Eventually I learned that such boats are driven by the camphor molecules that rush from the stern and spread as a layer one molecule thick on the otherwise clean surface of the water. The layer of camphor molecules lowers the surface tension of the water behind the boat, and the higher surface tension in front of the boat pulls it forward. Other substances, such as oils, also form monolecular layers on water. Molecules of oil naturally present in my uncle's skin spread on the water from his finger and exerted enough pressure on the boat to change its course.

The pressure developed by such monolayers varies with the size and structure of their constituent molecules. By measuring the pressure the experimenter can determine how the atoms in the molecule are grouped and how the monolayer interacts with other subitances [see "Monomolecular Films," by Herman E. Ries, Jr.; SCIENTIFIC AMERICAN, March]. Although some of these studies are beyond the reach of the amateur, he can perform many engrossing and even useful monolayer experiments with apparatus assembled from materials ordinarily found in the kitchen and bathroom. By adding a few items from the hardware store and the drugstore it is possible to measure the thickness of films down to a small fraction of the wavelength of light, to determine the pressure exerted by a film, to learn the size and general shape of the molecules in a film, to deposit monolayers on glass that produce striking patterns of iridescence and to substitute surface tension for conventional standards of weight in the calibration of Laboratory balances.

The experiments are not difficult even for beginners. With information supplied by Robert B. Dean, a physical chemist of Bainbridge, N.Y., I set up the apparatus and made most of it work on the first try.

"To have any luck at all in experimenting with monolayers," Dean writes, "you must set up a clean place in which to work and give careful attention to detail. The quantity of material comprising a monolayer is so minute that our ordinary standards of cleanliness must be revised by at least two orders of magnitude. A flyspeck constitutes gross contamination.

"The disproportion between the area and the volume of a monolayer, which makes cleanliness so important in these experiments, was observed in 1765 by Benjamin Franklin. He spread oil on a pond in London to learn why oil slicks tend to 'quiet troubled waters.' 'At length being at Clapham,' he wrote, 'where there is on the common a large pond which I observed one day to be very rough with the wind, I fetched out a cruet of oil and dropped a little of it on the water. I saw it spread itself with surprising swiftness upon the surface; but the effect of smoothing the waves was not produced; for I had applied it first on the leeward side of the pond where the waves were greatest, and the wind drove my oil back upon the shore.

I then went to the windward side where they began to form; and there the oil, though not more than a teaspoonful, produced an instant calm over a space several yards square which spread amazingly and extended itself gradually till it reached the lee-side, making all that quarter of the pond, perhaps half an acre, as smooth as a looking glass.'

"By dividing the known volume of his oil by the estimated area of the pond that it covered, Franklin concluded that the thickness of the film must be on the order of one ten-millionth of an inch. It is interesting to repeat Franklin's experiment. He used olive oil, but any good salad oil should work as well. A cubic inch of oil (half an ounce) should calm 10 million square inches of water (about 1.6 acres). Consider how much the volume of the oil is reduced when the area of the pond is scaled down to laboratory proportions-say to a dish eight inches wide by 12 inches long. On the assumption that the monolayer is a ten-millionth of an inch thick, the volume occupied by a film covering the water in the dish would be only one hundred-thousandth of a cubic inch, which is small even as flyspecks go.

"The inch is rather an unwieldy unit for measuring films so thin. For this reason those who experiment with monolayers customarily turn to the metric system for a more convenient scale: the angstrom unit. (One angstrom unit is one ten-millionth of a millimeter, or about four hundred-millionths of an inch.)

"To determine the over-all dimensions of a molecule in a monolayer (the proportions of the volume that it occupies) the experimenter must know approximately how many molecules the layer contains. This is found by an indirect method. For example, standard reference texts show that glyceryl trioleate, the principal constituent of olive oil, has a molecular weight of 885 and a density of .915. One gram molecular weight (the weight of 6 X 10²³ molecules) occupies 967 cubic centimeters, or 967 X 10²⁴ cubic angstrom units. One molecule of glyceryl trioleate will therefore occupy 6 X 1023 of this volume, or 1,612 cubic angstrom units. Franklin's estimate of the thickness of his film-one ten-millionth of an inch-is equivalent to about 25 angstroms. If this had been the correct value, each molecule in his monolayer would have occupied an area of only 65 square angstroms. But relatively simple measurements that any careful amateur can make prove that a molecule of glyceryl trioleate has a cross-sectional area of 97 square angstroms, which corresponds to a monolayer thickness of only 16 angstroms. We must conclude that some of Franklin's oil was lost on the edges of the pond.

"All possible precautions are taken in the laboratory to avoid errors of this sort. The material under investigation is first mixed with an insoluble spreading agent, such as hexane, that evaporates quickly and completely. An accurately weighed amount of the mixture is then gently floated on the clean surface of a trough of water. To assure even distribution somewhat less material is applied to the water than is required for a complete monolayer. The monolayer is subjected to pressure by sliding a waterproofed strip of glass or metal, called a barrier, across the surface of the trough from one end toward the other. This squeezes the molecules into a continuous layer that presses against a floating barrier at the far end of the trough. The floating barrier is linked mechanically to a sensitive balance that registers the pressure transmitted by the monolayer. In troughs of typical dimensions (in excess of 600 square centimeters) many substances are capable of transmitting pressures on the order of half a gram. If pressure is applied beyond the amount needed to pack the molecules closely, the monolayer buckles; this establishes an upper limit to the pressure that can be applied to the floating barrier. When the pointer of the balance stops moving up the scale, the experimenter knows that a closely packed layer has been formed. The area occupied by each molecule is then computed by dividing the known area of the monolayer by the known number of molecules. The length of the molecules, which in many substances is equal to the thickness of the film, is determined by dividing the known volume of the monolayer by its area.

"As an introduction to monolayer experiments, clean a container, say a glass baking dish about six inches square, with scouring powder and rinse with tap water until the glass wets completely. Dry the dish with a paper towel. Keep your fingers off! Then with a paper towel apply a film of white vaseline around the rim of the dish. Set the dish level and fill it to the brim with water. Let the water overflow for a time to carry away as much surface contamination as possible. In spite of such measures the surface of the water will most certainly be covered by a contaminating monolayer of some sort. To detect the monolayer, dust some talcum powder on the surface and blow it around gently. (You must use unperfumed, unmedicated talcum powder of the kind supplied to hospitals. You can buy it at a drugstore.) Try to compress the monolayer. Pressure can be applied by the edge of a clean, untouched piece of wax paper. Place one edge of the paper against the water where it curves up from the edge of the dish and pass it across the surface toward the opposite side. Particles of powder some distance from the edge of the paper will move, as if acted on 'at a distance' by the paper.

"Sweep the powder (and monolayer) off the opposite side of the dish. Repeat the experiment, using clean paper each time, until the powder is not pushed 'at a distance.' The classical tool for removing contamination from water surfaces is a strip of glass or metal coated with paraffin. Usually several are used. The sweep is started with one barrier; when it has been advanced some distance, another sweep is started with a second barrier and so on. The strips are simply laid across the end of the container and advanced slowly toward the other end. The surface can also be sucked clean, in the case of Franklin's pond it had been blown clean by the wind. No method gets all the dirt in one sweep. I prefer to wipe the surface with ordinary paper and then sweep it with paraffined glass barriers. In this initial experiment, however, wax paper is adequate. Add powder as necessary during the sweeping so that you can follow your progress. When at last the movement of the paper does not push nearby particles of powder, you have a reasonably clean surface on which to deposit your first monolayer.

"Before depositing the monolayer take a pinch of powder with the untouched edge of a paper scoop and dust the surface lightly. Then touch the surface with the tip of a single hair from your head. Unless your hair is exceptionally free of oil, a circular area ranging upward from the size of a dime will dart out from the hair. If you have dry hair, it may be necessary to dip the hair in olive or salad oil. Wipe the hair clean with a paper towel. There will still be enough oil sticking to the hair to drive the talcum completely off the surface. A more elegant tool for applying the oil is a freshly drawn glass fiber. It is clean and so enables you to choose any material you wish for the monolayer.

"Oils and other materials that form v monolayers on water are characterized by comparatively large, insoluble molecules that include one group of atoms that is attracted to water and another group that is repelled by water. In many substances-for example, olive oil-the hydrophilic, or 'water-loving,' group of atoms is situated at the end of the molecule. It is the attraction between this end and the water that accounts for the formation. of the monolayer. The area occupied by a monolayer depends on how the molecules are arranged. If they are oriented at random, the area may be relatively large. But when the layer is compressed, as by pushing against its edge with a piece of wax paper, the molecules crowd against one another and are believed to upend in tight formation with the hydrophilic ends in contact with the surface. If more molecules are present than can be accommodated in the monolayer, they may clump together as drops. Or, to put it another way, the monolayer is in equilibrium with the floating drops. You can see such droplets on the surface of clear soup. When portions of the monolayer are skimmed off, the droplets supply replacing molecules to the monolayer. Conversely, such droplets can be made to grow by compressing the monolayer into a smaller area."

The editor of this department tried the above experiment with a monolayer of olive oil on a dish that measured eight by 16 inches.

Half an hour of skimming failed to diminish the apparent size of: the droplet, even when it was examined by a 10-power magnifying glass. When the layer was compressed between barriers and viewed at a low angle, however, random points of reflected light appeared on the surface, possibly indicating places where oil from the buckled monolayer gathered as minute droplets. My results were more gratifying in another experiment suggested by Dean for demonstrating differences in the spreading pressure of substances that form monolayers. "Each oil," he writes, "has its own spreading pressure. An oil with higher spreading pressure will displace one of lower spreading pressure. To observe the effect, prepare a clean surface of water. Then cut a length of fine silk thread (or a thread of nylon or rayon) about half again as long as the width of the dish. Smear the thread with a film of vaseline, using a piece of paper towel for the applicator, and lower it zigzag onto the surface of the water. The ends are draped over the sides of the dish about a quarter of an inch and stuck down with dabs of vaseline. (Use a clean toothpick to handle the vaseline.) Sweep the surface on each side of the thread with a barrier until the water is clean. A monolayer of egg albumin is now applied to the cleaned surface on one side of the thread by dipping a clean toothpick into the white of a freshly opened egg and touching it to the surface. The spreading monolayer will exert pressure on the thread, causing it to bend into the clean area. If a bit of olive oil is now applied to the clean surface, the resulting monolayer will cause the thread to exert a pressure of about 20 dynes (1/1,400 ounce ) per centimeter against the albumin and collapse it. In experiments of this sort, where one monolayer is used to apply pressure to another, the movable barrier (the vaseline-coated thread) is called the piston and the driving substance the piston oil. Tricresyl phosphate exerts a spreading pressure of about 10 dynes per centimeter of thread length, somewhat less than is needed to collapse a monolayer of albumin, and can therefore be used as a piston oil to keep an albumin monolayer under pressure.

"Tricresyl phosphate is a plasticizer used in gasoline and in vinyl plastics. The spreading pressure and surface behavior of most other plasticizers have not been reported in the literature. Many of them would doubtless make good piston oils and their measurement could well be undertaken by amateurs."

I had a lot of fun setting up pistons. It turned out that our medicine cabinet was stocked with a variety of monolayer-forming substances ranging from insect repellents to hand lotions. In many cases the name of the substance was printed on the label. By pairing these off and testing each against the others I found it possible to arrange them in order of ascending spreading pressure. Care must be taken not to apply too much piston oil to the water or the piston will snap into a tight bow and the piston oil will spill over into the compressed monolayer. The pressures that I recorded were merely relative, of course. With Dean's encouragement I next undertook the construction of a hydrophilic balance for making quantitative measurements of spreading pressures.

The apparatus consists of three principal subassemblies: a trough for holding the water and monolayer, a base for leveling the trough and a calibrated torsion balance [see illustration in Figure 2]. The construction can be varied dimensionally or otherwise to suit the whim of the builder and the materials at hand. One primary design consideration should be kept in mind, however: The experimenter will spend most of his time cleaning the apparatus. Keep it simple and easy to take apart.

The trough is a glass baking dish eight inches wide, 16 inches long and two inches deep. The rim of the dish is ground flat so that a glass barrier that spans the dish can be slid from one end to the other without rocking. Invert the dish on a piece of plate glass and grind it down with successive grades of wet carborundum. I used No. 180 grit for 1 minutes and No. 240 for about the same interval. Apply firm pressure as the dish is pushed back and forth. Add water and additional carborundum as the grit becomes gummy. Don't worry if a few pits made by the coarse grades of abrasive remain after grinding with the finer grades. After being ground the dish is first cleaned with scouring powder, the with household ammonia, and is rinse in running water for several minutes. According to Dean this procedure, if vigorously followed, should suffice. I gave the inner surface an additional scrubbing with a swab dipped in nitric acid and rinsed the acid away with running water-the final step used by amateur telescope makers in cleaning glass for silvering If acid is used, remember that it is corrosive, highly poisonous and should not be used in the kitchen. The rim of the dish is then coated with paraffin. Either of two methods works well. The glass can be heated in the oven to a temperature above the melting point of the wax and a block of paraffin of the grade used for sealing jelly glasses rubbed completely around the rim. No harm will be done if melted wax runs down the sides of the dish. Alternatively, the wax can be dissolved in hexane and painted on the rim with an oil-free brush. In making up this solution use about one part paraffin to 20 parts of hexane. If you are in a hurry, shave the paraffin. It dissolves slowly in hexane.

My sweeping barriers are half-inch strips of window glass 10 inches long. Half a dozen were cut with an ordinary glass cutter of the wheel type. The edges were rounded slightly by grinding on plate glass with No. 240 carborundum grit and water. They were cleaned with scouring powder, rinsed, dried and dipped in melted paraffin. Floating barriers of several types were tried; all of them worked. The simplest was a soda straw. After the straw was cut to a length of seven inches (an inch less than the width of the trough) an 11-inch length of vaseline-coated silk thread was run through the straw. The ends of the straw were then sealed with a few drops of melted paraffin. A wooden strip about 3/8 inch wide, seven inches long and 1/8 inch thick, coated with paraffin, was also successful. Vaseline-coated silk threads two inches long were attached to the underside of the strip with dabs of paraffin. The third version is the rectangular boat of aluminum foil shown in the illustration in Figure 2. It was also coated with paraffin and equipped with threads. It seemed to respond to surface pressures somewhat more promptly than the soda straw or stick, doubtless because it is less massive.

The balance consists of (1) a torsion fiber supported by and in axial alignment with a pair of shafts, (2) a beam in the form of a bell crank attached to the fiber and fitted with an optical lever and (3) a supporting framework of strap iron. One shaft can be rotated with a vernier dial for adjusting the torsion fiber. The other shaft can be rotated as well as shifted along its axis to any desired position. The arms of the beam are equal in length, and force applied to either arm twists the torsion fiber the same amount. The magnitude of the force can be measured either by applying a measured twist to the fiber in the opposite direction (by means of the calibrated dial) until the beam is restored to its original position or by observing the displacement of the optical lever. When calibrated with care, the balance is accurate to better than 1 per cent from one to 500 milligrams and is sensitive to a small fraction of a milligram.

The torsion fiber was made from the E string of a violin and is .01 inch in diameter. The beam was made from the G string of a guitar and is .027 inch in diameter. I used strings instead of bulk music wire because they are long enough, available in any music store and cost only 30 cents. The torsion fiber is 12 inches long. The apex of the beam is soldered to the middle of the fiber. Acid-core solder and a hot iron were used to make the joint. Some strings are chromium-plated but can be soldered easily if the plating is sanded off with 00 grade carborundum cloth. A small hook is bent in the horizontal arm and an inverted U, bent from G-string stock, is soldered to the bottom of the vertical arm. The U engages the floating barrier as shown in the illustration in Figure 3. A brace of the same wire is soldered between the horizontal and vertical arms of the beam.

The reflector of the optical lever was cut from a thin pocket mirror of the type available at the cosmetic counter of a novelty store. It is about a quarter of an inch square and is attached to the apex of the beam by quick-drying cement. It reflects the image of a silk thread from a 35-millimeter slide projector to a screen on the wall. The thread is stretched across an empty slide holder. The screen is a strip of white cardboard with a vertical scale divided into 10minute intervals of arc.

The assembled balance was set up on the bench and, after the torsion fiber was pulled tight, oriented so the mirror would reflect the image of the silk thread; onto the screen. The image was centered on the zero graduation in the middle of ft the scale by turning the clamped shaft. A coil of No. 36 bare copper wire that weighed one gram was then suspended from the hook of the horizontal arm and the dial was rotated until the beam was restored to its zero position. As luck would have it, this required precisely one full turn. (A 25-foot length of No. 36 copper wire should weigh approximately one gram according to the handbook, but the weight of commercial wire may vary as much as 5 per cent. A gram of wire should be weighed out on an accurate analytical balance.) The wire was then cut precisely in the middle and half of the coil was suspended by the beam. Zero was again restored, this time by turning the dial through 180 degrees. A quarter of the coil required 90 degrees of dial rotation, and so on. Each degree of rotation therefore corresponded to 2.78 milligrams.

To prepare the balance for measuring the surface pressure of a monolayer the glass trough is cleaned, the rim of the trough is given a fresh coat of paraffin and the trough is leveled so that it can be filled to the brim with water. The surface is then swept by the barriers until free of contamination. Next, the floating barrier is placed on the water about a quarter of the length of the trough from one end. The associated threads, which act as seals to prevent the monolayer from leaking past the barrier, are draped over the rim of the trough about a quarter of an inch and are stuck fast: with dabs of vaseline. The torsion balance is then placed over the trough so the ends of the inverted U just touch the floating barrier. The ends of the U are stuck to the barrier with dabs of vaseline. The projector and screen are placed so that the image of the thread falls on the zero graduation of the screen. Both areas of the water surface are again swept with the barriers.

A monolayer is now placed on the larger of the two uncontaminated surfaces. For the initial experiment Dean recommends measuring the spreading pressure of stearic acid. Chemicals of the highest purity must be used or the measurements will not agree with the values tabulated in reference texts. (To obtain these chemicals you may want to have your druggist write Distillation Products Industries, Rochester 3, N.Y., for ordering information. He should ask for list No. 42 of organic chemicals.) One cubic centimeter of stearic acid is mixed with 49 cubic centimeters of hexane. Five cubic millimeters of this solution are then carefully floated onto the water with a micropipette. This volume contains enough molecules to cover a surface of some 400 square centimeters with a monolayer. To make a calibrated micropipette, buy or draw a glass tube with an inner diameter of about one millimeter. This capillary tube is then pushed into a rubber stopper that in turn fits into a larger section of tubing fitted with a length of rubber hose and mouthpiece [see illustration in Figure 4]. Before inserting the capillary into the stopper weigh it on the torsion balance. Then dip the outer end in a container of water tinted with food coloring. Capillary attraction will draw water into the tube. Measure the length of the water column and immediately weigh the capillary. The difference in weight between the empty and filled capillary in milligrams is equal to the volume in cubic millimeters that is occupied by water. Paint a mark across the tube at the inner end of the water column. Work quickly so that water is not lost by evaporation. Use India ink for marking the graduations and apply it with a single hair. Divide the remainder of the tube (from the mark to the outer end) into as many equal intervals as the water weighed in milligrams and mark off the intervals with ink lines. This method of calibration is adequate if the tube does not taper. Finally, coat the inner and outer surfaces of the capillary with silicone oil.

The hexane will evaporate, leaving an uncompressed monolayer on the surface that has a volume of .1 cubic millimeter. A clean barrier is now slid gently onto the water at the end of the trough farthest from the floating barrier and gradually pushed toward the barrier.

As the area is thus reduced, the balance will begin to respond to growing pressure exerted by the monolayer; eventually the reading will mount sharply to a maximum value. The maximum pressure (in grams) indicated by the balance is divided by the length (in centimeters) of the floating barrier plus half the distance between the ends of the barrier and the edges of the tank. The quotient is multiplied by 980 to get spreading pressure in dynes per centimeter. The spreading pressure of stearic acid at 25 degrees centigrade is about 30 dynes per centimeter, and its molecule occupies 20 square angstroms.

Spreading pressure can also be determined by measuring the surface tension of a monolayer and that of water. Suspend a clean glass slide, such as a microscope cover slip, by a small chain or wire as shown in the illustration i Figure 5. Hook the chain to the horizontal arm of the torsion balance and adjust the balance to zero. Then bring a container of water up under the glass until the surface of the water touches the bottom edge of the glass. Surface tension will pull it into the water. Now turn the dial of the torsion balance until the bottom of the cover slip is raised precisely even with the surface of the water. (Ignore the meniscus of liquid that clings to the glass.) To find the surface tension of the water in dynes per centimeter, multiply by 980 the weight (in grams) that is indicated by the torsion balance when the bottom of the wet slide is even with the surface and divide by twice the length (in centimeters ) of glass in contact with the water. If the temperature of the water is 0 degrees C., the result should be 75.6 dynes per centimeter; at 20 degrees it should be 72.7 dynes per centimeter. By reversing the formula you can use this procedure as a primitive method of establishing a standard of mass for calibrating balances. At 20 degrees C. the downward pull on the balance (in grams ) caused by surface tension is equal to the product of 145.4 times the length of the glass in contact with the water divided by 980. A weight, such as a length of wire, is then cut to produce the same deflection and used as the standard of mass. I have found the method accurate to about one part in 200.

A clean slide is then fully immersed in water that has been swept free of contamination, a monolayer is floated on the surface and the slide is carefully raised by rotating the dial of the balance. The surface tension of the monolayer is then computed as in the previous experiment. The difference between the surface tension of the water and that of the monolayer is equal to the spreading pressure of the monolayer.

One final experiment: Set up a clean dish about four inches deep and fill it to the brim with water containing a few drops of barium hydroxide ( approximately a .01 normal solution). Divide the surface with a piston, float a monolayer of stearic acid on one side and a monolayer of olive oil on the other. Then, with a small windlass improvised from handy materials [see illustration in Figure 6], slowly lower a clean microscope slide into the stearic acid and pull it up again. With practice you can make a monolayer of acid cling to the glass surface on each pass. Observe the action by watching the piston. Continue dipping the slide and replenish the monolayer as required. After about 20 monolayers have been applied the glass will begin to take on color (caused by the interference of light waves). Continued dipping may produce a pattern of color like that observed on soap bubbles and oil slicks. If the layers are evenly deposited, the color will consist of a single hue and the slide will function as an interference filter. Certain thicknesses will act as nonreflecting coatings. According to Dean, by immersing the slide fully on the first dip and to successively shallower depths thereafter it is even possible to build up a step gauge for measuring thicknesses optically.

Bibliography

THE PHYSICS AND CHEMISTRY OF SURFACES. Neil Kensington Adam. Oxford University Press, 1941.

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.