Conflicting Results on a Long-Lived Nuclear Isomer of Hafnium Have Wider Implications

Since 1999, a collaboration led by Carl Collins (University of Texas at Dallas) has been reporting evidence for the x-ray induced release of energy stored in an unusually long-lived nuclear isomer of hafnium-178. Collins’s claims have aroused a public debate whose acrimony reflects issues that go far beyond the interesting but normally staid physics of nuclear isomers.

The Texas collaboration purports to have demonstrated that a 10-keV x-ray photon can precipitate a prompt 2.45-MeV gamma-ray cascade as the isomer ¹⁷⁸Hf^m2 relaxes to the stable nuclear ground state. Normally, the isomer’s halflife is 31 years. Such triggered release of nuclear energy, if it is indeed possible, raises the prospect of radically new weapons and energy-storage technologies.

But a number of prominent nuclear physicists publicly characterize Collins’s claims as quite implausible a priori and completely unproven. They look with dismay on the funds already committed, and on the much larger expenditures under active consideration, by the Defense Advanced Research Projects Agency (DARPA), the Air Force Office of Scientific Research, and other parts of the US defense establishment.

Per gram, the energy stored in the Hf isomer is intermediate between those of chemical high explosives and fissile materials. In a presentation last year to DARPA’s Hafnium Isomer Production Panel (HIPP), whose assignment is to consider large-scale production schemes for ¹⁷⁸Hf^m2, Collins pointed out that “a golf ball filled with the isomer would have the energy content of 10 tons of explosive.”

Because isomer weapons would not involve transmutation of nuclear species, they don’t come under the rubric of existing nonproliferation treaties. Although William Herrmannsfeldt (Stanford University), a dissident member of HIPP, is convinced that Hf weapons cannot work, he expresses concern about the possible effects that widely publicized Pentagon plans for isomer weapons might have on countries that don’t yet have conventional nuclear weapons. The Defense Technologies Information Center, for example, proclaims that Hf weapons have “the potential to revolutionize all aspects of warfare.” ¹

A few weeks ago, DARPA cut its two-year $30 million Hf-isomer budget for fiscal years 2004 and 2005 roughly in half. Perhaps that is a response to the rising chorus of expert voices raised against the project. “The FY 2004 program is now exclusively based on rigorous and reproducible proof-of-concept experiments,” says a spokesperson for DARPA Director Anthony Tether.

Isomeric states

Nuclear isomers are excited states that eventually decay to the ground state, mostly by gamma radiation. Hundreds are known, but only a few have halflives longer than a day. The only one found in nature is a 10¹⁵-year isomer of tantalum-180. ¹⁷⁸Hfm² was discovered in the 1970s, created inadvertently by neutron irradiation of a nuclear reactor’s Hf cladding. The iso-mer’s conveniently long life and its high excitation energy (2.45 MeV) quickly made it an object of speculation about exotic potential uses. Talk of a possible Hf-isomer gamma-ray laser surfaced in Star Wars missile-defense discussions of the 1980s.

The Hf state is an example of a so-called K isomer. Not only is the spin of the excited nucleus very high (J = 16), but its projection on the nuclear symmetry axis, denoted by the quantum number K, is also 16. Selection rules for low-multipole electromagnetic decay severely inhibit transitions that change K. That’s why the isomer takes so long to venture the first decay step—down to a band of different J states with K = 8 (see figure 1). The bottom state of that band is another isomer, with a halflife of only 4 seconds, that briefly bars the way to the K = 0 ground-state band.

In 1999, in their first reported demonstration of hafnium triggering, ² Collins and company irradiated less than a microgram of the 31-year isomer, embedded in a plastic target, with a beam from a dental x-ray unit. The machine’s broad x-ray spectrum peaked at about 30 keV, with an upper limit of 90 keV.

The group reported that the irradiation produced a few-percent increase in the counting rate for gammas at two energies (426 and 495 keV) that occur naturally in the isomer’s decay cascade. The excess count rates were less than 3 standard deviations above background. The paper speculated that the apparently accelerated decay was due to an unknown K-mixing level some 20–60 keV above the ¹⁷⁸Hf^m2 state. Kicked up into this putative nearby state of mixed K by a low-energy x-ray photon, the isomer would be free to promptly enter and descend the K = 8 band.

Within weeks of the appearance of the Texas collaboration’s paper in Physical Review Letters, the journal received three comments ³ pointing out that the nuclear x-ray absorption cross section implied by the reported triggering was six orders of magnitude higher than what one expects from nuclear theory. The comments did, however, leave open the outside possibility that such an inordinate cross section might be due to some role of the atomic electrons in coupling the x-ray photons strongly to the nucleus.

“Extraordinary claims demand extraordinary proof,” says Caltech nuclear theorist Steven Koonin. Given the controversy engendered by the Collins group’s paper, and the weighty implications if the paper turned out to be right, Peter Zimmerman, then chief scientist of the US Arms Control and Disarmanent Agency, asked Koonin in 1999 to head a JASON review of the evidence for Hf triggering. JASON is a prestigious independent study group funded at the time by DARPA. In 2002, Tether broke off DARPA’s long-time sponsorship of JASON (see Physics Today, July 2002, page 27 ).

‘Inconclusive, at best’

Koonin’s panel included, among others, physicists Sidney Drell (Stanford), Will Happer (Princeton), and Richard Garwin (IBM). After studying the Texas collaboration’s paper and being briefed by two of its authors, the panel concluded that “the experiment is poorly characterized and ill described.” Because the “result is a priori implausible,” and the experiment “inconclusive, at best,” the panel suggested that a definitive experiment be undertaken at a synchrotron light source.

The JASON challenge was taken up by John Becker (Lawrence Livermore National Laboratory) and collaborators at the Argonne and Los Alamos national labs. “I was getting so many inquiries about hafnium from industrialists,” recalls Becker. The new experiment, using Argonne’s Advanced Photon Source (APS), irradiated a Hf-isomer target with a “white” x-ray beam whose intensity exceeded that of Collins’s x-ray unit by about five orders of magnitude over a continuum of photon energies from 10 to 100 keV. “If Collins’s originally reported effect had been real,” says John Schiffer of the collaboration’s Argonne contingent, “we should have seen it amplified a hundred thousand times.” Instead, the collaboration reported in the spring of 2001 that it had found no acceleration of the Hf isomer’s 31-year decay rate. ⁴

But there would have to be at least one more round. Collins responded that additional experiments with his x-ray unit, ⁵ reported before the experiment at Argonne got under way, strongly suggested that the resonant triggering energy was less than 20 keV. Therefore, Collins argued, the APS experiment couldn’t see the effect because the aluminum in its “flawed” target design was shielding the Hf nuclei from x-ray photons below 20 keV. “The principals [of the APS collaboration] do not maintain cognizance of current literature,” Collins told last year’s HIPP meeting.

To close the loophole invoked by Collins, the APS collaboration repeated its experiment in 2002, with modifications designed to make it much more sensitive to possible triggering by low-energy x rays—down to 4 keV. To that end, the original target of mixed aluminum and Hf oxide powder was replaced by thin Hf layers electroplated onto beryllium disks. ⁶ But once again the experiment revealed no evidence of accelerated isomer decay (see figure 2).

Collins dismissed the second null result, arguing that the redesigned Argonne target contained too little of the isomer. He frequently refers to the new result as the APS collaboration’s “second published failure.” Sometimes he attributes these “failures” to incompetence, sometimes to purposeful intent. “This has not been a rational exchange of ideas,” says Schiffer. “Collins is extremely emotional on the subject.”

Meanwhile, Collins had moved his experimental effort from the modest x-ray unit at his Center for Quantum Electronics on the Dallas campus to the formidable SPring-8 synchrotron light facility near Osaka, Japan. Instead of the intense broadband x-ray beam favored by the APS group, Collins and company opted to use a monochromator and thus sacrifice intensity in favor of narrow-band tunability. Early in 2002, they reported a 4-standard-deviation enhancement of isomer-decay gammas just above the 9.56-keV threshold for x-ray ionization of electrons in atomic hafnium’s L shell (see figure 3(a)). ⁷ The paper attributed the coincidence to a mechanism called nuclear excitation by electronic transition (NEET).

Such an assist by atomic electrons might conceivably circumvent the theoretical limits on direct coupling of x-ray photons to the nuclear isomer. The SPring-8 data suggested that about one L-shell photoionization in 600 triggered a Hf-isomer decay. But theorist Evgenii Tkalya at Moscow State University soon pointed out that the proposed NEET mechanism was many orders of magnitude too weak to account for such triggering. ⁸

In 2002, the deputy undersecretary of defense for science and technology asked the Institute for Defense Analyses (IDA) to examine the evidence and prospects for controlled extraction of energy from the nuclear isomers. IDA is an independent research center funded by the Department of Defense (DOD). The IDA study, completed in September 2002, expressed the harsh conclusion that the Texas collaboration’s initial 1999 paper “was flawed and should not have passed peer review.” The paper’s publication, noted the IDA study, “was followed by the breakup of the Dallas group.” The most prominent defector, James J. Carroll of Youngstown State University in Ohio, went on to do several experiments on x-ray hafnium triggering with new collaborators. They found only null results.

‘Still alive’

“You’d think that the IDA and JASON reviews, and the two experiments at Argonne, would have driven a stake through the heart of this thing,” says Zimmermann. “But amazingly, it’s still alive.” To figure out how to make and separate the quantities of the isomer that military applications might require, DARPA created the HIPP panel at the end of 2002. The micrograms of Hf isomer needed for physics experiments can be extracted from tantalum beam stops at existing high-flux proton accelerators. But making macroscopic quantities would require enormously expensive new facilities.

For example, HIPP is considering a two-beam, high-current deuteron accelerator that could bombard a liquid lithium target to make energetic neutrons for the reaction $$\text{n}+{}^{179}\text{Hf}\to {}^{178}{\text{Hf}}^{\text{m}2}+2\text{n}.$$ Then one faces the daunting problem of separating the isomer, in quantity, from ground-state ¹⁷⁸Hf, which is lighter by only 2.45 MeV. That’s a very much smaller mass difference than the 3-GeV difference between ²³⁵U and ²³⁸U.

Collins was invited to address the first HIPP meeting in May 2003. “I literally begged Ehsan Khan [HIPP chairman, on loan to DARPA from the Department of Energy (DOE) Office of Science] to also invite someone from the Argonne team,” recalls Herrmannsfeldt, “so we could weigh the triggering evidence before looking into grandiose production schemes. Khan said ‘no, it would be too disruptive.’ That’s when I began my active opposition.”

At the May meeting, Collins told the panel that “the Argonne work is without merit.” In August, Herrmannsfeldt and four of his 12 HIPP colleagues sent a letter to Khan and Martin Stickley, director of DARPA’s isomer program, that urgently requested “an expedited independent review [of the triggering evidence] before proceeding to study applications that may make no physical sense.” Among the concurring signatories who were not HIPP panelists were Drell, Koonin, Zimmerman, Arthur Kerman (MIT), Morton Weiss (Livermore), and Wolfgang Panofsky (Stanford).

The letter has, thus far, elicited no action toward the desired review. Many in the nuclear-physics community are hoping that DOE will convene an expert panel, much as it did for cold fusion 15 years ago (see Physics Today, April 2004, page 27 ). But there seems to be a territorial hangup. “Hafnium triggering is a DOD matter,” says a spokesman for the DOE Office of Science. “DOE is not involved and doesn’t intend to get involved.”

In January of this year, Stickley sent the dissenters a written version of a talk Collins had given last summer at a laser conference in Hamburg. In his cover letter, Stickley urged “those of you who continue to question his results to please study this paper.” The most provocative graph presented by Collins (reproduced in figure 3(b)) shows new SPring-8 data with a much more pronounced enhancement at the Hf L-shell absorption threshold than was reported in the group’s 2002 paper (figure 3(a)). The new figure is to be published in the proceedings of a more recent conference. ⁹ But the unsatisfied skeptics, including Carroll (who chaired the isomer session at the Hamburg meeting), contend that adequate detail has not been presented. Garwin, for example, wonders how Collins chose the zero-excess baseline and what the data look like at x-ray energies above and below the narrow energy interval shown in the figure.

Even if 10-keV photons can trigger the decay of ¹⁷⁸Hf^m2, Herrmannsfeldt argues, there’s still very little prospect of useful applications. In a written response to Stickley’s January challenge to unbelievers, Herrmannsfeldt considers, in some detail, both explosive and controlled uses of Hf-isomer energy in the light of Collins’s experimental claims.

Explosive uses require a chain reaction, and that, Herrmannsfeldt calculates, is impossible. Too much of the potential triggering energy of the degraded gammas is dissipated in Compton and photoelectric interactions. The JASON review had pointed out that, even if a chain reaction were possible, the mediating photons, unlike slow neutrons, would propagate so fast that the Hf assembly must break up prematurely and the explosion was “guaranteed to fizzle.” Then there’s the disputed question of how the x-ray triggering could be short-circuiting the 4-second isomer, as Collins claims it does.

Controlled uses—for example, the air force’s vision of long-term flight without refueling—require that the released isomer energy provide sufficient excess to power the triggering x-ray beam. Assuming that half the Hf in the working target is isomeric, that L-shell photoionization is maximal, and that one ionization in 600 triggers an isomer decay, it would cost, on average, 16 MeV of x-ray energy to recover 2.45 MeV in gamma rays.

Carroll worries that the baby might be thrown out with the bathwater. Although the evidence for low-energy triggering may have evaporated, he contends, “there’s much good nuclear physics to be learned from the study of isomers and their interaction with radiation.” The triggering of the ¹⁸⁰ Ta isomer by 2.8-MeV photons is well established. It doesn’t violate accepted theory, but it has no obvious practical uses. Collins played a pioneering role in the investigation of tantalum triggering in the 1980s.