Measurements Of Fluctuations With Probes In The Edge Region Of Various Toroidal Plasmas
R. Schrittwieser,1 C. Ioniţă,1 P.C. Balan,1 J. Stöckel,2 J. Adámek,2 M. Hron,2
M. Tichý,3 E. Martines,4 G. Van Oost,5 H.F.C. Figueiredo,6 J.A. Cabral,6
C. Varandas,6 C. Silva,6 M.A. Pedrosa,7 C. Hidalgo,7 T. Klinger8
1Institute of Ion Physics and Applied Physics, LFUI, Association EURATOM/ÖAW, Innsbruck, Austria
2Institute of Plasma Physics, Association EURATOM/IPP.CR, AS CR, Prague, Czech Republic
3Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic
4Consorzio RFX, Associazione EURATOM/ENEA sulla Fusione, Padova, Italy
5Department of Applied Physics, University of Gent, Gent, Belgium
6Association EURATOM/IST, Centro de Fusão Nuclear, Instituto Superior Técnico, Lisboa, Portugal
7Association EURATOM/CIEMAT, Particle and Probe Diagnostic Group, CIEMAT, Madrid, Spain
8Association EURATOM/IPP, Max Planck Institute for Plasma Physics, Greifswald, Germany
Abstract. We report on investigations of electrostatic fluctuations in the edge plasma region which have been carried out
during the last few years at several European fusion experiments. Various methods and probe arrangements have been
used to determine fluctuations of the plasma potential, the electric field and the electron temperature. Investigations were
undertaken at ISTTOK (Instituto Superior Técnico TOKamak), Lisbon, Portugal, at CASTOR (Czech Academy of Science
TORus), Prague, Czech Republic, and at the TJ-II Flexible Heliac at CIEMAT in Madrid, Spain.
Keywords: Toroidal plasma, edge fluctuations, plasma turbulence, plasma potential, electric probes
PACS: 52.25.Fi, 52.25.Xz, 52.35.Ra, 52.55.Fa, 52.55.Hc, 52.70.-m
Fluctuations in the edge region of a toroidal plasma cause radial transport and loss of plasma. Radial particle
transport due to strong electric field fluctuations1 may account for a major part of the anomalous energy and particle
losses.2,3 A counteracting effect may be the Reynolds stress, a gradient of which might cause sheared poloidal
flows.4,5,6,7,8,9 This can cause turbulent eddies to be tilted and elongated poloidally, thereby reducing the radial turbulent
transport. This mechanism plays a key role in explaining the L-H transition.10,11,12
For these phenomena the most relevant parameters are the radial and poloidal electric field components Er,θ and
the electron temperature Te. We have used various types of probes for a direct determination of the plasma potential
Φpl and of Er,θ. Among them were (i) emissive probes13,14,15,16 and (ii) ball-pen probes.17 Ball-pen probes work only
in a magnetic field. For the measurement of electric field components and the electron temperature, various combinations
of emissive or ball-pen probes and cold probes were used. From these parameters also the radial fluctuationinduced
particle flux18 and the Reynolds stress19 could be derived. Such investigations were carried out on ISTTOK
(Instituto Superior Técnico TOKamak), Lisbon, Portugal, on CASTOR (Czech Academy of Science TORus), Prague,
Czech Republic, and on the TJ-II Flexible Heliac at CIEMAT in Madrid, Spain.
PLASMA POTENTIAL PROBES AND A FEW RESULTS
There are few diagnostic tools to determine the plasma potential with sufficient accuracy and spatial and temporal
resolution. The least expensive and most easily to handle tool is the cold probe (Langmuir probe). Usually the
well-known relation Φpl = Vfl + Teln(Ies/Iis) = Vfl + αTe between the floating potential Vfl of a cold probe and Φpl is
used.14 However, for this we need to know Te; and also the ratio of the electron to the ion saturation currents in α =
ln(Ies/Iis) depends on Te. In typical tokamak edge plasmas α ≅ 2 – 3. The electron temperature is not easily measured
with sufficient reliability and temporal resolution, in particular in the edge region of a magnetically confined toroidal
fusion plasma where there are strong gradients and fluctuations of Te. An additional often ignored fact is that the relation
Φpl = Vfl + αTe is only valid for a Maxwellian plasma. As soon as there is a considerable electron drift or electron
beam, the entire I-V characteristic, and thereby also Vfl, shifts to the left due to the drifting electrons.
Although emissive probe were discussed by Langmuir in the 1920-ies,20 one of the first such probe was presented
by Sellen et al.21 Since then emissive probes are standard tools in laboratory plasmas,13,14,15,22,23 but only our group
has started to use them also in fusion experiments.13,14,15,16,19 The emission current Iem can be observed as long as the
probe potential is more negative than the plasma potential, irrespective of electron drifts or beams. In this case, however,
also only in a Maxwellian plasma, the above relation becomes:
can be determined under the assumption that also the poloidal fluctuating
velocity is due to the drift E × B r
Fig. 3 shows the radial fluctuation-induced particle flux determined
according to Eq. (2) in the edge region of ISTTOK. In general
it turns out that the fluctuations of Vfl (measured with the probes
not heated) are considerably smaller than those of Φpl (measured
with heated probes). Consequently also the turbulent particle flux
measured with the emissive probes is significantly larger than that
measured with cold probes, since in the latter case also temperature
fluctuations are superimposed. These results indicate that temperature
fluctuations are correlated with density fluctuation but not with
those of the floating potential. This also suggests that in the ISTTOK
edge plasma temperature fluctuations are important for the estimation
of the particle flux and therefore the standard method based
on cold probe measurements is not valid.
Laser-Heated Emissive Probe
Another way of heating an emissive probe was investigated recently, namely to heat a piece of LaB6 or graphite
by a laser to sufficiently high temperatures for electron emission.25,26 Something similar was attempted only once before.
Small cylindrical pieces of 2 mm diameter and heights of 4 mm of LaB6 or graphite were used as probe tips. The
electrical connection was made with an Mo wire of 0.2 mm diameter wound around the probe tip. The tip was
heated from the front side through a quartz-glass window by an infrared high-power diode laser JenLas HDL50F
from JenOptik, Jena, Germany, with a maximum laser power of 50 W at 808 nm. The laser beam is coupled into a
fiber cable of 3 m length ending in an output head, with which a focal spot of 0.6 mm diameter is produced in a distance
of 20 cm. It turned out that the behavior of such a probe is the same as that of a conventional emissive wire
probe, but it has several advantages such as longer lifetime and higher electron yield.
Also triple probes can be used to measure the electron temperature.28 This was done in the edge region of the
flexible heliac TJ-II at CIEMAT in Madrid. The plasma is produced by electron cyclotron resonance heating
(ECRH), while starting by t = 175 ms additional heating by neutral beam injection (NBI) was applied. Fig. 4 shows
an example of the temporal evolution of the electron temperature during a discharge. As we can see, Te shows a
rather strong drop of more than 20 eV after NBI was switched on.
A ball-pen probe consists of a cylindrical collector with a conical
tip of 2 mm diameter, which can be moved up and down inside a
screening tube of boron nitride.29,30 In Fig. 5 an additional Langmuir
probe ring is mounted on the BN tube. This acts as conventional
cold probe to deliver the floating potential similar to the abovedescribed
method.31 The parameter h indicates the position of the
collector relative to the tube, with h = 0 indicating that the collector
tip lies exactly in the plane of the mouth of the tube. The measurement
of the plasma potential Φpl by means of the ball-pen probe utilizes
the different electron and ion gyroradii in a magnetized plasma.
Since the former are on the average much smaller, they are easily
FIGURE 3. Radial fluctuation-induced particle
flux measured by the system shown in Fig. 2 in the
edge region of ISTTOK; red dots and line: emissive
probes were heated; blue dots and line: probes
were not heated. The radius a of the LCFS was
FIGURE 4. Temporal evolution of the electron
temperature in the edge region of TJ-II in Madrid.
Starting for t = 175 ms also neutral beam heating
was turned on.
screened off by the BN tube, when the collector is withdrawn inside
(Fig. 5(a)), while ions can still reach the probe collector. When the
magnitudes of the electron and ion saturation currents are equal, the
floating potential of the collector becomes equal to Φpl.
We have shown that with special probes also in toroidal fusion
experiments the plasma potential and its fluctuations can directly be
measured, which makes it possible to derive also further important
parameters characterizing various features of edge plasma turbulence,
among them the radial fluctuation-induced particle flux and
the Reynolds stress.
This work was carried out within the Associations EURATOM-ÖAW, -IST, -IPP.CR, -CIEMAT, -ENEA and –
IPP. The content of the publication is the sole responsibility of its author(s) and it does not necessarily represent the
views of the Commission or its services. Part of the work was supported by project 202/03/0786 of the Czech Republic
1 T. Huld, A.H. Nielsen, H.L. Pécseli, J. Juul Rasmussen, Phys. Fluids B 3, 1609-1625 (1991).
2 B.A. Carreras et al., Phys. Plasmas 3, 2664-2672 (1996).
3 V. Naulin, New J. Phys. 4, 28.1-28.18 (2002).
4 P.H. Diamond and Y.B. Kim, Phys. Fluids B 3, 1626-1633 (1991).
5 B.A. Carreras, D. Newman, P.H. Diamond, Y.-M. Liang, Phys. Plasmas 1, 4014-4021 (1994).
6 C. Hidalgo et al., Phys. Rev. Lett. 83, 2203-2205 (1999); Plasma Phys. Control. Fusion 42, A153-A160 (2000).
7 C. Hidalgo, M.A. Pedrosa, B. Gonçalves, New J. Phys. 4, 51.1.-51.12 (2002).
8 G.G. Craddock, P.H. Diamond, M. Ono, and H. Biglari, Phys Plasmas 1, 1944-1952 (1994).
9 S.B. Korsholm et al., Plasma Phys. Control. Fusion 43, 1377-1395 (2001).
10 F. Wagner et al., Phys. Rev. Lett. 49, 1408-1412 (1982).
11 K. Burrell, Phys. Plasmas 4, 1499-1518 (1997).
12 P.W. Terry, Rev. Mod. Phys. 72, 109-165 (2000).
13 R. Schrittwieser et al., Contrib. Plasma Phys. 41, 494-503 (2001).
14 R. Schrittwieser et al., Plasma Phys. Contr. Fusion 44, 567-578 (2002).
15 J. Adámek et al., Czechoslovak J. Phys. 52, 1115-1120 (2002).
16 P. Balan et al., Rev. Sci. Instrum. 74, 1583-1587 (2003).
17 J. Adámek et al., Czechoslovak J. Phys. 55, 235-242 (2005).
18 P. Balan et al., 29th EPS Conf. Plasma Physics and Contr. Fusion (Montreux, Switzerland, 2002) Europhys. Conf. Abstr. 26B,
19 C. Ioniţă et al., Rev. Sci. Instrum. 75, 4331-4333 (2004).
20 I. Langmuir, Phys. Rev. 22, 357-398 (1923).
21 J.M. Sellen Jr., W. Bernstein, R.F. Kemp, Rev. Sci. Instrum. 36, 316-322 (1965).
22 J.R. Smith, N. Hershkowitz, P. Coakley, Rev. Sci. Instrum. 50, 210-218 (1979).
23 A. Siebenförcher, R. Schrittwieser, Rev. Sci. Instrum. 67, 849-850 (1996).
24 P. Balan et al., 16th IAEA Technical Meeting on Research Using Small Fusion Devices (Mexico City, Mexico, 2005).
25 R. Madani, C. Ioniţă, R. Schrittwieser, T. Klinger, Proc. 12th Int. Cong. Plasma Phys., Nice, France, Oct. 25 – 29, 2004,
26 R. Schrittwieser et al., Physica Scripta 73, 94-98 (2006).
27 M. Mizumura, S. Uotsu, S. Matsumura, S. Teii, J. Phys. D: Appl. Phys. 25, 1744-1758 (1992).
28 Sin-Li Chen, T. Sekiguchi, J. Appl. Phys. 36 , 2363-2375 (1965).
29 J. Adámek et al., Czechoslovak J. Phys. 54, C95-C99 (2004).
30 J. Adámek et al., Czechoslovak J. Phys. 55, 235-242 (2005).
31 J. Adámek et al., 32nd EPS Plasma Physics Conference (Tarragona, Spain, 2005), Europhys. Conf. Abst. 29C (2005), P-5.081
Опубликовано в рубрике Documents