neglect of the exchange integral K or by considering the anti-
ferromagnetic interaction through the charge (electron) transfer
integral h(CT).26
is to pursue quantitative complementarity between optical and
magnetic properties; elucidating the disagreement between the
magnetic interaction derived from the magnetic susceptibility
measurements and that estimated from the variable temper-
ature absorption band intensity changes.
JAF = (ISF/ICT)(∆ECT)2∝{εmax(SF)/εmax(CT)}(∆ECT)2/ECT (4)
As in eqn. (4),27 the integrated intensity ISF may be approx-
Acknowledgements
imated by the product of the molar absorption coefficient(εmax
)
We gratefully acknowledge support of this research by a Grant-
in-Aid for Scientific Research (A)(2) (No.10304056) from the
Ministry of Education, Science and Culture.
and the half-bandwidth (∆1/2). Actually, since ∆1/2 of the MLCT
components are similar to those of the formally spin-forbidden
bands, the ratios of εSF/εCT are taken for simplicity instead of
ISF/ICT in the following discussion. The ratios of εSF/εCT are
almost constant as seen above, and the MLCT transition energy
ECT and the energy difference between the MLCT and the spin-
forbidden transition energy ∆ECT do not change with different
β-diketonates. This indicates that the value of
References
1 H. U. Güdel, in Magnetic–Structural Correlations in Exchange
Coupled Systems, ed. R. D. Willet, D. Gatteschi, O. Kahn, Reidel,
Dordrecht, The Netherlands, 1985, p. 297.
2 P. U. McCarthy and H. U. Güdel, Coord. Chem. Rev., 1988, 88, 69.
3 (a) C. Bellitto and P. Day, J. Chem. Soc., Dalton Trans., 1978, 1207;
(b) C. Bellitto and P. Day, J. Chem. Soc., Dalton Trans., 1986, 847;
(c) C. Bellitto and P. Day, J. Mater. Chem., 1992, 2, 265.
4 (a) C. Mathonieres, O. Kahn, J. C. Daran, H. Hilbig and F. H.
Kohler, Inorg. Chem., 1993, 32, 4057; (b) C. Mathonieres and
O. Kahn, Inorg. Chem., 1994, 33, 2103; (c) C. Cador, C. Mathonieres
and O. Kahn, Inorg. Chem., 1997, 36, 1923.
JAF∝{εmax(SF)/εmax(CT)}(∆ECT)2/ECT
is almost constant for different β-diketonates, though the Jobs
values vary considerably from ca. Ϫ9 to Ϫ99 cmϪ1 (Table 3).
That is, the antiferromagnetic interaction is not affected by the
coligands [eqn. (5)].
5 C. Benelli, A. Dei, D. Gatteschi, H. U. Güdel and L. Pardi, Inorg.
Chem., 1989, 28, 3089.
Jobs = JAF ϩ JF
(5)
6 A. Sokolowski, E. Bothe, E. Bill, T. Weyhermuller and K.
Wieghardt, Chem. Commun, 1996, 1671.
7 A. Caneschi, D. Gatteschi and P. Rey, Prog. Inorg. Chem., 1991, 39,
331.
Since the Jobs values consist of the antiferromagnetic and
ferromagnetic contribution (JAF ϩ JF) as in eqn. (5), the
variation of the Jobs values is associated with that of the JF
values. In other words, the coligand effect is operative in the
ferromagnetic interaction, but not in the antiferromagnetic
one. Therefore, the magnetic interactions of the bis(β-
diketonato)Cr() complexes mainly vary with the change of
the ferromagnetic coupling of the second term (JF) of eqn. (5),
in contrast to the case of NIT2py Ni() complexes where vari-
ation of the Jobs values is governed through the antifer-
romagnetic coupling of the first term (JAF). The reverse role in
the coligand effect on the Jobs values may be ascribed to the
difference in the respective magnetic orbitals [t2g(dπ) for Cr()
and eg(dσ) for Ni()]. This point will be discussed in more detail
in connection with the magneto-optical properties for two series
of Ni() and Cr() IM2-py complexes.28
8 I. Dasna, S. Golhen, L. Ouahab, O. Pena, J. Guillevic and
M. Fettouhi, J. Chem. Soc., Dalton Trans., 2000, 129.
9 D. Luneau, F. M. Romero, P. Rey, A. Grand, A. Caneschi,
D. Gatteschi and J. Langier, Inorg. Chem., 1993, 32, 5616.
10 D. Luneau, F. M. Romero and R. Ziessel, Inorg. Chem., 1998, 37,
5078; F. M. Romero, D. Luneau and R. Ziessel, Chem. Commun.,
1998, 551.
11 (a) T. Yoshida, T. Suzuki, K. Kanamori and S. Kaizaki, Inorg.
Chem., 1999, 38, 1059; (b) T. Yoshida and S. Kaizaki, Inorg. Chem.,
1999, 38, 1054.
12 T. Yoshida, K. Kanamori, S. Takamizawa, W. Mori and S. Kaizaki,
Chem. Lett., 1997, 603.
13 Y. Yamamoto, T. Suzuki and S. Kaizaki, J. Chem. Soc., Dalton
Trans., 2001, 1566.
14 Y. Yamamoto, T. Suzuki and S. Kaizaki, J. Chem. Soc., Dalton
Trans., 2001, 2943.
15 J. H. Osiecki and E. F. Ullman, J. Am. Chem. Soc., 1969, 90, 1078.
16 S. Kaizaki, J. Hidaka and Y. Shimura, Inorg. Chem., 1973, 12, 135.
17 G. M. Sheldrick, SHELXS-86, Program for Crystal Structure
Determination, Univ. of Göttingen, Germany, 1986.
18 TEXSAN, Single Crystal Structure Analysis Software, Version 1.7,
Molecular Structure Corporation, The Woodlands, TX, USA,
1995.
19 H. Ogino, Y. Abe and M. Shoji,, Inorg. Chem., 1988, 27, 986.
20 H. R. Fischer, D. J. Hodgson and E. Pedersen, Inorg. Chem., 1984,
23, 4755.
21 W. Wojciechowski, Inorg. Chim. Acta, 1967, 1, 324.
22 E. F. Ullman, J. H. Osieccki, P. Rey and R. Sessoli, J. Am. Chem.
Soc., 1972, 94, 7049.
23 S. Kaizaki, J. Hidaka and Y. Shimura, Inorg. Chem., 1973, 12, 142.
24 C. Lescop, D. Luneau, G. Bussière, M. Triest and C. Reber, Inorg.
Chem., 2000, 39, 3740.
Conclusions
The present spectroscopic studies along with the magnetic
properties can elucidate in some detail the spectral behavior in
2
2
terms of position and intensity of the ligand field E and T1
states of the Cr() complexes which are coupled with the
NIT2py radical ligand. The Jobs values are found to vary with
the JF values, but we are not yet in a position to obtain a cor-
relation between them through the coligand effect. Although
our initial attempts to obtain any correlation between the mag-
netic and the spectroscopic properties through the coligand
effect are as yet unsuccessful, the hypothesis of variable JF
values with varying coligands may be rationalized by con-
sidering the lowest LMCT which generates only a quintet but
not a triplet according to the formulation for the ferromagnetic
interaction in superexchange for dinuclear metal complexes.29
25 J. Ferguson, H. J. Guggenheim and Y. Tanabe, J. Phys. Soc. Jpn.,
1966, 21, 692.
26 O. Kahn, Molecular Magnetism, VCH publisher Inc., 1993, ch. 8.
27 Eqn. (3) in our previous paper11a was oversimplified on the
assumption of ESF Ӷ ECT as presented in refs. 5 and 25. In the case
of the low energy MLCT (ESF < ECT), this should be expressed as the
present formulation given in eqn. (4). However, this change has no
effect on the subsequent discussion in either case of Ni() or Cr()
complexes.
28 Y. Tsukahara, T. Kamatani, A. Iino, T. Suzuki and S. Kaizaki, to be
submitted.
29 (a) J. Glerup, D. J. Hodgson and E. Pedersen, Acta, Chem. Scand.,
Ser. A, 1983, 37, 161; (b) H. Weihe and U. Güdel, Inorg. Chem.,
1997, 36, 3632; (c) H. Weihe, U. Güdel and H. Toftlund, Inorg.
Chem., 2000, 39, 1351.
The quintet LMCT is a π
eg transition to the quintet
excited state |t2g t2g t2gϩegϩπϪπ*ϩ| from the quintet ground
ϩ
ϩ
ϩ
ϩ
state |t2g t2g t2gϩπϪπϩπ*ϩ|. The stabilization of the quintet
level in the ground state or the variation of JF values could
result from a difference in the configurational interaction
(<|egϩ|h|πϩ|>) between the quintet ground and excited states.
Our future goal will be to explore more detailed or quantitative
correlations between JF and the electronic properties of the
coligands as inferred from the relation between Jobs or JF and
the acid dissociation constants of the β-diketones. Another aim
J. Chem. Soc., Dalton Trans., 2002, 181–187
187