it to the party-balloon business, he notes.

McMaster University’s Brockhouse Institute for Materials Research has been recycling liquid helium since the 1960s, when physicists were looking at Fermi surfaces of metals and helium cost just Can$3–Can$4 ($3.20-$4), says Paul Dube, manager of research facilities. He cites a recovery rate as high as 95%.

The Brockhouse recapture system allows researchers who may need just 10 L of helium at a time for their experiment to get it without having to buy a 60 L or 100 L dewar. “Purchasing 100 liters and using only 20% of it never made fiscal sense,” Dube says.

Halperin has been managing a recovery and liquefaction facility at Northwestern since 1983. Today it recovers about 70% of the helium used by physicists, chemists, and materials scientists on campus. He keeps an inventory of about 5000 L. Installing a similar system today would cost a medium-to-large campus $3–$5 million, he says.

Smaller-scale recovery and liquefaction systems can capture the boil-off from three or more NMRs. Texas A&M’s chemistry department installed a $240 000 helium recovery and liquefaction system three years ago. It recycles the helium boiled off from nine instruments. The installation lowered the department’s annual helium expenditures from $40 000 to $8000. That includes the purchase of amounts needed to maintain three NMRs that are too distant to be connected to the system.

The payback for the chemistry department was immediate, says Wylie. A National Institutes of Health grant paid for most of the system cost, while the university administration picked up the rest. Wylie says the combined costs for equipment, staffing, and maintenance will likely be too large for many small universities that have three or fewer instruments.

Martha Morton, director of research instrumentation at the University of Nebraska–Lincoln, operates a liquefaction system that recovers 1200 L per year, about 80% of what’s needed for her four NMR spectrometers, a Fourier transform ion cyclotron resonance mass spectrometer, and a scanning tunneling microscope. The system was installed in 2021 with a $250 000 grant from NIH. The university kicked in $50 000 for the 400 m of pipes that collect gas from four floors in the laboratory building. “It’s all about plumbing,” Morton says, noting that leaks are often hard to find because much of the piping is hidden in the walls.

NIH has discontinued its support for helium-recovery systems, but NSF received new authority under last year’s CHIPS and Science Act for grants in support of them. As Physics Today went to press, NSF was reviewing proposals it solicited early this year offering anywhere from $100 000 to $4 million for new helium-recovery systems. An NSF spokesperson declined to comment on the number of grants the agency expects to award, saying the grants will be announced by this fall.

For the single-NMR institution, manufacturers offer a recovery system as an option for a new machine. But the feature costs around €100 000, says Spyros, and it would reduce, but not eliminate, the need for helium replenishment. He estimates his payoff period for such a system would be 15–20 years.

Liquefaction systems do have drawbacks: They consume lots of power, need chilled water lines and other supporting infrastructure, and require servicing. “NMR spectroscopists are struggling to keep up” with the additional costs, Morton says. “They’ve been asked to install the systems. They were expecting them to be more turnkey, and they’re not.” She says that keeping the liquefier system functioning properly takes 4–10 hours each week—time that could have been spent on research.

Cryogen-free systems can eliminate the need for helium altogether. Also known as dry fridges, they are well-suited for certain applications, such as in quantum information science, where the vibration they create isn’t an issue. But some other low-temperature applications, such as scanning tunneling microscopy, are vibration sensitive. “Maybe commercial suppliers will be able to better mitigate vibration, but it’s not available now,” says Halperin. “If you work at the nanoscale, dry fridges may not be usable.”

Cryogen-free systems are also expensive. A “bare-bones” system will cost $500 000, Halperin says. Still, manufacturers Blue Force and Oxford Instruments are now building them at the rate of one a day, he says.

, “
Helium is again in short supply
Phys. Today