MÖSSBAUER STUDY OF Xe ATOMS IN ANION VACANCIES OF NaI

HAL Id: jpa-00215406

https://hal.archives-ouvertes.fr/jpa-00215406

Submitted on 1 Jan 1973

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.

MÖSSBAUER STUDY OF Xe ATOMS IN ANION VACANCIES OF NaI

J. Biersack, U. Morfeld, D. Schneider

To cite this version:

J. Biersack, U. Morfeld, D. Schneider. MÖSSBAUER STUDY OF Xe ATOMS IN AN- ION VACANCIES OF NaI. Journal de Physique Colloques, 1973, 34 (C9), pp.C9-163-C9-165.

�10.1051/jphyscol:1973931�. �jpa-00215406�

~ O U R N A L DE PHYSIQUE C~Iloque C9, supplhtnent au no 11-12, Tome 34, Novembre-Dkcetnbre 1973, page C9-163

MOSSBAUER STUDY OF Xe ATOMS IN ANION VACANCIES OF NaI

J. P. BIERSACK, U. MORFELD and D. S C H N E I D E R Hahn-Meitner Institute, D 1 Berlin 39, F R G

RkumC. ^- L'effet Mossbauer au niveau 39,6 keV du Xe-129 dans un reseau Nal a ete mesurk avec un absorbeur a xenon solide a 4,2 K. Les atomes de Xe-129 sont engendres par la decrois- sance p- des ions 1-129 du reseau ; I'energie de recul de 0,6 eV est trop faible pour chasser le xenon.

Ainsi, l'atome de xenon est suppose adherer a la lacune anionique creee.

Abstract. - The Mossbauer effect at the 39.6 keV level of Xe-129 in a NaI lattice was measured against a solid xenon absorber at 4.2 K. The Xe-129 atoms are generated by 8- decay of 1-129 lat- tice ions ; the recoil energy of 0.6 eV is too low to dislocate the Xe. Thus, the Xe atom is expected to adhere to the newly created anion vacancy.

The experimental results yield - under the assumption of quadrupole splitting - the following data (using a least square fit)

isomeric shift IS = (- 0.01 ::: 0.19) mm/s , electric field gradient V,, = ( 5 1 1 & 3) e/A3

.

As there exists no measurable isomeric shift, one must conclude that Xe in the ionic crystal has the same electron configuration as in the absorber (neutral atoms in the ground state).

The electric field gradient of I I e / A b t the nucleus corresponds to an external field gradient of 0.08 e/Al (using the Sternheimer factor 7 = 143 of the isoelectronic Cs+). From this value the lattice position of the Xe atom is derived to be slightly off the vacancy center.

1. Introduction. - The Mossbauer effect in " ' ~ e atoms located in anion vacancies in Nal, has been studied in order to investigate the trapping of rare gas atoms t o such vacancies. In particular the electronic structure and the lattice locatiun of tlie rare gas atoms are of interest. T o the authors knowledge, there exists no other direct experimental check on these data.

The sodium iodide under investigation contains 1-129 which decays by a

p-

transition into Xe-129, thus simultaneously creating tlie Xe atom and the anion vacancy. The nuclear recoil from electron and neutrino emission is 5 0.6 eV. This means that the Xe atom cannot escape from tlie surrounding ionic cage which would require a minimum energy of about 1.4 eV. Theoretical estimates indicate a binding energy t o the vacancy of the ordel- of 1 eV with respect to a regular interstitial position of Xe, and in addition a potential barrier of about 0.4 eV for escaping to neighbouring cells through tlie most f:ivour:tble face centered positions (energies arc concluded by extrapolation from data of Norgett, Lidinsd [I], and Biersack, Ernst [2]).

After the nuclear tr;~nsmutation the Xe-I29 remains in the 39.6 keV excited level for about

lo-''

^{s which}

is sufficient time for phonons to movc out o f the region ; therefore the measurement is supposed to be relevant for the equilibrium state. The excited level of spin

+

is susceptible to quiid~-upole splitting thus probing the electric field gradient, whilc the ground state of spin

3

does not split. Therefore a simple line doublet is expected to yield the dcsired informution On the Xe position. Theoretically the energy of the

excited state is shifted due t o hyperfine interaction between the nuclear quadrupole moment Q =

-

0.41 b and the electric field gradient V,, by a n amount

As the level splits symmetrically in this case, and the intensities of the two lines are expected t o be equal (no nuclear alignment), only the absolute value of

V,, - not the sign - can be determined.

2. Experiment. - The Nal Mossbauer source and a solid Xe absorber are both placed inside a liquid He bath (4.2 K). The source performs a sine wave motion with less than 1

7;

higher harmonics by means of an electromagnetic drive system with negative feed back.

The 39.6 keV 7 quanta are observed through thin films of aluminum and capton at the bottom of the cryostot by a Ge(Li) detector of own production. The y counts within equidistant time intervals are collected into subsequent channels of a multichannel analyser

(SO called ⁽⁽ tiille i~rorlc ^D). The analyser is triggered by

:I start signal 30 before maximum velocity is reached in each oscillation. The calibration of speed throughout the relevant interval is accomplished with a Fe-57 magnetic hyperfine spectrum ~vhicli is measured simultaneously with the Xe-129 spectrum (using the othcr end of the oscillating rod). Thus velocities are well controlled over thc 700 hours of data acquisition.

Speeds are found to be stable within 0.3

%.

Thc absosbcr ~iscd in the cxpcriment is solid Xe

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1973931

C9-164 J. P. BIERSACK, U. MORFELD A N D D. SCHNEIDER

frozen onto a thin capton foil. The thickness obtained from 39.6 keV y absorption is (26 f 4) pm, the content of Xe-129 is 26.4

%

(natural abundance).

Although there exist no other measurements with solid Xe, this kind of absorber is chosen to ensure an unperturbed 5s2 5p6 electronic configuration to com- pare with the Xe atom in the Nal trap.

3. Results. ^- The data acquired in the experiment are depicted in figure 1. The least square fit with two Lorentz curves of equal intensities yields the following results :

effect (single dip) = 0.70

5

0.02

7;

full width at half min. = 7.0 f 0.9 mm/s isomer shift ^{= -} 0.014 f 0.190 mm/s el. field gradient

I

Vz,

I

= 11.4

F

2.9 e/A3

.

The resulting Lorentz curves and the sum curve are shown in the graph.

Obviously the efect is very small whicl~ must be attributed mainly to the low Debye-Waller factor f of the solid Xe absorber. Here, f = 0.05

+

0.01 is found,

FIG. la.

-

Mossbauer spectrum of Xe in NaI, experimental data and least square fit with two Lorentz curves of equal

intensity.

.

^'

. . .

Normalized 1:

..-... . . . . .

DwiaBm 0 . ;

... . . . . . . . .

- . : . _{. .} :.

FIG. lb. - Magnetic hyperfine spectrum of iron for velocity calibration and long term stability test.

and hence an effective Debye temperature of onl!

38 K. This low value is typical for inert gas van-der Waals crystals (from specific heat data, 0, = 55 K i.;

obtained ; this kind of discrepancy has also beep observed for solid Kr [3], [4]).

The lilie n ~ i ~ l t h of 7.0 f 0.9 mm/s agrees well wit11 the theoretically expected value of 7.01 6 mm/s deducec from the life time.

The ison7er sllifi is virtually zero, from which i t i,, concluded that the trapped Xe atom is in the sam~, electron configuration as in solid Xe, i. e. in the ground state 5s' 5p6 (missing or excited s electrons waul('

cause considerable isomer shifts, p electrons goin::

into molecular orbitals yield quadrupole splittings e. g. about 40 mm/s for XeF,, XeF, or about 30 rnm/,.

for XeCI, [5]).

The observed small C J I I N L / ~ L I / ~ O / L J splittil~g may b,:

explained by a Xe position (mean over

w

_lo-' s l

slightly off center. The field gradient

I

V,,

I

= 1 1.4 e/P\' at the nucleus corresponds to an external field gradien;

of

Here, the antisl~ielding factor y = 143.5 of the isoelec- tronic Cs' is used, as calculated by Sternheimer [6].

In order to derive the lattice position from the measu- rement, the field gradients along lines of simple sym- metry are calculated. As V,, decreases as 1/r3 (r. dis- tance from a point charge), it is sufficient to neglect contributions from outside a cube of 7 x 7 x 7 ion, (accuracy better than 1

%).

In figure 2 the experimental finding (hatched lines) is compared with the calculated Vz, (full drawn lines). This indicates a position about dl4 zz 0.8

A

away from the central vacancy. This final

-a

01 0.2 0.3 0.L 0 5

FIG. 2. -- Electric field gradients along lines of simple synlnietry, calc~tlated for NaCl structures in universal coordinatcs (full lines). The experimental values are depicted by hatched liner (< is the projection on the

<

100

>

direction). Interceptions

indicate possible positions of the rare gas atom.

MOSSBAUER STUDY OF Xe ATOMS I N ANION VACANCIES OF NaI C9- 1 65

result, however, should be accepted with sollie caution Besides, the experiment yielded data on solid Xe as it was derived from an unresolved doublet fitted absorbers (narrow line width and rather low effective into data of some statistical scatter. Debye temperature).

4. Conclusion. ^- It could be shown that the Moss- Acknowledgment. ^- The authors are very much bauer effect is suitable for obtaining information indebted to Dr. S. Roth for calculating the electric about the electronic structure and also - with some field gradients. They also wish to thank Mrs. S. Franke ambiguity - about the position of rare gas atoms in for her kind assistance in performingthe measurements

alkali halide crystals. and evaluating the data.

References

[ I ] NORGETT, M . J . , LIDIARD, A. B., Pliil. A4ag. 18 (1968) [4] GILBERT, K . , VIOLET, C. E., Piiys. Lett. 22 (1966) 52.

1193. [5] PERLOW, G . J., PERLOW, M. R., Rev. Mod. Phys. 36(1964) (21 BIERSACK, J. P., ERNST, H., to be published. 353 ; J . Chern. Plrys. 41 (1964) 1157.

[3] PASTERNAK, M., SIMOPOULOS, A,, BUKSHPAN, S., SONNINO, T., [6] STERNHEIMER, R . M., P11ys. Rev. 84 (1951) 244.

Ph.vs. Lett. 28a (1968) 381.

DISCUSSION

R. J. FRIAUF. - This is a comment on the calcu- tial, up to terms in rlO, which include essentially lation of the electric field gradients. For the [loo], all ions in the crystal. These would probably make [I 101, and [ I l l ] directions, there exist in the litera- very little difference in the calculated values however.

ture very accurate expansions of the Madelung poten-