X-RAY SPECTROSCOPY OF HIGHLY-IONIZED ATOMS IN AN ELECTRON BEAM ION TRAP (EBIT)

HAL Id: jpa-00229349

https://hal.archives-ouvertes.fr/jpa-00229349

Submitted on 1 Jan 1989

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished 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.

X-RAY SPECTROSCOPY OF HIGHLY-IONIZED

ATOMS IN AN ELECTRON BEAM ION TRAP

(EBIT)

R. Marrs, C. Bennett, M. Chen, T. Cowan, D. Dietrich, J. Henderson, D.

Knapp, M. Levine, K. Reed, M. Schneider, et al.

To cite this version:

JOURNAL DE PHYSIQUE

Colloque Cl, suppl6ment au nol, Tome 50, janvier 1989

X - R A Y S P E C T R O S C O P Y OF H I G H L Y - I O N I Z E D A T O M S IN AN E L E C T R O N B E A M I O N TRAP ( E B I T )

R.E. MARRS, C. BENNETT, M.H. CHEN, T. COWAN, D. DIETRICH,

J.R. HENDERSON, D.A. KNAPP, M.A. LEVINE", K.J. REED, M.B. SCHNEIDER and J.H. SCOFIELD

Lawrence L i v e m o r e National Laboratory, P.O.Box 808, L-296, Livermore CA 94550, U.S.A.

" ~ a w r e n c e Berkeley Laboratory, Berkeley CA 94720, U.S.A.

~ 6 s u m 6

-

Au laboratoire national de Lawrence Livermore, un pisge 2 ions EBIT (Electron Beam Ion Trap) est utilisi afin de produire et de piEger des ions t r k fort c h a r g k (q 5 70+), permettant de realiser des mesures de spectroscopie rayons-X. Des mesures r6centes de recombinaison di&lectr?nique, d excitation par impact 6lectronique. et d16nergies de transitions sont presentees.

Abstract

-

An Electron Beam Ion Trap at Lawrence Livermore National Laboratory is being used t o produce and trap very-highly-charged ions (q 1 7 0 + ) for x-ray spectroscopy measurements. Recent measurements of dielectronic recombination, electron impact excitation and transition energies are presented,.

1. INTRODUCTION

Almost two years ago a new device called an Electron Beam Ion Trap (EBIT) was brought into operation at Lawrence Livermore National Laboratory for the purpose of studying very- highly-charged ions using x-ray spectroscopy. During this period EBIT has demonstrated its ability t o make several different kinds of measurements for ionization stages that have not been accessible before.(ls2*3) The measurements may be grouped into three general types: (1) Measurement of electron collision cross sections using line radiation, (2) Precise measurement of energy levels (e.g. Lamb shifts), and (3) Measurement of properties of the

.

trap itself (such as ion temperature and confinement time). In order to give an indication of the current EBIT research program, we present a summary of measurements of the first two types.

2. EBIT OPERATION

The EBIT at LLNL works by trapping ions for long times (up to several hours) inside an electron beam compressed to a density of order 2000 ~ / c m ~ . Electron collision cross sections are determined from x-ray spectroscopy of the trapped ions excited by the electron beam; and energy level information is obtained from calibrated high-resolution spectrometers. The method of successive ionization of ions trapped in an electron beam is also used in the electron-beam ion sources (EBIS) developed t o provide highly charged ions for injection into accelerators. ( 4 * = ) In contrast t o the EBIS1s, the EBIT at LLNL uses a different and much smaller geometry, which is optimized for x-ray spectroscopy.

As shown schematically in Fig. 1, the ion trap consists of cylindrical copper drift tubes, which contain the trapped ions. The electron beam, which follows the central magnetic field line of t h e superconducting Helmholtz coils, is injected vertically from a Pierce gun and travels along the axis of the drift tubes. The beam is adiabatically compressed in the Helmholtz coil field; and at the peak magnetic field of 3 T the electron beam diameter is 7 0 urn. The electron beam current is usually operated in the range of 30 to 120 mA. The electron beam energy can be set very precisely and changed within a few milliseconds to any value up t o 30 kV. The electron energy spread is roughly 50 eV FWHM. X-rays are observed at 90' t o the electron beam through four different x-ray ports.

Ions in low charge states are loaded into the trap by injecting them downward through the electron collector (i.e. antiparallel t o the electron beam). The ions are obtained from a MEVVA source(6) which is fired periodically t o refill the trap. Highly charged ions are then obtained by successive ionization in the electron beam. Usually the ionization stage is selected by setting the electron beam energy t o be just below the ionization potential of

JOURNAL DE PHYSIQUE

To Electron Collector

Liquid Helium High Voltage

Drift Tubes Super- conducting

h

Beryllium Helmholtz X-Ray Coil Windows Trapped Ions 77'Ok Shield Vacuum Chamber Steel Plate U Bucking Coil 5 cm

Fig. 'l

-

The electron beam ion t r a p . The beryllium windows can be removed f o r s o f t x-ray or

VUV spectroscopy. Two d i f f e r e n t drift-tube geometries have been used. In t h e one shown here t h e d r i f t tube s t r u c t u r e operates a t a s i n g l e e l e c t r i c a l potential. I n t h e other version (not shown), t h r e e e l e c t r i c a l l y isolated d r i f t tubes can be biased a t d i f f e r e n t p o t e n t i a l s t o form a higher axial b a r r i e r .

t h e desired charge s t a t e . This works best f o r closed s h e l l ions, where charge-state p u r i t i e s g r e a t e r than 80% have been obtained i n some cases.

The ions a r e trapped r a d i a l l y by t h e space-charge potential of t h e electron beam and a x i a l l y by a potential b a r r i e r applied t o t h e end d r i f t tubes. Surprisingly. €BIT operates more successfully with t h e most highly charged ions even though they a r e much harder t o ionize. This i s because high-charge ions a r e more t i g h t l y bound i n t h e trapping potential and can be held i n t h e t r a p f o r much longer times. A more extensive discussion of t h e LLNL EBIT, including t h e important r o l e of col1 isional ion cooling, can be found elsewhere. ( 2 ) 3. QjELECTRONIC RECOMBINATION MEASUREMENTS

Dielectronic recombination (OR) i s t h e resonant capture of an incident electron i n t o a doubly excited s t a t e , followed by x-ray emission:

The related process of resonant excitation (RE) can a l s o produce x-ray emission a t similar energies:

energy in EBIT and the good energy resolution. EBIT is a powerful tool for measuring a resonant process such as DR. We have performed DR measurements for two very different ions: He-like ~i 26f and Ne-l ike A U ~ ~ + .

Since the DR cross sections being measured are quite large, setting the electron beam energy on a DR resonance soon destroys the charge state being measured. We overcome this problem by rapidly switching the electron energy between the resonant energy and a higher nonreson- ant energy which restores the ionization balance. Typically only

-

10% of the time is spent on resonance. An excitation function is obtained by taking successive data runs in which the lower electron energy (i.e. the DR energy) is changed by a small amount (typically 20-eV steps) and normalizing each run to a common nonresonant upper energy.

(a) Helium-like Nickel

We have measured a DR excitation function for He-like ~ i by detecting the K x-rays from ~ ~ + the recombined ~ i ions. For He-like target ions DR produces one and only one K x-ray ~ ~ + photon; so the number of K x-rays produced at a given incident electron energy is propor- tional t o the DR cross section. This technique was also used by Briand et al. in an experi- ment with an electron beam ion source in which an x-ray signal from DR of ~ r 1 4 + ions was observed. (10)

Examples of EBIT spectra obtained in a Ge detector a t three different electron energies are shown in Fig. 2. Since the n = 2 -, n = 1 and n = 3 + n = 1 transitions are well resolved from each other and from the higher members of the K series, we present separate excitation functions for these three energy bands in Fig. 3. Three interesting features which have not been observed before are apparent in these data. First, the centroid of the n = 2 -, n = 1 component of the KLM resonance strength is clearly displaced from the centroid of the n = 3

-, n = 1 component, reflecting the different distribution of resonance strength for the two decay channels. Second, the n 2 4 -, n = 1 band shows a shelf at the Ku direct excitation threshold,corresponding to t h e change from an electron bound in a high Rydberg level t o a free electron. Third, the RE process is clearly observed for the n = 2 + n = 1 transitions as the peaks above threshold in the excitation function. In this case our results imply that, for He-like Ni, RE is a significant contributor to the excitation of line radiation in plasmas.

A complete analysis of the above features is beyond the scope of the present paper. How- ever, by focusing on the KLL DR resonances it is possible t o make a limited comparison between the experimental measurements and theoretical calculations. In particular, our data provide a check of both the absolute magnitude and the distribution of the theoretical resonance strength. This is illustrated in Fig. 4, where the measured x-ray cross section is compared to a multiconfiguration Dirac Fock calculation(ll) for the KLL resonances. (b) Neon-like Gold

We have extended our study of DR t o an L-shell target farther up in the periodic table, Ne- like A U ~ ~ + . In this case the cross section is more complicated because of the larger number of configurations available for the doubly excited resonant states. In addition, compared to He-like Ni, it is more difficult to obtain a pure charge state in the trap, and the Na- like ionization stage of gold appears t o be as prevalent as the Ne-like.

Preliminary results for the LMN DR resonances in gold are shown in Fig. 5 along with a theoretical model. The measurements seem to support the predicted distribution of resonance strength. However the overall magnitude is not yet well determined experimentally.

4. ELECTRON IMPACT EXCITATION MEASUREMENTS

Electron impact excitation (IE) cross sections may be obtained in a manner similar to the DR measurements described above. In this case, however, the electron energies are carefully chosen t o avoid resonances, since they can influence the observed line intensities. To date, we have obtained impact excitation data for two Neon-like ions in different regions of the periodic table, ~ a and ~ A U ~ ~ + . ~ + This should help provide an understanding Of the role of relativistic and QED effects in electron-ion collisions. Since some of the fine

JOURNAL DE PHYSIQUE

Ni2" Target ions

600

r I n = 2 - + 1 Direct - n = 3 + 1 excitation Ee

+

Off DR

X-ray energy (keV)

Fig. 3 - Measured DR electron excitation function with ~ itarget ions f o r three ~ ~ +

different x-ray energy bands. The peaks above threshold for the n = 2 n = 1 x-rays are due t o RE. J - 4 I

2

O -

8 -

6 -

4

2 -

I I I I KLL

-

!

n=2

+l:

X-Rays

1 I

-

I

r

KLM K, Direct

-

i Excitation I

4

Threshold

2

-

I I I I I I KLN l I # ' m .

+

KLO m I

0

' L - -l2

-

I I I t

-

I I I I

-

I PC. I I I

-

KMM KMN

_-

4

2

0

0

I

-.

.

;

'2-.'.

5

6

7

8

9

10

Cl-450 JOURNAL DE PHYSIQUE

Electron

Beam Energy

(keV)

Fig. 4

-

Structure of the KLL DR feature compared to theory for He-like Ni target ions. The vertical bars are the theoretical resonance strengths. The curve is the theoretical crass section obtained by folding the calculated resonance strengths with the experimental electron energy resolution function. The data points at electron energies 2 5.4 keV show a contribution from DR on lower-charge-state Ni ions, which are also present in the trap.

(a) Neon-Like Barium

Our ~ a IE cross section measurements have been published elsewhere(l) and we only ~ ~ + summarize the results here.

A typical Si(Li) spectrum is shown in Fig. 6. The three lines of greatest intensity around 5 keV are the n = 3 -, n = 2 transition lines. The weak features between 6 and 8 keV are due to DR onto ~ a or tungsten (a trap contaminant) in several charge states and radiative ~ ~ + recombination (RR) t o excited states. The feature just above 9 keV i s R R t o the five

unresolved n = 3 levels in ~ a ~ ~ + . A high resolution crystal-diffraction spectrum obtained in a longer (5 h) run at the same electron energy'is shown in Fig. 7.

The three strongest L x-rays are resolved in the crystal spectrum. By using the relative line intensities from the crystal spectrum and the ratios of the (unresolved) n = 3 + n = 2 lines to the RR peak from the Si(Li) spectrum, we obtain experimental values for some of the n = 2 to n = 3 IE cross sections normalized to the RR cross section, which can be calculated more reliably.

The decay scheme for the relevant ~ a levels is shown in Figure B. ~ ~ + Direct excitation from the ground state dominates the feeding of all the levels chosen for study. Small

corrections for other feeding, as shown in Fig. 8 were made with use of theoretical

I I I I

-

Theory

-

-

_--

-

Ne-like

---m----

-

Na-like

-

Sum

-

-

-

-

-

-

-

l -

Experiment

-

-

₁

-

I ,

l ,

7

-

1

1

-

-

i

i

-

_{f i i i I}

-

f f l I i I I

-

-

l I l ' m r n

.

I I I I

JOURNAL DE PHYSIQUE

0 2 4 6 8 10 12

X-ray energy (keV)

Fig. 6

-

Si(Li) spectrum for Ne-like ~ a ~ ~ + . The spectrum is cut off below = 2.5 keV by absorption in the beryllium windows. The feature labeled kl is attributed t o RR on tungsten ion::, which were a contaminant in the trap. The spectrum has been multiplied by 10 above 6 keV for display purposes.

2pi;

core

I l

X-ray energy (keV)

Fig. 8

-

Diagram of t h e ~ a l e v e l s involved i n t h e cross-section measyrements. The decay ~ ~ + branching r a t i o s and feeding r a t i o s shown f o r t h e e x c i t e d l e v e l s a r e derived from

t h e o r e t i c a l r a t e s a t E,

7

5.69 keV. Upward arrows i n d i c a t e e l e c t r o n IE from t h e ground s t a t e . Downward arrows i n d i c a t e cascade feeding and decay, which sometimes involves l e v e l s t h a t a r e not shown.

Cl-454 JOURNAL DE PHYSIQUE

Table I. Comparison of measured and theoretical electron IE cross sections for neon-like ~ a 4 6 + . Units are 10-21 cm2. In addition to the tabulated errors there is an overall systematic uncertainty associated with the normalization to RR which we estimate t o be

+

5%.

a a a a

Energy ~ h e o r y ( ~ )

he or^(^)

Measured ~ h e o r y ( ~ ) iheory(b) Measured (ev)

(2~;:~ 3s112)0 5303 0.67 0.68 0.50 0.52

Sum J=O 2.58 2.60 2.50

+

-35 1 .89 1.94 2.27

+

.32 (2~:~ 3d512)1 4937 3.44 3.56 3.98

+

-56 2.99 3.23 3.30

+

.46

(a) Obtained by interpolation from values tabulated in Ref. 15 for Z = 5 4 and 5 8 at different collision energies.

(b) Relativistic distorted wave calculation for ~ a 4 6 + , Ref. 13.

In order t o provide a more complete characterization of the impact excitation process, w e have begun measurements of both the angular distribution and the polarization of the decay x-rays (which give equivalent'information for dipole transitions). In addition to the normal observation angle of 90" the EBIT apparatus also permits x-ray observation at an angle of 0". although the solid angle is much smaller. Preliminary ~ a 4 6 + spectra at 0 and 9 0 degrees confirm the expected anisotropy of the RR x-rays. For both energies listed in Table I, the ( 2 ~ 3 ) ~ 34512 1 transition is less intense at O0 than at 90' by a factor of 0.83

.L

0.07, and the (2pij2 363/2)1 transltion is less intense at 0° by a factor of 0.77

+

0.07.

(b) Neon-like Gold

Electron impact excitation of ~u~~~ was studied at an excitation energy of 1 8 keV in the same manner described above for ~a46+. Gold x-ray spectra from a Ge detector and a Bragg

diffraction crystal are shown in Figs. 9 and 10, respectively. Analysis of these data is still in progress. However some interesting observations can already be made. The (2ps)23p1/2)2 transition shows a relative intensity much stronger than expected from the calculated direct collision strengths alone. Presumably this is due to cascade feeding of this line from higher levels.

5. PRECISION SPECTROSCOPY MEASUREMENTS

X-ray energy (keV) 1 I I I I n = 2 - 3 Au at Ee

=

18 keV (g

-

69+) - Y Radiative a m recombination I I V) c u c u n = 6 5 4 3

Fig. 9

- X-ray spectrum from h i g h l y i o n i z e d g o l d (mostly

q = 69+ and 68+) o b t a i n e d - i n a Ge d e t e c t o r a t an i n c i d e n t e l e c t r o n energy o f 18.0 keV.

I

9 .O 9.5 10.0 10.5 11.0

X-ray energy (keV)

-

JOURNAL DE PHYSIQUE Curved Crystal Pivot Point Calibration Source

Fig. 11

-

Johann spectrometer arrangement. F o r s o f t x-ray measurements a p o s i t i o n - s e n s i t i v e microchannel-plate d e t e c t o r i s l o c a t e d tangent t o t h e Rowland c i r c l e , and t h e spectrometer operates i n u l t r a h i g h vacuum. For h i g h e r x-ray energies a p o s i t i o n s e n s i t i v e p r o p o r t i o n a l counter i s placed on t h e Rowland c i r c l e f a c i n g t h e c r y s t a l , and a vacuum chamber i s unnecessary. For scanning o p e r a t i o n t h e c r y s t a l and d e t e c t o r move t o g e t h e r w h i l e €BIT and t h e c a l i b r a t i o n source remain f i x e d .

(a) The Lamb S h i f t i n H- and He-Like Hiqh-Z Ions

P r e c i s i o n measurements o f t h e s p e c t r a o f H - l i k e and He-like high-Z i o n s a r e motivated by t h e d e s i r e t o t e s t QED p r e d i c t i o n s i n t h e s t r o n g f i e l d l i m i t . We have a l r e a d y performed a p r e l i m i n a r y experiment on H - l i k e Ni, and have measured t h e 2p t o 1s t r a n s i t i o n energies i n o r d e r t o compare t h e measured Lamb s h i f t w i t h theory. F i g u r e 12 d i s p l a y s a p a r t i a l spectrum f o r H - l i k e Ni. Our p r e l i m i n a r y measurements a r e i n agreement w i t h theory, and promise eventual p r e c i s i o n much b e t t e r t h a n has been p r e v i o u s l y obtained f o r f a s t high-Z i o n beams produced a t a c c e l e r a t o r s .

(b) H-Shell SDectroscoDy i n Hiqh-Z Ions

Channel number

F i g . 12

-

Data from t h e Johann spectrometer. Top: The d i f f e r e n t c a l i b r a t i o n l i n e s shown were measured i n se a r a t e scans w i t h d i f f e r e n t exposure times. Bottom: The 2 ~ 3 1 2 +

JOURNAL DE PHYSIQUE

6. CONCLUSIONS

The e l e c t r o n beam ion t r a p , with i t s c a p a b i l i t y f o r x-ray spectroscopy, i s now e s t a b l i s h e d a s a unique t o o l f o r t h e study of highly charged ions. In t h i s paper we have given,examples of e l e c t r o n c o l l i s i o n c r o s s s e c t i o n measurements and p r e c i s i o n transition-energy measure- ments performed with t h e EBIT a t LLNL. The i o n i z a t i o n s t a g e s a v a i l a b l e f o r x-ray spectres-

copy with EBIT a l r e a d y exceed t h o s e observed i n t h e most advanced tokamaks, and e s s e n t i a l l y match t h e h i g h e s t charge s t a t e s s t u d i e d i n beam-foil spectroscopy. I t i s l i k e l y t h a t f u t u r e upgrades of t h e e l e c t r o n beam energy i n EBIT w i l l a l l o w t h e x-ray measurements t o be extended a l l t h e way t o hydrogen-like uranium.

7. ACKNOWLEDGEMENTS

Work performed under t h e a u s p i c e s of t h e U. S. Department of Energy by t h e Lawrence Livcrmore National Laboratory under c o n t r a c t number W-7405-ENG-48.

REFERENCES

--

/ l / R . E. Marrs, M. A. Levine, D. A. Knapp, and J. R. Henderson, Phys. Rev. L e t t .

a,

171 5 (1988).

/ 2/ M. A. Levine. R . E. Marrs, J . R . Henderson. D. A. Knapp, and M. B. Schneider. Phys. Scr.

m,

157 (1988).

/ 3/ R. E. Marrs. N. A. Levine. D. A. Knapp, and J . R . Henderson, i n E l e c t r o n i c and Atomic C o l l i s i o n s , e d i t e d by H. B . Gilbody, W. R. Newell, F. H. Read, and A. C. H.

Smith (North-Holland, Amsterdam, 1988). p. 209.

/ 4/ E. D. Donets. Phys. Scr. T3, l1 (1983).

/ 5/ E. D. Donets and V. P. Ovsyannikov. Zh. Eksp. Teor. Fiz. 80, 916 (1981) [Sov. Phys. JETP 53, 466 (1981)l.

/ 6 / I. G. Brown, 3 . E. Galvin. R. A. MacGill, and R. T. Wright, Appl. Phys. L e t t . 49.

1019 (1986).

/ l / F. Bombarda. R. Giannella, E. Kallne, G. J. T a l l e n t s , F. Bely-Dubau. P. Faucher, M. C o r n i l l e , J . Dubau, and A. H. Gabriel, Phys. Rev. A 37, 504 (1988).

/ 8/ H. Hsuan. M. B i t t e r , K. W. H i l l . S. von Goeler, B. Grek. D. Johnson. L. C. Johnson, S. S e s n i c , C. P. Bhalla. K. R. Karim, F. Bely-Dubau, and P. Faucher, Phys. Rev. A 35. 4280 (1987).

-

/ 9 / K. Tanaka, T. Watanaabe, K. Nishi, and K. Akita, Ap. J . , 254, L59 (1982). /10/ J . P. Briand. P. Charles. J . Arianer. H. Laurent, C. Goldstein, J . Dubau,

M. Loulergue, and F. Bely-Dubau, Phys. Rev. L e t t .

2,

617 (1984). 1 M. H. Chen, Phys. Rev. A

33.

994 (1986).

l The RR s e c t i o n s were c a l c u l a t e d using t h e r e l a t i v i s t i c Hartree-Slater model. See

E. B. Saloman, J . H. Hubbell, and J. H. S c o f i e l d , A t . Data Nucl. Data Tables 38, 1

(1988).

/13/ K . J. Reed. Phys. Rev. A 37. 1791 (1988).

/14/ J. H. S c o f i e l d , unpublished. Radiative r a t e s were obtained from t h e RAC

r e l a t i v i s t i c atomic s t r u c t u r e code used a t Lawrence Livermore National Laboratory f o r many years.

l H. Zhang, D. H. Sampson. R. E. H. Clark, and J . B. Mann, A t . Data Nucl. Data Tables 37, 17 (1987).