Neutron inelastic scattering experiments on uranium antimonide

(1)

HAL Id: jpa-00218809

https://hal.archives-ouvertes.fr/jpa-00218809

Submitted on 1 Jan 1979

HAL

is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific 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.

Neutron inelastic scattering experiments on uranium antimonide

G. Lander, W. Stirling, O. Vogt

To cite this version:

G. Lander, W. Stirling, O. Vogt. Neutron inelastic scattering experiments on uranium antimonide.

Journal de Physique Colloques, 1979, 40 (C4), pp.C4-36-C4-37. �10.1051/jphyscol:1979413�. �jpa-

00218809�

(2)

JOURNAL DE PHYSIQUE Colloque C4, supplLment au no 4, Tome 40, avril 1979, page C4-36

Neutron inelastic scattering experiments on uranium antimonide

G. H. Lander, W. G. Stirling (+) and 0. Vogt ( t t )

Centre d'Etudes Nucltaires, 85X, 38041 Grenoble, France and Argonne National Laboratory (*), Argonne, Illinois 60439, U.S.A.

( 7 ) Institut Laue Langevin, 156X, 38042 Grenoble, France

( t t ) Laboratorium fur Festkorperphysik, ETH, CH 8093, Ziirich, Switzerland

Rhsurnh. - Nous prksentons la premibre observation par diffusion inklastique de neutrons d'une excitation collective BlCmentaire dans un composk d'actinide mktallique. Cette excitation est polariske longitudinalement par rapport 2 la direction du moment magnktique. Les seuls autres modes magnktiques observks s e situent ii des Cnergies supkrieures B 6 T H z et ne prksentent pratiquement pas de dispersion.

Abstract. - We report the first observation by neutron inelastic scattering of a collective elementary excitation in a metallic actinide compound. The excitation is longitudinally polarized with respect to the direction of the magnetic moment. The only other magnetic modes observed are at energies above 6 THz, and are essentially dispersionless.

1. Introduction.

-

At the present time models for the behaviour of metallic actinide (Sf) systems are based on the (4f) systems. Thus, USb with the NaCl crystal structure (a, = 6.197

hi)

and the type-I magnetic structure (T, = 241 K) might be expected to have a similar electronic structure as NdSb, which has the same crystal (a, = 6.235

h;)

and magnetic (T, = 13.6 K) structure [I]. Both compounds would then be basically f3 configurations, developing almost the full free-ion moment of 3.27 p B at low temperature [2] (2.98 p B in NdSb and 2.82 p B in USb). The aim of the present experiment is to measure the microscopic exchange and crystal-field interactions.

2. Experimental and results. - The experiments were performed on a crystal assembly, consisting of six oriented single crystals (total volume of 0.2 cc), with the triple axis spectrometers IN8 and IN3 at the High Flux Reactor, Institut Laue Langevin, Grenoble.

The type-I antiferromagnetic structure [2] of USb consists of ferromagnetic (001) sheets stacked in the simple alternating sequence

+

^-

+

-

.

The moments are parallel to [OOl] and the crystal contains three domains corresponding to the moments being aligned along each cube axis. In the present experiments, since most of the measurements have been done at or near the X points [I101 and [OOl], where the structure factors for two out of three domains are very small, the excitations may be identified with a single (001) domain [I]. In an exact analogy with the critical scattering experiments [3] on USb, the excitations at wavevectors [0, 0, 1

+ 51

are transver-

(*) Work supported by the U.S. Department of Energy.

U S b 8 K

4

PG (002)-PG (002)

I

'

^[1,1,01

Fig. 1.

-

(a) Constant Q scan at the [I, 1,0] point. ( b ) Constant Q scans at and near the [0, O,1] point. ( c ) Constant energy scan at 1 THz around the [I, 1,0] point.

se only, whereas those at [I

+

5, 1

+

(,0] consist of both transverse and longitudinal components. The first important observation is shown in figure 1, in which the excitation at the X point is clearly seen at [ l l O ] (Fig. 1 a), but not at, or nor, [001] (Fig. 1 b).

We conclude that the collective excitation is almost entirely longitudinal in nature.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1979413

(3)

NEUTRON INELASTIC SCATTERING EXPERIMENTS ON URANIUM ANTIMONIDE C4-37

The dispersion relations as measured to date are rate with the zone boundary phonon at [I101 with an summarized in figure 2. The following points are energy of 1.53 ²0.05 THz. In figure l c we show a

important : constant energy scan at 1 THz through the [I101

1) Two types of magnetic excitations are found.

At low energies we see a longitudinal mode with considerable dispersion. At higher energies we see at least two broad, and essentially dispersionless, magnetic modes. The polarization of these modes is not simply longitudinal, since they are also seen at wave

position. This confirms that no lower branch of the magnetic excitation exists.

3) At wave vectors not too close to the point X the magnetic excitation has a dispersion of (6.7

+

0.2) THz -

A

in both the [&&0] and [00&1 directions.

vectors [O, 0,

&I.

These high-energ~ modes have 3. Discussion.

-

The most important point to very little temperature dependence. emerge from the present studies, which are still in 2) he collective excitation is Completely degene- progress, is that the dynamic behaviour of the collective excitation in USb cannot be understood with simple crystal-field theory. This can be demonstrat- ed by recalling that the polarization of the wave functions necessary to obtain the full ordered moment tends to produce eigenvectors composed main- ly of single I M, ) components, as for example in the heavy rare-earth metals. Since a longitudinal excitation corresponds to a J, matrix element between similar components, this situation occurs when the exchange is small, as for example [4] in Pr. Usually the longitudinal modes are very difficult to observe [ l , 51.

Our inability to understand the dynamics does not mean that the crystal-field interactions are negligi- ble. The dispersionless high-energy modes at

-

6 THz may well be related to single-ion excitations.

The magnitude of the energy splitting,

-

^{6 THz, is}

almost identical to that between the T:' and T , states predicted by our form-factor measurements [2], so that both the static and dynamic measurements show that an excited state exists at

-

6 THz above the ground state.

In discussing the spin correlations [3] observed in the critical regime in USb we introduced the idea of a molecular orbital approach involving the hybridi- zation between the magnetic 5f electrons and the anion p orbitals. The excitation observed in USb at low temperatures would seem further evidence for the need for such a model. The excitation is truly Ising in nature corresponding to correlations in the J,

r 0.2 0.4 0.6 0.8 x 0 8 0.6 0.4 0.2 components.

kJO <I

^{[ E E}

01

Acknowledgments.

-

We are grateful to K. Mat-

tenburger for help in growing the crystals and to

Fig. 2. -The dispersion curves for USb ; energy plotted against

wave vector transfer Q (in units of 2 ~ / a ) . The dashed lines KnOtt and M- Henry them. It is

revresent the vhonon disversion and are based on the measured a wleasure to thank W. J. L. Buvers. - .B. R. C O O V ~ ~ , - . open points as well as our knowledge of phonons in NaCl

T.

M. ~ ~ land d S. ~K. ~sinha for a nu&,er , of

structures. The magnetic modes are represented by solid squares

(the collective excitation) and the hatched boxes (dispersionless We thank the Institut Laue Lan-

excitonic levels). gevin for the use of its facilities.

References

[I] FURRER, A., BUYERS, W. J. L., NICKLOW, R. M. and VOGT, [4] HOUMANN, J. G., CHAPPELLIER, M., MACINTOSH, A. R., BAK, O . , Phys. Rev. B 14 (1976) 179. P., MCMASTERS, 0. D., and GSCHNEIDER, K. A., Phys.

[2] LANDER, G. H., MUELLER, M. H., SPARLIN, D. M. and VOGT, Rev. Lett. 34 (1975) 587.

O . , Phys. Rev. B 14 (1976) 5035. [5] HOLDEN, T. M., SVENSSON, E. C., BUYERS, W. J. L. and [3] LANDER, G. H., SINHA, S. K., SPARLIN, D. M. and VOGT, O., VOGT, O., Phys. Rev. B 10 (1974) 3864.

Phys. Rev. Lett. 40 (1978) 523.