Super-thermal and runaway electrons at reconnection events during JET disruptions. V.V. Plyusnin 1, B. Alper 2, J. Mlynar 3, P. Helander 2, R J. Hastie.

Super-thermal and runaway electrons at reconnection events during JET disruptions.

V.V. Plyusnin1, B. Alper2, J. Mlynar3, P. Helander2, R J. Hastie2, F. Salzedas1, F. Serra1, R. Jaspers4, T.C. Hender2, V.G. Kiptily2,

M. F. Johnson2, E. de La Luna5 and JET EFDA contributors*.

1Association Euratom-IST, Centro de Fusão Nuclear, Lisbon, Portugal2Euratom/UKAEA Fusion Association, Culham Science Centre, Abingdon, UK3Association Euratom - IPP.CR, Prague, Czech Republic

4Association Euratom-FOM, Nieuwegein, The Netherlands 5Association Euratom-CIEMAT, Madrid, Spain*See appendix in J. Pamela et al., Fusion Energy 2002 (Proc. 19th Int. Conference Lyon, 2002).

8th IAEA TCM on Energetic Particles in Magnetic Confinement Systems 6-8 October 2003 San Diego, California USA

Disruptions and runaway electrons A series of the experiments on major disruptions provoked by intense gas puff, as well as detailed analysis of

spontaneously occurring disruptions have been carried out to further understanding of the trends of disruption

induced runaway process. Helium, argon and neon were used to achieve the density limit conditions and disruption

initiation. Other reasons for disruptions in JET were a consequence of the device operation near instability

thresholds on safety factor, plasma pressure or plasma inductance [1]. Runaway electrons generation has been

observed in disruptions that occurred in a wide variety of different experimental conditions, namely, various

triangularities and elongations, current rise stage, gas puff, Vertical Displacement Event (VDE), etc [2,3].

#53790 #54047 #54048

Runaway electrons have been detected using the hard X-ray and neutron diagnostics. The estimates of the runaway

electron energy have been carried out from the measurements of scattered -ray radiation with 250 and 300 ms time-bin

integration. The spectra of scattered hard X ray emission were analyzed and evaluated energies E have been taken as a

low-limits for the runaway electron energy. Neutron emission bursts (Figure 1) are the direct evidence that runaway

electrons possess the energy E >11 MeV, if these neutrons are the result from Fe(,n)-reaction, or, even, E >19 MeV, if

12C(,n)-reaction is the main neutron source. However, obtained data doesn’t allow reliable evaluation of contributions

from each source. In some cases, especially at low energies, the runaway beam also produced an observable soft X-ray

image by exciting plasma impurity ions [3].

Evolution of magnetic fields structure during disruption.Soft X-ray inverse reconstruction has been used to highlight the changes in magnetic field structure during disruptions

(Figures 2-4). This reconstruction is based on the method developed at JET and applied to study in detail transport

processes of injected trace impurities and the very localized SXR emission during ELMs [4]. Carried out in [4] an error

analysis on the tomographic reconstructions shows that, although the uncertainties become larger towards the centre of

the plasma, even hollow profiles can be reconstructed reliably. Similarly to issues of [4], the tomographic reconstruction

of soft X-ray emission in T-10 tokamak [5] has demonstrated the possibility to detect the evolution of magnetic field

structure. This evolution has revealed the formation of large m=1 island and hollow emission profiles during disruption.

50.94 50.941 50.942 50.943 50.944 t (sec) 1.6

1.7

1.8

1.9

2

Ipl (MA)

Figure 3. Example 2 of the magnetic field re-

arrangement. Pulse#54047. The appearance of axis-

symmetric configuration is detected after t=10.9436 s.

Figure 1. Typical examples of the disruptions in discharges #53790 (B0=3 T, Ipl=1.87 MA), #54047 (B0=3 T,

Ipl=1.9 MA) and #54048 (B0=3 T, Ipl=1.9 MA). All discharges disrupted under density limit conditions due

to intense argon (#53790 and #54047) and helium (#54048) puff.

Figure 2. Example 1 of the magnetic field evolution during disruption.

Pulse #53790. An axis-symmetric confining configuration has been

created again in a very short time-scale (~200 microseconds) after

incomplete reconnection (t=10.5394) and abrupt loss of the plasma energy.

The square in the lower corner of all frames indicates the grid size and the

green line separatrix. Intensities of soft X-ray emission in all frames are

shown in arbitrary units.

#53790: Te at t=10.5392 sec

Ipl (MA)

neutrons (a.u.)

Hard X-rays (a.u.)

neutrons (a.u.)

Hard X-rays (a.u.)

neutrons (a.u.)

Ipl (MA)

Hard X-rays (a.u.)

Ipl (MA)

Figure 4. Example 3. Pulse #54048. Despite existence of axis-symmetric configuration of magnetic fields the runaway

electrons generation hasn’t been observed. Relatively high Te (~100 eV, see contour plot in time vs. radius) is considered to

be the main reason. This resulted in low values of the electric fields appeared at thermal quench and low plasma current

derivative at current quench stage.

Evidence for super-thermal electron generation at reconnection event during disruption.

0cm

Tln

4

7

E

2E

E4

E

lnΛ I

I

Rμ

L

3

2πH 2

e

e

||

D

||

D

A

0

0

where , IA = 0.017 MA – Alfven current, L- plasma column inductance, 0 =4*10-7, R

- major plasma radius, ED=e3ln neZeff/402Te – is the Dreicer’s field Zeff is the

effective ion charge and ln is the Coulomb logarithm, E|| - electric field.

Thermal quench and runaway generation process.Observed trend of the decrease of runaway generation efficiency at disruptions with increase of

the electron temperature [3] is in adequate agreement (Figure 5) with results of [6], where the

criterion for estimation of runaway probability at disruptions has been proposed in a form:

During disruptions short (order of 100 microseconds) bursts of electron cyclotron emission (ECE) have been detected in the second harmonic electron cyclotron frequency range by heterodyne radiometer (Figures 6-9). Unlike

[7], these ECE bursts have been observed during disruptions occurring in different experiments (reversed shear, density limit, etc). First analysis of this phenomenon suggests that tearing mode instability can be considered as a

main reason for re-arrangement of magnetic configuration and disruption. At strong re-arrangement of the magnetic configuration the super-thermal (or even runaway) electrons could be generated due to acceleration in electric

field determined by the rate of change of the perturbed helical magnetic field during reconnection: E1~-dB*/dt, where B*=Bpol*(1-q). Appearance of such short and intense ECE emission bursts has been resulted from the

scattering of non-thermal electrons on background plasma due to Coulomb collisions.

Figure 6. The intensity of ECE bursts (contour plot) has achieved from one to two orders of the magnitude above the thermal level. Bursts have been observed during the thermal quench at the time, when soft X-ray diagnostic indicates that a reconnection event occurs in magnetic configuration

Figure 7. Detailed analysis of the bursts appearance trends has revealed that they mainly are localized in the vicinity of q=2 surface immediately after the TeECE signal vanishes due to the energy quench.

Figure 9. The intensity of the ECE bursts has been found depending on the pre-disruption plasma current values. Obtained trend can be added by experimental observation of the bursts of the hard X-rays signal at the beginning of the minor and major disruptions in JET, especially, in early JET pulses when the detector sensitivity was much higher than now.

Figure 8. Pulse #54047. Usually, the bursts have been remaining narrow, but sometimes the broadening of ECE profile has been observed with maximum intensity location near q=2.

Acknowledgements. This work has been carried out within the framework of the European Fusion Development Agreement and supported by the European Communities and “Instituto Superior Técnico” under the Contract of Association between EURATOM and IST. Financial support was also received from “Fundação para a Ciência e Tecnologia” in the frame of the Contract of Associated Laboratory. The views and opinions expressed herein do not necessarily reflect those of the European Commission, IST and FCT.

References.[1] J. Wesson, R.D. Gill, M. Hugon et al. Nuclear Fusion 29 (1989) 641[2] R.D. Gill et al. Nuclear Fusion 42(2002)1039[3] V.V.Plyusnin et al. 30th EPS Conference, July 7-11 2003, St. Petersburg, report P-2.094 (2003).[4] L.C. Ingesson, B.Alper, H.Chen et al. Nuclear Fusion 38(1998) 1675 - 1694.[5] P.V. Savrukhin, E.S. Lyadina, D.A. Martynov, D.A. Kislov, V.I. Poznyak. Nuclear Fusion 34 (1994) 317[6] P.Helander et al. Plasma Phys. Control. Fusion 44 (2002) B247–B262[7] A.Janos et al. Rev. Sci. Instruments 68 (1) January 1997 505-508

#53790

#56955#58383

#54047

Te (eV)

H

0 100 200 300 400

-20

-10

0

10

20

1 MA 2 MA

3 MA

Figure 5. H parameter is calculated in

0-D approximation for 3 values of

pre-disruption plasma currents in JET

ensuring that runaway generation

could be expected if the electron

temperature is lower than certain

value (H>0) for the given plasma

currents.

Radius, mRadius, m

Ipl (MA)

EC

E I

nte

ns

ity

(a.

u.)

0 0.5 1 1.5 2 2.5 3 3.5 4 102

103

104

105

106

49.12 49.128 49.136 49.144 49.152 time (s) 0

Hard X ray (a.u.)

0.8

1.6

Iplasma (MA)

53.66 53.67 53.68 53.69 t (s) 0

1.2

2.4

3.6

Hard X rays (a.u.)

Iplasma (MA)

q-profile (EFIT)

q-profile (EFIT)

Radius, m

8th IAEA TCM on Energetic Particles in Magnetic Confinement Systems 6-8 October 2003 San Diego, California USA

minor disruption #6591 #7118 B0= 3.4 T