Superionic iron hydride turns up in a wired diamond anvil, and then barely moves

The wiring is the interesting part. A kink in thermal expansion tells you a transition happened; it does not tell you which sublattice started moving. Putting a bias across a compressed, laser-heated sample and then finding, after a fast quench, that the hydrogen has piled up along one axis ^[9] is a transport measurement, and transport is what the superionic label actually asserts: light atoms moving through a lattice whose iron stays put ^[3].

The transport numbers then argue against any excitement. Mobility comes out on the order of 1 micrometre squared per joule per second, a diffusion coefficient near 10^3 micrometre squared per second ^[10], which is quick for an atom inside a solid. Under Earth's geomagnetic field, the same hydrogen manages about 0.1 micrometres in 10,000 years ^[11]. The gap is in the driving force, not the carrier. A mobility quoted per joule is a statement about flux per unit work supplied, and the geomagnetic field supplies very little work to a proton in an iron lattice, so a fast carrier still delivers almost no net flux.

The release bounds this loosely and then leaves the bound where it is. It says crossing the inner core's 1,200 km radius would take more than 100 times Earth's age ^[12]. Take the quoted rate at face value and the number is around 1.2 x 10^17 years ^[13], which is true to the letter of "more than 100 times" and hides the scale. Directed drift is also not the only route open to a mobile ion: a random walk at 10^3 micrometre squared per second spreads roughly 25 metres in 10,000 years ^[14], which is fast next to a crystal grain and nothing next to a planetary core.

Two assumptions sit inside the extrapolation to core conditions. The phase boundary was mapped over the pressures the cell could reach and then extended to inner-core pressure, where the transition temperature is reported to fall well below the estimated core temperature ^[8]; the longer that extension, the more the conclusion depends on the fitted shape rather than the data. And Ohta says iron hydride may be hexagonal close-packed or face-centred cubic under core conditions depending on how much hydrogen it holds ^[5], while the measurement was made on the face-centred cubic phase ^[6]. If the core's hydride is the other structure, the boundary has to be found again.

The seismology remains an inference. Shear softening is why geoscientists care about superionic iron alloys at all, and that argument has rested on molecular dynamics ^[4]. What this experiment put on the record is lattice volume and hydrogen mobility ^[6][9]. The slow shear waves of the inner core now have a mechanism with laboratory standing behind its premise, which is not the same as a measured elastic constant.