ELECTRONIC STRUCTURE OF FeII IN MAGNETIC THIOSPINELS

(1)

HAL Id: jpa-00218682

https://hal.archives-ouvertes.fr/jpa-00218682

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 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.

ELECTRONIC STRUCTURE OF FeII IN MAGNETIC THIOSPINELS

H. Oudet, L. Brossard, J. Dormann, P. Gibart, L. Goldstein

To cite this version:

H. Oudet, L. Brossard, J. Dormann, P. Gibart, L. Goldstein. ELECTRONIC STRUCTURE OF FeII IN MAGNETIC THIOSPINELS. Journal de Physique Colloques, 1979, 40 (C2), pp.C2-258-C2-260.

�10.1051/jphyscol:1979291�. �jpa-00218682�

(2)

JOURNAL DE PHYSIQUE Colloque C2, supplément au n° 3, Tome 40, mars 1979, page C2-258

ELECTRONIC STRUCTURE OF F e1 1 IN MAGNETIC THIOSPINELS

H. Oudet, L. Brossard, J . L . Dormann, P. Gibart and L. Goldstein

Labovatoire de Magnetisms, CN.R.S., 1, Place Aristide Briand, 92190 Bellevue, Franee

Résumé.- Les composés Mni_xFex7Cr2Sitet Cdi_xl?ex7Cr2Si» avec x = 0.01 et x = 0.005 ont été étudiés par spectroscopie Mossbauer. Les résultats présents diffèrent du précédent travail de Van Diepen et Van Stapele /3/. 0 et n ont été trouvés différents de zéro et une distribution de sites apparaît pour T > 30K. Ces caractéristiques peuvent être reliées à la variation anormale en température du para- mètre de réseau des deux composés.

A b s t r a c t . - The compounds l*u_xFex7Cr2Su and C d i - x F e ^ C r ^ w i t h x = 0.01 e t x = 0.005 were studied by Mossbauer spectroscopy. The present r e s u l t s depart from the previous work of Van Diepen and Van

Stapele / 3 / . 6 e t n were found d i f f e r e n t from zero and a d i s t r i b u t i o n of s i t e s appears for T > 30K.

These features may be r e l a t e d to the unusual temperature dependence of the l a t t i c e parameter of the two compounds.

1 . I n t r o d u c t i o n . - In a previous work on the ferrima- gnetic t h i o s p i n e l FeCr2Si,, we showed t h a t the tempe- r a t u r e dependence of the hyperfine parameters cannot be described in a localized model / l / . To set i n f o r - mations on the e l e c t r o n i c s t r u c t u r e of F e1 1 in A s i t e in t h i o s p i n e l , we studied by Mossbauer spectroscopy the following compounds :

Cd. Fe5 7Cr2S», Mn. Fe5 7 C^S* with x = 0.01 and

1- x x J -x x

x = 0.005

The single ion anisotropy due to d i l u t e d Fe^* i n CdCr2S^ was studied t h e o r e t i c a l l y and experimentally by Hoekstra e t a l . / 2 / . A Mossbauer study s t a r t i n g from these r e s u l t s was carried out by Van Diepen e t a l . / 3 / on Cdo.9sFeo702Cr2S^. Between 20 and 55K, a broadening of the Mossbauer peaks i s observed, however t h i s broadening was not c o r r e l a t e d with any de- v i a t i o n from the a x i a l symmetry of the e l e c t r i c field gradient nor with a non zero angle between H, . and V . The s t a t i s t i c s of t h e i r s p e c t r a did not allow zz r

to point out these f e a t u r e s . To avoid these broadenings of the peaks we used the lowest concentrations of Fe5 7 consistent with well resolved Mossbauer s p e c t r a , i . e . x = 0.01 and x = 0.005. At the same time, with these low concentrations, p a i r s e f f e c t s are lowered.

2..Experimental.-Data were taken from 4.2 up to 50K.

T = 86K for Cdo 99FeQ. oiC^S^ and T = 74 K for c • c Mno.99Feo>oiCr2Si(. Typical spectra are reported on figure I (MnCr2Si, : 0.01 Fe5 7) and figure 2 (CdCr2S^ : 0.01 Fe57).The present results differ fairly from the previously reported data /3/. The main features of the spectra obtained in this study are the following:

- For all temperatures and for the four samples two

sites with relative intensities 0.11 and 0.035 (for x = 0.01) and 0.06 and 0.02 (for x = 0.005) do exist (figure 3a and figure 3b).

- Furthermore, for T > 30K, spectra are solved assuming a distribution of hyperfine parameters which is simulated by several sites with hyperfine parameters close to the main site.

- For the main site, 0 and r\ are different from zero. In Cdo. 99Fe0. oiCr2Si,, 0 decreases from 16° at 4.2K to 12° at 50K , whereas n varies between 0.15 and 0.35. In the same temperature range in

Mno.99Feo.01 Cr2S>», 19° < 0 < 10° and 0.1 < r\ < 0.4.

3.Fe^I j[n tetrahedral site in magnetic thiospinel.- We used the crystal field theory which was applied to the ferromagnet CdCr^^ /3/ and we extend the calculation to the ferrimagnet MnCr2Sl(.

This lowest SE state is split by the exchange field and spin-orbit coupling described by the hamilto- nian :

51 = - A (M/M) . ? + A L.l

The first term results from the exchange field arising from magnetic chromium in CdCr2Si, and from chromium and manganese in MnCr2Sit. As a result, 5E is split in five doublets, then further split by second order spin-orbit coupling into ten singlets.

4. Theoretical fit ofH, ,(T)and V (T).-

• nr—; zz From the energy levels and the eigehfunctions, it is possible to calculate the mean value of H, „ and V . From the calculation, n = 0 and the main axis

zz

of the EFG is parallel to R (0 = 0) . The expres- sion of H^fCT) and V (T) are given in /3/.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1979291

(3)

2.73

-a o I

V

' ? rnrnl.

Fig. 2 : Mijssbauer spectra of Cdo.99Fe~lol~r2~~

To compare these relations with experimental data, Aex must be calculated as a function of T. A

ex (T) is proportional to the magnetization M(T) in a molecular field approximation (MFA)

.

In CdCrnSs, M(T) is easily calculated using the Brillouin function B312. In MnCrzSli the experimental curve M(T) presents a maximum around 40K. TO fit correctly the experimental curve, we calculated M(T) in a MFA with two exchange interaction JAB and JBB 141. Following Menyuk and Dwight / 5 / , we took JBB = -4 JAB. Using T = 74 K, we got JAB = -9 em-', JBB = 36 cm-I and calculated M(T).

~lthough 0 and 0 are different from zero, we tentatively applied the Van Diepen's model to fit the experimental temperature dependence of E(T) = e and

%f

(T).

The parameters which give the best fits are : K = 6 (-

x

+p) = 21 cm-', Ae,(0) = 50 cm-I for

A CdCrzS4 and

K = 17 cm-I and A (0) = 20 cm-I for MnCrnS,,.

ex

The figure 3 shows the experimental values of H hf (T) and E(T) in Mno .gs~epZ O I C ~ ~ S ~ and the calculated Fig. 1 : ~Zssbauer spectra of M n o . 9 g ~ e ~ l o l ~ r Z ~ 4 values of \f(T) and E ( T ) ( ~ d o . s s ~ e ~ ~ o l ~ r z ~ , , leads at different temperatures. Solid line : fitted cur-

ve; vertical bars : experimental data. to similar results / 6 / .

(4)

JOURNAL DE PHYSIQUE

- -

Fig. 3: Variation of Hhf (T) and E(T) in ~ n o . 9 9 ~ e a T o 1 ~ r i ~ s Solid line : theoretical curve calculated according to 15 4; dotted lines : experimental curves; dashed lines : extra sites.

5. Discussion.- 5.1.- The two extra sites are easily 6 . Conclusion.- A single ion crystal field scheme interpreted when considering neighbouring effects. of FeI1 in A site with an exchange field calculated The experimental intensities of these extra sites in MFA gives a fair agreement with experimental are in a good agreement with the probability of fin- data.

ding one Fe nearest neighbour or one Fe next nearest However some features

(@,n

# 0, appearence of neighbour. The

Hhf

of these extra sites is greater a distribution of sites for T > 30K) are not explai- thanthe Hhf of the main site : this can be explained ned. These features may be related to the unusual

assuming an exchange Fe-Fe interaction greater than temperature dependence of the lattice parameters of

Mn-Fe

.

the thiospinels. Furthermore, an Heisenberg Hamilto-

5.2.- The Van Diepen analysis which nian is not adequate to describe the exchange inte- fairly explained the temperature dependence of

Hhf

raction between dn ions with orbital degeneracy /8/.

and E for both compounds does not explain the de- parture from axial symmetry of the EFG, the exis-

tence of non zero 0 and the appearence of a distri- References bution of hyperfine parameters for T > 30K. This

/ I / Brossard, L., Thesis Paris-Sud (1977).

can be related to the existence in these compounds

/2/ Hoekstra, B., Van Stapele, R.P. and Voermans, either of exchange striction or strains which indu- A.B., Phys. Rev. B,

5

(1972) 2762.

ce an orthorombic component of the crystal field. _{/ 3 /} Van Diepen, A.M. and Van Stapele, R.P., Phys.

Another possibility is to consider coupling between Rev. B,

5

(1972) 2462.

phonons and electronic states of ~ e " diluted in ^{/ 4 /}This is slightly incorrect, JAA is non zero. In present case a fair agreement only is necessary the thiospinel matrix. to fit the experimental Miissbauer data. For more

G b e l 171 showed that CdCr9Sb expands anoma- details about the magnetic structures of kCrzSr see for instance Castets, A., ThSse 3Gme cycle lously below IOOK; Kleinberger (private communica- (1973); Mejai, M., These 31me cycle (1977).

tion) got similar results on MnCr2St,. Furthermore _{/ 5 /} Menyuk, N., and Dwight, K., J. Appl. Phys.

36

&be1 /7/ found unusual broadenings of the diffrac- (1965) 1088.

tion peaks at low temperature and an important 161 More details may be found in Oudet, H., Thesis Paris VII (1978).

variation of the u parameter which changes the atomic

distances and bonding angles. These features may be ^{/ 7 /} Gbel, H., J. Magn. and MM.

3

(1976) 143.

related to the two possibilities suggested below. /8/ Hartmann-Boutron, F., J. Physique

2

(1968) 212.