Every Object Glows
Everything above absolute zero emits electromagnetic radiation. You, your desk, the walls around you — all of them are radiating infrared photons right now. As Feynman puts it: a warm stove cools "by radiating the light into the sky, because the atoms are jiggling their charge and they continually radiate, and slowly, because of this radiation, the jiggling motion slows down."
If you put the stove inside a box with mirror walls, the radiation bounces back, and eventually the light and the matter reach thermal equilibrium. The radiation filling that box has a very specific spectrum that depends only on the temperature — not on what the box is made of, not on the oscillator's charge or mass. All material-specific properties cancel out. This is blackbody radiation — so called because, as Feynman notes, "the hole in the furnace that we look at is black when the temperature is zero."
The ultraviolet catastrophe
Classical physics predicted that the intensity of this radiation should grow as the square of the frequency — forever. More ultraviolet than violet, more x-rays than ultraviolet, without limit. The total energy in the box would be infinite. As Feynman describes it: "Of course we know this is false. When we open the furnace and take a look at it, we do not burn our eyes out from x-rays at all. It is completely false... Therefore, something is fundamentally, powerfully, and absolutely wrong."
In 1900, Planck resolved this by proposing that oscillators cannot take up energy continuously
— they must absorb and emit it in discrete chunks of size ℏω. At high frequencies,
these chunks are so large that thermal energy can't excite the oscillators at all, and the
spectrum cuts off. This was, in Feynman's words,
"the beginning of the end of classical mechanics."
At room temperature (~300 K), the peak wavelength is around 10 µm — deep in the infrared, invisible to our eyes. But the radiation is very real. Your body emits roughly 800 watts of thermal infrared right now — though you absorb nearly as much back from your surroundings, so the net heat lost by radiation is about 100 watts.
At 5500 K — the temperature of the sun's surface — the peak lands right in the visible spectrum. This is not a coincidence. Our eyes evolved to see the wavelengths where sunlight is brightest.
The factor of T4 can be derived from thermodynamics alone, as Feynman shows — but the value of σ requires Planck's formula. To go from energy density inside the box to the flux escaping through a hole, you multiply by c/4: a factor of 1/2 because only outward-moving radiation escapes, and another 1/2 from averaging the cosine of the angle to the hole.
Not All Materials Glow Equally
The Planck spectrum describes a perfect blackbody — one that absorbs every photon that hits it and emits the maximum possible radiation at every wavelength. Real materials fall short.
The ratio of a real material's emission to the theoretical blackbody maximum is called
emissivity, written ε. It ranges from 0 (emits nothing) to 1
(perfect blackbody). And here's the key: emissivity can vary with wavelength.
The difference is dramatic. At the same temperature, soot radiates nearly twenty times more power than polished aluminum. This isn't because soot is hotter — it's because soot is a better emitter.
But why? What makes one surface a better emitter than another?
Kirchhoff's Law
In 1860, Gustav Kirchhoff proved something remarkable. At thermal equilibrium, for every wavelength:
A material's ability to emit radiation at a given wavelength is exactly equal to its ability to absorb radiation at that same wavelength. This is Kirchhoff's law of thermal radiation, and it's one of the most profound results in physics.
Why must this be true? Imagine two plates facing each other at the same temperature. If one could emit more than it absorbs, it would heat the other plate up — violating the second law of thermodynamics. The only way to maintain equilibrium is if absorption and emission are perfectly balanced at every wavelength. This is what physicists call the principle of detailed balance — as Feynman puts it, "every process must, in thermal equilibrium, be balanced by its exact opposite."
Einstein's three processes
In 1916, Einstein showed that light interacts with atoms through exactly three processes: absorption (proportional to light intensity), spontaneous emission (happens even in darkness — an excited atom has some chance of falling to a lower state and releasing a photon), and stimulated emission (also proportional to light intensity — incoming light induces an excited atom to emit). Einstein proved that the absorption and stimulated emission coefficients must be exactly equal: "the induced emission probability and the absorption probability must be equal." This is Kirchhoff's law derived from quantum mechanics.
So the question "why do dark objects emit more?" becomes the question "why do dark objects absorb more?" And that question has a clear answer in quantum mechanics.
What Happens When a Photon Hits a Surface
When a photon arrives at a material surface, exactly three things can happen:
For opaque materials (no transmission), these fractions must sum to one:
A surface that reflects most light looks bright or shiny. A surface that absorbs most light looks dark. And by Kirchhoff's law, the one that absorbs more also emits more.
This is the core answer: dark surfaces absorb more photons, so they must also emit more photons. Color is a visible signature of a material's ability to exchange energy with the electromagnetic field.
The Quantum Mechanics of Absorption
Why do some materials absorb photons and others reflect them? The answer lies in quantum mechanics — specifically, in the electronic structure of the material.
Every atom has discrete energy levels that its electrons can occupy. When a photon arrives, it can only be absorbed if its energy matches the gap between an occupied level and an empty one. The photon's energy is determined by its wavelength:
Isolated atoms vs. solids
In a single atom, energy levels are sharp and discrete — hydrogen absorbs at exactly 121.6 nm, 656.3 nm, and a few other precise wavelengths. This is why hot hydrogen gas glows with distinct spectral lines.
But in a solid, billions of atoms pack together and their energy levels interact. The discrete levels broaden into bands — continuous ranges of energy that electrons can occupy. This is band theory, and it explains everything from why metals are shiny to why soot is black.
Why Metals Shine and Carbon Absorbs
Band theory reveals three classes of materials, each with different optical behavior:
Metals — free electrons, high reflectivity
In metals, the highest occupied band (the conduction band) is only partially filled. Electrons can move freely and respond to incoming electromagnetic waves by oscillating in sync with them. These oscillating electrons re-radiate the light back out — this is reflection.
This is why polished metals are shiny and have low emissivity. The free electrons that make metals good electrical conductors also make them good reflectors and poor emitters. Polished silver reflects about 97% of visible light, leaving only 3% absorbed.
Insulators — band gaps determine color
In insulators, the valence band is completely full and the conduction band is completely empty, separated by an energy gap. Photons can only be absorbed if their energy exceeds this gap.
If the gap is larger than any visible photon's energy (> 3.1 eV), no visible light is absorbed — the material is transparent or white. Glass (gap ~9 eV) and diamond (gap ~5.5 eV) are examples.
If the gap falls within the visible range, only certain colors are absorbed, and the material appears colored. Cadmium sulfide has a gap of 2.4 eV, absorbing violet, blue, and much of green (wavelengths below ~515 nm) while transmitting yellow and red — hence its bright yellow color.
Carbon (soot, graphite) — overlapping bands, total absorption
In materials like carbon soot, the band structure has no gap at all — or many overlapping bands that cover a continuous range of energies. Photons of every visible wavelength can find an available transition. The material absorbs broadly across the spectrum and appears black.
The Surprise: Visible Color ≠ Infrared Emissivity
Here's where it gets subtle. Kirchhoff's law says emissivity equals absorptivity at each wavelength. But the wavelength that matters for thermal radiation depends on the temperature.
At room temperature, thermal radiation peaks around 10 µm — far infrared. The color you see (400–700 nm) tells you about absorption in the visible band, but it says almost nothing about absorption at 10 µm.
This is why white paint has an infrared emissivity of ~0.90, nearly as high as black paint (~0.95). Both absorb infrared well because their molecular bonds (C–H, O–H, C=O) have vibrational modes that couple strongly to infrared photons.
The difference between them is only in the visible range. White paint contains titanium dioxide particles that scatter visible light (high reflectivity → low absorptivity in the visible). But in the infrared, those particles are too small relative to the wavelength to scatter effectively, so the paint's polymer matrix absorbs just as well as black paint's.
The takeaway: for objects near room temperature, visible color is a poor predictor of thermal emission. What matters is the infrared emissivity, which depends on molecular bond vibrations and crystal lattice modes rather than electronic transitions. Most non-metallic surfaces have high infrared emissivity regardless of color.
When Color Does Matter
Color does determine emissivity when the thermal radiation overlaps with the visible spectrum — that is, at high temperatures.
At 300 K, the difference is small — maybe 5%. But at 2000 K, a black surface emits up to four times more than a white one, because a significant fraction of the radiation is now in the visible range where their emissivities diverge.
This is directly visible: heat a piece of iron in a forge and it glows red, then orange, then white. A piece of white ceramic at the same temperature would glow much more dimly, because its visible-wavelength emissivity is lower.
The Full Picture
Let's trace the complete chain of reasoning, from quantum mechanics to the macroscopic world:
1. Electrons in solids occupy energy bands. The band structure — whether there are gaps, how wide they are, whether bands overlap — is determined by quantum mechanics and the arrangement of atoms.
2. A photon can only be absorbed if its energy matches a gap between a filled state and an empty state. Materials with many available transitions across a wide energy range absorb broadly and appear dark. Materials with few available transitions reflect light and appear bright.
3. By Kirchhoff's law, a material that absorbs well at a given wavelength also emits well at that wavelength. This isn't a coincidence but a thermodynamic necessity.
4. Which wavelengths dominate thermal emission depends on temperature (Planck's law). At room temperature, it's the infrared — and most non-metallic surfaces absorb infrared well regardless of visible color. At high temperatures, the visible spectrum contributes significantly, and visible color directly determines emissive power.
5. Metals are the exception at all temperatures. Their free electrons reflect across a wide range of wavelengths (visible and infrared), giving them low emissivity everywhere. A polished metal surface radiates poorly regardless of temperature — which is exactly why thermos flasks use silvered walls.
The deepest insight is that emission and absorption are the same physical process running in opposite directions. Einstein showed that the probability of absorbing a photon and the probability of stimulated emission are exactly equal — any quantum state that can absorb a photon of energy E can also emit one. This is not an approximation: it is forced by the requirement that thermal equilibrium be self-consistent at every frequency. As Feynman notes, the quantum theory of light interacting with an oscillator "gives exactly the same result as that given by the classical theory" — the relationship between emission and absorption is that robust.