DeepScience

Research Roadblocks

The major open problems across our research verticals — nuclear fusion, mental health, and artificial intelligence. Our AI pipeline continuously scans new research for breakthroughs that could move these forward.

Open
Partial
Progressing

Divertor Thermal Management

Open

Managing heat fluxes exceeding 10 MW/m² at the divertor. Liquid metal and advanced tungsten concepts are being explored.

thermal engineeringmaterials scienceplasma physics
divertorheat fluxliquid metaltungstenerosion

ELM Control

Partial

Edge-Localized Modes (ELMs) cause periodic bursts of energy that erode plasma-facing components. Resonant magnetic perturbations and pellet injection are promising but incomplete.

plasma physicscontrol systems
ELMedge-localized modeRMPpellet injectionpedestal

First Wall Materials

Open

Developing materials that can withstand 14 MeV neutron bombardment for the reactor first wall. Current candidates include tungsten and RAFM steels, but degradation under irradiation remains unsolved.

materials sciencenuclear engineering
neutron damagetungstenRAFMirradiationfirst wallblanket

HTS Superconducting Magnets

Progressing

Scaling REBCO high-temperature superconductor magnets to reactor size while maintaining field strength >20T. Critical for compact fusion reactor designs.

superconductivityengineering
REBCOHTSsuperconductormagnethigh field

Long Duration Confinement

Progressing

Sustaining H-mode plasma confinement beyond 300 seconds with acceptable energy loss. Current record is ~1000s in tokamaks but not at reactor-relevant parameters.

plasma physicsheating systems
H-modeconfinement timesteady-statecurrent drivebootstrap current

Plasma Disruption Control

Open

Controlling plasma disruptions and MHD instabilities remains one of the most critical challenges. Uncontrolled disruptions can damage reactor components and halt operations.

plasma physicsMHDcontrol systems
disruptionMHDinstabilityrunaway electronsvertical displacement

Plasma-Wall Compatibility

Open

Managing the complex interactions between the plasma edge and reactor wall materials. Impurity transport, sputtering, and redeposition affect both plasma performance and wall lifetime.

plasma physicsmaterials science
plasma-wall interactionsputteringimpurityerosionredeposition

Predictive Turbulence Modeling

Partial

Developing validated models for plasma turbulence that can predict transport at reactor scale. Current gyrokinetic simulations are computationally expensive and not fully validated.

computational physicsplasma physicsAI/ML
turbulencegyrokinetictransportsimulationmachine learning

Q Engineering > 1

Open

Achieving net energy gain (Q > 1) when accounting for all engineering systems, not just plasma physics. Requires simultaneous solutions to multiple other roadblocks.

systems engineeringplasma physics
Q factorenergy gainnet energypower plantefficiency

Tritium Breeding (TBR > 1)

Partial

Achieving a tritium breeding ratio greater than 1 is essential for fuel self-sufficiency. No breeding blanket design has been validated experimentally at reactor scale.

nuclear engineeringmaterials scienceneutronics
tritiumbreeding blanketTBRlithiumneutron multiplier