by Ben Russell
Just as the laboratory chemist spent much time labouring over the furnace, so great ‘bottle’ kilns, named after their shape, loomed large in the working lives of potters. Firing was the most risky part of the pottery process: wares were placed in the kiln packed inside saggars, containers made of fire-clay positioned to allow heat to circulate around them, and left for 50 hours or more. Some of Wedgwood’s work, ‘the cameos, intaglios, bas-reliefs, and the majority of the vases, lamps, and candelabra’, only needed a single firing.75 But other products, including the creamware, were referred to only as ‘biscuitware’ after its first firing because of the colour, porosity and absorbent qualities attained. After the first firing, the ware would be covered in glaze, a mixture of chemical ingredients, and then fired a second time. This time, the pieces were each separated with tiny stools or ‘spurs’ so that the glazed surfaces would not fuse together. And rather than a long, gentle cooling period, the kiln had to be opened at exactly the right moment to obtain the best effect from the glaze as its chemical constituents were transformed by the heat. Now the firemen rushed in, at risk of serious burns, and the wares were removed while still at high temperatures to be cooled, placed in storerooms and then packed, ready for dispatch to customers.
All applications of heat shared the problem of how precisely to measure the temperatures attained. A pottery kiln might have an eye-hole set into its wall, and by peering through it at the colour of the fire an experienced potter could gauge whether the temperature was right. The terms for describing the look of heat – ‘red, bright red, and white heat’ – were ‘indeterminate expressions’ with ‘numerous gradations, which can neither be expressed in words, nor discriminated by the eye’.76 But gauging the temperature wrongly could be costly: in firing his ‘Jasperware’ pieces, Wedgwood at one time suffered a 75 per cent loss rate, and the roads of the Staffordshire potteries were routinely repaired using misfired wares from the local kilns.77 Observations had to be within very close tolerances: Wedgwood once remarked of a firing, ‘Every Vase in the last Kiln were spoil’d! & that only by such a degree of variation in the fire as scarcely affected our creamcolour bisket at all.’78 And the same temperature had to be obtained repeatedly. Wedgwood complained that
we are told . . . that such and such materials were changed by fire into fine white, yellow, green or other coloured glass; and find, that these effects do not happen, unless a particular degree of fire has fortunately been hit upon, which degree we cannot be sure of succeeding in again.79
Accurately gauging temperature by eye under these conditions was a skill only obtained by hard-won experience over a period of years.
The problem of measuring temperatures was solved in philosophical chemistry by, for instance, using thermometers of the type made by Daniel Gabriel Fahrenheit from 1724, or by Anders Celsius from 1742. James Watt made them for Joseph Black, and lent him the dividing plate that he used to quickly mark them with graduated scales.80 However, delicate glass-and-mercury thermometers would not survive for long in a pottery kiln, where temperatures routinely passed above 1,000°C. Instead Josiah Wedgwood developed his ‘pyrometer’, which measured the amount clay will shrink as it is heated. Round clay tablets, each an inch across, were placed in a kiln alongside the wares being fired. At intervals, tablets were removed and placed in a scale consisting of ‘two pieces of brass, twenty-four inches long . . . divided into inches and tenths, fixed five-tenths of an inch asunder at one end, and three-tenths at the other, upon a brass plate’.81 The scale was calibrated for temperature on the basis of careful experiments; as each tablet would decrease in diameter by as much as one-quarter during a firing, the diameter read against the pyrometer scale told the fireman what the temperature inside was: the smaller the diameter, the higher the temperature.82
Wedgwood’s pyrometer embodies an interest in attaining the highest possible levels of precision across chemical experiments and processes. The need for precision extended beyond just the measurement of temperature. Equally careful attention was paid to recording changes in mass. As Black undertook experiments in the period from 1752 to 1755 which would lead to his discovery of carbon dioxide, he employed a balance that carefully measured the mass of the substances used to the nearest grain, or 64.8 milligrams.83 Henry Cavendish, working on the chemical composition of water in the early 1780s, could weigh small quantities to within six milligrams.84 The materials of pottery were also the subject of intense scrutiny. The old way of working, with wares made from clays ‘of the coarse yellow, red, black, and mottled kind . . . the body of the ware being formed of the inferior kinds of clay, and afterwards painted or mottled with the finer coloured ones mixed with water, separately or blended together’, was no longer adequate.85 There was now a move towards carefully obtaining the best available; Wedgwood specified precisely that clay obtained from Pensacola in North America ‘must be got as clean from soil, as if it was to be eat, & if they were to get several parcels, at different depths, & put them in separate Casks, properly number’d, I could by that means easily ascertain what depth of the mine is best for our purpose’.86
A Wedgwood pyrometer set, 1786.
And chemists could play a key role in validating the quality of such materials; as William Cullen wrote, ‘the philosophical chemist is the assay master to arts in general. He brings all new projects to the test, & . . . works trials in small for their improvement.’87 Although new means of precision measurement were becoming available, their use required more trust to be placed in those instruments than their users were sometimes willing to give. Michael Combrune introduced thermometers into brewing, publishing an account of how he did so in 1758, but his father objected to such ‘experimental innovations’ and Combrune had to keep the offending instruments hidden away.88 The senses remained the chemist’s preferred diagnostic tools across a spectrum of potential projects.89 In his ‘Brief Analysis of Water by Precipitants’ (1810), George Smith describes how ‘Good water has no smell; such as abounds in aerial acid, diffuses a penetrating odour. Such as contains any sulphur, yields a smell resembling that of rotten eggs.’ Taste was also employed:
Acid occasions a gentle pungent acescent taste; a bitterness accompanies those waters which contain . . . nitre, or magnesia; a slight austerity proceeds from lime and gypsum; a sweet astringency from alum; a saltishness from common salt; a lixivious flavour from alkali; a bitter astringency from copper; and an inky taste from iron.90
When Watt and Black corresponded about corrosion inside a pumping engine caused by the water, Black concluded that ‘the Water . . . contains a quantity of fixed air tho only a small quantity . . . it is perceptible by . . . a person who has a delicate taste’.91 The senses were amplified by specialized laboratory equipment, but they remained the foundation of much chemical work into the twentieth century.
In this respect the experience in pottery matched that in philosophical chemistry. The success of pottery really was in the hands of the potter; after being away from the wheel for even a short time, the ‘essential smoothness and flexibility’ of his hands was lost, and it would take them a number of days to become reacquainted to the feel of the clay.92 Pressing delicate relief detail, handles and decorations onto a piece of pottery without applying too much pressure took a well-trained, delicate touch. Even the process of applying glazes depended entirely on the dexterity of the dipper, immersing each piece up to his or her elbow, remembering the rule that ‘the portion of the article which enters first shall come out first, so that each portion shall receive the same amount of glaze.’93
This is not entirely to play down the role of new chemical equipment. Antoine Lavoisier’s laboratory contained around 6,000 pieces of glassware alone, for example, and improved apparatus was manufactured in both glass and ceramic, able to withstand corrosive substances, sudden changes in temperature, contain ‘elastic vapours’ and even bear violent heat ‘without melting or being otherwise injured’.94 Alongside his vases and dinnerwares, Josiah Wedgwood made a popular range o
f ‘Useful Wares’ for chemists: pestles and mortars for mixing and grinding purposes, evaporating pans, basins, funnels, siphons, retorts and tubes.95 Watt’s workshop contains some of these; beneath the central workbench rest long earthenware tubes used for experiments in which gases were passed over red-hot solids, and a pair of jasperware mortars sit on the open shelves.96 The manufacture and use of this apparatus also suggests further crossovers between experimental chemistry and industry of a more robust nature: the mechanical means by which chemical solids were broken into fragments or ground to powders in the laboratory involved ‘mortars, sieves, files, the hammer & anvil, &c.’ ‘Cementation’ of materials – laying them layer upon layer so that they reacted together – was named after a ‘resemblance between the arrangement of the materials, and the laying of bricks in mortar’.97 And the capabilities of industrial workshop practice impinged directly onto the manufacture of some experimental apparatus. Joseph Black complained about having to make his portable furnace ‘of the simplest rectilinear shapes, because workmen find great difficulty in executing curved and uncommon forms; and not one of a score of them will do it with accuracy’.98
Just as there were extensive crossovers between the practices of philosophical and industrial chemistry, so the equipment used by Watt and Black in their experiments on heat bridged both worlds. Watt’s work on the engine required his models and small full-size prototypes, as well as some specially commissioned instruments to measure the quantity of heat contained within steam, for example. But in addition to these, he employed readily available chemical balances to measure the amount of fuel burned and water evaporated, thermometers and barometers for measuring temperature and air pressure and, most prosaic of all, a ‘common tea kitchen’ to generate steam for his experiments, while Black’s laboratory at Glasgow ‘extended little beyond basic apparatus such as furnaces, stills and glassware’.99
To summarize, then, chemistry in many areas of experimentation and application was based on the accurate manipulation and measurement of heat. These techniques evolved from being based on the senses – although this tradition would long endure in other aspects of chemistry – to being approached in a more carefully quantifiable way. Though these developments were played out in potteries and chemical laboratories, they were of profound significance to the development of the steam engine, and the key players in realizing this were Black and Watt. Having outlined how heat was seen as a chemical substance by both men, and their contemporaries, we will assess their perceptions of its properties and how they could be applied to the engine.
Although their philosophies of the chemical nature of heat were similar, Black and Watt approached the nature of heat from subtly different starting points – Watt with the particular objective of making the engine work, and Black from the perspective of explaining philosophically how a substance absorbs heat. Black carried out his research before Watt, who referred to Black to validate his findings. But this is not to say that the two had a ‘master and pupil’ relationship. Far from it: Watt’s ideas about heat progressed beyond Black’s.100 The historian Arthur Donovan claimed that Black’s move from Glasgow to take up a post at the University of Edinburgh in 1766 was ‘the end of his career as a creative philosophical chemist’.101
Black wrote to Watt in March 1772, ‘I have no chemical news, my attempts in Chemistry at present are chiefly directed to the exhibition of Processes and experiments for my Lectures, which require more time and trouble than one would imagine.’102 Black’s interest in heat, despite his close associations with industry, would always remain primarily an interesting subject for lectures and debates. By contrast, in 1766 Watt was only a short way into the engine project.103 He would make a much greater personal commitment to understanding the engine, and stood to lose out in a very real way if his understanding faltered. As he later wrote, ‘I, having become the father of a family, and loaded with cares of many kinds, had less time for mere philosophical conversations.’104 Black had begun his research on heat in 1755 when he was first appointed to the college at Glasgow. It is traditionally said that he first turned to heat as a fruitful area of research after considering that snow is not melted instantly on a sunny winter’s day but lingers for some time, and that an overnight frost does not immediately cause ice to form – both phenomena that would be widely visible during a Glaswegian winter.105 His observations showed that while ice melted, its temperature remained constant, and this raised uncertainties over the relationship between temperature and heat: if the former did not necessarily reflect the latter, how could the amount of heat absorbed by ice be measured? Black devised an experiment to find out in December 1761: he placed two glasses in front of a fire, one filled with water at a temperature of 33°F and the other with ice. With the room temperature standing at 47°F, the water temperature reached 40°F, exactly halfway between its original temperature and that of the room, in 30 minutes. The ice, however, took over ten hours to reach the same temperature, during which time, Black calculated, the extra heat it absorbed would have heated the water by 140°F. Black called this considerable increment the ice’s latent heat. Black then turned his attention from melting ice into water, to turning water into steam. Noting that it takes longer to entirely boil off a quantity of water than it did just to raise it to boiling point, he found that a pan of water with a temperature of 50°F took four minutes to reach boiling point, but then another twenty minutes to boil off entirely – and this time the latent heat of the steam generated was calculated as being 810°F. Black quickly realized that steam was a tremendously effective medium for storing heat.
Watt’s experiments with engines led him to similar conclusions. The atmospheric engine used a piston that could slide up and down inside an open-topped cylinder. Steam was injected beneath it, and the piston rose. Then cold water was injected below, and atmospheric pressure pushed the piston down into the partial vacuum thus created underneath. When working on Anderson’s model engine, Watt found that at each working stroke, the amount of steam required amounted to several times the volume of the cylinder, and that ‘an enormous quantity of injection water’ was needed to subsequently condense it and create the working vacuum.106 This latter was, Watt subsequently concluded, because of the immense amount of heat that the steam contained: he found that one volume of water turned to steam could boil six times the initial volume of water.
As for the amount of steam required being considerably more than the volume of the cylinder, this brings us to the capacity a body has to absorb heat, or its specific heat. As Watt described it, ‘all bodies are sponges in respect of heat and some bodies can contain more of it and some others less.’107 A long series of experiments with his model and prototype engines showed Watt that excessive steam consumption was caused not, as he initially thought, by heat escaping through the cylinder walls, but by the metal cylinder itself having to be warmed prior to every working stroke. Because the cylinder of a steam engine could easily consist of several tonnes of cast iron, it had the capacity to absorb a large quantity of heat that was wasted as soon as cold water was sprayed inside the cylinder to form a vacuum.
The nature of specific heat had also been explored through smaller-scale experimental means. Daniel Fahrenheit had found that when equal volumes of water and mercury were mixed together, if the mercury was initially hotter than the water, then the temperature of the mixture was less than the average of the two, but if the water was hotter then the opposite was true. Scottish physician George Martine had also found that when two glasses of equal volume, one containing water and the other mercury, were placed in front of a fire, the temperature of the mercury rose faster. Black solved the puzzle posed by both experiments by suggesting that mercury had a lower capacity to absorb heat than water, meaning it could be heated and cooled more rapidly.
Watt’s responses to the nature of specific and latent heat shaped the design of his steam engine. His initial approach to the heat wasted by the metal cylinder being alternately heated and cooled was to p
ropose that it be built from a material that would absorb less heat, like wood. But later he realized that the problem could be entirely avoided by condensing the steam in a vessel separate from the cylinder – the separate condenser. This way, the cylinder could be kept at a constant high temperature, reducing the waste of heat to an absolute minimum, and the condenser could be kept as cool as possible to create the vacuum that the engine depended upon to operate. As for the movement of the steam between the cylinder and condenser, Watt appreciated that steam, perceived as a chemical compound of water and heat, was an elastic substance – more so as the amount of latent heat was increased.108 He realized the implications of this while walking on Glasgow Green in May 1765, later writing how ‘the idea came into my mind, that as steam was an elastic body it would rush into a vacuum, and if a communication was made between the cylinder and an exhausted body [containing a vacuum], it would rush into it, and might there be condensed without cooling the cylinder’.109 With the benefit of hindsight, we can say that Watt had grasped the principle which every condensing steam engine subsequently built would depend on.