LPP Fusion’s research team successfully completed the bake-out of the FF-1 experimental fusion device during March, reducing the oxygen in the device by about a thousand-fold. The oxygen had been an obstacle in reducing impurities in the plasma that produces fusion reactions, because tungsten oxide and its compounds with hydrogen are easily evaporated into the plasma. The bake-out took place from March 2 to March 23 at a temperature of 120° C, low enough to ensure that no damage was done to the plastic Mylar insulators.
During February, the team took a number of steps needed to prepare for the bake-out. LPPF Research Physicist Dr. Syed Hassan replaced the Rogowski coils that measure the ion beam. The new ones have a glass, rather than copper enclosure, to better protect the plastic insulation on the wires. The insulation itself was upgraded to Kapton, which can withstand the bake-out temperatures. Dr. Hassan also repaired the roughing pump that pumps most of the air out of the vacuum chamber, and a pressure gauge. In addition, he and LPPF Chief Scientist Eric Lerner tested the repairs of the main Mylar insulator to ensure that they could withstand the full 45 kV that they will be exposed to in firing the FF-1 device.
Fig. 1. FF-1 gets a warm blanket for bake-out. The vacuum chamber and drift tube were first wrapped with aluminum foil and heating tape, (left) and then more foil and an insulating blanket (right). Similar blankets with separately controlled heaters were made for the anode (heated from above the device) and the gas lines.
Hassan and Lerner then re-assembled FF-1, using a higher-purity ceramic insulator that has 50% more strength against electric breakdown. They installed a heating blanket with improved multiple thermostats to independently control the heating of the anode, cathode, vacuum chamber, drift tube and gas lines. Thick layers of aluminum foil distributed the heat evenly.
In the first day of bake-out about 40 milligrams (mg) of water vapor was removed, mostly the layer bound to the vacuum chamber—water that in the past released oxygen into the plasma. (This may not sound like a lot, but the total mass of the plasma carrying the current during FF-1 shots is only 3 mg.) In the subsequent nearly three weeks of bake-out a comparable amount of water was baked out, mainly from a large disk-shaped silicone gasket that helps seal the chamber.
We estimate that water remaining, mostly deep within the gasket, is only 100 micrograms. In addition, a single atomic layer of oxygen with a mass of about 40 micrograms is tightly bound to the surface of the tungsten. This will only come off when the plasma lifts it off during a shot. However, as the plasma cools, about three-quarters of the oxygen will be absorbed by the titanium nitride coating on the vacuum chamber, so the remaining oxygen on the tungsten will drop below our goal of 10 micrograms in only a few shots.
Of course, oxygen leaks from the atmosphere can’t be entirely eliminated, but we have reduced that source to less than 2 micrograms per hour. During the course of a day of firing, oxygen should be reduced to very low levels, reducing one main source of impurities to less than 1% of the mass of the current sheath.
