Hold your hand near a hot stove without touching it and you still feel the heat. There is air in the way, but the warmth reaches you faster and more directly than slow-moving air currents could carry it. What you are feeling is thermal radiation: electromagnetic energy streaming off the hot surface, crossing the gap at the speed of light, and depositing itself on your skin. Every object warmer than absolute zero does this, all the time, in every direction.
This article explains the Stefan-Boltzmann law, the relation that quantifies that radiated energy. We will look at where the steep fourth-power temperature dependence comes from, work a concrete net-exchange example, and clear up the mistakes that trip people up when they apply it to real surfaces.
Why this calculation matters
Radiation is one of the three modes of heat transfer, alongside conduction and convection, and it is the only one that needs no medium at all. That makes it the dominant pathway in vacuum and the controlling factor whenever surfaces run hot. Spacecraft shed all of their waste heat by radiation because there is nothing to convect into. Furnace and boiler design lives or dies on radiative exchange between flames, walls, and load. Incandescent lamps, electric heaters, the cooling of turbine blades, and the thermal balance of buildings all hinge on it.
The fourth-power dependence is what makes the calculation indispensable rather than optional. Because radiated power scales as temperature to the fourth, doubling the absolute temperature of a surface multiplies its emission by sixteen. At room temperature radiation is often a modest term you might neglect; at red heat it can dominate everything else. An engineer who skips the radiation term in a hot-surface energy balance is not making a small error — they are missing the largest channel in the budget.
The core formula
A perfect emitter, called a blackbody, radiates a total power per unit area that depends only on its absolute temperature:
E_b = sigma * T^4
Here E_b is the blackbody emissive power in watts per square metre, T is the absolute temperature in kelvin, and sigma is the Stefan-Boltzmann constant, 5.67e-8 W/m^2K^4. The temperature must be absolute — kelvin, never celsius — because the law counts energy from absolute zero upward.
Real surfaces emit less than a blackbody. The factor that scales them down is the emissivity, epsilon, a dimensionless number between 0 and 1. A polished metal might sit near 0.05; oxidised metal, paint, and most non-metals fall between about 0.8 and 0.95. The emissive power of a real surface is then epsilon times the blackbody value.
The case engineers usually need is not raw emission but the net exchange between a surface and its surroundings. A surface at temperature T1 both emits radiation and absorbs radiation arriving from an environment at temperature T2. The net radiative heat transfer is the difference:
Q = epsilon * sigma * A * (T1^4 - T2^4)
Here Q is the net radiated power in watts, A is the radiating area in square metres, T1 is the surface temperature, and T2 is the surrounding temperature, both in kelvin. When T1 is greater than T2 the surface loses heat; when T2 is greater it gains heat. The fourth-power terms are the heart of it: small temperature changes move Q a great deal, because each temperature enters raised to the fourth.
A worked example
Take a surface with an area of A = 1 m^2 and an emissivity of epsilon = 0.9 — a value typical of paint or an oxidised surface. It sits at a temperature of T1 = 400 K and radiates to surroundings at T2 = 300 K. Find the net radiated power.
Step 1 — raise both temperatures to the fourth power.
T1^4 = 400^4 = 2.56e10
T2^4 = 300^4 = 8.10e9
Step 2 — take the difference.
T1^4 - T2^4 = 2.56e10 - 8.10e9 = 1.75e10
Step 3 — apply the Stefan-Boltzmann law.
Q = epsilon * sigma * A * (T1^4 - T2^4)
Q = 0.9 * 5.67e-8 * 1 * 1.75e10
Q = 893 W
So this surface loses about 893 watts to its surroundings by radiation alone. That is a substantial heat flow from a single square metre with only a 100-kelvin temperature difference — and it comes purely from radiation, with no help from any airflow.
The fourth-power dependence is what makes that number large, and it is worth feeling its strength. If the surface temperature rose from 400 K to 500 K while the surroundings held at 300 K, the bracket would jump from 1.75e10 to about 5.45e10, and the net power would more than triple, to roughly 2780 W. Radiation does not climb gently as a surface heats — it climbs steeply, which is why hot equipment can dump heat far faster than intuition based on convection suggests.
Common mistakes
Using celsius instead of kelvin. This is the single most damaging error. The law counts energy from absolute zero, so the temperatures must be absolute. Putting 400 in as a celsius figure, or forgetting the 273-kelvin offset, corrupts the answer completely because the number is then raised to the fourth power.
Forgetting to subtract the surroundings. A hot surface is also being irradiated by everything around it. Computing only epsilon*sigma*A*T1^4 gives gross emission, not the net loss. The (T1^4 - T2^4) difference is what governs the actual heat the surface sheds.
Assuming every surface is a blackbody. Real materials emit less, and the emissivity varies widely — by more than an order of magnitude between polished metal and matte paint. Dropping the emissivity factor, or guessing it carelessly, can be the largest error in the whole estimate.
Treating radiation as negligible because it feels minor. Near room temperature the radiative term is often small next to convection, so it gets dropped by habit. At a few hundred degrees that habit fails badly — radiation can become the dominant term. Always scale it before deciding to ignore it.
Confusing emissivity with absorptivity in spectral problems. For exchange between surfaces at similar temperatures, treating them as equal is a fair approximation. For surfaces seeing sources at very different temperatures — sunlight on a wall, for instance — the absorptivity for the incoming spectrum can differ from the emissivity at the surface's own temperature.
Try the interactive NovaSolver calculator
Raising temperatures to the fourth power by hand is error-prone, and the real value of the law shows when you watch how steeply the numbers move. The Stefan-Boltzmann Radiation Calculator on NovaSolver lets you set the surface temperature, the emissivity, the area, and the environment temperature, then returns the emittance, the total radiated power, the peak wavelength from Wien's displacement law, and the net radiative exchange — with a Planck spectrum that shifts and grows as you change the temperature.
Related calculators
- Biot Number Simulator — checks whether a cooling solid can be treated as a single uniform temperature, the first question before any transient heat analysis.
- Lumped Capacitance Method — Cooling/Heating Curve Simulator — traces the temperature history of a body that loses heat to its surroundings over time.
- 1D Transient Heat Conduction Simulator — for the conduction side of the problem, when temperature varies through the thickness of a solid.
You can browse the full set in the thermal engineering tools hub.
Closing note
Thermal radiation is the heat transfer mode that asks for nothing in between — no fluid, no contact, just temperature and a line of sight. The Stefan-Boltzmann law captures it in one expression, Q = epsilon*sigma*A*(T1^4 - T2^4), and the fourth-power terms are the part to internalise: radiation grows steeply as a surface heats, so it can quietly become the dominant term in a hot-equipment energy balance. Keep the temperatures in kelvin, never forget the surroundings, choose the emissivity with care, and the radiation term in your heat budget will be the one you trust rather than the one you guess.
Top comments (0)