In an infinite universe, even the most bizarre thought experiments by astronomers—perhaps conceived late at night, perhaps proposed simply to see how weird stars can get—can come to pass. Imagine a massive star, near the end of its life and puffed up to the red supergiant phase, with a tiny neutron star, the skeletal remnant of an even more massive star, at its core. No one knows quite how this Frankenstar might form or how long It would live, and the fusion process would be anything but normal, yet the physics checks out. This mysterious star, called a Thorne-Żytkow object (TZO), could exist. But does it? Amazingly, 40 years after its conception, astronomers think they might have found one of these stars, and it has the potential to upend our understanding of stellar evolution.
A working theory
TZOs are named after Kip Thorne and Anna Żytkow, two astronomers who worked out detailed calculations of what this strange system would look like in 1977 at the California Institute of Technology. They proposed a completely new class of star with a novel, functional model for a stellar interior. Scientists had explored the idea of stars with neutron star cores when neutron stars were first thought of in the 1930s, but their work lacked a detailed analysis or any firm conclusions.
The origin of a TZO goes like this: For reasons not yet clear, the majority of the massive stars we observe in the universe are in binary systems. These stars are several times more massive than our Sun (at least eight times bigger, though stars as large as hundreds of solar masses have been observed) and spend their fuel much more quickly. The largest stars in the universe burn all their fuel in just a few million years, while a star the size of our Sun burns for several billion. In a binary system where the two stars’ masses are unequal, then, the larger of the two runs out of fuel and dies before its partner. The massive component explodes in a fiery supernova as bright as an entire galaxy. When the fireworks are over, this future TZO system is already exotic—the normal, lower-mass star is now paired with a rapidly rotating neutron star with a radius as tiny as 6 miles (10 kilometers), composed entirely of neutrons packed so tightly that they test the extremes of quantum mechanics,
Astronomers already have observed many such neutron star/normal star systems. As the two orbit each other, gas from the normal star can flow onto the outer layers of the neutron star, causing bright x-ray flares. These flares are millions of times more luminous than the x rays emitted by normal stars and are in fact some of the brightest sources of x rays in our galaxy.
But such systems raise a question: What ultimately happens to a system where a neutron star and a regular star orbit each other, but their orbits are unstable? This could occur for a variety of reasons, such as the supergiant’s puffed off gas layers dragging down the neutron star and causing it to spiral in or as a result of the super¬nova explosion that tore apart the first star. In many cases, the neutron star will get a gravitational “kick” that ejects it from the system. But for others, the binary system may reach a final stage of evolution wherein the neutron star orbits closer and closer to its companion, which by this stage is nearing the end of its own life and is a red supergiant star. Eventually, the two stars merge, the red supergiant swallowing the neutron star, and a TZO is born.
In a galaxy the size of our Milky Way, containing hundreds of billions of stars, such mergers should be happening routinely. In fact, scientists have proposed that as many as 1 percent of all red supergiants might actually be TZOs in disguise. “Mergers between a neutron star and a star are common,” confirms Selma de Mink, an astronomer at the University of Amsterdam whose research focuses on stellar evolution. “The question is, what does that look like? For me, that is the big excitement—this happens all the time, but we have no clue.” She explains that some sort of transient and observable event should occur at the moment of the merger—perhaps there is a flare of energy in the x ray or a nova explosion in visible light. Theorists are working on various models, but as yet there is no consensus on what scientists would see at the birth of a TZO.
Made of star stuff
TZOs are important because they have the potential to tell astronomers where some of the more exotic elements in the universe come from. Hydrogen, helium, and trace amounts of lithium were created immediately after the Big Bang. All the heavier elements in the universe, though, formed not at the dawn of the cosmos, but within the heart of a star. Some of these elements we know and love from our daily lives—carbon, oxygen, and iron, to name a few—are produced inside stars through regular processes that are fairly well understood. But the origin of some particularly heavy elements, such as molybdenum, yttrium, ruthenium, and rubidium, is less clear. “These elements are not household names, but still you might want to know where the atoms that make up our universe came from,” jokes Philip Massey, an astronomer at Lowell Observatory in Arizona whose research includes the evolution of massive stars,
Theory suggests that these elements might be created in TZOs. A neutron star inside a red supergiant leads to an unusual method for energy production: The object’s burning is dominated not by the standard nuclear fusion that occurs in other stars, but instead by thermonuclear reactions where the extremely hot edge of the neutron star touches the puffy supergiant’s gas layers. These reactions power the star and also create those heavy elements. Convection that circulates hot gas in the star’s outer layers transports these new elements throughout the star and ultimately even to its surface, where a keen-eyed observer with the right telescope might just spy them.
Hunting for TZOs
But tracking these mysterious objects down is not an easy task. “To an outside observer, TZOs look very much like extremely cool and luminous red supergiants,” explains Żytkow, now at the Institute of Astronomy at the University of Cambridge in England. This means they are nearly indistinguishable from the thousands of other normal, bright supergiant stars that many surveys observe. “However, they are somewhat redder and brighter than stars such as Betelgeuse in the constellation Orion,” she says, naming the famous red supergiant familiar to stargazers.
The only way to distinguish a TZO from a regular bright super-giant is to look at high, resolution spectra—patterns of light astronomers use as stellar fingerprints—to find the specific lines caused by the unusual elements more abundant in TZOs than in typical stars. Such work is severely complicated by the massive number of complex spectral lines from other elements and molecules in the star, which easily number in the thousands. “It is a needle in a haystack kind of problem,” says de Mink.
Despite this, a team of astronomers thinks they might have found the first needle. Nearly four decades and several unsuccessful searches have passed since Żytkow initially worked on the theory behind TZOs. When she saw new research on some unusually behaving bright red supergiants, however, she was intrigued. Emily Levesque, an astronomer at the University of Colorado at Boulder, spearheaded the work with Massey, whom she has been researching red supergiants with ever since an undergraduate summer internship in 2004. Two years later, they discovered several red supergiant stars in the Magellanic Clouds—satellite galaxies of our own—that were unusually cool and variable in brightness. This avenue of research eventually attracted Żytkow’s attention, so she asked whether the team had considered the possibility that these stars might be TZOs.
The potential to find the first TZO was exciting, but identifying a candidate from within the sample of red supergiants would require higher-resolution spectra than ever taken before. Levesque, along with her former mentor Massey and additional collaborator Nidia Morrell of the Carnegie Observatories in La Serena, Chile, secured time to observe a sample of several dozen red supergiants both in the Milky Way and in the Magellanic Clouds using the 3.5-meter telescope at Apache Point Observatory, New Mexico, and the 6.5-meter telescope at Las Campanas Observatory, Chile, respectively. They observed each of the stars with some of the most powerful spectrographs available and then began the meticulous task of identifying the various emission lines in the data and working out the relative elemental abundances in each star.
“It wasn’t immediately obvious at a glance if we had a TZO,” Levesque recalls, “but there was one star that jumped out at us.” A star called HV 2112 in the Small Magellanic Cloud had a particularly bright hydrogen emission line astronomers saw even in the raw data they glanced at as it came in. In fact, it was so unusual that it prompted Morrell to joke at first look, “I don’t know what it is, but I like it!”
It turns out there was much more to like about HV 2112—it had unusually high concentrations of the elements lithium, molybdenum, and rubidium, which are predicted TZO signatures. While finding a star with an unusual abundance of one key element can happen for a variety of reasons, this was the first time astronomers saw all the critical elements in the same star; the team published their results identifying HV 2112 as a TZO candidate in the summer of 2014. “It could still turn out not to be a TZO in the long run,” explains Levesque, “but even if not, it’s definitely a very weird star.”
This discovery was also satisfying for Żytkow, who was instrumental in pushing for telescope time and analysis of the spectral lines. “Work on the discovery of a candidate object which Kip Thorne and I first predicted many years ago is great fun,” Żytkow says. “Since we proposed our models of stars with neutron cores, people were not able to disprove our work. If theory is sound, experimental confirmation shows up sooner or later.”
Revisiting stellar evolution
While finding a “star within a star” sounds intriguing in itself, the discovery of a TZO is particularly interesting to astronomers for what its existence can tell them about stellar evolution. Major research advances in recent years in areas such as stellar convection allow astronomers to update their models for TZOs. These changes may yield new elemental abundances for observers to watch for. Astronomers also want to know whether TZOs can explain where some of the heavy elements come from: Rough estimates so far suggest there could be enough TZOs to explain their formation, but the numbers are highly uncertain.
With only one observed TZO in their stable, how do astronomers estimate how many TZOs are still in the wild, waiting to be discovered? This is not easy to answer: For one thing, no one is sure how long TZOs can be stable. Some models predict that they would be very short-lived objects—lasting only a few thousand years—either due to being torn apart by extremely strong stellar winds or collapsing into a black hole. “Computationally, this is one of the hardest things out there to model,” says de Mink, “so we aren’t sure.”
Research also has focused on finding the remnant of a TZO after it has died. Recently, an international team of astronomers examined the abstrusely named x-ray source 1E161348-5055, which has perplexed scientists since its discovery several years ago. Initial results suggested its power comes from a neutron star—1E161348¬5055 is in fact located in a supernova remnant estimated to be just 2,000 years old—but its rotation period is 6.67 hours. Such a young neutron star should be rotating thousands of times faster; this slow period is more indicative of a neutron star that is several million years old. Several theories have been suggested over the years—perhaps the neutron star has a stellar companion, or perhaps it has an unusually high magnetic field—but no one has explained this mysterious x-ray source to everyone’s satisfaction.
A TZO ghost may fit the bill. As a TZO, it might have burned for up to a million years. But a TZO’s outer layers are not as dense as a normal star’s, meaning this envelope of material is prone to dissipating over reasonably short time scales. The strong stellar wind common in larger stars could be all that’s needed to blow the outer envelope away. This would leave behind a shell similar to a supernova remnant and a neutron star that is far older than its environment suggests—exactly what astronomers see in 1E161348-5055.
Astronomers also are considering whether some parts of our galactic neighborhood might be easier hunting grounds for TZOs. Globular clusters present a particularly appealing target. Stars in a globular cluster all formed around the same time, are densely packed, and are old, meaning they have few of the heavy elements that enrich newer stars. A crowded globular cluster hosts the ideal circumstances to give a neutron star the needed “kick” to merge with a red supergiant star, and the unusual spectroscopic lines would stand out more easily in the metal-poor population.
As spectrographs and telescopes improve and surveys probe ever deeper into our celestial surroundings, TZO-hunters will keep trying to learn more about these weird stars, how they form and how they die, and how many others are waiting to be discovered. As Levesque explains, “It is very exciting to see what’s out there.”
How to make a TZO
A Thorne-Żytkow object starts its life as a normal binary star. One partner is close in mass to the Sun while the other is significantly hotter and more massive (the images to be shown are not at all to scale.) The heavier star buns through its fuel quickly and explodes as a supernova.
After the supernova, the massive partner leaves behind a tiny neutron star (even less to scale!). The Sun-like star consumes its hydrogen fuel more slowly and expands into a red supergiant. At some point , the stars’ orbits become unstable, and they begin to spiral toward each other.
The stars circle each other on decreasing orbits until they merge. The moment of the merger should be observable, but astronomers aren’t sure exactly what to look for. From most perspectives, the newly formed TZO now appears as a normal, if bright, red supergiant.
 See Yvette Cendes, “The Weirdest Star in the Universe,” Astronomy (43, 9, 2016, pp. 50-55). At the time of writing, Yvette Cendes was a Ph.D. candidate in radio astronomy at the University of Amsterdam.