Among the various instruments which have been devised for the measurement of temperature, the air thermometer has the distinction of being the first form of any value. It was invented probably by Galileo about the year 1593, and was used to a considerable extent by physicians; but its readings were deceptive, for at that time the influence of atmospheric pressure was unknown. Galileo invented the alcohol thermometer eighteen years later, and this more accurate and at the same time more simple instrument almost entirely superseded the older form.
Fig. 170 - A convenient scale for barometers.
In some ways however the air thermometer is more efficient than either the mercury or alcohol thermometer. Since it is based on the principle of expansion of a gas, the air thermometer is very sensitive, and offers a large register for a small change in temperature. The reason for this greater susceptibility to heat is evident from the following data. The coefficient of expansion of air is 0.003665, or approximately 1-273 of the volume; the coefficient of mercury is 0.0001815, or 1-5510. Thus, a cubic centimeter of air, upon the application of one degree Centigrade of heat, will expand about twenty times as much as an equal volume of mercury. Besides this, a greater quantity of air than mercury can be conveniently utilized for expanding.
A simple air thermometer can easily be made. The materials needed are: A thin, hollow sphere or bulb of glass, about two inches in diameter, having as an outlet a glass stem from eight to twelve inches long, of about one-eighth inch inside diameter. A bottle of considerable weight, about three inches in diameter and from three to five inches high, is necessary. ( Any ordinary rather small bottle will do.) This should be half filled with eosin solution or otherwise colored water. A cork stopper for the bottle, having a hole through it large enough to admit the glass stem. The stem must now be partially filled with the eosin solution. This can be done by warming the sphere with the hand, and holding the end of the stem under the surface of the liquid. Some of the expanded air is expelled, and when the hand is removed from the bulb, the eosin solution rises gradually in the tube to fill the sphere made vacant by the contraction of the cooling air. If temperature changes not far from the normal are to be registered, the eosin should stand finally at somewhat over half way up the tube. It is rather difficult to reach a satisfactory result sometimes, and several trials may be necessary. They are easily repeated, of course, for the liquid already in the tube can be driven out by warming the bulb again.
Two grooves, running lengthwise, should be cut into the sides of the stopper to provide for free communication between the air in the bottle and the outside atmosphere. It is essential that the bottle should not be corked airtight, since this condition would cause a counter pressure of the air in the bottle whenever the air in the bulb is expanded. When the cork stopper has been put in, and the stem of the glass sphere inserted so that the end of the tube is under the surface of the liquid, the air thermometer is complete. A scale of degrees marked on cardboard may be put back of the tube, or the gradations may be scratched on the glass itself, but the readings will be inaccurate, for they will vary with every barometric variation, since the air pressure on the liquid in the bottle fluctuates. In only a modified and rather complex form can the air thermometer be relied upon for exact measurement.
The delicacy in action of the air thermometer makes it very useful in detecting sudden local changes in temperature. Interesting experiments can be performed with it; for instance, if a piece of filter paper saturated with ether is placed on the bulb, the eosin quickly rises because of the heat absorbed in evaporation. Because of its inconsistencies in readings, however, it is wrongly named as a definite measurer of temperature, for it is really only a thermoscope.
Fig. 171 - A home-made air thermometer.