Most "grow your own snowflake" projects produce frost, not snowflakes. You put dry ice in a cooler, hang a string inside, wait an hour, and get a fuzzy white coating that looks like the inside of an old freezer. I've done exactly that in front of a classroom of twelve-year-olds, and nobody was impressed.
The problem is that those instructions don't create a temperature gradient. Without a warm layer at the top and a cold layer at the bottom, water vapor has no direction, no supersaturated zone forms, and you never get the six-armed dendrites everyone actually wants to see.
This guide covers the chamber layout I use now: the temperature zones, how I prepare the nylon line, what each crystal shape tells you about the conditions, and the specific mistakes that turn a crystal chamber into a frost box.
Why the "Dry Ice and a String" Method Gives You Frost Instead of Snowflakes
The common advice skips the physics that decides whether your setup works at all.
Snow crystals grow when the air is supersaturated with respect to ice, meaning it holds more water vapor than ice at that temperature can sit next to without growing. In clouds, that happens when warmer, moist air cools. A plain cold box can't do this. Everything inside settles to about the same temperature, the air dries out by depositing vapor on every cold surface, and humidity sits below ice saturation for most of the run. The walls, the lid and the string all collect a thin, disorganized crust. That's frost.
A working chamber needs three things running at the same time:
- A warm, wet vapor source at the top. Warm air holds far more water vapor than cold air, so this is your supply.
- A cold sink at the bottom. The cold pulls vapor downward by diffusion.
- A middle zone where the two meet. Vapor arriving from the warm top into the cold middle ends up well above ice saturation, and that's where dendrites grow.
Warm on top and cold on the bottom isn't arbitrary. It's a stable arrangement because cold air is denser and stays down, so you don't get convection currents mixing everything into an even, useless temperature. Flip it, with cold on top, and the chamber churns. I tried a "freezer pack on the lid" version early on and got turbulent fog plus a few lopsided needles. Stability matters more than how cold you can get.
How to Build the Vertical Temperature Gradient Chamber
You don't need lab equipment. I've built three versions, and the one that worked best was the cheapest.
Parts List
- A small insulated foam cooler, about 40 cm tall inside
- A clear plastic cylinder or a 2-liter bottle with the bottom cut off (this is the growth column)
- A thin foam or cellulose sponge, cut to fit the top of the column
- 0.1 mm clear nylon monofilament (fine fishing line works)
- A small weight for the bottom of the line, like a paperclip or a metal washer
- Dry ice pellets, roughly 1–2 kg for a 45-minute session
- A thin metal plate or baking tray lid to sit over the dry ice
- Two cheap digital probe thermometers, or one probe you can reposition
- A small seedling heat mat or warm-water bottle for the top
Assembly That Holds a Gradient
- Put the dry ice in the bottom of the cooler and lay the metal plate over it. The plate spreads the cold evenly. Without it you get cold spots and crystals growing on only one side of the line.
- Stand the clear column upright on the plate. Pack crumpled newspaper or foam around it so the walls stay insulated. Walls that get too cold grow their own frost and steal the vapor your line needs.
- Wet the sponge with warm water (around 30°C, warm to the touch but not hot) and fix it inside the top of the column.
- Thread the nylon line through the center of the sponge so it hangs straight down the middle, with the weight just above the plate.
- Rest the heat mat or warm-water bottle on top of the column to keep the sponge warm. Without it, the sponge cools within about fifteen minutes and growth stops.
- Close the cooler partway. A fully sealed cooler collects CO₂ from the subliming dry ice, so leave a gap and run the project in a normally ventilated room. Use gloves or tongs for the dry ice. It's cold enough to cause skin damage, the same reason you'd use oven mitts on a hot tray.
What Each Zone of the Chamber Should Read, and What Grows There
This is the reference I wish someone had given me before I wasted a dozen sessions. Probe each zone before you judge results. "It didn't work" almost always comes down to a missing zone.
Home-scale cloud chamber setup and crystal growth specification
| Chamber Height Zone | Temperature Profile (°C) | Relative Humidity (%) | Substrate Setup | Observed Crystal Habit After 30 Mins |
|---|---|---|---|---|
| Top layer (0–10 cm) | +25 to +30 | 100% (warm water sponge) | Suspended nylon thread top mount | Zero growth (moisture evaporation zone) |
| Middle growth zone (15–25 cm) | -12 to -15 | >110% ice supersaturation | Bare nylon filament, center | Active stellar dendrite branch formation |
| Bottom chill zone (30–40 cm) | -20 to -25 | Moderate saturation | Base dry ice plate | Small plate and column crystal deposition |
| Uncalibrated chamber (no gradient) | Uniform -10 | 80% | Unseeded cotton string | Unstructured frost crusting (no individual crystals) |
Two notes on reading the table. The bottom temperatures are air readings a few centimeters above the plate. The dry ice itself sits near -78°C, so don't touch the probe to the plate. Also, the 110% figure is supersaturation relative to ice, not liquid water. Air at -15°C can be slightly below water saturation and still far above ice saturation. That gap is what drives fast growth.
The last row is the setup most tutorials describe. Cotton string has thousands of loose fibers, and each one becomes a nucleation point. You end up with a mat of competing ice where no single crystal can collect enough vapor to grow branches.
How to Prepare the Nylon Line So Crystals Grow in Isolation
The line matters more than people assume. You want a few nucleation sites, spaced apart, not hundreds.
Bare 0.1 mm monofilament is smooth, which is exactly why it's good. Ice has trouble starting on it, so the crystals that do form have room and vapor to themselves. The catch is that a perfectly clean line sometimes grows nothing for the first ten minutes.
My fix is to seed it deliberately:
- Take a fine-cell sponge (makeup sponges work) and dampen it so it isn't dripping.
- Pinch the line lightly against the sponge at three or four points spaced 2–3 cm apart in the middle-zone section.
- Each pinch leaves microscopic water traces in the line's surface texture. When the cold reaches them, they freeze into tiny ice seeds, the same job dust particles do in a real cloud.
Spacing is the whole point. When I seeded continuously along the line, the crystals crowded each other and grew into a single feathery strip. At 2–3 cm apart, each seed had its own vapor supply, and I started seeing proper six-fold symmetry within about 20 minutes.
Leave the top section near the sponge bare. That's the warm evaporation zone and nothing grows there anyway.
How to Read Crystal Shape as a Temperature Gauge
This is the part I find most useful in practice, and generic project pages almost never mention it. The shape of a crystal tells you the temperature it grew at. Ukichiro Nakaya mapped this in the 1930s with lab-grown crystals, and Kenneth Libbrecht's later work refined it. The pattern holds up well at home.
| Growth Temperature (°C) | Typical Habit | What It Means in Your Chamber |
|---|---|---|
| 0 to -3 | Thin hexagonal plates | Zone is too warm. Check the insulation near the top |
| -3 to -8 | Needles and hollow columns | You're in a transition band. Normal just under the sponge |
| -10 to -12 | Plates, sectored plates | Close to target. Supersaturation is decent |
| -13 to -16 | Stellar dendrites, fernlike branches | Target zone with high supersaturation |
| -20 to -25 | Small plates, short columns | Bottom chill zone. Lower vapor reaches here |
Temperature controls the basic shape, plate versus column. Supersaturation controls how elaborate it gets. At -15°C with weak vapor supply you'll see a plain hexagon. Raise the supply and the same temperature produces branches. So if the middle zone reads -15°C and you're only getting plates, the thermometer isn't the issue. Your sponge has gone cold or dry.
I once spent an afternoon convinced my probe was broken because every crystal came out as a thick hexagonal plate. The real cause was a sponge I'd wetted with tap-cold water. Once I re-soaked it at about 30°C, branches showed up on the next run.
Fixing the Problems That Ruin Most Sessions
These are the failures I've hit most often, ranked by how much time they cost me.
- Frost crust over everything. The gradient collapsed, usually because the column walls got too cold. Add insulation around the column and make sure the top stays warm.
- Nothing grows for 20+ minutes. The line is too clean or the middle zone isn't cold enough yet. Seed the line and give the dry ice 10 minutes to pull the base down before you start timing.
- Fog fills the column. The sponge is too hot, so excess vapor condenses into droplets instead of feeding crystals. Bring it back down to around 30°C. Hotter isn't better.
- Crystals grow lopsided. The cold is uneven. Check that the metal plate covers the dry ice fully and the line hangs dead center.
- Crystals vanish when you look at them. Your breath and hands are warming the chamber. Pull the line out quickly and view it immediately, or photograph it through the clear wall.
How to Photograph and Log What You Grow
A crystal you don't record is just a story. I keep a simple log for each session:
- Sponge temperature at start and at 30 minutes
- Middle-zone and bottom-zone air temperatures
- Seeding spacing on the line
- Crystal habit observed at each height
A clip-on phone macro lens through the clear column wall works surprisingly well. Shoot against a dark background (black felt taped behind the column) and light from the side so the edges of the branches catch the light. Direct front lighting flattens the crystals into white smudges.
After four or five logged sessions, you'll be able to look at a crystal and estimate its growth temperature within a couple of degrees. For a science class, that's the real lesson: the crystal is a record of its own conditions.
Frequently Asked Questions
Can I make this work without dry ice?
Yes, but you'll mostly get plates and columns instead of dendrites. A household freezer typically sits around -18°C, and you can build a weaker version of the gradient by putting the column in the freezer with a warm sponge on top. The base won't get cold enough to create a strong vapor gradient, so expect simpler shapes and slower growth.
Why do my crystals look like feathers instead of six-armed stars?
Your seeds are too close together, or supersaturation is too high near the line. Feathery growth usually means crystals are merging or branching wildly on one face. Space the seed points wider apart and drop the sponge temperature a few degrees.
How long do the crystals last once they've grown?
Only seconds to a few minutes after you take them out. Room air is far below ice saturation, so the crystals start to sublimate right away. View them inside the chamber or photograph them fast.
Is cotton or wool thread ever better than nylon?
No, not for isolated crystals. Natural fibers have too many loose ends and nucleation points, so you get dense frost. Use them only if you specifically want to show students what uncontrolled deposition looks like next to a nylon line.
What does a session cost to run?
The recurring cost is the dry ice, and that's it. It sublimates whether you use it or not, so buy it the same day and plan for about a 45-minute window. The cooler, line and thermometers last for years.
Does room humidity affect results?
Barely, as long as the chamber is mostly closed. The sponge supplies nearly all the vapor inside the column. Room humidity mostly affects how fast crystals sublimate once you take them out.
What to Do Next
Build the chamber once with no seeding and run it for 30 minutes as a control. Then run it again with a seeded line and compare the two. That side-by-side comparison teaches the gradient concept better than any diagram.
After that, change one variable per session. Try a sponge at 25°C versus 30°C, or seeds spaced 1 cm versus 3 cm apart. Log the temperatures and shapes in a small table like the habit chart above. Within a week you'll have your own morphology map built from crystals you grew yourself.
0 Comments