The light on the back of your hand crossed 150 million kilometers of empty space to get there. It left the Sun’s surface, the photosphere, and flew in a straight, uninterrupted line to Earth in eight minutes and twenty seconds. That is the part of the story everyone knows, and it is true.
It is also the last eight minutes of a journey that began before there were cities, or writing, or agriculture — and almost none of it looked anything like the calm crossing at the end.
The light was born at the surface
Run the clock backward from your hand and the first surprise is that the photon warming your skin is young. It was born at the photosphere, the relatively cool 5,500-degree-Celsius surface where the Sun finally becomes transparent and a photon can fly free. Before that moment it did not exist as the soft light you can see. The energy it carries is ancient; the photon itself is not.
Push back further, below the surface, and the straight line ends. The eight-minute crossing is the only part of the trip that looks like the light we imagine. Everything before it was a stagger through a wall of plasma.
The seep, not the flight
The energy did not so much travel out of the Sun as seep out of it, and the reason is that the Sun is not empty. The core and the layer above it, the radiative zone, are packed so densely with ionized matter that light cannot travel in a straight line for any meaningful distance.
A gamma ray streaks off at the speed of light and, within a fraction of a millimeter, runs into something. Sometimes it glances off a free electron and is simply knocked sideways, a little poorer in energy than before. Sometimes it is swallowed outright, and a moment later a new photon leaves in its place. Either way it comes off in a random direction, with no memory of where the first one was heading, and the outward line has to start over. Then it happens again. And again. These deflections come over distances well under a millimeter near the core, lengthening only gradually in the thinner reaches of the radiative zone above. Each leg points nowhere in particular. The photon staggers outward the way a drunk staggers home: every step the same short length, every direction random, progress made only because there are unimaginably many steps and the only open space is, on average, outward.
This is a random walk, and its mathematics are unforgiving. To cover a real distance you do not need a few times more steps than a straight line would take; you need the square of that. The radiative zone reaches nearly half a million kilometers out from the center; the photon’s stride through it averages under a millimeter. The number of bounces required to cross that gap, by a path that doubles back on itself constantly, is beyond intuition.
The crawl does not last all the way to the surface. At roughly seven-tenths of the way out, the gas has cooled enough to turn stubbornly opaque, and the bouncing stops paying its way; instead of handing energy up the line photon by photon, the plasma starts simply carrying it — great columns of hot gas rising, dumping their heat near the top, and sinking back down. That is the convection zone, and it moves energy in something like a week. Nearly all of the long wait happens below it.
How long the seep takes
The honest answer is that nobody can hand you a stopwatch number, because the Sun is not uniform. Density and temperature change enormously from the core to the surface, and so does the mean free path — the average distance a photon covers between collisions. Near the dense core it is well under a millimeter; higher up it stretches. The transit time depends entirely on assumptions about that varying interior, which is why the published estimates scatter so widely.
The figure you will see most often is around 170,000 years, and it comes from a 1992 calculation by Robert Mitalas and Kenneth Sills. Their contribution was to stop assuming the photon’s stride was the same everywhere. Earlier estimates had used a step length of half a centimeter to a centimeter throughout and arrived at something between three thousand and thirty thousand years. Feed in a realistic density profile instead — one where the average step works out to about nine-tenths of a millimeter, and stays under a millimeter through more than half the Sun’s radius — and the answer stretches by an order of magnitude, to about 170,000 years.
So the textbook number moved up, not down, and the older, shorter figures still in circulation are the ones that have been superseded. Even 170,000 is a model-dependent estimate rather than a measured fact: change your assumptions about the Sun’s interior and the number moves with them.
There is one more wrinkle, and it cuts the other way. What Mitalas and Sills computed is how long radiation takes to diffuse out — the photon-diffusion time. That is not quite the same as how long a given parcel of heat takes to work its way to the surface. In 2003 the solar physicist Michael Stix pointed out why: almost all of the Sun’s heat is not stored in its light at all but in the thrashing motions of the electrons and ions, a reservoir vastly larger than the radiation field. At every collision the photons trade energy with that enormous heat bank, and the bank has to be filled and emptied too. Reckoned that way, the transport time is the Kelvin-Helmholtz timescale — of order thirty million years, roughly a hundred times longer. So the honest summary is that 170,000 years is the light’s crawl, not the last word on the heat’s, and anyone quoting millions of years is not simply wrong; they are answering a slightly different question.
Either way, the scale is hard to absorb. On the 170,000-year figure, the radiation now leaving the Sun’s surface set out from the core before the last ice age had even begun — back in the glacial period before that one, when our species had already been around for well over a hundred thousand years and still had no cities, no writing, no agriculture. It has spent the entire span of human civilization, many times over, bouncing around inside one star.
It was never the same photon
Here is the part that gets quietly mangled in the retelling. People say a photon takes 170,000 years to get out of the Sun, as though some heroic particle survives the whole ordeal. It does not. A photon may survive a few glancing scatters and still be itself, but it takes only a handful of steps before an absorption destroys it outright, a fraction of a millimeter from where it started. What continues outward is a relay, handed off from one short-lived photon to the next through countless re-emissions. Photons are created and destroyed the whole way up; what makes the journey is the energy, not the messenger.
And the energy changes as it goes. With every absorption and re-emission in cooler and cooler material, it tends to get parceled into more photons of lower energy. By the time it reaches the surface, the energy of a single core-born gamma ray has been spread across several million ordinary photons of visible and infrared light. The violent, invisible packet from the core has become a soft handful of sunlight.
The furnace at the bottom of the well
Follow the relay all the way down and you reach the core, the source of everything above it. There the temperature is about 15 million degrees Celsius and the pressure is past anything that has a useful comparison on Earth. Hydrogen nuclei, stripped of their electrons and moving fast enough to overcome their mutual repulsion, slam together and fuse into helium. Each fusion leaves a little less mass than it started with, and that missing mass comes back out as energy. Not all of it as light: some becomes raw motion imparted to the particles themselves, and a couple of percent leaves as neutrinos that flee the Sun without touching anything on the way. But some of it emerges as photons, and they are violent ones — gamma rays, carrying far more energy than anything your eye can register.
That gamma ray is the beginning of the light that will, someday, warm your face. It is not the light that warms your face. Almost nothing about the photon that escapes the Sun resembles the one that was born in the core, and the journey between the two is one of the strangest commutes in physics.
The man who guessed the furnace
For most of human history nobody knew what powered any of this. The Sun was obviously burning something, but coal or any ordinary fire would have consumed it in a few thousand years, and the Sun was plainly far older than that. The energy source was a genuine mystery.
On 24 August 1920, an English astronomer named Arthur Eddington stood up in Cardiff to give the presidential address to the mathematical and physical science section of the British Association. The talk was called “The Internal Constitution of the Stars,” and in it Eddington made a leap.
“A star is drawing on some vast reservoir of energy by means unknown to us,” he said. “This reservoir can scarcely be other than the sub-atomic energy which, it is known, exists abundantly in all matter.” He had a candidate mechanism, too. The chemist Francis Aston had recently measured that four hydrogen atoms weigh slightly more than the single helium atom they could in principle combine to make — a loss of about one part in 120. Eddington saw what that meant: fuse hydrogen into helium, let the missing mass become energy, and a star has fuel enough to burn for ages. “We need look no further for the source of a star’s energy,” he said.
He knew how far out on a limb he was. “I should not be surprised if it is whispered that this address has at times verged on being a little bit speculative,” he admitted. And he saw, even then, where the idea pointed. If humans ever learned to tap subatomic energy, it might bring “a little nearer to fulfilment our dream of controlling this latent power for the well-being of the human race — or for its suicide.” That was said a quarter of a century before Trinity.
The man who proved it
Eddington had the idea but not the machinery. He could not write down the actual nuclear reactions, because in 1920 the necessary physics did not exist. It took two more decades.
In 1939 a German-born physicist named Hans Bethe, by then at Cornell, published a paper in the Physical Review with the flat title “Energy Production in Stars.” It was the second half of a two-part answer. The year before, working with Charles Critchfield, he had cracked the hardest step of the proton-proton chain — the reaction sequence that dominates in a star like the Sun — by computing how often two protons actually stick together to make deuterium. That first collision is the bottleneck; everything downstream of it runs fast enough not to matter for the energy budget, though the chain’s later branches would not be worked out in detail until the 1950s. The 1939 paper supplied the other route, the carbon-nitrogen cycle, in which carbon and nitrogen act as catalysts that shepherd four protons into a helium nucleus and emerge unchanged, and which takes over in stars hotter than ours. Between the two he had closed the loop Eddington opened. The furnace was real, the fuel was hydrogen, and the steps were now written down. Bethe received the Nobel Prize in Physics in 1967, the citation pointing specifically at his discoveries concerning the energy production in stars.
The energy he described is the same energy that, on some particular afternoon, finishes its long crawl, takes eight minutes and twenty seconds to cross the dark, and lands on the back of a hand. The hydrogen that made it fused before anyone was there to wonder about it.