Ball-pen probe

Ball-pen probe used on tokamak CASTOR in 2004. A stainless steel collector moves inside a ceramic (boron nitride) shielding tube.

Schematic picture of a single ball-pen probe. Ions (in red) have a large gyromagnetic radius and can reach the collector more easily than electrons (in blue).

A ball-pen probe is a modified Langmuir probe used to measure the plasma potential in magnetized plasmas. The ball-pen probe balances the electron and ion saturation currents, so that its floating potential is equal to the plasma potential. Because electrons have a much smaller gyroradius than ions, a moving ceramic shield can be used to screen off an adjustable part of the electron current from the probe collector.

Ball-pen probes are used in plasma physics, notably in tokamaks such as CASTOR, (Czech Academy of Sciences Torus) ASDEX Upgrade, COMPASS, ISTTOK, MAST, TJ-K, RFX, H-1 Heliac, IR-T1, GOLEM as well as low temperature devices as DC cylindrical magnetron in Prague and linear magnetized plasma devices in Nancy and Ljubljana.

Principle

If a Langmuir probe (electrode) is inserted into a plasma, its potential is not equal to the plasma potential $\Phi$ because a Debye sheath forms, but instead to a floating potential $V_{fl}$. The difference with the plasma potential is given by the electron temperature $T_{e}$:

$\Phi -V_{fl}=\alpha *T_{e}$

where the coefficient $\alpha$ is given by the ratio of the electron and ion saturation current density ($j_{e}^{sat}$ and $j_{i}^{sat}$) and collecting areas for electrons and ions ($A_{e}$ and $A_{i}$):

$\alpha =ln\left({\frac {A_{e}j_{e}^{sat}}{A_{i}j_{i}^{sat}}}\right)=ln(R)$

The ball-pen probe modifies the collecting areas for electrons and ions in such a way that the ratio $R$ is equal to one. Consequently, $\alpha =0$ and the floating potential of the ball-pen probe becomes equal to the plasma potential regardless of the electron temperature:

$V_{fl}=\Phi$

Design and calibration

Potential and ln(R) of the ball-pen probe for different positions of the collector.

A ball-pen probe consists of a conically shaped collector (non-magnetic stainless steel, tungsten, copper, molybdenum), which is shielded by an insulating tube (boron nitride, Alumina). The collector is fully shielded and the whole probe head is placed perpendicular to magnetic field lines.

When the collector slides within the shield, the ratio $R$ varies, and can be set to 1. The adequate retraction length strongly depends on the magnetic field's value. The collector retraction should be roughly below the ion's Larmor radius. Calibrating the proper position of the collector can be done in two different ways:

1. The ball-pen probe collector is biased by a low-frequency voltage that provides the I-V characteristics and obtain the saturation current of electrons and ions. The collector is then retracted until the I-V characteristics becomes symmetric. In this case, the ratio $R$ is close to unity, though not exactly. If the probe is retracted deeper, the I-V characteristics remain symmetric.
2. The ball-pen probe collector potential is left floating, and the collector is retracted until its potential saturates. The resulting potential is above the Langmuir probe potential.

Electron temperature measurements

Using two measurements of the plasma potential with probes whose coefficient $\alpha$ differ, it is possible to retrieve the electron temperature passively (without any input voltage or current). Using a Langmuir probe (with a non-negligible) and a ball-point probe (whose associated $R$ is close to zero) the electron temperature is given by:

$T_{e}={\frac {\Phi -V_{fl}}{\alpha }}$

where $\Phi$ is measured by the ball-pen probe, $V_{fl}$ by the standard Langmuir probe, and $\alpha$ is given by the Langmuir probe geometry, plasma gas composition, the magnetic field, and other minor factors (secondary electron emission, sheath expansion, etc). It can be calculated theoretically, its value being about 3 for a non-magnetized hydrogen plasma.

In practice, the ratio $R$ for the ball-pen probe is not exactly equal to one, so that the coefficient $\alpha$ must be corrected by an empirical value for $R$:

$T_{e}={\frac {\Phi _{BPP}-V_{fl}}{\bar {\alpha }}},$

where ${\bar {\alpha }}=\alpha -ln(R).$

External links

• PhD thesis, Jiri Adamek (Czech and English language)
• Overview: Ball-pen probe design, theory and first results at different fusion devices.
• Examination of plasma current spikes and general analysis of H-mode shots in the tokamak COMPASS
• Electrical Probe Measurements on the COMPASS Tokamak
• Probe Measurements on the COMPASS Tokamak
• Scanning ion sensitive probe for plasma profile measurements in the boundary of the Alcator C-Mod tokamak
• Development of Probes for Assessment of Ion Heat Transport and Sheath Heat Flux in the Boundary of the Alcator C-Mod Tokamak
• Video: Temporal evolution of Type-I ELM in the divertor region on the COMPASS tokamak.
• The new divertor Ball-pen and Langmuir probes on the COMPASS tokamak.