Experiments with Thin Electron Beam at GOL-3

8 th International Conference on Open Magnetic Systems for Plasma Confinement July 8, 21, Novosibirsk, Russia Experiments with Thin Electron Beam at GOL-3 V.V. Postupaev, A.V. Arzhannikov, V.T. Astrelin, V.I. Batkin, A.V. Burdakov, V.S. Burmasov, I.A. Ivanov, M.V. Ivantsivsky, K.N. Kuklin, S.A. Kuznetsov, M.A. Makarov, K.I. Mekler, S.V. Polosatkin, S.S. Popov, A.F. Rovenskikh, A.A. Shoshin, S.L. Sinitsky, V.F. Sklyarov, N.V. Sorokina, A.V. Sudnikov, Yu.S. Sulyaev, L.N. Vyacheslavov Budker Institute of Nuclear Physics, 639, Novosibirsk, Russia Novosibirsk State University, 639, Novosibirsk, Russia Novosibirsk State Technical University, 6392, Novosibirsk, Russia V.V.Postupaev@inp.nsk.su

Outline of the Talk 1. Introduction 2. Relaxation of the thin electron beam in the plasma 3. Heating of the plasma and of non-ionized deuterium 4. Stabilization of the beam 5. Summary V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 2/26

GOL-3

Motivation of This Activity GOL-3 experiments are encouraging for a fusion prospects of such systems. Several new physical effects were found which benefit confinement. Future multiple-mirror reactor must be a steady-state system. The relevant physics and technology is to be developed. Feasible reactor will have moderate power of the electron beam: OS-21: A.V.Burdakov, Tuesday, 11:1 OS-21: V.T.Astrelin, poster 47 tens GW tens MW (dictated by economics) Development of fusion technology: long-pulse electron beams. Development of physics in a new domain of parameter space at the existing device: - lower specific parameters of the electron beam, - smaller physical size (radius) of the beam-heated zone this paper OS-21: I.V.Kandaurov, poster 51 OS-21: V.V.Postupaev, poster 45 V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 4/26

Physical Problems for a Thin Beam Experiments with decreased radius of the electron beam in the plasma: Theory and most of experiments deal with a large plasma ( Ø >> 2πc/ω p ) will the beam-plasma interaction change at comparable spatial scales? Up to date axial heat losses were prevailing. Smaller beam size worsens ratio of the plasma volume to its surface will transverse losses dominate? The beam current creates highly sheared magnetic field due to turbulent plasma resistance in the beam-heated zone. This benefits confinement of axisymmetric plasma, but is dangerous due to large axial currents which make a q < 1 plasma. will this stabilization by the magnetic shear work for a lower beam current? Current GOL-3 system strictly prohibits experiments with the beam injection into a neutral gas (1% disruptions with a heavy damage to in-vacuum components). Low beam current will make such experiments possible. Some plasma diagnostics benefit from smaller beam size (geometry of ports). V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 5/26

Multiple-Mirror Mirror Trap GOL-3 U-2 generator of the electron beam «Standard» regime B, T 8 6 4 2 sheet beam diode corrugated magnetic field Thomson NBI solenoid exit unit -4-2 2 4 6 8 1 12 14 axial coordinate, m 8PO1E Magnetic Field multiple-mirror 52 mirror cells 4.8/3.2 T (max/min) Electron Beam Energy ~.8 MeV Current <2 ka Duration ~12 μs Diameter 4 cm @ 3.2 T Energy ~.12 MJ Plasma Length ~12 m n ~ 1 2-1 22 m -3 T ~ 1 kev Lifetime ~.5 ms V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 6/26

Thin Beam Configuration at GOL-3 primary storage U c=42 kv Q =.5 MJ c U I shape transformer slit channel compression zone 2 2 cm Ø 6 cm limiter Initial uncompressed beam 75 3.5 cm 2 E =.8 MeV I =3kA τ =9 12 μs OS-21: S.L.Sinitsky, poster 43 cathode 75 3.5 cm Two regimes of the beam injection Vacuum chamber Limiters and foreplasma Ø 1 mm Ø 8 mm «thick» (or standard) 4cm «thin» 1.3 cm j ~ 1 ka/cm 2 Q ~7kJ Hot core plasma becomes much smaller than the cold edge plasma thin beam. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 7/26

Conditions in Confinement Zone Kr D 2 D 2 Initial gas density distribution concentration, 1 2 atoms/m 3 1 1.1 maximum and minimum density regimes 2 4 6 8 1 12 14 distance from the input mirror, m gas-puffing technology enables producing a nonuniform density profile along the solenoid; three valves are in operation; typical n max = (2 8) 1 2 m -3 Three different initial conditions of the experiment: the beam injection into vacuum at ~3 1-3 Pa residual pressure; the beam injection into neutral deuterium at shown density profile; the beam injection into a low-temperature preliminary plasma. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 8/26

Net Plasma Current U, MV injection into the neutral gas net current corresponds to the beam current 1 diode voltage.5 PL9216 injection into the plasma the beam current is screened by the plasma the beam current counters the discharge one 6 PL97 4 beam injection J, ka 1.5 net (beam) current J, ka 2-2 4 8 12 time, microseconds -4 2 4 6 8 time, microseconds Note: a special low-current shot is shown. Standard shots are of 1.5-2 ka, but measurements are impossible due to emerging plasma currents which heavily distort the waveform. waveforms of 8 Rogowski coils at different coordinates are shown V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 1/26

Efficiency of the Beam Relaxation Dynamics of the mean energy of the beam electrons after passing the plasma U diode,mv E av,mev.8 PL 822.8.6 53%.6.4.4.2.2 5 1 t, μs OS-21: S.L.Sinitsky, poster 44 At the beam injection into the vacuum the mean electron energy is ~9% of the initial one some beam-plasma interaction exists due to gas in compression area. Decrease of the mean energy reaches ~5% at the beam injection into the plasma. Analyzer signals are continuous and last for the expected period at the beam injection into vacuum. At the beam injection into the gas or into the plasma, the waveforms experience simultaneous breaks and recoveries that may evidence large beam displacement. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 11/26

Plasma Emission at 2ω p Process: non-linear merging of two Langmuir plasmons with emission of 2ω p photon ( k l 1,ω P ) ( k ph 3, 2 ω P ) f = 25 41 GHz, 4 spectral channels ( k l 2,ω P ) OS-21: A.V.Arzhannikov, Tuesday, 12:25 U, V 3/2nTS, 1 19 ev/cm Ud,MV.8.4 1.2.1 PL946 cathode voltage diamagnetism microwaves 312 GHz fragment of two microwave waveforms is it an evidence of localized emission? signal, a.u.,2,1 PL946 275 GHz 2.4 1 2 m -3 312 GHz 3. 1 2 m -3-2 2 4 6 8 1 12 time, microseconds 4.1 4.2 4.3 4.4 4.5 4.6 time, microseconds V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 12/26

Profile of Diamagnetic Pressure 3/2(neTe+niTi)S, 1 19 ev/cm 1.2 1.8.6.4.2 PL97-PL9163 (statistics for 3 shots is shown) Q dia = 45 J n max = (3 4) 1 2 m -3 the green line was drawn by hand for convenience only! 2 4 6 8 1 12 Z, m The profile consists of a decreasing pedestal and two hot spots at ~.8 and ~2 m. Slight differences in the regime lead to slightly different axial pressure profile. The hot spots are the features of the beam relaxation in the nonuniform plasma. Large spread of measurements at.8-2.1 m is due to displacements of the peaks. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 14/26

Plasma Heating: Diamagnetism Beam-heated area is 1% of the standard one Waveforms for 3 different locations are shown PL8626: standard beam diamagnetism, a.u. (raw signals) PL9125: thin beam 5 126 cm 4 3 2 1 3 29 cm 2 1 1.2 524 cm.8.4 5 1 5 1 time, microseconds 5 1 The plasma heating by the thin beam is as good as before. Slight differences in the regime lead to slightly different axial pressure profile. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 15/26

Neutron Emission from D-D D D Reaction flux, 1 9 cm -2 s -1 1 1 Blue: reference standard regime, shot PL8626 Red: thin beam, shot PL9142 The signals are normalized for the beam area.1 1 = S thick S thin.1 2 4 6 time, microseconds All earlier observed at GOL-3 features of neutron emission are reproducing. Decay time at the confinement stage is good at least the hottest zone is stable. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 16/26

Peak Plasma Pressure 3/2(neTe+niTi)S, 1E19 ev/cm 2.5 2 1.5 1.5 β vac? 3 35% B = 3.2 T coordinate Z = 2.9 m shot PL9142 5 1 15 2 time, microseconds if the energy is kept within a formal flux tube beam size If the energy stays within the formal beam flux tube, then β vac 3 35% (!) If the energy spreads to farther radii, the beta value is lower. Origin of this pressure: is there a lot of trapped fast electrons? Position of this pressure peak along the axis is easily controllable by axial density profile. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 17/26

Thomson Scattering Data Diagnostics is located at Z = 415 cm (far from the hottest zone) Two radial points: and 6 mm OS-21: S.S.Popov, poster 39 edge center detection Three observed cases: <E> center > <E> edge laser dn e/de e e,, cm -3 /nm 1 13 1 12 1 11 1 1 PL959 <E> ~4 ev center <E> ~1 ev edge <E> center ~ <E> edge <E> center < <E> edge 1 9.5 1 E, kev Electron distribution function is non-maxwellian. Hot core displacement comparable to its radius is observed. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 18/26

Beam Injection Into Neutral Deuterium At the injection into the gas the beam relaxation efficiency is ~3%, i.e. about the half of the plasma case. Final plasma energy is also ~5% of the plasma case. The plasma size is wider than the expected beam diameter. OS-21: S.L.Sinitsky, poster 44 intensity, a.u. Profile of OV line at 76, nm 6 4 2 beam size -15-1 -5 5 1 15 transverse coordinate, mm OS-21: N.Sorokina, poster 37 Density increment at 1 μs after the beam start (NBI data) nl, 1 15 cm -2.5.4.3.2.1 beam size -6-4 -2 2 4 6 X, cm Excursions of the beam are detected (VUV, NBI, Thomson). The beam cross-section is close to the expected one (better than to a factor of 1.5) After the beam shot the plasma emerges anywhere within the vacuum chamber. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 19/26

Stabilization of the Beam Пучок Some possible reasons for the beam instability: decompensation by current density, (q < 1 within the beam cross-section); beam asymmetry at the input into the plasma; asymmetry of the return plasma current. B gas puffing Possible ways of stabilization: improve frozening of the magnetic field into the beam receiver; improve plasma conductivity near the beam receiver for more uniform return current. Technology: puffing of a heavy gas near the exit receiver. beam + plasma target bremsstrahlung imaging system OS-21: I.A.Ivanov, poster 5 V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 21/26

Beam Footprint at the End Receiver X-ray images of the beam footprint 1 μs frame vertical position, mm? with????? gas PL967 #967???? window?????????????????????,????? no gas???? PL9629 #9629 window horizontal position, mm?????????????? horizontal position, mm Asymmetry of the footprint at different Krypton puffing variance of normalized intensity, a.u. 3 2 1 9 8 2 4 delay after Kr puffing, ms n Kr = 1.5 1 21 m -3 V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 22/26

Improvement of Stability is Statistical Statistics for the net energy lost to the wall during the shot (probability distribution) P(Qw).4.3.2 n Kr = no gas puffing n Kr = 1.5 1 21 m -3.1.4.3 5 1 15 Qw,J n Kr = 1.5 1 21 m -3 good shots can be found in both regimes; P(Qw).2 no catastrophic events with Kr;.1 improvement is statistical but valuable; losses to the wall are comparable with peak plasma energy; what does this mean combined with good confinement? 5 1 15 Qw,J V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 23/26

Summary Main results from GOL-3 experiments with the thin beam are the following: the measured beam relaxation efficiency reaches ~5%; the beam relaxation in the plasma is twice better than in the gas; transient events of 2ω p emission with ~1 ns duration are detected; in general diamagnetic energy content is consistent with the decreased beam cross-section, the confinement is good in best shots; electron distribution function is non-maxwellian one, it should be studied better in the hot spots ; radial excursions of the beam are detected; in the gas shots the plasma emerges anywhere in the vacuum chamber; stabilization by the gas-puffing statistically works. V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 25/26

The Really Last Slide Thank You! V. V. Postupaev Experiments with Thin Electron Beam at GOL-3 8th Open Systems Conference Novosibirsk, July 8, 21 26/26