H2O as a chemical link for ferromagnetic interactions between aminoxyl units

Article abstract of DOI:10.1002/1099-0690(200301)2003:1<167::AID-EJOC167>3.0.CO;2-G

A bis(radical) derivative based on a triphenylphosphane core substituted by two nitronyl nitroxide units is reported. This compound was isolated in two forms. One consists of an H₂O adduct where H²O is hydrogen-bonded to the radical units, whereas the second lacks this bridging unit. Magnetic studies revealed that the presence of the H²O bridge induces ferromagnetic exchange interaction between the spin carriers. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/1099-0690(200301)2003:1<167::AID-EJOC167>3.0.CO;2-G

FULL PAPER
H2O as a Chemical Link for Ferromagnetic Interactions Between
Aminoxyl Units
Corinne Rancurel,[a] Nathalie Daro,[a] Oscar Benedi Borobia,[a] Eberhardt Herdtweck,[b] and
Jean-Pascal Sutter*[a]
Keywords: Hydrogen bonds / Through-bond interactions / Radicals / Magnetic properties
A bis(radical) derivative based on a triphenylphosphane core
substituted by two nitronyl nitroxide units is reported. This
compound was isolated in two forms. One consists of an H2O
adduct where H2O is hydrogen-bonded to the radical units,
whereas the second lacks this bridging unit. Magnetic stud-
ies revealed that the presence of the H2O bridge induces fer-
romagnetic exchange interaction between the spin carriers.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Purely organic magnets are desirable materials and the lated both to the building block (such as the chemical na-
design of open-shell supramolecular architectures is cur- ture of the molecule itself, the substitution pattern i.e. ami-
rently attracting an increasing interest.[1,2] The approach in- noxyl unit versus H-donor or -acceptor group positions,
volving molecular crystals of aminoxyl radicals is certainly and the main exchange pathway among several intermol-
the most investigated. It has quickly become evident that to ecular connections) and to the crystal lattice organization.
reach the objective it is essential to gain control over the Consequently, it is difficult to manage the dominant ex-
construction of the supramolecular network as well as over change between the magnetic centers.
the exchange pathways between the spin carriers. Indeed,
An alternative approach is to link the paramagnetic cen-
the bulk magnetic properties will depend on the dimen- ters by an independent molecule which allows for ferromag-
sionality of the system. The basic tools of organic supramo- netic exchange. Obtaining materials with a remnant mag-
lecular chemistry provide solutions for a directed material netization by a homo-spin strategy requires a ferromagnetic
synthesis.[3,4] Among them, the hydrogen-bonding strategy interaction between the molecular units. A prominent ex-
attracts special interest because it has been suggested from ample of this approach involves the use of phenylboronic
both theoretical[5Ϫ7] and experimental[8Ϫ11] results that a acid.[19,20] In this report, we show that a simple hydrogen-
hydrogen bond might mediate the intermolecular magnetic bonded water molecule may be the driving force behind fer-
communication.
romagnetic interaction between aminoxyl radical units.
An increasing number of supramolecular frameworks
consisting of H-bonded paramagnetic subunits bearing
either OH,[12Ϫ14] NH,[15Ϫ17] or acetylenic CH[18] groups as
H-donor moieties were found to exhibit long-range mag-
netic correlation. However, whereas the chemical dimen-
sionality of the solid-state structure and the intermolecular
links between the subunits can be controlled by the means
of the molecular synthon, it remains difficult to anticipate
the final properties of the supramolecular material. In these
homomolecular architectures, the exchange interactions be-
tween the paramagnetic units proceed through the supram-
olecular links and the core of the molecules. The resulting
dominant exchange will depend on several parameters re-
Results and Discussion
Synthesis and Crystal Structures
The bis(radical) derivative 5 considered in this study was
synthesized as outlined in Scheme 1. The synthesis of bis(al-
dehyde) 2 was adapted from original procedures,[21,22] ex-
cept for the deprotection step (Scheme 1, step 3), which was
modified to improve the yield (up to 97%). The condensa-
tion of phosphane oxide 3 and 2,3-bis(hydroxyamino)-2,3-
dimethylbutane, affording 4, gave best results in refluxing
methanol. This reaction should not exceed 6 h as a longer
reaction time results in decomposition of the product. The
crystalline bis(radical) was obtained by diffusion of either
Et2O or hexane into a solution of 5 in CH2Cl2. Either red
prismatic or needle-like crystals (5a and 5b, respectively)
were formed depending on whether dry or ‘‘wet’’ solvent
was used, i.e. dried (see Exp. Sect.) or ‘‘for synthesis’’ qual-
[a]
`
´
Institut de Chimie de la Matiere Condensee de Bordeaux
CNRS-UPR No. 9048,
33608 Pessac, France
E-mail: jpsutter@icmcb.u-bordeaux.fr
Anorganisch-chemisches Institut, Technische Universität
München,
[b]
85747 Garching bei München, Germany
Eur. J. Org. Chem. 2003, 167Ϫ171  2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/03/0101Ϫ0167 $ 20.00ϩ.50/0
167

C. Rancurel, N. Daro, O. Benedi Borobia, E. Herdtweck, J.-P. Sutter
FULL PAPER
Figure 1. ORTEP drawing for compound 5a (30% probability el-
˚
lipsoids); selected bond lengths [A] and angles [°]: O2ϪN42
1.287(2), O3ϪN45 1.281(2), O4ϪN52 1.275(2), O5ϪN55 1.283(2),
C41ϪN42 1.343(2), C41ϪN45 1.338(2), C26ϪC41 1.473(2),
C21ϪC26 1.408(2), PϪO1 1.486(1), PϪC11 1.810(2), PϪC21
1.825(2), PϪC31 1.822(1); O1ϪPϪC11 112.41(7), O1ϪPϪC21
111.75(7), O1ϪPϪC31 112.88(7); mean plane angles:
radicalϪphenyl 68.22(5) and 78.99(5), radicalϪradical 67.28(6)
Scheme 1
ity solvent. These crystals were suitable for X-ray struc-
tural analysis.
The X-ray crystal structure of 5a confirmed that the mo-
lecule is indeed the anticipated bis(radical) and revealed
that the crystal lattice includes CH2Cl2. An ORTEP view is
presented in Figure 1 with selected bond lengths and angles.
The main interatomic distances are very similar to those
found for the related mono(radical) derivative.[23] The dis-
tance between the two nitronyl nitroxide units in the molec-
ule can be evaluated by the C41ϪC51 separation, which is
˚
5.220(2) A; the shortest separation involving the NO units
of the two radical moieties is found for O2ϪO5 which is
˚
4.213(2) A. The angle between the mean plane of the rad-
Figure 2. ORTEP drawing for compound 5b (30% probability el-
ical moieties is 67.28(6)°. The shortest intermolecular con-
tacts involving the aminoxyl O atoms are O2 to the CH2Cl2
solvate [O2···H2C, 2.33(2) A], O3 and O4 to the phenyl
rings of neighboring molecules [O3···HC22, 2.45(2) A;
O4···HC24, 2.32(2) A], and O5 to a methyl group
[O5···HC46, 2.44(3) A].
The X-ray crystal structure analysis for the needle-shaped
crystals of 5b revealed that an H2O molecule had co-crystal-
lized with compound 5. No other solvate molecules were
found in the crystal lattice. Interestingly, the H2O molecule
˚
lipsoids); selected bond lengths and distances [A] and angles [°]:
O2ϪN42 1.288(2), O3ϪN45 1.278(2), O4ϪN52 1.283(2), O5ϪN55
1.280(2), C41ϪN42 1.334(2), C41ϪN45 1.351(3), C26ϪC41
1.476(3), C21ϪC26 1.409(3), C51ϪN52 1.341(2), C51ϪN55
1.341(3), C36ϪC51 1.477(3), PϪO1 1.480(2), PϪC11 1.811(2),
PϪC21 1.821(2), PϪC31 1.827(2) O2···H2 2.03(3), O5···H1,
1.94(3); O2ϪOϪO5 94.35(7), O1ϪPϪC11 110.74(8), O1ϪPϪC21
113.96(8), O1ϪPϪC31 114.79(9); mean plane angles:
radicalϪphenyl 51.56(7) and 87.26(7), radicalϪradical 77.09(8)
˚
˚
˚
˚
is actually linked to 5, as depicted in the ORTEP view of ical moieties [77.09(8)°] remain very similar to those found
this adduct (Figure 2). Two hydrogen bonds are established in 5a. Most of the intramolecular geometric characteristics
between the H2O molecule and an O atom of one NO unit are also similar for the two compounds 5a and 5b (see cap-
˚
of each radical moiety, H1ϪO5 [1.94(3) A] and H2ϪO2 tion to Figure 2). The shortest intermolecular contacts in-
˚
[2.03(3) A]. Despite the presence of the H2O in 5b, the volving the aminoxyl O atoms of 5b (containing the
˚
˚
separation of the two aminoxyl units [4.170(2) A for bridging H2O), are found for O2···HC56 [2.70(2) A],
˚
˚
O2···O5], and the angle between the mean plane of the rad- O3···HC49 [2.79(2) A], O4···HC14 [2.37(3) A], and
168
Eur. J. Org. Chem. 2003, 167Ϫ171

H2O as a Chemical Link for Ferromagnetic Interactions Between Aminoxyl Units
FULL PAPER
˚
O5···HC47 [2.47(3) A]. The crystal structure shows two to the H2O molecule linking the two aminoxyl moieties,
close, statistically disordered positions for two methyl providing an alternative pathway for the intramolecular ex-
groups (C58 and C59) of one radical moiety.
change interaction between the two radical units. The field
dependence of the magnetization for 5b, measured at 2 K,
compares favorably to the calculated value for an S ϭ 1
spin state (as calculated by the Brillouin function; Figure 4).
An estimate of the intramolecular exchange parameter, J,
for 5b was obtained by analyzing the magnetic data with a
dimer-model derived from the spin Hamiltonian H ϭ
ϪJS1·S2. Least-square fitting to the experimental data led
to J ϭ 1.58 Ϯ 0.02 cmϪ1 with g ϭ 2.00 (Figure 3). This
ferromagnetic interaction between the aminoxyl units via
the H2O molecule is accounted for by the mechanism based
on electron spin polarization. Indeed, an alternate polariza-
tion on the atom sequence NO···HϪOϪH···ON results in
the same polarization on each of the aminoxyl groups, a
situation for which ferromagnetic interaction between both
spin carriers is anticipated.
Magnetic Properties
The solid-state temperature dependence of the molar
magnetic susceptibilities, χM, for 5a and 5b were investig-
ated on polycrystalline samples with a SQUID magneto-
meter in the temperature range 2Ϫ300 K. The plot of χMT
versus T for both derivatives is shown in Figure 3. In the
temperature region of 50Ϫ300 K, the χMT value for each
compound is 0.75 cm3·K·molϪ1 and remains almost invari-
ant as expected for two S ϭ 1/2 spins in a Curie regime. At
lower temperatures, however, compounds 5a and 5b behave
quite differently. For 5a, the χMT value decreases with tem-
perature and reaches 0.67 cm3·K·molϪ1 at 2 K, indicating
antiferromagnetic interaction between the spin carriers. For
5b, the value of χMT steadily increases up to 0.91
cm3·K·molϪ1 as T is lowered to 2 K, a feature characteristic
for ferromagnetic interactions between the aminoxyl units.
Figure 4. Experimental (black dots) field dependence of the mag-
netization for compound 5b at 2 K compared to the magnetization
given by the Brillouin function for one S ϭ 1 spin (squares) and
two uncorrelated S ϭ 1/2 spins (triangles) for the same temperature
(the lines are just eye guides)
Figure 3. Experimental χMT versus T behavior for compounds 5a
(tilted squares) and 5b (squares); the solid line represents the calcu-
lated best fit to the experimental data for 5b
Concluding Remarks
The magnetic behavior for compound 5a indicates that,
overall, weak antiferromagnetic interactions are operative
between the radical units. These may originate either from
the intramolecular exchange interaction between the two
aminoxyl groups or from intermolecular exchange interac-
tions through the close contacts to neighboring molecules.
Most probably both take place concomitantly. More im-
portantly, it reveals that the intramolecular exchange
through the phosphane core in bis(radical) 5a is rather
weak. This observation can be attributed to the large dihed-
ral angles (68.22 and 79.00°) between the nitronyl nitroxide
heterocycles and phenyl rings to which these are linked. It
has been shown that this reduces the spin density on the
atoms linking the aminoxyl moieties and, consequently, re-
duces the exchange interaction through the molecular
core.[24]
Crystal engineering based on the self-assembly of mo-
lecular subunits utilizing hydrogen bonds has proven to be
a valuable approach for the preparation of magnetic supra-
molecular materials. A possible strategy consists of linking
the paramagnetic subunits by using an independent molec-
ule which allows for ferromagnetic exchange between the
magnetic centers. The results gathered with the bis(radical)
derivatives 5a,b show that a hydrogen-bonded water molec-
ule may be the driving force behind ferromagnetic interac-
tion between aminoxyl radical units. These two compounds
demonstrate the usefulness hydrogen bonds in the propaga-
tion of exchange interactions.
Experimental Section
No significant differences in the geometrical features dis-
tinguish the bis(radical) unit in 5a from that of 5b. Thus,
General: Unless otherwise specified, commercially available sol-
the ferromagnetic behavior found for 5b may be attributed vents and reagents were used without further purification. Dry
Eur. J. Org. Chem. 2003, 167Ϫ171 169

C. Rancurel, N. Daro, O. Benedi Borobia, E. Herdtweck, J.-P. Sutter
FULL PAPER
THF was obtained by distillation from Na/benzophenone, CH2Cl2
was distilled from CaH2, MeCN was dried with P2O5, DMF was
dried with MgSO4 and distilled under vacuum, MeOH was dried
bis(radical) 5 was obtained by diffusion of Et2O into a CH2Cl2
solution. IR (KBr): ν˜ ϭ 1399 (s), 1367 (s), 1197 (m), 1171 (w), 1134
(m), 754 (m), 728 (m), 714 (m), 694 (m), 596 (s) cmϪ1. EPR
with Mg(OMe)2 prior to distillation. Reactions involving nBuLi (CH2Cl2, 295 K): g ϭ 2.006 (1 broad line). 5a: C33H39Cl2N4O5P
were carried out under N2 in oven-dried glassware. PPhCl2 was
distilled prior to use. NMR spectra were recorded with a Bruker
instrument operating at 200 MHz for 1H. IR spectra were recorded
in the range 4000Ϫ400 cmϪ1 using a Perking Elmer FT-IR Perga-
mon 1000. Elementary analyses were performed by the ‘‘Service
Central d’Analyse du CNRS’’ at Vernaison, France.
(673.5): calcd. C 58.84, H 5.84, N 8.32; found C 58.75, H 5.83, N
8.51. 5b: C32H39N4O6P (606.6): calcd. C 63.35, H 6.48, N 9.23;
found C 63.23, H 6.41, N 9.29.
Crystal Structure Analysis of 5a: C32H37N4O5P·CH2Cl2 (673.55),
red prism, 0.10 ϫ 0.36 ϫ 0.36 mm, monoclinic, Cc (no. 9), a ϭ
˚
10.1228(2), b ϭ 29.4913(7), c ϭ 11.5810(3) A, β ϭ 104.241(1)°,
Bis(o-formylphenyl)phenylphosphane (2)[21,22]
3
V ϭ 3351.08(14) A , Z ϭ 4, ρ ϭ 1.335 g·cmϪ3, F000 ϭ 1416, µ ϭ
˚
0.288 mmϪ1. A suitable single crystal for the X-ray diffraction
study was obtained by diffusion of Et2O into a solution of 5 in
CH2Cl2. The selected crystal was coated with perfluorinated ether,
fixed onto a capillary and placed in a cold nitrogen flow (Oxford
Cryosystems) in the diffractometer. Preliminary examination and
data collection were carried out with a Kappa CCD device (NON-
IUS MACH3) at the window of a rotating anode (NONIUS
FR591) with graphite-monochromated Mo-Kα radiation (λ ϭ
Step 1: 2-(o-Bromophenyl)-1,3-dioxolan was prepared from 2-bro-
mobenzaldehyde (1), and ethylene glycol according to procedures
described in the literature.[21,25]
Step 2: 2-(o-Bromophenyl)-1,3-dioxolan (10 g, 44 mmol) was dis-
solved in freshly distilled THF (120 mL) and cooled to Ϫ78 °C. A
solution of 1.6  nBuLi in hexane (30 mL; 48 mmol) was added
and the resulting mixture stirred at Ϫ78 °C for 1 h. Neat PCl2Ph
(3.0 mL; 22 mmol) was then added dropwise and the reaction mix-
ture was stirred at room temperature overnight. Water (15 mL) was
cautiously added, the organic and aqueous phases were separated
and further extracted and washed with water and benzene respect-
ively. The combined organic fractions were dried with anhydrous
MgSO4 and, after filtration, the solvent was removed in vacuo. The
residue was recrystallized from MeOH affording white crystals of
bis[2-(1,3-dioxolan-2-yl)phenyl](phenyl)phosphane (66%, 5.8 g).
˚
0.71073 A). Data collection was performed at 143 K within the Θ
range of 2.19° Ͻ Θ Ͻ 27.48°.[27] A total of 12904 reflections were
integrated and corrected for Lorentz and polarization effects. A
correction for absorption effects and/or decay was applied during
the scaling procedure.[28] After merging (Rint ϭ 0.0175), 6858 [6697:
Io Ͼ 2σ(Io)] independent reflections remained and were all used to
refine 562 parameters. The structure was solved by a combination
of direct methods[29] and difference Fourier syntheses.[30] The hy-
drogen atom positions were found in the difference Fourier map
calculated from the model containing all non-hydrogen atoms. The
hydrogen positions were refined with individual isotropic displace-
ment parameters. Full-matrix least-squares refinements were car-
ried out by minimizing w(F2o Ϫ Fc2)2 and converged with R1 ϭ
0.0258 [Io Ͼ 2σ(Io)], wR2 ϭ 0.0654 (all data), GOF ϭ 1.042 and
shift/error Ͻ 0.002. The residual electron density between Ϫ0.30
Step 3: A solution of bis[2-(1,3-dioxolan-2-yl)phenyl](phenyl)phos-
phane (5.9 g, 14 mmol) and p-toluenesulfonic acid monohydrate
(0.58 g) in acetone (160 mL) was refluxed for 70 min. Water
(40 mL) was then added and the acetone evaporated under reduced
pressure leading to the formation of a yellow crystalline precipitate
(2) which was immediately filtered,[26] washed with water and dried
(4.45 g, 97%).
3
˚
and 0.24 e/A showed no distinctive features. The drawings, listings
and checking for correctness were performed with the program
PLATON.[31]
Bis(o-formylphenyl)(phenyl)phosphane Oxide (3):[22] An aqueous so-
lution of H2O2 (0.1 mL, 35%, 3.44 mmol) was added to a solution
of 2 (1 g, 3.44 mmol) in acetone (40 mL) and stirred for 1 h. The
solvent was removed under reduced pressure and the residue dis-
solved in CH2Cl2. This solution was then washed with H2O until
the oxidation test of the aqueous fractions with Mohr’s salt proved
negative. After drying the organic phase with Na2SO4, the solvent
was removed in vacuo quantitatively affording 3 as a white powder.
Crystal Structure Analysis of 5b: C32H37N4O5P·H2O, (606.64), red
fragment, 0.03 ϫ 0.18 ϫ 0.38 mm, monoclinic, P21/n (no. 14), a ϭ
˚
17.6592(2), b ϭ 10.3019(1), c ϭ 18.7393(2) A, β ϭ 116.8956(5)°,
3
V ϭ 3040.36(6) A , Z ϭ 4, ρ ϭ 1.325 g.cmϪ3, F000 ϭ 1288, µ ϭ
˚
0.142 mmϪ1. A suitable single crystal for the X-ray diffraction
study was obtained by diffusion of Et2O into a solution of 5 in
‘‘wet’’ CH2Cl2. The selected crystal was coated with perfluorinated
ether, fixed onto a capillary and placed in a cold nitrogen flow
(Oxford Cryosystems) in the diffractometer. Preliminary examina-
tion and data collection were carried out with a Kappa CCD device
(NONIUS MACH3) at the window of a rotating anode (NONIUS
FR591) with graphite-monochromated Mo-Kα radiation (λ ϭ
Bis[o-1,3-dihydroxy-4,4,5,5-tetramethylimidazol-2-yl)phenyl]-
(phenyl)phosphane Oxide (4): A solution of 3 (0.5 g, 1.49 mmol)
and 2,3-bis(hydroxyamino)-2,3-dimethylbutane (0.44 g, 3 mmol) in
dry MeOH (40 mL) was refluxed under N2 for 6 h. After reducing
the volume in vacuo to ca. 4 mL, H2O (20 mL) was added leading
to the precipitation of a white solid which was filtered, washed
˚
0.71073 A). Data collection was performed at 143 K within the Θ
1
with H2O and dried in vacuo to afford 4 (0.54 g, 60%). H NMR
range of 2.14° Ͻ Θ Ͻ 25.03°.[27] A total of 55322 reflections were
integrated and corrected for Lorentz and polarization effects. A
correction for absorption effects and/or decay was applied during
the scaling procedure.[28] After merging (Rint ϭ 0.0214), 5378 [4481:
Io Ͼ 2σ(Io)] independent reflections remained and were all used to
(200 MHz, CDCl3, 25 °C): δ ϭ 8.15 (m, 2 H, Ar); 7.65Ϫ7.1 (m, 9
H, Ar), 7 (m, 2 H, Ar), 6.35 and 5.12 (2 s, 4 H, OH), 5.38 (s, 2 H,
HCN), 1.11, 1.06 and 1.04 (3 s, 24 H, Me) ppm.
Bis(radical) Derivative 5: A solution of NaIO4 (0.58 g, 2.73 mmol)
in water (10 mL) was added to a stirred suspension of 4 (0.54 g, refine 571 parameters. The structure was solved by a combination
0.91 mmol) in CH2Cl2 (35 mL). After 20 min, the violet phases
of direct methods[29] and difference Fourier syntheses.[30] The hy-
were separated and the aqueous phase further extracted with drogen atom positions were found in the difference Fourier map
CH2Cl2. The combined organic fractions were washed with H2O
(15 mL) and dried with anhydrous Na2SO4. Purification was
achieved by flash chromatography under N2 on silica gel (SiO2:
35Ϫ70 micron, 30 ϫ 2 cm, CH2Cl2/Me2CO (7:3)]. Crystalline
calculated from the model containing all non-hydrogen atoms. The
hydrogen positions were refined with individual isotropic displace-
ment parameters. Full-matrix least-squares refinements were car-
ried out by minimizing w(F2o Ϫ Fc2)2 and converged with R1 ϭ
170
Eur. J. Org. Chem. 2003, 167Ϫ171

H2O as a Chemical Link for Ferromagnetic Interactions Between Aminoxyl Units
FULL PAPER
[11]
G. Maruta, S. Takeda, R. Imachi, T. Ishida, T. Nogami, K.
Yamaguchi, J. Am. Chem. Soc. 1999, 121, 424Ϫ431.
M. M. Matsushita, A. Izuoka, T. Sugawara, T. Kobayashi, N.
Wada, N. Takeda, M. Ishikawa, J. Am. Chem. Soc. 1997, 119,
4369Ϫ4379.
J. Veciana, J. Cirujeda, C. Rovira, E. Molins, J. J. Novoa, J.
Phys. I 1996, 6, 1967Ϫ1986.
J. Cirujeda, E. Hernandez-Gasio, C. Rovira, J. L. Stanger, P.
Turek, J. Veciana, J. Mater. Chem. 1995, 5, 243Ϫ252.
N. Daro, J.-P. Sutter, M. Pink, O. Kahn, J. Chem. Soc., Perkin
Trans. 2 2000, 1087Ϫ1089.
N. Yoshioka, N. Matsuoka, M. Irisawa, S. Ohba, H. Inoue,
Mol. Cryst. Liq. Cryst. 1999, 334, 239Ϫ246.
A. Lang, Y. Pei, L. Ouahab, O. Kahn, Adv. Mater. 1996, 8,
60Ϫ62.
F. M. Romero, R. Ziessel, M. Drillon, J. L. Tholence, C.
Paulsen, N. Kyritsakas, J. Fischer, Adv. Mater. 1996, 8,
826Ϫ829.
T. Akita, Y. Mazaki, K. Kobayashi, J. Chem. Soc., Chem. Com-
mun. 1995, 1861Ϫ1862.
T. Akita, K. Kobayashi, Mol. Cryst. Liq. Cryst. 1997, 306,
257Ϫ264.
G. P. Schiemenz, H. Kaack, Justus Liebigs Ann. Chem. 1973,
1480Ϫ1493.
D. Hellwinkel, W. Krapp, Chem. Ber. 1978, 111, 13Ϫ41.
C. Rancurel, J.-P. Sutter, O. Kahn, P. Guionneau, G. Bravic, D.
Chasseau, New J. Chem. 1997, 21, 275Ϫ277.
F. Kanno, K. Inoue, N. Koga, H. Iwamura, J. Am. Chem. Soc.
1993, 115, 847Ϫ850.
A. K. Hoffman, W. M. Thomas, J. Am. Chem. Soc. 1959, 81,
580.
E. F. Landvatter, T. B. Rauchfuss, Chem. Commun. 1982, 1170.
Data Collection Software for Nonius KappaCCD devices, Delft,
The Netherlands, 2001.
Z. Otwinowski, W. Minor, Methods Enzymol. 1997, 276, 307.
SIR92: A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagli-
ardi, M. C. Burla, G. Polidori, M. Camalli, J. Appl. Crystallogr.
1994, 27, 435Ϫ436.
0.0394 [Io Ͼ 2σ(Io)], wR2 ϭ 0.0906 (all data), GOF ϭ 1.021 and
shift/error Ͻ 0.001. A disorder (51:49) at C58 and C59 clearly re-
solved. The residual electron density between Ϫ0.35 and 0.33 e/ A
showed no distinctive features. The drawings, listings and checking
for correctness were performed with the program PLATON.[31]
[12]
3
˚
[13]
[14]
[15]
[16]
[17]
[18]
CCDC-190126 (5a) and -190125 (5b) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge at www.ccdc.cam.ac.uk/conts/retrieving.html or from
the Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK [Fax: (internat.) ϩ 44-1223/336-033;
E-mail: deposit@ccdc.cam.ac.uk].
Acknowledgments
The TMR Research Network ERB-FMRX-CT-98Ϫ0181 of the
European Union, entitled ‘‘Molecular Magnetism: From Materials
toward Devices’’, is gratefully acknowledged for financial support.
[19]
[20]
[21]
[1]
K. Itoh, M. Kinoshita, Molecular magnetism: new magnetic
materials, Gordon and Breach, Amsterdam, 2000.
P. M. Lahti, Magnetic properties of organic materials, Marcel
[22]
[23]
[2]
Dekker Inc., New York, 1999.
[3]
[24]
[25]
V. A. Russell, M. D. Ward, Chem. Mater. 1996, 8, 1654Ϫ1666.
[4]
G. R. Desiraju, Chem. Commun. 1997, 1475Ϫ1482.
R. Improta, K. N. Kudin, G. E. Scuseria, V. Barone, J. Am.
[5]
Chem. Soc. 2002, 124, 113Ϫ120.
K. Yamaguchi, M. Okumura, J. Maki, T. Noro, H. Namimoto,
[6]
[26]
[27]
M. Nakano, T. Fueno, K. Nakasuji, Chem. Phys. Lett. 1992,
190, 353Ϫ360.
[7]
[28]
[29]
T. Kawakami, S. Takeda, W. Mori, K. Yamaguchi, Chem. Phys.
Lett. 1996, 261, 129Ϫ137.
F. M. Romero, R. Ziessel, M. Bonnet, M. Pontillon, E. Ressou-
[8]
che, J. Schweizer, B. Delley, A. Grand, C. Paulsen, J. Am.
Chem. Soc. 2000, 122, 1298Ϫ1309.
[30]
[31]
G. M. Sheldrick, Programs for Crystal Structure Analysis, re-
lease 97-2, University of Göttingen, Germany, 1998.
A. L. Spek, PLATON, A Multipurpose Crystallographic Tool,
Utrecht University, Utrecht, The Netherlands, 2001.
Received July 21, 2002
[9]
Y. Pontillon, T. Akita, A. Grand, K. Kobayashi, E. Lelievre-
´
Berna, J. Pecaut, E. Ressouche, J. Schweizer, J. Am. Chem. Soc.
1999, 121, 10126Ϫ33.
H. Heise, F. H. Köhler, F. Mota, J. J. Novoa, J. Veciana, J. Am.
Chem. Soc. 1999, 121, 9659Ϫ9667.
[10]
[O02417]
Eur. J. Org. Chem. 2003, 167Ϫ171
171