But Leo Szilard at least was not yet done with protest.
* * *
Enrico Fermi took the initiative at least once during the war. Perhaps influenced by the enthusiasm he found at Los Alamos for weapons-making, he proposed at the time of the April 1943 conference—privately to Robert Oppenheimer, it appears—that radioactive fission products bred in a chain-reacting pile might be used to poison the German food supply.1941
The possibility of using radioactive material bred in a nuclear reactor as a weapon of war had been mentioned by Arthur Compton’s National Academy of Sciences committee in 1941. German development of such a weapon began worrying the scientists at the Met Lab late in 1942, on the assumption that Germany might be a year or more ahead of the United States in pile development.1942 If CP-1 went critical in December 1942, they argued, the Germans might have had time by then to run a pile long enough to create fiercely radioactive isotopes that could be mixed with dust or liquid to make radioactive (but not fissionable) bombs. Germany might then logically attempt preemptively to attack the Met Lab, if not American cities. German development of radioactive warfare, another vision in a dark mirror, seemed to the leaders of the Manhattan Project to require countering by examination into parallel U.S. development; the S-l Committee gave such assignment to a subcommittee consisting of James Bryant Conant as chairman and Arthur Compton and Harold Urey as members. That subcommittee went to work sometime before May 1943, probably before February.1943
Fermi would have known of the Met Lab discussions. His proposal to Oppenheimer at the April conference was different from those essentially defensive concerns, however, and clearly offensive in intent. He may well have been motivated in part by his scientific conservatism: may have asked himself what recourse was open to the United States if a fast-fission bomb proved impossible—it could not be demonstrated by experiment for at least two years—and have found the answer in the formidable neutron flux of CP-1 and its intended successors. Oppenheimer swore Fermi to intimate secrecy within the larger secrecy of the Manhattan Project; when the Italian laureate returned to Chicago he went quietly to work.
In May Oppenheimer traveled to Washington. Among other duties he reported Fermi’s ideas to Groves and learned of the Conant subcommittee. Back at Los Alamos on May 25 he wrote Fermi a warm letter reporting what he had found. He attributed the subcommittee assignment to a request from the Army Chief of Staff, George Marshall, although it seems far likelier that the study originated within the Manhattan Project. “I therefore, with Groves’ knowledge and approval, discussed with [Conant] the application [i.e., poisoning German food supplies] which seemed to us so promising.”1944
Oppenheimer had also discussed Fermi’s idea with Edward Teller. The isotope the men identified that “appears to offer the highest promise” was strontium, probably strontium 90, which the human body takes up in place of calcium and deposits dangerously and irretrievably in bone. Teller thought that separating the strontium from other pile products “is not a very major problem.” Oppenheimer wanted to delay the work until “the latest safe date,” he told Fermi further, so that they would have “a much better chance of keeping your plan quiet.” He did not even want to include Compton in any immediate discussion. Summarizing, he wrote in part:
I should recommend delay if that is possible. (In this connection I think that we should not attempt a plan unless we can poison food sufficient to kill a half a million men, since there is no doubt that the actual number affected will, because of non-uniform distribution, be much smaller than this.)
There is no better evidence anywhere in the record of the increasing bloody-mindedness of the Second World War than that Robert Oppenheimer, a man who professed at various times in his life to be dedicated to Ahimsa (“the Sanscrit word that means doing no harm or hurt,” he explains) could write with enthusiasm of preparations for the mass poisoning of as many as five hundred thousand human beings.1945
Mid-1943 was in any case a season of great apprehension among the atomic scientists, who saw Nazi Germany beginning to lose the war and sensed that country’s desperation. The Manhattan Project expected to produce atomic bombs by early 1945; if Germany had begun fission research in 1939 at similar scale it should have bombs nearly in hand. Hans Bethe and Edward Teller wrote Oppenheimer in a memorandum on August 21:
Recent reports both through the newspapers and through secret service, have given indications that the Germans may be in possession of a powerful new weapon which is expected to be ready between November and January.1946 There seems to be a considerable probability that this new weapon is tubealloy [i.e., uranium]. It is not necessary to describe the probable consequences which would result if this proves to be the case.
It is possible that the Germans will have, by the end of this year, enough material accumulated to make a large number of gadgets which they will release at the same time on England, Russia and this country. In this case there would be little hope for any counter-action. However, it is also possible that they will have a production, let us say, of two gadgets a month. This would place particularly Britain in an extremely serious position but there would be hope for counter-action from our side before the war is lost, provided our own tubealloy program is drastically accelerated in the next few weeks.
The memorandum goes on to criticize the handling of production “entirely by large companies”—the Hungarian threnody Szilard and Wigner also sounded—and to propose a crash program directed by Urey and Fermi to build heavy-water piles. Nothing seems to have come of the Bethe-Teller proposal—Hitler’s secret weapons proved to be the V-l and V-2 rockets then in development at Peenemünde, the first of which crossed the English coast on June 13, 1944—but it captures the mid-war mood.
Less worrisome was radioactive dusting. Conant’s subcommittee considered the possibilities and concluded that they were “rather remote.”1947 Conant emphasized that he thought it “extremely unlikely that a radioactive weapon will be used against the U.S. and unlikely the weapon will be used at all.” Groves eventually proposed to George Marshall that a handful of officers be trained in the use of Geiger counters and sent to England to observe. Preparing for the Normandy invasion, Marshall approved.
* * *
It was easier for Americans guarded by the wide moat of the Atlantic than for the British to dismiss the possibility of radioactive attack. Sir John Anderson, Chancellor of the Exchequer, a scientist and the member of Churchill’s cabinet responsible for the Tube Alloys program, discussed the question with Conant at lunch at the Cosmos Club in Washington in August 1943.1948 He was concerned particularly about German heavy-water production because British scientists believed they had found a way to separate light from heavy water at five times the efficiency of existing processes and feared their German counterparts might have made the same discovery. Heavy water would certainly work to moderate a chain-reacting pile. And such a machine might be used to breed radioactive isotopes for dusting London.
The British therefore kept closer watch on the High Concentration Plant at Vemork in Norway.1949 It had not been damaged beyond repair. To the contrary, intelligence sources reported that summer, it had begun production again in April; German scientists had shipped heavy water from laboratory stocks in Germany to refill the various cells and speed restoration of the cascade.
When Niels Bohr escaped from Stockholm to Scotland on October 6, 1943, he carried with him Werner Heisenberg’s drawing of an experimental heavy-water reactor. Bohr met more than once in London that autumn with Sir John Anderson; Anderson matched up Bohr’s information with the Conant subcommittee’s radioactive-warfare study and the Norwegian underground’s news of Vemork’s renewed production and concluded that the plant once again urgently required attack. The Nazis had significantly increased security at Vemork, which ruled out another commando raid. After British and American representatives discussed the problem in Washington George Marshall authorized precision bombing.
American Eighth Air Force B-17’s
climbed northeast from British bases before dawn on the morning of November 16. To minimize Norwegian casualties the aircraft were scheduled to drop their bombs during the Norsk Hydro lunch period, between 11:30 A.M. and noon. No German fighters came up from the defensive airfields of western Norway to delay them and they elected to circle over the North Sea to kill time before penetrating the Scandinavian peninsula. That alerted German flak, which took a limited toll as the bombers crossed the coast. One hundred forty got through to Vemork and released more than seven hundred 500-pound bombs. None hit the aiming point but four destroyed the power station and two damaged the electrolysis unit that supplied hydrogen to the High Concentration Plant, effectively shutting it down.
Abraham Esau of the Reich Research Council decided then to rebuild in Germany. To expedite construction the council planned to dismantle the Vemork plant and remove it to the Reich. The Norwegian underground reported that decision to London. Anderson was less concerned with the plant itself—Germany had only limited hydroelectricity to divert to its operation—than with the heavy water preserved in its cascade. British intelligence asked the Norwegians to keep watch.
Word came by way of clandestine shortwave radio from the Rjukan area on February 9, 1944, that the heavy water would be transported under guard to Germany within a week or two—not enough warning to prepare and drop in a squad of saboteurs. Knut Haukelid, who had spent the past year living on the land and organizing future military operations, was the only trained commando in the area except for the radio operator. He would have to destroy the heavy water alone with whatever amateur help he could assemble.
Haukelid slipped into Rjukan at night and met secretly with the new chief engineer at Vemork, Alf Larsen. Larsen agreed to help and they discussed possible operations. The heavy water, of enrichments varying from 97.6 down to 1.1 percent, would be transferred to some thirty-nine drums labeled potash-lye.1950 “A one-man attack on Vemork,” writes Haukelid, “I considered out of the question. . . . The only practical possibility, therefore, was to try to carry out an attack on the transport in one way or another.”1951 He and Larsen, joined later by the Vemork transport engineer, considered the various stages of the journey. The drums of water would go by train from Rjukan to the head of Lake Tinnsjö. From there the cars would be run onto a rail ferry to travel the length of the lake, proceeding beyond Tinnsjö again by train to the port where they would be loaded aboard a ship bound for Germany. Blowing up the trains would be difficult and bloody, since they would be crowded with Norwegian passengers; Haukelid finally decided to attempt to sink the ferry, which also carried passengers, into the 1,300-foot lake. The transport engineer agreed to arrange to dispatch the heavy water on a Sunday morning, when the ferry was usually least crowded.
Sabotaging the boat would almost certainly mean the deaths of some of the shipment’s German guards, which would call down heavy reprisals in the Tinnsjö area against the Norwegian population. Haukelid radioed London for permission, emphasizing that his engineer compatriots had questioned if the results were worth the reprisals:
The fact that the Germans were using heavy water for atomic experiments, and that an atomic explosion might possibly be brought about, was a thing we now talked of openly. At Rjukan they doubted very much whether the Germans had come in sight of a solution. They also doubted whether an explosion of the kind could be brought about at all.1952
The British begged to differ:
The answer came from London the same day:1953
“Matter has been considered. It is thought very important that the heavy water shall be destroyed. Hope it can be done without too disastrous results. Send our best wishes for success in the work. Greetings.”
So Knut Haukelid laid his plans. He put on workman’s clothes, packed his Sten gun into a violin case, identified which ferry would make the run on Sunday, February 20, 1944, the appointed day, and rode it with one eye on his watch. The Hydro was flat and bargelike with twin smokestacks jutting up side by side through its boxy superstructure. It reached the deepest part of the lake about thirty minutes after sailing and took twenty minutes then to cross to shallower waters. “We had therefore a margin of twenty minutes in which the explosion must take place.”1954 For even such generous leeway Haukelid needed something better than a time fuse: he needed electric detonators and a clock. He visited a Rjukan hardware-store owner at night for the detonators but was suspiciously turned away. One of his local compatriots had better luck. A handyman retired from Norsk Hydro donated one alarm clock to the cause; Alf Larsen supplied a backup. Haukelid modified them so that their hammers struck not bells but contact plates, closing a battery-powered electrical circuit that could fire the detonators.
Months earlier the British had dropped supplies to the Norwegian commando that included sticks of plastic explosive. Haukelid strung the stubby sticks together to make a circumferential loop to cut a hole in the bottom of the ferry. “As the Tinnsjö is narrow, the ferry must sink in less than five minutes, or else it would be possible to beach her. I . . . spent many hours sitting and calculating how large the hole must be for the ferry to sink quickly enough.”1955 To test his timing mechanism he hooked up a few spare detonators at his cabin on the mountain above Rjukan after a long night’s work, set the alarm for evening and lay down to sleep. The detonators went off on schedule; he bolted bewildered from bed, grabbed the nearest gun and reflexively covered the door. “The timing apparatus seemed to be working properly.”1956
On Saturday Haukelid and a local compatriot, Rolf Sörlie, slipped into Rjukan. It was crowded with German soldiers and SS police. An hour before midnight “Rolf and I went over to the bridge which crossed the river Maan and had a look at our target.” The freight cars “had been run up under some lamps, and were guarded. . . . The train was to go at eight next morning, and the ferry was due to leave . . . at ten.”1957
From the bridge the two men slipped to a back street where they met their driver in a car Haukelid had arranged with its owner to steal in the name of the King and return on Sunday morning. The owner had modified the car to run on methane and they were a long hour starting it. They picked up Larsen, who was prepared to escape Norway to avoid arrest after the work was done. He brought a suitcase of valuables and had come directly from a dinner party where he had heard a visiting concert violinist mention plans to leave on the morning ferry and had tried unsuccessfully to convince the musician to stay in the area one more day to sample its excellent skiing. Another Rjukan man also joined them. They drove to the lake well past the middle of the night:
Armed with Sten guns, pistols and hand-grenades, we crept . . . down toward the ferry. The bitterly cold night set everything creaking and crackling; the ice on the road snapped sharply as we went over it. When we came out on the bridge by the ferry station, there was as much noise as if a whole company was on the march.1958
Rolf and the other Rjukan man were told to cover me while I went on board to reconnoitre. All was quiet there. Was it possible that the Germans had omitted to place a guard at the weakest point in the whole route to the transport?
Hearing voices in the crew’s quarters, forward, I stole to the companion[way] and listened. There must be a party going on down there, and a game of poker. The other two followed me on to the deck of the ferry. We went down to the third-class accommodation and found a hatchway leading to the bilges. But before we had got the hatch open we heard steps, and took cover behind the nearest table or chair. The ferry watchman was standing in the doorway.
Haukelid thought fast. “The situation was awkward, but not dangerous.” He told the watchman they were escaping the Gestapo and needed a place to hide:
The watchman immediately showed us the hatchway in the deck, and told us that they had several times had illicit things with them on their trips.
The Rjukan man now proved invaluable. He talked and talked with the watchman, while Rolf and I flung our sacks down under the deck and began to work.
It was an anxious job, an
d it took time.
Haukelid and Sörlie found themselves standing on the bottom plates of the boat in a foot of cold water. They had to tape the two alarm-clock timers to one of the steel stringers that braced the ferry’s hull, attach four electric detonators to the timers, attach high-speed fuses to the loop of plastic explosive, lay the charge of explosive on the bottom plates and then, most dangerously, hook up batteries to detonators and detonators to fuses.
“The charge was placed in the water and concealed. It consisted of nineteen pounds of high explosive laid in the form of a sausage. We laid it forward, so that the rudder and propeller would rise above the surface when water began to come in [to prevent navigating the boat to shallower water]. . . . When the charge exploded, it would blow about eleven square feet out of the ship’s side.”1959 The sausage was some twelve feet around.
Sörlie went up on deck. Haukelid set his alarms to go off at 10:45 A.M. “Making the last connection was a dangerous job; for an alarm clock is an uncertain instrument, and contact between the hammer and the alarm was avoided by not more than a third of an inch. Thus there was one third of an inch between us and disaster.”1960 Everything worked and he finished at 4 A.M.
The Rjukan man had convinced the watchman by then that the escapees he had sheltered needed to return to Rjukan to collect their possessions. Haukelid considered warning their benefactor but decided that might endanger the mission and only thanked him and shook his hand.
Ten minutes from the ferry station Haukelid and Larsen left the car to ski to Kongsberg, forty miles away around the lake, where they would catch a train for the first leg of their escape to Sweden. Sörlie carried a report for London to the clandestine radio. The driver returned the stolen car and he and the Rjukan man strolled home. At Haukelid’s suggestion the Norsk Hydro transport engineer had arranged a foolproof alibi: over the weekend doctors at the local hospital operated on him for appendicitis, no questions asked.