Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Bringing supernovae down to Earth

Nuclear physicists, accelerator physicists and astrophysicists are planning a journey into uncharted territory — studying the nuclear processes that occur when massive stars explode. Alexander Hellemans reports.

When the great Russian chemist Dmitri Mendeleev pieced together his periodic table of the elements, little did he know that many of its members were created in huge cosmic explosions called supernovae. Elements lighter than iron can be formed by nuclear fusion in stars. But heavier elements are mostly forged in the extreme environments of some of the most violent explosions in the Universe.

Fast track: the ISAC heavy-ion accelerator in Vancouver will experiment with radioactive ions. Credit: TRIUMF

The exact nature of those nuclear processes remains mysterious. Theorists have derived models to explain how unstable heavy nuclei are formed in violent stellar explosions, and then decay into the range of heavy elements that are found in the Universe today. But given the complexity of the processes involved, these models are far from perfect. “Almost every aspect of nuclear physics and particle physics comes to bear in a supernova explosion,” observes Dieter Frekers, a nuclear physicist at Münster University in Germany.

But a series of accelerator projects that aims to bring the nuclear physics of supernovae down to Earth could, within a decade or so, provide the means to test and refine the theoretical models. The accelerators will collide unstable radioactive ions with other nuclei to form a range of exotic nuclei. In charting the wild frontiers of nuclear physics, they could also help astrophysicists better understand the dynamics of supernovae.

As large stars burn up their nuclear fuel and form elements up to iron, they end up with an onion-like structure. Towards the end of their life, they have an iron core that is surrounded by concentric layers containing increasingly lighter elements towards the outside — silicon, oxygen, carbon, and so on, with helium and hydrogen at the surface. Eventually, the core reaches a mass of about 1.4 times that of our Sun. It then collapses under its own gravity in less than a second.

When the core can collapse no more, it rebounds, and collides with the still-collapsing layers that lay immediately above it. This creates a cataclysmic shock wave that — together with an energetic outpouring of the neutrinos that are formed as protons and electrons in the compressed iron atoms get converted to neutrons — blows away the star's outermost gas layers in a violent explosion.

In with a bang

Aftershock: the remnant of a type II supernova in which exotic nuclei would have been formed. Credit: NASA

During these extreme events, known as 'type II' supernovae, nuclei fuse and can also capture neutrons to form highly unstable exotic nuclei that exist momentarily before decaying into more stable isotopes. “The explosion is energetic enough that the shock wave that moves out can, by thermonuclear processes, produce other elements,” says Adam Burrows, an astrophysicist at the University of Arizona in Tucson.

Usually, nuclei made unstable by the addition of a neutron undergo beta decay — the neutron expels an electron and is converted into a proton. The nucleus can then accept another neutron and the process continues. This 'slow' neutron capture occurs in certain types of star, and probably also in more violent events such as novae — bursts of thermonuclear activity that occur as a white-dwarf star siphons off material from a companion star by gravitational attraction.

But in the intense conditions of a type II supernova, additional neutrons can be added to the nucleus before beta decay occurs. This is called 'rapid' neutron capture, and causes highly exotic, unstable nuclei to form. These then decay into familiar heavy elements such as thorium, gold and silver. But the lifetimes of the exotic nuclei produced by rapid neutron capture, and their subsequent radioactive decays, are very poorly understood.

Exotic star performers

If physicists can create the exotic nuclei formed by rapid neutron capture in type II supernovae, they might learn more than simply how the heavy elements in the Universe were made. As much of the violence of a supernova explosion is thought to depend on interactions between the outpouring of neutrinos and the nuclei they encounter, knowing more about the exotic nuclei present should allow theorists to make better simulations of stellar explosions.

Investigating such questions will depend on physicists' ability to create and study unstable nuclei. Put too many protons, or neutrons, into a nucleus and the excess will immediately be expelled. In theory, nuclei can exist in more than 6,000 different combinations of proton and neutron numbers (see figure). Less than 300 of these are stable enough to exist naturally on Earth. But using accelerator technology to fragment stable nuclei, physicists have so far created about 3,000 different unstable nuclei.

Figure 1: Combinations of protons and neutrons form a variety of stable (black) and unstable nuclei.
figure 1

The magic numbers of these particles (shown) confer additional stability to the nuclei.

Shooting gallery: CERN's ISOLDE fires high-energy protons at heavy nuclei to make unusual isotopes. Credit: CERN

These exotic nuclei are produced using two main processes. The first, known as the Isotope Separator On-Line (ISOL) method, was pioneered in the 1960s at a facility called ISOLDE at CERN, the European Laboratory for Particle Physics near Geneva. It involves firing a beam of high-energy protons at a target containing heavy, stable nuclei such as uranium-238, so that some of their neutrons and protons are dislodged. The target is heated to around 2,000 °C, allowing the desired isotopes to escape and be separated using chemical or physical methods. For instance, lasers tuned to specific frequencies can be used to ionize particular elements, which can then be manipulated using electric fields.

The second method, called projectile fragmentation, is used by several facilities around the world. Heavy nuclei are accelerated up to 60% of the speed of light and hit a thin target of a light material such as beryllium. As they pass through the target, protons or neutrons are stripped from the nuclei, forming a variety of unusual isotopes. Magnetic and electric fields can then be used to separate the isotopes according to their mass, charge and velocity.

Most of the proton-rich nuclei that can exist in theory have already been made, because they are what ISOL and projectile fragmentation tend to produce. As a result, nuclear astrophysicists have been able to study many of the nuclei created by another process that occurs in extreme astrophysical environments, called rapid proton capture.

Drip fed

But about 3,000 nuclei that exist between the stable region and the upper limit for neutron-richness — known as the neutron-drip line — remain unstudied. Indeed, the exact position of the neutron-drip line on the landscape of possible nuclei remains unclear for heavier elements. And it is the lifetimes and decays of exotic nuclei in the uncharted territory near this drip line that will provide the key to understanding the formation of heavy elements in type II supernovae.

The challenge is to produce enough exotic neutron-rich nuclei to be able to measure their properties. Not only are they difficult to produce, but they often disintegrate before they can be captured and studied. ISOLDE, for instance, has over the years produced more than 600 isotopes of 68 elements, and can churn out hundreds of billions of exotic nuclei every second. But neutron-rich isotopes such as nickel-78 are produced at rates of only one nucleus every few seconds.

By measuring the lifetimes of these isotopes — which can be shorter than a millisecond — physicists can compare the behaviour of unstable nuclei with theoretical models used to describe nuclear structure. For stable nuclei, these models are well established. The neutrons and protons — collectively known as nucleons — that make up a nucleus are arranged into 'shells'. Within each shell, neutrons or protons have similar energy levels and, according to the laws of quantum mechanics, only a certain number of neutrons or protons can exist within each shell.

A nucleus that consists entirely of full neutron or proton shells is said to contain a 'magic' number of protons or neutrons. Many of these nuclei lie outside the zone of stable nuclei, but they are markedly more stable that their immediate neighbours in the landscape of possible nuclei. Some nuclei are doubly magic — they are magic for both proton and neutron numbers — and are more stable still.

When they are formed, magic nuclei persist for longer — up to a few seconds — than other nuclei outside the stable zone. So as neutron capture proceeds in a type II supernova, these magic nuclei will accumulate. “These are milestones in the pathways, and if we are able to find the milestones, we will understand how they are produced — how nuclear synthesis proceeds, step by step, from the iron to the superheavy elements,” says Sidney Gales, director of the Institute for Nuclear Physics in Orsay, near Paris.

But the problem is that, as you move towards the neutron-drip line, the nuclear shell model seems to break down. “By moving out of stability you are modifying the shell structure in such a way that well-known models do not work any more,” says Gales. Because of quantum-mechanical effects, the loosely bound neutrons in exotic, neutron-rich nuclei do not form clear shells, and the outer neutrons form a 'halo' around the core. “This has enormous consequences for the energy level of the nucleons, resulting in different magic numbers,” says Piet Van Duppen of the Catholic University of Leuven in Belgium. It might also help explain the abundances of heavy elements in the Solar System.

Getting heavy

To study neutron-rich isotopes under conditions that are relevant to type II supernovae, it is necessary to accelerate them to high energies. Accelerating beams of unstable, radioactive nuclei was pioneered in 1989 at the Cyclotron Research Centre at Louvain-la-Neuve in Belgium, run by the Catholic University of Louvain, and has been developed further at the Holifield Radioactive Ion Beam Facility at the Oak Ridge National Laboratory in Tennessee, which opened in 1996. So far, the experiments at these facilities have not delved deeply into the exotic territory of heavy, neutron-rich nuclei. But the Oak Ridge facility is now beginning to work with beams of neutron-rich nuclei, such as tin-132 and nickel-56.

Inspired by the examples set at Louvain-la-Neuve and Oak Ridge, other groups are also gearing up to begin accelerating radioactive ion beams. At TRIUMF, Canada's National Laboratory for Particle and Nuclear Physics in Vancouver, a new heavy-ion accelerator called ISAC will in the spring start experiments with a beam of sodium-21 ions, created by the ISOL method. From 2003, an upgrade will allow ISAC to study nuclei with atomic masses of more than 150. At CERN, the ISOLDE facility is now being fitted with a post-accelerator called REX-ISOLDE, which should begin accelerating radioactive beams later this year. Meanwhile, France's GANIL ion-accelerator laboratory in Caen, Normandy, has built a radioactive ion beam facility called SPIRAL. It is essentially a souped-up version of the Louvain-la-Neuve facility, and is currently awaiting permission to start accelerating radioactive ions.

None of the present facilities has the capacity to produce and accelerate the full range of heavy, neutron-rich nuclei produced by rapid neutron capture in type II supernovae. But larger projects now on the drawing-board should provide the means to explore this territory. These 'second generation' facilities will produce radioactive beams some 100 times as intense as those that can be achieved today. Their beams will also be more pure, containing precisely defined nuclei with few contaminants. To investigate the rapid neutron-capture process, these facilities will fire neutron-rich nuclei at targets containing deuterium nuclei, which consist of a proton and a neutron. In the resulting collisions, some of the nuclei should gain a neutron.

Projectile splintering

EURISOL, the European ISOL Facility, is currently the subject of a feasibility study being conducted by ten laboratories across Europe. It could be completed — at a site to be determined — by 2010 and will produce exotic nuclei by ISOL. A comparable facility being planned at the GSI National Research Centre for Heavy-Ion Research in Darmstadt, Germany, will accelerate beams produced by projectile fragmentation. Both production techniques will be needed to investigate the full range of neutron-rich nuclei: ISOL has a higher yield, but is limited to isotopes with relatively long half-lives; projectile fragmentation can produce very short-lived isotopes, but has a lower yield. “We are now investigating which targets we will use, which ion sources, which purification methods, and how we will accelerate the ions,” says Van Duppen.

In the United States, the Argonne National Laboratory near Chicago and Michigan State University in East Lansing are bidding to host the Rare Isotope Accelerator (RIA), which could be in operation by the end of the decade. This proposed $500-million facility will accelerate radioactive ions produced by both projectile fragmentation and ISOL. The RIA should be able to create the majority of theoretically possible isotopes for any element. “What is really exciting and interesting is the ability to reach very far into the neutron-rich terrain,” says Konrad Gelbke of the National Superconducting Cyclotron Laboratory at Michigan State University.

If these second-generation facilities get the go-ahead, nuclear astrophysicists will begin an unprecedented voyage of discovery. Within a decade, predict researchers in the field, exotic beasts that have so far only existed in the thermonuclear cauldron of a type II supernova will finally start to be brought down to Earth.


Louvain-la-Neuve cyclotron

Holifield Radioactive Ion Beam Facility




GSI Future Facility

Rare Isotope Accelerator

Author information

Authors and Affiliations


Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hellemans, A. Bringing supernovae down to Earth. Nature 409, 448–450 (2001).

Download citation

  • Issue Date:

  • DOI:


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing