第7回原研若手研究会
2004年3月19日
電磁加速プラズマ流の制御と
マッハプローブの特性評価
電磁加速プラズマ流の制御と
マッハプローブの特性評価
東北大学 大学院工学研究科
戸張 博之
e-mail : tobari@ecei.tohoku.ac.jp
共同研究者:渡辺 俊明,渡辺 貴史,安藤 晃,服部 邦彦,犬竹 正明Keywords
:
Plasma Flow
, Mach Probe, Plasma Acceleration,
Electric Propulsion, MPD Thruster
Outline
Outline
1. Introduction
2. Electric Propulsion
3. Mach Probe Experiment in the HITOP Device
4. Magnetic Nozzle Acceleration of MPDA Plasma
5. Summary
Introduction
Introduction
プラズマの“流れ”と電磁場との相互作用
天体プラズマの物理
宇宙ジェットの形成 太陽フレア (太陽のパラドクス)プラズマの閉じ込め改善
電気推進機の開発
人工衛星の姿勢制御 径電場とポロイダルフローシア 流れによる渦の分断 惑星間航行 有人火星探査 核融合プラズマ推進 断熱層の形成Image of MUSES-C ion engine
動圧を利用した高ベータ プラズマ閉じ込め
TOHOKU UNIV. 100 102 104 106 102 103 104 105 T h ru s t d e n s it y [N /m 2]
Specific im pulse [sec]
M P D Thruster The rm a l Arcjet C hem ical P P T Ion E n gine Ha ll T hruster
Hall thruster MPD thruster
Thermal arcjet Ion engine
zIonization by electric power
zPower source: solar cell, nuclear reactor zHigh Isp (≧103-104sec)
with small consumption of propellant
zUseful for interplanetary mission
long-term station-keeping manned Mars mission
Chemical Propulsion (CP)
zLarge thrust densityto lift off the earth gravity
zLow Isp (≦500sec)
Parameters for Thrust Performance U m F = : Thrust
[ ]
sec : Impulse Specific g U g m F Isp = =Introduction to Electric Propulsion(EP)
Introduction to Electric Propulsion(EP)
Recent Achievement of EP in Japan
Recent Achievement of EP in Japan
Image of MUSES-C ion engineGround test of MPDT
>>May 9, 2003
The MUSES-C ( Hayabusa ) spacecraft mounting four ECR ion
thrusters was successfully
launched. Asteroid sample return mission is now under progress.
>>March 18, 1995
The MPD thruster onboard the Space Flyer Unit (SFU) was successfully pulse-operated in space with few misfirings.
Outline
Outline
1. Introduction
2. Electric Propulsion
3. Mach Probe Experiment in the HITOP Device
4. Magnetic Nozzle Acceleration of MPDA Plasma
5. Summary
Plasma Flow Measurement
Plasma Flow Measurement
¾Laser Induced Fluorescence (LIF)
¾Visible-Light Spectroscopy (Doppler shift) ¾Mach Probe →簡便で,空間分解能に優れる.
流れに対して異なった方向に捕集面を向け,そのイオン飽和電流値 jisの信号値の
違いから流速,イオンマッハ数 ( Mi = 流速 / 音速 )を求める
Theoretical Model of Mach Probe
~up-down~
Theoretical Model of Mach Probe
~up-down~
< up-down タイプ>
Hudis and Ridsky model (1970) 非磁化プラズマ
最初にマッハプローブを提案
1次元のエネルギー保存より導出 Mi << 1, Ti << Te
Chung and Hutchinson (1988) 磁化プラズマ 1次元Kinetic model, 粘性の効果を考慮 Hutchinson model (1987) 磁化プラズマ 1-D fluid model,粘性による輸送効果を考慮,Mi < 1 Stangeby model (1984) 磁化プラズマ 1-D fluid model,粘性の効果は無視,Mi < 1 ρi< rp : 磁化プラズマ ρi> rp : 非磁化プラズマ Mc を決めるために… LIFを用いた較正実験:Gunn(2001) 磁化プラズマ中で粘性を考慮したモデ ルとよい一致(実験は Mi < 0.4 ) 非磁化プラズマ中ではモデルが確立さ れていない!!
const.
:
exp
c i down up cM
M
M
J
J
=
TOHOKU UNIV.Hutchinson PIC Simulation
Hutchinson PIC Simulation
PICコードを用いて,非磁化プラズマ中に球プ ローブがある際,プラズマの流れ ( vf )に対して ある角度(θ )をもった点に流れ込んでくるイオン フラックス Γ (vf , θ ) を計算. シミュレーションより得られたモデル式
(
)
( )
=
=
Γ
Γ
c i down up f fexp
0
,
,
M
M
J
J
v
v
π
(example) イオンフラックスのcosθ依存(Ti / Te =1); 曲線上の数値はイオンのドリフト速度を表す ただし vf : (Te / mi )1/2 で規格化 Γ (vf , θ ) : ni∞(Te / mi )1/2 で規格化Theoretical Model of Mach Probe
~perp-para~
Theoretical Model of Mach Probe
~perp-para~
< perp-para タイプ> 通常のプローブの理論よりperp-tipのイオン飽和電流 J⊥は, i i i e e i
m
T
T
en
J
⊥=
κ
γ
+
γ
( κ : Ti / Te で決まる定数 ) Mi << 1 では… Mi > 1 では…
=
⊥2
exp
2 i ||M
J
J
κ
γ
γ
i i i i e e ||M
m
T
T
U
J
J
=
+
=
⊥U
en
J
||=
i Kuriki and Inutake (1974)• • •(ii)
Stangeby and Allen(1971)
(i) ,(ii) が Mi = 1 でなめらかに接続す るように下式のようにαを導入 • • • (i) TOHOKU UNIV.
(
α
κ
)
α
ln
exp
1 i ||=
−
⋅
=
⊥ αM
J
J
他の計測方法と比較し,比例定数 を決める較正実験が必要!!HITOP(High density TOhoku Plasma) Device
HITOP(High density TOhoku Plasma) Device
MPDA Magnetic Coils 2D Movable Probe TMP Plasma 0 1 2 3 Z(m)
Mach Probe Array Spectrometer Segm ented End-Plate Y X Z
X(m)
0 +0.4 -0.4 TMP Laval Nozzle CoilCylindrical chamber : length = 3.3 m, inner diameter = 0.8 m
Magnetic field : up to 0.1 T Plasma source : MPD Arcjet
Ion temperature : 20-40eV Electron temperature : 3-10 eV
Plasma density : ~1015cm-3(near the MPDA) Ionization degree : 50-90%
TOHOKU UNIV.
(b) With Externally-Applied Field
(Anode) (Cathode) j jz jr j Fθ=jrBz B z Bz (Anode)
Principle of Plasma Acceleration
Cross Section of MPDA
Anode Plasm a 0.1 (m ) Ga s Rese rvo ir Ga s Flo w Cathode φ 0 .0 3 (m ) φ 0 .0 1 (m )
Fast Acting Valve
Y
X Z
In sula to r
MPD(Magneto-Plasma-Dynamic) Arcjet
MPD(Magneto-Plasma-Dynamic) Arcjet
(a) Self-Field Acceleration
(A node) (Cathode) j jz jr j Bθ Fz=jrBθ F Fr=jzBθ (Anode)
The MPDA has a coaxial structure with a center tungsten rod cathode and an annular molybdenum anode.
By use of a fast-acting gas-puff valve, a quasi-steady ( ~1 msec ), high-density
(up to 1015cm-3 near the MPD outlet),
Hutchinsonモデルの検証実験
Hutchinsonモデルの検証実験
Ion saturation current distribution as a function of θ (cos θ)
Hutchinsonモデルの検証実験
Hutchinsonモデルの検証実験
TOHOKU UNIV. Hutchinsonモデルとの比較
(a) Mi = 1.3 (U =28km/s, Ti =8.6eV, Te =5.3eV)
(b) Mi = 0.8 (U =19km/s, Ti =11eV, Te =6.2eV)
実験条件
cosθ = -1 での値で 規格化 Good agreement !! <MPDプラズマ> ¾ rp < ρi (非磁化プラズマ) ¾1.5 < Ti / Te < 3 実験値と Hutchinson シミュレーション (Ti / Te = 2 の場合)を比較Mach Probe Calibration
~
分光計測による較正~
Mach Probe Calibration
~
分光計測による較正~
κ i || M J J = ⊥ = ⊥ α exp α | | Mi J J 亜音速流 ( Mi < 1 ) : 超音速流 ( Mi > 1 ) : =− κ ln 1 α
κ = 0.33
=
c i down upexp
M
M
J
J
perp - para タイプ up - down タイプM
c= 0.40
⊥ ⋅ = J J Mi κ ||
⋅
=
down up c iln
J
J
M
M
(Mi > 1) α α α ⋅ = ⊥ J J|| i ln 1(
α
=
−
ln
κ
)
∝
⊥ down up ||ln
J
J
J
J
α
∝
⊥J
J
J
J
|| down upln
ln
perp - para タイプ (κ = 0.33 (α = 1.1)) up - down タイプ M (Mi < 1)M
i> 1
M
i< 1
TOHOKU UNIV.up - down タイプ perp - para タイプ
Comparison between exp. and PIC simulation
0.4 < M
i< 1.5
DLPによる計測結果
PIC シミュレーションによる結果
0.4 < M
i< 1.5 の広い範囲で
良い一致を示した
Outline
Outline
1. Introduction
2. Electric Propulsion
3. Mach Probe Experiment in the HITOP Device
4. Magnetic Nozzle Acceleration of MPDA Plasma
5. Summary
Anomalous ion heating in the MPDA Plasma
Anomalous ion heating in the MPDA Plasma
2 5 0 1 0 2 0 3 0 H e I(ato m ) H e II(io n ) u z [km /s e c ] 0 5 1 0 1 5 2 0 F low E n e rgy [ e V ] 0 1 0 2 0 3 0 T [ e V ] 0 0 .2 0 .4 0 .6 0 .8 1 0 2 4 6 8 1 0 M D isch a rg e C urrent I [k A ] Id=8.6kA, dm/dt=0.06g/s(He), B0=1kG(uniform) at Z=4cm Ti increases steeply.
(
)
2 1 2 e e i B 2 i s i < + = = T T k u m C u M iγ
γ
¶ Why is the Mach number limited?
¶ What is the mechanism of the
conversion the input energy to the thermal energy?
Detailed measurement of ¾ j×B force field
¾ flow field
Spatial Distribution of jxB Force
Spatial Distribution of jxB Force
5 0 -5 0 1 0 2 0 3 0 -1 0 -5 0 5 10 Z [cm ] X [cm ] Y [c m ] Bθ:300[G ] 0 200 400 600 800 1000 0 50 100 150 B Z [G ] Z [cm] ∆B Z
net field strength
Id=7.2kA, Vd=200V,dm/dt=0.1g/s(He), B0=870G(uniform)
TOHOKU UNIV.
Magnetic Field distribution in the Vicinity of MPDA
deformation of magnetic field
Spontaneous formation a
helically-converging magnetic nozzle in the
Spatial Distribution of jxB Force
Spatial Distribution of jxB Force
X [cm] M PDA Bθ jr
Fz
+
Fz
-B r jθ θ θ
B
j
B
j
F
Z=
−
r+
rF
z-F
z+Drag force
Accelerating force Plasma Current DistributionSchematic of the drag force generation
Energy Balance in the MPDA Plasma Flow
Energy Balance in the MPDA Plasma Flow
P la sm a Y X Z M P D A sc an to s p ec trom e ter len s e Laval Nozzle Coil Generalized Bernoulli’s Equation Related
to the Applied-Field Acceleration**
(
)
const. 1 2 1 0 0 2 2 2 + − = − + + z Z Z u u B B B P u u ρ µ ρ µ ρ γ γ θ θ θ θ Flow energy ~108 Thermal energy ~109 Self-field energy ~107 Additional energy ~107 Mi<1 Mi>1 A M U TThermal > Flow. Ti / Te ~ 2. …. Why?
Flow energy Magnetic nozzle
Thermal energy
Momentum Conversion
Magnetic Laval Nozzle Formation
Magnetic Laval Nozzle Formation
0 10 20 30
Z[cm ]
M PDA Laval Nozzle Coil 0 2 4 6 8 I d [k A ] 0 1 2 0 0.5 1 1.5 2 I LN [k A ] Time [m s] (a) Discharge Current: Id (b) Nozzle Coil Current: ILN Quasi-Steady Discharge ~1ms Plasm a Y X Z M PDA s ca n
to spe ctro m ete r
le ns e 0 0.1 0.2 uniform Laval 0 10 20 30 B [ T ] Z [cm ] BLN Nozzle Throat
Characteristics in Magnetic Laval Nozzle
Characteristics in Magnetic Laval Nozzle
TOHOKU UNIV. Improvement of Acceleration Performance
Id = 7.2kA, dm/dt = 0.1g/s (He), Nozzle Throat at Z=17cm.
assuming γi = 5/3 The thermal energy is converted to the
flow energy by passing through the Laval
nozzle and a supersonic plasma flow is
Various Behavior in the Laval Nozzle
1-D Isentropic Flow Model
1-D Isentropic Flow Model
The MPD plasma flow is modeled
by
a one-dimensional adiabatic
flow with a constant entropy
at any
cross section along a flux tube.
U Nozzle Throat A