Caramel in Five Stages: Sugar, Heat and Chemistry

A candy thermometer is not measuring temperature in the way a child assumes. It is measuring how much water is left. That is the idea at the centre of this caramelisation experiment kids can genuinely follow: one pan of sugar syrup, five samples drawn at five temperatures, and five completely different materials in the bowl at the end. Soft thread, chewy ball, shatterable glass, amber caramel, bitter black. Same sugar every time. Read first, though, because this is the one article in our kitchen series with a real injury risk, and it is not the heat you expect.

Key Takeaways

• You will cook one pan of sugar syrup from 110 °C to 175 °C, pulling a small sample into cold water at five stages, and compare the five results.
• The concept is boiling point elevation and then caramelisation: temperature tracks concentration until the water runs out, after which the sugar starts breaking apart.
• Ages 10 to 15 as observers. Nobody under 14 touches the pan. An adult does all heating, sampling and pouring, with no exceptions.
• Hot sugar syrup causes worse burns than boiling water because it sticks. At 175 °C it is 75 degrees above boiling and it will not shake off.
• The cooled samples are edible candy, made of nothing but sugar and water in clean kitchen gear.

What You’ll Need

Item	How much	Where to get it	If you don’t have it
Granulated white sugar	200 g	Supermarket	Caster sugar; do not use brown sugar, honey or panela, which brown before the stages separate
Water	60 mL	Tap	—
Heavy-based saucepan	1, small	Kitchen	Any thick pan; a thin pan scorches in patches and ruins the comparison
Candy or instant-read thermometer reading to 200 °C	1	Supermarket, hardware shop	The cold-water test alone, which is the traditional method and works
Metal teaspoon	1	Kitchen	Never plastic, which softens and fails at these temperatures
Bowl of cold water	1 large	Kitchen	Refill between samples so it stays cold
Five small plates or a muffin tray	1	Kitchen	Baking paper on a tray, labelled
Oven gloves	1 pair	Kitchen	A folded dry dish towel, never a damp one
Lemon juice	5 mL, optional	Fridge	Prevents the syrup recrystallising on the pan sides

Time: prep 10 min, cooking 20 to 30 min, cooling 10 min, cleanup 20 min Cost: everything is probably in your kitchen except the thermometer Mess level: medium, and the cleanup is slow. Dried sugar comes off with a pan of simmering water, never with scrubbing.

Do the Experiment

Step 1 — Set up before any heat goes on

Fill the bowl with cold water and put it beside the stove. Lay out the five plates, labelled 1 to 5. Put on the oven gloves. Tie back hair, wear closed shoes, and move children to the far side of the kitchen, standing, with a clear view and nothing in their hands.

Step 2 — Dissolve, then stop stirring

Put the sugar, water and optional lemon juice in the pan over medium heat and stir only until the sugar has completely dissolved and the liquid is clear. Then stop stirring entirely. Stirring a boiling syrup throws crystals onto the pan walls, and they seed the whole batch.

Step 3 — Pull sample one at the thread stage, 110 to 112 °C

Let it boil. When the thermometer reads 110 °C, the adult lifts out a quarter teaspoon and drizzles it into the cold water. The child records the temperature and describes what happens. Return the spoon to a safe spot, never to the pan rim.

Step 4 — Pull samples two, three and four

Repeat at 114 °C for soft ball, 125 °C for hard ball, and 150 °C for hard crack. The intervals get faster as you go: 110 to 125 takes several minutes, 125 to 150 takes about one, and after 150 the temperature climbs very quickly indeed.

Step 5 — Watch the colour at 165 to 175 °C, then stop

Past 160 °C the syrup starts to turn straw, then gold, then amber. Pull sample five at about 170 °C and immediately take the pan off the heat, because residual heat keeps it climbing. If you want the sixth, burnt sample for comparison, know that 185 °C smells acrid and sets off smoke alarms.

What you should see: At 110 °C the syrup falls off the spoon in a thin thread and in cold water it stays soft and sticky. At 114 °C a sample squeezed underwater forms a soft ball you can flatten between two fingers. At 125 °C that ball is firm and chewy, the texture of a caramel sweet. At 150 °C it hits the water with a crack and shatters like glass. From 160 °C the colour arrives suddenly: straw at 160, gold at 165, proper amber at 172, and dark enough to taste bitter by 180. The colour change takes seconds, not minutes, which catches almost everyone out the first time.

What’s Actually Going On in This Caramelisation Experiment

For the first four stages, nothing is happening to the sugar at all. You are boiling water off. A sugar solution boils above 100 °C because dissolved sugar lowers the escaping tendency of the water, and the more concentrated the syrup gets the higher that boiling point rises. So the thermometer reading maps directly onto concentration. Harold McGee’s standard reference on kitchen science lists the correspondence: thread stage is roughly 80 percent sugar, soft ball about 85, hard ball about 90, and hard crack around 99 percent. The thermometer is a concentration meter wearing a disguise.

That is why the five samples behave so differently. A sample at 85 percent sugar still has enough water to stay plastic and chewy. One at 99 percent is essentially pure amorphous sugar, a glass, and glasses shatter. Nothing chemical changed between those two. Only the water content.

Past about 160 °C there is effectively no water left, and now the sugar itself starts to break down. Caramelisation, which Lothar Kroh reviewed in 1994 in Food Chemistry, is a cascade of dehydration, fragmentation and polymerisation reactions. Sucrose splits into glucose and fructose, those lose water, the fragments recombine, and you get hundreds of volatile compounds along with the large brown polymers that give caramel its colour. Note what is absent: no protein is required. That is the clean dividing line between caramelisation and the Maillard browning that makes toast brown, which does need an amino group.

Here is where a lot of sources are wrong. Sugar is routinely described as melting at 186 °C. Joo Won Lee and colleagues showed in 2011, in the Journal of Agricultural and Food Chemistry, that what looks like melting in sucrose is actually thermal decomposition: the apparent melting point shifts depending on how fast you heat the sample, which a true melting point does not do. So the honest statement is that sucrose does not melt. It comes apart, and the liquid you see is a mixture of its decomposition products. If your child’s textbook gives a melting point for sugar, that is a genuinely good thing to question.

One practical piece of chemistry is worth adding because it explains the lemon juice. Pure sucrose syrups are prone to recrystallising into a grainy mess. A small amount of acid inverts some of the sucrose into glucose and fructose, and those mismatched molecules cannot pack neatly into a sucrose crystal lattice, so crystallisation is suppressed. This is the same reason commercial recipes use corn syrup, and it is the reverse of what you want when growing rock candy, where crystals are the goal.

How to Explain It to Your Kid

Ages 5-8: Not recommended at the stove

Honestly, this one is for watching from a distance with young children, and the right framing is simple: the pot is getting rid of the water, and the less water there is, the harder the sugar gets when it cools. Let them handle the cooled samples once an adult says they are safe. Ranking five candies from softest to hardest is a real observation task and plenty for this age.

Ages 9-12: The thermometer tells you how wet it is

Give them the table and the job of recording it. At each stage they write the temperature, then the cold-water result in their own words. Then reveal the concentration numbers and ask them to match each description to a percentage. This connects a number they measured to a property they felt, which is the core move of quantitative science.

Ages 13+: Two different kinds of change

Draw the line explicitly. Stages one to four are a physical change, the removal of water, fully reversible by adding water back. Stage five is a chemical change: new molecules exist that did not before, and no amount of water brings the sucrose back. Have them test that claim by dissolving a hard-crack sample and an amber caramel sample in equal hot water. One goes clear. One stays brown.

The question to ask: “If I added a tablespoon of water to the pan at 150 °C, would the temperature go up or down, and what would happen to the sample I pulled next?”

Why Teaching This Early Pays Off

This experiment teaches the difference between physical and chemical change better than any worksheet, because the dividing line is a temperature you can watch the needle cross. That distinction is on every science syllabus for ages 11 to 14 and is one of the most commonly muddled ideas in school chemistry. Having a memory of the exact moment the syrup stopped being wet sugar and started being something else is worth a lot.

It also teaches instrument literacy, which is underrated. A thermometer that is secretly measuring concentration is a perfect small example of how most measurements work: you measure what is easy and infer what you want. Thermometers reading concentration, voltages reading temperature, pressures reading altitude. Once a kid has met one proxy measurement knowingly, the idea transfers.

The honest limit: this is a demonstration with a strong observation component, not an open investigation, and that is deliberate given the hazard. The National Research Council’s 2012 framework lists “obtaining, evaluating and communicating information” among its eight practices, and that is the one this trains, along with careful recording. Open-ended experimentation with 170 °C sugar is not something to encourage.

Before You Start: What Parents Should Know

Best age	10 to 15 as recorders and observers. Below 10 the wait and the hazard both argue against it. Nobody under 14 should handle the pan or the spoon, and even then an adult pours.
Supervision	Total. The adult holds the pan, the spoon and the thermometer for the entire session, and never turns away from the stove.
Real hazards	Sugar syrup at 150 to 175 °C is the serious one. It is 50 to 75 degrees hotter than boiling water, it adheres to skin, and it keeps delivering heat after contact, so burns are deeper than a water scald. The American Burn Association identifies hot liquids as a leading cause of childhood burns. Steam from the pan, a hot pan handle and a hot thermometer are secondary risks.
Don’t	Never add water to hot caramel. It flashes to steam instantly and throws syrup out of the pan. Do not use a plastic spoon or bowl for sampling. Do not substitute honey, syrup or brown sugar, because they brown before the stages separate and you lose the comparison entirely.

If someone gets burnt: hold the area under cool running water for 20 minutes, do not pick off stuck sugar, and get medical advice for anything larger than a coin or on a face or hand.

On eating it: yes, once cool. The samples are nothing but sugar and water, handled with clean kitchen utensils, so they are food. Let each sample sit in the cold water for a full minute and have the adult confirm it is cool before anyone picks it up. The hard-crack and caramel stages are genuinely good. The burnt sample is bitter, and a crumb is enough.

When It Doesn’t Work

• The syrup turned grainy and white partway up. Crystallisation, triggered by stirring after boiling began or by crystals on the pan wall. Add 5 mL of lemon juice next time, stop stirring once dissolved, and brush the pan sides with a wet pastry brush in the first minute.
• It went from gold to black in ten seconds. Normal and the most common surprise. Caramelisation accelerates sharply with temperature, and the pan keeps heating after you remove it. Pull the pan at 170 °C, not 175, and sit it in a shallow tray of cold water to stop it.
• The temperature stalled around 105 °C and would not climb. You have too much water, which simply needs more time, or your thermometer tip is touching the pan base rather than sitting in the liquid. The second is far more common.
• Samples all felt about the same. The water in the bowl was too warm after two or three samples, so later ones never set. Refresh it each time.
• Burnt black patches with pale syrup elsewhere. Thin pan, uneven hot spots. A heavy base is not a luxury here, it is the difference between a result and a mess.

Take It Further: A Caramelisation Experiment Kids Can Extend

Plot temperature against time. Record the thermometer every 30 seconds from the moment it boils. The curve is almost flat early, then steepens dramatically past 140 °C. That shape is the story of the water running out, and it makes a genuinely interesting graph rather than a decorative one.

Measure the colour against temperature. Pull samples every 3 °C from 158 to 178 and line them up on white paper. Photograph the row in consistent light. You now have a calibration strip for caramel, which is exactly how commercial caramel colour is graded, and it shows that the reaction rate roughly doubles every few degrees.

What to Watch For Over the Next 3 Months

• Week 4: They notice a caramel sauce going darker and comment on the temperature rather than the time.
• Month 2: The phrase “physical change” or “chemical change” gets used correctly in a situation you did not set up, often about melting versus burning.
• Month 3: They question a measurement they are given, asking what the instrument is really detecting. That is the transferable skill, and it is worth more than the candy.

Frequently Asked Questions

What temperature is caramel, exactly?

Light caramel starts around 160 °C, a good amber sits near 170 to 175 °C, and past about 185 °C it is burnt and bitter. The usable window is narrow, which is why colour matters more than the clock.

Can kids do this experiment at all?

Yes, as recorders and observers, from the other side of the kitchen. They read the thermometer, write the results and handle the cooled samples. An adult handles everything hot, every time. There is no version where a child under 14 holds the pan.

Is the candy safe to eat afterwards?

Yes. It is sugar and water, cooked in a clean pan with a clean metal spoon. Let each sample cool fully in the cold water first and have an adult check it. The burnt sample is edible but unpleasant.

Why did my sugar turn into a crystallised lump instead of caramel?

A stray crystal seeded the batch, usually from stirring after it started boiling or from sugar stuck to the pan wall. Dissolve it again with a little added water before the pan gets hot, add lemon juice, and leave it alone once it boils.

Do I really need a thermometer?

No, and the cold-water test is how confectioners worked for centuries. It is also more informative for a child, because they feel the stage instead of reading it. A thermometer makes the lesson about concentration much clearer, so use both if you can.

---

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. Kroh, L. W. (1994). “Caramelisation in food and beverages.” Food Chemistry, 51(4), 373-379. https://doi.org/10.1016/0308-8146(94)90188-0
2. Lee, J. W., Thomas, L. C., Jerrell, J., Feng, H., Cadwallader, K. R., & Schmidt, S. J. (2011). “Investigation of thermal decomposition as the kinetic process that causes the loss of crystalline structure in sucrose.” Journal of Agricultural and Food Chemistry, 59(2), 702-712. https://doi.org/10.1021/jf104235d
3. Mathlouthi, M., & Reiser, P. (eds.) (1995). Sucrose: Properties and Applications. Springer US. https://doi.org/10.1007/978-1-4615-2676-6
4. National Center for Biotechnology Information. “Sucrose, CID 5988.” PubChem Compound Summary. https://pubchem.ncbi.nlm.nih.gov/compound/5988
5. McGee, H. (2004). On Food and Cooking: The Science and Lore of the Kitchen, 2nd edition. Scribner.
6. American Burn Association. “Prevention Resources.” https://ameriburn.org/prevention/
7. National Research Council. (2012). A Framework for K-12 Science Education. National Academies Press. https://doi.org/10.17226/13165