Z. Phys. D - Atoms, Molecules and Clusters 23, 67-70 (1992)

Atoms, Molecules zo,,=r,, fiJr Physik D and Clusters © Springer-Verlag 1992

Laser and radio-frequency spectroscopy of the 5d36s 2 4p3/2,5/2 states in Ta I J. Persson, U. Berzinsh*, T. Nilsson, and M. Gustavsson Department of Physics, Chalmers Universityof Technology,and Universityof Gfteborg, S-41296 G6teborg, Sweden Received 23 December 1991

Abstract. The hyperfine structure of the metastable atomic states 5d36s z 4p3/2 and 4Ps/2 in 181Ta has been studied using high resolution laser spectroscopy and laser radiofrequency double-resonance methods. PACS: 35.10.F; 32.30.B; 32.30.J

Introduction The transition elements with an unfilled d-shell have in most cases many low-lying metastable states of the evenparity configurations nd N, ndN-l(n+l)s and nd N-2 (n + 1)s 2. These metastable states can in general be sufficiently populated, by different methods, for measurements of hyperfine structure and isotope shifts. Therefore many of the transition elements are especially favourable for extensive investigations. In the case of the 5d-elements measurements are of particular interest since the nuclei lies in the deformed region, yielding large nuclear quadrupole moments. A review of the hfs theory, experimental techniques and hyperfine structure (hfs) measurements in the 4d and 5d elements can be found in [I]. 181Ta is the only stable tantalum isotope with a nuclear spin of 7/2. The isotope lS°Ta is radio-active with a very large half-life time (> 1 1012 y) and thus it has a natural abundance. However, having a low abundance (0.012%), makes observations of this isotope difficult in natural samples. The nuclear magnetic dipole moment of lSlTa is & = 2.341 #N [2] and the nuclear electric quadrupole moment has earlier been studied by Konijn et al. [3] using pionic and muonic X-ray measurements obtaining, Q=3.30(6) and 3.28(6)barn, respectively. The nuclear electric quadrupole moment of 181Ta is thus one of the * P e r m a n e n t address: Department of Spectroscopy, University of Latvia, 19 Rainis Blvd, 226098 Riga, Latvia

largest for stable isotopes, giving rise to a very large quadrupole coupling constant in the hfs. In Fig. 1 the low lying energy levels of Ta together with the studied transitions are shown. The designations of the states are given according to Moore [4]. Hfs has earlier been measured with high precision in the five lowest atomic states 5d36s24F3/2,5/2,7/2,9/2 and *P1/2 by Bfittgenbach et al. [5] and Bfirger et al. [6] using the Atomic Beam Magnetic Resonance (ABMR) method. In this work we report on high-precision if-measurements of the hfs in the 5d36s24P3/2 and 4p5/2 states. In addition we also report on hfs in the excited state 5dS6s6p 4F5/2 and the unidentified J = 7 / 2 state at 26 960 cm- 1. Ta 2s 2.p 2D 2F 2(3 2H4 P 4D 4F 413 4 H 6S 6D 6F 6G XX M -l

30000-

-

_

2013139,

-

"__

J, S i 782

]00oo-

M

m

Fig. 1. The lower part of the energylevel diagram of Ta

68

Experimental In general one of the major difficulties in measurements of properties in the refractory elements such as Ta, is to obtain an intense atomic beam and also a sufficient population of the metastable states. To accomplish this the molten ball technique [7, 8] was used, due to its capability of producing high temperatures and also due to its relative simplicity. This method utilizes electron bombardment on the tip of a thin Tantalum wire (1 mm) where a drop of metal is formed from which atoms are evaporated. At the melting point of Ta (3269 K) the vapour pressure is 5.10-3 torr, which is sufficient to produce an atomic beam. The electron bombardment of the wire also gives rise to inelastic collisions between the evaporated atoms and the electrons, producing a sufficient population of the high lying metastable states. A schematic view of the experimental set up is shown in Fig. 2. The atomic beam, with a collimation ratio of about 1 : 100, was intersected perpendicularly with a laser beam from a ring dye laser (CR 699-21), pumped with an argon ion laser (Coherent Innova 100), and operated with the dye Rhodamine 6G. The fluorescence light following the laser excitation, was imaged on and registered by a photomultiplier tube (PMT). In the High Resolution Laser Spectroscopy (HRLS) experiment the laser frequency was scanned over the hfs components. Simultaneously the transmission fringes from an actively stabilized 51.24 MHz Fabry-Perot interferometer (FP) [9] was recorded, giving the necessary frequency scale. The laser frequency was controlled by the output voltage from a D/A converter in the PCcomputer. The fluorescence signal and the FP fringes were simultaneously A/D converted and stored in the PC [10]. In the Laser-Radio-Frequency Double-Resonance experiment (L/RF) the laser beam was tuned to one of the hfs components and the laser beam intersected the atomic beam twice. In the first interaction region the population of the lower state could be reduced by 50%90% due to optical pumping, resulting in a reduction of the fluorescence light in the second interaction region. Between these two regions a radio-frequency (rf) field was applied. If the frequency of the if-field fulfilled the resonance condition the depleted state was repopulated from a neighbouring state. As a result an increase in the fluorescence signal was then detected in the second

interaction region by the PMT. The radio-frequency was repetitively scanned by a PC-computer [10] via a GPIB interface and the PMT signals were A/D-converted and signal averaged. Three pairs of Helmholtz coils were used in order to reduce the influence of the earth magnetic field in the rf interaction region.

Measurements Two transitions were studied, 5782.29 ~ and 5647.47 ]k connecting the metastable 5d36s24p3/z a n d 4P5/2 with 5da6s6p 4F5/z and an unidentified J = 7 / 2 state (26960 cm- 1), respectively. A spectrum of the transition 5 d a 6 s 2 4P3/2 -- 5 d 3 6 s 6 p 4_/75/2 (5782 ]k) is shown in Fig. 3. The line width in the high resolution laser spectroscopy experiment was about 40 MHz. The main contribution to the width arises from the natural line widths and doppler broadening due to the divergence of the atomic beam. Neutral density filters were used to decrease the laser power in order to eliminate saturation broadening. For the odd-parity states the hfs separations and the evaluated magnetic dipole and electric quadrupole coupling constants A and B are presented in Table 1. The quoted limits of error represent two standard deviations and a systematic error, due to drifts of the Fabry Perot interferometer locking, of 0.4%0 of the measured hfs-sepa-

@

® ®

®

Q

00

Fig. 3. -- 5d a

A

high-resolution

spectrum

6s6p4Fs/2transition (5782 A)

of

the

5d 3 6S24P3/2

C°mp°rl Fig. 2. Schematic diagram of the experimental set-up

69 F=6

Table 1. Measured lffs separations and evaluated hfs constants in the excited states State

hfs-separation

Measured (MHz)

5d3 6s6p 4F~/2

1-2 2-3 3-4 4-5 5-6 A

959.7(3.2) 1444.0(3.0) 1921.6(2.0) 2403,2(5.8) 2885,6(22,2) 480.5(8) 480.2(2) ~ --0.7(10.1) -4.3(2,9)"

F=5

B

5d36s6p 4F3/2 F=4

F=3

F=2 F=I

XX7/z

1-2 ~3 3-4 4-5 5-6 6-7 A B

837.7(1.9) 1273.5(1,5) 1733.0(1.7) 2230.3(2,5) 2760.4(3.4) 3345.0(2.4) 453.7(3) 392.2(3.6)

F=5 F=4

5d36s 2 4P3/2

[11]

F=3 Table 2. Measured and corrected hfs separations in the metastable

states together with the evaluated hfs constants State

hfsseparation

Measured (MHz)

Corrected (MHz)

'~P3/2

2-3 3-4 4-5 1-2 2-3 3-4 4-5 5-6

2102.6941(37) 1900.9308(32) 933.6896(38) 11. t 708 (36) 200.8533(43) 611.4649(21) 1316.6665(38) 2390.0857(40)

2101.06(43) 1902.16(32) 934.11 (12) 11.1676 (44) 200.8512(48) 611.4641 (23) 1316.6664(38) 2390.0919(56)

4P5/2

hfs constants 4P3/~ A B 4P5/~ A B

379.268 (90) - 1348.74(64) 251.0593 (4) 1718.3566(60)

379.346 (39) - 1347.70(27) 251.0595 (4) 1718.3665(64)

ration. In the unidentified state one s e p a r a t i o n was not evaluated due to unresolved hfs c o m p o n e n t s . T h e Aand B-factors for the 4Fs/2 state agree with m o r e accurate m e a s u r e m e n t s m a d e by D u q u e t t e et al. [ 1 t ] using Doppler-free i n t e r m o d u l a t e d s a t u r a t i o n spectroscopy. In T a b l e 2 the m e a s u r e d hfs-separations of the lower states are given trogether with the c o r r e s p o n d i n g A a n d B factors using radio frequency spectroscopy. T h e radiofrequency resonances of the 4P3/e state are s h o w n in Fig. 5. T h e error in these values is taken as 1/10 of the linewidth. The line width is typically 20-45 k H z and the m a i n contributions to the linewidth is the m a g n e t i c field b r o a d e n i n g caused by an i n h o m o g e n e o u s m a g n e t i c field in the rf interaction region a n d the transit-time b r o a d e n ing due to the time of flight t h r o u g h the rf loop. F r o m the hfs splittings the experimental A- a n d B-constants were evaluated.

F=2

Fig. 4. Energy level diagram indicating the transitions in Fig. 3 F=5 - rF=4

i

F=5

o., ;'k \

y

%

.~9"~

F=4 933.60

933.65

933.70

933.75

933.80

F=4 - F=3

5d36s24P3/2

.A :

..

.:~o4~ F=3

,

1900.85

i

"~.~, ..........

1900.90

t

i

1900.95

1901.00

1901.05

F=3 - F=2

i

~t

I

F=2 21o2.oo Fig. 5.

Rf

resonances

21o2.65

in

2102.70

2102.75

the 5d 36s 2 '~P3/2state

2102,80

70 The if-resonances have been corrected for second order effects, due to the hyperfine interaction, which is non-diagonal with respect to the quantum number J. These corrections are quite large in the 4P3/2 state caused by the Close lying 4P1/2 state. The corrections have only been calculated within the 4p states, since they are quite close in energy and the contribution from other states are small. The radial integrals used in the calculation were taken from B/ittgenbach [6], and these were assumed to reproduce the true values with less than 25%. The rather large errors of the corrected hfs separations in Table 2 are due to the uncertainties in the radial integrals. The second order corrections of the hfs separations improved the fit of the A- and B-factors considerably. The number of states where the hyperfine structure is known with high accuracy, is not large enough to justify an extensive analysis, since these states belong to only one configuration. The aim of this investigation has been to increase the available amount of high accuracy hfs data in tantalum. In order to get an accurate value of the nuclear quadrupole moment further experimental data is needed.

This work was financially supported by the Swedish Natural Science Research Council (NFR). One of us (U.B.) acknowledges the Swedish Institute (Sl) for support during his visit to Sweden.

References 1. Biittgenbach, S.: Hyperfine structure in the 4d- and 5d-shell atoms. Springer Tracts in Modern Physics. vol. 96. Berlin, Heidelberg, New York: Springer 1982 2. Lederer, C.M., Shirley, V.S.: Table of isotopes. (7th edn.) New York: Wiley 1978 3. Konijn, V., Van Doesburg, W., Ewan, G.T., Johansson, T., Tibell, G.: Nucl. Phys. A 360, 187 (1981) 4. Moore, C.E.: Atomic energy levels.Washington D.C.:US Goverment Printing Office 1971 5. Btittgenbach, S., Meisel, G.: Z. Phys. 244, 149 (1971) 6. B/irger, K.H; Biittgenbach, S., Dicke, R., Gebauer, H., Kuhnen, R., Tr~iber, F.: Z. Phys. A -Atoms and Nuclei 298, 159 (1980) 7. Pendelbury, J.M., Ring, D.B.: J. Phys. B 5, 386 (1972) 8. Rubinsztein, H., Lindgren, I., Lindstr6m, L., Riedl, H., Ros6n, A.: Nucl. Instrum. Methods 119, 269 (1974) 9. Hanstorp, D., Carlberg, C., Berzinsh, U.: (to be published) 10. Gustavsson, M.: (to be published) 11. Duquette, D.W., Doughty, D.K., Lawler, J.E.: Phys. Lett. A 99, 307 (1983)