Copper does not collapse at its superheating limit, and fusion material screening was built on the assumption

Researchers at SLAC National Accelerator Laboratory and collaborators imaged copper atoms step by step through extreme heating and found the lattice melts steadily rather than collapsing all at once, which is what earlier simulations had predicted ^[1][2]. Those same simulations, run at volume with AI and machine learning, are how candidate materials for fusion heat-handling components get shortlisted before anyone machines one ^[7].

The measurement was done at MeV-UED, SLAC's electron camera, which resolves atomic and molecular motion down to the femtosecond; the team heated a thin copper film with a laser and then imaged it with an electron beam as it heated ^[9]. The starting sample was pure copper ^[20]. The prediction going in was specific: surface melting beginning around 1,085 C (1,985 F), the sides and edges progressively melting as temperature rose, and the central region, under higher pressure, holding its lattice longer ^[10][11]. At about 1,424 C (2,595 F), roughly 1.25 times the melting temperature and copper's superheating limit, the remaining lattice was supposed to fall apart instantaneously into a fully disordered liquid ^[12]. It did not. Melting stayed gradual past the superheating limit ^[13].

Two pieces of arithmetic are worth keeping. The predicted window between the onset of surface melting and the predicted collapse is 339 C ^[18]. And the 1.25 factor is an absolute-temperature ratio, not a Celsius one: 1,085 C is 1,358 K, 1.25 times that is 1,697 K, which is 1,424 C, while the same two numbers in Celsius give 1.31 ^[19]. Anyone porting these thresholds into a design margin should check which scale they inherited.

The reason this went unnoticed is methodological. Qualification often runs on what the SLAC account calls cook and look: heat the sample, then inspect the aftermath, which resembles a metallic brown puddle whether the melting was gradual or sudden ^[15]. The test cannot distinguish the two failure modes, so a model that got the mode wrong could keep passing. "We needed a time-resolved, step-by-step look at the melting process," said Mianzhen Mo, the SLAC staff scientist who led the work ^[16][3].

The stake is not academic. Copper and its alloys are primary candidates for handling the punishing transient heat loads on fusion structural components, with leading models casting copper alloys as heat sinks that pull heat out of materials sitting closer to the plasma ^[4][5]. The core plasma burns at hundreds of millions of degrees while the surrounding structure absorbs sudden loads comparable to spacecraft reentry ^[6]. Mo's group had earlier worked on tungsten after it was flagged for the same duty ^[8].

The correctable part is named. Existing simulations assumed the melting copper sat in static conditions, with uniform pressure on all sides holding atoms in place ^[14]. The published account credits a key parameter for the steady melting without, in the material available here, spelling out what replaces the static assumption ^[2]. Mo's claim is that the results "greatly improve the simulations we use to predict which materials have the best shot at surviving the extreme conditions of future fusion reaction chambers" ^[3].