Frost Crystal Experiment: Why Ice Grows Six-Sided

Nobody expects to grow frost in a warm kitchen. This frost crystal experiment does it in about four minutes using a soup can, crushed ice and three tablespoons of salt, and if you hand your kid a magnifying glass at the right moment they will see hexagons. Not approximate hexagons. Six-sided plates with six feathery arms, the same geometry as every snowflake that has ever fallen, growing in front of them because the water molecules have no other option. The explanation sits one level deeper than most articles go, and it is worth having.

Key Takeaways

• You pack a clean metal can with crushed ice and salt, which drops the can’s outside wall well below 0 °C, and water vapour from the room deposits straight onto it as frost.
• The six-sided shape comes from the hexagonal lattice of ordinary ice, not from the cold or from any kind of mould.
• Best for ages 7 to 13, and the magnifier is not optional. Without it this is a white can.
• What goes wrong: a humid room gives you a wet can first and crystals later, or never. Air dryness matters more than you expect.
• Ice and salt reach about -18 °C. That is cold enough to hurt bare skin held against it.

What You’ll Need for the Frost Crystal Experiment

Item	How much	Where to get it	If you don’t have it
Empty metal can, label removed	1, standard soup or bean size	Recycling. Rinse it and tape the cut rim	A stainless steel cup or a thin-walled tin mug
Crushed ice	Enough to fill the can, roughly 300 g	Freezer. Crush cubes in a sealed bag with a rolling pin	Bagged crushed ice from a shop
Table salt	3 tablespoons, about 45 g	Kitchen cupboard	Coarse or rock salt; it works slightly slower
Magnifying glass or jeweller’s loupe	1, 5x to 10x	Stationery shop, phone clip-on macro lens	A drop of water on clear plastic as a crude lens
Tray or baking sheet	1	Kitchen	A plate with a rim; this will drip salty water
Spoon	1, long	Kitchen	Any spoon
Thermometer that reads below zero	1, optional but good	Kitchen shop, or an infrared thermometer	Skip it, and describe the cold instead
Dark paper or cloth	1 sheet	Craft drawer	Any dark surface; frost shows up far better against it

Time: prep 10 min, experiment 15 min, cleanup 5 min Cost: everything is probably in your kitchen Mess level: medium, because salty meltwater gets everywhere

Do the Experiment

Step 1 — Prepare the can

Peel the label off, rinse the can, and dry the outside completely. A film of washing-up water will freeze into a smooth glaze instead of growing crystals, which is the single most common reason this fails. Tape over the cut rim so nobody runs a finger along it.

Step 2 — Crush the ice properly

Put ice cubes in a sealed freezer bag and crush them with a rolling pin until the pieces are pea-sized or smaller. Whole cubes leave air gaps, air insulates, and the can wall never gets cold enough. This step takes two minutes and makes the difference between frost and a damp can.

Step 3 — Layer ice and salt

Fill the can a third full with crushed ice, add a tablespoon of salt, stir, and repeat twice. Stirring matters: the salt has to contact the ice surfaces to work. Put the can on the tray.

Step 4 — Ask for the prediction, then wait and watch

Before anything happens, ask your kid what they think will appear on the outside of the can and why. Then watch the lower third of the can. Within one to two minutes it will go misty, and within three to five minutes a white velvet bloom spreads upward.

Step 5 — Look at it with the magnifier

This is the whole experiment. Hold the lens 2 to 3 cm from the frost against the dark paper background. Look for flat six-sided plates, needles, and branched arms that come off at 60 degrees. Have your kid draw three different shapes they can find.

Step 6 — Breathe on one side and compare

Breathe steadily on one patch for ten seconds, then compare that patch to the untouched side. Your breath carries far more water vapour than room air, so that patch will grow thicker, coarser frost. Same temperature, different supply of water. That contrast is a one-variable experiment you did by accident.

What you should see: A dry, fluffy white layer, not a wet glaze. Under 10x magnification the crystals should show straight edges meeting at 120 degrees. If you only get a smooth clear coating or beads of water, the room air is too humid or the can started damp; see the troubleshooting section. On a dry winter day indoors you can get frost 2 mm thick in ten minutes.

What’s Actually Going On

Two separate pieces of physics are running here, and separating them is the point.

The first is why the can gets so cold. Ice at 0 °C in contact with salt forms a brine, and the brine’s freezing point is far below that of pure water. The mixture therefore keeps melting, and melting absorbs energy from its surroundings, so the whole slurry drops in temperature until it reaches equilibrium. For table salt and water that floor is -21.1 °C at the right concentration; in a kitchen can you will realistically hit -10 to -18 °C. The can wall, being thin metal, follows the slurry closely.

The second is how frost arrives. This is where most explanations get sloppy. Frost is not frozen dew. The American Meteorological Society’s Glossary of Meteorology defines hoarfrost as “a deposit of interlocking ice crystals formed by direct deposition on objects,” and specifies that the surface must be below freezing and that it forms when air with a dewpoint below freezing is brought to saturation by cooling. Deposition means the water goes straight from gas to solid, skipping liquid entirely. If instead the air near your can saturates while still above 0 °C, you get liquid condensation first, and that freezes into a smooth glaze rather than crystals. Both happen in kitchens. The dryness of the room decides which one you get.

Now the six sides. A water molecule can form four hydrogen bonds arranged roughly tetrahedrally, two through its hydrogens and two through lone pairs on the oxygen. When water freezes into ordinary ice, those tetrahedral bonds tile into a lattice whose oxygen atoms sit in hexagonal rings, a structure crystallographers call ice Ih, where the h is for hexagonal. A growing crystal adds molecules fastest along six equivalent directions in that lattice, so the macroscopic shape inherits the microscopic symmetry. Six-fold, every time, on every planet.

Which kind of six-sided shape you get depends on temperature and on how much water vapour is available. Kenneth Libbrecht’s review in Reports on Progress in Physics lays out the morphology diagram: thin plates near -2 °C, slender columns and needles near -5 °C, and elaborate branched plates again near -15 °C, with branching increasing as supersaturation rises. Your can has a temperature gradient from bottom to top, so a careful look often shows different habits at different heights on the same can. That is a genuinely impressive observation for a kitchen.

How to Explain It to Your Kid

Ages 5–8: water molecules holding hands

Have six of their fingers or six pencils meet at a point to make a star. Tell them water molecules are shaped so that each one can hold hands with four others, and when they get cold enough to stop sliding around, they always stack into rings of six. The shape of the frost is the shape of the stack. Then let them look through the magnifier and count arms.

Ages 9–12: the word is deposition

Introduce three words they already half know: melting, freezing, and the new one, deposition, gas straight to solid without a liquid stage. Then give them the number: the ice and salt mix reaches about -18 °C, which is colder than a home freezer. Ask them to predict which part of the can grows frost first and check whether they were right.

Ages 13+: lattice symmetry and crystal habit

Give them ice Ih properly: tetrahedral hydrogen bonding, hexagonal rings of oxygen atoms, and growth rates that differ between the basal face and the prism faces. Then introduce habit as a function of temperature and supersaturation, and have them design a way to test it, for example by varying the salt fraction to change the surface temperature and photographing the crystals each time. That is close to what research groups actually do with diffusion chambers.

The question to ask: “If frost always comes out six-sided, why is every snowflake different?”

Why Teaching This Early Pays Off

This experiment teaches that macroscopic shape can be a direct readout of invisible structure, which is one of the most productive ideas in chemistry and materials science. A kid who has seen hexagonal ice has a concrete anchor for the lattice diagrams that arrive in year nine and for why graphite, snowflakes and honeycomb all land on hexagons for different reasons.

The research base for this kind of activity is worth stating accurately. Alfieri and colleagues’ meta-analysis found that unassisted discovery learning does not reliably help, while guided discovery, where an adult explains alongside and feeds back, does. In practice that means the magnifier plus your explanation, not the can alone. The National Research Council’s Framework for K-12 Science Education would classify what you are doing here under “constructing explanations” and “using models,” and the honest version of the claim is that one afternoon gives a hook, not a transformation.

One more specific payoff: you have pre-loaded a correct answer to a question that gets answered wrongly all the time. Frost is deposition. Dew that froze is something else. That distinction will still be useful to them in a geography exam ten years from now.

Before You Start: What Parents Should Know

Best age	7 and up. Younger children enjoy it but will not sustain attention through the magnifier stage, which is where the learning is.
Supervision	Adult required. The adult handles the can, checks the rim, and manages the ice crushing.
Real hazards	The ice and salt slurry reaches roughly -18 °C. Skin held against the can for more than a few seconds can get a cold burn. Keep fingers off it or use a cloth. Cut can rims are sharp.
Don’t	Do not substitute dry ice to get colder. Dry ice sublimates at -78 °C, causes rapid tissue damage, and asphyxiates in enclosed spaces. The salt route is cold enough.

When It Doesn’t Work

• The can just gets wet and stays wet. Room air is too humid, so vapour condenses as liquid before the surface can collect it as ice. Try again in a heated room in winter, or in an air-conditioned room, both of which have low absolute humidity.
• You get a clear glaze instead of crystals. The can was damp when you started, or liquid condensed first and froze. Dry the can thoroughly, and add the salt in layers so the surface cools fast rather than slowly.
• Frost appears but no visible shapes. Almost always the lens. A 3x reading magnifier is not enough. A 10x loupe or a phone clip-on macro lens transforms this experiment, and cheap ones cost very little.
• It works for thirty seconds then melts. The slurry has warmed because the ice was in large pieces or you ran out of salt. Pour off the excess water, add fresh crushed ice and another tablespoon of salt.
• The frost is only at the bottom. That is correct, not a fault. The slurry settles, so the lower wall is coldest. Use it: the gradient is why different crystal shapes appear at different heights.

Take It Further

Run the control. Prepare a second identical can with crushed ice and no salt. It sits at 0 °C, so it will get wet but should grow little or no frost. Two cans side by side turn a demonstration into an experiment with one variable changed.

Measure, then vary. With an infrared thermometer, record the can’s surface temperature at 1, 5 and 10 minutes for three different salt amounts, and photograph the crystals each time through the loupe. Plot temperature against which crystal habit dominates and compare your results against the published morphology diagram. This is the version that wins a science fair.

Connect the cold. The same freezing-point depression powers making ice cream in a bag, and the same crystal-growth ideas show up in growing borax crystal snowflakes overnight and growing sugar crystals into rock candy.

What to Watch For Over the Next 3 Months

• Week 4: They notice the frost pattern on a car windscreen or a freezer wall and look closely instead of walking past.
• Month 2: The word “crystal” starts being used for something specific, a repeating arrangement, rather than meaning “shiny.”
• Month 3: They ask why salt melts ice on roads, or why a snowflake is symmetrical if its two arms never touched. Both questions are deep and both deserve a proper answer.

Frequently Asked Questions

Why does salt make ice colder rather than just melting it?

Both happen. Salt lowers the freezing point of the water, so the ice starts melting even below 0 °C, and melting absorbs energy from the surroundings. With nothing adding heat fast enough, the slurry’s own temperature falls. The cold is a by-product of forced melting.

Can we do this in summer in a humid climate?

Probably not well. High absolute humidity means liquid condensation wins and you get a wet can. An air-conditioned room is your best shot, because cooling the air to run the air conditioner removes most of its water.

Is the salty meltwater a problem?

For your worktop and for plants, a little. Work over a tray and tip the waste down a drain with plenty of water. Do not pour it on the lawn or in a flower bed, because salt accumulates in soil and damages roots.

Are real snowflakes really all different?

Effectively yes, and that is a probability result rather than a law. Each arm’s fine detail depends on the exact path of temperature and humidity the crystal travelled, and a large crystal has an enormous number of molecules arranged along that path. The six-fold symmetry comes from the lattice; the uniqueness comes from the journey.

Our crystals look like needles, not stars. Did we do it wrong?

No, you found the morphology diagram the hard way. Needles and columns are the normal habit near -5 °C, with plates and stars appearing at other temperatures. Different parts of your can are at different temperatures, so check the top as well as the bottom.

---

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. American Meteorological Society. “Hoarfrost.” Glossary of Meteorology. https://glossary.ametsoc.org/wiki/Hoarfrost
2. Libbrecht, K. G. (2005). “The physics of snow crystals.” Reports on Progress in Physics, 68(4), 855–895. https://doi.org/10.1088/0034-4885/68/4/R03
3. National Snow and Ice Data Center. “Science of Snow.” https://nsidc.org/learn/parts-cryosphere/snow/science-snow
4. Alfieri, L., Brooks, P. J., Aldrich, N. J., & Tenenbaum, H. R. (2011). “Does discovery-based instruction enhance learning?” Journal of Educational Psychology, 103(1), 1–18. https://doi.org/10.1037/a0021017
5. National Research Council. (2012). A Framework for K-12 Science Education: Practices, Crosscutting Concepts, and Core Ideas. National Academies Press. https://doi.org/10.17226/13165
6. White, R., & Gunstone, R. (1992). Probing Understanding. London: Falmer Press.