The Kavli Foundation Q&A: How Did Nature’s Heaviest Elements Form?

A unique galaxy loaded with hard-to-produce heavy elements sheds light on stellar histories and galactic evolution. The Kavli Foundation hosted a Q&A with three astronomers to probe this discovery.

Neutron stars colliding

An artist's impression of two neutron stars colliding.
Dana Berry / Skyworks Digital, Inc.

Researchers have solved a 60-year-old mystery regarding the origin of the heaviest elements in nature, conveyed in the faint starlight from a distant dwarf galaxy.

Most of the chemical elements, composing everything from planets to paramecia, are forged by the nuclear furnaces in stars like the Sun. But the cosmic wellspring for a certain set of heavy, often valuable elements like gold, silver, lead and uranium, has long evaded scientists.

Astronomers studying a galaxy called Reticulum II have just discovered that its stars contain whopping amounts of these metals — collectively known as "r-process" elements. (See "What is the R-Process?") Of the 10 dwarf galaxies that have been similarly studied so far, only Reticulum II bears such strong chemical signatures. The finding suggests some unusual event took place billions of years ago that created ample amounts of heavy elements and then strew them throughout the galaxy's reservoir of gas and dust. This r-process-enriched material then went on to form Reticulum II's standout stars.

Based on a new study, from a team of researchers at the Kavli Institute at the Massachusetts Institute of Technology, the unusual event in Reticulum II was likely the collision of two, ultra-dense objects called neutron stars. Scientists have hypothesized for decades that these collisions could serve as a primary source for r-process elements, yet the idea had lacked solid observational evidence. Now armed with this information, scientists can further hope to retrace the histories of galaxies based on the contents of their stars, in effect conducting "stellar archeology."

Courtesy of The Kavli Foundation, Sky & Telescope is featuring an in-depth Q&A with three astrophysicists on how this discovery can unlock clues to galactic evolution. The participants were:

  • Alexander Ji – is a graduate student in physics at the Massachusetts Institute of Technology (MIT) and a member of the MIT Kavli Institute for Astrophysics and Space Research (MKI). He is lead author of a paper in Nature describing this discovery.
  • Anna Frebel – is the Silverman Family Career Development Assistant Professor in the Department of Physics at MIT and also a member of MKI. Frebel is Ji's advisor and coauthored the Nature paper. Her work delves into the chemical and physical conditions of the early universe as conveyed by the oldest stars.
  • Enrico Ramirez-Ruiz – is a Professor of Astronomy and Astrophysics at the University of California, Santa Cruz. His research explores violent events in the universe, including the mergers of neutron stars and their role in generating r-process elements.

The following is an edited transcript of The Kavli Foundation's roundtable discussion with these three astronomers. The participants have been provided the opportunity to amend or edit their remarks.


THE KAVLI FOUNDATION: What was your reaction to discovering an abundance of heavy elements in the stars in the galaxy called Reticulum II?

ALEX JI: I had spent some time looking at stars in other galaxies like this, and in every one of those, the content of this type of element – which we call r-process elements – was very low. So we went into this whole project thinking we would get very low detections as well with this galaxy. When we read off the r-process content of that first star in our telescope, it just looked wrong, like it could not have come out of this galaxy! I spent a long time making sure the telescope was pointed at the right star. Then I called Anna — actually, I had to wake her up, it was 3 a.m. — and we started doing instrument checks to make sure we were looking at the right thing. It turns out we were.

ANNA FREBEL: It was quite funny, because usually when I get a call in the middle of the night from someone at the telescope, it means something really bad has happened! [Laughter] In this case, we were all super-excited because Alex had found something in the data that was really unexpected and also was a smoking gun. We pretty quickly confirmed that at least that first star he was looking at really had all these heavy elements in rather large quantities.

Then another star showed the same kind of signature. I was like, "Oh my god—we've hit the lottery . . . twice!" We would have been happy walking away with just one awesome star, and then it turned into two, then into three, and four, five and so forth. The universe had thrown us a really big bone!

ENRICO RAMIREZ-RUIZ: I've been working on neutron star mergers for a while, so I was extremely excited to see Alex and Anna's results. Their study is indeed a smoking gun that exotic neutron star mergers were occurring very early in the history of this particular dwarf galaxy, and for that matter likely in many other small galaxies. Neutron star mergers are therefore probably responsible for the bulk of the precious substances we call r-process elements throughout the universe.

Supernova forging heavy elements

An artist's conception of a supernova forging heavy elements.
Supernova illustration: Akihiro Ikeshita; Particle CG: Naotsugu Mikami (NAOJ)


What Is the R-Process?

The r-process stands for "rapid neutron-capture process." This phenomenon, first theoretically described by nuclear physicists in 1957, creates elements in nature that are heavier than iron. In the supernova explosions of massive stars and in neutron star collisions, tremendous numbers of freely moving neutrons bind with iron atoms. As more and more neutrons pile up in the atom's nucleus, the neutrons undergo a radioactive decay, turning into protons. Accordingly, new, heavier elements are formed, because elements are differentiated by the number of protons in their nucleus. As its name implies, this process must occur rapidly in order to build up to very heavy, neutron-rich nuclei that then decay into heavy elements, such as uranium, which has 92 protons compared to iron's 26. While a theoretical understanding of the r-process is sound, scientists have debated over the astrophysical conditions and sites where the process can actually occur.


TKF: Why has the provenance of these elements been such a tough nut to crack?

FREBEL: The question of the cosmic origin of all of the elements has been a longstanding problem. The precursor question was, “Why do stars shine?” Scientists tackled that in the early part of the last century and solved the mystery only around 1950. We found out that stars do nuclear fusion in their cores, generating heat and light, and as part of that process, heavier elements are created. That led to a phase where a lot of people worked on figuring out how all the elements are made.

Understanding how heavy, r-process elements, are formed is one of hardest problems in nuclear physics. The production of these really heavy elements takes so much energy that it's nearly impossible to make them experimentally, even with current particle accelerators and apparatuses. The process for making them just doesn't work on Earth. So we have had to use the stars and the objects in the cosmos as our lab.

JI: As Anna just mentioned, we have been mostly stuck with astronomy, trying to measure what could have made all of these elements out in the stars. But it's also very difficult to find stars that give you any information about the r-process.

RAMIREZ-RUIZ: Right, it is very difficult to see these elements shine when they're created in the universe because they are very rare. For example, gold is only one part in a billion in the Sun. So even though the necessary physical conditions needed to make these elements were clear to physicists more than 50 years ago, it was a mystery as to what sort of objects and astrophysics would provide these conditions, because we couldn't see r-process elements being produced in explosion remnants in our own galaxy.

Two competing theories did emerge, which are that these elements are produced by supernovae and neutron star mergers. These phenomena are very different in terms of how often they should happen and in the amount of these elements they should theoretically produce. Just to give you an example, the explosion of a star with more than eight times the mass of the Sun is thought to produce about a Moon's mass-worth of gold. A neutron star merger, however, is thought to produce a Jupiter's mass-worth of gold. That's over 25,000 times more! So just one neutron star merger can provide the gold we would expect to find in about six million to 10 million stars.

Alex and Anna's observations are so unbelievably useful because they really show that the phenomenon which created these elements is something rare, but that produces a lot of these elements, as a neutron star merger should.

FREBEL: It took 60-something years of work to figure this out, and a variety of astronomers — observers as well as theorists — have all put in their share. That's exactly what we and Enrico are continuing to do.

TKF: Enrico, you study the ionized gas called plasma that composes stars. How is the material in neutron stars different than the plasma in run-of-the-mill stars like the Sun, and how does this provide the raw ingredients for making r-process elements?

RAMIREZ-RUIZ: Neutron stars are only about the size of San Francisco Bay, which I live close by, yet they pack in as much mass as the Sun — about 330,000 times the mass of the Earth. Neutron stars are the densest objects in the universe. A neutron star the size of a Starbucks cup would weigh as much as Mount Everest! We call them neutron stars because they are neutron-rich, and that's a key aspect for making r-process elements, as I'll let Alex and Anna explain.

JI: So the nuclear fusion in stars can only make the elements up to iron. That's because iron is the most stable nucleus. If you try to fuse two things to make elements heavier than iron, it actually takes more energy than the fusion reaction itself releases. A neutron that gets close enough to this dense iron nucleus can join it thanks to one of the fundamental forces of nature, the strong force, which binds protons and neutrons together.

You can keep increasing the size of this nucleus by adding more neutrons, but there’s a trade-off. That nucleus will undergo a radioactive decay called a beta decay. Specifically, one of those added neutrons will spontaneously release some energy and turn into a proton. The r-process is what happens when you capture neutrons faster than the beta decays happen, and in that way you can build up to heavier nuclei.

FREBEL: This process can only happen when you have lots and lots of free neutrons outside of an atomic nucleus, and that's actually a difficult thing to do, because neutrons only survive for about 15 minutes before they decay into a proton. In other words, almost as soon as you have free neutrons, they just disappear. So it's really hard to find places where there are even free neutrons to undergo this neutron capture. As far back as the 1930s, neutron stars had been postulated as something that could exist, and it wasn't until the late 1960s that we knew they were real.

RAMIREZ-RUIZ: As we learned more about neutron stars, we found out that about two percent of them have companion stars, and a very small fraction have another neutron star orbiting around them. If the neutron stars are close enough, they will merge within several billion years or less because they produce gravitational waves as they spin around each other. These waves simply carry off energy and angular momentum, so the stars get closer and closer, and eventually they touch each other.


"Neutron stars are the densest objects in the universe. A neutron star the size of a Starbucks cup would weigh as much as Mount Everest!"  — Enrico Ramirez-Ruiz


TKF: What happens to these heavy elements once two neutron stars collide? 

RAMIREZ-RUIZ: As these neutron stars come together, the stars eject some material in their tidal tails into space at very close to the speed of light. So the atoms of these elements are moving very fast when they are first formed. By the time the ambient gas and dust in the galaxy is able to slow these elements down, they have probably mixed with about a thousand Sun-masses worth of material, enriching it atom-by-atom.

FREBEL: Everything gets nicely mixed, like dough. And from that mixed material, the next generation of stars then forms. This stellar generation contains many, little, low-mass stars that have very long life times. It's these low-mass, long-lived stars that we observed today in Reticulum II for this study.

TKF: Anna, you published a book last year called "Searching for the Oldest Stars: Ancient Relics from the Early Universe." How do these results demonstrate what you call “stellar archeology?”

FREBEL: Finding these elements at Reticulum II thoroughly illustrates the concept of stellar archaeology. The idea is that we can use the composition of individual stars to trace the processes that created the elements in the early universe. Because the elements that we observe in our stars today were made prior to the stars' birth — the stars inherited these heavy elements like "cosmic genes" — we have this incredible opportunity to look back in time to study the early chemical and physical processes that ushered in stars and galaxy formation soon after the Big Bang.

Reticulum II is actually a perfect example of what we now call dwarf galaxy archaeology. It's pretty much the same thing I just described, but now we are able to add other dimensions by not just using individual stars, but the entire dwarf galaxy and all the information that comes with it. We can use galactic environmental conditions and the star formation history to trace what happened very early on in that galaxy that provided the various elements we see today.

It's very nice that despite all the progress we have made in this field, there is more to come. I really think these findings have opened a new door for studying galaxy formation with individual stars and to some extent individual elements. We are seriously connecting the really small scales of stars with the really big scales of galaxies. I'm very excited to see what else we find. I don’t think we'll find another galaxy like Reticulum II anytime soon, but hey, we're going to keep looking!

JI: The way I like to think about this is, imagine if you were an actual archaeologist and you wandered around on the surface of the Earth picking up artifacts whenever you found them. You'd find a collection of random artifacts from different periods and places, and you wouldn't be sure how to associate them. In contrast, looking at galaxies like Reticulum II is like digging into a coherent, subterranean layer and finding a collection of artifacts that are all telling the same story . . .

FREBEL: Like Pompeii!

JI: Yeah, like Pompeii!

TKF: Ah yes, the Roman town, and its residents, who were completely buried under volcanic ash. That was not a very nice outcome . . .

FREBEL: Not for the people, no.

RAMIREZ-RUIZ: But the archeological evidence did remain pristine . . .


"One of the things that I think attracts people to astronomy is understanding the origin of everything around us." — Alex Ji


TKF: Bad for Pompeians, but good for archeologists. Shifting gears here, what tools do you need to dig even deeper, if you will, into how elements like gold and silver originate, and otherwise find more cosmic archeological clues?

Reticulum II galaxy

A Dark Energy Survey image of Reticulum II. The nine stars, described in a recent study, are circled in red, seven of which have high r-process element abundances.
Alex Ji; Background image: Fermilab / Dark Energy Survey)

JI: There are two types of things that we need. First, we have to find dwarf galaxies and that requires very large sky surveys like the Dark Energy Survey—which discovered Reticulum II—as well as surveys conducted by the Large Synoptic Survey Telescope, which will start operations in the 2020s. The second thing is we have to look at the stars in those galaxies. The problem with galaxies is that they are far away, so we need pretty large telescopes to do that.

FREBEL: The stars that Alex has been observing are actually really, really faint. We had to work very hard to squeeze out whatever information we could about them. It was only because these stars had such a strong signal of r-process elements that we could see those signals in their light, very little of which we're actually able to capture with current telescopes.

So that really shows why we need larger telescopes. Multiple telescope projects are underway and are scheduled to open in the 2020s. They will have mirrors more than twice as big as today's best ground-based telescopes. These include the Giant Magellan Telescope, the Thirty Meter Telescope and the European Extremely Large Telescope. They promise more light per unit of time hour, which means we can observe fainter stars, but we can also go back to brighter stars and get insanely high quality data. That is what we need for these r-process stars because there is so much information in their light. I think the next five to 10 years will be very exciting in this regard.

RAMIREZ-RUIZ: I want to make a plug for the Laser Interferometer Gravitational-Wave Observatory, or LIGO. The ultimate dream of mine would be to detect the gravitational wave signal of a neutron-neutron star merger. When we have multiple gravitational wave observatories in operation, such as when LIGO India is built next decade, we will be able to pinpoint the location of these rare events. We can then use our conventional, light-based telescopes to look at the transient light signals from the merged neutron star, which we actually think will be powered by the decay of these precious elements. That would be the ultimate direct evidence that these mergers are indeed producing all of these elements.

FREBEL: Pinpointing the location of neutron star mergers might become possible for events in the nearby universe. But I don't think we'll go back far enough in space, and therefore time, to see a merger like in Reticulum II that went off billions of years ago. I agree with Enrico, though, it would be really great to have a nearby example that shows us, right in front of our eyes, how this really all works.

RAMIREZ-RUIZ: Anna's absolutely right. We won't see the r-process enrichment events that took place at the time when a galaxy like Reticulum II was being formed, but hopefully we'll see the newly synthesized gold closer to home! [Laughter]


"We are seriously connecting the really small scales of stars with the really big scales of galaxies." — Anna Frebel


TKF: Let's take a moment to consider that most of the gold, silver and platinum in our valuable jewelry, as well as the uranium in our nuclear reactors, was created when mind- bogglingly dense neutron stars crushed into each other at incredible speeds. As you're doing your research, does this sort of notion ever stop you in your tracks?

JI: It does stop you in your tracks, right? Definitely one of the things that I think attracts people to astronomy is understanding the origin of everything around us. The other part of it for me is these neutron stars mergers are happening on really small scales, but these events are explosive enough to affect a whole galaxy! Imagining that event, then zooming out to the whole galactic scale, then zooming back down to us on Earth—I think it's pretty cool to be able to follow the consequences of the production of these elements from beginning to end.

RAMIREZ-RUIZ: Something to think about is that all the gold originally here on Earth sank into the planet's center because the early Earth was molten. So all the gold we have today on or near the surface is from asteroid impacts!

FREBEL: As we've been saying, the gold wasn't made in the asteroids, it was probably made in a neutron star merger. It then mixed into the cloud of gas and dust in which all the asteroids and planets formed. That gold was then transported to us on Earth as a special delivery. [Laughter]

RAMIREZ-RUIZ: We have some gold atoms in our bodies, too. If we were to "talk" to one of these atoms, it would tell us a story how it was formed in billions of degrees, how it flew through space. Because just one of these neutron star mergers produced so much gold, probably all of the gold atoms that are in the four of us in this roundtable discussion came from the same event. So we're not only linked by genetics, but by these exotic phenomena that happen in the universe.

2 thoughts on “The Kavli Foundation Q&A: How Did Nature’s Heaviest Elements Form?

  1. Shepsters

    Seems to me that the authors are really only surmising that
    The super heavy elements were formed via neutron
    star collisions. What is their evidence for such an event. Also only 5 or so stars in this dwarf Galaxy have super abundancies, why not all?

  2. GerryP

    I know that newtron stars are very small with extreme densities. I also assume that when 2 such stars collide that a small black hole would result unless I am missing something. If 2 newtron stars collide resulting in a larger newtron star I can sort of understand the article’s main premise but if a black hole is produced would it not capture all the super-heavy element atoms created? Or do the atoms created somehow manage to escape the gravity of the black hole created? Is there some maximum newtron size that is not a black hole? Weird stuff this.

Comments are closed.