Synthesis, crystal structure and magnetic properties of o-carboranyl nitronyl nitroxide biradical: A prototype of a three-dimensional analogue of o-benzoquinodimethane

Article abstract of DOI:10.1039/b511579e

We have designed and synthesised a new carborane derivative containing two nitronyl nitroxides. This molecule can be considered as a prototype of a three-dimensional analogue of o-benzoquinodimethane. The magnetic susceptibility of a crystalline sample revealed that this biradical possesses a weak antiferromagnetic interaction ( = -1.05(4) K) and an ESR study in frozen matrix in the 5.4-104 K range gave an intramolecular antiferromagnetic interaction of -27(2) K. The synthesis, physical properties and DFT calculation result are also reported. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b511579e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis, crystal structure and magnetic properties of o-carboranyl nitronyl
nitroxide biradical: a prototype of a three-dimensional analogue of
o-benzoquinodimethane
Fumiyasu Iwahori,a,b Yoshikazu Nishikawa,c Ken-ichiro Mori,c Masahiro Yamashitab,d and Jiro Abe*a
Received 15th August 2005, Accepted 20th October 2005
First published as an Advance Article on the web 14th November 2005
DOI: 10.1039/b511579e
We have designed and synthesised a new carborane derivative containing two nitronyl nitroxides. This
molecule can be considered as a prototype of a three-dimensional analogue of o-benzoquinodimethane.
The magnetic susceptibility of a crystalline sample revealed that this biradical possesses a weak
antiferromagnetic interaction (h = −1.05(4) K) and an ESR study in frozen matrix in the 5.4–104 K
range gave an intramolecular antiferromagnetic interaction of −27(2) K. The synthesis, physical
properties and DFT calculation result are also reported.
Introduction
Considerable scientific efforts have been made on the chemistry
of carboranes.1 Especially, owing to the characteristic icosahedral
structure having remarkable stability, 1,2-dicarba-closo-dode-
caborane derivatives have often been used as modules for supra-
molecular chemistry, organometallic chemistry, biochemistry and
so on.2 The substitution of the C–H hydrogen on the carborane
cage has been achieved with the aim of the development of func-
tionality. For example, a porphyrin-labelled carborane derivative
has been synthesised as a pharmaceutical agent,3a and a poly-
(hydroxyl) substituted carborane was prepared for the improve-
ment of the solubility in water.3b Recently, we proposed a para-
magnetic carborane derivative, the nitronyl nitroxide substituted
carborane OCB-NN (= 1ꢀꢀ-(4ꢀ-phenylene(3,3,4,4-tetrameth-
ylimidazolyl-2-oxyl-5-oxide))-1ꢀꢀ,2ꢀꢀ-dicarba-closo-dodecaborane,
Chart 1(a); ortho-, R1 = NN, R2 = H).4 The head-to-tail dimer
form of OCB-NN via hydrogen bonding was found in the crystal
structure and an intradimer ferromagnetic interaction (J/kB
+4.26(2) K) was observed.
=
From the viewpoints of molecular design and crystal engineer-
ing, the study on carboranyl biradicals should be important for the
understanding of the physical character of paramagnetic carbo-
rane derivatives. There are large number of reports concerning
the syntheses and physical properties of organic biradicals having
various spin-coupling units.5 For instance, o-, m- and p-phenylenes
are common p-conjugated aromatic spin-coupling units, and
benzoquinodimethane (BQDM) radicals are shown in Chart 1(b).
The molecular orbital (MO) theory predicts that only non-Kekule´
Chart 1
hydrocarbon m-BQDM has non-disjoint NBMOs, and the spin
ground state will be a triplet while the other two isomers having
disjoint MOs are expected to be singlet.6a The valence bond
(VB) theory by using the alternating spin-polarisation model is
another approach for the qualitative understanding of the spin
ground state. The VB theory also predicts that only m-BQDM is
in high-spin ground state whereas o-, p-BQDMs are closed-shell
molecules.6b These theories well represent the spin ground states
of the simple alternative hydrocarbon biradicals.
aDepartment of Chemistry, The 21st Century COE Program, Aoyama
Gakuin University, 5-10-1 Fuchinobe, Sagamihara, Kanagawa, 229-8558,
Japan. E-mail: jiro_abe@chem.aoyama.ac.jp; Fax: +81 42 759 6225;
Tel: +81 42 759 6225
bCREST, Japan Science and Technology Agency, Kawaguchi Center Build-
ing, 4-1-8 Honcho, Kawaguchi, Saitama, 332-0012, Japan
cDepartment of Applied Chemistry, Graduate School of Engineering, Tokyo
Metropolitan University, 1-1 Minami-Osawa, Hachioji, Tokyo, 192-0397,
Japan
dDepartment of Chemistry, Graduate School of Science, Tohoku University,
6-3 Aramaki-Aza-Aoba, Aoba-ku, Sendai, Miyagi, 980-8578, Japan
One of the unique characters of carboranes are their
three-dimensional aromaticity.1d,7 The B–B and B–C bonds in
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 473–477 | 473
©

View Article Online
carboranes are non-classical, electron-deficient multi-centre
bonds. Therefore carborane molecules have resonance energy
resulting from the electron delocalisation as found in aromatic
hydrocarbons. Taking advantage of the unique structure and
the three-dimensional aromaticity, we propose a new series
of the magnetic materials by use of carboranes. Herein we
addressed our interest to the synthesis and investigation of
the ortho-carboranyl bis(nitronyl nitroxide) derivative OCB-
BNN (= 1ꢀꢀ,2ꢀꢀ-bis(4ꢀ-phenylene(3,3,4,4-tetramethylimidazolyl-2-
oxyl-5-oxide))-1ꢀꢀ,2ꢀꢀ-dicarba-closo-dodecaborane; Chart 1(a); or-
tho-, R1 = R2 = NN) that can be considered as a prototype
compound of this class of biradicals.
made on a Rigaku Saturn 70 CCD area detector with graphite-
monochromated Mo-Ka radiation; M = 680.89, monoclinic,
˚
space group P21/c (no. 14), k(Mo-Ka) = 0.7107 A, l(Mo-Ka) =
0.75 cm−1, T = 123(1) K, a = 13.198(4), b = 21.760(5), c =
◦
3
˚
˚
13.410(4) A, b = 102.438(4) , V = 3761.1(1) A , Z = 4, Dc =
1.202 g cm−3.
Hydrogen atoms were located on the basis of geometrical con-
sideration except B–H hydrogen atoms. All non-hydrogen atoms
were refined anisotropically and the least-squares refinement
(490 parameters) on 11054 independent reflections converged at
R1(F) = 0.070 (data for I > 2.0r(I)), and wR2(F2) = 0.149 for all
data; GOF = 1.009.
In this paper, the synthesis, structural and magnetic characteri-
sation, and DFT calculation results of OCB-BNN willbe described
with the aim for the investigation of the character of the carborane
cage as the building unit of the spin network.
CCDC reference number 280804.
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b511579e
Magnetic measurements
The magnetic susceptibility at 0.1 T in the 2–300 K temperature
range was carried out with a Quantum Design MPMS-XL SQUID
magnetometer. The polycrystalline sample of OCB-BNN was
contained in a diamagnetic gel-capsule.
The ESR signals of OCB-BNN in frozen toluene matrix were
recorded on a JEOL JES-TE200 spectrometer operating at the X-
band equipped with a digital temperature controller model 9650
(Scientific Instruments Inc.). Toluene was distilled on CaH2 under
nitrogen prior to use and the solution was degassed by the freeze–
pump–thaw method.
Experimental
Preparation of OCB-BNN
OCB-BNN was synthesised from 1,2-bis(4ꢀ-formylphenyl)-1,2-
dicarba-closo-dodecaborane8 and 2,3-dimethyl-2,3-bis(hydroxyl-
amino)butane, followed by oxidation using aqueous sodium meta-
periodide.5c
To a methanol solution of 250 mg (0.709 mmol) of 1,2-
bis(4ꢀ-formylphenyl)-1,2-dicarba-closo-dodecaborane, 200 mg
(1.35 mmol) of 2,3-dimethyl-2,3-bis(hydroxylamino)butane was
added under nitrogen atmosphere at room temperature. The
reaction mixture was continuously stirred for 10 days with
exclusion of the light and the precipitate was filtered off. The short
reaction time gave the mono-substituted derivative as the major
product. The filtrate was washed with methanol and dried under
reduced pressure. The radical precursor, 1ꢀꢀ,2ꢀꢀ-bis(4ꢀ-phenyl(2,5-
dihydroxy-3,3,4,4-tetramethylimidazolyl))-1ꢀꢀ,2ꢀꢀ -dicarba-closo-
dodecaborane was obtained as a white powder (375 mg, 86.3%).
The radical precursor (370 mg, 0.604 mmol) was oxidised in
dichloromethane by excess of aqueous sodium meta-periodide at
0 ◦C under nitrogen. After 30 min, OCB-BNN was extracted with
dichloromethane and the combined organic layer was dried over
Na2SO4. After the solvent was removed under reduced pressure,
the purification was performed using a column chromatograph
packed with silica gel in Et2O eluent. A blue powder of OCB-
BNN·Et2O (273 mg, 74.2%) was obtained. An X-ray-quality single
crystal was prepared by the slow diffusion of diethyl ether into
acetonitrile solution of OCB-BNN.
Results and discussion
The molecular structure obtained by an X-ray crystallographic
analysis is shown in Fig. 1(a). The bond lengths and angles are
listed in Table 1 with the result of DFT calculation that will be
described later.
All bond lengths and angles of OCB-BNN are reasonable
as compared with those found in the literature. The least-
square planes defined by O–N–C–N–O fragments of two nitronyl
nitroxides lead to dihedral angles of 36.2(1) and 22.9(1)◦,
Table 1 Selected bond lengths and angles. Calculated values were
obtained from the DFT calculation (UB3LYP/6-31G*)
Experimental
Calculated
˚
Distances/A
N–O•a
1.285
1.359
1.389
1.775
1.721
1.726
1.275
1.365
1.401
1.778
1.717
1.762
N–C(sp2)a
C–C (phenylene)a
B–Ba
IR (KBr pellet): mmax/cm−1 2593 (B–H) and 1368 (N–O). UV
absorption: kmax/nm (CH2Cl2) = 277 (e/dm3 mol−1 cm−1 (26100),
285 (26000), 294 (26100), 377 (22500), 599 (580). The intense
absorption band ascribed to the p–p* transition band of nitroxides
was observed below 400 nm. The weak and broad absorption band
around 600 nm is attributable to the n–p* transition of nitronyl
nitroxides.
B–Ca
C–C(carborane)
Angles/◦
N–C–N
108.14
126.43
22.9
108.2
128.0
3.4c
O–N–C(sp2)
Dihedral, Ph–NNb
36.2
X-Ray crystallographic analysis
a The mean length of the corresponding bonds. b Dihedral angles between
the least square planes defined by O–N–C–N–O fragments of nitronyl
nitroxide and phenylene rings. c The two dihedral angles were identical.
A
blue block crystal of OCB-BNN·Et2O (C28H42B10N4O4·
(C2H5)2O) having approximate dimensions of 0.20 × 0.10 ×
0.10 mm3 was mounted on a glass fibre. All measurements were
474 | Dalton Trans., 2006, 473–477
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
i
i
oxygen atoms O1 · · · O2 and O3 · · · O4 (symmetry operator (i)
+x, −y + 1/2, z − 1/2) with distances of 4.592(3) and 4.911(3)
˚
A are shorter than the intramolecular oxygen–oxygen distance,
thereby forming a one-dimensional triangular column parallel to
the crystallographic c-axis as shown in Fig. 1(b) and (c). From
the crystallographic viewpoint, we note that the intramolecular
magnetic interaction would not be predominant in the solid state
by comparison with the intermolecular interaction.
A theoretical calculation was carried out in order to obtain
insight into the electronic ground state of OCB-BNN. Unfortu-
nately, the SCF single-point energy calculation at the B3LYP/6-
31G* level for the isolated OCB-BNN itself using the experimental
geometry could not converge because the experimental structure
might be far from the theoretical equilibrium structure. The
detailed results for the DFT calculation have been reported in
our previous paper.9 The geometry optimisations were carried
out at the unrestricted B3LYP level of the theory using the 6-
31G* basis set, and converged to identical structures in singlet and
triplet states. The calculated bond lengths and angles are listed in
Table 1. The dihedral angles of the least-square mean-planes of two
radical fragments (O–N–C–N–O) with respect to benzene rings are
optimised as 3.4◦ by the DFT calculation, while the experimental
angles are36.2(1)and 22.9(1)◦. This difference might arise from the
steric hindrance of diethyl ether in the crystal lattice. The energy
levels of singlet and triplet states were also calculated. The spin-
squared expectation value ꢁS2ꢂ after spin projections are calculated
as ꢁS2ꢂsinglet = 1.04 and ꢁS2ꢂtriplet = 2.01, respectively. The large spin
eigenvalue of the singlet state indicates the spin contamination
of the wavefunction from higher spin states. Similarly to other
nitroxides in the literature, most spin population is located on the
O–N–C–N–O part and the spin distributions on the benzene ring
and carborane cage are small.10 The total energies are obtained
as Esinglet = −1860.6229193 Hartree for the singlet and Etriplet
=
−1860.6229177 Hartree for the triplet state. The singlet–triplet
energy gap (DES–T = Esinglet − Etriplet) was 1.6 × 10−6 Hartree.
The intramolecular magnetic exchange interaction was estimated
for the spin Hamiltonian defined as H = −2JcalcS1·S2 according
to the literature method.11 The calculated energy difference of
OCB-BNN corresponds to the exchange interaction of 2Jcalc/kB =
−0.53 K predicting a weak antiferromagnetic interaction between
the nitronyl nitroxides.
The magnetic susceptibility of the polycrystalline sample was
measured by a SQUID magnetometer (Fig. 2). The temperature
dependence of the magnetic susceptibility was measured under 0.1
T of applied magnetic field.
At 300 K, the vmT value was 0.777 emu K mol−1, slightly higher
than the theoretical value for two moles of S = 1/2 radicals.
The vmT value was constant in the temperature range between
300 and 100 K, then began to decrease gradually below 100 K
and showed abrupt decreasing in lower temperature. This fact
indicates the presence of antiferromagnetic interaction in the solid
state. The reciprocal susceptibility was fitted by the Curie–Weiss
law with parameters C = 0.781(6) emu K mol−1 and h = −1.05(4) K
between 10 and 300 K. The obtained Curie constant corresponds
to twice the isolated S = 1/2 value and the negative Weiss constant
indicates a weak short-range antiferromagnetic interaction in the
crystal.12
Fig. 1 (a) ORTEP drawing of OCB-BNN (50% probability). Et2O
molecules and geometrically calculated hydrogen atoms are omitted. (b)
The molecular alignment in the crystal packing. Diethyl ether molecules
are displayed by balls and sticks. (c) Intermolecular contacts along the
˚
c-axis. The intermolecular distances (A) are as follows: a = 5.74, b = 3.85,
c = 4.59, d = 5.13, e = 4.91, f = 3.70, g = 3.28, h = 4.07, i = 4.03, j =
5.54, k = 4.59, 1 = 4.91, m = 3.70, n = 3.40, o = 4.47.
respectively, for the mean planes using the benzene rings (C3–
C8 and C16–C21). The shortest intramolecular oxygen–oxygen
˚
distance between nitronyl nitroxide fragments is 6.86(1) A (O2–
The ESR spectrum in toluene solution shows nine lines of
hyperfine structure which corresponds to four equivalent nitrogen
O4). On the one hand, the intermolecular contacts between
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 473–477 | 475
©

View Article Online
data was fitted by the modified Bleaney–Bowers equation (eqn
(1); where Cꢀ is a fitting coefficient) for the Hamiltonian H =
−2JESRS1·S2,13 giving the antiferromagnetic interaction JESR/kB =
−27(2) K. The calculated curve is shown in Fig. 3.
Cꢀ
IESRT =
(1)
3 + exp(−2JESR/kBT)
A larger exchange interaction was observed for the isolated
molecule in frozen matrix, whereas the Weiss constant of the
crystalline state is smaller. This difference probably arises from
the structural difference in both states. In the crystalline state,
diamagnetic molecules, i.e. diethyl ether is interposed between
nitronyl nitroxide arms, and the magnetic centres are separated
˚
by approximately 7 A. On the other hand, in solution, diethyl
ether would be released into bulk solution and the two nitroxide
arms are allowed to rotate around the C6–C9 and C19–C22 bonds.
The rotation of the imidazole rings might induce the proximity of
the nitroxides that increases the dipolar interaction. The magnetic
interaction between spin sites provides the most important insights
for our goal of understanding the magnetic exchange through a
carborane cage.
Taking into account the results of X-ray crystallography, ESR
and DFT studies, the antiferromagnetic interaction should be a
through-space dipole interaction among radicals. It is obvious that
the phenylene rings between the radical and carborane cage impair
the intramolecular exchange interaction through a carborane cage.
However, initial attempts to synthesise a radical without chemical
coupler gave unsatisfactory results because of poor stability of
the target compound, and the synthesis is continuously under
investigation. More progressive researches and syntheses of other
analogues are also now in progress.
Fig. 2 vmT vs. T plot (ꢀ) and Curie–Weiss plot (᭺) with theoretical
fitting (solid line) for OCB-BNN.
atoms at room temperature. The hyperfine coupling constant (aN)
is 0.363 mT (= 2.4 × 10−4 K) which is comparable with those of the
nitronyl nitroxide family having four equivalent nitrogen nuclei in
the literature.5c This ESR spectrum means that the intramolecular
magnetic exchange interaction (JESR/kB) is significantly larger than
the hyperfine coupling constant (|JESR/kB| ꢃ aN). The variable-
temperature ESR spectra of OCB-BNN in frozen toluene matrix
(9.6 × 10−6 mol dm−3) were measured in order to evaluate the
intramolecular magnetic interaction with exclusion of intermolec-
ular interaction. The ESR signal intensities (= IESR) were obtained
by using double integration of the first-derivative spectra. The
temperature dependence of the IESRT product is shown in Fig. 3.
Conclusion
In summary, we have demonstrated here the first example of a
carboranyl biradical molecule, OCB-BNN. The ESR measure-
ment in frozen matrix evaluated the intermolecular magnetic
interaction as −27(2) K.
Our magnetic measurement revealed the spin ground state of
OCB-BNN is a singlet, and the DFT calculation that predicts
the singlet spin ground state is consistent with the experimental
results. This result gave insights on the physical properties of the
paramagnetic carborane family. In this case, we can conclude that
o-carborane framework having nitronyl nitroxide radicals exhibits
intramolecular antiferromagnetic interaction.
Acknowledgements
This work was partly supported by a Grant-in-Aid for young
scientists B (No. 16750125) from the MEXT, Japan. F. I. and M. Y.
thank the Japan Science and Technology Agency (JST) for
financial support. F. I. and J. A. are grateful for financial support
of the 21st Century COE program of Aoyama Gakuin University.
Fig. 3 Temperature dependence of IESRT in frozen toluene matrix. The
solid curve shows the fitting result described in the text.
References
At liquid-helium temperature, a very weak ESR signal indicat-
ing an antiferromagnetic intramolecular coupling was observed,
1 (a) Z. Xie, Acc. Chem. Res., 2003, 36, 1; (b) R. N. Grimes, Angew. Chem.,
Int. Ed., 2003, 42, 1198; (c) S. O. Kang and J. Ko, Adv. Organomet.
and no half-field signal around g = 4 was found. The IESR
T
476 | Dalton Trans., 2006, 473–477
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Chem., 2001, 47, 61; (d) R. B. King, Chem. Rev., 2001, 105, 1119;
(e) V. I. Bregadze, Chem. Rev., 1992, 92, 209; (f) J. Plesek, Chem. Rev.,
1992, 92, 269; (g) L. A. Leites, Chem. Rev., 1992, 92, 279, and references
therein.
2 (a) M. F. Issac and S. B. Kahl, J. Organomet. Chem., 2003, 680, 232; (b) J.
Johnsamuel, Y. Byun, T. P. Jones, Y. Endo and W. Tjarks, J. Organomet.
Chem., 2003, 680, 223; (c) I. B. Sivaev, S. Sjoberg and V. I. Bregadze,
J. Organomet. Chem., 2003, 680, 106; (d) J. F. Valliant, K. J. Guenther,
A. S. King, P. Morel, P. Schaffer, O. O. Sogbein and K. A. Stephenson,
Coord. Chem. Rev., 2002, 232, 173; (e) A. Maderna, R. Huertas, M. F.
Hawthorne, R. Luguya and M. G. H. Vicent, Chem. Commun., 2002,
1784; (f) J. F. Valiant, O. O. Sogbein, P. Morel, P. Schaffer, K. J. Guenther
and A. D. Bain, Inorg. Chem., 2002, 41, 2731.
3 (a) M. G. H. Vicent, S. J. Shetty, A. Wickramasinghe and K. M. Smith,
Tetrahedron Lett., 2000, 41, 7623; (b) H. Nemoto, J. G. Wilson, H.
Nakamura and Y. Yamamoto, J. Org. Chem., 1992, 57, 435.
4 F. Iwahori, K. Kamibayashi, Y. Nishikawa, M. Yamashita and J. Abe,
Chem. Lett., 2004, 11, 1460.
5 For example;(a) A. Rajca, Chem. Rev., 1994, 94, 871; (b) D. A. Shultz,
A. K. Boal, H. Lee and G. T. Farmer, J. Org. Chem., 1999, 64,
4386; (c) the following papers especially deal with bis-nitronylnitroxide
radicals: E. F. Ullman, J. H. Osiecki, D. G. B. Boocock and R. Darcy,
J. Am. Chem. Soc., 1972, 94, 7049; (d) M. Tanaka, K. Matsuda, T. Itoh
and H. Iwamura, Angew. Chem., Int. Ed., 1998, 37, 810; (e) L. Catala,
J. L. Moigne, N. Kyritsakas, P. Rey, J. J. Novoa and P. Turek, Chem.
Eur. J., 2001, 11, 2466.
7 P. v. R. Schleyer and K. Najafian, Inorg. Chem., 1998, 37, 3454; J.
Aihara, J. Am. Chem. Soc., 1978, 100, 3339.
8 1,2-Bis(4ꢀ-formylphenyl)-1,2-dicarba-closo-dodecaborane was synthe-
sised from 4,4ꢀ-bis(2H-1,3-dithiolane)diphenylacetylene and decabo-
rane (B10H14) followed by the cleavage reaction of the protective groups
by Hg(ClO4)2·6H2O. The mercury(II) perchlorate is explosive and it
should be handled carefully.
9 Proceedings of the Conference on the Science and Technology of
Synthetic Metals: F. Iwahori, K. Kamibayashi, K. Mori, Y. Nishikawa,
M. Yamashita and J. Abe, Synth. Met., 2005, in press.
10 Y. Pontillon, T. Akita, A. Grand, K. Kobayashi, E. Lelievre-Berna,
J. Pe´caut, E. Ressouche and J. Schweizer, J. Am. Chem. Soc., 1999,
121, 10126; A. Zheldev, V. Barone, M. Bonnet, B. Delley, D. Grand, E.
Ressouche, P. Rey, R. Subra and J. Schweizer, J. Am. Chem. Soc., 1994,
116, 2019.
11 K. Yamaguchi, Y. Takahara and T. Fueno, Applied Quantum Chemistry,
ed. V. H. Smith, et al., 1986, p. 155.
12 Because of the comparable intra- and intermolecular contact distances
between the radicals, the data fitting by a singlet–triplet two-spin model
based on the molecular structure is insufficient to obtain physically
meaningful parameters for the crystalline sample. Tentative treatment
by using the modified Bleaney–Bowers equation (ref. 11) gave Jcryst/kB =
ꢀ
−1.59(1) K and zJcryst /kB = −0.33(1) K, where Jcryst/kB means a
ꢀ
predominant intradimer magnetic interaction and zJcryst /kB describes
interdimer magnetic interaction by the mean field approximation (C. J.
O’Connor, Prog. Inorg. Chem., 1982, 29, 203).
13 B. Bleaney and K. D. Bowers, Proc. R. Soc. London, Ser. A, 1952, 214,
451.
6 (a) W. T. Borden and E. R. Davidson, J. Am. Chem. Soc., 1977, 99,
4587; (b) A. A. Obchinnikov, Theor. Chim. Acta, 1978, 47, 297.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 473–477 | 477
©