TRANSVERSE SPIN FLUCTUATIONS IN γ - Mn (37 % Fe) ALLOY

(1)

HAL Id: jpa-00229030

https://hal.archives-ouvertes.fr/jpa-00229030

Submitted on 1 Jan 1988

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.

TRANSVERSE SPIN FLUCTUATIONS IN γ - Mn (37

% Fe) ALLOY

J. Milczarek, K. Mikke, E. Jaworska

To cite this version:

(2)

JOURNAL DE PHYSIQUE

Colloque C8, Suppl6ment au no 12, Tome 49, dkembre 1988

TRANSVERSE SPIN FLUCTUATIONS IN

7 -

Mn

(37

%

Fe) ALLOY

J. J. Milczarek, K. Mikke and E. Jaworska

Institute of Atomic Energy, Swierk, 05-400 Otwock, Poland

Abstract. - The neutron scattering studies of the sublattice magnetization in the vicinity of the N&l temperature in 7

-

Mno.6Feo.37Cuo.03 alloy reveal its marked critical behaviour. The staggered transverse susceptibility and correlation length remain finite at TN and some anisotropic features of critical scattering are observed.

Although the critical magnetic scattering in y-Mn alloys was studied for the first time quite a long time ago [I, 21, there was no continuation of these studies. In view of the growing interest in the phase transi- tion phenomena in 7-Mn alloys [3, 41 we would like to present the results of our neutron scattering studies for the Mn0,~Feo.37Cuo.o3 alloy. The main aim of this work was to determine the temverature behaviour of the sublattice magnetization and the spin fluctuations in the vicinity of the NBel temperature.

The sample was a 10 x 10 x 10 mm cube produced by the Bridgrnan method. The addition of 3 % Cu improves the crystal quality and facilitates the crystal growth. The FWHM of the crystal rocking curve was N 2'. The measurements were performed with the triple axis neutron spectrometer for the incident neutron energy Eo = 50 meV and for collimations of 45' and 50' before and after the sample, respectively. Most measurements were performed for the energy transfer A E = 0 but some scans were taken in the double- crystal arrangement (with analyzer crystal removed). The sample was mounted in the vacuum cryostat with the heating coil wound on the cold finger. The temperature was stabilized within

f

0.1 K and the temperature gradient on the sample did not exceed 0.6 K. The magnetic structure of the investigated sample is in good approximation of AF1 type allowing in princi- ple the studies of both longitudinal

x""

(q) and transverse

xXx

(q) (in respect to the magnetic ordering axis) components of the generalized susceptibility tensor [I]. The temperature dependence of the main magnetic re- flection (110) allowed the determination of the NBel temperature TN = 470 K and the critical exponent

p =

0.38 f 0.02. Unfortunately the longitudinal (parallel to the direction of the spin ordering) fluctuations could not be studied with desired accuracy because of very weak critical scattering detected in the vicinity of (110) Bragg peak. However, the critical scattering easily detected near the (100) rlp, allowed detailed analysis of the transverse fluctuations. The temperature dependence of I(loo) is shown in figure 1. We have performed two

Grids

of scans with the scattering vector q = Q - - T ( ~ O ~ ) parallel and perpendicular to the r(100). Both types of I (q) distributions were fit-

Fig. 1. - The temperature dependence of the intensity at

(100) Bragg point. Open circles come from AE = 0 scans, black ones from scans taken in double-axis arrangement.

ted with the convolution of standard critical scattering cross-section

with four dimensional resolution function of the neutron spectrometer (n denotes the inverse of the correlation length

[,

w is the energy transfer). The energy linewidth

r

was assumed in the dynamic scaling form

with the dynamic scaling function f r ( x ) given by RBsibois and Piette [5]. From our fits we have found that the constant C = 100 meV is the most suit- able for all scans. It was found that the n values es- timated in such fits depend on the scan direction of

q. The results, presented in figure 2, for n, ( a

=jl

for

q

11 T (1001

and a =

I

for q l r ( l O O f ) indicate that the correlation length in the (100) direction

Ell

is about 10 % bigger than that in the (010) direction t L , for

T

>

T N . At TN both correlation lengths remain finite:

6

= 24 f 1

A

and [I = 21 f 1

A.

One should notice the different slope of nil (T) and KL ( T ) below

TN.

Be-

(3)

JOURNAL DE PHYSIQUE

Fig. 2. - The inverse correlation length dependence on

temperature.

cause of scarce data we could not estimate the power law parameters for the transverse fluctuations correlation length (transverse fluctuations critical tempera ture Tl and the critical exponent v l ) -only the linear fit seemed to be reasonable.

In view of the marked anisotropic character of the correlation length we have studied the intensity dis- tribution in the (001) scattering plane. The measurements were performed for Eo = 25.9 meV with collimations 10' and 20' before and after the sample, respectively. The results shown in figure 3 reveal the addi- tional feature of the critical scattering: apart from the alongation of isointensity contours along [010] direc-

Fig. 3. - Constant-intensity contours near (100) rip. No-

tation for q, : a = x for q Ilr(loo) and a = y for qlr(loo).

tion, the small cups pointing at [I101 type directions were discovered. This kind of the critical scattering anisotropy could be dealt with the anisotropic carrel* tion functions of the type introduced by Kocinski and Marzec [6] but for detailed comparison the temperature dependence of the constant-intensity contours is needed. Since for such correlation functions the sub- stantial shift of the temperature maximum of critical scattering should be observed [7] for larger q values, we have performed the measurements of the temperature dependence of the intensity at (qz, q,) = (0,0.2)

A-'

but no such effect was found.

The discovered anisotropy of transverse fluctuations could be attributed to the strong anisotropy of spin interaction which produces the large energy gap in the spin was spectrum (7 meV at 300 K). However, the observed anisotropy

(ell

>

[I)

is at difference with that found for the longitudinal fluctuations in the cubic USb compound [8] with AF1 structure, where

tII

<

tL.

This feature may stem from the different character of the magnetic anisotropy in both systems.

Aknowledgements

We are grateful to Mr. J. Zoladek for experimental assistance.

This work was performed under the contract CPBP 01.12.

[I] Ishikawa, Y., Endoh, Y. and Ikeda, S., J. Phys.

Soc. J p n 35 (1973) 1616.

[Z] Ikeda, S. and Ishikawa, Y., J. Phys. Soc. Jpn 39

(1975) 332.

I31 Jo, T. and Kunitomo, H. J., J. Phys. Soc. Jpn 55 (1985) 2017.

[4] Long, M. W. and Yeung, W., RAL-87-035 (1987). [5] R6sibois, P. and Piette, C., Phys. Rev. Lett. 24

(1970) 514.

[6] Kocinski, J. and Marzec, W., Physica 86-88B (1977) 1105.

[7] Kocinski, J. and Wojtczak, L., Critical Scattering Theory (PWN-Elsevier) 1978.