C

B

D

by pressure alone at ordinary temperatures, since the critical temperature is 31°, this is allowed to evaporate quickly by reducing the pressure, and the latent heat absorbed by a portion of the liquid evaporating reduces the temperature of the remainder to such an extent that it solidifies. On mixing the solid carbon dioxide with ether, a temperature of -77° C. is obtained. This temperature is below the critical temperature of ethylene (13°), so that on pumping ethylene gas into a receiver, immersed in the solid carbon dioxide and ether, liquid ethylene is produced, and can be stored as a liquid, if the receiver is kept packed in ice. The liquid ethylene is passed from the receiver A (Fig. 189), through a fine spiral copper tube B, which is immersed in a bath of solid carbon dioxide and ether. The liquid, cooled down to about -70°, then passes into the thin glass testtube c, which is surrounded by another glass vessel D, the two communicating through a side-hole E. The tube F is connected to an exhaust-pump, which draws off the gaseous ethylene as it is produced. The liquid ethylene, vaporising rapidly, produces a temperature of about 150° C. The gas to be condensed is contained in a steel bottle O, compressed to a pressure of about 200 atmospheres, and is conducted by means of a tube G to the strong glass test-tube H, which is surrounded by the boiling ethylene. Under the influence of the low temperature and the high pressure the gas condenses, and collects as a liquid in the bottom of the tube H. The temperature at which condensation takes place is obtained by means of a hydrogen thermometer, a thermo-electric junction, or platinum thermometer T. By allowing the liquid oxygen which forms in H to evaporate rapidly, by releasing the pressure, or even connecting G with an exhaust-pump, a yet lower temperature can be obtained.

W

V

C

FIG. 190.

(From Newth's "Chemistry.")

As an example of another class of apparatus for the liquefaction of gases, we may take that used by Dewar, and shown in Fig. 190. This form of apparatus depends, for the production of a low temperature, on the fact that when a gas expands against pressure it does work, and hence becomes cooled (§ 252). Thus when a gas is allowed to escape into the atmosphere from a receiver, in which it has been compressed, it has to force back the atmospheric pressure, and in doing so becomes cooled. The inside of the apparatus is first about half filled with liquid carbon

dioxide through the inlet on the left, the supply being regulated by the valve W, which is worked by the head B. This liquid can evaporate freely, and in doing so becomes so much cooled that the remainder solidifies. The oxygen, which is stored under pressure in a steel cylinder, enters the apparatus by the right-hand inlet, and passes up through the tube O. It then passes round the spiral s, which is immersed in the solid carbon dioxide, and thus becomes cooled to about -70. Next the oxygen passes down the spiral tube D to the tube U, in the side of which there is a very small jet, which can be closed by the rod v and screw A. The compressed gas escaping by this jet and expanding becomes cooled, and this cooled gas passes up, as shown by the arrows, between the spirals of the tube D, through which the oxygen is descending, and then escapes

into the air. In its

passage up between the spirals, the cooled oxyE gen cools the spirals and the contained oxygen, so that the oxygen escaping at the jet becomes colder and colder. Each portion of oxygen as it travels down the spiral is cooled down by the escaping gas to the temperature this has acquired by its expansion at the jet, and this oxygen, when it in turn reaches the jet and expands, becomes further cooled. This regenerative process goes on till the escaping gas

yet

at the jet is cooled down to its liquefying point, when liquid oxygen collects in the vessel G. This vessel is of particular construction, so as to reduce the conduction of heat from surrounding objects to the liquefied gas to a minimum. It consists of a double-walled glass test tube, the space between the walls being exhausted to the highest attainable vacuum. In such a "vacuum vessel," particularly if the outside is silvered, so as to be a very bad absorber of radiant heat (§ 246), it is possible to preserve liquid air for many hours. A third method of liquefying such gases as oxygen and air will be described in § 254.

If oxygen is caused to evaporate rapidly, by connecting a closed vessel containing the liquid to an exhaust-pump, such a low temperature is obtained that the air in contact with the vessel containing the boiling

liquid oxygen is liquefied at the ordinary pressure, and may be collected in a vessel placed to catch it as it drips down.

By allowing hydrogen which was cooled to 205° C., by passing first through a coil in a vessel B (Fig. 191) containing solid carbon dioxide, then through a coil in a vessel C containing liquid air, which was caused to boil rapidly by reducing the pressure, and under a pressure of 180 atmospheres to escape through the nozzle G of an apparatus somewhat similar to that shown in Fig. 190, the vessel D being itself placed in a space kept below - 200°, liquid hydrogen has been found by Dewar to collect. The liquid hydrogen was thus collected in the form of a liquid even at atmospheric pressures. By introducing a glass tube filled with helium into the liquid hydrogen, a distinct drop of liquid, presumably helium, formed in the tube. Thus all the known gases have been condensed into liquids, and the term "permanent gas" has no meaning.

T

CHAPTER IV

CONDUCTION OF HEAT

236. Transference of Heat.-When defining the higher of two temperatures, we said that if, when two bodies are brought near each other, heat passes from the one to the other, the one from which the heat passes is said to have the higher temperature. We have now to consider the laws that govern the passage of heat from one body to another. Heat may be propagated in three ways. In the first place, heat may travel from one portion of matter to another by what is called radiation, and in this process the transference can take place without the intervention of matter. It is by radiation that heat (and light) reach us from the sun. In the second place, heat may be propagated by the actual visible transference of matter, as in the case when a building is heated by the flow of hot water through pipes. This method of propagation is called convection. Thirdly, heat may be propagated by conduction. In this case the heat is conveyed by matter, but no visible motion of the matter itself takes place; the heat is usually considered as propagated by the warmer molecules heating the neighbouring colder molecules, and so on. Thus, when one end of a metal rod is placed in a flame and the other is placed in melting ice, it is found that heat is conducted along the rod, causing the ice to be melted.

237. Conduction. In order to define the conductivity for heat of a body, let us suppose we had a slab of the material of thickness d, with parallel faces each of area A, and that the opposite faces are kept at the temperatures, and to respectively. Then heat will be conducted by the material of the slab from one face to the other. Let Q units of heat pass from one face to the other through the slab in a time 7. Then it is found that

Q=kA(ty-t1)r

d

where is a constant for any one substance, independent of the thickness, area of the faces, and the difference of temperature (so long as this is not too great), but varies from one substance to another. If we make each of the quantities A, d, T, and the difference of temperature (t − t) unity, we have that is equal to the quantity of heat which would pass

1 When considering the subject of light, we shall show that the energy, in the case of radiant heat, is propagated by a wave motion in the ether.

between the opposite faces of a slab of the material of unit area and of unit thickness in unit time, when the temperatures of the faces differ by unity. The quantity k is called the thermal conductivity of the substance. The difference of temperature between the faces, (t-t1), divided by the thickness, gives the change of temperature per unit length in the direction in which the heat is flowing, and is called the temperature gradient. The rate at which the temperature of a body, say a metal rod, rises when it is heated at one end, depends not only on the conductivity of the material, but also on the specific heat. Let c be the specific heat of the material and p its density, then the heat required to raise the temperature of unit volume through one degree is cp. Now the thermal conductivity k is the quantity of heat which would pass through a slab of the material of unit thickness and unit cross section, in unit time, when the temperatures of the two faces differ by one degree. This quantity of heat would raise the temperature of unit volume of the material through where t is given by the equation

The quotient k/cp, or the coefficient of conductivity divided by the heat required to raise the temperature of unit volume through one degree, is called the diffusivity or coefficient of thermometric conductivity of the material.

238. The Measurement of the Conductivity of Solids. If one end of a long bar is heated, and a series of thermometers are placed in small holes drilled in the bar, the readings of the thermometers will increase. The thermometer nearest the heated end will rise first, the others following in succession. After a time the temperature of all parts of the bar will become constant, but diminishing gradually from the heated end to the other end. When this occurs, the heat supplied to the bar at the hot end during each second is exactly equal to that lost by radiation and conduction from the sides and the cold end. Let a curve AB (Fig. 192) be drawn such that the abscissæ represent distances along the bar, measured from the heated end, and the ordinates represent the corresponding temperatures. If we consider two cross sections of the bar at M and N, the temperatures at these points being represented by MR and NS, the difference in temperature between these two sections is equal to MRNS or to RP. Of the heat which crosses the section of the bar at M, part is conducted on and crosses the section at N, while the rest is radiated from the outside surface of the bar between the two sections. By taking the distance MN between the two sections sufficiently small, the proportion of heat lost by radiation from the edges of this small section of the bar bears so small a proportion to the heat conducted through the section, that we may neglect it. Also, since the points R