A NUMBER of substances can be identified by measuring their boiling temperature. For example, oleic acid and stearic acid are chemically similar animal fats that to the eye appear identical, but oleic acid boils at 286 degrees Celsius and stearic acid boils at 383 degrees C. Either one can be identified by heating it to a boil in an open vessel at sea level and measuring the temperature with a thermometer. The altitude where the measurement is made is significant because the boiling temperature of a fluid is determined not only by the nature of the substance but also by the pressure exerted on the surface of the fluid by the atmosphere.

As a fluid is heated some of its molecules acquire enough energy to escape through the surface and exert a vapor pressure in the space above the liquid. The temperature at which this pressure equals the pressure exerted by a column of mercury 760 millimeters high is by definition the boiling point of the fluid. Chemists routinely measure the boiling point of certain fluids with an isoteniscope, an apparatus that excludes the atmosphere. With an isoteniscope the experimenter can investigate a number of properties of fluids, including their vapor pressure at any temperature, the theoretical boiling point of fluids that would be destroyed if heated to boiling, the individual pressures exerted by each gas in a mixture of gases and related information of interest, particularly to those who experiment with such instruments as gas lasers and gas chromatographs. An isoteniscope that amateurs can make at home has been designed by Walter Averett, a chemist of Vicksburg, Miss. He discusses its construction and operation as follows:

"In effect the isoteniscope consists of a closed vessel supplied with a sensitive gauge for measuring vapor pressure and a means for heating or cooling the vessel to a desired temperature. The specimen is placed in the vessel and frozen. Air is then exhausted from the system with an air pump. The specimen is warmed just to thawing. At this temperature some-air absorbed by the specimen escapes and is removed by the pump after the specimen is frozen again. Customarily the specimen is frozen, thawed and exhausted several times to remove all the absorbed air. Finally, the frozen specimen, now free of unwanted gases, is thawed and warmed to a selected temperature. There a measurement is made and the vapor pressure in the space above the specimen is recorded.

"Measurements are similarly made and recorded at several arbitrarily selected higher temperatures. The tabulated results show that vapor pressure increases with temperature for all substances and by an amount that is characteristic of each substance. One need not heat a fluid to its boiling point to learn the temperature at which it boils. The boiling point can be determined graphically by plotting vapor pressure against temperature at several temperatures below boiling and extrapolating the graph to find the temperature that would correspond to a vapor pressure of 760 torr, or one atmosphere.

"The isoteniscope consists of a 10milliliter flask connected through a tapered glass joint to a manometer [see illustration at rught]. A bypass tube that includes a stopcock connects the sample side and the outlet side of the manometer and is used for equalizing pressure between the two. The outlet side of the manometer also includes a stopcock for isolating the apparatus from the vacuum system and a tapered glass joint for connection to the accessory apparatus.

"The accessory apparatus includes a mercury manometer for measuring a maximum pressure of 760 torr; a ballast flask of about one-liter volume that cushions abrupt changes in air pressure; an air pump; a needle valve for isolating the pump from the system; a needle valve for admitting air to the system; a thermostatically controlled water bath for maintaining the isoteniscope at a desired temperature, and a cathetometer for measuring pressure as indicated by the mercury manometer [see illustration below left]. The system can be simplified if the experimenter is willing to sacrifice accuracy in the final results. For example, a meter stick can be substituted for the cathetometer for measuring the height of the mercury column. A reasonably effective substitute for the unsilvered Dewar flask can be made by nesting a pair of ordinary beakers and circulating thermostatically controlled water between the two. An adequate vacuum pump can be improvised from the compressor of a refrigerator if the check valve is removed from the inlet port of the compressor by the procedure described in The Scientific American Book of Projects for the Amateur Scientist (Simon and Schuster, Inc., 1960). The dimensions of the apparatus are not critical.

"I made the mercury manometer from barometer tubing that has a bore three millimeters in diameter. One end of the tube is heated in a gas burner until the bore closes. A U bend is made 800 millimeters from the closed end. When you make the bend, heat the glass until the bore shrinks at that point to a diameter of about one millimeter. The constriction limits the velocity of mercury in the tube and is intended to prevent the mercury from striking the closed end of the manometer with sufficient impact to fracture the glass should the vacuum be accidentally broken during operation. A right-angle bend is made in the open end of the manometer about 600 millimeters from the constriction. Shops that repair neon signs can construct the apparatus for those who do not go in for glassblowing.

"The manometer was filled by connecting the open end to a mercury still that contained some six ounces of metal. The system was evacuated to the limit of the mechanical air pump. During the filling operation the manometer was placed horizontally but tilted so that the closed end was at the lowest point. The closed end was chilled with ice. Condensed mercury filled the tube progressively from the closed end. The still was shut down when mercury had accumulated about two centimeters beyond the constriction. When the manometer was removed from the still and placed upright, a void appeared at the closed end. This space is a vacuum and must be present.

"Depending on the barometric pressure of the atmosphere, the net difference in the level of mercury between the two arms of the manometer is about 760 millimeters. If a perfect vacuum were now created in the open end of the manometer, mercury in the closed arm would fall and mercury in the open arm would rise until the net difference would be zero, indicating zero pressure in the open arm. To make a pressure measurement you need to measure the level of metal in only one arm, because as metal rises in one arm it falls an identical distance in the other. To find the net difference measure the movement in either arm and multiply by two.

"A still can be improvised from a Pyrex distillation flask that has a ground neck mating with a tapered glass joint. The distilling operation should be conducted in a fume hood because the fumes of mercury are toxic. Incidentally, another potential hazard is the ballast flask. Apply a layer of adhesive tape to the ballast flask for confining glass fragments in case the vessel shatters while it is evacuated.

"The manometer that forms part of the isoteniscope is partly filled with oil. It indicates inequalities of pressure between the space above the specimen and the space in the manifold. The device is more responsive to changes in pressure than the mercury manometer because the density of the oil is much lower than the density of mercury. I used fluorocarbon oil to minimize the possibility of error arising from the chemical reaction of the oil with vapor from the specimen. The oil is known as Fluorolube MO-10 and is available from the Hooker Chemical Company.

"The vapor pressure of fluorocarbon oil is low, but the oil may contain air and volatile lighter oils. To remove these substances I exhausted the filled manometer to the limit of the air pump. There was considerable frothing as absorbed air escaped. A beaker of hot water was then raised around the U-tube, and pumping was continued until the oil stopped bubbling. The glass was also tapped lightly to encourage the release of additional gas. Subsequently I learned that the oil absorbs gases from acid specimens, particularly at elevated temperatures. For this reason I routinely degassed the manometer between tests and after a few runs refilled it with new oil.

"The isoteniscope must be maintained at a temperature as nearly constant as reasonably possible. Ordinarily the temperature is deliberately varied through a range during a test, but the temperature must be closely controlled at selected points. The isoteniscope is rigidly mounted with enough neighboring space so that the specimen flask can be immersed by raising a vessel of water under the apparatus.

"This arrangement enables the experimenter to conveniently submerge the specimen in refrigerant or in a constant- temperature bath. Depending on the nature of the specimen, I do the freezing with either a slurry of dry ice in acetone or with liquid nitrogen. Frozen specimens are warmed by a water bath.

"Originally I submerged the apparatus so that water covered the stopcock in the bypass line, but I discovered that I could not keep the stopcock vacuum tight at elevated temperatures. I had to lower the water level to clear the stopcock, thus exposing a small section of the vapor line to the atmosphere. The top of the vessel was covered with a sheet of aluminum foil to reduce the exchange of heat between the bath and the room.

"After the apparatus has been assembled the system must be tested for air leaks. Initially I used Kel-F grease No. 90 on all stopcocks and ground-glass joints. It turned out that the grease tended to leak when the apparatus was operated at 85 degrees C., the maximum temperature in my experiments. The problem was solved by coating the ground joint of the specimen flask with a mixture of Kel-F grease and Type 25-20M grease, a product of the Halocarbon Products Company. Avoid using excess grease: thick layers tend to flow and invite a leak. When the system has been exhausted to the limit of the pump and sealed off, the rise of pressure over a period of 12 hours should not be more than two millimeters.

"The procedure for a test run will be apparent if I give an example in which the boiling point of nitric acid is determined. It is next to meaningless to try to obtain a specimen of pure nitric acid, heat it to boiling and measure the boiling point with a thermometer, because the acid continuously decomposes as it is heated. On the other hand, it is relatively easy to measure the vapor pressure of the acid at several temperatures below boiling, to make a graph by plotting vapor pressure against temperature and so by extrapolation to estimate what the temperature of the mixture would be at a pressure of 760 torr.

"With a pipette fill the specimen flask almost to the top with acid. The object of almost filling the flask is to minimize the empty space in the isoteniscope. (The flask could be filled with a substance that is normally solid at room temperature, such as paraffin wax, by melting the wax.) Exhaust the system to the limit of the air pump.

"With the bypass stopcock open freeze the acid by immersion in liquid nitrogen. (Were the specimen an aqueous solution the freezing could be accomplished by a slurry of dry ice.) Close the valve to the air pump and warm the specimen until it becomes slushy. This procedure will release absorbed air and carbon dioxide but not any significant amount of nitric acid, water or nitrogen dioxide because the freezing point of the mixture ranges from -20 to -40 degrees C.

"Repeat the freezing, thawing and pumping three times. Make a test to determine if the specimen has been adequately degassed by again freezing the acid and, after closing the valve to the air pump, closing the stopcock in the bypass. Any gas released by the specimen will be indicated by the oil manometer.

"Record the level of the mercury in the closed-end arm of the mercury manometer. Make the reading by observing the top of the meniscus in relation to the meter stick. If a cathetometer is used set the hairline of the telescope eve with the top of the meniscus. This reading represents zero pressure.

"Remove the freezing bath and let the specimen warm. As the temperature rise vapor will be released by the specimen and will displace the oil in the oil manometer. Carefully crack the needle valve that opens to the atmosphere. Admit just enough air to establish equilibrium in the two legs of the oil manometer. The adjustment is a bit tricky, but you will soon develop the knack of admitting air at a rate that establishes and maintains the equilibrium. If air is admitted too rapidly, oil may be blown into the specimen, ruining the test.

"It is assumed that the specimen is surrounded by a water bath at some minimum temperature above freezing and that this temperature is measured by a thermometer. When the oil manometer remains in equilibrium for a few minutes, you can assume that the temperature of the specimen equals that of the water bath. At that temperature read the pressure as indicated by the mercury manometer and record the temperature and the pressure.

"Raise the temperature of the water bath a few degrees. As the specimen warms and liberates vapor again admit air to maintain equilibrium in the oil manometer. After the specimen has reached the higher temperature read the vapor pressure as indicated by the mercury manometer. With nitric acid I usually measure the pressure in intervals of 10 degrees from 0 to 80 degrees C.

"If the specimen were a stable liquid, such as water, the measurement would now be complete. A graph of the results could be made and extrapolated to find the boiling point. As previously mentioned, however, nitric acid continuously decomposes when it is heated. As the temperature rises the specimen becomes a mixture of nitric acid, nitrogen dioxide, nitrogen tetroxide and other oxides of nitrogen plus water. The proportions of nitrogen dioxide and nitrogen tetroxide change with both temperature and pressure. Indeed, the proportions can change even though the specimen is held at constant temperature, with the result that the vapor pressure may change at constant temperature. Hence the pressure readings as recorded can only be a best approximation at a given temperature.

"An interesting check of chemical decomposition can be made by recording a series of measurements as the specimen is cooled from an elevated temperature. If the specimen is stable, the resulting measurements will match those made when the specimen was warmed. If the specimen decomposes, the pressures will be somewhat greater during the cooling run. A graph of the results of both the heating run and the cooling run will take the form of a hysteresis loop [see illustration at right]. The area enclosed by the loop varies with the extent of the decomposition.

"Because of the decomposition the pressure measurements actually are measurements of partial pressures. The total pressure is the sum of each of the components in the mixture. The measurements are not misleading in the case of decomposing nitric acid because one is interested in the behavior of the mixture. Measurements could be misleading if one were under the false impression that one was testing a pure solution of nitric acid in water.

"When the vapor pressures have been recorded for the selected temperatures, the next step is to draw a graph of the results. I prefer a plot of the type first suggested in 1923 by the chemist Edwin R. Cox. It displays pressure v. temperature, but the temperature axis is laid off in units of vapor pressure of a standard substance [see illustration at left]. For example, the vapor pressure of water is well established: 760 torr of water vapor appears at 100 degrees C., hence the temperature point of 100 degrees C. on the plot can be designated as 760 torr. At each lower temperature water has a known vapor pressure, and so the axis can be calibrated in units of pressure that are equivalent to units of temperature. If the vapor pressure of the specimen is plotted against these pressure equivalents of temperature, a straight line results. A plot of vapor pressure against temperature is not linear, but a graph in the form of a straight line results when two vapor-pressure curves are plotted against each other. The Cox chart lends itself nicely to extrapolation because it is easier to extend a straight line than a curved one. A little time spent in plotting the data will soon lead the experimenter to the form of plot that seems most useful to him. Another form often used is the plot of vapor pressure against the reciprocal of temperature.

"In addition to its usefulness as an instrument for measuring vapor pressure the isoteniscope is a handy tool for preparing gas mixtures according to a set of prescribed partial pressures. My work in gas chromatography and infrared spectrophotometry requires the preparation of gas standards of known composition. I find it convenient to use a manometric method, wherein a sample bulb is evacuated and then a specified gas is admitted to obtain a certain pressure.

"A Bourdon gauge is adequate in the general range of .25 atmosphere and higher, but the reading error soon becomes impossibly large at lower pressures. For example, my gauge is calibrated for a vacuum ranging from 0 to 30 inches of mercury, but there are only three divisions in the interval from 0 to five inches, which corresponds to about 40 torr per division. In contrast, the isoteniscope is reasonably accurate to .5 torr.

"Assuming that the air pump exhausts the system to zero pressure, the introduction of gas to a pressure of 76 torr amounts to .1 atmosphere, and the system contains 10 percent of the amount of gas it is capable of holding at a pressure of one atmosphere. Another gas can be introduced to bring the pressure up to exactly one atmosphere, in which case the concentration of the first gas would be 10 percent by volume. (It is assumed that the temperature remains constant, that no chemical reaction occurs, that there is no association or dissociation and that the gas compressibility can be neglected.)

"Thus by using gas sources such as cylinders or lecture bottles any desired mixture of compatible gases can be prepared, including helium-neon, helium-argon, helium-carbon dioxide and similar mixtures used in gas lasers. My infrared spectrophotometer has a cylindrical Pyrex cell, 10 centimeters long by about five centimeters in diameter, with a sodium chloride window at each end and with two sidearms complete with stopcocks and 18/9 ball joints. If the Dewar flask and the sample flask are eliminated from the isoteniscope and its associated equipment, a simple adapter makes it possible to connect one sidearm of the infrared cell to the 19/38 joint where the sample flask would normally be. One or more gas cylinders can be attached to the other sidearm of the cell through a manifold.

"The entire system is then evacuated, up to the leaktight valve on the gas cylinder, and the vacuum pump is isolated. The bypass stopcock is closed, the cathetometer is set to read the calculated pressure and gas is slowly bled into the cell while air is admitted to the vacuum side to keep the oil manometer in balance. When the mercury meniscus reaches the correct point, the calculated pressure of the gas will have been introduced into the cell. The sidearm stopcocks are closed and the gas cylinder is removed, or additional gas is added to the next calculated pressure.

"The bypass stopcock can then be opened and air can be bled in until the vacuum in the system has been relieved. This can be done so that the corrosive gas will not escape from the system. A liquid-nitrogen trap can be installed to protect the manometer. A trap should be used in any event to protect the vacuum pump, unless some other expedient can be devised."

Bibliography

TECHNIQUE OF ORGANIC CHEMISTRY, VOL. I: PHYSICAL MFTHODS OF ORGANIC CHEMISTRY, PART I. Edited by Arnold Weissberger. Interscience Publishers, Inc., 1959.

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.