Calories on your plate and kilowatt-hours on your electricity bill seem to belong to two separate worlds: one rules nutrition, the other the power station. Yet to physics they are exactly the same phenomenon. Energy is a scalar quantity describing a system's ability to do work — and it does not care in the slightest whether that work is done by an electric motor or by the contraction of a muscle fuelled by the sugar in a doughnut. A calorie, a joule and a kilowatt-hour are three different rulers for the same thing. To prove it, we will take one ordinary doughnut and find out how much electricity is hiding inside.
One quantity, many masks
The confusion comes from history, not from physics. Each field invented its own unit long before anyone proved they all measure the same thing.
The calorie was born alongside steam engines. One thermochemical calorie (1 cal) is the amount of heat needed to raise the temperature of 1 gram of water by 1 °C — exactly 4.184 joules. In nutrition we use its thousandfold, the kilocalorie (1 kcal = 1000 cal = 4184 J); that is what we colloquially call a "calorie" and print on packaging.
The joule is the SI derived unit of energy — the work done when a force of one newton acts over a distance of one metre. It is to the joule, as a common denominator, that NebulaMath reduces every other unit of energy.
The watt-hour and its multiple, the kilowatt-hour, come from the world of electricity. A watt is a power of one joule per second (1 W = 1 J/s), so a watt-hour is simply the work done at one watt for one hour:
1 Wh = 3600 J, and 1 kWh = 3,600,000 J = 3.6 MJ
Once you substitute the definitions into one another, the supposed gulf between food and electricity vanishes. Here are the same values in a single table:
| Unit | joules (J) | kilocalories (kcal) | watt-hours (Wh) | kilowatt-hours (kWh) |
|---|---|---|---|---|
| 1 J | 1 | 2.39 × 10⁻⁴ | 2.78 × 10⁻⁴ | 2.78 × 10⁻⁷ |
| 1 kcal | 4184 | 1 | 1.1622 | 1.1622 × 10⁻³ |
| 1 Wh | 3600 | 0.8604 | 1 | 0.001 |
| 1 kWh | 3.6 × 10⁶ | 860.42 | 1000 | 1 |
One kilocalorie from your plate is about 1.16 watt-hours on the meter. Everything else in this article is just a consequence of that single equality.
The quarrel over heat that gave birth to thermodynamics
Before science arrived at that simple equality, heat and work were the subject of fierce disputes. The very word "calorie" (from the Latin calor, heat) was coined between 1819 and 1824 by the French chemist Nicolas Clément during his lectures on steam engines. It only entered English in 1863, with the translation of a popular physics textbook.
The crucial proof that heat and work are interchangeable came from James Prescott Joule — a brewer from Manchester by trade, who was optimising the machinery in the family business. In his famous experiment a falling weight turned paddles submerged in an insulated vessel of water, and Joule measured how much its temperature rose. In this way he determined the mechanical equivalent of heat at 772 ft·lbf (today's value is 778 ft·lbf) — an error of less than 0.8%. He also fought a priority dispute with the German physicist Julius Robert von Mayer, who had reached similar conclusions theoretically as early as 1842. Today the unit of energy bears Joule's name, and the popular anecdote that he measured the temperature of water at the foot of an Alpine waterfall during his honeymoon has no support in the sources.
The hardest step was extending these laws to living organisms — for a long time people believed the body obeyed separate, "vital" rules. The breakthrough came from the American chemist Wilbur Olin Atwater. Together with his collaborators, he built the first chamber calorimeter for humans at Wesleyan University: a copper chamber in which one could precisely measure the oxygen, carbon dioxide and heat given off by a person performing various activities. The results left no doubt: the first law of thermodynamics applies to a human being exactly as it does to a dead machine. The energy balance closes.
How food calories are actually counted
The energy value on the packaging is not the same as the energy released in a furnace. You have to distinguish gross energy — all the heat released when a sample is completely burned — from metabolizable energy, meaning what the body can actually use.
Gross energy is measured in a bomb calorimeter: a homogenised, dried sample is ignited by an electric spark in an atmosphere of pure oxygen, inside a steel "bomb" submerged in water, and the rise in the water's temperature is read off. The human digestive tract, however, does not work that way. Atwater noticed that unused chemical energy remains in the excreta, and that different components burn with different efficiency — so he introduced the notion of availability rather than simple digestibility.
In the end he averaged these differences into the so-called Atwater factors, which stand behind every label to this day: 4 kcal/g for carbohydrates and proteins, 9 kcal/g for fats and 7 kcal/g for alcohol. It is a useful approximation, though not a perfect one — it ignores the energy spent on digestion itself and the differences between carbohydrate sources. "Calories in equal calories out" is a sensible signpost in nutrition, but at the level of the cell it is a considerable simplification.
How much work a single doughnut holds
Take a traditional jam doughnut, about 70 grams. It delivers on average 350 kcal — made up of under 6 g of protein, over 50 g of carbohydrate and about 13 g of fat (home-baked versions drop to ~170 kcal, while heavily glazed or chocolate ones exceed 400 kcal). Let us convert those 350 kcal into "electrical" units:
E = 350 kcal × 4184 J/kcal = 1,464,400 J ≈ 1.46 MJ
E = 1,464,400 J ÷ 3600 s = 406.78 Wh ≈ 0.41 kWh
More than 400 watt-hours in one pastry sounds abstract until you turn it into how long a bulb would glow. Assuming perfect conversion of chemical to electrical energy:
| Light source | Power draw | Light output | Runtime on 1 doughnut (406.78 Wh) |
|---|---|---|---|
| Incandescent (tungsten) | 60 W | 800 lm (room) | 6.8 hrs |
| Incandescent (tungsten) | 100 W | 1600 lm (bright) | 4.1 hrs |
| LED (60 W replacement) | 10 W | 800 lm | 40.7 hrs |
| LED (100 W replacement) | 15 W | 1600 lm | 27.1 hrs |
The energy of a single doughnut would keep a bright, efficient LED (1600 lm) alive for more than 27 hours — well over a full day of light from one pastry.
The doughnut as a power bank
Portable electronics store energy in lithium-ion batteries at a nominal voltage of usually 3.7–3.8 V, which is why their capacity is given in milliamp-hours (mAh). To compare different devices you have to reduce those mAh to watt-hours, accounting for the voltage:
Wh = (mAh × V) ÷ 1000
A typical flagship smartphone (4500 mAh at 3.8 V) therefore stores about 17.1 Wh; a rugged phone with a huge 33,000 mAh battery — as much as 125.4 Wh. Let us set our 406.78 Wh from the doughnut against them, adding an electric car at the end (assuming a Tesla Model 3 consumption of 150 Wh/km):
| Device / vehicle | Energy capacity | Equivalent from 1 doughnut (406.78 Wh) |
|---|---|---|
| Flagship smartphone | 17.1 Wh | ≈ 23.8 full charges |
| Rugged phone | 125.4 Wh | ≈ 3.2 full charges |
| Office laptop | 50.0 Wh | ≈ 8.1 full charges |
| Tesla Model 3 (150 Wh/km) | — | 2.7 km driven |
| A box of 12 doughnuts (≈ 4.88 kWh) | — | 32.5 km driven |
One doughnut is nearly 24 full phone charges — but a mere 2.7 kilometres of driving. A whole box would take a Tesla around 32 kilometres. It is a good lesson in scale: the energy content of food is enormous compared with pocket electronics, and vanishingly small compared with transport.
Boiling water: who wastes the energy
Since the doughnut is now our unit of energy, let us price an everyday task in it: boiling 1.5 litres of water from 10 °C to 100 °C (ΔT = 90 K). The theoretical minimum follows directly from the specific heat of water (≈ 4184 J/(kg·K)):
E = 1.5 kg × 4184 J/(kg·K) × 90 K = 564,840 J = 135 kcal = 156.9 Wh
That is barely 0.39 of a doughnut — but only in an ideal world of 100% efficiency. In reality some heat escapes through the vessel walls, convection and evaporation, so the real cost rises the less efficient the method:
| Method (1.5 L, ΔT = 90 K) | Efficiency | Real energy use | Cost in doughnuts |
|---|---|---|---|
| Theoretical limit (100%) | 100% | 156.9 Wh | 0.39 doughnut |
| Induction hob | 83–86% | 184.6 Wh | 0.45 doughnut |
| Electric kettle | 80–90% | 196.1 Wh | 0.48 doughnut |
| Microwave oven | 43% | 364.9 Wh | 0.90 doughnut |
| Gas hob (with lid) | 27% | 581.1 Wh | 1.43 doughnut |
| Gas hob (no lid) | 16% | 980.6 Wh | 2.41 doughnut |
The conclusion is surprising: even though natural gas is often a cheaper primary energy carrier than electricity, a gas burner without a lid is only about 16% efficient — most of the heat warms the kitchen around the base of the pot rather than the water. Boiling water on it consumes, in real terms, more energy than a whole doughnut holds. The same 1.5 litres on a decent electric kettle costs less than half a pastry.
Why the human body is a poor engine
If the energy balance closes just as it does in a machine, you might expect us to be efficient engines. Not at all. The mechanical efficiency of skeletal muscle is typically 18–26%: 22–25% when cycling, 18–22% when running. As much as 75–80% of the energy released from ATP is irretrievably lost as heat — which is why exertion makes us hot.
There are illusions. In dynamic running the apparent efficiency can reach 40%, but that is thanks to the passive elasticity of tendons (the Achilles tendon acts like a spring, storing and returning energy). It can also go the other way: in eccentric work — walking down a steep slope against gravity — metabolic "efficiency" takes on negative values, as the muscles spend energy even though the external work acts against them.
Let us translate this into doughnuts, taking two real efforts: a marathon by a 70 kg runner, and a climb of Rysy (a 1500 m ascent, a hiker with a backpack, 90 kg):
| Effort | Net mechanical work | Real expenditure | Cost in doughnuts |
|---|---|---|---|
| Marathon (42.195 km, 70 kg) | — (running flat) | 2953.7 kcal | 8.44 doughnuts |
| Rysy — theoretical minimum | 316.5 kcal | 316.5 kcal | 0.90 doughnut |
| Rysy at 25% efficiency | 316.5 kcal | 1266.1 kcal | 3.62 doughnuts |
| Rysy at 20% efficiency | 316.5 kcal | 1582.6 kcal | 4.52 doughnuts |
The pure physics of climbing Rysy is plain work against gravity: E = mgh = 90 kg × 9.81 m/s² × 1500 m ≈ 1.32 MJ, that is 316.5 kcal — less than a doughnut. But because the muscles work at 20–25% efficiency, the hiker must really "burn" between 3.6 and 4.5 doughnuts — of which more than three turn into heat warming their body and surroundings. The marathon comes out even dearer: almost 3000 kcal, or more than eight doughnuts.
The same energy, different stories
A calorie on your plate, a joule in the lab and a kilowatt-hour on your bill all measure one and the same quantity — the ability to do work. The human body resembles an engine not only in its 20–25% efficiency but also in the fact that it never switches off: even at rest it draws about 100 watts just to keep its basic functions running.
The difference lies elsewhere. In a machine we can account for that energy almost down to the joule. In a living cell its release is so tangled and adaptive that the simple equation "calories in = calories out" remains a useful approximation rather than a law. But every time you look at a doughnut and think "407 watt-hours", you are holding proof that food, electricity and work are three words for the same thing.
Further reading
- Physics LibreTexts, Efficiency of the Human Body — where the 20–25% figure comes from and where the rest of the energy goes.
- Wikipedia, Atwater system — the origin of the 4/9/4 kcal factors and the difference between gross and metabolizable energy.
- PMC / NIH, Indirect Calorimetry: History, Technology, and Application — from Rubner's dogs to Atwater's chamber calorimeter.
- Encyclopædia Britannica, Mechanical equivalent of heat — Joule's experiments and the priority dispute with Mayer.
- BIPM, The International System of Units (SI Brochure), 9th ed. (2019) — the definitions of the joule and the watt in the SI system.