SOFT X-RAY EMISSION FROM GAS PUFF
IMPLOSIONS
P. Burkhalter, G. Mehlman, F. Young, S. Stephanakis, V. Scherrer, D.
Newman
To cite this version:
P. Burkhalter, G. Mehlman, F. Young, S. Stephanakis, V. Scherrer, et al.. SOFT X-RAY
EMISSION FROM GAS PUFF IMPLOSIONS. Journal de Physique Colloques, 1986, 47 (C6),
pp.C6-247-C6-252. <10.1051/jphyscol:1986631>. <jpa-00225874>
HAL Id: jpa-00225874
https://hal.archives-ouvertes.fr/jpa-00225874
Submitted on 1 Jan 1986
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
JOURNAL DE PHYSIQUE,
C o l l o q u e C6, suppl6ment
au no 10, T o m e 47, octobre 1986
SOFT X-RAY EMISSION FROM GAS PUFF IMPLOSIONS
P.G. BURKHALTER, G. MEHLMAN*, F.C. Y O U N G , S. J. STEPHANAKIS,
V.E. S C H E R R E R and D.A. NEWMAN*
Naval Research Laboratory, Washington, DC 20375-5000, U.S.A.
' ~ a c h s / ~ r e e m aAssociates,
n
Inc., Landover, MD 20785, U.S.A.
Resume - Un generateur d'impulsions de 2TW produit, B partir d'une bouffee de gaz,
une colome de plasma de densite BlevBe. On a etudie ces implosions de bouffees de
neon et d'argon du point de vue de la realisation d'un milieu laser. L'appareillage
utilise pour les diagnostics permet l'observation radiale du rayonnement pour estlmer
l'uniformite et le degre d'ionisation des plasmas. L'interpretation des donnees
spectroscopiques obtenues sur le neon introduit sous forme de trace dans les plasmas
d'argon. indique une temp6rature trop &levee (200-300 eV) pour une production
efficace d'ions Ar IX (isoelectronique du neon). Le rayonnement des plasmas de neon produits B des pressions s6lectionnees fournit 2 & 3 kJ dans les transitions de
1'Blectron Is, selon une repartition uniforme dans l'espace interelectrode de 4 cm.
On discute des efforts d'optimisation de cette source. On a obtenu, avec une decharge
dans un tube capillaire associee au g6n6rateur d'impulsions, une emission de 30 GGI
pour la raie 1s-2p de l'ion Na X, B 11 A .
-
Abstract
Dense plasma columns are produced in a 2-TW pulsed-power generator
retrofitted with a gas-puff valve. Z-pinch implosions of neon and argon gas puffs are
studied as potential x-ray lasing media. X-ray diagnostics record radially the
emitted radiation to evaluate the plasma uniformity and the degree of ionization.
/
Interpretation of spectra from argon implosions containing neon tracer indicates
plasma temperatures too high (200-300 eV) for efficient production of Ne-like ArIX
transitions. Neon plasmas at selected gas pressures emit 2-3 ~ JofI K-shell line
radiation with good uniformity along the 4-cm interelectrode gap. Erforts towards
source optimization are discussed. Replacing the gas puff with a capillary discharge
source for sodium implosion experiments produced -30 GW of the 11 A He-a line from Na X
I - INTRODUCTION
High current pulsed discharges of
exploded wires and gas puff columns,
provide intense sources of keV x-ray and
subkeV (XW) radiation.
Gas-puff
implosions of neon driven by the 2-TW
pulsed power generator (Gamble 11) at
the Naval Research Laboratory have been
studied.l2 Annular gas columns of
argon, neon, or mixtures are imploded
with 1.0-1.5 MA driving currents to
produce 4-cm long 2-pinch implosions in
the interelectrode gap. In this work,
x-ray intensities, plasma parameters,
uniformity of implosion, and the degree
of ionization are evaluated with
diagnostics which provide temporal,
spatial, and spectral measurements.
This effort was directed towards
producing a suitable XW laser gain
medium by studying the plasma implosions
obtained with Gamble I1 as the load
parameters were varied. Two schemes for
population inversion (in XW laser
scenarios) are potentially achievable:
excitation of neon-like argon for lasing
at long wavelength (3p-3s at 460 A) ,l3
and coincidence photopumping of Ne IX by
the 11 A line from NaX for lasing at 230
A (n=4 to 3 transitions) .l4
The Gamble I1 generator is configured
with a gas puff valve and a supersonic
nozzle, as shown in Fig. 1 in order to
inject a hollow annular gas column
across the 4-cm interelectrode gap. The
2.5 cm diameter gas puff nozzle is
located on the machine axis and is
surrounded by the return-current posts
on the 7 cm diameter circle. Wire
stretched across these posts serves as
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1986631
JOURNAL DE PHYSIQUE
C6-248
the anode. The gas is preionized with
two W flashboards prior to current
discharge. Normally, the driving
current has a 70 ns risetime; however
the risetime can be reduced to 20 ns
with the use of a lasma erosion opening
The PEOS consists
switch (PEOS) .l5
of a low density plasma injected across
the vacuum gap of the electrode assembly
as indicated in Fig. 1. The use of a
PEOS to eliminate prepulse and to
sharpen the current risetime has led to
improved plasma uniformity in argon and
neon implosions.
3
-Fig. 2. Photograph viewing the plasma
forming region (nozzle) and the diagnostic instrumentation on Gamble 11.
Fig. 1. Schematic drawing of the
pulsed-power load assembly for
generating imploding plasma.
Diagnostics are mounted around the
vacuum chamber to view radially the
imploded plasma. The location of the
following diagnostics is shown in a
photograph of the front end of Gamble I1
(Fig. 2). The diagnostics include: 1)
filtered x-ray pinhole cameras for time
integrated images of the implosion, 2)
filtered x-ray diodes (XRD's) for
intensity and time history measurements,
3) convex-curved crystal spectrographs
to collect x-ray spectra in the 4-14 A
region, and 4) a l-m grazing incidence
spectrograph to record 500-40 eV (20-300
A) XUV spectra. A fast closure valve is
used as a debris shield for the grazing
incidence spectrograph.
Filtered XRD's were used to record X-ra
emission in selected spectral regions.
For argon implosions an Al-cathode XRD
filtered with 0.7-pm Ti was used to
measure L-shell emission from 0.25 to
0.46 keV (Fig. 3). This filter was also
used on a pinhole camera with Kodak 10105 film to record L-shell images of the
imploded plasma. For neon implosions, a
Ni-cathode XRD with a 1.0-pm Cu filter
was used to measure the neon He-a line.
The crystal spectrographs employed
curved mica with good reflectivities in
the 4-17 A region from the 002 planes
and curved KAP that has strong
reflections from the 001 planes and also
from the 013 and 014 diffraction
planes.l8 The different diffraction
planes in mica and KAP gave a wide
recording sensitivity for soft x-ray
I
I
ARGON
13
01
0
I
20
I
I
40
60
PLENUM PRESSURE
I
(PSI
80
A)
I
100
Fig. 3. The argon L-shell emission as a
function of filling pressure.
emission. This emission was recorded
through a 10-pm-thickBe window and a
0.25 mm spatial slit positioned
perpendicular to the axis of the plasma
column. Spectra were recorded on Kodak
direct exposure film (DEF) for which the
film response is known.l9 Integral
reflection coefficients of the crvstal
were computed from a model for the
curved KAP crystal2' and taken from
published values for mica.21 Spectral
traces were digitized with a scanning
microdensitometer and processed by
computer to yield x-ray intensities.
XW spectra collected with the grazing
incidence spectrograph were recorded on
uncalibrated Kodak type 101-01 film.
The scans are presented as film density
traces. To aid in the identification of
spectral lines, atomic structure
calculations were performed with a code
from the Los Alamos National
Laboratory.22 Besides, argon and neon
transitions have been observed and
classified for example in the early work
using a plasma
by Peacock et
focus device and more recently by a
group at the Imperial College in London
using a gas puff device. (A.E. Dangor,
private communication)
I11
-
RESULTS
A. Argon Plasma
Pinhole images of the argon L-shell
plasma emission shown in Fig. 4 were
collected with and without the PEOS near
the optimum gas filling pressure of 55
psiA (see Fig. 3). Images recorded
without the switch (Fig. 4-upper) show
considerable flaring and large emission
diameters (>l cm) along the interelectrode gap. An axial channel of
intense emission, about 1 mm in diameter
is observable in the upper image of Fig.
4. Discharges with the PEOS produce
narrow, more uniform images of Ar Lshell emission (Fig. 4 bottom image).
An absence of Ar K-shell emission was
noted in shots with the PEOS operated
with gas loadings near or above the
optimum pressure for Ar L-shell
emission. The observable plasma length
for argon implosions both with and
without the PEOS was about 3.5 cm. The
Ar K-shell emission, which is maximum
for a gas filling pressure of about 15
psiA, is an order of magnitude less
intense chan the maximum L-shell
emission.
Fig- 4. Pinhole images of argon ~lasma.
The grazing incidence spectrograph was
used to record the argon L-shell spectra
in order to identify the stages of
ionization formed in argon implosions.
Spectra were recorded in the 25-300 A
wavelength region. The spectral portion
between 25 and 50 A (250-460 eV)
corresponds to the range recorded by the
Ti-filtered XRD (An=l transitions).
Detailed identifications revealed arrays
of 2p-3s, 2p-3d transitions in Ne-like
ArIX through N-like Ar XI1 as indicated
in Fig. 5. At longer wavelengths, (Fig.
6) intense 2s-2p transitions from Ar X
and Ar XI are observed. A few weak
lines believed to be Na-like ArVIII
exist in this wavelength range. These
lines were observed with both machine
conditions but were slightly enhanced in
shots with the PEOS. The existence of
such ionization stages indicates that
the argon implosions were well beyond
the desired 50-70 eV plasma temperature
for efficient excitation of Ne-like
argon (J. P. Apruzese, P. C. Kepple, and
J. Davis, private communication).
A technique which used neon as a tracer
in argon was developed to determine the
temperature of imploded argon plasmas.
Spectra of argon implosions with 5% and
1% neon tracers were collected with a
curved KAP crystal spectrograph. Figure
7 shows the argon K lines diffracted by
the 001, 002, 013, and 014 planes in
KAP. The plasma-to-spectrograph
distances were 150 and 100 cm for the 5%
Fig. 5. XW argon spectrum showing
2p-nl transitions.
JOURNAL DE PHYSIQUE
C6-25 0
Ar XI
1
2s-2p TRANSITIONS
Ar X
I
Fig. 6. X W argon spectrum showing 2s-2p
transitions.
and 1% neon tracers respectively. Under
these conditions the argon K lines are
overexposed in first order but readable
in the higher order planes (particularly
the 002 planes). Spectral line
densities were recorded in first order
between 10 and 14 A. The neon K-line
intensities from Ne X and Ne IX were
measured to interpret plasma
temperatures from the a to p line ratio.
The lower spectrum in Fig. 7 recorded
with the spectrograph positioned on the
vacuum chamber housing as seen in Fig. 2
(at a distance of about 50 cm)
registered, in the 9-12 A region, copper
L-lines from the brass nozzle (cathode).
As the gas filling pressure was varied
between 15 and 60 psiA, the neon La/Hea
lfne intensity ratio was found to vary
from 0.26 to 1.1. The imploded argon
plasma temperatures were estimated
between 200 and to 350 eV. These
temperatures are significantly higher
than needed for Ne-like Ar production
but are consistent with the ionization
stages identified in the grazing
incidence spectrum.
A narrow space-resolving slit was not
used to collect these spectra in order
to record readable neon a and P lines.
B. Neon Plasma
Spatially-resolved spectra were
collected to study and measure the x-ray
line emission in neon gas implosions.
Time-integrated spectra were recorded on
DEF film using both curved KAP and
curved mica diffraction crystal
spectrographs and the derived spectral
intensities were used to determine
plasma conditions. Spectra were
acquired for machine currents of 1.0 1.5 MA with and without a PEOS using the
largest diameter (2.5 cm) nozzle. Data
were also acquired for neon plasma
implosions generated using smaller
nozzles and no switch. Pinhole images
record a plasma column of 3.5-4 cm
length.
Neon spectra were processed to obtain
integrated line intensities. Figure 8
shows the neon plasma spectral intensity
for two typical shots. The integrated
line intensities obtained for both shots
are comparable within a factor of two
and the total yields are similar except
that more continuum emission
(bremsstrahlung) occurs in the shots
without the PEOS; also the spectral
lines are noticeably narrower when the
850
Fig. 7 . Spectrograms from argon with
neon tracer plasmas (curved KAP crystal).
1000
1150
ENERGY (ev)
1300
Fig. 8. Intensity trace from neon
(upper: w/o PEOS, peak current 1.2 MA,
lower: with PEOS, peak current 0.83 MA)
PEOS is used. The most pronounced
feature of neon plasma emission, for
both shots, is that most of the line
radiation, i.e., at least 90%, is
contained in both a lines. Neon
implosions yield high radiation emission
and, under the appropriate machine
conditions, uniform plasmas. In shots
without the PEOS, we measured energies
of 2 to 3 kJ due to K-shell line
radiation into 4s from the whole
neon plasma column. The total radiated
energy was 4-5 W. The use of the PEOS
significantly reduces the plasma column
diameter and improves the plasma
uniformity. The factor of two intensity
difference between the traces in Fig. 8
is related to the difference between
peak currents; the latter was about 30%
less in the shot with the PEOS.
Improved implosion stability and plasma
uniformity is observed in neon gas
implosions generated with the PEOS or
with a smaller diameter nozzle. This
insight was gained from the degree of
plasma flaring or pinching in pinhole
images and from spatial variations of a
line intensities. With the largest
diameter (2.5 cm) nozzle and no switch
the La/Hea line ratio decreased from
around 1.0-1.3 near the cathode to
nearly unity at midgap, and to about 0 . 3
near the anode. With the PEOS, this
ratio was about 1.1 with a uniformity of
-+ 30% across the entire interelectrode
gap. Without the PEOS, but with a
smaller diameter nozzle (1.75 cm), the
plasma uniformity improved. In this
case the a-line ratio varied from 1.0 to
1.3 near the cathode to a value of 0.70.8 near the anode. As shown in pinhole
images (Fig. 9) the plasma flaring is
reduced with the smaller nozzle.
C. Sodium Plasma
A capillary discharge source has been
developed to produce a plasma from solid
dielectric materials such as sodium
fluoride. A small capacitor bank
generates a discharge along the interior
wall of a NaF capillary tube and the
generated plasma is injected axially,
through a 2 cm diameter nozzle, into the
4 cm electrode gap of Gamble 11. (1.2
MA current) This plasma implodes and
emits x-rays about 100 nsec after the
beginning of the current pulse. This xray emission was recorded with pinhole
cameras, x-ray diodes, and crystal
spectrographs. The pulse duration,
(FWHM), as recorded by an aluminum
filtered XRD detector is 20 nsec for
this K-shell radiation. Pinhole images
indicate that the imploded plasma fills
the 4 cm interelectrode gap.
The NaF spectrum was collected with
moderate resolution using the mica
crystal spectrograph with spatial
resolution. Figure 10 shows the sodium
K-shell radiation (above 1 keV) from a
typical NaF spectrum obtained with this
instrument. The spectra were processed
to obtain the sodium line intensities
from which plasma conditions and
uniformity could be estimated. The
fluorine emission below about 900 eV was
absorbed strongly by the 10 pm Be window
on the spectrograph and only F IX higher
Rydberg transitions were recorded. A
few Cu L-lines were observed originating
from the brass nozzle of the capillary
discharge. The Na X and Na XI a-lines
dominate the spectra in the same manner
as in the neon gas puff spectra.
The sodium a transitPons were recorded
siniultaneously with high resolution
using the second-order diffraction from
a curved KAP crystal spectrograph. A
typical spectrogram is shown in Fig. 11.
A narrow slit was used to provide
spatial resolution along the axis of the
N a X Ib
Fig. 9 . Pinhole images of neon plasmas
for two different diameter nozzles
(upper 2.5 cm, lower 1.75 cm).
i i
Fig. 10. Mica crystal spectrogram from
NaF plasma generated with the capillary
discharge.
JOURNAL DE PHYSIQUE
C6-252
implosion. The spectral line shapes can
be correlated to the plasma diameter and
therefore exhibit some resemblance with
the pinhole images (see Fig. 11).
PINHOLE
h%
L,
Heg
Fig. 11. KAP crystal (002) spectrogram
showing spatial uniformity along the
sodium lines. The associated plasma
pinhole image is shown on the left, the
capillary nozzle being on top.
Sodium line intensities were obtained
from both spectrograph images to measure
line emission variations along the
interelectrode gap and line ratios. We
obtained good agreement between both
spectrograph measurements for intensity
variations. The He-like transitions
show a maximum near midgap and decrease
towards either electrode. The La
emission shows two broad intensity
maxima. Both a line integrated
intensities vary by a factor of 2-2.5.
The La/Ha intensity ratio varies from
around 1 along a plasma region of more
than 1 cm length on the nozzle side, to
about 0.5 on the cathode side over about
1.8 cm of the plasma column. The
estimated sodium plasma temperature
appears in excess of 200 eV based on the
measured line intensity ratios (a, p and
-y transitions).
Absolute line intensities were obtained
by integration over the line profile
with an average value reached by
summation over the individual scans.
Most of the sodium line radiation occurs
in the a transitions with an energy
between 0.7 and 1 W. The corresponding
power in these two lines is 42 + 10 GW
based on a 20 nsec pulse duration from
the XRD measurements. The power in the
Hea line (11.0 A) is 27 GW in agreement
with an independent measurement of 25 GW
obtained with a Ge-filtered x-ray diode.
IV
-
CONCLUSIONS
Imploding plasmas driven by pulsed power
generators are found to be efficient
emitters of keV energy x-rays for both
the gas puff and the capillary
discharges studied in this work.
Appropriate configurations and
conditions can produce uniform
implosions of several cm in length. The
plasma uniformity and stability can be
improved by decreasing the risetime of
the current during the implosion and by
adjusting the diameter of the gas-puff
nozzle. Argon is efficiently imploded
with 1-MA driving currents, but is
overheated and ionized well beyond the
Ne-like argon stage that is desired for
x-ray laser experiments; therefore,
higher atomic number elements, e.g.,
iron or chromium are indicated for
future work. Implosions of sodium
plasmas using a capillary-discharge
source have produced an intense pump
source of 11.0 A photons appropriate for
Na/Ne line coincidence x-ray laser
experiments.
REFERENCES
1. D. Mosher, S . J . Stephanakis, I . M . Vitkovitsky. C.M.
Dozier, L.S. Levine, and D.J. Nagel. Appl. Phys. L e t t .
23, 419 (1973).
2. C. S t a l l i n g s , K . Nielsen, and R. Schneider, Appl.
?hys. L e t t . 29, 404 (1976).
3. P.G. Burkhalter, C.M. Dozier, and D.J. Nagel. Phys.
ev. A l5, 700 (1977).
4. P.G. Burkhalter, C.M. Dozier, C. S t a l l i n g s , and
Robert D . Cowan, J . Appl. Phys. 49, 1092 (1978).
5 . J . Shiloh, A. F i s h e r , and N. Rostoker, Phys. Rev.
515 (1978).
Lett.
6. P.G. Burkhalter, J. Shiloh, A. F i s h e r , and R.D.
Cowan, J. Appl. Phys. 50, 4532 (1979).
7. C. S t a l l i n g s , K . Childers, I. Roth, and R.
Schneider, Appl. Phys. L e t t . 35, 524 (1979).
8 . J.C. Riordan, J.S. Pearlman, M. Gersten, and J.E.
Rauch. Subkilovolt X-Ray Emission f r m Imploding
Wire
.
Plasmas, AIP Conf. Proc. No. 75, 35, (1981).
9. R.D. Bleach. P.G. Burkhalter, D.J. Nagel, and R.L.
Schneider, 3 . Appl. Phys. 4, 1273 (1983).
10. C.D. C h a l l i s , A.E. Dangor, E.S. Wyndham, D. Mosher.
W.G. B e s s e l l , and D.J. Bond, i n Proc. 11th European
Conf. Controlled Fusion and Plasma Physics, Aachen,
(1983).
11. R.B. Spielman. D.L. Aanson, M.A. Palmer, M.K.
Matzen, T.W. Hussey, and J . M . Peek, J . Appl. Phys. 57,
830 (1985).
12. S . J . Stephanakis, J . P . Apruzese, P.G. Burkhalter,
J . Davis, R.A. Meger, S.W. MeDonald, G. Mehlman, P.F.
O t t i n g e r , and F.C. Young, Appl. Phys. L e t t . 48, 829
(1986).
13. A.V. Vinogradov, I.Yu. Skobelev, and Yudov, Sov. J.
@anturn Electron. 2, 630 (1975).
14. J . P . Apruzese and J. Davis. Phys. Rev. A. 31, 2976
(1985).
15. R.A. Meger, R.J. Comnisso, G. Cooperstein, and S.A.
Goldstein, Appl. Phys. L e t t . 42, 943 (1983).
16. P.F. O t t i n g e r , S.A. Goldstein, and R.A. Meger. J .
Appl. Phys. 56. 774 (1984).
17. F.C. Young, S . J . Stephanakis, and V.E. Scherrer,
F i l t e r e d X-Ray Diodes f o r Imploding Plasma Experiments,
Rev. S e i . I n s t r . ( t o be published).
18. P.G. Burkhalter. D.B. Brown and M. Gersten, J.
Appl. Phys. 52, 4379 (1981).
19. P.D. Rockett, C.R. Bird, C.J. Hailey. D. Sullivan.
D.B. Brown, and P.G. Burkhalter, Appl. Optics 2,2536
(1985).
20. D.B. B r m and M. Fatemi, J. Appl. Phys. 1
,
2540
(1980).
21. B.L. Henke and P.A. Jaanimani. Rev. S c i . Instrum.
56, 1537 (1985).
22. R.D. Cowan, J. Opt. Soc. Am. 3, 808 (1968).
23. N.J. Peacock, R.J. Speer, and M.G. Hobby, J. Phys.
a,
~