As a child, I've read that stars ignite during a so-called "nuclear flash". I understood this as a very violent outburst of energy from the sun, causing all dust between the protoplanets to be blasted away out of the solar system. The gas and dust clots which were too heavy stayed but were melted because of the high temperature, they collided and formed the aforementioned protoplanets, who then cooled down, the fact that they started as melted clots also explains the reason that planets are round.

Now that I'm (a bit) older, one thing is puzzling me: I have the impression of the "nuclear flash" as a very strong explosion, followed by the sun behaving normally almost immediately after, but that does not feel right anymore: I now have the idea that both processes (the "nuclear flash" and the sun's temperature decrease) both happen together, so I wonder:

  • Is the "nuclear flash" indeed a very short explosion, which is finished in a matter of minutes or even seconds or is the birth of a star more a slow process, taking thousands of years before even reaching the "normal" temperature?
  • Is the sun's temperature decrease after the "nuclear flash" very short or does it also take very long?

1 Answer 1


There is no "nuclear flash". The approach to a stable, hydrogen-burning star is rather smooth and in fact the Sun was more luminous during its first few million years as a protostar than it is now. The commencement of hydrogen fusion does not increase the luminosity of the Sun, it just gradually takes over from the luminosity provided by gravitational contraction. The Sun's temperature immediately before and after the gradual initiation of hydrogen burning is more-or-less constant.

The picture below shows a model evolutionary track for a 1 solar mass star in the luminosity vs effective temperature plane (the Hertzsprung-Russell diagram, taken from Stuart-Morrison et al.). The paths of the stars are the red and black lines. The green lines are "isochrones" joining points of equal age. You can see that a 1 solar mass star gets less luminous for about the first 10 million years, then warms up and becomes a little more luminous before dropping onto the "zero age main sequence" at around 30-50 million years old. It is during this 10-50 million year old period that the hydrogen fusion ramps up.

Track for a 1 solar mass star

Most of the planet formation, certainly of the giant planets, is already done by the time hydrogen burning begins in the Sun.

The reason that the protoplanetary disc disperses is not due to the beginning of nuclear fusion in the star. The two are not directly connected.

  • $\begingroup$ I could do with a good one myself, too - thus asking :) Thanks. $\endgroup$ Nov 15, 2023 at 9:16
  • $\begingroup$ Is MIST the MESA Isochrones and Stellar Tracks website/package any use ? $\endgroup$ Nov 15, 2023 at 23:54
  • $\begingroup$ So, you say that during millions of years the inner part of the solar system was that hot that even stone was melting, resulting in a "soup" of glots who combined into the protoplanets (who got round because of their fluidity), and only in a later stage the sun cooled down, resulting in those protoplanets to start freezing from the outside, causing those protoplanets to become real planets with a crust on the outside? (A situation which is not fulfilled in Jupiter's moon Io, as the tidal forces of Jupiter on that moon block the freezing and crusting) $\endgroup$
    – Dominique
    Nov 16, 2023 at 10:00
  • $\begingroup$ Protoplanets melt because the conglomeration of rock into bigger and bigger chunks is a series of violent impacts that have to bleed off vast quantities of kinetic energy as heat, not because the star is just so bright that it's melting the planets. $\endgroup$ Nov 16, 2023 at 15:24
  • $\begingroup$ I have also learnt that the nuclear flash is responsible for cleaning the solar system from all the rubbish between the (proto-)planets. If there is no such a thing as a nuclear flash, then how did the dust between the different planets disappear? $\endgroup$
    – Dominique
    Nov 21, 2023 at 14:52

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .