by Brian Clegg
What made me special – particularly demonic, I guess you would say – is that I can do my business of sorting the thermodynamic sheep from the goats using the door without the need to put energy into the system. I am operating a frictionless, inertia-free door (not available at your local DIY store – this is a thought experiment) and no energy is added to the system by my actions. If you aren’t comfortable with a frictionless door that has no mass, because that could never exist in the real world, bear in mind that this is just a convenience. The only essential as far the second law is concerned is that I do not put any energy into the system of molecules, which I don’t.
So how do I perform my trick? How could someone, even with my unrivalled brilliance, break the unshakeable second law of thermodynamics? That was James Clerk Maxwell’s challenge to the world – and himself. It’s perhaps the only physics challenge he ever took on and failed at. As did his friends, like William Thomson. I was a conundrum. They couldn’t see how it was possible for me do my job, yet equally they couldn’t see why I would fail. Some people tried to argue that I was pointless – me, pointless! – because in the real world there couldn’t be such a demon. But physics doesn’t work like that. If a law’s a law, it should hold up, whatever you throw at it. And I managed to beat it.
Or so it seemed back then, though I would have to face some challenges further on along the way. But I suppose we need to get back to my creator to see what happened when he ventured to the mighty metropolis that was London in 1860.
Notes
1 – Maxwell’s postcard to Peter Tait from London, dated 23 October 1871, featuring ‘O T′! R. U. AT ’OME?’ is quoted in Peter Harman (ed.), The Scientific Letters and Papers of James Clerk Maxwell, Vol. 2 (Cambridge: Cambridge University Press, 1995), p. 682.
2 – Thomson’s statement that the second law made the end of the world inevitable is quoted in Stephen Brush, The Kind of Motion We Call Heat: A History of Kinetic Theory in the 19th Century (Amsterdam: North Holland, 1976), p. 569.
3 – Maxwell’s comparison of the second law to pouring a tumbler of water into the sea is from a letter to John Strutt (Lord Rayleigh), noted in Robert Strutt, John William Strutt: Third Baron Rayleigh (London: Edward Arnold, 1924), pp. 47–8.
4 – Maxwell’s first mention of the demon is in a letter to Peter Tait from Glenlair, dated 11 December 1867, reproduced in Peter Harman (ed.), The Scientific Letters and Papers of James Clerk Maxwell, Vol. 2 (Cambridge: Cambridge University Press, 1995), pp. 328–33.
5 – Maxwell’s description of the demon as intelligent and exceedingly quick was in a letter to John William Strutt from Glenlair, dated 6 December 1870, reproduced in Peter Harman (ed.), The Scientific Letters and Papers of James Clerk Maxwell, Vol. 2 (Cambridge: Cambridge University Press, 1995), pp. 582–3.
6 – William Thomson’s paper introducing the term ‘demon’ was William Thomson, ‘The Kinetic Theory of the Dissipation of Energy’, Nature, 9 (1874): 441–4.
* In letters, JCM tended to refer to people by initials. T was already allocated to William Thomson, so Tait became T′.
† It may seem there is no imaginable reason for having such a box – and this is often the case in physicists’ thought experiments – but, as it happens, you could imagine this as a simplified version of putting an icebox into a warm room.
‡ A little derogatory, I feel, that ‘who can do no work’. Perhaps it would have been better to have said, ‘who rightly felt that work was beneath him’.
§ This is better.
¶ The other side of the box is not just at room temperature – it warms up because the refrigerated side of the box has a radiator on the outside of it. Look at the back of any fridge.
|| Assuming that refrigerator manufacturers, other than those in the fiction of Terry Pratchett, do not, in fact, employ demons as the operative part of their devices.
Chapter 4
A capital adventure
Every university that Maxwell had attended or worked at up until now was an ancient institution, still clinging on to traditions and even aspects of the syllabus that dated back to medieval times. But his new academic home, King’s College, prestigiously located on London’s Strand, took its first students only in the year that Maxwell was born (remarkably, before the 1820s, London did not have a single seat of higher education). The university’s management prided itself on its modern values, with explicit courses in individual scientific disciplines, rather than traditional, classical-heavy, ‘bit of everything’ undergraduate courses; they even covered the upstart, hands-dirty topic of engineering.
Science at King’s
We can get a feel for Maxwell’s attitude to science and how it should be approached from part of his stirring inaugural lecture at King’s. He told the students:
In this class I hope you will learn not merely results or formulae applicable to cases that may possibly occur in our practice afterwards, but the principles on which those formulae depend, and without which the formulae are merely mental rubbish. I know the tendency of the human mind is to do anything rather than think. But mental labour is not thought, and those who have with labour acquired the habit of application, often find it much easier to get up a formula than to master the principle.
Maxwell was emphasising the importance of doing more than ticking the boxes and mechanically working the equations – as we now might allow a computer to do for us. His approach to science was always to look for underlying principles, to get as close as he could to the true ‘laws’ of nature.* To get a flavour of the duties expected of him, Maxwell was required to be in college three mornings a week (10.00am to 1.00pm) and to give one evening class aimed at working men – leaving a comfortable amount of time for working on his own projects.
His pay from the university was directly linked to the number of students he taught, receiving 5 guineas (£5.25) for each day student, 18 shillings (90p) for each evening class student and £2, 7 shillings and threepence (just over £2.36) for each ‘occasional student’, who did not matriculate but were enrolled in individual classes to gain expertise. This made him around £450 a year, the equivalent of around £39,000 now,† a little more than he had earned in Aberdeen. In practice, though, he was significantly better off, as one of the conditions of the Aberdeen merger was that he would receive a remarkably generous annuity for life of his salary when he lost his position, giving him an additional £400 a year, making his annual income a handsome £850 – around £76,000 in purchasing power or £579,000 in proportion to earnings.
Something of Maxwell’s approach to teaching while at King’s College can be seen in an incident recorded in the Campbell and Garnett biography written shortly after his death:
The professors had unlimited access to the [College] library, and were in the habit of sometimes taking out a book for a friend. The students were only allowed two volumes at a time. Maxwell took out books for his students, and when checked for this by his colleagues explained that the students were his friends.
While this early biography has a tendency to paint Maxwell in something of a saintly light, it illustrates here an unusual attitude for a professor at the time and perhaps reflects both Maxwell’s youth – at 29, he was still only a few years older than his students – and his unusual upbringing for someone of his class, having mixed with the working class far more than would be the norm for the landed gentry.
The students Maxwell taught at King’s College had a strong focus on practical, applied science – many of the young men in his physics and astronomy classes would become engineers, a subject that was hardly recognised at other universities, and they received training that would benefit them in such roles. This meant that their education was seen as more of a boost to their practical skills than a means of obtaining a degree,‡ and the majority did not stay for a complete three-year course, typically averaging just over four terms before moving on. The fees at King’s were over seven times those of Marischal College at £12, 17 shillings (£12.85) for each
of the three terms, though those wanting only to attend natural philosophy classes had a cut-price rate of just 3 guineas (£3.15) a term.
Maxwell and Katherine had a comfortable home to entertain guests at 8 Palace Gardens Terrace in Kensington (oddly, this is now number 16). It was a fair walk to reach the Strand, though Maxwell often would, when he wanted to exercise his country-bred legs with a four-mile stroll. The house may well have reminded him of his aunts’ houses in Edinburgh – Palace Gardens was a little grander, with five storeys and pillars at the entrance, but it was still a solid city townhouse.
After Aberdeen’s relative social backwater, Maxwell was looking forward to having more opportunity to meet with like-minded individuals, as had been possible in Edinburgh and Cambridge. Where the entire staff of Marischal College dealing with all disciplines numbered just twenty, Maxwell’s department of Applied Science alone, one of four at King’s College, had a similar number. And his colleagues were only the start.
Both the Royal Society and its more practically-minded little brother, the Royal Institution, offered lectures and discussions in the city attended by many of the leaders of British science – and in May 1861, Maxwell was thrust to the fore with an invitation to give a lecture at the Royal Institution (RI).
Maxwell in the 1860s, during his post at King’s College London or shortly afterwards.
Getty Images
Bring colour to the Institution
The RI was the spiritual home of Maxwell’s hero, Michael Faraday. As we have seen, Faraday started at the Institution as an assistant to Humphry Davy and now, though 70, Faraday was still associated with the venue where he had given so many lectures and set up the famous Christmas Lecture series for children. Following on from his Rumford Medal, Maxwell was asked to give a lecture at the RI on colour theory, a topic that would remain a lifetime interest for him.
One of the traditions of the Institution was to give demonstrations during lectures – the more dramatic presentations were popular draws, sometimes bringing in royalty among the audience. The spinning colour wheel that Maxwell had used in his experiments was too small for the audience to see it at a distance, so he decided to produce something that had never been seen before – he would project a large-scale image of a true colour photograph.
It was common enough for black and white photographs to be hand-tinted to give the effect of colour, but Maxwell’s plan was to produce a full colour image by combining three monochrome photographs taken and then projected using red, green and blue filters, demonstrating that these three primary colours were sufficient to generate all the colours that we see. He had first conceived this idea back in 1855, when he had briefly discussed it at the Royal Society of Edinburgh, but in the best manner of Royal Institution demonstrations, he intended to bring the theory alive before his audience.
Maxwell was no photographer himself – at the time, a distinctly specialist activity. But luckily, one of the country’s top experts, Thomas Sutton, himself a Cambridge Wrangler in his day, had been employed as the official King’s College photographer, a role that was primarily a teaching one. Sutton was able to help Maxwell with the tricky (and potentially dangerous) photographic medium of wet collodion. This was no simple matter of buying a roll of film and having it processed, let alone the ease of modern digital photography. The photographer had to be a deft chemist before there was any possibility of taking an exposure.
The wet collodion process started with cotton being soaked in a toxic and highly corrosive mixture of nitric and sulfuric acids. The product – gun cotton§ – had to be washed and dried before dissolving it in ether or alcohol to produce a terrifyingly flammable gel. The would-be photographer then added halogen salts (iodine or bromine) to the gel and spread the resultant suspension carefully onto a clean glass plate. This part of the process required particular skill to get a consistent, level layer of the ‘collodion’ gel. The plate was then dipped in a bath of silver nitrate, which reacted with the halogen to produce a light-sensitive silver-based coating. After this ‘activation’ process, the plate, still wet, was placed in the camera and exposed. Finally, the plate had to be developed, fixed, rinsed, dried and varnished before the final product was available. This was anything but ‘point and shoot’.
Under Maxwell’s direction, Sutton took three separate photographs of a multicoloured tartan-like ribbon,¶ using red, blue and green filters, made by mixing different coloured dyes in water and placing the camera behind glass troughs containing the liquids. Maxwell was then able to superimpose the three separate images projected onto a large screen at the RI with three magic lanterns, each shining through one of the red, blue and green troughs.
Making it happen on the night must have given Maxwell some bitten fingernails. His talk was a Friday Night Discourse, the most fearsome of Faraday’s public event programme. The audience were a stern mass of black ties and the speaker was required (as they still are) to crash through the doors of the lecture theatre on the second of the starting time and begin speaking immediately without introduction. But everything went well; Maxwell pronounced himself satisfied with the outcome. The three light beams combined on the screen to give a relatively realistic-looking colour image of the ribbon,|| though Maxwell did remark that a better result could be obtained with materials that were more sensitive to colours away from the blue end of the spectrum.
It was a few weeks later in May 1861, still only 29, that Maxwell was elected a Fellow of the Royal Society, cementing his position as a member of the London scientific establishment. In the present day, election to a fellowship of the ‘Royal’ is probably the highest honour available to a British scientist, though in the Society’s early days, most fellows weren’t scientists per se, but rather wealthy individuals who had an interest in science. By Maxwell’s time the shift in the membership to actual scientists was well under way, though he would have qualified on both counts.
Electromagnetism goes mechanical
By the time of his Royal Institution lecture, around five years had passed since Maxwell had last worked on electromagnetism. Whether inspired by appearing at Faraday’s lecture table or simply coming naturally back to a topic that would always fascinate him, while at King’s College he picked up the subject where he had earlier left off. His previous model, treating electricity and magnetism as fluids, was workable only for fields that did not move. But to cope with the likes of generators and electric motors, now becoming relatively common as a result of Faraday’s work, Maxwell had to deal with movement. Just as he had transformed his picture of the rings of Saturn from solid to fluid to particles, he now modified his approach to electromagnetism by resorting to a mechanical model.
This ‘model’, like many that physicists would construct from Maxwell’s time to the present day, was not a model in the sense of a small-scale physical construction, looking like the real thing. Instead he used a theoretical construct that reflected what was observed in nature and that could be used to make predictions to see whether the model was an effective representation of the phenomenon, or whether it needed refining. This was not an actual mechanical device, but one that used the principles of mechanics to try to reproduce the effects of electromagnetism.
Electromagnetism has one fundamental difference from the other force of nature that we experience directly – gravity. All gravity attracts.** But electromagnetism comes in two flavours, known as negative and positive for electricity, and as north and south for magnetism. The rule is that like flavours (say negative and negative for electricity, or south and south for magnets) repel while opposite flavours (negative and positive, or north and south) attract. Bring together electrical charges or magnetic poles and this becomes very obvious. Maxwell set out to model this process, starting with magnetic poles, using a model of the magnetic field that worked in a purely mechanical fashion.
There are a couple of other requirements he needed to cope with in designing a model to work for magnetic poles. One is that they always seem to come in opposi
ng pairs – unlike electrical charges which are happy to be standalone negative or positive, we have never seen a separate north or south pole†† – and the force that is felt between poles, whether repulsive or attractive, obeys an inverse square law as does gravity, dropping off at the same rate as the square of the distance between the two poles that are attracting or repelling each other.
The biggest problem that Maxwell faced was the same one that had caused many to struggle in trying to find a model to explain how gravity could work. It’s relatively easy to have a mechanical model that produces the effect of repulsion, because it’s easy for one object to push another. But it’s harder to have a model that produces attraction, as without involving something like magnetism it’s difficult to see how one object can pull another to which it has no direct physical connection.
Ever since Newton’s time this problem had been got around by devising mechanical models for gravitational attraction which were based on the idea that space was filled with invisible high-speed particles heading in all directions which did not interact with each other, but which pushed on massive bodies. Usually the impact from all directions balanced out, but when a second body was nearby, it blocked some of the particles heading towards the first body, producing the effect of attracting one body towards the other. These models needed a fair amount of tweaking, as the obvious implication is that gravitational pull would depend on the size of a body rather than its mass. There seems no evidence that Maxwell ever considered this type of model for electromagnetic attraction.
