Sugar Does Not Really Melt, It Falls Apart

A small saucepan sits on the stove in a sunny kitchen, holding a heap of white sugar. Turn the heat on low and wait. For a while nothing happens. Then the crystals at the edges turn glassy, a clear syrup creeps across the pan, and slowly it deepens to gold and then to amber, filling the room with the smell of toffee. It looks like the most ordinary thing in cooking: sugar melting. According to food scientists at the University of Illinois, it is not melting at all, at least not the way ice melts. The sugar is falling apart.

What people often get wrong

We usually talk as if sugar were like ice or butter: warm it to the right temperature and it simply turns from solid to liquid, still the same stuff. Scientific tables even list a melting point for table sugar, which chemists call sucrose. A true melt does work like that. It happens at one steady temperature, however fast you heat, and the liquid is chemically the same as the solid. Melt ice and you get water; freeze the water and you get ice back.

Sugar does not behave like that. In 2011 the food chemist Shelly Schmidt and her graduate student Joo Won Lee reported that what looks like melting is really thermal decomposition: the heat breaks the sugar molecules into other substances, and the crystals collapse as they come apart. They gave it a new name, apparent melting, and showed that glucose and fructose, two other common sugars, do the same.

How the scientists caught it

The first clue was a number that would not sit still. Published melting points for sucrose varied widely, and scientists had long blamed impurities or differences between instruments. The Illinois team found instead that the temperature depended on how fast the sugar was heated. Heat it quickly and it seemed to melt at one temperature; heat it slowly and it gave way at another. A true melting point does not care about speed.

Then they checked what the liquid actually was. Using a method that separates and measures every substance in a sample, they looked at sucrose just as it began to turn liquid. Breakdown products, including glucose and a compound called 5-HMF, were already there. When they instead held sugar at a steady, lower temperature, the first breakdown products appeared after about 50 minutes, and the crystals went on to give way well below the temperature usually listed as sugar's melting point. Time, it turned out, could do the work of heat.

Not everyone liked the way the results were described. In 2012 a group of nine scientists published a comment on the work in the same journal, and the Illinois team answered with a long reply, so experts still debate the best way to put it. The research also added a twist of its own: by heating very fast, faster than the sugar could break down, the team managed a true melt for fructose.

Why it matters in your kitchen

For cooks, the idea is quietly useful. If caramel comes from sugar breaking down rather than from reaching one magic temperature, then time and heat work together. Schmidt pointed out that some of the new flavour compounds taste of rich caramel while others taste burnt or bitter, and that holding sugar at a lower temperature for longer could favour the pleasant ones, much as dairies heat milk to a set temperature for a set time. So the next time the pan turns amber, you are not watching sugar melt. You are watching it become something new.