In January 1999, physicist Carl Collins at the University of Texas at Dallas published a paper claiming he had triggered the release of stored nuclear energy from Hf-178m2, a metastable isomer of hafnium, using nothing more than a dental X-ray machine. The implications were staggering: if true, a single gram of this material held roughly 1,330 megajoules — about 60 times the energy density of TNT, in a clean, non-fissile package that could be triggered on command. DARPA took notice immediately.

The physics is genuine, even if Collins's experiment wasn't. Hf-178m2 is a real nuclear isomer with a 31-year half-life, sitting in an extraordinarily long-lived excited state at 2.446 MeV above ground. It decays through a spin-16 transition so forbidden it took until 1968 to identify. The dream: bombard it with low-energy photons, induce stimulated emission down the gamma cascade, and trigger an avalanche release of all that stored energy. A gamma-ray laser. A weapon with no critical mass, no fissile signature, no fallout — just a flash of hard radiation from something the size of a sugar cube.

From 1998 to 2004, the Stimulated Isomer Energy Release program funneled around $30 million through DARPA and the Air Force. The Pentagon convened a JASON panel. They imagined isomer-triggered warheads small enough for hand grenades, isomer batteries for satellites, isomer-powered drones with year-long endurance.

Then the physics community pushed back. Replication attempts at Argonne National Laboratory (2001) and the European Synchrotron Radiation Facility using vastly more sensitive instruments found no triggering effect at cross-sections at least five orders of magnitude below Collins's claim. By 2004 the JASON report was scathing. Funding evaporated. The field became radioactive in a different sense — no serious researcher would touch it.

Why it deserves another look in 2026:

• The stored energy is real. Hf-178m2 absolutely exists and absolutely contains that energy. The only question is whether we can coax it out.
• The original experiments were photon-starved. Collins used a broadband dental source delivering perhaps 10^8 photons/sec into a microgram sample. Modern X-ray free-electron lasers — European XFEL, LCLS-II, SACLA — deliver 10^20 coherent photons per pulse at tunable, monochromatic energies. We can now probe the exact resonance levels that 1999 instruments couldn't even resolve.
• Production is no longer the bottleneck. The 1990s relied on scavenging Hf-178m2 from spent reactor targets at picogram scale. Modern spallation neutron sources (SNS at Oak Ridge, ESS in Sweden) combined with laser isotope separation could plausibly reach milligram inventories.
• Triggering mechanisms have multiplied. The 1999 program only considered photon-induced triggering. Recent theoretical work on NEEC (nuclear excitation by electron capture) and NEET (electron transition) — confirmed experimentally in Mo-93m in 2018 — opens entirely new triggering pathways unavailable two decades ago.

The civilian payoff dwarfs the weapons application everyone fixated on. A controllable isomer release would give us gamma-ray lasers — a regime of coherent light we've never accessed — enabling sub-angstrom imaging, nuclear waste transmutation, and medical isotope production at a fraction of current costs. A 5-watt isomer battery the size of a watch could power a deep-space probe for a century without plutonium.

The 1999 experiment was almost certainly wrong. The 1999 idea was almost certainly right.