Detached Divertor Control
Divertor detachment spreads exhaust heat over volume by radiation and neutral collisions; control holds the detachment front in a safe location.
The exhaust problem
The divertor receives the plasma exhaust, and in an attached state the heat lands on a narrow strip that can exceed material limits. Detachment cools the plasma before it reaches the target so that heat is radiated and transferred to neutrals over a larger region, dropping the peak surface heat flux to tolerable levels.
The detachment front
Detachment is characterized by a radiation and ionization front that sits somewhere between the target and the confined plasma. If the front stays near the target, exhaust is handled and the core is unaffected. If it moves too far upstream toward the main plasma, it cools the confined region and degrades performance, and can trigger instabilities. The front's position is the quantity to control.
Sensing and actuating
- Sense front position with divertor spectroscopy, Langmuir probes, or thermocouples
- Actuate with impurity seeding gas rate
- Trim with divertor neutral pressure via fueling and pumping
- Fall back on reducing input power if the front runs away
Why it is hard
The relationship between seeding rate and front position is nonlinear and can be bistable: near full detachment a small change in seeding can move the front a long way. Controllers use gain scheduling or nonlinear observers so they do not overshoot into the confined plasma. The loop must react before the front crosses into the core.
In the Kronos program
The Hyperion breeder is designed to run its divertor in a controlled detached state so the plasma-facing surfaces see spread-out heat rather than a concentrated strip. Detachment control shares seeding actuators with radiation and impurity control, so the supervisor coordinates them. These schemes are validated against edge models before hardware operation.