Uncle Tungsten: Memories of a Chemical Boyhood (2001)

by Oliver Sacks

  There were only a handful of these ‘coloring’ elements – titanium, vanadium, chromium, manganese, iron, cobalt, nickel, and copper, so far as I could see, being the main ones. They were, I could not help noticing, all bunched together in terms of atomic weight – though whether this meant anything, or was just a coincidence, I had no idea at the time. It was characteristic of all of these, I learned, that they had a number of possible valency states, unlike most of the other elements, which had only one. Sodium, for instance, would combine with chlorine in only one way, one atom of sodium to one of chlorine. But there were two combinations of iron and chlorine: an atom of iron could combine with two atoms of chlorine to form ferrous chloride (FeCl2) or with three atoms of chlorine to form ferric chloride (FeCl3). These two chlorides were very different in many ways, including color.

  Because it had four strikingly different valencies or oxidation states, and it was easy to transform these into one another, vanadium was an ideal element to experiment with. The simplest way of reducing vanadium was to start with a test tube full of (pentavalent) ammonium vanadate in solution and add small lumps of zinc amalgam. The amalgam would immediately react, and the solution would turn from yellow to royal blue (the color of tetravalent vanadium). One could remove the amalgam at this point, or let it react further, till the solution turned green, the color of trivalent vanadium. If one waited still longer, the green would disappear and be replaced by a beautiful lilac, the color of divalent vanadium. The reverse experiment was even more beautiful, especially if one layered potassium permanganate, a deep purple layer, over the delicate lilac; this would be oxidized over a period of hours and form separate layers, one above the other, of lilac divalent vanadium on the bottom, then green trivalent vanadium, then blue tetravalent vanadium, then yellow pentavalent vanadium (and on top of this, a rich brown layer of the original permanganate, now brown because it was mixed with manganese dioxide).

  These experiences with color convinced me that there was a very intimate (if unintelligible) relation between the atomic character of many elements and the color of their compounds or minerals. The same color would show itself whatever compound one looked at. It could be, for example, manganous carbonate, or nitrate, or sulfate, or whatever – all had the identical pink of the divalent manganous ion (the permanganates, by contrast, where the manganese ion was heptavalent, were all deep purple). And from this I got a vague feeling – it was certainly not one that I could formulate with any precision at the time – that the color of these metal ions, their chemical color, was related to the specific state of their atoms as they moved from one oxidation state to another. What was it about the transition elements, in particular, that gave them their characteristic colors? Were these substances, their atoms, in some way ‘tuned’?«10»

  A lot of chemistry seemed to be about heat – sometimes a demand for heat, sometimes the production of heat. Often one needed heat to start a reaction, but then it would go by itself, sometimes with a vengeance. If one simply mixed iron filings and sulphur, nothing happened – one could still pull out the iron filings from the mixture with a magnet. But if one started to heat the mixture, it suddenly glowed, became incandescent, and something totally new – iron sulphide – was created. This seemed a basic, almost primordial reaction, and I imagined that it occurred on a vast scale in the earth, where molten iron and sulphur came into contact.

  One of my earliest memories (I was only two at the time) was of seeing the Crystal Palace burn. My brothers took me to see it from Parliament Hill, the highest point on Hampstead Heath, and all around the burning palace the night sky was lit up in a wild and beautiful way. And every November 5, in memory of Guy Fawkes, we would have fireworks in the garden – little sparklers full of iron dust; Bengal lights in red and green; and bangers, which made me whimper with fear and want to crawl, as our dog would, under the nearest shelter. Whether it was these experiences, or whether it was a primordial love of fire, it was flames and burnings, explosions and colors, which had such a special (and sometimes fearful) attraction for me.

  I liked mixing iodine and zinc, or iodine and antimony – no added heat was needed here – and seeing how they heated up spontaneously, sending a cloud of purple iodine vapor above them. The reaction was more violent if one used aluminium rather than zinc or antimony. If I added two or three drops of water to the mixture, it would catch fire and burn with a violet flame, spreading fine brown iodide powder over everything.

  Magnesium, like aluminium, was a metal whose paradoxes intrigued me: strong and stable enough in its massive form to be used in airplane and bridge construction, but almost terrifyingly active once oxidation, combustion, got started. One could put magnesium in cold water, and nothing would happen. If one put it in hot water, it would start to bubble hydrogen; but if one lit a length of magnesium ribbon, it would continue to burn with dazzling brilliance under the water, or even in normally flame-suffocating carbon dioxide. This reminded me of the incendiary bombs used during the war, and how they could not be quenched by carbon dioxide or water, or even by sand.

  Indeed, if one heated magnesium with sand, silicon dioxide – and what could be more inert than sand? – the magnesium would burn brilliantly, pulling the oxygen out of the sand, producing elemental silicon or a mixture of silicon with magnesium silicide. (Nonetheless, sand was used to suffocate ordinary fires that had been started by incendiary bombs, even if it was useless against burning magnesium itself, and one saw sand buckets everywhere in London during the war; every house had its own.) If one then tipped the silicide into dilute hydrochloric acid, it would react to form a spontaneously inflammable gas, hydrogen silicide, or silane – bubbles of this would rise through the solution, forming smoke rings, and ignite with little explosions as they reached the surface.

  For burning, one used a very long-stemmed ‘deflagrating’ spoon, which one could lower gingerly, with its thimbleful of combustible, into a cylinder of air, or oxygen, or chlorine, or whatever. The flames were all better and brighter if one used oxygen. If one melted sulphur and then lowered it into the oxygen, it took fire and burned with a bright blue flame, producing pungent, titillating, but suffocating sulphur dioxide. Steel wool, purloined from the kitchen, was surprisingly inflammable – this, too, burned brilliantly in oxygen, producing showers of sparks like the sparklers on Guy Fawkes night, and a dirty brown dust of iron oxide.

  With chemistry such as this, one was playing with fire, in the literal as well as the metaphorical sense. Huge energies, plutonic forces, were being unleashed, and I had a thrilling but precarious sense of being in control – sometimes just. This was especially so with the intensely exothermic reactions of aluminium and magnesium; they could be used to reduce metallic ores, or even to produce elemental silicon from sand, but a little carelessness, a miscalculation, and one had a bomb on one’s hands.

  Chemical exploration, chemical discovery, was all the more romantic for its dangers. I felt a certain boyish glee in playing with these dangerous substances, and I was struck, in my reading, by the range of accidents that had befallen the pioneers. Few naturalists had been devoured by wild animals or stung to death by noxious plants or insects; few physicists had lost their eyesight gazing at the heavens, or broken a leg on an inclined plane; but many chemists had lost their eyes, limbs, and even their lives, usually through producing inadvertent explosions or toxins. All the early investigators of phosphorus had burned themselves severely. Bunsen, investigating cacodyl cyanide, lost his right eye in an explosion, and very nearly his life. Several later experimenters, like Moissan, trying to make diamond from graphite in intensely heated, high-pressure ‘bombs,’ threatened to blow themselves and their fellow workers to kingdom come. Humphry Davy, one of my particular heroes, had been nearly asphyxiated by nitrous oxide, poisoned himself with nitrogen peroxide, and severely inflamed his lungs with hydrofluoric acid.

  Davy also experimented with the first ‘high’ explosive, nitrogen trichloride, which had cost many people fingers and ey
es. He discovered several new ways of making the combination of nitrogen and chlorine, and caused a violent explosion on one occasion while he was visiting a friend. Davy himself was partially blinded, and did not recover fully for another four months. (We were not told what damage was done to his friend’s house.)

  The Discovery of the Elements devoted an entire section to ‘The Fluorine Martyrs.’ Although elemental chlorine had been isolated from hydrochloric acid in the 1770s, its far more active cousin, fluorine, was not so easily obtained. All the early experimenters, I read, ‘suffered the frightful torture of hydrofluoric acid poisoning,’ and at least two of them died in the process. Fluorine was only isolated in 1886, after almost a century of dangerous trying.

  I was fascinated by reading this history, and immediately, recklessly, wanted to obtain fluorine for myself. Hydrofluoric acid was easy to get: Uncle Tungsten used vast quantities of it to ‘pearl’ his lightbulbs, and I had seen great carboys of it in his factory in Hoxton. But when I told my parents the story of the fluorine martyrs, they forbade me to experiment with it in the house. (I compromised by keeping a small gutta-percha bottle of hydrofluoric acid in my lab, but my own fear of it was such that I never actually opened the bottle.)

  It was really only later, when I thought about it, that I became astonished at the nonchalant way in which Griffin (and my other books) proposed the use of intensely poisonous substances. I had not the least difficulty getting potassium cyanide from the chemist’s, the pharmacy, down the road – it was normally used for collecting insects in a killing bottle – but I could rather easily have killed myself with the stuff. I gathered, over a couple of years, a variety of chemicals that could have poisoned or blown up the entire street, but I was careful – or lucky.«11»

  If my nose was stimulated in the lab by certain smells – the pungent, irritating smell of ammonia or sulphur dioxide, the odious smell of hydrogen sulphide – it was much more pleasantly stimulated by the garden outdoors and the kitchen, with its food smells, and its essences and spices, inside. What gave coffee its aroma? What were the essential substances in cloves, apples, roses? What gave onions and garlic and radishes their pungent smell? What, for that matter, gave rubber its peculiar odor? I especially liked the smell of hot rubber, which seemed to me to have a slightly human smell (both rubber and people, I learned later, contain odoriferous isoprene). Why did butter and milk acquire sour smells if they ‘went off,’ as they tended to do in hot weather? What gave ‘turps,’ oil of turpentine, its lovely, piney smell? Besides all these ‘natural’ smells, there were the smells of the alcohol and acetone that my father used in the surgery, and of the chloroform and ether in my mother’s obstetric bag. There was the gentle, pleasant, medical smell of iodoform, used to disinfect cuts, and the harsh smell of carbolic acid, used to disinfect lavatories (it carried a skull and crossbones on its label).

  Scents could be distilled, it seemed, from all parts of a plant – leaves, petals, roots, bark. I tried to extract some fragrances by steam distillation, gathering rose petals and magnolia blossoms and grass cuttings from the garden and boiling them with water. Their essential oils would be volatilized in the steam and settle on top of the distillate as it cooled (the heavy, brownish essential oil of onions or garlic, though, would sink to the bottom). Alternatively one could use fat – butter fat, chicken fat – to make a fatty extract, a pomatum; or use solvents like acetone or ether. On the whole my extractions were not too successful, but I succeeded in making some reasonable lavender water, and extracting clove oil and cinnamon oil with acetone. The most productive extractions came from my visits to Hampstead Heath, when I gathered large bags of pine needles and made a fine, bracing green oil full of terpenes – the smell reminded me a little of the Friar’s Balsam that I would be set to inhale, in steam, whenever I had a cold.

  I loved the smell of fruits and vegetables and would savor everything, sniff at it, before I ate. We had a pear tree in the garden, and my mother would make a thick pear nectar from its fruit, in which the smell of pears seemed heightened. But the scent of pears, I had read, could be made artificially, too (as was done with ‘pear drops’), without using any pears. One had only to start with one of the alcohols – ethyl, methyl, amyl, whatever – and distill it with acetic acid to form the corresponding ester. I was amazed that something as simple as ethyl acetate could be responsible for the complex, delicious smell of pears, and that tiny chemical changes could transform this to other fruity scents – change the ethyl to isoamyl, and one had the smell of ripening apples; other small modifications would give esters that smelled of bananas or apricots or pineapples or grapes. This was my first experience of the power of chemical synthesis.

  There were, besides the pleasant fruity smells, a number of vile, animally smells that one could easily make from simple ingredients or extract from plants. Auntie Len, with her botanical knowledge, sometimes colluded with me here, and introduced me to a plant called stinking goosefoot, a species of Chenopodium. If this was distilled in an alkaline medium – I used soda – a particularly vile-smelling and volatile material came off, which stank of rotten crabs or fish. The volatile substance, trimethylamine, was surprisingly simple – I had thought the smell of rotting fish would have a more complex basis. In America, Len told me, they had a plant called skunk cabbage, and this contained compounds that smelled like corpses or putrefying flesh; I asked if she could get me some, but, perhaps fortunately, she could not.

  Some of these stinks incited me to pranks. We would get fresh fish every Friday, carp and pike, which my mother would grind to make the gefilte fish for shabbas. One Friday I added a little trimethylamine to the fish, and when my mother smelled this, she grimaced and threw the lot away.

  My interest in smells made me wonder how we recognized and categorized odors, how the nose could instantly delineate esters from aldehydes, or recognize a category such as terpenes, as it were, at a glance. Poor as our sense of smell was compared to a dog’s – our dog, Greta, could detect her favorite foods if a tin was opened at the other end of the house – there nevertheless seemed in humans to be a chemical analyzer at work at least as sophisticated as the eye or the ear. There did not seem to be any simple order, like the scale of musical tones, or the colors of the spectrum; yet the nose was quite remarkable in making categorizations that corresponded, in some way, to the basic structure of the chemical molecules. All the halogens, while different, had halogenlike smells. Chloroform smelled exactly like bromoform and (while not identical) had the same sort of smell as carbon tetrachloride (sold as the dry-cleaning fluid Thawpit). Most esters were fruity; alcohols – the simplest ones, anyway – had similar ‘alcoholic’ smells; and aldehydes and ketones, too, had their own characteristic smells.

  (Errors, surprises, could certainly occur, and Uncle Dave told me how phosgene, carbonyl chloride, the terrible poison gas used in the First World War, instead of signaling its danger by a halogenlike smell, had a deceptive scent like new-mown hay. This sweet, rustic smell, redolent of the hayfields of their boyhood, was the last sensation phosgene-gassed soldiers had just before they died.)

  The bad smells, the stenches, always seemed to come from compounds containing sulphur (the smells of garlic and onion were simple organic sulphides, as closely related chemically as they were botanically), and these reached their climax in the sulphuretted alcohols, the mercaptans. The smell of skunks was due to butyl mercaptan, I read – this was pleasant, refreshing, when very dilute, but appalling, overwhelming, at close quarters. (I was delighted, when I read Antic Hay a few years later, to find that Aldous Huxley had named one of his less delectable characters Mercaptan.)

  Thinking of all the malodorous sulphur compounds and the atrocious smell of selenium and tellurium compounds, I decided that these three elements formed an olfactory as well as a chemical category, and thought of them thereafter as the ‘stinkogens.’

  I had smelled a bit of hydrogen sulphide in Uncle Dave’s lab – it smelled of rotten eggs and farts and (
I was told) volcanoes. A simple way of making it was to pour dilute hydrochloric acid on ferrous sulphide. (The ferrous sulphide, a great chunky mass of it, I made myself by heating iron and sulphur together till they glowed and combined.) The ferrous sulphide bubbled when I poured hydrochloric acid on it, and instantly emitted a huge quantity of stinking, choking hydrogen sulphide. I threw open the doors into the garden and staggered out, feeling very queer and ill, remembering how poisonous the gas was. Meanwhile, the infernal sulphide (I had made a lot of it) was still giving off clouds of toxic gas, and this soon permeated the house. My parents were, by and large, amazingly tolerant of my experiments, but they insisted, at this point, on having a fume cupboard installed and on my using, for such experiments, less generous quantities of reagents.

  When the air had cleared, morally and physically, and the fume cupboard had been installed, I decided to make other gases, simple compounds of hydrogen with other elements besides sulphur. Knowing that selenium and tellurium were closely akin to sulphur, in the same chemical group, I employed the same basic formula: compounding the selenium or tellurium with iron, and then treating the ferrous selenide or ferrous telluride with acid. If the smell of hydrogen sulphide was bad, that of hydrogen selenide was a hundred times worse – an indescribably horrible, disgusting smell that caused me to choke and tear, and made me think of putrefying radishes or cabbage (I had a fierce hatred of cabbage and brussels sprouts at this time, for boiled, overboiled, they had been staples at Braefield).

  Hydrogen selenide, I decided, was perhaps the worst smell in the world. But hydrogen telluride came close, was also a smell from hell. An up-to-date hell, I decided, would have not just rivers of fiery brimstone, but lakes of boiling selenium and tellurium, too.