Homage to Gaia (33 page)

Now here was a straightforward need and there must be a simple answer. The obvious scientist’s response would be to use a
heat-sensing
thermocouple and a meter. I dismissed this thought because it was neither simple nor suitable for a battlefield. Then I considered crystals of organic chemicals with different melting points stuck on a sheet of dark paper. The idea was that when the radiant heat reached a certain level, perhaps enough to cause a first-degree burn, some of the lower-melting-point crystals would melt. I knew the levels of heat radiation that would produce first-, second-or third-degree burns. Therefore, it was easy to test the idea. It worked after a fashion but was messy. It was hard to see if some crystals had melted, for they
crystallized
again as the paper cooled down. Then it occurred to me that if, instead of just dark paper, I used paper coated with gas-detector paint things might be easier. First, I must explain that gas-detector paint was a green paint applied to boards mounted horizontally all around London and its purpose was to reveal instantly use by the Germans of mustard gas or similar chemical warfare agents. Although called gases, these are usually sprayed as liquid, and when the droplets hit the gas-detector paint, they produce bright red circles, which show up clearly on the green background. This happens because the mustard gas liquid dissolves particles of a red pigment suspended in the green paint; the pigment particles are insoluble in the paint but soluble in the mustard liquid. There is nothing singular about mustard gas as a solvent and it occurred to me that any of the crystals I was using might
serve as well when they melted. I tried sticking the crystals on paper painted with gas-detector paint and then exposing the paper to
radiant
heat. Whenever the crystals melted there was a bright red spot. Things were looking much better, but then I noticed that the crystals were not needed. When the heat flux was large enough to cause a
first-degree
burn, the paper itself went pink. At the flux that would cause a second-degree burn, it went bright red. At the flux that would produce a third-degree burn, it went yellow. Therefore, simply by painting a piece of paper with gas-detector paint, I had solved the problem. My boss, Robbie Bourdillon, was delighted and told me that it was just what they needed.

Much later, I heard from an Institute staff member, GL Brown, who had been present at the meeting in Whitehall, just what a
gentleman
Bourdillon was. It would have been so easy for him to go there and take credit for the radiation-detection paper, after all it was one of his staff that invented it, and leave it at that. Instead, he stood up and said, ‘Gentlemen, thanks to the ingenuity of my young colleague, Mr Lovelock, I have a solution to your problem.’ Working for men like this is a joy. There is that feeling of warmth and trust that no amount of payment or any other kind of reward can replace. I spent my formative years as an apprentice in this altruistic environment where my bosses were gentlemen in the old-fashioned usage of the word. Consequently, I spent a lifetime inventing without receiving more than a few per cent of the value of my inventions, indeed my belief in public service, inspired by the idealism of the Second World War, made the idea of profiting from my inventions abhorrent. The MRC paid me well, why should I expect more?

This attitude persisted even when I became independent and I patented personally only one of my inventions. Perhaps if I had been less successful and money had been short I would have tried harder to exploit my inventions. As it was, there always seemed to be enough for our needs and there were so many other things to think about.

7

The ECD

Perhaps the most important event in my life as a scientist was the moment in 1957 when, as a staff member of the National Institute for Medical Research, I stumbled on the electron capture detector (ECD). This simple device that fits easily into the palm of my hand was without doubt the midwife to the infant environmental
movement.
Without it we would not have discovered that chlorinated pesticides like DDT and dieldrin had spread everywhere in the world. The ECD was not limited to finding traces of pesticides; it soon found important trace quantities of other pollutants, notably the polychlorobiphenyls (PCBs), the chlorofluorocarbons (CFCs), and nitrous oxide. It made us aware for the first time of the global extent of pollution. Without the ECD, the appearance of environmentalism and green politics might have been delayed by as much as ten years.

Two things make the ECD special: first, its exquisite sensitivity—at least 1,000 times greater than that of other instruments at the time of its invention; and second, the fact that it is specifically sensitive to pollutants, poisons, and carcinogens. When I discovered the ECD I had no idea how much it would change the world, neither did I realize how much it would change my own life by enabling my independence and so setting me free to become aware of Gaia. It also led me to find how the CFCs were accumulating in the atmosphere—a discovery that culminated in the Montreal Protocol and the ban on the release of CFCs to the atmosphere. The simple and inexpensive nature of the ECD made it appealing to scientists in disciplines ranging from meteorology to geology. When I showed them how to use it to solve their problems it made it possible for me in return to discuss
with them Gaia theory in the context of their own discipline. In this way the ECD was a passport that let me cross the frontiers of scientific disciplines which otherwise are guarded as jealously as national boundaries. In spite of its extraordinary influence on the course of both science and politics, the ECD was not seen as in the forefront of science. This was, I think, because it was seen as a mere invention and not as main-stream science. It was not until the 1990s, more than thirty years after its invention, that its significance was internationally recognized by the award of three environmental prizes: the
Amsterdam
Prize for the Environment in 1990, the Volvo Prize in 1996, and the Blue Planet Prize in 1997.

I will explain in more detail how the ECD works at the end of this chapter, but for now it is enough to know that when a gas like nitrogen is exposed to nuclear radiation, free electrons are torn from a few of its molecules, leaving positively charged nitrogen atoms behind. With pure nitrogen inside the ECD, it is easy to collect all of these electrons by making one of the ECD electrodes positive, and the flow of electrons registers as a small but easily measured electric current. When a trace of DDT vapour enters the ECD in a stream of nitrogen, close to one electron is removed for each molecule of DDT, and the current decreases. By this means as little as 200,000 molecules can be detected and this is equivalent to just over one tenth of a femtogram of DDT. A femtogram is an infinitesimal quantity; it is one thousand million times smaller than a microgram, which itself is a millionth of a gram.

In a way, I first became involved with the ECD in 1948, when I was working on the problem of the common cold. In those days, we knew precious little about the science of this subject, but, as often is the way in states of scientific ignorance, the man in the street knew all that there was to know about the common cold. To him it was quite simple: you caught colds in the winter by getting cold, hence the name. Now, the Medical Research Council was a government
institution
and therefore not immune to public opinion and political
pressure
. We thought it might be wise, therefore, to consider the possibility that we caught colds by getting cold. My job was to determine the extent of chilling objectively, and then compare it with clinical data on the frequency of colds. The three factors
important
in chilling are temperature, humidity, and air movement. The first two are easy to measure, but the air movements in a closed room—draughts, as the English call them—were so slight as to be
undetectable by the simple anemometers then available. As usual, I had no option other than to invent a sensitive anemometer.

In those days, it was customary to build, not buy, instruments. Indeed, scientists were expected to invent, and I soon found myself with two novel anemometers. The first was an ultrasonic device that exploited the change in wavelength of sound due to air motion. It worked well, but was still too insensitive to detect the slight air motion we needed to measure. The second method I tried was an
ion-drift
anemometer. Positive ions move in air at a speed as slow as ten millimetres per second or, if you prefer the older units, about half an inch in a second, in a field of one volt per centimetre. Draughts easily perturb the drift of these slowly moving ions. It was great fun to make such an anemometer and to find that it worked even better than expected.

When I say ‘make’, I mean it literally. I had to make everything from the electronic amplifier to the sensor itself by hand. Remember also that in those days we used vacuum tubes, not solid-state electronics. I even made the radioactive source needed to ionize the air by scraping the dial paint from gauges taken from the flight deck of old wartime aircraft. These gauges provided a rich harvest of radium. I made the sources by ashing this paint, resuspending the ash in lacquer, and then painting this radioactive lacquer onto the anemometer ion source. It worked well and I was able to take it on my Arctic expedition in the winter of 1949. Its only drawback was that cigarette smoke perturbed its response; it was as sensitive to this as is a reformed smoker. To discover the cause of this disturbance by smoke and perhaps find a cure, I exposed the anemometer to a number of different gases and smokes and found that, in addition to smoke, CFCs disturbed its function. At the time, we did not need a device to detect low levels of halocarbons and so the electron capture detector was, in a sense, prematurely discovered.

In 1951, having by then found out little more about the cause of the common cold—other than that it was not chilling—I was moved back to our parent institute in London. My new task was to work on the preservation of life in the frozen state as described in Chapter 4. As I did my freezing experiments, I became aware that the fatty acid composition of cell membrane lipids was important in their sensitivity to freezing damage. Archer Martin and Tony James were in a lab one floor above me and I knew of their newly invented gas chromatograph. I asked them what chance there was of analysing 
the fatty acid composition of my cell lipids. At first, they were
enthusiastic
, but when they saw how small were my samples, just a few hundred micrograms, they advised me to extract larger samples. As an afterthought, Martin added that perhaps I could invent a more
sensitive
detector than his gas density balance. Larger-scale experiments with cell membranes would have required about two months’ work. To go back to inventing seemed much more fun. I remembered the sensitive ionization anemometer I had made in 1949 and how easily the presence of CFCs disturbed it. I wondered how I could turn this disadvantage of the anemometer to advantage as the basis for an ionization detector.

At the National Institute it was the tradition of those days never to read the literature, especially textbooks. Senior scientists asserted that our job was to make the literature, not read it. This recipe worked well for me. Had I read the literature of ionization phenomena in gases before doing my experiments, I would have been hopelessly
discouraged
and confused. Instead, I just experimented. Fortunately, the excesses of the health and safety bureaucracy did not hamper us, like now. The Institute expected its scientists to be personally responsible when they used dangerous chemicals or radioactive materials. There was some risk, but I doubt if under the stifling restrictions of today I would have had the persistence to carry on with so uncertain a project as the infant ECD.

I modelled my first detector on the Dutch scientist H Boer’s design of the ionization cross-section detector. The Boer detector was, in effect, a gas thickness gauge so that the denser the gas, the greater the numbers of ions, and consequently the larger the flow of current. I tried a detector made from a simple cylindrical ion chamber, about two millilitres in volume and which contained a twenty-millicurie strontium-90 beta-particle source.

I remember bending the stiff silver foil of this radioactive source behind a sheet of thick glass to protect me from its hard beta rays. Eventually its curvature was narrow enough to fit the detector cavity. I then joined the case of the detector to a source of negative electrical potential and the central electrode to a home-made electrometer. This used a pair of vacuum tubes in a balanced cathode follower circuit and I made the electrometer on our kitchen table at home. I
purchased
the electronic components from surplus equipment vendors in downtown London. The chromatograph itself consisted of a 1.2 metre-long straight glass column filled with a powder coated with
a non-volatile hydrocarbon mixture called apiezon. I mounted the column vertically within a solid rod of aluminium, 2.5-centimetre in diameter and electrically heated. It ran at a temperature of about 100°C.

This ionization cross-section detector works best with light carrier gases such as helium or hydrogen; it is similar in this respect to the thermal conductivity detector in its performance. Helium was then expensive in Europe and hydrogen was unacceptable for use in a
high-temperature
apparatus expected to run overnight unattended. I was obliged, therefore, to use nitrogen as the carrier gas—as did Martin and James with their gas density balance. It was easy enough to confirm Boer’s performance figures, but these were miserably
insensitive
when compared with those of Martin’s gas density balance. The first ionization detector did not seem to be promising. Sometimes when confronted with a failed experiment or an unsatisfactory device, it is better to cut one’s losses and move on to something else, but in this instance I remembered how well the ion anemometer worked and how its sensitivity was very dependent upon the applied potential. I thought it at least worth trying a few experiments to see if different ranges of applied potential would improve the performance of the ionization cross-section detector.

It was easiest to try low potentials first. I soon found that if I polarized the detector with less than thirty volts, the ion current in pure nitrogen became a little less, but that when other substances were present it became much less. Tony James had supplied me with a test mixture containing methyl propionate, methyl butyrate, methyl valerate and methyl caproate—these are pleasant smelling volatile liquids, chemicals like those that give fruit its flavour. With the detector operating at 100–300V, one milligram of this mixture gave four small peaks. When I tried it with only ten volts and reversed the recorder connection in order to reveal negative peaks positively, the one-milligram sample gave what seemed to be a never-ceasing range of off-scale peaks. I thought that the search was over and that we now had a truly sensitive detector. I asked James and Martin to come and try it, which they did, bringing with them an allegedly pure sample of methyl caproate. I shall never forget the look of amazement on Tony James’s face as peak after peak was drawn from a small sample of this substance. Worse, none of them had the
retention
time of methyl caproate or of any other fatty-acid ester. We now know that what we saw were traces of electron-absorbing impurities
in the sample, but at that time, it seemed to be a useless and wholly anomalous device.

In spite of this disappointment, I continued to play with it whenever there was time, and by trying compounds chosen at random from the lab shelves, I discovered a certain sense in its behaviour. It seemed to respond to reactive compounds like ketones and alcohols but not to hydrocarbons and ethers. When I tried a mixture of compounds made up in the relatively unreactive solvent, carbon tetrachloride, the ion current fell to zero and remained there, resisting all attempts to restore normal operation. I did not realize that the ECD is so sensitive to compounds like carbon tetrachloride that mere traces evaporating from surfaces within the chromatograph were enough to overload it for a week.

For the ordinary gas chromatograph, we clearly needed something more sensitive than the original ionization detector, but less
temperamental
than the electron capture detector. Nevertheless, I continued to experiment with the ECD, and by 1959 I had reduced it to the practical form now used. It was then, and still is, the most sensitive, easily portable and inexpensive analytical device in existence. It is not easy to describe the exquisite sensitivity of the ECD. One way is to imagine that you had a wine bottle full of a rare perfluorocarbon liquid somewhere in Japan and that you poured this liquid onto a blanket and left it to dry in the air by itself. With a little effort, we could detect the vapour that had evaporated into the air from that blanket here in Devon a few weeks later. Within two years, it would be detectable by the ECD anywhere in the world.