COOKING IS ALL PHYSICS (including, of course, chemistry), but most people (including me) cook largely on the basis of instinct guided by experience. Still, it is great fun when physical principles can be enlisted to make cooking better or easier. Recently I have encountered several kitchen contrivances based on principles that are not obvious. Among them are a heat pipe that speeds the cooking of meat, a liquid crystal that monitors the boiling of an egg, a butter dish that keeps butter cool in a warm place and an Oriental pot that makes it possible to cook with steam even if the food is partly liquid.

A heat pipe designed for the kitchen is a closed hollow tube, usually pointed at one end so that it can pierce meat. It contains a small amount of fluid that transports heat from the oven to the cool interior of the roast. Since the transfer is faster than the rate at which heat is conducted through the meat, the cooking time is reduced, diminishing the shrinkage caused by prolonged cooking.

The outer end of my favorite heat pipe is embedded in a solid metal cylinder. I have bought a dozen more heat pipes from Jerryco, Inc. (601 Linden Place, Evanston, III. 60202), that are longer and have smaller cylinders at the end. Heat pipes made specifically for cooking are marketed by Thermo Pin Manufacturing Corporation (67 Sheer Plaza, Plainview, N.Y. 11803). These pipes have heat fins. The cylinders and the fins are intended to absorb heat from an oven so that the fluid inside the pipe is vaporized. As the vapor spreads through the pipe some of it condenses in the part of the pipe surrounded by the relatively cool interior of the meat, releasing heat to the pipe wall and the surrounding meat.

Normally the pipe is put in the meat at an upward angle so that the condensed fluid is returned to the bottom of the pipe by gravity. To assist the process the pipe usually has an inner lining of porous material that draws the liquid by capillary action. At the bottom the liquid vaporizes and the cycle begins again.

When the upper section of the pipe gets warm, the condensation diminishes and the cycle slows. When the interior of the meat is almost as hot as the oven, the circulation through the pipe stops.

The liquid in such a pipe can be water, methanol or something else that requires a large amount of heat to vaporize. Consider a pipe containing a gram of water at room temperature (say 19 degrees Celsius). To heat the water to the normal vaporization temperature of 100 degrees C. requires about 340 joules of energy, which is not much compared with the amount of heat needed to vaporize the water (2,256 joules). That much energy is needed to free the molecules from one another so that they can form a gas.

The energy needed to vaporize water is often called the latent heat of vaporization. This is the energy the heat pipe conveys to speed the cooking of meat. Every time the water condenses on a cool section of the pipe it loses its latent heat as the forces between the molecules re-form the liquid. The heat released by the water is conducted through the wall of the pipe to the surrounding meat. If one gram of water deposits its latent heat every second, 2,256 joules per second are released into the meat.

For a comparison I calculated the amount of heat that would be conducted into the meat by a solid rod made of an aluminum alloy. A rod with the same diameter as my heat pipe would deposit only 35 joules per second,- considerably less than the heat pipe.

I ran two trials. One night I cooked a small (2.22 pounds) sirloin roast. The heat pipe was my old one with the solid cylinder on the end. It passed through the meat at an upward angle, with the pointed end sticking out.

With a nail I made a hole in the to of the meat so that I could install a heat-detecting thermocouple. The hole which was snug for the wires of the thermocouple, was 3.5 centimeters deep The temperature was measured at about two centimeters from the pipe, which at that point was 2.2 centimeters from the top of the roast and about midway across its length and width. I monitored the temperature of the thermocouple with a thermocouple thermometer from the Cole-Parmer Instrument Company (7425 North Oak Park Avenue, Chicago, IL 60648). The thermocouple probe was an unmounted Type K.

I put the roast on the middle rack of my electric oven, which had been preheated to 400 degrees Fahrenheit (204 degrees C.). The temperature measured at the tip of the thermocouple rose by 25 Celsius degrees in the first 25 minutes. It increased by a total of 58 degrees in 44.5 minutes.

At that point I removed the roast from the oven and sliced it into several vertical sections. One clue to the extent of cooking could be seen in the color of the meat in the sections. The slice through the middle was brown for only a few millimeters below the top of the roast but for two centimeters above the bottom. A circular brown region about two centimeters in diameter surrounded the hole made by the heat pipe. The rest of 5 the slice was still red. (At this point I should state in passing that this expensive experimental material did not go to waste. Clearly the situation was not conducive to serving it, but it found its way into some very fine dishes that need not be described here.)

A slice near the point of entry of the heat pipe was browner around the hole made by the heat pipe. A slice at the other end of the roast was less brown around the hole. Thus the transfer of heat into the roast is more efficient near the lower end of the pipe than it is at the upper end, presumably because the liquid condensing at the lower end has a shorter distance to travel to return to the bottom and so returns quickly.

The next night I cooked without the heat pipe a similar roast weighing 2.18 pounds. I buried the thermocouple at about the same place and kept the cooking procedures as nearly the same as I could make them. This time the temperature at the thermocouple rose only 12 degrees in the first 25 minutes of cooking. An increase of 58 degrees took one hour and nine minutes, about 50 percent longer than it had with the heat pipe. (This time we ate the roast. Incidentally, in case the reader is wondering if a heat pipe tends to make a roast well done throughout, the answer is yes. This may sound terrible for a beef roast but is of course an advantage for a pork roast. Each to his own taste.)

Thermo Pin also sells a device for baking potatoes. The one I bought has six vertical heat pipes about nine centimeters long. Potatoes are speared on the pipes and the rack is placed in the oven. Since potatoes conduct heat poorly, I have often put nails in them, thinking that the rapid heat conduction by the metal would speed up the baking. I thought a heat pipe would transfer heat even faster.

To check these assumptions I did an experiment with five potatoes of about the same size (11 by six by five centimeters). All were placed close to one another on the middle rack of an oven preheated to 400 degrees F. I pierced the sides of each potato with a nail so that the thermocouple could record the temperature at a depth of two centimeters. One potato was horizontal, one was vertical, one was pierced with two aluminum nails intended for baking potatoes and one was speared on a vertical heat pipe. The fifth potato was pierced with my old heat pipe and placed horizontally on the rack. (Because it was too large for the potato, the pipe ripped open a small part of one end.)

After 10 minutes I checked the internal temperature of each potato. The one with the large heat pipe had warmed by 28 Celsius degrees, the horizontal potato by 23, the potato on the vertical heat pipe by 19 and the other two by 17. After 20 minutes the potato with the large heat pipe was still in the lead but the others were closely clustered in temperature. The vertical potato remained the coolest.

I continued the race until the potatoes reached approximately 100 degrees. The potato with the large heat pipe won, taking about 55 minutes. The other four took about 74 minutes to finish.

The baking nails did not decrease the time of baking. Although they conduct heat rapidly, they do not bring much into the potato. A potato cooks faster with a heat pipe but is at a disadvantage if it is vertical. Then it presents a small cross section to the heating coils at the bottom of the oven and intercepts asmall amount of the direct infrared radiation. The heat delivered by the heat pipe might be offset by this loss in absorption of direct radiation. One might do better by placing the potato horizontally over the coils. My conclusions are tentative. If you try this experiment, I would enjoy hearing about your results. (We sacrificed the potatoes.)

Until recently my method of timing eggs as they boiled was guesswork. I was never sure what state the eggs had reached. A new device neatly solves the problem. This clever egg timer is sold by the Wahl Company (5750 Hannum Avenue, Culver City, Calif. 90230) under the trademark of Eggrite. I bought one from a mail-order company (Williams-Sonoma, P.O. Box 7456, San Francisco, Calif. 94120). The Eggrite consists mostly of transparent plastic. Horizontally through the middle is a thin red film marked with a scale reading "soft," "medium" and "hard."

The Eggrite is put in the water along with the eggs. As the temperature rises the red film in the Eggrite begins to change color from bright to dark red. The color change appears first at the edge of the film. As heat is conducted through the plastic more of the film darkens. Soon the color change reaches the label "soft," indicating that the eggs are soft-boiled. Further heating drives the color change toward the center of the film, passing "medium" and "hard." With some experimentation one can calibrate the scale to get eggs with any degree of firmness. One can also calibrate the scale in terms of the size of the eggs.

The Eggrite is as sensitive as the eggs are to the initial temperature of the water, the number and size of the eggs in the pan and the rate at which the water is heated. Suppose I decrease the rate of heating by turning down the heat or by adding eggs. The conduction of heat into the eggs and the Eggrite is slowed and the color change is slower.

The Eggrite even allows for a change in air pressure. Suppose I carry an Eggrite from sea level to a high altitude. Since air pressure decreases with altitude, the boiling temperature of water is lower. An egg conducts less heat into its interior each second, hence taking longer to cook. The same thing happens with the Eggrite.

To investigate the film I cracked open an Eggrite with a small sledgehammer. (I was careful to protect my eyes from pieces sent flying.) The film consists of two layers. The top layer is a clear flexible plastic. The bottom layer is a spongy red material that appears to have been painted on the top layer. I dipped the film into water with tongs and heated the water while measuring its temperature with a thermocouple.

When the water near the film reached about 68 degrees, the layer quickly turned from bright red to dark red, maintaining the dark hue up to the boiling point of water. I lifted the film from the water and touched one edge. The cooling effect of my finger immediately changed the edge back to bright red. Cool water restored all the bright red.

I believe the spongy bottom layer of the film consists of a liquid crystal, probably of the cholesteric type. In such a substance the molecules have a certain type of crystalline order even though the substance is fluid. My guess is that the liquid crystal in the Eggrite strongly reflects red light until it reaches approximately 68 degrees. At higher temperatures it absorbs most of the light, becoming dark red.

A cholesteric liquid crystal has a layered structure in which rodlike molecules lie with their long axes in the plane of each layer. The orientation of the molecules gradually shifts from layer to layer. If a vector is made to point in the direction of the long axes of the molecules for each layer, the vector rotates through a helical path as it travels through the layers.

The crystal can be thought of as consisting of uniformly spaced planes that reflect light from the environment. The separation between the planes is the distance one must bring the pointing vector up through the layers so that it turns through 180 degrees. Since the molecules are considered to be rods with indistinguishable ends, those two planes have identical molecular alignments.

Suppose white light shines on such an arrangement of uniformly spaced planes. Light reflecting from one layer interferes with the light reflecting from another layer. For most wavelengths of light these reflected rays interfere destructively to yield darkness or at best a dim color. For other wavelengths the rays interfere constructively to yield a bright color that corresponds to those wavelengths.

Often cholesteric liquid crystals are noticeably fluid and iridescent. The colors from different sections of the fluid depend on the orientations of the crystal planes in those sections. Warming the sample changes the distance between planes and thus the color that is most strongly returned to the observer. If the temperature is high enough, the molecular vibrations disrupt the structure of the crystal, and the fluid loses its ability to reflect strongly at certain colors.

In the Eggrite I believe a liquid crystal has been mixed with some other substances and then painted on the flexible plastic layer. The mixture is not iridescent, but when it is illuminated with white light, it does strongly return red light to the observer. At about 68 degrees the selective return of red light is lost. Instead most of the incident light is absorbed, leaving the film dark red. As soon as the film is cooled the ordered structure and the selective reflection of red are reinstated.

The terra-cotta butter cooler is a clay container from Italy that retards the tendency of butter to melt on a hot day. I use it for backyard picnics. Without the container the butter soon gets too soft. With the container it remains firm through most of the festivities.

The bottom dish of the container is glazed so that the butter does not penetrate the clay. The unglazed cover is responsible for the temperature control. About half an hour before dinner I invert the cover and fill it with cool water.

By dinnertime the clay is saturated. I pour out the remaining water and put the cover over the butter (fresh from the refrigerator) and the bottom dish.

As soon as the container is taken outdoors it begins to warm because it absorbs heat from its surroundings. Even if it is not in direct sunlight, it still absorbs visible and infrared radiation. It also receives heat from the convection of warm air past it. Even more heat is conducted into the container from the table unless it is put on an insulator such as a potholder.

If the container is completely dry, the energy gained from the radiation and air at the exterior is gradually conducted through the walls to the interior. When the cover is wet, the transfer of heat through the wall is retarded because much of the energy is consumed in the evaporation of water. As the exterior dries, more water is drawn from the interior of the wall to replenish the supply on the exterior. Because of the delay in heat conduction, the interior air and the butter remain cool.

To check my explanation I set up an experiment in which the butter cooler was warmed by a quartz heater, which delivers infrared radiation and some visible light. I first soaked the clay cover by inverting it and filling it with water. The external surface was noticeably dry. After a few minutes the water inside the cover developed a great many air bubbles as water displaced air from the clay. After about 45 minutes water had soaked through the wall and wet the external surface.

After emptying the cover I put it on the dish and ran a thermocouple wire into the interior of the container. The tip of the thermocouple was near the center of the dish and a few millimeters above it. With the thermocouple I monitored the temperature of the air inside the container. The container and the thermocouple were then placed about 30 centimeters in front of the quartz heater, which provided more heat than would be expected on a summer day. One side of the container was bathed with infrared and visible radiation; the other side faced a cool basement room and so received comparatively little infrared.

As the container received heat the temperature of the interior air began to rise. Initially the increase was at the rate of one Celsius degree per five minutes. After half an hour the rate slowed to about one degree per eight minutes. By then the air temperature had risen by only five degrees and the side facing the heater looked and felt dry. The side shaded from the heater still appeared to be wet.

After letting the container dry for a day I repeated the experiment with a dry cover. This time the interior air warmed at the rate of one degree every three minutes. After half an hour the temperature rose by 10 degrees, twice as much as it had when the cover was wet.

Later I weighed the cover before and after soaking it with water. The soaked cover holds about 85 grams of water (after it has been emptied, of course). To evaporate that much water approximately 192 kilojoules of energy must be provided. If I assume that the intensity of the sunlight at my picnic table is about one kilojoule per second over a square meter, the cover should take about two hours to absorb enough sunshine to dry completely. It actually dries faster because nearby objects also provide infrared and visible radiation and because convection and conduction are delivering heat.

The Yunnan pot, which I obtained from Williams-Sonoma, also employs water, but for a different purpose. The glazed clay pot has a central chimney through which steam rises to heat food inside the pot. When I cook chopped meat or vegetables, I place the pot in a deep pie pan filled with water. As the water heats up, steam rises through the chimney and gently cooks the food.

The heat transfer is somewhat like that in a heat pipe except that the water does not return to the heat source. Instead it collects in the pot with the food keeping the food moist. Steam cooking is more often done with a perforated metal basket mounted just above boiling water. The steam rises through the holes of the basket and condenses on the bottom of the food, transferring the latent heat of vaporization to it.

Some of the condensed water clings to the food, but most of it drains back to the bottom of the pan and is revaporized. The food remains moist but is not soaked. The advantage of the Yunnan pot over the metal basket is that the pot enables one to steam a food that is already liquid. In the basket the liquid would be lost to the pool of water.

I checked the pot with two simple experiments relying on a thermocouple probe mounted inside the pot about three centimeters from the bottom and about midway between the wall and the chimney. The temperature of the air inside the pot was initially 21 degrees. When I put the pot in a pan of boiling water, the air temperature began to rise at the rate of about .6 Celsius degree per second, reaching 90 degrees after slightly more than three minutes.

After cooling the pot I covered the top of the chimney with heavy tape and repeated the measurements. This time the increase in temperature was about four times slower. After three minutes the temperature was still less than 55 degrees. Clearly the normal rapid rise in the temperature inside the pot is driven by the latent heat of vaporization of the steam in the chimney. The conduction of heat through the bottom of the pot is less important.

Bibliography

THE HEAT PIPE. G. Yale Eastman. Scientifc American, Vol. 218, No. 5, pages 38-46; May, 1968.

EXPERIMENT IN THE BRAGG REFLECTION OF LIGHT FOR THE UNDERGRADUATE USING CHOLESTERIC LIQUID CRYSTALS. A. Olah and J. W. Doane. American Journal of Physics, Vol. 45, No. 5, pages 485-488; May, 1977.

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.