4

The Sceptical Chymists

The different quantities of the same element contained in different molecules are all whole multiples of one and the same quantity, which, always being entire, has the right to be called an atom.

Stanislao Cannizzaro1

Newton was nothing if not ambitious. The Mathematical Principles set the foundations for a classical mechanics that could be applied to matter in all its forms. But missing from its scope was that other ‘everyday’ phenomenon, light. He set out to remedy this situation in 1704, with the publication of another treatise, titled Opticks.

At this time there were two competing theories of light. Newton’s Dutch contemporary Christiaan Huygens had argued in favour of a wave theory in a treatise published in 1690. According to this theory, light is conceived to be a sequence of wave disturbances, with peaks and troughs moving up and down much like the ripples that spread out on the surface of a pond where a stone has been thrown.

But it’s obvious that waves are disturbances in something. Throwing the stone causes a disturbance in the surface of the water, and it is waves in the water that ripple across the pond. What, then, were light waves meant to be disturbances in? The advocates of the wave theory presumed that these must be waves in a tenuous form of matter called the ether, which was supposed to fill the entire universe.

Newton was having none of it. Tucked at the back of the first edition of Opticks is a series of sixteen ‘queries’, essentially rhetorical questions for which Newton provides ready answers. By the time of the publication of the fourth edition of Opticks in 1730, the number of queries had grown to thirty-one, the later additions each running to the length of a short essay. Much of Opticks is concerned with the properties of light ‘rays’, and in the very first definition Newton makes it clear that he regards these as discrete things: the ‘least parts’ of light.2

This leaves their status somewhat ambiguous, but in Querie 29 Newton makes clear what he thinks they really are: ‘Are not the Rays of Light very small Bodies emitted from shining Substances?’3 This ‘atomic’ description of light is rather less ambiguous, and has the virtue of bringing light within the scope of his mechanics.

But it is Newton’s Querie 31 that allows us a further glimpse of his audacity. He observes that large objects act on each other through forces such as gravity, electricity, and magnetism, and he ponders on whether these or indeed other, unknown, forces of attraction might also be at work on the ‘small particles of bodies’.4

Newton was ready to take the step that the Greek atomists had thought unnecessary. He was ready to replace the notion that the atoms interacted and joined together because of their shapes with the alternative idea that atoms interact and combine because of forces that act between them.* He refused to speculate on precisely what kinds of force might be involved, or how these might work. But in posing the question he opened the door to the idea that the motions of atoms and their combinations might be governed by forces that could be already known to us—gravity, electricity, and magnetism. Newton also dabbled in alchemy, and it is clear from the way that Querie 31 unfolds that he suspected that such ‘atomic’ forces, working with great strength at small distances, are responsible for the great variety of chemical reactions.5

We shouldn’t get too carried away. Our hindsight shouldn’t blind us to the simple truth that there was no empirical evidence to support Newton’s suspicions. He had sought to account for already-known chemical facts based on an atomism now augmented by the notion of some form of inter-atomic force. This was an idea, not a theory, and he was consequently unable to use it to rationalize any existing chemical facts except in the most general terms. He was certainly unable to use the idea to make any testable predictions.

Any and all arguments relating to the properties and behaviour of atoms remained firmly limited to speculation, of the metaphysical kind. Although Newton’s Querie 31 might seem to us tantalizingly prescient, there was nothing in it that could help the new breed of eighteenth-century scientists pursue these ideas any further through observation and experiment. They simply didn’t have the wherewithal.

Instead, the founding fathers of modern chemistry simply went about their business, unravelling the mysteries of the deliciously complex chemical substances, their compounds, their combinations, and their reactions, that were now being revealed in the laboratory. They did this without much (if any) acknowledgement of the existence and likely role of the atoms of the mechanical philosophers. Their purpose was rather to try to make sense of this complexity and establish some order in it from the top down, and to develop and elaborate some fundamental (though entirely empirical) chemical principles based on what they could see, and do.

Central to the chemists’ emerging logic was the concept of a chemical element. In the seventeenth and eighteenth centuries, the term ‘element’ had the same kind of position in the hierarchy of material substance as the ancient Greek elements earth, air, fire, and water. Of course, the chemists had by now determined that there was much more to material substance than these four elements. They had adapted the term to mean individual chemical substances that cannot be decomposed into simpler substances. And, although elements can be combined with other elements or compounds in chemical reactions, they do not thereby lose their identity, much as Boyle had explained in The Sceptical Chymist, first published in 1661.6 The metallic silver, recovered in its ‘pristine state’ following the series of chemical manipulations described by Sennert, is an example of a chemical element.

Could chemical elements nevertheless be atoms? Recall that the atomic theory of Boyle and later of Newton afforded atoms spatial extension (although distinct shapes no longer featured), hardness, impenetrability, motion, and inertial mass. Querie 31 hinted at a possible role for some kind of force between them, but it didn’t ascribe to them any chemical properties. It seems reasonable to suppose that, for those mechanical philosophers who dabbled in chemistry, it was anticipated that atoms sat somewhere even lower down in the hierarchy.

The unfolding history of eighteenth-century chemistry shows science doing what science does best: building a body of evidence that helps to dismantle previously received wisdom and replace it with new, more robust, ways of thinking about the world. However, this progress was (as it always is) rather tortuous. The chemists were practical men. (There were very few women engaged in this enterprise, except in supporting roles.) What they learned about chemistry would have very practical commercial implications and would help to establish the beginnings of the industrial revolution in the latter half of the eighteenth century. However, for the purposes of telling this story I will limit my choice of highlights to the efforts of the chemists to understand the different ‘affinities’ of the elements for each other, the ‘rules’ for combining them in chemical compounds, and thence to understand the nature of the elements themselves. As it happened, this turned out to be a lot easier when working with chemical substances in their gaseous form.

In the 1750s, whilst working as Professor of Anatomy and Chemistry at the University of Glasgow, Joseph Black discovered that treating limestone (calcium carbonate) with acids would liberate a gas which he called ‘fixed air’. This is denser than regular air, and so would over time sink to the bottom of any vessel containing regular air. It would snuff out a candle flame, and the life of any animal immersed in it (don’t ask). Passing the gas through a solution of limewater (calcium hydroxide) would cause calcium carbonate to be precipitated. We would eventually come to know this new gas as carbon dioxide.

The English chemist Joseph Priestley found that he could make fixed air by slowly dripping ‘oil of vitriol’ (sulphuric acid) onto a quantity of chalk (another form of calcium carbonate). In 1772, he published a short paper which explained how fixed air could be encouraged to dissolve in water, thereby producing ‘carbonated water’.

He speculated on its possible medicinal properties, suggesting that it might be useful as a means to combat scurvy during long sea voyages. (He was wrong.) Nevertheless, like natural spring water, such artificially carbonated water has a pleasant taste. The German-born watchmaker and amateur scientist Johann Jacob Schweppe used Priestley’s methodology to develop an industrial-scale production process and founded the Schweppes Company in Geneva in 1783.

Priestley went on to perform a variety of observations and experiments on different ‘kinds of air’ (different gases), published in six volumes spanning the years 1774–1786. These included ‘nitrous air’ (nitric oxide), ‘diminished nitrous air’ (nitrous oxide), ‘marine acid air’ (hydrochloric acid), ‘vitriolic acid air’ (sulphur dioxide), and ‘alkaline air’ (ammonia). But it was his experiments on ‘dephlogisticated air’ that were to resonate in science history.

According to the theory of combustion that prevailed at the time, all combustible materials were thought to contain the element phlogiston, released when these materials burn in air. The theory had been established in 1667 by the German alchemist Johann Joachim Becher: the term phlogiston is derived from the Greek word for ‘burning up’. So, when Priestley used a lens to focus sunlight on a sample of ‘mercurius calcinatus per se’ (mercuric oxide), he liberated a gas in which ‘a candle burned … with a remarkably vigorous flame’.7 Whatever this new gas was, it appeared to encourage a much more vigorous release of phlogiston, suggesting that it must be somehow more depleted of phlogiston than regular air. Priestley called it ‘dephlogisticated air’.

Priestley’s French contemporary, Antoine-Laurent de Lavoisier, disagreed with these conclusions. Lavoisier’s approach differed somewhat from the more descriptive or qualitative approaches of his fellow chemists, bringing to his science elements of quantitative measurement and analysis more typical of the mechanical philosophers. Specifically, Lavoisier took pains carefully to weigh the substances he started with and the substances produced subsequently in a chemical reaction. For reactions involving gases, this meant using sealed glass vessels that could trap the gas and hold it secure. In some 1772 studies of the combustion of phosphorus and sulphur, he had observed that these substances gain weight as a result of burning. This was rather puzzling if burning involves the release of phlogiston to the air.

It seems that Lavoisier learned of ‘dephlogisticated air’ directly from Priestley himself, during the latter’s visit to Paris in October 1774. He repeated and extended Priestley’s experiments, and a few years later published a memoir in which he argued that combustion does not involve the release of phlogiston, but rather the chemical reaction of the combustible material with ‘pure air’, which is a component of regular air. He called it oxygen.

Combustion, then, is a process involving the oxidation of chemical substances, a process which may occur spontaneously or with the assistance of heat or light. In 1784, Lavoisier showed that oxygen could be reacted with ‘inflammable air’ (which he called hydrogen) to form water, thereby demonstrating unequivocally that a substance that had been regarded as an ‘element’ at least since Plato’s Timaeus is, in fact, a chemical compound of hydrogen and oxygen.

In 1789, Lavoisier published Traité Élémentaire de Chimie (‘Elementary Treatise of Chemistry’), widely regarded as the first textbook on ‘modern’ chemistry. It contains a list of chemical elements which includes hydrogen, nitrogen, oxygen, phosphorus, sulphur, zinc, and mercury, organized into ‘metals’ and ‘non-metals’. The list also includes light and caloric (heat), still thought to be distinct elements at that time.

Alas, Lavoisier’s story does not end happily. He was a powerful aristocrat, and an administrator of the Ferme générale, essentially a private customs and excise (or ‘tax-farming’) operation responsible for collecting taxes on behalf of the royal government. His work on combustion had earned him an appointment to the Gunpowder Commission, which came with a house and laboratory at the Royal Arsenal.

The French revolution changed the political order in 1789, and the rise of Maximilien Robespierre led to the Reign of Terror four years later. The Ferme générale was particularly unpopular with the revolutionaries, and an order for the arrest of all the former tax collectors was issued in 1793. Lavoisier was sent to the guillotine in May 1794. He was exonerated eighteen months after his execution.

Lavoisier’s careful experiments had enabled him to establish an important principle. In a chemical transformation, mass (as measured by weight) is neither lost nor created: it is conserved. The total mass of the products of a chemical reaction will be the same as the total mass of the reactants. And, although they may become incorporated in different kinds of compounds as a result of some reaction, the identities of the chemical elements must also be conserved. This suggested rather strongly that the property of mass or weight might be traced to the individual chemical elements involved.

In October 1803, the English chemist John Dalton read a paper to the Manchester Literary and Philosophical Society, in which he famously observed that he had entered on the study of the relative weights of the ‘ultimate particles of bodies’, with remarkable success.8 It seems that Dalton was inspired by aspects of Newton’s atomism, but I’m going to take care to distinguish Dalton’s ‘ultimate particles’, or chemical atoms (the atoms of chemical elements) from the physical atoms of Boyle and Newton. As I noted earlier, the chemical atoms have chemical properties that never formed part of the mechanical philosopher’s atomic theory.

Dalton’s work on relative weights culminated in the publication of his New System of Chemical Philosophy in 1808. This work features a table that now extends to twenty chemical elements, including carbon (or charcoal), sodium (soda), potassium (potash), copper, lead, silver, platinum (platina), and gold. The table also includes ‘compound atoms’ (which, to avoid confusion, I’m hereafter going to call molecules), consisting of combinations of between two and seven chemical atoms, assembled in whole-number ratios.

He devised an elaborate symbolism to represent the chemical atoms, for example using ⊙ for hydrogen and ⚪ for oxygen, and representing a molecule of water as the combination of one atom of hydrogen with one of oxygen, or ⊙⚪.

Dalton had focused on relative weight. His French contemporary Joseph Louis Gay-Lussac observed similar whole-number regularity in the volumes of gases that were combined. He found, for example, that two volumes of hydrogen would combine with one volume of oxygen to produce two volumes of water vapour. This just didn’t fit Dalton’s recipe. If we write 2⊙ + 1⚪ gives 2⊙⚪ we see that this chemical equation doesn’t ‘balance’, there aren’t enough oxygen atoms on the left-hand side. Dalton was dismissive.

The Italian scientist Amadeo Avogadro was sufficiently compelled by Gay-Lussac’s work to turn these observations into a hypothesis (also sometimes referred to as Avogadro’s law), which he published in 1811: equal volumes of all gases, at the same temperature and pressure, contain the same number of atoms or molecules.* But confusion still reigned. Avogadro noted that hydrogen and oxygen combine in the ratio 2:1 (not 1:1 as Dalton had assumed), but the combination produces two volumes of water vapour. This could only be possible if the oxygen atoms were somehow divisible. Few took any notice, and those who did weren’t sure what to do with Avogadro’s hypothesis.

A few years later, the Swedish physician and chemist Jöns Jacob Berzelius worked to sharpen and extend Dalton’s system of atomic weights and introduced a chemical notation which, with one modification, we still use today. Instead of Dalton’s exotic symbols, Berzelius proposed to use a simple letter notation. Hydrogen is represented by H, oxygen by O. Berzelius wrote the 2:1 combination of these elements in water as H2O. Today we write it as H2O. Berzelius dodged the problem posed by the fact that two water molecules are produced by suggesting that Avogadro’s hypothesis applies only to atoms, not molecules.

This is how science works. A detailed examination of science history shows that discoveries are very rarely—if ever—‘clean’, with a single individual scientist or a small group of collaborators leaping directly to the ‘truth’. Instead, glimpses of the truth are often revealed as though through a near-impenetrable fog, with one scientist who has grasped part of the truth often arguing vehemently with another who has grasped a different part of the same truth. Progress is secured only when a sense of order can be established, and the fog is cleared.

Some historians have awarded Italian chemist Stanslao Cannizarro the role of clarifier, and there can be little doubting the clarifying nature of his treatise Sketch of a Course of Chemical Philosophy, published in 1858. At that time, Cannizzaro was Professor of Chemistry at the University of Genoa and had worked across all of chemistry’s emerging sub-disciplines, physical, inorganic, and organic. He is known today largely for his 1851 discovery of the Cannizzaro reaction, involving the decomposition of a class of organic compounds called aldehydes to produce alcohols and carboxylic acids.

In the Sketch, Cannizzaro synthesized information available from studies of the densities of gases, specific heat capacities (the different capacities of substances to absorb and store heat), and the burgeoning chemistry of organic compounds, particularly in relation to the elucidation of their chemical formulae. All this information afforded opportunities to deduce an internally consistent set of atomic and molecular weights. But first, Cannizzaro had to establish that Avogadro’s hypothesis makes sense only if we acknowledge the difference between atoms and molecules. He goes on:9

‘Compare,’ I say to them, ‘the various quantities of the same element contained in the molecule of the free substance and in those of all its different compounds, and you will not be able to escape the following law: The different quantities of the same element contained in different molecules are all whole multiples of one and the same quantity, which, always being entire, has the right to be called an atom.’

The mystery of Gay-Lussac’s measurements of the combining volumes of hydrogen and oxygen was now resolved. Setting the relative atomic weight of hydrogen equal to 1 based on hydrogen chloride (HCl), Cannizzaro observed that the relative weight of hydrogen present in hydrogen gas is actually 2. Hydrogen gas consists of molecules, not atoms: ‘The atom of hydrogen is contained twice in the molecule of free hydrogen.’10 Likewise, if the relative atomic weight of oxygen in water (H2O) is taken to be 16, this ‘quantity is half of that contained in the molecule of free oxygen’.11

Clearly, hydrogen and oxygen are both diatomic gases, which we write as H2 and O2, and the combining volumes now make perfect sense: 2H2 + O2 → 2H2O. The equation balances—four hydrogen atoms in two molecules of hydrogen gas combine with two oxygen atoms in one molecule of oxygen gas to give two molecules of water.

Of course, all these elaborate combining rules don’t prove the existence of chemical atoms, and some scientists remained stubbornly empiricist about them. But nevertheless when speculative theoretical entities are found to be useful across a range of scientific disciplines, over time scientists will inevitably start to invest belief in them.

In 1738, the Swiss physicist Daniel Bernoulli had argued that the properties of gases could be understood to derive from the rapid motions of the innumerable atoms or molecules in the gas. The pressure of the gas then results from the impact of these atoms or molecules on the surface of the vessel that contains them. Temperature simply results from the motions of the atoms or molecules. This kinetic theory of gases bounced around for a few decades before being refined by German physicist Rudolf Clausius in 1857. Two years later, Scottish physicist James Clerk Maxwell developed a mathematical formulation for the distribution of the velocities of the atoms or molecules in a gas. This was generalized in 1871 by Austrian physicist Ludwig Boltzmann, and is now known as the Maxwell–Boltzmann distribution.

The Maxwell–Boltzmann distribution can be manipulated to yield an estimate for the mean or most probable velocity of the atoms or molecules in a gas. For molecular oxygen (O2) at room temperature, the mean velocity is about 400 metres per second. This is roughly the muzzle velocity of a bullet fired from an average rifle. (Fortunately for us, oxygen molecules are a lot lighter than bullets.)

This is all very well, but we still can’t see these motions, and there may be other explanations for the properties of gases that have nothing to do with atoms or molecules. This final stumbling block was removed by Einstein in 1905, as I mentioned in Chapter 1. He suggested that tiny particles suspended in a liquid would be buffeted by the random motions of the molecules of the liquid. If the particles are small enough (but still visible through a microscope) it should be possible to see them being pushed around by the molecules (which remain invisible). He further speculated that this might be the explanation for the phenomenon of Brownian motion, but the experimental data were too imprecise to be sure.12

Better data were eventually forthcoming. French physicist Jean Perrin’s detailed studies of Brownian motion in 1908 subsequently confirmed Einstein’s explanation in terms of molecular motions. Perrin went on to determine that the value of what he suggested should be called Avogadro’s constant, which scales the microscopic world of atoms and molecules to the macroscopic world in which we make our measurements.* There could now be no doubting the reality of atoms.

One small fly in the ointment: British physicist Joseph John Thompson had discovered and characterized a new particle, which he called the electron in 1895. New Zealander Ernest Rutherford discovered the proton in 1917. It seems that evidence for the existence of atoms was being established just as physicists were working out how to take these same atoms apart.

Five things we learned

1. Newton speculated that forces (of an unspecified nature) acting between the atoms of chemical substances are responsible for the great variety of chemical reactions.
2. Priestley and Lavoisier helped to clarify the nature of chemical substances in terms of the different chemical elements they contain, such as hydrogen, carbon, oxygen, etc.
3. Dalton devised a system of chemical atoms, each with a different atomic weight. This was translated by Berzelius more or less into the system we use today, in which a molecule of water is represented as H2O, where H and O represent hydrogen and oxygen atoms.
4. Cannizzaro clarified the rules for combining different atoms to form molecules. Hydrogen and oxygen are actually both diatomic gases, H2 and O2. In the reaction between hydrogen and oxygen, two molecules of hydrogen combine with one of oxygen to form two molecules of water: 2H2 + O2 → 2H2O.
5. Einstein suggested, and Perrin demonstrated, that the Brownian motion of small particles suspended in a fluid is the result of the motions of the molecules of the fluid. It was finally accepted that atoms and molecules really do exist.
* In fairness to the Greeks, their atoms also ‘swerved’ and collided with each other, which I guess can be interpreted in Newton’s mechanics in terms of forces acting between them.
* Confusingly, Avogadro referred only to molecules, calling what we would now call atoms ‘elementary molecules’. Don’t worry, we’ll get there in the end.
* Avogadro’s constant is 6.022 × 1023 ‘particles’ (such as atoms or molecules) per mole. The mole is a standard unit of chemical substance defined in relation to the number of atoms in 12 grams of carbon-12. Thus, a mole of oxygen (which has a volume of about 22 litres) will contain about 602 billion trillion molecules. That’s a lot.