Sunday, March 8, 2015

The Philosopher's Bomb, Part 3

Those Magnificent Men and their Atomic Machines

The Philosopher's Bomb: The AEC Effort to Create New Elements with Nuclear Explosions

Part III

Back to Part I, Part II

With Special Thanks to Dr. Stephen A. Becker and Dr. David W. Dorn


The AEC fired four more heavy element tests in the spring of 1966: DURYEA (April 14 1966, 70 kT, LRLL), CYCLAMEN (May 5 1966, 12 kT, LASL), KANKAKEE (June 15 1966, 20 to 200 kT, LRLL), and VULCAN (June 25 1966, 25 kT, LRLL).[GURC][Be2] DURYEA was a failure, but I haven't been able to find out anything else about it. The other shots left more concrete records.

CYCLAMEN's target included americium-243 as well as the usual uranium-238. By this point, the scientists had decided that a side reaction was the most likely reason for the even-odd mass abundance reversal: deuterium nuclei were colliding with the uranium-238 target, ejecting a neutron and forming neptunium-239. J. Carson Mark's team included the americium to exploit this effect: deuterium fusing with americium would form curium-244, six nucleons heavier than uranium-238. If, like the neptunium-239, it then captured 19 neutrons, it would reach mass 263, 6 neutrons heavier than any previous test.[Do2][Ec]

Figure 24: Samples from CYCLAMEN Test[Wa]
(US Government)

Unfortunately, the trick didn't work. CYCLAMEN produced a flux of 18 moles/cm2, 50% higher than their previous best, but the debris samples didn't yield any new isotopes. The higher concentration of curium-244 and -245 in the debris indicated that some deuterium atoms had fused with the americium, but the heaviest isotope recovered was fermium-257, the same as in BARBEL.

KANKAKEE was designed to minimize this neptunium-producing side-reaction, to test whether this was the real explanation for the odd-even reversal. The device matched TWEED's neutron flux but had a lower yield of heavy elements above mass 253; not quite conclusive, but strong evidence that the side reaction was the real cause of the even-odd reversal.[Do2][Ec]

VULCAN was a repeat of the TWEED test, using a uranium-238 target instead of plutonium and neptunium. The results were essentially identical to the PAR test, confirming that TWEED had reached 12 moles/cm2, and its disappointing results were not due to a lower-than-calculated neutron flux. TWEED's plutonium target had probably fissioned rather than absorbing neutrons. VULCAN also included scandium in the sample; since its neutron capture cross-sections were well-known, it would provide an easy way to calculate the precise neutron flux.[Bel][Ec][EH]

Mathematical models suggested heavier isotopes should have been formed, at least by CYCLAMEN. At the very least, either fermium-259 or mendelevium-259 should have been recovered. Why weren't they? The scientists considered a number of explanations:

  • Mendelevium-259 might have been accidentally separated out in the chemical processing.

  • The chemists were looking for new isotopes by searching for alpha particle radiation signatures that did not match known isotopes. Perhaps the 259 product was so long-lived they hadn't detected its alpha decay.

  • Alternatively, perhaps it decayed so fast that it was gone before the samples reached the laboratory.

  • Neutron-induced fission might overwhelm neutron capture. When a neutron strikes a uranium or neptunium nucleus, there is always a chance the nucleus may split instead of absorbing the neutron, but below mass 257 the ratio of fissions-to-captures was small enough that plenty of nuclei survived. Perhaps the ratio climbed steeply above mass 257.

  • Finally, as the neptunium-259 atoms beta-decayed, they might pass through a region with a very short spontaneous fission half-life, resulting in all of the mass 259 atoms being lost before they reached fermium.[HH][Hof]

Unfortunately, while none of these could be ruled out, the final two possibilities seemed most likely. The first possibility might account for the absence of mendelevium-259, but they should still have found fermium-259. The second and third possibilities would mean the theoretical predictions for their half-lives were wrong by about a factor of 100, one way or the other; half-life prediction is an extremely inexact science, but usually not that inexact. That left neutron-induced fission or a spontaneous fission “catastrophe” somewhere in the decay chain – a bad sign. If true, it would mean the whole enterprise was impossible.[Hof][Co4]

But Dorn's team would give it one more try. Code-named HUTCH, it would take three years for new designs and fabrication methods to be developed and implemented, for one last great assault on the stronghold of the atomic nucleus.[Do]


Los Alamos's next neutron physics experiment, PERSIMMON, was fired 300 meters underground on February 23rd of 1967.[Si][DoE] PERSIMMON featured experiments to measure the fission cross-sections of plutonium-238 and curium-244 and capture cross-sections of plutonium-238, promethium-147, europium-151, lutetium-175, and niobium-43.[HDB] These last four were of special interest as fission products formed in nuclear reactors.

The Promethium-147 sample was transported in a lead-lined canister and only loaded into the experimental rig with hydraulics fifteen seconds before detonation.[Fa] This complicated maneuver was a rehearsal for a planned later experiment on promethium-148; Pm-148 is so radioactive it could damage the instruments if left in place for too long.[BDS] In fact, with the technology of the day, the Pm-148 cross-section could not be accurately measured by any other way – its natural radioactivity was so high, it introduced substantial error into measurements made with weaker neutron sources. Besides Los Alamos, British scientists from the Aldermaston Atomic Weapons Research Establishment also mounted experiments on PERSIMMON, including making their own measurements of the Pu-238 fission cross-section.[Si]

Unfortunately, all these experiments were spoiled by an error in the design of the neutron moderator, which reduced the energy resolution to inadequate levels.[He]

The next experiment, POMMARD, was fired March 14th, 1968, yielding 1.5 kT[DoE]. While the problems with PERSIMMON's moderator had been fixed, this time noise in the signal cables lost all of the data produced between 50-250 microseconds after the shot, covering the intermediate energy range, while other problems introduced significant noise into the rest of the data.[He][BAPS]

But, after two strikes, the next shot hit a home run. This experiment was called variously PLIERS or PHYSICS 8. PHYSICS 8 had several major differences from earlier shots. The 100-foot-tall experimental tower, which took several months to assemble, was actually mounted on railroad tracks so it could be slid out of the way before the subsidence crater formed.[He]

Figure 25: PLIERS Experiment Tower Prior to Test[He]
(US Government)

Also, the neutron pipe was not aimed at the bomb itself; instead, it was aimed at a block of moderating material to its side. Neutrons would be reflected off the moderator up the pipe, but the gamma rays, which added noise to the experimental signals, would not be.[BCo] Los Alamos detonated PLIERS on August 27th, 1969.[DoE]

The PLIERS shot generated so much data it took months to just finish reading the film records.[He] The shot included 32 separate experiments:

  • Fission cross-sections of americium-243, berkelium-249, californium-249 and -252, curium-243, -244, -245, -246, -247, and -248, einsteinium-253, neptunium-237, plutonium-239, -242, and -244, and uranium-232, -234, -236, and -238;

  • Capture cross-sections of gold-197, thorium-232, uranium-238, curium-244 and -246, and plutonium-239;

  • Scattering cross-sections of ytterbium-89, tantalum-181, and plutonium-239;

  • Transmission cross-section of plutonium-239;

  • Fission symmetry of uranium-235;

  • Isomer production of cesium-134;

  • And a feasibility experiment in neutron polarization.[CD]

Figure 26: PLIERS Experiment Tower After Test[LANL2]
(US Government)

PLIERS was the last identified neutron physics shot of the 1960s, but there was at least one, and possibly two, additional tests which I have been unable to identify. At a conference in 1968, J. A. Farrell of Los Alamos briefly discussed two different approaches to neutron physics tests, unlike any used in the known physics shots:

While the technique of measuring cross sections by means of a vacuum flight path to the surface has been well developed, there are other possibilities for using the nuclear explosion that are also being explored. One is the use of a short flight path with the entire apparatus underground. This would permit even higher fluxes for measurements on very small samples. A first attempt at this technique was unsuccessful but there is no reason to believe it will not work. Another possible experiment involves a very long flight path in a horizontal tunnel with an unmoderated source for high resolution in the kilovolt region. A test experiment of this method is in progress.”[Fa]

These experiments are more unusual than they seem. The known neutron physics tests could be added to existing weapons tests with relatively little additional expense, since the line-of-sight pipes leading from the bomb to the experimental tower could be installed in the vertical drillhole containing the bomb. Farrell's proposals would require a more complicated tunnel geometry, with horizontal shafts leading to the device. Such tests were generally mounted by the Defense Nuclear Agency, but not by Los Alamos. Unfortunately, I could not locate any other records of these experiments, so for now they will have to remain mysterious.

Cowan's team planned a further test for 1970.[Di]

Plowshare in the Late '60s

By 1969, Plowshare was clearly in trouble. And since the heavy elements program was part of Plowshare, politically and budgetarily, that meant it was in trouble too.

The AEC had fired three nuclear cratering tests by 1965: SEDAN, SULKY, and PALANQUIN. The next test was codenamed CABRIOLET, a 2.7-kiloton device buried 170 feet under the Nevada Test Site. The AEC wanted to fire CABRIOLET around March 1st, 1966. The AEC submitted the proposed shot to the 269 Committee, the interagency committee which approved all nuclear tests that could potentially violate the Partial Test BanTreaty (PTBT), in November of 1965.

The Committee did not even meet to consider CABRIOLET until after it was supposed to have been fired, and when it did meet, it did not reach a conclusion. The Arms Control and Disarmament Agency (ACDA) and the State Department were deeply opposed to the test, seeing it as a threat to the PTBT and to other arms control initiatives. The matter bounced around Washington for months. Finally, in September, the argument reached all the way to the White House, and President Lyndon Johnson personally authorized the shot for November.

Then, in October, he decided to reevaluate the shot a second time. In December, he authorized CABRIOLET again, this time for February of 1967. Then – one day before the shot was fired, with the device already emplaced underground – he changed his mind again. This cycling continued for the rest of 1967. The AEC finally fired CABRIOLET on January 26th of 1968.

Figure 27: CABRIOLET Blast
(US Government)

The blast formed a 400-foot-wide, 125-foot-deep crater. Glenn Seaborg triumphantly noted in his diary that “no radioactivity attributable to CABRIOLET was detected by the Canadians” - the PTBT had not been violated. It was a sour victory. The repeated delays had tripled the planned cost of the experiment.

Figure 28: CABRIOLET Subsidence Crater
(US Government)

The AEC managed two more cratering shots on the heels of CABRIOLET, BUGGY and SCHOONER, both in 1968. But those were the last. The objections to nuclear excavation raised by ACDA and the State Department were growing into a chorus, both within and outside the government. The estimated cost of nuclear excavation had risen sharply, as previously neglected costs were factored in. And the hoped-for amendment to the PTBT to permit peaceful nuclear explosives had proved elusive – the Soviets were indeed embarking on their own version of Plowshare, but apparently felt no need to seek a revision to the treaty.[SL]


HUTCH would be LLRL's biggest, boldest heavy element experiment. Part of the plan was to increase the neutron flux by using a bigger bomb: the more D-T fusions, the more neutrons. The neutron flux, they hoped, should be strong enough to also confirm or disprove the hypothesis that the odd-even reversal was due to protactinium from a side reaction; the strong flux should mean the main reaction would dominate the yield until past mass 260.

The target was a mixture of 17.8 grams of uranium-238 and 8.8 grams of thorium-232. It was also spiked with argon, phosphorous, and iron as a side experiment; theoretical analysis suggested that the previously-unknown isotopes phosphorous-35, silicon-34, argon-46, iron-62, and thorium-236 might be produced by neutron absorption. Like VULCAN, the HUTCH target also included scandium to measure the neutron flux.[EH][Ec] If it succeeded in finally getting past mass 257, they planned to fire a follow-on test in fiscal year 1970.[JCAE70]

Dorn's team detonated HUTCH on July 16th, 1969.[DoE]

Figure 29: HUTCH Subsidence Crater[Ec]
(US Government)

Post-shot analysis of the debris showed the device had reached 35 moles/cm2, almost twice the previous best. The first sample of rock from the test contained seven times as much fermium-257 as had ever been produced before. And, unlike every previous test, there was no odd-even reversal, confirming the side reaction theory.[EH][Ec]

Figure 30: HUTCH and CYCLAMEN Element Abundances[Ec]
(US Government)

But they still didn't find any new elements.

The End of the Project

Some of the Lawrence Livermore scientists still wanted to press on. Theory said that new elements must have been produced in HUTCH; that they were not found meant they were decaying before they could be recovered. Prompt sampling systems could get samples to the lab fast enough to spot the new elements before they decayed.

Others disagreed, and, in the end, they won. Dr. Dorn, the head of the effort at Livermore, believed that further shots were futile: if HUTCH had not reached new elements, there was no reason to expect any other shot would, and he was not alone.[Do][Ec] And any shot with a prompt sampling system would not only be difficult and expensive, it would risk a leak of radioactive fallout through the sampling tube. In 1960, when the PTBT was still new, the AEC would have been willing to run the risk, but not in 1970.[Be]

The AEC's 1971 budget proposal asked for money to fire a follow-up to HUTCH. They didn't get it. LASL's next neutron physics experiment was cancelled too. All Plowshare funding for scientific applications was eliminated in that year's budget – along with all funding for excavation research. Only natural gas extraction was funded. Plowshare's supporters still hoped this would be only a temporary defeat. John Kelly, head of the AEC's Division of Peaceful Nuclear Explosives, called it a “hiatus”.[JCAE71]

It wasn't. The AEC asked for money to reopen Plowshare scientific work in 1972, but didn't get it then either. They asked again in 1973 – including requesting money for an “open” neutron physics experiment – and got the same answer. That was the last time they asked.[JCAE73][PWA74] Plowshare itself didn't survive much longer. By 1975, it was officially dead.

Attempts at Resurrection
However, the idea resurfaced periodically through the 1970's and 80's.

In 1972, a group of Livermore scientists, including the famed Edward Teller, suggested a “sequential exposure” process in which one HUTCH-style device would irradiate a U-238 target, and the plasma then “squirted” into the blast of a second bomb.[Me] A year later, another group suggested replicating HUTCH in a salt formation to produce large quantities – a tenth of a milligram – of Cm-250, then bombarding it with U-238 ions in a particle accelerator.[Ho] Still another group proposed a complicated scheme in 1974 involving a target of mixed U-238 and Pu-242, prompt sample recovery, and subsequent re-exposure in a pair of “laser-energized fusion micro-explosives”, the aim being to chart a course of neutron captures and beta-decays through the periodic table, avoiding regions with high spontaneous fission rates.[MNW] Still another proposal suggested leaving an open horizontal tunnel between a HUTCH-style device and a chamber full of salt, which would capture heavy elements shot produced in the explosion for faster, easier recovery.[Hec] Besides such sometimes Rube Goldbergian ideas, there were a number of proposals to simply replicate HUTCH to produce Cm-250 and Fm-257 for more conventional physics experiments.

The final attempt to resurrect the program came at Los Alamos in the early 90s, and it came closer then any of the others. Dr. Stephen Becker, a Los Alamos physicist, proposed fielding a new series of shots, beginning with a replication of HUTCH using a Th-232 target.[Be3] He managed to convince the chief designer of a suitable upcoming nuclear test to incorporate his heavy element experiment. But the shot was delayed by unrelated fabrication issues – and, by the time the problem had been resolved, Congress had passed a nine-month moratorium on all nuclear testing. The moratorium has not ended to this day.[Be][Ea]

The End of the Story

But that's not quite the end.

Extrapolation from the initial samples of HUTCH debris indicated it had made more then a quarter of a milligram of fermium-257 – ten billion times more Fm-257 then had ever before been made by humans – along with 40 milligrams of curium-250.[EH] From August 22nd to September 14th 1969, half a ton of additional debris was recovered by reaming the sample boreholes, and processed to recover the precious isotopes.

In 1971, a group of radiochemists at LLRL bombarded samples of fermium-257 from the HUTCH test with deuterium ions in a particle accelerator, producing the first samples ever detected of fermium-258.[HWLEQ] That was only the beginning. Over the next decade, at least nine scientific papers were published on experiments using Fm-257, Pu-244, and Pu-246 recovered from HUTCH, and three new isotopes were discovered:

1971: Fm-258 produced for the first time by deuterium bombardment of Fm-257[HWLEQ]
Measured the number of neutrons produced by spontaneous fission in Fm-257[CBHT]
Measured the mass symmetry of the Fm-257 spontaneous and neutron-induced fission fragments[JHLW]
Measured the gamma-ray decay of Am-246 and Cm-246[MTMM]
1972: Measured the thermal neutron destruction cross-section of Fm-257[WHL]
1973: Measured the kinetic energy of spontaneous fission fragments of Cm-250 and Cf-250[HFB]
1976: Measured the gamma-ray and electron-conversion decay of Am-246 and Cm-246[MTM]
1978: Cm-251 produced for the first time by neutron irradiation of Cm-250, and the Cm-250 neutron capture cross-section measured.[LWHHL]
1981: Cf-255 produced for the first time by neutron irradiation of Cf-254, and the Cf-254 neutron capture cross-section measured.[LHWQED]

Obviously, there is no chance of these experiments being revived as long as the nuclear testing moratorium holds, and hopefully it will hold for many, many years to come. But fate and international politics have a way of coming back around, and I would honestly not be too shocked if, some day, the US did resume nuclear weapons testing.

I am not a physicist, and I do not have the expertise to say if reviving this project would be worthwhile. The actual physicists I have spoken to disagree: Dr. Dorn believes they are unlikely to get any further then they did in the '60s, while Dr. Becker believes it is worth trying. But I do know this: while Dr. Dorn and the other AEC scientists may not have achieved their primary aim, they succeeded in pushing back the boundaries of human knowledge, and that is a valuable prize.

The Plowshare scientific applications program added to our knowledge of how heavy elements are formed in supernovae.[Ho3] It generated data on reaction cross-sections and other aspects of particle physics. It provided a supply of impossibly-rare heavy isotopes that could not have been obtained in any other way. A program that is still producing papers twelve years after its last experiment is not a failure.

None of the other Plowshare projects left a legacy like this. Nuclear excavation is a fascinating idea, but the fact remains that the Plowshare excavation program consumed a great deal of money and man-power and, in the end, produced very little. The reasons for its failure are complicated, and not entirely the fault of the technology itself, but still: no canals were dug, no mountain passes blasted open, no harbors excavated. But while nuclear excavation got all the attention and the press, the scientific shots actually produced something of value.

There's a moral in there, somewhere.

 The End

Citations can be found here.

1 comment:

  1. ...very interesting. I had no idea of these elaborate experiments to produce new elements.