6.3.1 Latent heat
The origins of Black's interest in the phenomenon of melting have been the subject of some debate. John Robison remarked, in his edition of Black's lectures, that Black had been struck by the simple fact that snow does not melt instantly on a sunny winter's day nor does a sharp night-time frost cause ponds to form thick layers of ice immediately (Robison, 1803, vol. 1, pp. xxxvi-xxxvii). It is now generally agreed, however, that Black's interest in heat arose from his study of the temperature changes which take place when salts dissolve in water. Some salts give out heat, while others produce cold, and these differences forced him to think about the more general question of aggregation and heat.
Several scholars, notably Henry Guerlac (1982, pp. 15–16), regard Black's reflections on the observation of supercooling by Daniel Fahrenheit (1686–1736) as the crucial factor. (Supercooling is the phenomenon whereby the temperature of undisturbed chilled water can fall below 32°F without freezing, but when the water is shaken, the thermometer rises to 32°F and remains there until all the water has frozen.) Arthur Donovan (1975, pp. 224–5) argues that Black would have perceived a link between the fixing of ‘air’ by quicklime and the fixing of heat (so that it is no longer registered by the thermometer) by ice.
However he came to the question of why ice does not melt immediately the temperature rises above freezing, Black's experimental programme is clear. If the temperature – as measured by a thermometer – does not change while the ice is melting, can we be sure that the thermometer bears any relationship to heat at all, and if the temperature does not change, how can we measure the quantity of heat taken up by the ice? Black was able to confirm that a mercury thermometer was a reasonably accurate record of heat changes when no change of state occurred, by mixing equal volumes of hot and cold water and assuming that the temperature of the mixture was the average of the initial temperatures.
But how could the heat entering the melting ice be measured with the thermometer? Fortunately, Black recalled an experiment that a Scottish physician George Martine (1702–41) had published in 1740. He had put two thin glasses, one containing water and the other mercury, in front of a fire; if the fire is a steady one, the quantity of heat entering each vessel should be the same. Black adapted the idea by measuring the rise in temperature of water in one glass, while ice was melting in another one.
He had to wait for the winter to arrive so he could obtain the necessary ice, and the key experiment was made in December 1761. One glass contained water that had been frozen using a snow and salt mixture and the other held water that had been chilled to 33°F; the room temperature was 47°F. After half an hour, the water temperature had risen to 40°F, but the ice took ten and a half hours to reach the same temperature. Black calculated that the extra heat required to melt the ice – its latent heat – was equal to the heat required to raise the temperature of the water by 140°F. The term ‘latent heat’ was devised by Black from the Latin latet, ‘hidden’ (Robison, 1803, vol. 1, pp. xxxvii).
He then carried out a different experiment, which he later described as an ‘obvious method’ (Black, 1803, vol. 1, p. 122). He made a small block of ice, which was placed in hot water. Within a few seconds, the ice had melted and the temperature of the water had fallen from 190°F to 53°F. The ice, the mixture of melted ice and water, and the empty glass were all weighed. With this information, Black recalculated the latent heat of ice and the result this time was 143°F. The average was therefore 141.5°F, or 330 KJ/Kg in the modem SI system, close to the currently accepted value of 336 KJ/Kg.