Space Charge Effects Of Emissive Probes, Investigated In A DP-Machine

R. Gstrein1, A. Marek2, C. Ionita1, P. Kudrna2, S.B. Olenici3,
P.C. Balan1, R. Schrittwieser1, M. Tichý2
1Institute for Ion Physics and Applied Physics, Leopold-Franzens University of Innsbruck, Technikerstr. 25,
A-6020 Innsbruck, Austria
2Charles University in Prague, Faculty of Mathematics and Physics, V Holešovičkách 2,
18000 Prague, Czech Republic
3Al. I. Cuza University, Faculty of Physics, 11 Carol I Blvd., RO-700506, Iasi, Romania
Abstract. We report on a systematic investigation of electron-emissive probes in a DP-machine plasma. The rationale are
perturbing effects observed with emissive probes in various plasmas and with emissive probes of various designs. These
effects might impair the accuracy of emissive probes as diagnostic tools for the direct determination of the plasma potential.
The two effects are a deviation of the floating potential of an emissive probe from the plasma potential even for a
strongly heated probe and a variation of the electron saturation current with the probe heating. We have therefore tested
conventional emissive wire probes with different wire diameters and materials and different loop lengths in the Innsbruck
Keywords: Plasma diagnostics, emissive probes, plasma potential, DP-machine, space charges
PACS: 52.75.-d, 52.70.-m, 52.70.Ds, 52.80.Tn
Emissive probes are well-known diagnostic tools for the determination of the plasma potential. The principle is
that the probe emits an electron current into the plasma which can be detected as long as the probe voltage is more
negative than the plasma potential. In contrast to the conventional cold Langmuir probe this method is insensitive to
electrons drifts and beams and does not depend on the electron temperature.
In this contribution we present systematic investigations on the variation of the current-voltage characteristic
with the heating current in a DP-machine (Double Plasma Machine). An accompanying paper is devoted to analogous
investigations in a magnetron discharge.1 The rationale of our investigations is related to two observed discrepancies
of the emissive probe:
♦ According to simple theory the floating potential of a sufficiently emitting probe should be identical to the
plasma potential. However, often it is observed that even with strong emission the floating potential remains
somewhat below the value of the plasma potential as determined from a cold probe current-voltage characteristic
(assuming a Maxwellian plasma). This effect seems to be related to a space charge forming around the probe
wire even with sufficient electron emission.2,3
♦ Whereas the magnitude of the current on the negative side of the emissive probe characteristic naturally increases
due to the electron emission current which superimposes on the ion saturation current, the electron saturation
current on the positive side should in principle not be affected by the electron emission. Nevertheless, many investigations
frequently do show an increase of the electron saturation current with the heating. Our investigations
(also Ref. 1) show that this effect seems to be related to the shape of the probe.
We point out, however, that another recent investigation did not show such effects but that in this case the current-
voltage characteristic was claimed to behave strictly according to the text book, i.e., without the above-mentioned
perturbing effects.4
The Emissive Probe Design
A conventional electron emissive probe, according to our design,2,5,6 consists of a ceramic or boron nitride tube
of a few mm outer diameter and suitable length according to the necessity of the experiment. The tube has at least
two bores of around 0,5 mm diameter. Through these bores, a wire of a refractory metal (tungsten, thoriated tungsten
or tantalum) with a diameter in the range of 0,05 to 0,2 mm is inserted in such a way that on one side of the tube (at
the “hot end”) a wire loop of a total length of a few mm is formed. In each of the bores, the W-wire extends a few
cm towards the other end (the “cold end”) of the ceramic tube. Before insertion, the two ends of the wire are spliced
equally with about 8 – 10 copper threads with diameters of 0,05 mm on a few cm length so that in the centre a piece
of the desired loop length remains uncovered. In this way, inside the bores the two ends of the probe wire are
densely covered with a thin layer of Cu so that the conductivity of these parts is larger.
If the wrapping of the W-wires with Cu-threads is performed in the right way the electrical and mechanical contact
between the probe wire and the copper is excellent. By careful choice of the number of Cu-threads for wrapping
the wire, the thickness of the combined wire ends can be adjusted so that it tightly fits into the bores of the probe
tube. This increases the electric and mechanical contact. On the cold end of each tube, only the twisted Cu-wires are
protruding and can there be connected easily to any further electrical leads and eventually to a battery or power supply.
This treatment has the effect that only the exposed loop of the emissive probe is heated when a current is passed
through the probe wire. Compared to other constructions our design is much less bulky and thus such a probe can be
constructed very small. See Fig. 1 for a schematic of our probe construction. Our group was the first to use such
probes even in fusion experiments,2,6,7,8,9 where we also found the two effects described in the Introduction.2
We have tested emissive probes with three different total lengths of the probe wire loop (5 mm, 10 mm and
20 mm), various wire diameters and the three above mentioned FIGURE 1. Schematic of the probe construction. Inside the ceramic the probe wire is spliced with thin copper wires. The connection
between the two materials is very tight and provides an excellent mechanical and electrical contact. The covering of the
probe wire with copper threads is denser than shown here for clarity.
Great care has to be taken to make meaningful measurements with emissive probes. They are more complicated
to handle than cold probes and are much more sensitive to external influences. A conventional emissive wire probe
as described here obviously can quickly melt when it is overheated. Therefore also the available emission current,
which in turn depends on the plasma density and temperature, is limited. Already before melting, the wire diameter
diminishes which leads to an increase of the electric resistance and thereby of the heating power and the emission, if
no special measures are taken. The reduction of the diameter of the wire might also have other effects which impair
the accuracy of the probe, but not so much concerning the determination of the plasma potential.
The Plasma Apparatus
DP-machines are well-established tools for basic plasma investigations.10 The Innsbruck DP-machine consists of
a vacuum cylinder of 44 cm diameter and 90 cm length. The chamber is separated into a source chamber and a target
chamber by a fine-mesh grid. The grid is isolated from the walls and usually biased to about –100 V. In each part of
the DP-machine a heated double filament of 0.2 mm diameter tungsten wire serves as hot cathode for a lowtemperature
discharge in argon. The inner side of the entire chamber is covered by rows of strong permanent magnets
with opposite polarity to increase the efficiency of the discharge.
Ceramic (Al2O3)
Probe wire (W, or BN
W+Th or Ta)
Copper threads
(0,05 mm diam.)
With a background pressure between 10–4 and 10–3 mbar and a discharge current between 50 and 300 mA, the
achievable plasma density lies in the range of 109 and 1010 cm–3. In the present experiments, plasma was produced
only in the target chamber.
Fig. 2 a,b shows typical sets of current voltage characteristics of an emissive probe taken in the Innsbruck DPmachine
Ar-plasma for a background.

cold probe, however, with the geometry rather of a cylindrical probe.
The characteristic for 0 A heating current in Fig. 2 a with a loop length of only 5 mm could be due to coating
since its electron saturation part is considerably higher than of the following characteristics. However, this probe
was then used for the first time. From the following curves we see that expectedly the magnitude of the emission
current on the left-hand side increases while the floating potential shifts to the right-hand side. In this case, extremely
high heating currents were applied which for the highest value (4 A) led to a fast reduction of the wire diameter
and thereby of the geometrical collection surface of the probe until the probe melted. For very high heating
currents we also see a clear increase of the fluctuations superimposed on the entire characteristic. But in the range
from 1 A to 3.9 A there is only a small variation of the electron saturation current.
Fig. 2 b with a loop length 4 times as long obviously also produces a much larger emission current. On the other
side the electron saturation current shows a much stronger effect with the heating current; we see a strong increase,
which behavior is closer to the expected one described in the Introduction.
In general our results also show that it is advantageous to fire a probe for a while until bright red hot before using
them for measurements. This not only cleans the surface but apparently affects also deeper layers beneath it so that
the work function has a better constancy after this treatment.
Also Fig. 3 shows the principle behavior of emissive probes when the floating potential is measured. We clearly
see that for a certain heating current the floating potential suddenly increases, approaching the value of the plasma
potential that here was determined according to the textbook from the inflection point of the cold probe characteristic.
For both loop lengths the behavior is qualitatively the same. However, very surprising is the result that in both
cases the floating potential can also exceed the cold probe value! This phenomenon was definitely not seen in denser
and hotter plasmas, as for instance on that of the CASTOR tokamak.2 But also in the case of the recently investigated
laser-heated probe in the plasma of VINETA, also for strong heating the floating potential remains clearly be low the value of the cold probe.11 On the other hand, the transgression of the floating potential of an emissive probe
over the value of the cold probe was seen sometimes also in other thin cold plasmas such as in a Q-machine.

Obviously further intensive investigations are needed to fully explain the above described phenomena, in particular
the question whether or not, and if yes, to what extent a space charge forms around an emissive probe even in
the floating case.
The work was supported by the Austrian Science Foundation (Fonds zur Förderung der wissenschaftlichen Forschung)
under grant No. L302-N02, Austrian-Czech Scientific-Technical Collaboration project A-14/2004, the
Czech Science Foundation, grants 202/03/H162, 202/06/0776 and 202/04/0360, by the Ministry of Education, Youth
and Sports, Research plan MSM 0021620834, by the and by EURATOM.
1. A. Marek, R.P. Apetrei, S.B. Olenici, R. Gstrein, I. Picková, P. Kudrna, M. Tichý, R. Schrittwieser, 13th Int. Cong. Plasma
Phys., Kiev, May 22-26, 2006, submitted.
2. R. Schrittwieser, J. Adámek, P. Balan, M. Hron, C. Ioniţă, K. Jakubka, L. Kryška, E. Martines, J. Stöckel, M. Tichý, G. Van
Oost, Plasma Phys. Contr. Fusion 44, 567-578 (2002).
3. M.Y. Ye, S. Takamura, Phys. Plasmas 7, 3457-3463 (2000).
4. N. Mahdizadeh, F. Greiner, M. Ramisch, U. Stroth, W. Guttenfelder, C. Lechte, K. Rahbarnia, Plasma Phys. Contr. Fusion
47, 569-579 (2005).
5. A. Siebenförcher, R. Schrittwieser, Rev. Sci. Instrum. 67, 849 – 850 (1996).
6. R. Schrittwieser, C. Ioniţă, P.C. Balan, Jose A. Cabral, H.F.C. Figueiredo, V. Pohoaţă, C. Varandas, Contrib. Plasma Phys.
41, 494 – 503 (2001).
7. J. Adámek et al., Czechoslovak J. Phys. 52, 1115-1120 (2002).
8. P. Balan et al., Rev. Sci. Instrum. 74, 1583-1587 (2003).
9. C. Ioniţă et al., Rev. Sci. Instrum. 75, 4331-4333 (2004).
10. R.J. Taylor, K.R. MacKenzie, H. Ikezi, Rev. Sci. Instrum. 43, 1675-1678 (1972).
11 R. Schrittwieser, C. Ioniţă, P.C. Balan, C.A.F. Varandas, C. Silva, J. Stöckel, M. Tichý, E. Martines, G. Van Oost, T. Klinger,
R. Madani, R.M.O. Galvão, the ISTTOK team, the CASTOR team, Proc. 2nd German-Polish Conf. Diagnostics of Plasmas
(Krakow, Poland, 2004), Inv12.
12 D. Strele, private communication.

Опубликовано в рубрике Documents