HAL Id: jpa-00217862
https://hal.archives-ouvertes.fr/jpa-00217862
Submitted on 1 Jan 1978
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.
CRYSTAL-FIELD EFFECTS IN (La, Nd)Sn3
C. Bredl, F. Steglich, W. Schmid, E. Umlauf
To cite this version:
C. Bredl, F. Steglich, W. Schmid, E. Umlauf. CRYSTAL-FIELD EFFECTS IN (La, Nd)Sn3. Journal
de Physique Colloques, 1978, 39 (C6), pp.C6-882-C6-883. �10.1051/jphyscol:19786393�. �jpa-00217862�
JOURNAL DE PHYSIQUE Colloque C6, supplément au n" 8, Tome 39, août 1978, page C6-882
CRYSTAL-FIELD EFFECTS IN ( U - N d ) S n3
C D . Bxedl, F . S t e g l i c h , W. Schmid t and E. Umlauf t
II. Phys. Inst-itut Universitat Koln u. SFB 126., D-S KoVn, Germany t Zentrdlinst. f. Tieftemperaturforschung, D-8046 Garaking, Germany
Résumé.- Des mesures de chaleur spécifique et de la résistivité électrique des alliages (La,Nd)Sn.
à l'état normal sont présentées. En accord avec des résultats récents sur les propriétés supracon- ductrices, nos résultats confirment l'absence de l'effet Kondo et peuvent révéler les niveaux de champ cristallin de Nd3 dans LaSn,.
Abstract.- We report measurements of the specific heat and the electrical resistivity of (La,Nd)Sn.
alloys in the normal state. In agreement with recent results on the superconducting properties our data confirm the absence of the Kondo effect and can reveal the crystal field scheme of Nd3 in LaSn .
Schmid et al. / I / have recently started to investigate the superconducting properties of (La, Nd)Sn3. From these experiments they conclude Nd3
to possess a stable magnetic moment when diluted in LaSn,. This is very remarkable since in the related
(La, Pr)Sng and (La, Sm)Sn, alloys Kondo anomalies have been observed e. g. in the electrical resisti- vity / 2 , 3, 4/.
In this paper we present the first results of the specific heat and the resistivity o f U B ^ S n
(1 at%<_x^7 atZ) in the normal state between 0.4 K<^T£14 K and in external magnetic fields up to 5 tesla. These results will i) support the absen- ce of the Kondo effect and ii) reveal the CF-sche- me in (La, Nd)Sn3. The latter is so far only known for the NdSn, compound and there consists of a rg
ground state doublet and two excited r„ quartets,
o
seperated from V. by 22 K and 117 K, respectively /5/.
Figure 1 shows for La#g3Nd>nySn3 the magnetic contribution to the specific heat in atomic units vs. temperature on a logarithmic scale in different magnetic fields B. At B = 0 we observe a broad peak around 5 K, a second but incomplete anomaly with a peak below our temperature limit of 0.4 K, and a small jump at T . Upon application of B-fields the lower peak is shifted to higher temperatures. For B > 3 tesla it cannot be resolved from the anomaly at 5 K. The analysis of the specific heat data, in- cluding other Nd-concentrations and other B-fields not shown in figure 1, allows to determine an en- tropy SM,/k„ = An 2, corresponding to the lower
peak, whereas the extrapolation to higher tempera- tures, is consistent with a value of Jln6 for both peaks.
1 i 1 r — 1 — 1
B=
La Nd Sn / T N
C . -*-0 T .93 .07 3 //"~^\
- -~,— Schottky ///?/ ^'"'- ~
0.5 - 4=9-
9K(s-^—Z/J^
rV
0L 1 „„.~*^——^i—. 1 — 3 : —
0.3 1 3 T ( K ) 10
Fig. 1 : Specific heat per Nd-ion in units of k as function of T in various magnetic fields B.
Dashed lines : calculated Schottky-anomalies with different splitting energies A of the rg ground state
Thus, our data can be qualitatively described by a level scheme consisting of a Zeeman split T doublet as ground state and a degenerate r„ quar- tet with an excitation energy of 14 K. The sequen- ce of the CF levels is the same as determined in NdSn, by neutron scattering /5/. However, for our dilute alloy the energy scale of the CF splitting is reduced by a factor of 14/22. Compared to the calculated Schottky-anomaly at B = 0 which implies a Zeeman split Tg with a splitting energy of h = 0.9 K (figure 1) the experimental peak is con- siderably broader. This implies that there is a distribution of molecular fields due to the Nd-Nd interactions. We note, that this broadened peak is much sharper than a Kondo anomaly in corresponding magnetic fields /6/.
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19786393
Fig. 2 : Resistivity p vs. T for LaSn3 and (g,Nd) Sn3 alloys. Data points below Tc were obtained at B = 0.4 tesla and are shifted by 0.003 @cm to cor- rect for the positive magneroresistance of LaSn3 (at LaSn3 I tesla must be applied below 5 K, with a shift of 0.016 ~Qcm). The inset shows the incre- mental resistivity pNd = &(pLa -P
Nd Sn3 La Sn3) 1-x x
In figure 2 the normal state resistivity p(T) of LaSn3 and the alloys with 3 and 7 at% Nd is plot- ted on expanded scales vs. temperature. The promi- nent features of this plot are : (i) an increasing resistivity with the Nd-concentration, (ii) a strong temperature dependence of p(T) below 6 K, where LaSn3 has reached the residual resistivity,
(iii) a significant change in the curvature of p(T) with the Nd-concentration. The variation of the in- cremental resistivity per at% Nd, pNd with tempera- ture is shown in the inset. First we note that no Kondo anomaly is vislble in the resistivity, either.
For the 3 at% Nd sample, pNdcanbe qualitatively understood as the resistance anomaly due to CF splitting of the single Nd-ion 171-: the "plateau"
around 10 K marks the depopulation of thchighest CF-level. This gives a rough estimate of Q 70 K for the excitation energy of the uppej:
r8
level in di- lute (La, Nd)Sng. Considering the p(T) data in addi- tion to the specific heat results we conclude that in our diluted alloys, Nd3+ lias the same CF level scheme as in NdSng i.e. the same LLW-parameter x /8/but the splitting energy is smaller by a factor of about 2/3. A second plateau in pNd is expected at around 2 K corresponding to the first excited T8 le- vel (at 14 K). However, a further decrease of p(T) is found at the low temperature end, presumably due
to the Nd-Nd-interactions. These interactions beco- me much more important for the 7 % Nd sample. The
typical negative magnetoresistivity is superposed to the single-ion CF resistivity, which is there- fore smeared out. In addition, a 2 % Gd sample was
~easured, which reaches a constant residual resisti- vity at 6 K and shows no anomaly due to the "reverse"
Kandoeffect, as observed in
(5,
Gd)A12 191.Therefore, this effect is not important for the in- terpretation of our p(T) data, either.
To summarize, experiments on normal state (gNd)Sn confirm the conclusion from superconduc-
3
ting properties /I/ that in this system the Kondo effect is not relevant. In the dilute alloys the Nd3+ ions exhibit the same CF level scheme but with splitting energies reduced by a factor of 2/3 compared to the NdSn3 compound.
References
/I/ Schmid, E., Umlauf, E., Bredl, C.D., and Steglich, F., subm. to J. Physique (LT15) /2/ Lethuillier, P. and Haen, P., Phys. Rev. Let-
ters
2
(1975) 1391/3/ Abou-Aly, A.I., Bakanowski, S., Berk, N.F., Crow, J.E. and Mihailisin, T., Phys. Rev. Let- ters
35
(1975) 1387/4/ Bakanowski, S., Crow, J.E. and Mihailisin, T., Sol. State Connnun.
22
(1977) 241151 Lethuillier, P., Pierre, J., Knorr, K. and Drexel, W., J. Physique
36
(1975) 329 161 Schotte, K.D. and Schotte, U., Phys. Lett.55A (1975) 38
-
/7/ Hirst, L.L., Sol. State Commun. z(1967) 751 /8/ Lea, K.R., Leask, M.J.M. and Wolf, W.P., J.
Phys. Chem. Solids
23
(1962) 1381/9/ Steglich, F., in : Festkiirperprobleme (Adv.
in Solid State Phys.) Vol. XVII, P 319, ed.
J. Treusch, Braunschweig, Vieweg (1977)
