John T. Costello

1
VUV Photoabsorption Imaging

• John T. Costello
• National Centre for Plasma Science Technology
  (NCPST) and School of Physical Sciences, Dublin
  City University
• www.physics.dcu.ie/jtc john.costello_at_dcu.ie

QuAMP - Open University -September 8th 2003
2
Outline
1. Centre for Laser Plasma Research /NCPST 2.
VUV Photoabsorption/ionization Imaging
Principle 3. VPIF - VUV Photoabsorption Imaging
Facility 4. Charge State Selected Plasma Specie
Images 5. Time Resolved Column Density Maps
(Ba) 6. Conclusions and Current/Proposed
Applications
3
NCPST/CLPR Who are we ? What do we do ?
4
NCPST What is it ?

• 1. NCPST established with Government/Benefactor
  funding
• (Euro 8M) in 1999. Now EU Training Site.
• 2. Consortium of new and existing laboratories in
  plasma
• physics, chemistry and engineering
• 3. Fundamental and Applied Scientific Goals

5
The CLPR node comprises 5 (soon to be 6)
laboratories focussed on PLD photoabsorption
spectroscopy/ imaging
Staff John T. Costello, Eugene T. Kennedy,
Jean-Paul Mosnier and Paul van Kampen PDs John
Hirsch , D Kilbane (post Xmas) PGs Kevin
Kavanangh Adrian Murphy (JC), Jonathan Mullen
(PVK), Alan McKiernan Mark Stapleton (JPM),
Eoin OLeary Pat Yeates (ETK) MCFs Jaoine
Burghexta (Navarra) and Nely Paravanova (Sofia)
Vacancies PDRA-1 XUV FEL Experiments
(ETK) PDRA-2 Pulsed Laser Deposition
(JPM)PhD Dual Laser Plasma Experiments
(PVK/JC)
6
NCPST/ CLPR - What do we do ?
DCU Pico/Nanosecond Laser Plasma Light
Sources VUV, XUV (X-ray) Photoabsorption
Spectroscopy VUV Photoabsorpion Imaging VUV LIPS
for Analytical Purposes ICCD Imaging and
Spectroscopy of PLD Plumes Orsay/Berkeley
Synchrotrons Photoion and Photoelectron
Spectroscopy Hamburg - FEL Femtosecond IRXUV
Facility Development
7
Whats a Laser-Plasma ?
8
How do you make a laser plasma ?
Vacuum or Background Gas
Target
Plasma Assisted Chemistry
Laser Pulse- 1 J/ 10 ns
Lens
Spot Size 100 mm (typ. Diam.) F gt 1011
W.cm-2 Te 100 eV (106 K) Ne 1021
cm-3 Vexpansion ? 106 cm.s-1
Emitted - Atoms, Ions, Electrons, Clusters, IR -
X-ray Radiation
9
What does a Laser Plasma look like ?
10
Intense Laser Plasma Interaction
S Elizer, The Interaction of High Power Lasers
with Plasmas, IOP Series in Plasma Physics
(2002)
11
Part II - VUV Photoabsorption Imaging
POSTER P45 - Kevin Kavanagh
John Hirsch et al, Rev.Sci. Instrum. 74, 2992
(2003)
12
VUV Photoabsorption Imaging Principle
John Hirsch et al, J.Appl.Phys. 88, 4953 (2000)
VUV CCD
Sample
Io(x,y,l,Dt)
I(x,y,l,Dt)
Pass a collimated VUV beam through the plasma
sample and measure the spatial distribution of
the absorption.
13
Laser Plasma VUV/XUV Continua
P K Carroll et al., Opt.Lett 2, 72 (1978)
E T Kennedy et al., Opt.Eng 33, 3894 (1994)
14

• Motivations
• 1. To add to the DCU Laboratory a new diagnostic
  to work
• alongside the existing spectroscopic systems
• 2. Pulsed Laser Deposition (PLD) and Dual Laser
  Plasma (DLP)
• photoabsortion expeiments require increasingly
  detailed knowledge
• of the spatio-temporal characteristics of plasma
  plumes
• 3. Lots of photoionization cross sections due
  (Aarhus/ALS)
• Limitations of existing imaging methods
• 1. Direct imaging of light emitted by a plasma
  using gated array detectors
• (e.g., ICCD) provides information on excited
  species only
• 2. Probing plasma plumes using tuneable lasers
  provides information on non-
• emitting species but is limited to wavelengths gt
  200 nm or so

15

• Why a pulsed, tuneable and collimated beam ?
• Pulsed
• 1. Automatic time resolution the VUV pulse
  laser pulse duration (15 ns)
• 2. By varying the delay between the lasers the
  plasma can be probed at
• different times after its creation
• Tuneable
• 3. One can access all resonance lines of all
  atoms and moderately charged
• ions with resonances between 30 nm and 100 nm
• Collimated
• 4. Light path identical for all rays can derive
  the eqn of radiative transfer
• 5. The detector can be located far away from the
  sample plasma, reducing
• the sample plasma signal on the detector, and
  improving SNR

16
Q. Anything Else ? A. Yes, its a VUV beam
1. VUV light can probe the higher (electron)
density regimes not accessible in visible
absorption experiments 2. The refraction of the
VUV beam in a plasma is reduced compared to
visible light with deviation angles scaling as l2
3. The images analysis is not complicated by
interference patterns since the VUVcontiuum
source has a small coherence length (mms) 4. VUV
light can be used to photoionize atoms and ions -
this process simplifies greatly the equation of
radiative transfer (no bound states). 5.
Fluorescence to electron emission branching ratio
for many inner shell transitions can be 10-4 or
even smaller, almost all photons are converted
to electrons
17
VUV Photoabsorption Imaging Facility- V-P-I-F
18
The obligatory picture !!
19
Another one !
VUV Monochromator
Mirror Chambers
LPLS Chamber
Sample Plasma Chamber
VUV-CCD
20
VPIF - Design Considerations Measured
Characteristics
21
Final Design Parameters
22
VUV Photoabsorption Imaging Facility- Ray Tracing
with Light Path Simulation
Computed point spread distributions at entrance
slit for various apertures.
23
Ray Tracing with Light Path Simulation Beam
Footprints
Computed and measured VUV beam footprints (A)
0.5m (B) 1.0 mNOTE LOW DIVERGENCE !!
24
Spectral Resolution at 54 nm
Resolution
LPS
Wavelength (nm)
He, 1s - 2p line 50mm/50mm slits Rgt1000
Iint (Arb. Units)
Wavelength (nm)
25
Spatial Resolution (100mm/100mm slits l 50 nm)
Vertical Plane (150 mm)
Horizontal Plane (120 mm)
26
VPIF Specifications Time resolution 20 ns
(200 ps with new EKSPLA) Inter-plasma delay
range 0 - 10 ?sec Delay time jitter
1ns Monochromator Acton VM510 (f/12, f1.0
m) VUV photon energy range 10 - 35 eV VUV
bandwidth 0.025 eV _at_25 eV (50mm/50mm
slits) 0.05 nm _at_ 50 nm Detector Andor
BN-CCD, 1024 x 2048/13 ?m x 13 ?m
pixels Spatial resolution 120 ?m (H) x 150 ?m
(V)
27
VUV Photoabsorption Imaging Principle
VUV CCD
Sample
Io(x,y,l,Dt)
I(x,y,l,Dt)
Pass a collimated VUV beam through the plasma
sample and measure the spatial distribution of
the absorption.
28
What do we extract from I and Io images ?
Absorbance
Equivalent Width
dl
29
Equivalent Width (nm)
1 - exp-s(l)NL 1 -I/Io 1 -T
Io
Wl
l
30
Some Preliminary Results
Time resolved Wl maps of Ca plume species
Tune system to 3 unique resonances Ca 3p64s2
(1S) - 3p54s23d (1P) Ca 3p64s (2S) -
3p54s23d (2P) Ca2 3p6 (1S) - 3p53d (1P)
31
VUV Absorption Spectra of Ca Plasma Plumes
32
Maps of equivalent width of atomic calcium using
the 3p-3d resonance at 39.48 nm (31.4 eV)
33
Maps of equivalent width of singly ionized
calcium using the 3p-3d resonance at 37.34 nm
(33.2 eV)
34
Maps of equivalent width of doubly ionized
calcium using the 3p-3d resonance at 35.73 nm
(34.7 eV)
35
Plume Expansion Profile of Singly Charged Calcium
Ions
Plume COG Position (cm)
Delay (ns)
Ca plasma plume velocity experiment 1.1 x 106
cms-1 simulation 9 x 105 cms-1
Ba plasma plume velocity experiment 5.7 x 105
cms-1 simulation 5.4 x 105 cms-1
36
Extracting maps of column density,e.g.,Barium
We measure resonant photoionization, e.g.,
Ba(5p66s 2S)h? ? Ba(5p56s6d 2P) ? Ba2 (5p6
1S)e- h? 26.54 eV (46.7 nm) AND The
ABSOLUTE VUV photoionization cross-section for
Ba has been measured,Lyon et al., J.Phys.B 19,
4137 (1986) Ergo ! We should be able to extract
maps of column density - 'NL' ?n(l)dl
37
Maps of equivalent width of singly ionized Barium
using the 5p-6d resonance at 46.7 nm
38
Convert from WE to NL
Compute WE for a range of NL and fit a function
f(NL) to a plot of NL .vs. WE
Apply pixel by pixel
dl
dl
39
Result - Column Density NL Maps

• 100 ns
• 150 ns
• (C) 200 ns
• (D) 300 ns
• (E) 400 ns
• (F) 500 ns

40
Summary
VPIF - Provides pulsed, collimated and tuneable
VUV beam for probing dynamic and static
samples Spectral, spatial, divergence etc. all
in excellent agreement with ray
tracing Recorded time and space resolved maps of
equivalent width of Ca and Ba plasma
species Extracted time and space resolved maps
of column density for various time
delays Measured plume velocity profiles which
compare quite well with simple simulations based
on self similar expansion
41
Current Future Applications
Space Resolved Thin Film VUV Transmission and
Reflectance Spectroscopy - PVK Colliding-Plasma
Plume Imaging Combining ICCD Imaging/Spectroscop
y PI Photoion Spectroscopy of Ion Beams
? Non-Resonant Photoionization Imaging Lots of
new measurements from Aarhus ALS
42
Collaborators - VPIF
Univ. Padua Giorgio Nicolosi Luca Poletto
DCU John Hirsch Kevin Kavanagh Eugene Kennedy
Collaborators - Proof of Principle _at_ RAL
QUB Ciaran Lewis Andy McPhee R ORourke
RAL Graeme Hirst Waseem Shaikh
DCU John Hirsch et al
43
Ideally we would like a VUV/ XUV source with
lots of photons to do these experiments !!
44
And there is one in Germany ! (and coming to the
UK and US)
45
X-VUV FELs Femtosecond OPAs- The Ultimate
Photoionization Setup ?

• Tuneable NOW! 80 - 110 nm (20 - 60 nm in 2004)
• Ultrafast 100 fs pulse duration
• High PRF 1 - 10 bunch trains/sec with up to
  11315pulses/bunch
• Energy Up to 1 mJ/bunch
• Intense 100 mJ (single pulse) /100 fs /1 mm gt
  1017 W.cm-2
• Moving to XUV (2005) and X-ray (2010)
• Need a Linac insertion devices gt Fraction of a
  GigaEuro !!
• Project TitlePump-Probe with DESY-VUV-FEL
  (EU-RTD)
• Aim FEL OPA synchronisation with sub ps
  jitter
• URL http//tesla.desy.de/new_pages/TDR_CD/start.
  html
• Personnel MBI, DESY, CLPR-DCU, LURE, LLC, BESSY

46
Femtosecond X-VUV IR Pump-Probe
Facility,Hasylab, DESY
DESY, MBI, LURE, BESSY, LLC NCPST-DCU