Tikalon Blog by Dev Gualtieri

Fondant Physics

August 7, 2023

There's a strong connection between cooking and chemistry. Chemists supposedly make good cooks, since cooking, just like chemical synthesis, involves selection of the correct ingredients, careful measurement of their quantities, selection of the right utensils and containment vessels, and processing mixtures at the proper temperature for the proper time. analytical chemists will carefully sample the product at stages and adjust things accordingly.

Cooking likewise entails some important chemical reactions, such as the Maillard reaction. This reaction between amino acids imparts flavor when certain foods are browned. This reaction flavors such foods as breads, biscuits and French fries. Pretzels are purposely coated with lye to accelerate this reaction to give them a deep brown color.

"Do you want fries with that?"

(Photo by Evan-Amos, via Wikimedia Commons.)

Perhaps I'm biased because of my profession, but I see much more physics in a modern household kitchen than chemistry. Going beyond the simple mechanical devices, such as can openers and blenders, there are devices dependent on physical principles. Every house has a refrigerator that uses the expansion of a gas to cause cooling by the well-known thermodynamic process of free expansion. Other cooling principles have been proposed to replace this current technology. These include the elastocaloric effect and the magnetocaloric effect, as I've described in some previous articles (Elastocaloric Effect, January 6, 2020, Giant Magnetocaloric Effect, September 10, 2018, and Magnetic Refrigeration, September 3, 2014).

The Joule heating of the resistance elements of conventional ovens is a simple process that's been known since 1840, and the transfer of heat from the resistance elements involves radiation, conduction, and convection. The usual oven temperature is below 300 °C, and radiation follows a T⁴ law; so, radiative transfer of heat is inconsequential. Heat transfer is by conduction in the air, and the more efficienctt convection. More than a century after Joule's resistance heating experiments, most kitchens have an additional type of oven, the microwave oven. More than 90% of United States households, my own included, have a microwave oven.

Most foods contain a lot of water, and water is an excellent absorber of microwaves (see figure). Microwaves induce rotation of polar molecules, such as water, and the resulting thermal energy heats the food in a process known as dielectric heating. The cavity magnetron of a microwave oven is likely the sole vacuum tube now present in a house that typically contains billions of the transistors that made the vacuum tubes in consumer electronics obsolete. The frequency of microwaves for cooking, 2.45 GHz, is in the middle of the lowband Wi-Fi spectrum, 2.401 to 2.484 GHz.

Dielectric loss of water at microwave frequencies. The 2.45 GHz frequency of microwave ovens is indicated by the red line.

(Data extracted from a Wikimedia Commons image. Click for larger image.)

Just as high-tech as a microwave oven is the induction stovetop. Induction heating of pots and pans made of cast iron or ferromagnetic stainless steel is done by alternating current in the frequency range between 25-50 kHz, just above audio frequency. The low frequency electromagnetic field induced by a copper induction coil (see figure) causes generation of large eddy currents in the cooking vessel to create resistive heating. Copper or aluminum pots and pans won't heat by induction at these low frequencies, but a less efficient work-around is the placement of a disk of a suitable metal between the stovetop and the vessel.

Large copper induction coil and associated electronics for induction cooking.

The coil is made from a heavy copper ribbon, since a kilowatt of power is typically required.

(Wikimedia Commons image by Walter Dvorak.)

There's physics in such kitchen devices, but there's also physics in food preparation as well. The June, 2023, issue of Physics of Fluids is a special issue on food physics. This issue contains an article about fondant making by physicists from the Max Planck Institute for Polymer Research (Mainz, Germany), and the Technische Universität Berlin (Berlin, Germany).[1-3] The article's content is freely available under a Creative Commons Attribution License.

The scientific study of fondant making goes back at least a century with a 1919 publication by Mary Stephens Carrick in the Journal of Physical Chemistry.[4] Carrick writes that fondant is made by heating sugar in water and acid, cooling this syrup, and then beating it vigorously.[4] Carrick writes that cream of tartar (potassium acid tartrate) is preferred over vinegar and lemon juice as an acid, and that heating causes a portion of the sugar to be inverted.[4] Relatively small changes in the amount of invert sugar have a marked effect on the quality of the fondant.[4]

Fondant in the recent study was made by the reduction of hot 80-92 wt-% sucrose syrup, cooling the syrup while preventing preliminary crystallization, and then stirring the supersaturated system.[1] This stirring causes rapid nucleation and crystal growth to form a material in which crystals make up between 40-75% wt-%, the percentage being dependent on the composition and temperature.[1]

The crystals, which can be regular or irregular in shape, are ideally between 10-20 μm in diameter.[1] The physics of fondant creation are difficult to analyze, since the initial supersaturated solutions are metastable, and crystallization happens under conditions that are far from equilibrium.[1] Sucrose crystallization is inhibited by the low mobility and diffusivity of the sucrose molecules.[1] The kinetic energy of agitation, however, causes rapid nucleation and crystal growth.[1]

Fondant torque peak time as a function of sucrose content at 25 °C (top curve) and 44 °C (bottom curve).

The error bars are the standard deviation.

(Graphed using Gnumeric using data from the Supplementary Materials for Ref. 1.[2] Click for larger image.)

For proper experimental control, the research team used a laboratory kneader; that, combined with light microscopy allowed them to precisely observe the fondant creation process and develop models using parameters of agitation, temperature, and concentration.[1] Primary attention was paid to torque and its influence on crystallization.[1] Torque exhibited a characteristic minimum followed by a sharp peak during crystallization.[1] This effect is supposed to arise from an initial change in concentration of the liquid phase, followed by the formation of large crystalline conglomerates, and their eventual breakage.[1] Low torque indicates a smoother texture, and the peak torque indicates a thicker solution that's characteristic of a fondant.[3]

Says Thomas A. Vilgis, a professor at the Max Planck Institute for Polymer Research and an author of the paper,

"It was surprising to see the sugar crystals in the solution grow first during the early stages of stirring, and then the biggest get smaller again due to the stirring. Finally, their sizes adjust themselves, which leads to this fine creamy texture of fondants, provided the concentration, temperatures, and stirring speed are chosen correctly."[3]

Vilgis and co-author, Ph.D. student Hannah Hartge intend to examine the behavior of fondants created with sugar alternatives, such as erythritol and isomalt.[3]

Torque and crystallization of fondant as a function of mixing time. The torque curve is the average during the kneading process of a solution with 85 wt-% sucrose. (Image by Hannah M. Hartge, enhanced to show detail.)

Fondant Physics

References: