A new paper details results from Century, Zap Energy’s first fully integrated 100 kW sheared-flow-stabilized (SFS) Z-pinch prototype. Unlike Zap’s earlier FuZE plasma experiments, Century is not designed to pursue breakeven directly.
Instead, it tackles the other half of the fusion challenge: whether the engineering subsystems needed for a power plant—pulsed power, plasma-facing components, and thermal management—can operate together, repetitively, and reliably. The answer, at least at this early stage, appears to be yes.
An Integrated Testbed
Century is a vertical machine, with its cylindrical pinch chamber and stacked capacitor banks filling a footprint the size of a double-decker bus. Hydrogen plasmas are fired repetitively, each pulse carrying ~100 kA of current through the pinch column.
Importantly, the device is fueled with protium, not deuterium or tritium. That design choice strips out neutrons, activation, and tritium handling risks, while preserving plasma conditions close enough to fusion shots to meaningfully stress the hardware.
The platform integrates three critical technologies:
Repetitive pulsed power: capacitor banks delivering one plasma discharge every ten seconds (0.1 Hz).
A flowing liquid-metal wall: ~70 kg of molten bismuth circulating around the plasma chamber to absorb heat.
Electrode protection: a novel “wet” electrode tip, cooled with a thin liquid-metal film, designed to resist erosion.
In its first year of operation, Century fired over 8,000 shots, including a continuous run of 1,080 consecutive pulses over three hours. No major failures occurred. For a field where many “proof of concept” results are single-shot, one-off experiments, Century’s multi-hour run is notable for its focus on reliability and operability.
Artist’s conception of the overall Century system based on the engineering models of the components.
Engineering Results
The peer-reviewed data offer several takeaways:
Subsystem Integration Works. Each component had been tested separately; Century demonstrated they can operate in concert. Pulsed power, plasma, liquid wall, and electrodes all maintained integrity over hours of testing.
The Liquid Metal Wall Held Up. Bismuth circulated smoothly, absorbed heat, and transferred it to air-cooled exchangers without overheating. This is one of the largest liquid-metal plasma-facing systems yet tested in fusion.
Electrodes Still Need Work. The wet electrode design reduced wear significantly compared to solid tips, but erosion after thousands of shots remains visible. Scaling to millions of pulses, a baseline for economic power plants, remains unsolved.
Limitations and Open Questions
Century’s success is not without caveats.
No Fusion Output. By design, the device produced zero neutrons. Net energy (Q>1) was not tested; plasma gain remains the domain of FuZE-Q, Zap’s parallel physics experiment.
Low Repetition Rate. The 0.1 Hz cadence is two orders of magnitude below commercial requirements. A plant module must fire at 5-10 Hz, continuously, for months.
Sub-Megamp Currents. At 100 kA, Century is far from the 1-2 MA currents projected for breakeven. Demonstrating durability at those loads remains a future challenge.
Neutron Effects Untested. With no D-T fuel, Century cannot validate neutron-driven issues: material embrittlement, tritium breeding, or activation. These will require a fusion-capable follow-on device.
Despite these limits, Century does what it was designed to: uncover engineering issues before Zap couples them with high-gain plasma physics.
Strategic Roadmap
Zap’s strategy is to run physics and engineering programs in parallel, rather than sequentially. FuZE-Q pushes for scientific breakeven; Century tests power plant subsystems. Together, they sketch a path toward a pilot plant in the late 2020s.
2025–2026: Ramp Century to the full 100 kW average input and increase rep-rate toward 1 Hz. In parallel, pursue Q~1 shots in FuZE-Q.
2027–2028: Combine advances, possibly in an upgraded Century testbed capable of both repetitive operation and fusion fuel.
2028–2030: Deliver a pilot-scale module (~50 MWe) and demonstrate net electricity.
2030+: Replicate modules into a multi-hundred-MW commercial plant.
Zap is already exploring siting options for its pilot, including converting TransAlta’s retiring Centralia coal plant in Washington state—a site with grid ties, cooling water, and a workforce in place.
Closing Thoughts
Century materially improves Zap’s position in a crowded field. It’s one of the first private prototypes to integrate plant-relevant subsystems and to publish peer-reviewed engineering results. That places Zap in a small, credible cohort and strengthens its case with investors and government partners.
But the company’s 2030 grid-connection goal remains aggressive. Achieving it will require merging Century’s engineering with FuZE-Q’s physics in a single, repeatable, net-power module—very much unproven territory.