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.


Chance discovery sheds light on exploding stars

Even with the increasing amount of data archived in all-sky surveys, astronomical discovery still relies heavily on serendipity. Theory can guide the process (and sometimes lead it astray), but there is no ‘standard model’ for the Universe and everything in it, so progress is often driven by the unexpected. It takes a good observer to see something unusual and interpret it as noteworthy. Writing in Nature, König et al.1 have turned serendipity into success by correctly characterizing X-rays, detected in just 35.8 seconds, as the product of an exploding white dwarf — thus contributing to our understanding of the physics of these stars.

The detection came from eROSITA, an instrument that scans a swathe of the sky for X-ray emissions every four hours from an orbit around Earth’s L2 point, which also hosts the James Webb Space Telescope’s orbit. L2 is shorthand for the second of Earth’s five Lagrange points; these are areas in which the gravitational pulls of the Sun and Earth are balanced by the centripetal acceleration due to the satellite’s orbit, so that the satellite remains there.

eROSITA takes six months to build up an image of the X-ray sky, and before the joint German–Russian mission was interrupted by Russia’s invasion of Ukraine, it had completed four full scans. These data allow us to study how X-ray sources vary on timescales from seconds to years.

König et al. recognized that a very bright X-ray source, which was detected only once, in a single scan on 7 July 2020, coincided in space and time with the outburst of the classical nova YZ Reticuli2. From this, they inferred that a decades’-old prediction3 about the physics of such novae was correct.

Classical novae are runaway thermonuclear reactions (essentially, hydrogen bombs) that occur on the surfaces of white dwarf stars — the cinders that remain after low-mass stars have spent their nuclear fuel. White dwarfs are inert bodies with masses close to that of the Sun, packed into volumes comparable to that of Earth. Their densities are around one million times that of ordinary matter.

When a white dwarf in a binary system gains mass by accreting matter from its companion star (Fig. 1), the immense pressure at the base of the added hydrogen layer forces the gas into a state in which its pressure is set by the quantum properties of its electrons, rather than by the thermal pressure. This means that the accreted hydrogen does not expand, even though its temperature continues to increase. A thermonuclear reaction occurs when this layer becomes hot enough to trigger fusion reactions.

Figure 1

Figure 1 | The onset of a nova explosion. As a white dwarf star accretes hydrogen from a companion star (not to scale), the pressure and temperature in the accumulated layer increase, but the gas does not expand. As the temperature continues to increase, a runaway thermonuclear reaction is triggered. Some of the energy released contributes to ejection of the outer layers of the white dwarf, resulting in a nova (not shown) that eventually becomes visible to the naked eye. However, it was previously predicted3 that some of the energy released in the explosion immediately leaks through the accreted layer, and would be detectable as a soft (low-energy) X-ray flash. König and colleagues1 have now verified this prediction.

Because fusion rates increase extremely rapidly with temperature, and because increasing the temperature does not relieve the pressure, the reaction quickly gets out of control. The star’s immediate response is to blow off much of the accreted layer, increasing its radius enormously and relieving the pressure.

In 1990, the runaway reaction was predicted3 to produce a bright flash when the blast from the explosion reaches the photosphere (the outer layer of the white dwarf). At this point, the star shines at the highest possible luminosity, known as the Eddington luminosity, because the outward acceleration of its material due to radiation pressure matches its inward acceleration due to gravity. Before the accreted layer can be ejected, and when its radius is close to that of the white dwarf, the object’s temperature will be several million degrees3. But it will cool rapidly so that the emitted electromagnetic radiation moves out of the X-ray frequency band as the radius increases at several thousand kilometres per second. The nova becomes visible to the naked eye some hours to a few days later, when the temperature has decreased and the emitted radiation moves to optical frequencies.

Later, days to months after the explosion, the expanding ejected layer becomes so dilute that soft X-rays (those with energies of less than one kiloelectronvolt) can pass through it. In principle, this should make the surface of the white dwarf visible again, because it is still radiating at around the Eddington luminosity. But the flash of soft X-rays had never been seen — until now.

Astronomical observation demands patience, but it also requires savvy interpretation. On observing a brief flash of X-rays in the eROSITA data, König et al. realized that it coincided with the position and expected timing of the outburst from YZ Reticuli, providing long-awaited verification of the 1990 prediction.

As well as confirming a theory, this serendipitous discovery makes two new inferences possible. The first is an estimate of the exact time of the thermonuclear reaction. The time lag between the reaction and the beginning of the brightening at optical frequencies informs us of the dynamics of the expanding photosphere. And the temperature of the surface at the time of the explosion constrains the mass of the white dwarf.

The lesson from this observation is that a brief, unexpected detection can allow an astute team to figure out that a rare event has occurred and capitalize on it. Lest it sound too easy, the X-ray source was actually too bright for the detector, which was severely affected by ‘pile-up’ — photons arrived faster than the detector could count them. This severely complicated the data analysis. But by overcoming the problem, König and colleagues have filled a gap in our understanding of how classical novae occur. And all this from 35.8 seconds in the cross hairs.

Nature 605, 230-231 (2022)



  1. König, O. et al. Nature 605, 248–250 (2022).

    Article  Google Scholar 

  2. McNaught, R. H. CBET 4811 (2020).

  3. Starrfield, S., Truran, J. W., Sparks, W. M., Krautter, J. & MacDonald, J. in Physics of Classical Novae (eds Cassatella, A. & Viotti, R.) 306–310 (Springer, 1990).

    Google Scholar 

Download references

Competing Interests

The author declares no competing interests.


Nature Careers


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


Quick links