Can Crush Experiment: Air Pressure Is Huge

It happens faster than a blink and it is loud. A can that was sitting quietly on the stove, upended into cold water, folds in on itself with a bang. Nobody touched it. The force came from the air that is pressing on everything in the room, including your kid, all the time. That is the whole lesson of the can crush experiment air pressure demos are supposed to deliver, and most versions get the mechanism wrong. The can does not collapse because cold air shrinks. It collapses because steam turns back into water and leaves almost nothing behind. This one involves boiling water and live steam, and the heat is adult-only.

Key Takeaways

• An adult boils a splash of water in an empty aluminium can, then inverts it fast into ice water; atmospheric pressure crushes it instantly.
• It teaches that air exerts about 101 kilopascals, and that a phase change, not cooling, is what removes the gas inside.
• Best fit is ages 9 and up as observers, standing back. Nobody under about 12 handles the hot can, and the tongs stay in adult hands at any age.
• The usual failure is not boiling long enough. If air is still inside, nothing dramatic happens.
• Steam burns far worse than boiling water. The numbers are below, and they are the reason this is supervised.

What You’ll Need

Item	How much	Where to get it	If you don’t have it
Empty aluminium drinks can, 330 ml	2–3 (they’re single use)	Recycling bin	Must be aluminium; a steel food tin is too strong to collapse
Water for inside the can	15–20 ml, about a tablespoon	Tap	No substitute; too much water takes forever to boil off
Ice water	A bowl at least 15 cm deep	Freezer plus tap	Cold tap water works less well; ice matters
Long kitchen tongs	1 pair	Kitchen drawer	Locking pliers. Never oven gloves alone
Hob or camping stove	1	Kitchen	An electric hotplate. Not a microwave
Oven mitt	1	Kitchen	A folded dry tea towel
Safety glasses	1 pair each	Hardware shop	Required for everyone within 2 m
Kitchen scale and ruler	1 each	Kitchen	Optional, for the force calculation

Time: prep 5 min, experiment 15 min including boiling, cleanup 5 min Cost: everything is probably in your kitchen Mess level: medium, and do it at the sink

Do the Experiment: The Can Crush Experiment, Air Pressure in Action

Step 1 — Put a tablespoon of water in an empty can

Rinse the can and tip out any remaining drink. Add 15 to 20 ml of water, no more. Stand the can upright on the hob.

What you should see: a shallow puddle covering the bottom. If you can hear it sloshing, you have too much, and the boil will take several minutes longer than it needs to.

Step 2 — Fill the bowl with ice water and set it within arm’s reach

Deep enough that the can’s opening will go well under the surface. Put it beside the hob, not across the kitchen.

What you should see: ice floating. The transfer in Step 4 takes about one second, and you should not be walking anywhere during it.

Step 3 — Boil hard until steam has driven the air out

Adult only from here. Glasses on, everyone else at least a metre back. Heat the can until you see steam streaming from the opening, then keep heating for a further 30 to 60 seconds.

What you should see: a visible plume from the mouth of the can, steady rather than puffing. That plume is the air being pushed out and replaced by water vapour. This is the step people cut short, and cutting it short is why the demo fails.

Step 4 — Grip, flip, and plunge the mouth underwater

With the tongs around the can’s middle and an oven mitt on the other hand, lift the can, turn it upside down and push the opening about 5 cm below the water surface in one continuous movement.

What you should see: the can crumples with a bang, almost instantly, and water is drawn up inside it. It will look as though something squeezed it in a fist.

Step 5 — Run the control that proves the mechanism

Take a second can with the same splash of water. Heat it only until it is hot but before steam streams out, then flip it into the ice water exactly as before.

What you should see: a dent, a creak, perhaps a partial dimple, but no collapse. Same cooling, same can, same water. The difference is that the first can was full of steam and this one was still full of air. That comparison is the experiment.

What’s Actually Going On

At sea level the atmosphere presses on every surface with about 101.3 kilopascals, which the NOAA National Weather Service JetStream pages and NASA’s standard atmosphere model both give as the sea-level reference value. That is a little over 10 tonnes on every square metre. A 330 ml can has roughly 0.02 square metres of outer surface, so the atmosphere is already pushing on it with something like 2,000 newtons, the weight of a large motorbike. The can does not care, because an identical 101 kPa is pushing back from inside. Remove the inside pressure and the arithmetic stops being balanced.

Boiling does the removing. As water boils, vapour fills the can and pushes the original air out through the opening. After half a minute of hard boiling, the inside is mostly water vapour at roughly atmospheric pressure. Now flip it into ice water. The vapour touches cold metal and cold water and condenses, and here is the number that matters: liquid water occupies about one sixteen-hundredth of the volume of the steam it came from. The gas does not shrink, it ceases to be a gas. Interior pressure falls to the vapour pressure of water at the new temperature, which the NIST Chemistry WebBook puts at roughly 2.3 kPa at 20 °C and under 1 kPa near freezing. The outside is still at 101. The can buckles.

Now the correction, because it is the thing most explanations get wrong. People say the cold makes the air inside contract. Cooling does reduce pressure, but only in proportion to absolute temperature. Going from 373 K to 278 K cuts pressure by about a quarter, leaving roughly 75 kPa inside against 101 outside. That is a real imbalance and it will dent a can, which is exactly what your control in Step 5 shows. It will not crush one. The factor of fifty you need comes from the phase change, not the temperature change.

Two details that decide whether it works. The can must go in mouth-down, because the water seals the opening; mouth-up and air rushes straight back in to refill the void as fast as the steam condenses. And aluminium cans are built as thin pressurised cylinders, strong when inflated and extremely prone to buckling when the pressure difference reverses. A steel food tin has the same physics acting on it and simply does not deform.

How to Explain It to Your Kid

Ages 5–8: The air is leaning on everything

Watch, do not touch, from a metre away. Afterwards, hold their palm flat and press gently on it with yours: that is what air does to their hand all day, except harder, and they do not notice because it pushes from every side at once. The can had air pushing from inside too, until the steam went away.

Then let them hold the crushed can. The physical object does a lot of the work at this age. Ask them what they think squeezed it, and take the answer seriously before correcting it.

Ages 9–12: Steam takes up a lot of room, water does not

Introduce the 1,600-to-1 ratio, which is the most memorable number in the whole demo. A thimble of water makes about a litre and a half of steam. When the steam turns back to water, the room it was taking up is simply gone, and the can is what was holding that room open. Use the Step 5 control to show that cold alone is not enough, which is the part their classmates will get wrong.

Ages 13+: Put numbers on the force

Give them p = F/A and the ideal gas law together. Have them measure the can’s diameter and height, work out its lateral surface area, and compute the net inward force for an interior pressure of 2 kPa against 101 kPa outside. The answer, around 2,000 newtons, is startling when you hold the crumpled can at the same time. The extension that turns this into an investigation: repeat the control at several boiling durations, 5, 15, 30 and 60 seconds past the first visible steam, and record how completely each can collapses. You are measuring how thoroughly the air has been displaced. Pair it with Boyle’s law with a syringe to separate pressure from temperature cleanly.

The question to ask: “If we did this on top of a mountain where the air pressure is lower, would the can crush harder, less hard, or the same?”

Why Teaching This Early Pays Off

Atmospheric pressure is one of those facts children are told and do not believe, because nothing in daily experience suggests that air weighs anything. The misconception is not stupidity; it is a reasonable generalisation from a life spent in equilibrium. Smith, diSessa and Roschelle argued in 1994 that such intuitions are productive raw material rather than errors to be erased, and this demo works precisely because it takes the intuition seriously and then breaks it in one second with a sound.

The concept pays off across several later topics. Phase change and latent heat appear in chemistry; vapour pressure appears in cooking, in weather and in why water boils cooler up a mountain; pressure differences explain siphons, straws, barometers, breathing and aircraft cabins. It also seeds a habit worth more than the content: running a control. The second can is what turns a party trick into evidence, and most published versions of this demo do not include one.

Do not oversell a single afternoon. The honest claim is narrower: the National Research Council’s K-12 framework lists planning and carrying out investigations among the eight core science practices, and a controlled comparison is the simplest real example of that practice a kitchen can produce.

Before You Start: What Parents Should Know

Best age	9 and up to watch and reason. Under 12, no handling of the hot can. The adult does Steps 3 and 4 regardless of the child’s age.
Supervision	Constant and hands-on. One adult, one task, nobody crowding the hob. Clear the counter first.
Real hazards	Live steam is the serious one. Condensing steam releases about 2,257 kilojoules per kilogram, roughly eight times the heat that the same mass of boiling water gives up cooling to body temperature, which is why a steam burn is worse than a scald. Hot metal, boiling water, splashing and a loud bang. Keep faces away from the can’s mouth throughout.
Don’t	Don’t use a glass bottle, which can implode into fragments. Don’t use a sealed or capped container of any kind, and never microwave it. Don’t substitute a steel tin and keep heating in the hope it works. Don’t reheat a crushed can.

If anyone does get scalded, cool the burn under cool running water for 20 minutes and seek medical advice for anything larger than a coin or on a face, hand or joint, per the NHS guidance on burns and scalds and the American Academy of Pediatrics’ advice on treating burns.

When It Doesn’t Work

• The can only dents. Almost always insufficient boiling. Keep heating for a full 30 to 60 seconds after steam first streams out. A visible, steady plume is the signal, not the first bubble.
• Nothing happens at all. Check the water is actually boiling inside, not just warm. A can on a large burner can heat its sides without the puddle reaching 100 °C if there is too much water in it.
• It collapses weakly and slowly. Your bowl water is not cold enough, or not deep enough to seal the mouth. Add ice and use a deeper bowl.
• The can fills with water but keeps its shape. You went in mouth-up, or tilted it so the opening broke the surface. Air got back in. Go straight down, fast.
• It crushes but nobody saw it. This is common and annoying. Film it at 240 frames per second on a phone; the collapse sequence is genuinely beautiful slowed down, and it also lets you see which direction the buckling starts from.

Take It Further

Vary one thing: boiling time. Use four cans and stop heating at 5, 15, 30 and 60 seconds after steam appears. Photograph each result and rank the degree of collapse. You are plotting how completely steam has displaced air, and the curve flattens once the air is gone.

Change the coolant. Try room-temperature water, cold tap water, ice water and ice water with salt. Record the water temperature each time with a thermometer. Colder means lower final vapour pressure, so colder should crush harder, and the effect is measurable.

Try the gentle version with a plastic bottle. Pour 100 ml of just-boiled water into a 500 ml plastic bottle, swirl, pour it out, cap it immediately and let it stand. It slowly crumples over a minute or two with no bang at all. Same physics, much lower drama, and safe enough that a younger sibling can hold it while it happens.

What to Watch For Over the Next 3 Months

• Week 4: they notice a sealed snack packet puffed up on a mountain drive or a plane, and connect it to outside pressure.
• Month 2: “the air is pushing on it” turns up in a different explanation, often about straws, suction cups or a stuck lid.
• Month 3: they ask a question that requires the phase-change idea rather than the cooling idea, such as why a kettle’s lid rattles or why a pressure cooker cooks faster.

Frequently Asked Questions

Is this safe to do with a seven-year-old in the room?

Yes, from a metre back, in safety glasses, with the adult doing all the heat handling. The hazard is not the collapse; it is the hot can and the steam before it. Seven-year-olds get real value from watching and holding the crushed can afterwards.

Why does everyone online use a match or a candle instead?

Some versions burn something inside the can to consume oxygen. That works far less well, because burning replaces oxygen with carbon dioxide and water vapour rather than removing gas, and it adds an open flame for no benefit. The steam version is both safer and more dramatic.

Can we reuse the can?

No. Once the walls have buckled, the geometry that made it strong is gone, and reheating a can with water trapped in a fold is a splashing hazard. Flatten them and recycle them. The same pressure imbalance runs the show in popcorn, which is a pressure vessel.

Would this work in a vacuum chamber in reverse?

Good question, and yes: an inflated sealed can placed in a vacuum chamber bulges and can split, because the imbalance runs the other way. It is the same equation with the sign flipped, and it is why spacecraft structures are designed for internal pressure rather than external.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. National Oceanic and Atmospheric Administration, National Weather Service. “Air Pressure.” JetStream: An Online School for Weather. https://www.noaa.gov/jetstream/atmosphere/air-pressure
2. NASA Glenn Research Center. “Earth Atmosphere Model, Metric Units.” https://www.grc.nasa.gov/www/k-12/airplane/atmosmet.html
3. National Institute of Standards and Technology. “Thermophysical Properties of Fluid Systems.” NIST Chemistry WebBook. https://webbook.nist.gov/chemistry/fluid/
4. The Engineering ToolBox. “Water Properties: Vaporization Heat vs. Temperature.” https://www.engineeringtoolbox.com/water-properties-d_1573.html
5. National Health Service (UK). “Burns and scalds.” https://www.nhs.uk/conditions/burns-and-scalds/
6. American Academy of Pediatrics. “Treating and Preventing Burns.” HealthyChildren.org. https://www.healthychildren.org/English/health-issues/injuries-emergencies/Pages/Treating-and-Preventing-Burns.aspx
7. Smith, J. P., diSessa, A. A., & Roschelle, J. (1994). “Misconceptions reconceived: A constructivist analysis of knowledge in transition.” Journal of the Learning Sciences, 3(2), 115–163. doi:10.1207/s15327809jls0302_1
8. National Research Council. (2012). A Framework for K-12 Science Education. National Academies Press. doi:10.17226/13165