The absolute or thermodynamic scale of temperature
If volume-temperature and pressure-temperature graphs for a gas are plotted on extra large sheets of graph paper and then produced or extrapolated backwards they cut the temperature axis at – 273°C (Fig. 16.8 and 16.9). This suggested to the early
experimenters that – 273°C might be the lowest temperature attainable or the absolute zero. This must, however, be only an assumption, since we know that gases liquefy in most cases long before such a temperature is reached.
Thus by shifting the vertical axis 273°C to the left and renumbering the temperature scale, – 273 °C becomes zero, the ice point 273 and the steam point 373 on what is called an absolute scale of temperature. This comes very close to the thermodynamic scale of temperature proposed on theoretical grounds by Lord Kelvin about the middle of the nineteenth century. Temperatures on this scale are given in units called kelvins, instead of degrees. They are represented by T and denoted by K. Also the scale is defined in such a way that kelvins are equal to Celsius degrees in size.
It follows that temperatures on the Celsius scale can be converted to kelvins simply by adding 273. Thus 0 °C = 273 K, 8 °C = (273 + 8) K and so on.
It can be shown that the thermodynamic scale can be related to the products p
for a perfect or ideal gas, i.e., one which obeys Boyle’s law exactly. In the case of practical gas thermometers it is difficult to measure both p and V at the same time, so it is usual to keep V constant, in which case the temperature will be proportional
to the pressure of the gas. This explains why Jolly’s apparatus of the previous experiment is also called a constant-volume air thermometer. Now air is not a perfect gas any more than any other gas, but the one which comes closest to this requirement is hydrogen. An improved form of Jolly’s apparatus is called the standard gas thermometer, descriptions of which are to be found in
more advanced textbooks. This thermometer needs laborious corrections and is far too difficult and cumbersome for ordinary day-to-day use. When very high temperatures are to be measured it is filled with nitrogen and is employed only for the
purpose of obtaining accurate values for a number of other fixed points both above 373 K and below 273 K, e.g. the freezing-point of gold (1336 K) and the boiling point of oxygen (90 K). In practice, therefore, such instruments as the platinum resistance thermometer, thermos-couples, and so on are standardized by the fixed points so obtained and used to measure temperature over a wide range where mercury thermometers are unsuitable.