Hubble's 1999 cold case points to a planet born from a dead star

The star here is a white dwarf, the dense leftover core of a Sun-like star that ran out of fuel and shed its outer layers. The idea is that some of that cast-off material settled into a disc and clumped into a gas giant about the size of Jupiter. Because it was built from a dying star's leftovers, its chemistry looks nothing like a planet made from a star's birth material.

How did a 1999 mystery get solved?

Jamie Williams, a doctoral candidate at the University of Warwick, went back to the old Hubble data with updated chemical databases and found that niobium matched many of the unexplained features. Niobium had never been reported in a white dwarf's atmosphere before. Here it shows up at more than 1,000 times the level seen in the Sun, alongside zinc and copper.

That matters because of where niobium comes from. Elements heavier than iron "are not formed in the cores of stars by thermonuclear fusion," said Nicholas Stone, a theoretical astrophysicist at the University of Wisconsin-Madison and a member of the research team. Stone calls the pattern "a telltale sign" of the s-process, which forges heavy elements inside bloated red giants near the end of their lives. So the star appears to be eating material from that dying-star stage.

The case took shape over decades. Hubble recorded the unexplained features in 1999, and archival data from NASA's FUSE ultraviolet mission later backed up the niobium finding. NASA's TESS then watched the star for four months and picked up a repeating brightness signal, though the releases don't say when those observations took place. The study, "Discovery of a second-generation planet candidate accreting onto a white dwarf," came out in Nature Astronomy on Oct. 5, 2026 (doi:10.1038/s41550-026-02983-7).

Is the planet actually confirmed?

No, and the messaging isn't consistent on this. The paper's own title calls it a second-generation planet "candidate", and NASA and ESA say "suspected" and "candidate" throughout. The University of Warwick's press release is bolder. In the body text it calls this "the first second-generation planet ever found".

Nobody has photographed the planet. The case rests on two clues: the strange chemistry, and a faint brightness signal from TESS that repeats every 4.4 days. That points to a world orbiting about 3.7 million miles (6 million km) from the white dwarf, far closer than Mercury is to the Sun. The team thinks the white dwarf's heat is boiling off the planet's atmosphere, and that the stripped gas falls onto the star and leaves the niobium signature behind.

An outside expert finds the chemistry striking. "We know of a large number of white dwarfs that are polluted by planetary material which is similar in composition to rocks within our solar system. However, in this case, the white dwarf is polluted by incredibly unusual material," Sarah Casewell of the University of Leicester, who wasn't involved in the study, told CNN. Her comments were about the chemistry, though. We didn't find an independent researcher assessing the TESS signal itself.

There's another catch. A single dying star usually throws its material out evenly, so a disc doesn't form easily. Williams says this system "likely required a companion star" to pull the ejected material back into orbit, and he says that helps explain why such planets would be rare. The releases we read don't say whether that companion is still around.

What happens next?

Co-author Boris Gänsicke is already looking closer to home. One example, he said, raises the question of "how many more might be out there," and whether our own Sun might one day build a planet from its ashes. Despite losing atmosphere, the planet is expected to survive once the white dwarf cools.

The next step is more telescope time. Williams plans to use Hubble over several years to work out how these planets form, how common they are and how their orbits change. Until more data firm up that 4.4 day signal and the planet behind it, this stays a candidate.

The WattsUpNext take: Hubble's possible second-generation planet is a lovely piece of detective work on 27-year-old data. Treat it as a strong candidate for now, not a confirmed discovery.