Mining And Refining: Helium

With a seemingly endless list of shortages of basic items trotted across newsfeeds on a daily basis, you’d be pardoned for not noticing any one shortage in particular. But in among the shortages of everything from eggs to fertilizers to sriracha sauce has been a growing realization that we may actually be running out of something so fundamental that it could have repercussions that will be felt across all aspects of our technological society: helium.

The degree to which helium is central to almost every aspect of daily life is hard to overstate. Helium’s unique properties, like the fact that it remains liquid at just a few degrees above absolute zero, contribute to its use in countless industrial processes. From leak detection and welding to silicon wafer production and cooling the superconducting magnets that make magnetic resonance imaging possible, helium has become entrenched in technology in a way that belies its relative scarcity.

But where does helium come from? As we’ll see, the second lightest element on the periodic table is not easy to come by, and considerable effort goes into extracting and purifying it enough for industrial use. While great strides are being made toward improved methods of extraction and the discovery of new deposits, for all practical purposes helium is a non-renewable resource for which there are no substitutes. So it pays to know a thing or two about how we get our hands on it.

A Product of Decay

Despite the fact that it’s the second most abundant element in the visible universe, helium is surprisingly rare on Earth. While it was first discovered in spectrographs from the sun and other stars in the 1860s, getting enough helium to study and make the determination that it’s an element would wait another 30 years, when a gas with the same spectral signature was released by dissolving a sample of uranium ore in acid.

Uranium decay series. When U-238 decays into Th-234 (upper left), it liberates an alpha particle, which is a helium nucleus. The particle quickly picks up two electrons, creating a new atom of helium. Source: Tosaka, CC BY 3.0, via Wikimedia Commons

The discovery of helium on Earth came at an opportune time in the history of chemistry. The late 1800s and early 1900s saw the focus of chemistry expand from reactions involving atoms as a whole to the subatomic realm, on the level of the electrons, protons, and neutrons that make up atoms. Radioactivity had just begun to be explored, and the existence of alpha, beta, and gamma rays was already known at the time helium was first isolated. And so when Rutherford and Boyd discovered that alpha rays are actually particles consisting of two protons and two neutrons, which is identical to the nucleus of a helium atom, it immediately suggested a mechanism for how helium managed to become trapped within uranium ore.

Like all heavy radioactive elements, uranium decays along a specific series of elements. The Uranium Series starts with the isotope 238U, the naturally occurring and relatively abundant isotope of uranium. 238U has a half-life of about 4 billion years, and when it decays, it does so by releasing an alpha particle. The loss of a pair of protons and a pair of neutrons turns the 238U into 234Th, or thorium-234. The liberated alpha particle, which is really a helium nucleus, easily soaks up two electrons when it is absorbed by pretty much any matter it runs into, creating an atom of helium.

This neatly explains why helium was inside that sample of uranium ore — over time, uranium decay had released alpha particles that were absorbed by the rock, gaining the electrons needed to become helium atoms. Helium accumulated over time, collecting in the rock’s pores, only to be liberated when the minerals in the rock were finally dissolved. And this same process, albeit on a geological scale, is the key to industrial helium production.

A Gas in a Gas

Unlike most industrial gasses, helium is not present in the atmosphere in any significant concentration. Any helium that isn’t somehow sequestered after it is produced will find its way into the atmosphere and quickly be lost, rapidly ascending to the upper atmosphere and eventually out into space. So isolating helium from air as we do for oxygen, nitrogen, argon, and other gasses is not practical. Rather, we need to look under our feet for significant reservoirs of helium.

Luckily, the same geological conditions that tend to trap natural gas in underground reservoirs also tend to trap helium, and so natural gas wells are the biggest source of helium. Historically, the United States has been the main supplier of helium to the world’s markets, with most coming from natural gas wells in Oklahoma, Kansas, and Texas. Here the gas coming out of the ground is up to 7% helium, which is more than enough for profitable extraction.

Natural gas is a mixture of methane, nitrogen, carbon dioxide, and higher gaseous alkanes like ethane and propane. Where sufficient helium is mixed in — anything above 0.4% is considered profitable — extraction and purification of the helium are performed by fractional distillation. Helium has the lowest boiling point of any element, meaning that every other gas can be isolated by dropping the temperature and controlling the pressure.

The first step in helium production is to scrub any CO2 and hydrogen sulfide (H2S) from the natural gas. This is done in an amine treater, where the chemical monoethanolamine (MEA) is sprayed into the gas stream inside a reaction vessel. The MEA ionizes the acidic compounds and makes them soluble in water, allowing them to be scrubbed from the natural gas. The scrubbed gas is pretreated further by passing it over a molecular sieve, such as zeolite, and a bed of activated carbon, to remove water vapor and any of the heavier hydrocarbons.

Schematic of a generic industrial distillation process. Source: H Padlekas, CC-BY-3.0

What’s left after these pretreatment steps is mainly methane and nitrogen, but also some neon and helium. The gas is cooled by passing it through a heat exchanger, then through an expansion valve into a baffled fractionation column. The sudden drop in pressure lowers the temperature of the gas enough that the methane, which boils at -161.5°C, condenses into a liquid and drains to the bottom of the column.

The remaining gas, now mostly nitrogen and helium, is passed through a condenser that cools the stream even further. When the temperature of the mixture falls below -195.8°C, the nitrogen condenses out as a liquid. Along with the liquid methane, the liquid nitrogen is piped to the heat exchangers that were initially used to cool the incoming pretreated process gas. The now-gaseous nitrogen and methane, valuable products both, are piped to storage tanks.

About half of the remaining process gas is helium, with the rest being a mixture of contaminating methane and nitrogen, along with a little bit of hydrogen and neon. This mix is called cold crude helium, and must now undergo further purification to get to the purity level required for industrial use. Purification begins with another heat exchanger that drops the crude helium mix below the boiling point of nitrogen, to condense out the remaining nitrogen and methane contaminants. This step takes the crude helium to about 90% purity.

Final Purification

To get rid of the hydrogen, oxygen is introduced and the mix is heated in the presence of a catalyst. The hydrogen and oxygen form water, which can be separated out of the process gas stream before it heads to final purification by pressure swing adsorption, or PSA. Pressure-swing adsorption is the same process used in oxygen concentrators, including many of the DIY versions we’ve seen as a response to COVID-19. PSA uses the ability of materials known as molecular sieves to selectively adsorb a gas. In helium purification, the 90% pure gas is pumped into a pressure vessel containing a molecular sieve, usually Zeolite. The contaminating gasses are preferentially adsorbed into the Zeolite, leaving the output stream nearly pure helium. When the first column is saturated with contaminants, flow is switched to a second column that had been previously regenerated by backflushing it with pure helium. The gas flow switches back and forth between the two columns, one purifying the helium while the other is regenerated. The result is gaseous Grade-A helium at 99.995% purity.

The process described here is by no means the only way to extract helium from natural gas, but it does represent the most common way of producing the gas, mainly because most of the pretreatment and initial purification steps are already used to process natural gas for fuel and as a feedstock for the chemical industry. Other methods include a completely PSA process, which can use natural gas with a mere 0.06% helium concentration, and membrane separation, which relies on the fact that helium can penetrate a semipermeable membrane much easier than the much larger methane and nitrogen molecules. Membrane separation technology can be much more energy-efficient than traditional fractional distillation, since it doesn’t require phase changes and the energy they require.

But Are We Running Out?

Knowing the abundance of uranium-238 in the Earth’s lithosphere along with its half-life, it’s possible to estimate the amount of helium produced by the radiogenic process. It turns out to be not a lot — only about 3,000 metric tons a year. And almost all of that escapes into the atmosphere and out into space. So in much the same way as the natural gas in which it is usually found, helium is effectively a non-renewable resource.

But does that mean we’re running out? Yes, like any other limited resource, eventually we’ll extract all there is to extract. But that doesn’t necessarily mean we’ve found all the helium there is to find. Exploration has led to new deposits in the United States, and massive helium finds in places like Algeria, which became the second-largest helium producer in the world in the early 2000s. Qatar also had a huge helium find in 2013, moving it up to second place worldwide. These finds, along with the recent discovery of natural gas wells in South Africa with up to 12% helium, promise to address some of the concerns about losing access to this irreplaceable gas.

But at the end of the day, these new finds only push back the clock and forestall the inevitable day when the helium finally runs out. We may catch a break if commercial-scale fusion ever becomes a thing, but that breakthrough has been “only twenty years away” for the last 80 years.

46 thoughts on “Mining And Refining: Helium

  1. Can everybody please stop parroting that “commercial scale fusion has been 20 years away (for some long time). It’s nothing more then a catchy nonsense phrase, and it has never been true.

    Take for example ITER. ITER was “officially initiated” in 1988 ( ) which is over 34 years ago, and ITER and considering that timescale It’s remarkably close to it’s original time scale even though quite big and unknown problems had to be solved to get here. ITER is also not designed (and never was intended to!) be a commercial reactor. It is not even a goal of ITER to generate electricity from the fusion, and it never has been.

    DEMO is supposed to be the first fusion plant to generate a significant amount of electricity, and it’s idea was conceived in the ’70-ies ( Plans are to build DEMO after the research with ITER has been done, and even DEMO is not even supposed to be a commercial fusion reactor. IT’s only goal is to be a demonstration to show that it can be done and get accurate figures about cost and viability.

    And after DEMO, comes PROTO. ( PROTO is supposed to be the prototype for a commercial fusion plant, and It’s not even expected to be built before 2050.

    Also, if you do a bit of research, then the amount of energy “freed” in fusion experiments has been growing for 40+ years in a somewhat predictable rate. So please stop repeating that silly quote from long ago.

    That said, the Helium article was also engithening / intersting to read.

    1. If Elon Musk was doing fusion power his company would already have built a couple of dozen test reactors and would be about to start on the first commercial plant. Speed and “rapid fire” iteration to quickly work out what doesn’t work has definitely never been a thing in fusion research and development.

      Other than SpaceX, the one other big project that used such a strategy was the Soviet development of the closed cycle rocket engine. Their people designed and engine, tested it, watched it fail, figured out what was the likely cause, modified it, and tried again, and again, and again… until they got a design that would work. Unfortunately for the USSR they didn’t have people able to independently invent the alloys and metallurgical processes their engines needed to be super reliable. Nor were their spies able to steal that information. So while a successful design, the Soviets could not make it *reliable* enough to make their N-1 moon rocket work.

      The rocket engines Soviet scientists and engineers saved from being scrapped will mostly end up as museum pieces, or scrapped. After a couple of high profile kabooms of ‘vintage’ Soviet engines that had been refurbished and x-rayed and tested every which way for flaws, the rocket industry concluded that they should just use the design and make new copies rather than rely on the same iffy technology that exploded 100% of the N-1 rockets.

          1. Indeed. I’m amazed by some of Musk’s companies’ achievements. But I don’t think he’s necessarily the world’s savior. I don’ see why some think it has to be one or the other.

            I do believe a well-funded individual like Musk or even a company is much more likely to solve the fusion problem well before a giant conglomeration of nations manages to do. Even one government brings in huge cost overruns and way too much bureaucracy. Multiply that by ~12 governments and it’s exponentially worse.

            Unfortunately you’ll never see a publicly traded company spend the resources on it because they won’t have end results to boost the share price by the next quarter. So they’ll just contract with government to do it slowly and squeeze out as much money from the process as possible.

          2. TimT:

            ‘Publicly traded companies can’t see past the next earnings statement’ is just as true as ‘Politicians can’t see past the next election.’

            Both are just gross oversimplifications of complicated situations. Both are true on the edge cases.

            Fusion is just a hard, very expensive, problem.
            Single genius builds fusion power plant is just innumerate hollywood bullshit story time.

      1. Right, if he’d done that “80 ways not to make a lightbulb” Edison thing on Fusion reactors, maybe we’d have a fusion reactor, and enough of one piece of the planet left to retain an atmosphere.

    2. That’s not a “silly quote from a long time ago.” That’s my experience having lived now almost three score of years, and having followed science and technology closely since about 1975 or so. My perception, my memory of the promises made for fusion power have always consistently been “it’ll be commercialized in 20 years”. The sources that formed that impression may have been questionable — I wasn’t reading physics papers when I was 11 — and there’s always the problem of the popular press distilling scientific papers and looking for headlines that the general public can digest. But that’s what the general public has been lead to believe for as long as I can remember.

      I think your observations about DEMO and PROTO kind of prove my point — here we are 40 years after I started following commercial fusion, and even those projects are still 20+ years away from paying off commercially. And I never said that progress hadn’t been made — I know there’s been a ton of progress, and some experiments have nearly reached the break-even point. But that doesn’t change the fact that the message that gets out to the public has consistently been “20 years away” for almost as long as I’ve been alive.

      All that said, fusion is probably still a long way from plugging the helium gap. And glad you enjoyed the article.

    3. Why are you going on about rhetorical sarcasm? Fusion as a viable energy source isn’t here and it’s not going to be here any time soon. Certainly won’t be here in time to save civilization from rising sea levels. That’s the entire point of the phrase which you missed. Rhetorical tropes need not be entirely accurate, only widely understood in context – a context that apparently flew over your head.

  2. The essential info I’m missing here is… How many expired smoke detectors do I have to collect for their alpha sources to give me enough helium to fill a Zeppelin in about a year?

  3. The helium shortage was initiated by the US government deciding to sell off a large amount of the strategic helium reserve at low prices. That made it unprofitable to separate helium from natural gas, especially from the sub 1% wells.

    As for why the shortage wasn’t resolved after the US stopped selling from the reserve ?????

    Why does the US have a strategic helium reserve? I assume it was for all those military airships we don’t have, and haven’t had for a long time.

    1. “Why does the US have a strategic helium reserve? I assume it was for all those military airships we don’t have, and haven’t had for a long time.”

      Well, it is a good thing to have a “strategic reserve”, it can now be used for MRIs instead of airships.

      1. Unfortunately, we don’t have much of a strategic helium reserve anymore. The helium from the reserve in Kansas (or was it Oklahoma?) was all sold off. I think it closed in 2019 or 2020.

    2. They only held it to 1950 for that, subsequently it was held for the above top secret Project Chipmunk psy op. The idea was to issue a helium canister to every NATO vehicle, land, sea and air, and when given the signal, the commander of each was ordered to take a huge haul on the helium canister, and make a radio broadcast something along the lines of “The time manipulation device appears to have been a success, we will proceed cautiously to observe the results” whereupon to really sell it, they were supposed to bang the throttles wide open, full military power. Complete panic was meant to ensue in the Warsaw Pact countries.

      The plan was deprecated for the following reasons, i) Tankers and some aviation units were exercised with the equipment, and deliberately and accidentally and as a result of practical jokes with the service crews, the nitrous intended to accelerate the vehicles beyond normal speeds was switched with the helium… and it only took one guy giggling on the open mic to set everyone off. ii) It was found that the West Point study on the plan buried in it’s minutiae the fact that 70% of the effectiveness was anticipated to be from sepsis and disease from the opposing forces collectively crapping themselves in great amounts. Therefore it may have been seen as biological warfare and opening up the possibility of retaliation with anthrax or something else more nasty.


  4. “like any other limited resource, eventually we’ll extract all there is to extract”
    Not until we’ve extracted all the uranium and other radioactive heavy elements. Helium production from decay might be slower than our extraction, but not by a vast a factor as the death and deposition of ocean flora and fauna today is slower than the consumption of fossil fuels.

  5. Qatar, Algeria and Australia combined produce more helium than the USA does. It is a secondary product from the waste stream of LNG production, which is ramping up, not down, therefore any claims of a shortage are highly suspect, or the shortage is just a deliberate manipulation of the market and not a genuine resource scarcity.

  6. It’s a by-product of every natural gas well. When the price is high, they save it. When it’s low, they dump it.

    I would love to try aluminum welding but I can’t afford the helium and my grandson has never had a helium balloon. Thanks for selling off the reserve.

    1. Tig or mig welder with pure argon will weld aluminum with no problem. I weld aluminum every day and we have never had a helium bottle attached to any of my welders.

    2. You really don’t want to do aluminum welding unless you can afford mask with a powered respirator, which keeps nasty fumes away. Stick to welding steel using inexpensive argon gas.

      1. Great article. The fusion is only so unreachable as space rockets. We need only to reach high temperatures and pressures. That will take some time to develop, especially if lot of humans are looking for loot and murder, not for science etc. And it is much better to promise something in 20 years from now (and then:) than throwing the idea as impossible.

  7. If we tried to make enough helium through fusion, we would probably overheat Earth. Just collect helium from a gas planet. Simply collect helium in a very low orbit around such a planet. The technology to do so will not be that hard to come up with.

    1. “The technology to do so will not be that hard to come up with.”

      I just love it when someone say OK guys i’ve done the hard part in coming up with the concept, i don’t understand why why everyone is taking so long to implement it

      Lets look at this will we. So far we have sent 8 missions to the outer planet. The probes tend to be small because energy costs, even with some pretty impressive orbital mechanics are horrendous. To get helium from say Jupiter, not only do we need to get it there, but also get it back, and presumably the ship needs to be big enough to capture appreciable amounts of gas, so again more energy is required. Just getting it to orbit would be a challenge, so it would probably need to be constructed in orbit, a capability we do not have.

      Apart from getting there, there is the whole collecting gas bit. diving into a gas planet energy well is not exactly easy. Firstly to survive you will need some sort of heat shield probably, so more mass, and even then you would need to account for the vagaries of the atmosphere. Too low, you crash and burn, too high and no gas.

      then even if you solve that, what you would collect is a whole mixture of gases, only a small amount would be Helium.

      Finally if it works and you have a train of these flying around the sun collecting helium (yes I read Arthur c Clarke too), you some how need to get it down the energy to Earth, without crashing into somewhere populated or losing the gases you collected (which takes more energy)

      So yes the technology to do so will not be that hard to come up with :) However like most things studies have been done

        1. hoovers work by sucking in air with a big fan, so I am not sure how you can literally hoover up a vacuum. At the end of the day 8% of virtually nothing is not a lot

  8. “for all practical purposes helium is a non-renewable resource”

    Right and just as the peak oil catastrophe has been avoided by new technology and higher prices leading to more being found and extracted, how much helium do you think is present in the Earth’s crust that can be extracted from economically practical natural gas sources? I’d say, a absolutely vast amount, especially as prices go up. See:

    Helium Crisis or Helium Hype? – October 11, 2012

  9. @abjq, @hammarbytp: the worst thing about breaking a Dewar bottle is hoovering up the mess, because the vacuum scattered on the floor will suck out all the dust you already hoovered in due to its low pressure :o)

  10. How do we know all the helium on earth comes from radioactive decay of uranium and that none of it comes from the cloud out of which the solar system coalesced? Did all the gasses from the cloud end up in the atmosphere (so that astronomical helium would have drifted away), and none underground?

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.