Structural instability in new magnetic heusler compounds

Structural instability in new magnetic heusler compounds

Solid State Communications,Vol. 18, pp. 423—425, 1976. Pergamon Press. Printed in Great Britain STRUCTURAL INSTABILITY IN NEW MAGNETIC HEUSLER COMP...

285KB Sizes 2 Downloads 44 Views

Solid State Communications,Vol. 18, pp. 423—425, 1976.

Pergamon Press.

Printed in Great Britain

STRUCTURAL INSTABILITY IN NEW MAGNETIC HEUSLER COMPOUNDS J.C. Suits IBM Research Laboratory, San Jose, CA 95193, U.S.A. (Received 17 July 1975 by H. Suhi) The crystal structure and magnetic properties have been determined for a new series of compounds of the form Rh2 TSn. For T = Mn, Ni, or Cu, the room temperature structure is the fully ordered cubic Heusler structure. For T = V, Cr, Fe, or Co, a new structure is observed which is an excep. tionally large tetragonal distortion of the Heusler structure (c/a = 1.18— 1.27). The appearance of this tetragonal distortion is attributed to an electronic instability of the band Jahn—Teller type. 1 proposed a model to exIN 1966, Labbétoand Friedel structural transformation in plain the cubic tetragonal Al 5 type intermetallic compounds. In this model, lattice deformation lifts the degeneracy of d electron sub-bands causing a lowering of the energy of the system by electron redistribution. In the Al 5 compounds the 3d band structure is quasi one-dimensional. For the particular compound V3Si the deformation amplitude e = c~/aCUb~ 1 = 0.2%, with the cubic—tetragonal 2More recently, a transition temperature Tm 30°K. cubic—tetragonal transformation was observed in LaAg~In 3 In this system the cubic structure 1_~ is CsCl, with ealloys. 5% and Tm 200°K. This letter will describe a new tetragonal structure —

which appears among certain compounds of the group Rh2 TSn where T is a 3d transition element. The deformation amplitude for some of these compounds is as large as 17% and transition temperatures may be as large as 700—800°K.In addition to being new magnetic compounds and displaying one of the largest deformations yet observed, this is the first time a cubic—tetragonal instability has been observed in a system of compounds which allows a wide substitution of atomic species. The compounds were made by mixing together powders of the elements, the at mixture ated quartz ampoules andsealing annealing 250 C inforevacu7 h, followed by 700°Cfor 6 days. The samples were then pulverized and reannealed in vacuum for 5 days at 700°C. For T = Mn, Ni, or Cu the structure observed at room temperature is the fully ordered L21 Heusler structure shown in Fig. 1. The Rh atoms occupy the corners of a simple cubic sub-lattice. The atoms of the second simple cubic sub-lattice occupy the body center positions of the first sub-lattice. The second sub-lattice is occupied by T and Sn atoms ordered as shown. For T = Cr, Fe, or Co a new structure is observed. Analysis of X-ray powder diffraction data shows that

this structure is an exceptionally large tetragonal distortion of the L2 1 structure. This distortion shows no rearrangement of atoms and no motion of one set of atoms relative to another set of beyond a simple elongation of the unit cell along (001> and a corresponding shrinking along (100> and (010>. The tetragonal structure is P42 /ncrn (origin at center 2/rn) with Rh atoms at 4b sites and 4e sites (z = 1/4), Tatoms at 4d sites, and Sn atoms at 4cX-ray sites.diffractometer Figure 2 showspattern a comparison of an experimental with a pattern calculated for the above structure. The agreement is seen to be good, even down to fine detail. The presence and magnitude of the line at 20 = 24.8°indicates that the T and Sn atoms are fully ordered as in the L2 1 structure. For T = V the room temperature structure is tetragonal for a sample which has been slowly cooled, and is a two phase mixture ofL21 and tetragonal for a sample which has been quenched. Table 1 gives the lattice constants, c and a, for the Rh2 TSn compounds. To allow direct comparison, the lattice constants of the tetragonal compounds are given in terms of the 14/rnrnrn space group of the cubic L21 compounds. The c/a ratios indicate distortion which is comparable or larger than that observed in tetragonal 3~or Cu21.4 spinels Mn For T =containing V one hasthe theJahn—Teller opportunityions to compare the umt cell volume V 0 for the cubic and tetragonal structures in the same compound. It is seen that the large distortion occurs with essentially no change in atomic volume. Table 1 also gives ferromagnetic Curie temperature T0 and magnetic moment per formula unit ~ measured at 4.2°K,and also paramagnetic Curie temperature 0 and magnetic moment hi derived from reciprocal susceptibility vs temperature measurements. The data show that the T = Mn, Fe, Co and probably Ni compounds are ferromagnetic. The T = V and Cr compounds show only a very small net magnetic moment. The change in moment p with composition are suggestive of simple




Vol. 18, No.3

Table 1. Structure and magnetic properties of Rh2 TSn T

V t1~ L2

Structure (at 20°C)


a(A) c(A) c/a t2~(A3) v0 0(K) T 0(K)

6.192 6.192 1 238

— —














4.062 7.291 1.27 241

4.093 7.162 1.24 240

6.242 6.242 1 243 413

4.150 6.912 1.18 238 595

4.114 6.906 1.19 234

6.136 6.136 1 231

6.146 6.146 1 232


412 3.10 3.30

583 3.70 3.68

444 2.44


— — —

/~L1 (hBY~


— —

— —


0.0 —

(1) Tet = new tetragonal structure, L2 1 = Heusler structure. (2) Unit cell volume. (3) Moment per formula unit measured at 4.2°Kand 66 kOe. (4) Moment per formula unit calculated from 2~/(p/2)(1+ p). •Rh


: : :~~~~



Fig. 1. Heusler structure. filling of a 3d band. It should be remembered that some moment will probably reside on the Rh atoms as for 5 example in FeRh.alternation of structures with cornThe curious position shown in Table 1, as well as the coexistence of both structures in quenched Rh 2 VSn, suggests that t1~e cubic and tetragonal structures are closely related. Additional evidence for this the dependence of structure on temperature. Since one of the consequences of a band Jahn—Teller effect is a tetragonal cubic transition, a sample of Rh2 CoSn was heated on a high temperature X-ray diffraction stage. A tetragonal cubic transition was found and Fig. 3(b) shows a graph of the peak intensity of the strongest line of the cubic structure as a function of temperature. At the maximum temperature obtained (850°K),only about one-third of the tetragonal structure was transformed to the cubic structure. The structural transformation exhibits large hysteresis and is reversible on temperature cycling, Since the transformation temperatures here are well above the Curie temperature of the compound (444°K), —~











Fig. 2. Experimental and calculated polycrystalline X-ray diffractometer patterns. (Although not described in the text, Rh2MnSb also a new compound is closely related to the Rh2 TSn series, and does show the same tetragonal structure). —

this effect is not due to ~spin—orbit type mechanism. The c/a ratio of the untransformed tetragonal structure is also shown in Fig. 3. The decrease in this c/a with increasing temperature is large, and in contrast to the structural transformation, shows no hysteresis with temperature. In the band type of Jahn—Teller effect, bands which are degenerate in the cubic system are no longer degenerate in the tetragonal system, allowing a repopulation of electrons toward the lower energy bands. In contrast to the usual Jahn—Teller effect which shows a splitting of energy levels, in the band Jahn—Teller effect the main

Vol. 18, No.3 1.19 1.17 c/a









creased orbital overlap respectively.’3 This allows a net transfer 3d electrons d~2_~2band. It is of suggested here to thatthethe large structural transformation in the Rh 2 TSn compounds is due to an elecI








tronic instability of the band Jahn—Teller type. This is based on the following facts: (1) the crystallographic structure of the tetragonal phase can be described as a simple deformation along (001>of the L21 structure without requiring any changes in occupation of atomic sites, (2) the atomic volume of the two structures are essentially the same, (3) a reversible tetragonal cubic transformation has been observed for Rh2 CoSn on

___________________________________ I











heating, and (4) a rapid decrease in c/a with increasing temperature observed. band at Jahn—Teller effectand is sensitive toisthe densityThe of states the Fermi level










it may be that Rh2 MnSn is cubic (Table 1) because of a mmimum in 3d density of states at this composition. Since this tetragonal instability has not been reported

o—J.o 500





Fig. 3. (a) c/a ratio of tetragonal Rh2 CoSn as a function of temperature; (b) peak height of the principal X-ray line of the high temperature cubic structure as a function of temperature. For both sets of data, increasing (decreasing) temperatures are represented by open (closed) circles,

before for other Heusler compounds, it seems likely that the lifting of the degeneracy of the 4d bands of the Rh, as well as the 3d bands of the T atom, also plays an important role in these compounds. In summary, a new series of compounds has been found of the form Rh2 TSn. For some of these com-

effect is a modification of the width of the energy

pounds a new structure is observed which is a large distortion of the Heusler structure. The appearance of this

levels a narrowing of bands derived from orbitals which overlap in the direction of crystal elongation, and —

a broadening of bands which overlap in the direction of contraction. For example, the 3d eg narrow band orbitals are degenerate in the cubic structure, but in the tetragonal structure, for c/a> 1, the d~2 band will narrow and the d~2_~2band will widen because of de—

tetragonal distortion is attributed to an electronic instability of the band Jahn—Teller type. Acknowledgements The author wishes to thank G. Guthmiller, C. Breen, and RK.Van Valkenberg for technical assistance and T. Huang and S. Liu for very helpful discussions. —


LABBE J. & FRIEDEL J., J. Phys. 27, 153 (1966). BATTERMAN B.W. & BARRETT C.S., Phys. Rev. Lett. 13, 390 (1964).


IHRIG H., VIGREN D.T., KUBLER J. & METhFESSEL S., Phys. Rev. B8, 4525 (1973).


GOODENOUGH J.B.,Magnetism and the Chemical Bond. Interscience, NY (1963).


BERTAUT E.F., DELAPALME A., FORRAT F., ROULT G., DE BERGERIN F. & PAUTHENET R.,J. App!. Phys. Suppi. 33, 1123 (1962).