Solving Moisture Migration Using Water Activity
A fruitcake manufacturer wants to predict conditions over time in her fruitcake. She doesn't want the cake to get soggier as the fruit pieces get dry and hard. So she measures moisture content. The cake contains 30% water, while the fruit pieces have 50% water. She knows that water wants to come to equilibrium. Therefore, she assumes that the water will migrate from the wetter ingredient (the fruit) into the drier ingredient (the cake).
As she keeps her eye on moisture content rather that water activity, she'll suddenly be confronted with a surprise ending − dried up cake and soggy fruit, because the cake had a higher water activity than the fruit pieces at these moisture contents.
Why the soggy fruit?
In this case, water content is nothing more than a distraction. True, unless acted on by other forces, water does want to come to
equilibrium. But equilibrium occurs when the partial specific Gibbs free energy is the same everywhere in a system. Water activity is a
measurement of Gibbs free energy. Water content has nothing to do with the energy of the water.
To understand this concept better, think of two tanks of water. One is almost full at 10,000 gallons. The other huge tank is nearly empty, containing only one gallon. Which way will the water move? Knowing the water content of the tanks is completely misleading. The volume of water is irrelevant. Water moves from higher pressure to lower pressure, not from full to empty. If we raise the pressure of the nearly empty tank by putting it on top of the almost full tank, that last gallon of water will quickly leave its spacious tank.
Likewise, water activity − not water content-predicts how the water will migrate within a product.
The fruitcake manufacturer can develop a recipe in which cake and fruit pieces both have the same water activity. No moisture surprises
when this cake is stored and sold: it is a safe, palatable, and shelf-stable product.