Sunlight takes about eight minutes and twenty seconds to travel from the Sun to the Earth. That figure is well-known and accurate. The slightly less well-known part of the story is what happens before the eight-minute trip. The energy carried in any photon of sunlight currently reaching the Earth was generated by nuclear fusion in the Sun’s core, but it did not exit the Sun immediately afterwards. It spent somewhere between tens of thousands and a few hundred thousand years working its way outward through the Sun’s interior before reaching the surface and beginning its short trip across space. The eight-minute final leg is almost the only part of the journey that anyone tends to remember.
The standard short version of this fact, which appears in textbooks and on planetarium walls, is often phrased as “the photon you see today was born in the Sun’s core a hundred thousand years ago.” That is roughly right at the level of energy. It is not literally right at the level of photons. The distinction is worth making, because the actual story is more interesting than the simplified one.
What is happening inside the Sun
Energy is generated at the centre of the Sun by the fusion of hydrogen nuclei into helium, releasing high-energy gamma-ray photons at a temperature of roughly 15 million kelvin. The core occupies the inner quarter of the Sun’s radius. Beyond the core, the next layer outward is the radiative zone, which extends from about 25% to about 70% of the Sun’s radius. According to NASA’s “Layers of the Sun” overview, this is the region in which energy from the core is transported outward by radiation, with the outer 30% of the Sun being dominated by convection instead.
The radiative zone is where the slow part of the journey happens. The photons created by core fusion zigzag outward through this region in what is sometimes described as a “drunken walk,” repeatedly absorbed by atoms and re-emitted in random directions. The reason for the slowness is geometry. Each photon travels only a short distance, on the order of a millimetre or less, before colliding with an electron or ion in the dense plasma, being absorbed, and being re-emitted in a new random direction. The radiative zone is so optically thick that this absorption-and-re-emission cycle happens an enormous number of times before a photon reaches the outer edge of the zone.
The classic calculation by Robert Mitalas and Kenneth Sills, published in The Astrophysical Journal in 1992, used a solar model to determine that the average free path of a photon between absorptions is about 0.9 millimetres, which gives a diffusion time from core to surface of roughly 170,000 years. Earlier estimates, using simpler assumptions about the Sun’s interior, had given figures ranging from a few thousand years up to many millions. According to Princeton University’s introductory astrophysics course notes, a simple random-walk estimate using a 1-cm step length gives a photon diffusion time of about 30,000 years, which is roughly the lower bound of credible figures. The 170,000-year figure from Mitalas and Sills is the most-cited modern estimate, but the range from tens of thousands to a few hundred thousand years is all within the spread of defensible answers depending on the assumed density profile.
The thing the photon is not
The detail that gets lost in the popular telling is that a photon does not actually retain its identity through the journey. Each time a photon in the radiative zone is absorbed by an atom, that photon ceases to exist. Some short interval later, the atom emits a new photon, in a new random direction. The new photon is not the old photon. It carries some of the original photon’s energy, but it is a different particle. Over the course of the journey from core to surface, this absorption-and-re-emission cycle occurs roughly 1025 times. The energy is conserved through the chain. The particle identity is not.
This is why the standard phrasing “the photon you see now is 100,000 years old” is misleading. The photon you see now is a few microseconds old, emitted from the Sun’s outermost visible layer, the photosphere, after being processed through countless prior absorptions and re-emissions. The energy in that photon, however, was generated tens of thousands of years ago by fusion in the core, and has been migrating outward ever since through a long chain of intermediate photons. The original gamma-ray photon emitted by the fusion reaction has long since been absorbed and forgotten. What survives is the energy.
The frequency of the radiation also changes outward, not because individual photons are losing energy at each scattering, but because at each depth in the Sun the radiation field is in approximate local thermodynamic equilibrium with the surrounding plasma. Closer to the core, the plasma is hotter and the local population of photons peaks at gamma-ray frequencies. Closer to the surface, the plasma is cooler and the local population of photons peaks at visible-light frequencies. As energy migrates outward, the photon populations it occupies are thermalized to successively cooler local temperatures, and what arrives at the photosphere is a visible-light spectrum rather than the gamma-ray spectrum produced in the core.
An even longer timescale, when measured differently
There is one further refinement worth flagging, because it makes the article’s central editorial point sharper rather than softer. In 2003, the solar physicist Michael Stix pointed out in Solar Physics that the photon-diffusion timescale calculated by Mitalas and Sills is not the same as the time required for energy to migrate from the core to the surface. The reason is that most of the Sun’s thermal energy is not stored in its radiation field; it is stored in the thermal motions of the electrons and ions making up the plasma. The photons exchange energy continuously with that vastly larger thermal reservoir at every scattering. The relevant timescale for energy transport, Stix argued, is the Kelvin-Helmholtz timescale, of order 3 × 107 years, about 100 times longer than the photon diffusion time. By this measure, the energy reaching Earth today was generated by fusion not tens of thousands but tens of millions of years ago, with the photons merely the carriers along the way.
The popular version of the story, which is the version this article opened with, picks the photon-diffusion timescale of tens of thousands of years because that is the quantity that most people think they are being told about. The strictly accurate version, taking the heat capacity of the plasma into account, is even more extreme. Either way, the underlying point holds: the eight-minute trip from the Sun’s surface to Earth is a tiny final coda to a journey through the solar interior that took thousands to millions of years.
What happens at the surface
Above the radiative zone, the mode of energy transport changes. The plasma in this region, called the convective zone, is no longer dense enough for radiation to be the efficient way to move heat. Instead, hot plasma rises in vast convective cells, releases its energy near the surface, and sinks back down to be reheated. This convective transport is much faster than the radiative diffusion through the deeper interior. According to Montana State University’s Solar Physics group, hot material can carry its energy through the entire convective zone in a little over a week, compared with the tens of thousands of years required for the radiative zone underneath.
At the top of the convective zone, the plasma finally becomes thin enough for photons to escape the Sun freely. This thin outer layer is the photosphere, only about 500 kilometres thick, with a temperature of roughly 5,800 kelvin. The photons emitted from this layer are the photons that reach Earth. They are, in the strict sense, freshly minted at the photosphere. They are also, in the broader sense, carrying energy that has been migrating outward through the Sun for a very long time.
What this changes
The figure most often quoted, and the one quoted in the title of this article, is the photon-diffusion time from core to surface, not the lifetime of any individual particle and not the longer thermal-energy-migration timescale. That diffusion time was estimated at roughly 170,000 years by Mitalas and Sills, with credible alternative figures spanning tens of thousands to a few hundred thousand years depending on the modelling assumptions. The energy itself, on Stix’s analysis, has an even older provenance than that, going back tens of millions of years.
This is the kind of fact that does not change anything practical. The Sun continues to deliver light at the same intensity it has for billions of years. What it changes is the mental picture. The blank distance between the Sun and the Earth, eight minutes wide at the speed of light, is by far the shortest part of the energy’s journey. The long part happened inside the Sun, and it ended before the photon you can see right now was emitted.
