In the early days of my independent practice, I did not know how much to charge my customers and usually sold my services too cheaply. It did not matter much because the total returns were adequate for my needs and I gradually approached a fair price by adjusting the time spent on a customer’s problem to the amount paid.
The most valued lessons of my apprenticeship at Hampstead and Mill Hill were those in experimental science. The outstanding difference between these MRC laboratories and others I have known was the willingness, even the eagerness, of the scientists to build their own apparatus. From Mill Hill came the huge advances in separation science that enabled molecular biology, and it was all done with equipment the scientists themselves had invented and had made in the Institute workshop. It was a laboratory for medical research, yet the products of individual scientists made for their own personal research are now as important as the research itself. The Wright Spirometer, a simple instrument to measure the peak-flow rate when breathing, finds wide use in diagnosing asthma, and I well remember Wright and his wonderful inventive capacity, and fruitful arguments about the best way to measure something. He was, in the 1950s disdainful of electronics and preferred a mechanical solution if possible, and he was not alone in this. I remember Archer Martin arguing in favour of nanotechnology—ultra-microscopic mechanisms. Martin proposed a Babbage-style computer built of minute mechanical parts and said it would be as fast and as reliable as an electronic computer. Oddly, though, it is from the need to make ultra-microscopic electronics—the computer silicon chips—that has come the means of making minute mechanisms. At these medical labs, I learnt to blow glass, to make simple chemical apparatus, to braze and weld metal, and to use lathes and milling machines. I wonder why this is so rarely, if ever, in the science student’s curriculum. These skills, together with a set of tools, have given me the autarky I needed. I treasure most the small watchmaker’s lathe which was made by a firm called Pultra. I bought it in 1964 and it, with its wide range of accessories, is still in use here at Coombe Mill in 1999.
I find that I think clearly when my hands, as well as my brain, are involved. I frequently wake at five in the morning when it is too early to rise, so instead I stay in bed and try to model in my mind the invention of the day. By breakfast time, the model has taken shape as a three-dimensional image, and after breakfast I go to my lab to translate the mental model into something solid, a construction of metal, quartz, or plastic. This act seems to refine the idea behind it, but the mental experiments done in bed ensure that usually it works first time. Seen from outside, techniques can appear impossibly difficult; how could I ever paint a portrait or play the violin? In those examples, there are no satisfactory states in between starting to learn and becoming proficient as an artist. Experimental science is different: if a customer such as the JPL needed from me a device or an instrument to go on a spacecraft, they needed me merely to make a working model, what they called a ‘breadboard’. The name breadboard goes back to the days when they made electronic equipment, like radio receivers, by screwing down the components onto a kitchen breadboard. Neither they nor I expected to send crude homemade pieces of hardware like this into space or to a planet. Their engineers would use my crude working model as a solid sketch from which to model their own exquisitely perfect constructions, each as light and as strong as an Arctic Tern, a bird that migrates across half of the Earth’s surface. So long as my home-made apparatus works and demonstrates its principle, that is all that is needed. I never minded when at the JPL or at the Hewlett Packard laboratories, engineers would look at my handcrafted device and say, ‘We can do better than that.’ I knew that if my breadboard had been as good as their finished product, they would not have thanked me. Even if I could have done it, it would have been an absurd waste of my time. It also pays to think small: I find that the instrument companies who supply the present-day laboratory equipment tend to make their devices monstrously large. A gas chromatograph or a spectrometer is likely to be far too heavy for one person to lift and will occupy most of a laboratory bench, but I see no reason why these two instruments should be so large and so heavy. The combined gas Chromatograph and mass spectrometer sent to Mars on the Viking lander weighed only seven pounds, and that was twenty-five years ago. Is it necessary for such instruments still to fill a small room? A project close to my heart has been to make a gas chromatograph as small as a pocket calculator. It would enable anyone anywhere in the world to do the science now only possible in the wealthy first world, and I discuss this in more detail on page 184.
The everyday life of an independent scientist is wholly different from that of a typical scientist working in the laboratories of industry, the universities, or government service. To bring you this unusual flavour let me tell you how I invented one small component of the instruments carried to Mars in 1975 by the Viking landers. Sandy Lipsky, then a professor at Yale University, had persuaded JPL that they needed a gas chromatograph/mass spectrometer (GCMS) combination to analyse the soil of Mars. It was a much more powerful instrument than either of these instruments used alone. The combined instrument would not merely separate the substances present in the soil, it would identify them. Now, it was not easy at that time, the 1960s, here on Earth to join a gas chromatograph to a mass spectrometer. A gas chromatograph column delivers the substances it separates greatly diluted in a stream of gas such as nitrogen. The mass spectrometer operates with its interior kept as close to a perfect vacuum as possible, and the sample to be analysed must be introduced without breaking this vacuum. We needed a way to isolate the substances emerging from the column from the large volume of the gas that carried them. Sandy Lipsky brought to JPL a Swedish scientist who had invented a useful device in which the lighter, smaller molecules of helium gas were separated from the larger and heavier molecules of the substances to be analysed. He did it ballistically. If you take a handful of sand and stones and throw the mixture, the stones will travel much further than the sand. This is because, lacking inertia, the sand is quickly slowed by air resistance. In the same way, the heavier molecules projected in the stream of helium went on in a straight path, whereas the much lighter helium molecules diffused out sideways. It worked well as a separator, but the space engineers were unhappy about its power needs, particularly its requirement for a powerful pump to clear away the excess of helium. They did not think it would work on Mars. One of them said to me after trying the ballistic separator, ‘What we need is a separator that removes all of the carrier gas without needing a vacuum pump.’
I thought about this need for a perfect separator and it occurred to me that I could use the strange properties of the metal palladium. Palladium is a precious metal rather like platinum in colour and it does not easily corrode. It is one of the so-called noble metals, favoured by jewellers and others. When it is heated to moderate temperatures round about 200° C, palladium allows hydrogen gas to flow through it as if it were no more substantial than tissue paper. To all other gases and vapours, it remains a solid piece of metal. We could use a long tube of palladium, or some alloy of palladium, and use hydrogen as the carrier gas; the hydrogen would diffuse out through the walls and leave the substances behind. This could be the basis of a most effective separator. JPL provided me with a contract and some palladium alloy tubes and said, more or less, ‘Go home and try these and see if you can make a separator.’
This contract gave me a year of fascinating experiments. In a proper laboratory, a scientist would have set up the palladium tube in an evacuated glass vessel, and then with notebook at hand, he would have measured the flow of hydrogen through the walls of the tube at different temperatures and pressures. After this, he would repeat the measurements with the substances he wanted to separate added to the hydrogen; it would be a project taking several months to complete. In my thatched cottage laboratory at Bowerchalke, I did not have much equipment. My vacuum pump, a somewhat ancient one, was in the loft of the cottage and I could not be bothered to get it down. I was impatient to ha
ve a go and I connected the palladium alloy tubing to a variable transformer so that I could heat it to 200° C by passing an electric current along it, just as if it were the heating element of an electric fire. Then I passed hydrogen at a known rate of flow into it, with no more in mind than to see what happened. From a proper scientist’s point of view this was a terrible experiment, but I was encouraged to do it by earlier conversations with that distinguished scientist, Archer Martin. He and I agreed that it was a mistake ever to do the first experiments too carefully. One should just join up whatever one had to hand, do it, and learn from the experience before designing a proper scientific experiment. Therefore, I merely watched how much hydrogen escaped from the end of my palladium tube when it was cold and when it was hot. To do this I lit the gas escaping from the open end of the palladium tube and watched the height of the small flame of burning hydrogen. I had to use a trick here, for the flame of burning hydrogen is invisible. What I did was to make some smoke in the lab by igniting a small pile of sodium chlorate and sugar; the fine particles of sodium compounds present as smoke in the atmosphere made the flame bright yellow and clearly visible. I turned on the hydrogen with the palladium tube cold—the flow was about one cubic centimetre a second—and lit my tiny flame. Then I turned up the current through the tube until its temperature was about 200° C and, to my astonishment, the flame went completely out. The hydrogen was flowing into the tube but none was coming out, but I found that by turning up the hydrogen to 4 cubic centimetres a second I could now re-establish a tiny flame at the end of the tube. I turned off the heating current and slowly, as the tube cooled, hydrogen began to escape from the end of it and the flame grew in size to the expected height at a flow of 4 cubic centimetres a second. To make sure I was not being fooled, I joined a short length of silicon rubber tubing to the end of the palladium tube where the flame had been and dipped this into a beaker of water so that I could see the bubbles of hydrogen escaping. When I repeated the experiment, the results were even more astonishing. At a flow of 1 cc a second, when the palladium tube was at 200° C, the hydrogen flowed back into the tube and drew water up from the beaker into the silicone tubing as if there were a vacuum inside the palladium tube. Here was hydrogen flowing into both ends of it: where was it going? This was the stuff of magic. I knew that if the tube was merely permeable to hydrogen, a little would always have flowed from the far end, but it was behaving as if the solid palladium metal were a pump. Moments like this are what make the life of a scientist worthwhile. Here was a simple experiment that any schoolboy, or schoolmaster, for that matter, could do and, more importantly, just what was needed to solve the JPL problem. I had fulfilled my contract by the very first experiment.
It took me some time to realize why I could find no reports of this remarkable behaviour. The permeability of palladium to hydrogen was well known and discovered in the 19th century. In proper laboratories, no one would have heated a palladium tube in air. They would have placed it in a chamber and measured the leak of hydrogen gas through the walls of the tube. I tried doing this and found that the extraordinary permeability of my palladium tube vanished when another tube or vessel surrounded it. Even with a vacuum on the outside, it did not pass hydrogen anywhere near as efficiently as it did when it was simply heated in air. It dawned on me that what was happening was that the hydrogen passing through the walls of my tube encountered oxygen atoms at the surface of the palladium in the air. Now, palladium is a catalyst when hot and there must have been an instantaneous reaction between the hydrogen escaping to the surface and the oxygen of the air. This completely removed all hydrogen atoms and constituted the most powerful pump that one could imagine, and this was the explanation of its apparently magical properties. I had just what they wanted for Mars and, a few months later, with many more experiments behind me, I returned to JPL to show them what a splendid separator they had in the form of a palladium tube. When I told them my story, it was so strange that at first few believed me. They were well used to exaggeration from over-excited scientists or salesmen, and how could they tell that what I was saying was different from this? It did not help either to start talking about magic; there is nothing that puts off a scientist more than that. But they were prepared, like the good engineers that they were, to give it a trial by experiment.
The next morning when I came in from the hotel, I found that an engineer had connected one of my palladium tubes to an old mass spectrometer that was in operation and indicated an internal vacuum of 109 torr. This is an exceedingly good vacuum (a torr is a unit of pressure used by those who work with vacuums and equal to about one thousandth of an atmosphere). He had joined the other end of the tube to a pressure gauge, a flow regulator and a hydrogen cylinder. He heated the palladium tube to 250° C and turned on the hydrogen until the gauge said twenty torr. This was the inlet pressure to my palladium tube and it was 20 thousand million times greater than that inside the mass spectrometer, but the mass spectrometer happily kept its vacuum and said that no hydrogen was flowing into it. The engineer turned up the pressure until half an atmosphere, 400 torr, registered on his gauge. Still nothing happened, and the mass spectrometer showed no hydrogen at all coming through into it. The engineer turned to me and said, ‘Your tube is blocked—there’s no way that could happen otherwise.’ So we disassembled the apparatus and tried blowing hydrogen through the cold tube but found no blockage whatever. Puzzled, the engineer joined it all up again, and repeated our test at 400 torr; with just the same result as before. Hydrogen was flowing fast into the inlet of the tube but none was coming out. The engineer was so surprised that he said, ‘Wait a minute’, and went to telephone his friends, asking them to come over and watch the experiment. Soon some scientists, space scientists, and space engineers were drifting in to the small lab to see the experiment. We next turned the pressure up to a full Earth’s atmosphere, 760 torr, and the mass spectrometer gauge stayed obstinately at 10-9 torr; no hydrogen was flowing into it. It was pouring in at the other end of the metal tube and literally vanishing.
It was not until we raised the pressure at the input to a 1000 torr, well above atmospheric pressure, that suddenly the mass spectrometer vacuum failed and it failed catastrophically. Reducing the flow through the palladium tube re-established once more the high vacuum inside the mass spectrometer. There was a small cheer, and from that moment on, doubts vanished. The tube, which we called a transmodulator, was to be the means of joining the gas chromatograph to the mass spectrometer that six years later went to Mars on the Viking spacecraft. Sandy Lipsky’s original idea of a gas chromatograph/mass spectrometer combination as the ideal soil analysis experiment had been reduced to practice. We had some fun breaking the turgid rules of scientific paper writing when we described the separator in a paper in the Journal of Chromatographic Science in 1970. Here is the opening of the paper by JE Lovelock, PG Simmonds, GR Shoemake, and S Rich:
Maxwell’s sorting demon, which could segregate hot from cold molecules, has so far eluded discovery although some of its attributes are to be found in the vortex tube. Lower in the hierarchy of demons is one able to segregate molecules of different species. Such a demon, doorman at an orifice connecting two vessels filled with a gas mixture, could segregate them completely or, alternatively, if only one gas was present stand as a perfect barrier between high-pressure gas and a vacuum. This note reports an experiment performed at the Jet Propulsion Laboratory, the results of which may be used to design an apparatus in which the functions of this second demon are automatically performed. Our motivation in making the experiment was not the promotion of unemployment in the underworld but rather to gather design information for a separator…
The JPL took out a series of patents on the device. They gave me three awards for it and three plaques, which I proudly hang on the walls of my laboratory here at Coombe Mill. There has been no successful commercial exploitation of the transmodulator here on Earth. This was probably because the gas Chromatograph soon evolved to use narrow-gauge colu
mns through which a mere wisp of gas flowed and mass spectrometers evolved with much more powerful pumps. We no longer needed the palladium transmodulator, which had served as a kind of marriage broker to bring together these two powerful instruments.
My experiments with palladium and hydrogen gave me a clue as to what may have led astray those who thought that they had discovered ‘cold fusion’. Palladium will store as much as sixty per cent of its inner atomic space with hydrogen, and this readily happens when palladium is made the negative electrode in a watery solution. A piece of palladium filled with hydrogen is stable for a while in air, but then there is often a surge of heat as the stored hydrogen reacts suddenly and catalytically with the oxygen of the air. It was this, I think, that led to the false conclusion that cold fusion had occurred.
Homage to Gaia Page 39