It was not quite that simple, but the Atomic Energy Authority appreciated the heads-up, and thanks for not waiting until the Windscale facility was wiped out to let us know what had happened. The Windscale piles were very large, and Wigner Energy could collect in various regions or pockets in the graphite stack, meaning that the pile would have to be selectively heated and annealed by skillful use of the controls. There was never a written procedure for annealing. It was more of an art than a science. The first anneal on Pile No. 1 was in August 1953, and they were scheduled for once every 30,000 megawatt days of operation.[124] There were eight anneals from the first until July 1957, and during the first operation three scientists hovered over the operators, giving advice and asking questions. After that, it seemed like a routine task, and no science was needed.
The two Windscale piles, with some help from the Canadian NRX, produced enough plutonium for an official British atomic bomb test on October 3, 1952. The blast, occurring under water a few hundred yards from Trimouille Island, Australia, released a respectable 25 kilotons of raw energy, digging a crater 20 feet deep and 980 feet across. William Penney’s bomb design crew and the efforts of thousands working at Windscale had paid off in the most visible way. This would prove to the United States that Britain had achieved parity in the arms race, and an immediate summit meeting was expected.
Unfortunately for this motive, a month later the United States set off “Ivy Mike,” the world’s first thermonuclear bomb on Elugelab Island at Enewetak Atoll, releasing the energy equivalent of 11 million tons of TNT high explosive. Elugelab Island vanished, replaced with a crater 64 feet deep and 6,240 feet across. The British arms industry would have to go a bit farther to come up even with that.
Penney’s team had the imposing task of coming up with a thermonuclear weapon design quickly, starting from scratch. They had not so much as an encouraging word from the Americans, and making hydrogen isotopes fuse explosively was not an easy problem. It was only obvious that a source of tritium, the heaviest isotope of hydrogen and one most likely to fuse, would be necessary.
There is no natural source of tritium, and, like plutonium, it must be manufactured using fission reactors. To accomplish this, a rod of lithium-magnesium alloy, about half an inch in diameter, is encased in a sealed aluminum can and pushed into one of many special “isotope channels” in the reactor. Neutrons from the fission process are captured by the lithium-6 nuclide component of natural lithium, and the atom subsequently breaks down into one helium atom and one tritium atom. The helium pressurizes the can, and the tritium combines chemically with the magnesium component of the alloy, becoming magnesium-tritide. After being in the reactor for about a week, the rods in several cans are harvested and chemically processed to remove the pure tritium. So far, so good.
The tritium production cartridges were given the code name “AM,” and the Mark I cartridges were built using standard fuel cans with a lead weight added so that they would not fly away. The Mark II cartridges were an improvement, having the lithium-magnesium rod diameter increased to 0.63 inches. Under the extreme needs of the hydrogen bomb program, the even more improved Mark III cartridges were built using thinner cans, no lead weight, and alloy rods an inch in diameter. The Mark III’s were admittedly dangerous. The lithium-magnesium rods were pyrophoric, meaning they would burst into hot flame and burn like gasoline if they were exposed to air. The cans were as thin as possible, and the helium pressure could blow them open if the temperature was as high as 440 °C. In addition, there were now so many AM cartridges in the Windscale cores, they dragged the fission process down. Absorbing neutrons without adding any to the fission process made the reactor subcritical. To work at all, the reactor fuel had to be beefed up with a slight uranium-235 enrichment, coming from the new gaseous diffusion plant at Capenhurst.[125]
On January 9, 1957, the British Prime Minister, Anthony Eden, having just presided over the political disaster of the Suez Crisis, resigned from office after being accused of misleading the parliament. The British Army had done a splendid job of tearing up Port Said, Egypt, with French assistance, but internationally it was seen as the wrong force applied at the wrong time as a reaction to Egypt having nationalized the Suez Canal. Eden was succeeded by Harold Macmillan, who desperately wanted to recapture the benevolence and camaraderie of the United States, and the hydrogen bomb initiative was the point of the spear. It would proceed at an ever-accelerated pace.
By May 1957, Penney’s group had put together the first British thermonuclear test device, code named Short Granite, to be exploded in Operation Grapple I off the shore of Malden Island in the middle of the Pacific Ocean. Time was becoming crucial, as a nuclear test-ban treaty was in the works, and the end of unrestricted nuclear weapon experimentation was in sight. The expected yield was north of a megaton. On May 15 Short Granite was dropped from a Vickers Valiant bomber, and it was an embarrassing wipeout, giving an energy release of only 300 kilotons. The first attempt at a British hydrogen bomb had failed.[126] They would get it right the next time, but a lot more tritium was required.
After Grapple I, the Windscale piles were operating in emergency mode at the highest possible power level using flammable metallic uranium fuel and flammable lithium-magnesium in piles of flammable graphite with gale-force air blowing through them. Fire was prevented by coverings of thin aluminum on the fuel and AM cartridges sitting in graphite bricks that could become superheated on their own at an unpredictable time for reasons that were not entirely understood. The annealing interval was increased to once every 40,000 megawatt days for the sake of more plutonium and tritium production. It is not unreasonable to predict an eventual problem under these conditions.
Sunday night, October 6, 1957, was in the middle of a local influenza epidemic, and Windscale Pile No. 1 was well overdue for an anneal. The front lower part of the graphite had not really released any energy at the last anneal in July 1957, and it probably had 80,000 megawatt days of Wigner energy buildup ready to let go at any time. Production was halted, the pile was cooled down, and the ninth anneal would begin Monday morning.
The temperature of the pile was monitored using 66 thermocouples placed at three depths in the reactor face in selected fuel channels. Considering that the core occupied 62,000 cubic feet, and a detailed map of the temperature throughout the graphite was necessary to monitor an anneal, having only 66 thermocouples was pitifully inadequate. There were also 13 uranium-temperature monitors in the control room and another seven on top of the pile in the crane room. The operations staff had performed these measurements before, and there was no concern about this weakness in the instrumentation. On October 7, the main blowers were turned off at 11:45 a.m., and an extremely slow and cautious approach to criticality in the lower part of the core was begun. Stopping twice to look at the temperature readings, it took the operators seven and a half hours to withdraw the controls and reach criticality. There was a holdup to work on some faulty thermocouple connectors. Once the pile was maintaining a low power level, the crew wrangled with some control rod positions to concentrate the fissions in the lower front of the core, where annealing was obviously needed. The goal was to bring this section in the core up to 250 °C.
124
The operating power of a Windscale reactor has never been published, but near as I can tell it was 90 days between anneal no. 8 and anneal no. 9 on Pile No. 1. The plutonium/tritium production was so pressed at this time, it was decided to go 40,000 megawatt days between anneals for no. 9. Assuming that Pile No. 1 ran 24/7, a power rating of over 400 megawatts can be calculated. That is a lot of power for an air-cooled reactor.
125
The piles were also loaded with LM cartridges. These special canisters were used to produce polonium-210 from neutron capture in bismuth oxide for use in A-bomb triggers. The LM cans were not considered flammable.
126
I am told that Short Granite was a full-scale two-stage hydrogen bomb weighing 4,550 kilograms, and the active fusion component had to be the light solid, lithium deuteride. So what was all the furious tritium production for if not a liquid tritium-deuterium bomb? A tritium-based two-stage thermonuclear weapon, such as the American EC-16, would weigh about 32,000 kilograms, which would have nailed the Vickers Valiant bomber to the ground. As it turns out, the tritium production in the Windscale piles was not for the Short Granite bomb. It was used in the electrically driven neutron generator, or “initiator,” used in the A-bomb “Orange Herald,” tested over Malden Island on May 31, 1957. Short Granite used a beryllium-polonium solid-state initiator, made using the polonium-210 produced in the Windscale piles in the LM cartridges. The accelerated production at Windscale was to make bomb-trigger components.