Historical research guide · 0924 · 2020
Kinetic characteristics of ions in an inertial electrostatic confinement device
D. Bhattacharjee · N. Buzarbaruah · S. R. Mohanty · S. Adhikari
In one minute
An inertial electrostatic confinement machine is the simplest fusion device anyone builds: a vacuum chamber, a see-through wire cage in the middle held at a large negative voltage, and deuterium gas. Ions fall inwards through the gaps in the cage, overshoot, come back, and keep falling through the middle again and again until they hit something. Bhattacharjee and colleagues at the Centre of Plasma Physics in Assam modelled their own tabletop machine particle by particle and set the result against probe measurements made on the same device. The simulation reproduces what the experiment sees: the ions organise themselves into bright spokes through the cage openings, and their own space charge builds a virtual anode inside the real cathode — an electrode made of nothing but trapped ions, with a second, nested one made of electrons appearing at the highest voltages. Core ion density climbs an order of magnitude between one and four kilovolts, and the energy spread of the ions inside the cage tells you which ones are trapped and which are still recirculating.
Connection
Chapter 9 is about plasma that organises itself into structure — and the star mode, the virtual anode and the nested virtual cathode of an inertial electrostatic device are that behaviour on a benchtop, with a knob on the front. Chapter 12 keeps the ledger of fusion routes, and this is the one an ordinary laboratory can actually own: the same architecture Philo Farnsworth patented, run as a neutron source and instrumented well enough to be modelled.
From the source
Ions recirculating through the openings of the gridded cathode pick out dominant channels rather than filling the volume evenly — the star mode — and the same spokes appear in the simulated phase space and in a photograph taken through the bottom of the chamber during a minus one kilovolt discharge.Section IV, opening; Figure 3(a) and 3(b)
Trapped ions build their own electrode: their space charge forms a virtual anode inside the real cathode which reflects later ions before they reach the centre, and the depth of the potential well deepens as the cathode is driven from minus one to minus five kilovolts.Section IV A; Figures 4 and 5(a) to 5(e)
Secondary electrons knocked off the grid are themselves trapped by that virtual anode and can form a virtual cathode nested inside it — a second electrode made of nothing but charge — which the experiment shows clearly at minus five kilovolts while the simulation shows only its beginning.Section IV A, discussion of Figure 5(d) and 5(e)
The ion density in the core rises an order of magnitude with drive, from 6.3 times ten to the fifteenth per cubic metre at minus one kilovolt to 3.1 times ten to the sixteenth at minus four kilovolts, and the simulated profiles track the probe measurements on the same machine.Section IV A; Figure 7(a) to 7(d)
Grid transparency is the design parameter that sets core density: an eight-wire cathode about 92 per cent transparent gives a higher peak core ion density than a sixteen-wire cathode of the same diameter at about 85 per cent, in both the simulation and the experiment.Section IV A, final paragraphs; Figures 8 and 9
Inside the cathode the ion energy distribution shifts from a double-Gaussian to a single-Gaussian shape as the voltage rises, the high-energy peak being the trapped population and the low-energy one the ions merely recirculating, so drive converts recirculation into trapping — and resolving the multiple wells above minus four kilovolts is the named next step, needing finite-element or massively parallel particle-in-cell codes rather than the single-processor one used here.Section IV B, Figure 10; Section V, conclusion
The way in
https://doi.org/10.1103/physreve.102.063205How to cite it
D. Bhattacharjee, N. Buzarbaruah, S. R. Mohanty, S. Adhikari (2020) Kinetic characteristics of ions in an inertial electrostatic confinement device doi:10.1103/physreve.102.063205