Dark Matter and the Frontier of EUV Astronomy

How the discovery of dark hydrogen provides a mundane (and profound) resolution to the Dark Matter problem.

Brett Holverstott
Jan 6 · 16 min read
Dark matter detection from gravitational lensing. Illustration by Matt Schmidt.

In 1933, Caltech astronomer Fritz Zwicky noticed that the galaxies within the Coma cluster were orbiting one another too quickly. Much too quickly.

In the solar system, the orbit of an object is related to its mass and its velocity. If it is moving too fast, like the asteroid Oumuamua, the orbit becomes hyperbolic and the object leaves the solar system. If it is moving too slowly, it falls into the sun. The same rules apply for stars in a galaxy, and galaxies within a galactic cluster.

Based on the visible luminosity of the Coma galaxies, they should have been orbiting one another at about 80 km/s. Instead, they were moving over 1–2,000 km/s. He was led to speculate:

“dark matter is present in much greater amount than luminous matter.”

There was more stuff out there, and it was ‘dark.’ (Or, if the six-year-old Calvin had naming rights, it could have been the “invisible omnipresent lurking mass of doom.”)

Sinclair Smith found the same thing after studying the Virgo cluster. He theorized that the missing mass was contained in large clouds of gas that formed halos around galaxies.

While small inconsistencies between theory and experiment is acceptable and even encouraging, huge inconsistencies make us groan. It means there are likely a number of different possible ways we are wrong; or a combination of ways that we are wrong.

Most astronomers completely ignored the problem. Fifty years went by before there was serious discussion in the scientific community. Meanwhile the evidence continued to accumulate.

J. H. Ort, studying the rotation of the Spindle galaxy (NGC 3115) found that if you look above the equatorial plane, the starlight tapers off dramatically (by a factor of ten) but the mass, judging from the velocity of stars, holds about constant. At the outer edge of the galaxy, the mass was 250 times what he was seeing from luminous matter.

In 1964, Vera Rubin and W. Ford began studying Andromeda, our close neighbor in space (only 2.5 million light years). Ford had invented a new spectrograph that allowed them to measure velocity with a resolution 10 times better than before.

By 1968, they were able to plot the angular velocity of the stars, from the core to the galactic edge of Andromeda. They expected to find a diminishing curve. Instead, the line was mostly flat.

The stars in the metropolis of Andromeda’s central bulge, those in the suburbs, and those in the rural outskirts were all moving at the same velocity. Even those stars at the rim of the visible disk didn’t show any sign of slowing down. Radio observations confirmed the data. Perhaps it was the same problem Zwicky had seen in the Coma Cluster, Smith with the Virgo Cluster, and Oort with the Spindle Galaxy.

Over the next decade, every galaxy studied had a flat rotation curve.

Astrophysicists ran simulations of galaxies to study their motion. The early conclusion: if our theory of gravity is correct over scales vastly larger than our own solar system, the whole Andromeda Galaxy is surrounded, submerged, and stabilized by enormous halos of unseen matter, and these get more dense as you move outward from the galactic core.

This dark matter represents not just some of what is out there, but the overwhelming majority of matter in the universe — perhaps 10 times what can be seen.

Astronomers received this possiblity with incredible hostility. Some thought that Rubin and Ford were ruining their careers by pursuing the problem. Dark matter became the source of arguments at conferences. But eventually astronomers had no choice but to accept the overwhelming evidence that there was a serious problem.

With a sigh, they began to theorize.


Dark matter could be made up by bodies that give off little light: black holes or remnants, rogue Jupiters or (the slightly larger) brown dwarfs, even white dwarfs or neutron stars. Astronomers looked for these Massive Astrophysical Compact Halo Object (MACHO’s) in the galactic halo.

Although we have discovered supermassive black holes at the galactic core with the mass of millions of suns, this mass is in the wrong place and only compounds the problem. It can’t be in the core, it needs to be in the halo.

Perhaps it isn’t an object but a small particle. Maybe it is hot, moving at relativistic velocities through the galaxy, or maybe it is cold, moving slowly but only weakly interacting with normal matter. For the love of acronyms, astronomers call these Weakly Interacting Massive Particles (WIMP’s).

Perhaps neutrinos, once thought to be massless, do have mass. There are trillions flowing through you every second, so even a low mass would have a huge contribution.

Or, perhaps the missing mass is made up of theoretical but never-before-seen particles such as axions, neutralinos, or a particle predicted by string theory, the sneutrino.

Or perhaps there is something fundamentally wrong with our theories of gravity or inertia. After all, evidence for dark matter is indirect. Some suggest we need to modify Newton’s theory over long distances to explain the discrepancy.

What do we go on? Do we modify theory or invent particles, or both? What restraints are there on the speculation of the imagination?

Dark matter seems to be everywhere we look. Perhaps one way to learn more is to look where dark matter isn’t.


We experience time on a characteristic human scale. A starfish on the beach appears to barely move. But time-lapse film shows that starfish are active, even social creatures. To truly observe a starfish, we need to alter our experience of time.

Imagine the time scale of a redwood tree, in which the seasons pass like breaths. Or that of a continent, or a mountain. Galaxies live in something even longer than geologic time: cosmological time, and if we alter our perception to match, we find an equally social existence.

Galaxies often collide over millions of years, their stars interpenetrating and intermingling in a kind of cosmic mating ritual. The gravitational maelstrom of these encounters can often spin off a small eddy of stars that becomes a new galaxy; a cosmic birth.

These young dwarf galaxies carry off stars, dust, and gas that originated in small regions of one or both of the parent galaxies. It is as if we had drawn the sample out with a syringe.

In 2007, Frederic Bournaud noticed that most theories predict that dark matter surrounds galaxies in large halos. So he went looking for a good syringe to obtain material in the galactic interior.

He found NGC5291, a galactic collision surrounded by gas-rich collisional debris, including several dwarf galaxies. He found that these galaxies contained only about twice as much dark matter as luminous matter. And it must have been within the disk of the spiral.

Modeling the evolution of matter in the universe confirms that dark matter clumps similarly to normal matter. And it is likely born cold, with relatively low kinetic energy. It seems the WIMP hypothesis is winning the day. (See Ethan Siegel’s recent article for an overview.)

Perhaps, dark matter is almost ‘normal’ stuff; able to form stars much like normal matter.

We appear to live in a hydrogen universe. Of the matter that we can see, about 95% is made up of hydrogen. And what isn’t trapped in the inferno of a star lingers as clouds of gas throughout the galaxy. Bournaud speculated that the most natural candidate for dark matter (we might say the least imaginative possibility) is hydrogen gas.

Hydrogen is difficult to trace directly, so we estimate its existence via emission lines from carbon monoxide (CO). Bournaud suggested that either the missing hydrogen is very cold such that it doesn’t radiate, or CO lines are not a good indicator of the quantity of hydrogen in these regions. Bournaud needed about three times more hydrogen than could be traced with CO.

To be undetectable, hydrogen gas must be extremely cold, only a few degrees above absolute zero. But the hydrogen observed in these dwarfs was warm, over 400 K. Other gases seen also appeared to be quite warm.

If hydrogen is the most natural assumption for the identity of dark matter, why the hell can’t we see it?


A grazing incidence EUV spectrometer from the Extreme Ultraviolet Explorer. Illustration by Matt schmidt.

On April 22, 1986, Stuart Bowyer, professor at UC Berkeley, watched a rocket launch from White Sands, New Mexico, ten minutes after midnight. It carried a spectrometer capable of seeing in the band of wavelengths known as the extreme ultraviolet (EUV) range, what Bowyer called the “last frontier in observational astronomy.”

For many years, astronomers believed that EUV astronomy would be futile. The Earth’s atmosphere absorbs EUV light, making ground-based telescopes impossible. Any light reaching a space-born telescope would also need to pass successfully through gas that permeates interstellar space in all directions.

Observing in the EUV was also difficult, for logistical reasons: new mirrors had to be developed that could focus the light without absorbing it, that could be exposed to space without glass windows (glass absorbs EUV), and would be stable at extremely cold temperatures.

Unable yet to convince his peers that an EUV satellite was justified, Bowyer applied for smaller grants to launch sounding rockets into the upper atmosphere. With five minutes of observing time on each flight, he was able to test key equipment for later space missions while gathering useful data.

In a study in conjunction with graduate student Simon Labov, the spectrometer would not point at individual stars, but take light from broad regions of space to capture diffuse background radiation. Theoretically, this could tell us something about the intersteller medium, the space between stars.

The findings, published in 1991, identified several peaks that could only be assigned to highly unusual atoms of oxygen, silicon, neon, or iron.

Unusual, because the peaks were not a great match, and they were highly ionized, meaning the atoms had many electrons stripped off them; and the atoms must have been at a very high temperatures, hundreds of thousands of degrees. Further, the temperatures of the atoms assigned to different peaks did not match one another.

The idea that the intersteller medium is filled with gas at wildly different, very high temperatures, generated more questions than answers.

In 1995, when Bowyer finally launched the Extreme Ultraviolet Explorer (EUVE) telescope, it found EUV light, including stars hundreds of light years away, from white dwarves to stars with highly active corona. Astronomers counted thousands of EUV sources across the sky. Space was more transparent then anyone thought.

When the EUVE followed up with a study of the diffuse background however, it did not find the lines seen earlier. Instead it found a different set of broad peaks. The telescope was not specifically designed for diffuse observations, but it had the most sensitive equipment ever used.

When I followed up with Bowyer about the discrepancy, he felt that although all astrophysical data must be statistically significant to be published, most of it is overturned by later research. Perhaps the previous lines were merely artifacts.

Or perhaps they were real, a possibility Bowyer considered equally plausible. Space is filled with stellar processes that result in distinct environments. One emission may be due to a cataclysmic variable star. Another a white dwarf. The source is out there. We just don’t know what it is.


Our solar system lives in an interstellar cloud of gas, within a much larger, lower density cloud called the Local Bubble.

Astronomers have found that much of this gas within this bubble is ionized.

To split an electron from a proton in the hydrogen atom, you needed it to absorb a photon with an energy at least in the ultraviolet, but perhaps higher frequencies such as EUV, X-ray, or Gamma-ray range.

A team recently found that there is four times more ionized hydrogen than can be explained by the background ultraviolet light sources such as quasars and hot young stars. When you look right at a light source, the math works out, but when you look at darker areas of nearby space, there is a huge discrepancy.

Since UV and EUV emissions are quickly absorbed by interstellar gas such as hydrogen, the source for these emissions must be widely distributed across the sky.

The first explanation offered is that our bubble was produced by a supernova explosion some millions of years ago, which ionized most of the gas. But we see the same thing throughout the galaxy. If point-sources such as stars, nebulae, or supernovae are the source, they are not widespread enough. Now scientists are speculating that the solution to this “ultraviolet catastrophe” is some form of dark matter decay.

Could dark matter also be the source of hydrogen-ionizing emissions?

If so, then we should find that the distribution of hydrogen-ionizing sources throughout the galaxy should match the distribution of dark matter. This also appears to match well.

So we are looking for a cold particle that is dark and decays with emission of a high-enough energy light to ionize space gas.


In a laboratory in New Jersey, a little-known research team has discovered a new kind of hydrogen that could be a really, really big deal.

It began thirty years ago, when Randell Mills, fresh out of a graduate course in electrodynamics at MIT and a medical degree at Harvard, figured that there must be a way to understand the structure of the atom with classical physics, the laws of electricity and magnetism.

The hydrogen atom is the simplest atom, composed of one (negatively-charged) electron orbiting one (positively-charged) proton. Since the early 1900’s, physicists had sought to understand why the electron stays bound in its orbit, why it doesn’t rapidly lose energy as the rules of electricity and magnetism dictate it should.

This failure of theory is one of the pillars that led to the invention of quantum mechanics, a whole new view of nature that models the atom statistically, with a murky idea of the underlying structure. For Mills, this was unsatisfactory.

In 1986, Mills’s professor at MIT, Herman Haus, published a paper on how some configurations of charges may accelerate without losing energy. Haus was an engineer, not much interested in atomic theory, and most theoreticians were not willing to admit the possibility that classical physics could still be used to understand the atom. He may not have seen the potential of his own research.

But Mills felt Haus’s work could be the foundation for a new theory of nature, in which electrons were subject to the condition that they not radiate when bound to the atom.

Mills developed a new theory, in which the electron was a spherical shell of moving charge centered on the proton. It seemed to work.

A spherical shell has a kind of natural appeal to physicists. Fritz Zwicky even insulted a college thus: “he is a spherical bastard, a bastard every way you look at him.” But there is good reason to believe that the atom is a sphere. When light or other particles are fired at an atom, they glance off (scatter) in a pattern expected from a sphere.

Pretty much the entire history of classical electron theory gravitates around various forms of spherical shells or solid spheres, oscillating or orbiting. This was the only model for the electron that classical physicists ever found remotely plausible. But never before had it been centered on the proton.

The Potassium Atom, shown as a series of concentric atomic orbital shells using Mills’s model. Illustration by Matt Schmidt.

Mills’s model was the first to offer a good explanation for how the electron could jump from the ground state orbit to excited state orbits in the atom; and why the ground state orbit was stable to radiation, whereas the excited state orbits were not. The calculations all came out right; Mills’s model reproduced the well known Rydberg formula.

But it is here that Mills unexpectedly found another theoretical possibility: the atom could also remain stable at fractional integer multiples of the ground state.

These smaller atoms, what Mills called “hydrinos” had never been imagined before, and no evidence had ever been recognized by a century of research in hydrogen chemistry — although some of it was there. Mills rolled up his sleeves and dedicated his career to hydrino research.

After thirty years of research, and over a hundred papers published in peer-reviewed scientific journals offering analytical evidence for hydrinos, Mills’s team has a pretty good handle on the hydrino atom.

The chemistry of hydrino is unique. A hydrino doesn’t form by a conventional chemical reaction, instead, it must undergo a collision with another species capable of serving as a catalyst. The catalyst undergoes resonant coupling with the hydrogen atom, pulling some of the energy from its orbit, and releasing that energy as ionized electrons or bond breakages.

When the hydrogen atom shrinks to form a hydrino, it releases high energy light, ranging from EUV to soft X-ray light. This kind of radiation from hydrogen has never been seen before.

The reaction cells at Mills’ laboratory are impressive, and the result of decades of engineering. The reactions occur in a hot, ionized gas called a plasma, composed of hydrogen and argon. An oxide is used as a catalyst, and stream of injected gallium acts as a liquid electrode to deliver current to the plasma.

When the current is applied, there is a brilliant, blinding light as hydrino catalysis occurs. Spectrometers confirm most of the light is high energy.

Once formed, the hydrino atom (and resulting dimer of hydrino gas) can be analyzed chemically. The gas shows rovibrational transitions corresponding to molecules with an interatomic distance that is an integer fraction (1/2, 1/3, 1/4, etc…) of conventional hydrogen gas.

However, hydrino atoms and molecules, unlike anything else in nature, do not exhibit electronic excited state orbits. Once the electron drops to a hydrino orbit through catalysis, it is locked in position. This makes sense theoretically in Mills’s model, but the physics is complicated.

Weirdly, the hydrino states are what we might think of as a ladder of ‘ground states’ below the ‘ground state.’

In short, the hydrogen goes dark — other than transitions involving rotation and vibration of the molecules, or nuclear magnetic resonance (NMR) or spin-nuclear coupling (serendipitously seen by a graduate student studying far infrared absorption).

As Vera Rubin was fond of saying “Nobody told us that all matter radiated. We just assumed that it did.”

Mills had found something that didn’t.


When Mills’s team saw the data from Labov and Bowyer, they compared the EUV lines with that of their own plasma experiments. They found that they could explain most of the EUV lines with hydrino transitions (H to H(1/2), H(1/2) to H(1/3), H(1/3) to H(1/4), and so on), and some of these lines could be seen directly in experiments involving mixed helium-hydrogen microwave plasmas.

Other lines that had not directly observed could by explained by shifting them after being scattered by helium, which was present in the cell.

There are many different ways hydrino can form. It can be catalyzed by helium ion; it can be catalyzed by three-body collisions in which two other hydrogen atoms produce one hydrino. Hydrino can also be catalyzed to deeper states by other hydrino atoms, sometimes causing the other to ionize.

Hydrogen is the most abundant atom in the universe; helium is a distant second. Whip up a universe, start with a lot of hydrogen, add a pinch of helium, and shake for a few billion years. You may end up with lot of hydrino on your hands.

The existence of large intersteller quantities of dark ‘hydrino’ hydrogen species could be the most mundane — and profound — answer to the scientific mystery of dark matter.

It seems to fit with everything we know about dark matter. The energetic reaction of hydrino catalysis can explain why clouds of interstellar gas are warm; why there is far more hydrogen than we can visibly see via CO lines or through warm hydrogen emissions. Hydrino molecules will clump and form stars like conventional hydrogen, and respond to radiation pressure (see Rob Lea’s post on Dark Matter Heating).

We also ought to expect a higher proportion of dark hydrogen to normal hydrogen over time, as the universe evolves. But this should be depleted in the galactic core, as dark hydrogen clouds form active luminous stars. Only in the outer halo is the density too low for star formation.

The light produced by hydrino catalysis is also high frequency (EUV and X-ray), which allows it to ionize surrounding space gas. We know that EUV light does not travel far unless it is passing through a ‘window’ in space; but we observe ionized hydrogen across wide swaths of the sky.

In short: hydrino catalysis is happening everywhere, producing a diffuse glow throughout the universe.

Hydrino is telling us, in its own way, that it is out there.


Recently, a team of scientists decided to look for theoretical candidates for dark matter. They didn’t even need to rent a telescope. Instead, they went fishing in a pile of existing data.

X-ray data can be detected by large telescopes equipped with spectrometers; the team borrowed the spectroscopic data from several telescopes’ observations of 73 galaxy clusters. Each cluster, remember, contains dozens of galaxies, each galaxy contains hundreds of billions of stars.

Then they “stacked’’ the spectra, using well-known atomic emission peaks to align the data at different redshifts. This had the effect of amplifying any real signals and diminishing random noise.

They were looking for a theoretical particle called the sterile neutrino. Expecting to find it in a narrow band between 2 and 10 keV, they discovered a small blip in the data at about 3.56 keV (another team found it at 3.52 keV).

The line was extremely faint; and almost impossible to see when glancing at the full spectrum dominated by strong, well-known emission peaks. It would be lost in the noise if you analyzed any one spectra with the human eye. But taking all the data together, the blip emerged.

The blip was most interesting because there was no known corresponding atomic line.

The authors of the study were excited. Perhaps they had finally found the signature of dark matter. After all, good chance that whatever was making the line was happening everywhere in the universe. This remains a strong piece of evidence in favor of the theory.

Mills, however, had his own interpretation.

Mills calculated that a collision between an H atom and an H(1/4) hydrino — which is the most preferred hydrino seen in experiments — could undergo catalysis to generate a H(1/17) hydrino with the emission of a band of continuum radiation with a cutoff at 3.48 keV.

That’s pretty close.

It remains to be seen whether Mills’s team can reproduce the H(1/4) to H(1/17) transition experimentally. If they can, they may be able to convince the astrophysics community that dark matter is not an exotic new particle, but rather more of the mundane stuff from which the luminous universe is made.


Brett Holverstott is author of the book Randell Mills and the Search for Hydrino Energy. This is the first in a series of articles on Mills that adapt content from the book.

Dialogue & Discourse

News and ideas worthy of discourse. Fundamentally informative and intelligently analytical.

Brett Holverstott

Written by

Writer on topics in science & art; architect in Seattle, WA.

Dialogue & Discourse

News and ideas worthy of discourse. Fundamentally informative and intelligently analytical.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade