Plasma Diagnostics of Ion Sources...Plasma Diagnostics of Ion Sources Ursel Fantz CERN Accelerator...
Transcript of Plasma Diagnostics of Ion Sources...Plasma Diagnostics of Ion Sources Ursel Fantz CERN Accelerator...
Plasma Diagnostics of Ion Sources Ursel Fantz
CERN Accelerator School, Senec, Slovakia 29th May – 8th June 2012
Diagnostics – The Window to the Knowledge
Langmuir probes: φpl, ne, Te
Absorption techniques: ns → Cs, Hˉ
Emission spectroscopy: ns, Ts → e, H, H2, Hˉ
Monitoring and Quantification – Spatial and Temporal Resolution
Universität Augsburg Max-Planck-Institut für Plasmaphysik
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 2
Preliminary considerations
What do I want to know?
identify the quantity (and the reason for it) define the required precision temporal behaviour and required time resolution necessity for spatial resolution …
Accessibility of the ion source?
diagnostic ports test stand or continuous operation risks and feasibility reliability …
Adequate diagnostic technique?
extensive or simple setup data acquisition and evaluation reliability costs and time (manpower) needed …
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 3
General plasma diagnostics techniques
Method Standard Sophisticated Extras Langmuir probe single probe
(cylindrical or planar) double or triple
probe emissive probe
special method: Boyd-Twiddy
emission spectroscopy
optical wavelength range with fibre optics & survey spectrometer
extended wavelength range
VUV, UV or IF
sophisticated system spectral resolution,
type of detector absorption
spectroscopy white light absorption
technique tuneable laser
absorption cavity-ringdown spectroscopy
Laser methods laser induced fluorescence
TDLAS
mass spectrometry
residual gas analyser energy resolved mass spectrometry
invasive – non-invasive ; active – passive ; basic – specific parameters
Introduction into techniques and applications for example in [1] – [7]
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 4
Example case: IPP ion source for negative hydrogen ions
ICP: f = 1 MHz, P = 70 kW, p = 0.3 Pa
∅ = 24 cmℓ = 15 cm 32×59×23 cm3
∅ = 24 cmℓ = 15 cm 32×59×23 cm3
source bodyat 50°C
grid at150°C
oven at> 120°C
6s plasma (4s beam), every 3 min: BATMAN cw, up to 1 hour, every 3 min: MANITU
cesium followed by H2 + e- → H2(v) + e- + hν
H2(v) + e- → Hˉ + H
H2(B,C)
dissociative attachment H2‾
surface process H, Hx
+ + surface e- → Hˉ
caesium layer low work function
H‾ formation and losses ...
volume process
destruction processes Hˉ + e- → H + 2e- Hˉ + H+ → H + H Hˉ + H → H + H + e-
... determine source optimisation
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 5
Example case: sources for negative hydrogen ions
Ion sources for negative hydrogen ions:
gas feed
Filter field
N
S
Driver ∅ 24cm
Expansion region
Extraction region
e Hx
+
H Hˉ
RF power 100 kW
p = 0.3 Pa
≈ 10 mT
Plasma generation ionising plasma ionisation: α ≈ 0.1 dissociation: δ ≈ 0.3 Te ≈ 10 eV, ne ≈ 5×1018 m-3
H‾ generation recombining plasma Te ≈ 1 eV, ne ≈ 5×1017 m-3
Hˉ/ne ≈ 0.1 – 5 Cs+/ne ≈ 0.01 – 0.1
Cs evaporation
H, Hx+ + surface e- → Hˉ
Cs evaporation → low work function
focus of diagnostics (and modelling)
on plasma close to the grid
Main issues Production and destruction of negative ions Extraction of negative ions Reduction of co-extracted electrons
ionising – recombining plasma
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 6
Langmuir probes: φpl, ne, Te, (EEDF)
Main principle
Recommended text books [1], [2], [5], [6] & publications and programs by F.F. Chen [8]
probe tip (tungsten wire)
insulator (ceramics)
voltage supply (-50 V to + 50 V)
plasma chamber
resistor
voltage measurement
stick a wire into the plasma tungsten, ∅≈ 100 µm, l = 1 cm
choose a reference electrode potential of plasma chamber
apply a variable voltage typ. from - 50 V to + 50 V
I – V characteristics
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 7
Langmuir probes: φpl, ne, Te, (EEDF)
Main principle
probe tip φprobe
plasma chamber
φfloating
φplasma
5 10 15 20 25-202468
10121416
transition
electronsaturation
prob
e cu
rrent
[mA]
probe voltage [V]
ion saturation
φfl
φpl
ne ni
Te
EEDF cylindrical probe
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 8
Langmuir probes: φpl, ne, Te, (EEDF)
Floating potential
5 10 15 20 25-202468
10121416
transition
electronsaturation
prob
e cu
rrent
[mA]
probe voltage [V]
ion saturation
φfl
same fluxes for ions and electrons Γions = Γelectrons
same currents
no probe current
Iprobe = 0 → φfl
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 9
5 10 15 20 25
-1
0
1
2
seco
nd d
eriva
tive
[mA
/ V2 ]
probe voltage [V]
φpl
Langmuir probes: φpl, ne, Te, (EEDF)
Plasma potential
5 10 15 20 25-202468
10121416
transition
electronsaturation
prob
e cu
rrent
[mA]
probe voltage [V]
ion saturation
φpl
determined by ambipolar diffusion turning point of I-V characteristics zero-crossing of second derivative
d2 I / d φpr2 → φpl
for curves with high noise level crossing of linear fits to electrons saturation current and transition close to turning point
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 10
5 10 15 20 25
-1
0
1
2
seco
nd d
eriva
tive
[mA
/ V2 ]
probe voltage [V]
EEPF
φpl
0 1 2 3 4 5 610-4
10-3
10-2
10-1
100
EEP
F [e
V-3/2
]
energy [eV]
slope → Te
Langmuir probes: φpl, ne, Te, (EEDF)
Electron energy distribution function (EEDF) and electron temperature
5 10 15 20 25-202468
10121416
transition
electronsaturation
prob
e cu
rrent
[mA]
probe voltage [V]
ion saturation
φpl
distinguish: EEDF = sqrt(E) × EEPF
Te
EEDF
e.g.: Maxwell function probability function
plot log(Ipr) versus E = φpl − φpr
Te also from potential difference φpl−φfl = kbTe × ln(sqrt(mi/(2πme))) ≈ 2-3 × Te
d(ln I ) / dE) = e / ( kbTe )
for both: normalisation
slope yields Te for Maxwell EEDF
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 11
Langmuir probes: φpl, ne, Te, (EEDF)
The electron density
electron saturation current
5 10 15 20 25-202468
10121416
transition
electronsaturation
prob
e cu
rrent
[mA]
probe voltage [V]
ion saturation
φpl
ne
effeesate AvenI41
, =
problem: effective probe area due to increase of plasma sheath
take current at plasma potential → Aeff = Aprobe
eB
e
prpr
ple Tk
melr
In
πφ
2)(
=
needs Te
Probe geometry influences shape of electron saturation current cylindrical probe (standard case) planar probe → Apr >> sheath spherical probe
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 12
Langmuir probes: φpl, ne, Te, (EEDF)
Ion density (positive ions)
ion saturation current
basically three theories available OML: Orbital-Motion-Limited ABR: Allen-Boyds-Reynolds BRL: Bernstein-Rabinowitz-Langmuir
all of them assuming collision-less plasma sheath, i.e. λ(ions) > r(sheath)
ni
5 10 15 20 25-202468
10121416
transition
electronsaturation
prob
e cu
rrent
[mA]
probe voltage [V]
ion saturation
simplest case: OML (rpr /rsheath < 3)
i
eBprii m
TkeAnI = needs Te
needs mi often unclear, e.g. hydrogen → H+, H2
+, H3+
choose proper φpr
guide line: φ = φpl − 10 × kTe
check if ne = ni is fulfilled
Isat, i at fixed φpr: useful as monitor signal
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 13
Langmuir probes: φpl, ne, Te, (EEDF)
Specific features – to keep in mind
invasive method→ probe size versus plasma volume
level of noise → EEPF for typically three orders of magnitude average, smoothing, filtering
RF field → oscillating φpl
measured curve ≠ real curve
magnetic field → gyro motion use Iion instead of Ie
negative ions → Ineg.ion instead of Ie for same mass Isat,ion = I sat,neg. ion → symmetric curves
prob
e cu
rren
t
probe voltage -20 -10 0 10 20 30
measured curve
real curve
oscillation
RF compensation active or passive
For monitoring or quantification with spatial and temporal resolution !
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 14
Langmuir probes: φpl, ne, Te, (EEDF)
Double probe
-40 -20 0 20 40
-2
-1
0
1
2
transition region
ion saturationprobe 1
Ii2
curre
nt [m
A]
voltage between the tips [V]
Ii1
ion saturationprobe 2
probe 1
probe 2
A1
A2
volta
ge
pla
sma
two probe tips of same size distance > 2 Debye length
voltage between the probes both probes are floating no reference potential needed
compatible with quartz or ceramic chamber and RF field
symmetric curve: ion saturation current
i
eB
satii
mTkeA
In ,=
2,,
,−+ +
= isatisatsati
III
Te from transition region
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 15
Example: Langmuir probes in ion source & Boyd-Twiddy setup
B. Crowley, S. Dietrich PSST 18 (2009) 014010
Comparison of EEDF
0 5 10 15 20 25 3010-3
10-2
10-1
Langmuir probewithout RF compensation
f (E)
[eV-3
/2]
Energy [eV]
Boyd-Twiddy
Vpl= 21.5 V
MaxwellTe =6.6eV
d=5cmH2, 80 kW, 0.36Pa
driver exit
plasma grid 10-3
10-2
10-1
100
0 5 10 15 20 25 30
H2,80kW, 0.36Pa
Te=1.2eVTe=1.4eV
Te=2.9eVTe=6.6eV
f(E) [
eV-3
/2]
Energy [eV]
Te=14eV
distan
ce [c
m]
05
914
19
Maxwellian EEDF
voltage ramp is superimposed by AC modulated signal measure frequency spectrum of probe current
Boyd-Twiddy method: direct measurement of EEDF
McNeely et al. PSST 18 (2009) 014011
Sophisticated Langmuir probe system for RF ion sources
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 16
All about photons: emission and absorption
Radiation of low temperature plasmas
He pink
Ne red
N2, air
orange
H2, H
purple
Colourful plasmas ! Neutrals atoms and molecules Ions single charged Electrons ne << nn
a + ef → a* + es → a + hν + es
collisions and spontaneous emission
electronically excited state
radiation
level q
level p
e impact excitation
Emission of light in the UV-VIS-IR
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 17
All about photons: emission and absorption
Emission versus absorption spectroscopy
photon energy E = hν wavelength λ = (Ep-Ek)/hc Einstein coefficients Apk, Bkp
Emission passive
Absorption active
excited state
λpk, Apk
q
p
impact excitation k
λpq, Apq
excited state
λpk, Bkp
q
p
k
radiation field Lλ λpq, Bqp
806 808 810 812 814 816
g [ ]
Wavelength [nm]
Inte
nsity
Wavelength [nm]
806 808 810 812 814 816
Inte
nsity
VIS: simple equipment information of upper level p
expensive equipment information of lower level q
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 18
Absorption techniques: nspecies → Cs, Ar, He, ...
Main principle of line absorption
light source detector optics
absorbing medium
white light tunable laser
vacuum or plasma spectrometer with CCD diode with filter
Non-invasive and line of sight integrated method !
=
),()0,(ln1)(
lII
lk
λλλ
[ ]lIlI )(exp)0,(),( λκλλ −=Absorption in a medium with path length l
)(λκabsorption coefficient statistical weights gi, gk
with Ladenburg relation π
λλλκ8
)(4
0 ik
k
ik
lineki
Acg
gnd =∫
∫
=
lineiki
kk d
lII
lAggcn λ
λλ
λπ
),()0,(ln18
40
Density of lower state ground state, metastables
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 19
Absorption techniques: nspecies → Cs, Ar, He, ...
Example: Cs line at 852.1 nm resonance line 6 2S1/2 – 6 2P3/2 with hyperfine structure
852.105 852.110 852.125 852.130 852.135
FF
34
44
54
2333
T = 1000 K
Inte
nsity
Wavelength [nm]
43
sum of Doppler-broadenedhyperfine lines
take line broadening into account
six hyperfine lines overlap to two peaks: ∆λ ≈ 21 pm
mTk
cB
FWHMD2ln8
,λλ =∆
Doppler profile
852.080 852.100 852.120 852.140
0.0
0.5
1.0
1.5
0.00
0.05
0.10
0.15
Sign
al w
hite
light
abs
orpt
ion
Sign
al L
aser
abs
orpt
ion
Wavelength [nm]
Laserabsorption
white lightabsorption
appearance depends on technique
apparatus profile (spectrometer) Doppler profile
white light absorption ↔ laser absorption
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 20
Absorption techniques: nspecies → Cs, Ar, He, ...
Example: Cs line at 852.1 nm white light absorption versus laser absorption
U. Fantz, C. Wimmer J. Phys. D 44 (2011) 335202
Improved detection limit for laser absorption: factor > 40 ! sensitivity range: 3×1013 m-3 ̵ 1017 m-3 (path length = 15 cm)
being perfectly in the range required for the ion sources
851.6 851.8 852.0 852.2-0.02
0.00
0.02
0.04
nCs = 1.4 x 1015 m-3
Abso
rptio
n [a
. u.]
Wavelength [nm]
white light absorption
0.0 0.1 0.2 0.3 0.4
0.000
0.005
0.010
0.015
0.020
Abso
rptio
n [a
. u.]
Rel. wavelength [a. u.]
laser absorption
nCs = 3.5 x 1013 m-3
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 21
Absorption techniques: nspecies → Cs, Ar, He, ...
Straightforward analysis
∫
=
lineiki
kk d
lII
lAggcn λ
λλ
λπ
),()0,(ln18
40
... but ....
Line saturation
strong absorption: nk × l correction factors by profile fitting
... and .... 0
2
4
measured signal
Gaussian fit
heavy absorption
mediumabsorption
Abso
rptio
n [a
. u.]
Wavelength
lowabsorption
0
1
(b)
Rel.
inte
nsity
(a)
Depopulation effect strong intensity attenuation of laser to ≈1% → trade-off with temporal resolution
in vacuum with plasma → subtract emission
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 22
Laser absorption for Cs in ion sources
Tuneable diode laser – Fibre optics – Photo diode with interference filter
ND-filter 2.0 interference filter 852 nm
lens f = 25.4 mm
beamdiameter11.5 mm
diodelaser
fiber105 µm
NA = 0.22
photodiode
fiber200 µm
NA = 0.22
lens f = 25.4 mm
ND-filter 2.0 interference filter 852 nm
lens f = 25.4 mm
beamdiameter11.5 mm
diodelaser
fiber105 µm
NA = 0.22
photodiode
fiber200 µm
NA = 0.22
lens f = 25.4 mm
Simple and robust setup for application to ion sources !
single mode laser, 150 mW λ = 851-853 nm with ∆λFWHM = 0.01 pm
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 23
Example: laser absorption for on-line monitoring of Cs dynamics
On-line monitoring: vacuum – plasma (6s with 4s beam) - vacuum
5450 5455 5460 54650
1
2
3
4
0
1
2
3
4
5
RF off
Cs e
miss
ion
[a. u
.]
Cs emission
Cs-d
ensit
y [1
015 m
-3]
Time [s]
RF on
Cs densitylaser absorption
BATMAN #68562
3 s before pulse: vacuum 6 s plasma with 4 s beam: plasma 90% of Cs is ionized
30 s after pulse: Cs peak & decay time
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 24
Absorption techniques: nspecies → H‾, Cs, Ar, H3+...
Cavity – Ringdown – Absorption – Spectroscopy → CRDS
high reflectivity mirrors R>99.9%
dcavity detector
transmitted signal
plasma with length l
laser
pumping of an optical cavity by a (tunable) laser source measurement of signal decay after laser source is switched off
Measurement of laser light attenuation trapped in a high-finesse optical cavity transfers absorption signal from wavelength into time dependence
0t
0 e(t) τ−⋅= II
)1(0 Rcd−
=τ; )0
1'
1(ld
cσ1n
ττ−⋅⋅
⋅=
additional absorption ττ ′→0empty cavity with decay time τ0
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 25
Absorption techniques: nspecies → H‾, Cs, Ar, H3+...
Cavity – Ringdown – Absorption – Spectroscopy → CRDS
0 5 10
0.0
0.2
0.4
0.6
0.8
1.0
limited by time resolution
increasedensity
τ0 (reflectivity of mirrors)
norm
alize
d in
tens
ity
Time [µs]
empty cavity
)0
1'
1(ld
cσ1n
ττ−⋅⋅
⋅=
Hˉ density: line of sight averaged detection limit ≈ 1015 m-3
Hˉ = 5×1017 m-3 → t = 8 µs 0 50 100 150 200
0.2
0.4
0.6
0.8
1.0
plasma
τ = 47 µs
τ0 = 62 µssig
nal [
a.u.
]
µs
empty cavity (vacuum)
Example H‾ cross section: photodetachment
0.0 0.5 1.0 1.50
1
2
3
4
cros
sec
tion
σ [1
0-21 m
2 ]
wavelength [µm]
H- + hν → H + e
Nd-YAG laser
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 26
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
The main principle
spectrometer detector optics
plasma
Non-invasive and line of sight integrated method !
UV and VUV: vacuum system IF: detector
πλε
4/)( hcApn pkpk =
Measures density of excited state ...
... which depends on plasma parameters !
Recommended text books [1], [4], [7], [9], [10]
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 27
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
What information can be obtained from the line emission ?
499 500 5010.0
0.2
0.4
0.6
0.8
1.0
Inte
nsity
[a. u
.]
Wavelength [nm]
Imax n(p) Apk Intensity: plasma parameters
density and temperature of neutrals, ions, electrons
insight in plasma processes
∆λ Tgas Doppler broadening: particle temperature
Line profile: broadening mechanism
λ0 Element
Wavelength: species
Wavelength shift: particle velocity
⇒=∫ 1λλ dPline
λλ εε Ppk=
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 28
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
selection rules
Radiation
. . .
.
n(k)
Ion
n(p)
. . .
.
Energy
. . .
pkA
1,0 ±=∆L
0=∆S
Atoms
1,0 ±=∆J
Appearance depends on spectral resolution !
587 588 589 590 591 5920.00.20.40.60.81.01.2
3 2P1/2 - 3 2S1/2
Inte
nsity
[a.u
.]
Wavelength [nm]
Na D-line
3 2P3/2- 3 2S1/2
366 368 370 372 374 376 378 380 3820.0
0.2
0.4
0.6
0.8
1.0
3-52-4
1-3
Inte
nsity
[a.u
.]
Wavelength [nm]
Nitrogen N2C 3Πu - B
3Πgv'-v''
0-2
≈
SLS Ln +
+12 . . .
.
. . . .
.
.
.
.
n(p), v’, J’
n(k), v’’, J’’
Ion
Radiation '''''' JJvvpkA
Molecules
0=∆Σgu ↔
1,0''' ±=∆=− JJJP, Q, R branch
≈
Σ+Λ+ Λ12S −+,
,ug
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 29
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
Spectroscopic system
Optics Spectrometer Detector
photomultiplier λ scan ∆λ, ∆t diode array λ range CCD, ICCD pixel size - ∆λ intensity
focus length spectral resolution ∆λ grating spectral resolution Blaze - intensity slits entrance slit ∆λ exit slit - detector
fibre very flexible VIS: glass, quartz, UV enhanced lens and aperture Imaging optics solid angle
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 30
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
The spectroscopic system is determined by the purpose !
survey spectrometer pocket size ∆λ ≈ 1-2 nm 1m spectrometer good optics ∆λ ≈ 20 pm Echelle spectrometer high resolution ∆λ ≈ 1-2 pm
time resolution detector spatial resolution detector, lines of sight intensity detector, spec., optics spectral resolution detector, spec., optics
line profile line shift
line monitoring very simple ∆t, poor ∆λ
less information
relative intensities common technique poor ∆t, ∆λ, ∆x, flexible
moderate information
absolute intensities expensive technique poor ∆t, ∆λ, ∆x, flexible
powerful tool
Inte
nsity
Wavelength]
∆λImax
Inte
nsity
TimeLOS
1
4 y
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 31
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
Calibrated spectrum spectral radiance [W/m2/sr/nm]
Measurement intensity [counts/s]
Conversion factor spectral sensitivity
exposure time
=×
(counts/s)nmsm
photons4(counts/s)nmsrmW
22 hcλπ
400 500 600 700 8001
10
100 x10-6
Wavelength [nm]
0.00.10.20.30.4
x10602468
10
Ulbricht sphere
Calibration of the spectroscopic system
Wavelength: pixel → nm spectral lamps, plasma, λ tables
Radiance - Intensity counts → W/m2/sr, ph/m2/s
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 32
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
From intensity to plasma parameter pkpk ApnI )(=
Population models Boltzmann distribution (TE, LTE)
Collisional radiative model Corona equilibrium
high ne
low ne
relevance of photons
n(1)
. . .
.
n(k)
Ion
n(p)
. . .
.
≈
Energy α β S
A X
Rate equations for excitation and de-excitation processes electron impact excitation and de-excitation absorption and emission, heavy particle collisions, ....
CR model ,...),(0 eeeffpkepk nTXnnI =
0)()(d
)(d11 =−= ∑k pkepe ApnTXnn
tpn
∑== k pkpkeppkepkepk AATXXwithTXnnI /)()( 10
Corona model rate coefficient
emission rate coefficient 01 nn ≅
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 33
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
Electron temperature from absolute line emission
n0, ne known e
pkee
effpk nn
InTX
0,...),( =
,...),(0 eeeffpkepk nTXnnI =
1 2 3 4 5 6 7 810-19
10-18
10-17
10-16
+ 1 eV- 0.5 eV
+- 0.1 eVArI 750 nm
Emiss
ion
rate
coe
fficie
nt [m
3 s-1
]
Electron temperature [eV]
Find suitable gases and diagnostic lines
admixture of small amount of diagnostic gas
prominent example: Ar
Very sensitive for low Te !
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 34
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
Electron temperature from line ratio (relative calibration)
)()(
22
11
2
1
epke
epke
pk
pk
TXnnTXnn
II
= → ratio of rate coefficients for known densities
1 2 3 4 5 6 7 8 9 1010
100
1000
Ar750/He728
Ratio
of r
ate
coef
ficie
nts
Electron temperature [eV]
Find suitable gases and diagnostic lines
n1, n2 inert gases or n1 = n2
Ιpk undisturbed lines
ground state excitation
Xpk ratio depends on Te
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 35
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
Actinometry: density ratio from line ratio (relative calibration)
)()(
22
11
2
1
epke
epke
pk
pk
TXnnTXnn
II
= → ratio of densities for known rate coefficients
density ratio (nH/nH2)
density ratio (nAr/nH2)
density ratio (nHe/nH2) dependence on Te (factor of 10) →needs iteration
)()(
2
2
2
434434
eHFuleH
eH
eHHFul
H
TXnnTXnn
II
= independent on ne
)()(
2
2
2
750750
eHFuleH
eAr
eArHFul
Ar
TXnnTXnn
II
=
1 2 3 4 5 6 78910 20 300.01
0.1
1
10Ar750 / H2Ful
Hγ / H2Ful
Hγ / Ar750
Ratio
of r
ate
coef
ficie
nts
Electron temperature [eV]
Te dependence negligible
U. Fantz et al., Nucl. Fusion 49 (2009) 125007
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 36
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
Particle density from absolute line emission
ne, Te known ,...),(0
eeeffpke
pk
nTXnI
n =
,...),(0 eeeffpkepk nTXnnI =
Knowledge of dominant excitation mechanism is essential !
Example: Cs and Cs+ lines
)(852852 eCs
eCsCs TXnnI =
needs ne, almost independent of Te
)(460460 eCs
eCsCs TXnnI
+
+
+
=
needs ne, strong dependence on Te
Cs:
Cs+:
1 2 3 4 5 6 7 8 9 1010-15
10-14
10-13
10-12
Cs+: 460 nmBorn approximation
Cs: 852 nm
Rate
coe
fficie
nt [m
3 /s]
Electron temperature [eV]
×1000
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 37
Example: emission spectroscopy at the ion source
Survey spectrometer and on-line monitoring
300 400 500 600 700 800 9000
1000
2000
3000
H2
Hβ Hα
Cs
Hγ
Inte
nsity
[cou
nts]
Wavelength [nm]
Cu, OH, N2 O Cs+
impurities Cs, Cs+ H from Hγ ne from Hβ/Hγ
ne, Te diagnostic gas Ar Hˉ from Hα/Hβ
0 1 2 3 4 5 60
1000
2000
3000
#19776
#19776#19788
#19800
Cs 852
Hδ
#19800
#19788Inte
nsity
[cou
nts]
Time [s]0 2 4 6 8 10 12
0
10
20
30
40
50
0
2
4
6
off
Time [s]
Cs852
HγLi
ne in
tens
ity [c
nm
/ms] #64623
Uex on Dio
de s
igna
l [V]
Cs diode
820 840 860 880
Intensity
Cs852
Transmission Cs filter
diode & filter
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 38
Example: emission spectroscopy: ns→ H‾
A novel diagnostic technique for Hˉ volume density U. Fantz, D. Wünderlich
NJP 8 (2006) 301
H+ + Hˉ → H(n=3) + H
mutual neutralisation
H(n)
H
H+ H2+
H2 H-
recombination dissociative recombination
dissociative excitation
effective excitation
H2+/H < 10-3
H/H2 > 0.1
Te > 1 eV
Collisional radiative model Hα
Population mechanisms for H
10-5 10-4 10-3 10-2 10-1 11
10
100
5x10181018
5x10171017
ne [m-3]
5x101810185x10171017
Te = 3 eV
Hβ/Hγ
Hα/Hβ
Line
ratio
Density ratio H-/H
Line ratios depend on ne, Te and Hˉ/H
Measurement of Balmer line ratios Hα/Hβ depends on Hˉ/H Hβ/Hγ reflects ne and Te
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 39
Example: emission spectroscopy: ns→ H‾
A novel diagnostic technique for Hˉ volume density U. Fantz, D. Wünderlich
NJP 8 (2006) 301
H+ + Hˉ → H(n=3) + H
mutual neutralisation
H(n)
H
H+ H2+
H2 H-
recombination dissociative recombination
dissociative excitation
effective excitation
H2+/H < 10-3
H/H2 > 0.1
Te > 1 eV
Collisional radiative model Hα
Population mechanisms for H
high Hα/Hβ ratio: Hˉ = 1×1017 m-3 stable Hβ/Hγ ratio, i.e. stable ne and Te
-1 0 1 2 3 4 50
2
4
6
8
10
12
14
H-
Hα/Hβ Hγ
Lin
e ra
tio, l
ine
inte
nsity
[10
19 p
h/m
3 /s]
Hβ/Hγ
Hα/Hβ
Time [s]
extraction
H2, 80kW, 0.65Pa
Measurement of Balmer line ratios Hα/Hβ depends on Hˉ/H Hβ/Hγ reflects ne and Te
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 40
Emission spectroscopy: ns, Ts→ e, H, H2, H‾
Powerful diagnostic tool
?
in-situ, non-invasive
simple equipment
line of sight averaged results Analysis
simple quite complex supported by
collisional radiative models
based on atomic and molecular physics
identification of species particle densities particle temperatures on-line monitoring insight in plasma processes spatial resolution by several lines of sight
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 41
Summary
The three W’s What do I want to know ? → and why? What is the adequate technique ? → effort versus gain! What is the accessibility of the source ? → feasibility !
Plasma Diagnostics of Ion Sources
Diagnostics – The Window to the Knowledge !
The three examples Langmuir probes → φpl, ne, Te, (EEDF) Absorption techniques → nspecies → Cs, H‾ Emission spectroscopy→ ns, Ts→ e, H, H2, H‾
The three “keep-in-mind’s” Monitoring versus quantification → trends or full information Spatial resolution → averaged or x-resolved (step width!)) Temporal resolution → averaged or t-resolved (time scale!)
CAS on Ion Sources, 6th June 2012 Ursel Fantz, p. 42
References
[1] F. F. Chen, J.P Chang, Lecture Notes on Principles of Plasma Processing (Kluwer/Plenum, 2003)
[2] M. Lieberman, A. Lichtenberg, Principles of Plasma Discharges and Materials Processing (Wiley,1994)
[3] B. Chapman, Glow Discharge Processes (Wiley, 1986)
[4] R. Hippler, S. Pfau, Low Temperature Plasma Physics (Wiley, 2001)
[5] D. Flamm, O. Auciello: Plasma Diagnostics, Volume 1 (Academic Press, Inc., 1989)
[6] I.H. Hutchinson: Principles of Plasma Diagnostics (Cambridge University Press, 1987)
[7] A. Thorne, Spectrophysics: Principles and Applications (Springer 1999)
[8] http://www.ee.ucla.edu/~ffchen/
[9] U. Fantz: Basics of plasma spectroscopy Plasma Sources Sci. Technol. 15 (2006), 137
[10] U. Fantz: Emission Spectroscopy of Molecular Low Pressure Plasmas Contrib. Plasma Phys. 44 (2004) 508