2542
J. Phys. Soc. Jpn., Vol. 73, No. 9, September, 2004
K. TOMIYASU and I. KAGOMIYA
Table I. Experimental and best-fit calculated magnetic scattering inten-
sity. The multiplicity of the tetragonal domains is also given. The symbol
N.O. means that no reflection was observed within statistical uncertainty.
the 3d-metal chromite normal spinel systems MCr O4
2
(
M ¼ Mn, Fe, Co, Ni, Cu).
The magnetic ordering of FeCr2O4 is also described by
two magnetic components, the ferrimagnetic one and the
spiral one.
been determined yet.
FeCr2O4 is contracted with a tetragonal distortion of 3% at
hkl
Multiplicity
Experimental
Calculated
1
0,11)
0
01
10
01
12
2
4
8
8
N.O. (< 70)
499
0
533
423
100
The more precise magnetic structure has not
1
0,11)
1
2
1
However, since the lattice in
429
1
1)
N.O. (< 80)
4.2 K, the four B sublattices are expected to be grouped
into the B1–B2 and B3–B4 pairs, as for MnCr O , CoCr O
0
02 þ 200
11
20 þ 202
R-factor
2 þ 4
8
N.O. (< 80)
1430
0
1426
149
2
4
2
4
and CuCr2O4. We intend to reinvestigate the magnetic
ordering of FeCr2O4 in the future.
1
2
4 þ 8
—
N.O. (< 150)
The KDLM theory and the LKDM theory predicted some
magnetic structures by solving the Heisenberg Hamiltonian
—
1.0%
5
,7)
for the 3d-metal MCr2O4. However, a structure with the
unique B-sublattice pairs obtained in the present magnetic
[001]
5,7)
structure of NiCr O was not proposed. The reason may
4
2
be that a plausible magnetic structure of NiCr O was not
2
4
determined at that time. More detailed theory is required to
explain the magnetic structures of the 3d-metal MCr2O4
systematically.
0.26
0.78
0
.21
3.0
2.9
3.0
Anderson proposed the concept of magnetic geometrical
frustration coming from the tetrahedron (pyrochlore lattice)
composed by the B-sites for the cubic normal spinel
[
µ
B
]
2
+
3+
3+
A1-A2 Ni
B1-B3 Cr
B2-B4 Cr
100]
[
1
)
systems. The antiferromagnetic short-range correlation,
which is the sign of magnetic geometrical frustration, was
Fig. 9. The new magnetic structure model of NiCr2O4 below TS ¼ 31 K.
The transverse component exhibits collinear (only up or down)
antiferromagnetic ordering described by propagation vector Q ¼ ð0; 0; 1Þ.
1
7)
actually observed in other 3d-metal chromite ZnCr2O4.
However, the magnetic ordering of NiCr2O4 exhibits long-
range order below TS. The reason is most likely the large
tetragonal distortion of NiCr2O4. This distortion removes the
and Cr ions at their condition that all B–B distances and B–O–B angles are the
2
þ
3þ
magnetic moments of the Ni
6)
1
theoretical values of 3 ꢀB. For the magnetic form factors same, and completely suppresses the magnetic geometrical
2þ
3þ
2)
of the Ni and Cr ions, the values reported by Watson frustration, as discussed in the case of CuCr O by Prince.
2
4
4
)
and Freeman were used. The multiplicity of the tetragonal
domains for each scattering and the best-fit calculated
intensity are given in Table I.
Acknowledgment
We thank Professor K. Kohn, Professor K. Siratori,
Figure 9 shows the new magnetic structure model. This Professor Y. Tsunoda and Dr. K. Kamazawa for their fruitful
new magnetic structure possesses a spontaneous magnet- advice and discussion. We also thank Mr. H. Suzuki for his
ization of 0.3 ꢀB/formula, which is consistent with the support in the experiments. This work was supported by
magnetization measurements. The summation of the longi- Waseda University Grand for Special Research Projects
tudinal component of the magnetic moments at the A-sites is (No. 2003A-589).
larger than that at the B-sites, which is compatible with other
1
6)
magnetization measurements. The longitudinal component
the spontaneous magnetization) is perpendicular to the c-
axis, agreeing with a M o¨ ssbauer spectroscopy study of
1
2
3
)
)
)
P. W. Anderson: Phys. Rev. 102 (1956) 1008.
E. Prince: Acta Crystallogr. 10 (1957) 554.
I. S. Jacobs: J. Phys. Chem. Solids 15 (1960) 54.
(
6
1
15)
Ni.
4) R. E. Watson and A. J. Freeman: Phys. Rev. 120 (1960) 1134.
)
5
T. A. Kaplan, K. Dwight, D. H. Lyons and N. Menyuk: J. Appl. Phys.
32 (1961) 13S.
E. Prince: J. Appl. Phys. 32 (1961) 68S.
4
.
Discussion
6
)
)
The new magnetic structure model of NiCr O is
4
2
7
D. H. Lyons, T. A. Kaplan, K. Dwight and N. Menyuk: Phys. Rev. 126
characterized by the fact that the four B sublattices are
classified into the B1–B3 and B2–B4 pairs for the sponta-
neous magnetization along the ½100ꢃ direction (or the B1–B4
and B2–B3 pairs for the spontaneous magnetization along
the ½010ꢃ direction). The ferrimagnetic spiral ordering of
cubic MnCr2O4 and CoCr2O4 is described by the B1–B2 and
B3–B4 pairs, as shown in Fig. 2. Furthermore, the Yafet–
Kittel triangular magnetic configuration in the normal spinel
system CuCr O , whose lattice is contracted with a
(
1961) 540.
8) J. M. Hastings and L. M. Corliss: Phys. Rev. 126 (1962) 556.
N. Menyuk, K. Dwight and A. Wold: J. Phys. (Paris) 25 (1964) 528.
9
)
1
1
1
1
0) G. L. Bacchella and M. Pinot: J. Phys. (Paris) 25 (1964) 537.
1) G. Shirane and D. E. Cox: J. Appl. Phys. 35 (1964) 954.
2) R. Plumier: J. Appl. Phys. 39 (1968) 635.
3) E. F. Bertaut and J. Dulac: Acta Crystallogr., Sect. A 28 (1972) 580.
14) E. F. Bertaut and J. Dulac: Acta Crystallogr., Sect. A 36 (1980) 157.
1
5) J. Nakamura et al.: RIKEN Rev. 16 (1997) 25.
6) A. N. Goryaga, L. G. Antoshina, A. I. Kokorev and D. A. Chursin:
Phys. Solid State 44 (2002) 663.
1
2
4
tetragonal distortion of 9% at 4.2 K, is also expressed by
1
7) S.-H. Lee et al.: Nature (London) 418 (2002) 856.
8) S. Klemme and J. C. Miltenburg: Phys. Chem. Miner. 29 (2002)
663.
2
)
the B1–B2 and B3–B4 pairs. Thus, we consider that such
combinations with respect to the B sublattices is due to the
tetragonal lattice elongation (not contraction) of NiCr2O4 in
1