Synthesis of nitronyl- and imino-nitroxide-triradicals interconnected by phenyl ethynyl spacer

Article abstract of DOI:10.1016/S0040-4039(99)02218-2

The design and synthesis, as well as solid state and ESR characterisation of three new triradicals based on phenyl ethynyl backbone bearing nitronyl-nitroxide and imino-nitroxide groups, are reported. The synthesis of these compounds was based on palladium-coupling reactions between alkynyl and bromo derivatives. An ESR study showed magnetic interactions between the radicals through the phenyl ethynyl-coupling unit in solution. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02218-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1015–1018
Synthesis of nitronyl- and imino-nitroxide-triradicals
interconnected by phenyl ethynyl spacer
Laure Catala,a,b Philippe Turek,a,∗ Jacques Le Moigne,b André De Cian c and
Nathalie Kyritsakas c
aInstitut Charles Sadron, 6, rue Boussingault, CNRS, F-67083 Strasbourg, France
bInstitut de Physique et Chimie des Matériaux de Strasbourg, UMR 7504, 23, rue du Loess, F-67037 Strasbourg, France
cLaboratoire de Cristallochimie et Chimie Structurale, associé au CNRS, Université Louis Pasteur, Institut Le Bel,
4, rue Blaise Pascal, F-67000 Strasbourg, France
Received 31 August 1999; accepted 18 November 1999
Abstract
The design and synthesis, as well as solid state and ESR characterisation of three new triradicals based on
phenyl ethynyl backbone bearing nitronyl-nitroxide and imino-nitroxide groups, are reported. The synthesis of
these compounds was based on palladium-coupling reactions between alkynyl and bromo derivatives. An ESR
study showed magnetic interactions between the radicals through the phenyl ethynyl-coupling unit in solution.
© 2000 Published by Elsevier Science Ltd. All rights reserved.
In the area of molecular magnetism, purely organic materials based on radicals are still under active
investigation.1 High spin organic molecules may be obtained through interconnection of paramagnetic
centres using an active magnetic coupling unit (MCU). In order to achieve ferromagnetic coupling, the
positioning of unpaired electrons relatively to the MCU and the nature of the latter must fulfil electronic
as well as topological rules.2 The assessment of the nature and the strength of the magnetic interactions
through the MCU units requires model systems based on oligoradicals which may be of interest for
understanding more complex polyradicals.3
Here, we report the design and synthesis, as well as structural determination and ESR characterisation
of three new triradicals 1, 2 and 3, based on phenyl ethynyl backbone bearing nitronyl-nitroxide and
imino-nitroxide groups. The homotriradical 1 is based on three nitronyl-nitroxide units, whereas the
other two hetero systems 2 and 3 are based on a combination of nitronyl-nitroxide and imino-nitroxide
moieties. The phenyl ethynyl spacer has been previously used as a connector between two identical
radicals.4
For the synthesis of compounds 1 and 2, three different strategies based on palladium-catalysed co-
upling reactions5 between alkynyl and bromo derivatives were investigated (Scheme 1). The first strategy
was based on the Pd-coupling of compounds 8 and 9 both bearing the N,N0-dihydroxyimidazoline moiety.
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Published by Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02218-2
tetl 16146

1016
Scheme 1.
Compound 8 was obtained in four steps starting from the commercially available isophtalic acid 4.
Bromination of 4 by bromine in sulfuric acid in the presence of silver sulfate afforded the compound 5
in 86% yield.6 The reduction of the latter by BH3·THF gave the diol 6 in 93% isolated yield.6 Oxidation
of 6 using pyridinium chlorochromate at room temperature afforded the dialdehyde 7 in 82% yield.7
Condensation of 7 in methanol with a slight excess (2.3 equivalents) of N,N0-dihydroxy-2,3-diamino-2,3-
dimethylbutane, prepared in two steps following the published procedure,8 gave the compound 8 in high
yield. It is worth noting that the high yield of 90–95% for the compound 8 could only be obtained upon
slow bubbling of argon into the reaction mixture at room temperature and over a period of 3 days in order
to avoid the precipitation of the monocondensed product. On the other hand, compound 9 was obtained
by reacting at room temperature N,N0-dihydroxy-2,3-diamino-2,3-dimethylbutane with compound 12 in
MeOH (91% yield). The latter was obtained upon deprotection of 11 (94% yield) using K2CO3 in MeOH
at 25°C. Compound 11 was prepared by a Pd-coupling reaction between trimethylsilylacetylene (TMSA)
with the commercially available 4-bromobenzaldehyde 10 in 93% yield. Dealing with the palladium
catalyst used for the coupling reaction, in the case of the synthesis of diimino-nitroxide derivatives,
it has been previously observed4 that when PdCl2(PPh3)2 was prepared in situ by reacting PdCl2,
PPh3 in Et3N in the presence of Cu(OAc)2, the dehydration of N,N0-dihydroxyimidazolines into N-
hydroxyimidazolines systematically occurred. Interestingly, we found that for the coupling reaction of 8
with 9 (48 h, 80°C), when commercially available PdCl2(PPh3)2 complex was used in the presence of CuI
in Et3N, the side dehydration reaction could be considerably reduced. However, the reaction afforded a
mixture mainly containing the desired compound 13 and the mono dehydrated derivative. Unfortunately,
due to low solubility, they could not be separated. The oxidation by MnO2 at room temperature of the
mixture in suspension in CH2Cl2 gave a mixture of radicals which afforded both compounds 1 and 2 after
chromatography (SiO2, CH2Cl2:AcOEt:MeOH, 80:19:1) in 11 and 15% overall yields, respectively.
In order to increase the yield of compound 1, another strategy based on the preparation of the

1017
trialdehyde 16 by the Pd-coupling reaction prior to the introduction of the N,N0-dihydroxyimidazoline
functionality was exploited. The coupling of aldehydes 10 and 15 leading to the trialdehyde 16 was
achieved in 74% yield in the presence of 10% Pd(PPh3)4 in a 1:10 mixture of diisopropylamine:benzene.
Before this, compound 14 was obtained in 94% yield, by Pd-coupling reaction between TMSA and 7 in
diisopropylamine:THF, 1:5, mixture and in the presence of PdCl2(PPh3)2 and CuI. Deprotection of 14 by
K2CO3 in MeOH afforded the compound 15 in 99% yield. Using the same conditions as described above
for the preparation of 8, upon condensation of compound 16 with 3.3 equivalents of N,N0-dihydroxy-
2,3-diamino-2,3-dimethylbutane in MeOH, the precursor compound 13 could be obtained in 90% yield.
The oxidation at room temperature of the pure compound 13 by MnO2 in CH2Cl2 afforded the desired
compound 1 in 74% yield.
A third strategy, already employed for the synthesis of pyridine containing nitronyl-nitroxide groups,9
consisting in coupling preformed nitronyl-nitroxide derivatives such as 17, 19 or 18 and 20 was also
attempted. However, using Pd(PPh3)4 as the catalyst in an amine:benzene, 1:10, mixture, the coupling
of either 17 with 19 or 18 with 20 failed. Attempts by varying the nature of the amine (triethylamine,
pyrrolidine, diisopropylamine) were unsuccessful.
Dealing with the synthesis of the heterotriradical 3 bearing two imino-nitroxide and one nitronyl-
nitroxide groups, following the first synthetic strategy presented above, it could be obtained in 22% yield
upon coupling precursors 9 and 21 followed by oxidation by MnO2. Compound 21 was obtained in 85%
yield by catalytic dehydration of 8 using 15% SeO2 in methanol.10
All triradicals 1, 2 and 3 were characterised by IR, UV–visible spectroscopy, FAB spectrometry and
elemental analysis. Furthermore, they were also characterised in the solid state by X-ray diffraction on
single-crystal obtained upon slow evaporation of the desired compound in CH2Cl2/hexane mixture.11 All
three compounds were isostructural (monoclinic, P21/n space group). For compounds 1 and 3, the solid
state structure of the molecular units are presented in Fig. 1.
Fig. 1. Structure of the molecular units of: (a) compound 1; and (b) compound 3 (H atoms have been omitted for the sake of
clarity)
The conjugated ethynyl diaryl skeleton is close to planarity with a tilt angle of 11–12°, and the cyclic
radical units C1, C2 and C3 (see Fig. 1) are tilted from the plane of ethynyl diaryl with torsion angles of
46° (C1), 26° (C2) and 18° (C3) for compound 1 and 37° (C1), 22° (C2) and 15° (C3) for compound 3.
In the crystal, the molecular units are arranged as head-to-tail dimers with a distance of 3.6 Å between
the aromatic planes.
Complete oxidation of all three compounds was confirmed by SQUID measurements at 0.5 T, which
gave values of 1.15–1.18 emu.K.mol−1 at 298 K, close to the theoretical value of 1.125 emu.K.mol−1
.
ESR spectra of compounds 1, 2 and 3 were recorded at room temperature in diluted CH2Cl2:acetone,
1:1, solution (10−4 M). For both 1 and 2 the expected hyperfine patterns were observed, with 13 lines
and 12 lines, respectively, indicating magnetic interactions through the MCU within the strong exchange

1018
coupling limit J»a (with J=exchange coupling constant and a=hyperfine coupling constant). However, for
3, the presence of both weak and strong exchange processes was observed. The evolution of the magnetic
susceptibility χ has been determined on the 4–120 K temperature range by ESR studies on dilute frozen
solutions containing the triradicals 1 and 3. The χT versus T plots obtained revealed the presence of
ferromagnetic interactions in both cases, with weak coupling through the phenyl ethynyl coupler (7 K for
1, 0.7 K for 3), and larger magnetic exchange between two radicals through the m-phenylene part (23 K
for 1, 6.5 K for 3).
In conclusion, three new triradicals based on phenyl ethynyl backbone bearing nitronyl-nitroxide and
imino-nitroxide groups were prepared and structurally characterised in the solid state. ESR studies revea-
led the presence of magnetic interactions between the radical moieties in dilute solution, demonstrating
the ferromagnetic coupling efficiency of this phenyl ethynyl spacer. Magneto-structural correlations as
well as detailed ESR study of the intramolecular coupling processes in frozen solutions will be reported
elsewhere.
Acknowledgements
We thank CNRS and COST D4 for the financial support. Thanks to R. Poinsot for the SQUID
measurements.
References
1. Rajca, A. Chem. Rev. 1994, 94, 871–893; Molecular Magnetism: From Molecular Assemblies to the Devices; Coronado, E.;
Delhaès, P.; Gatteschi, D.; Miller, J. S. NATO ASI Series, Kluwer Academic Publishers, Dordrecht, 1996, 321; Magnetic
Properties of Organic Materials; Lahti, P. M., Ed.; Marcel Dekker: New York, 1999.
2. Longuet-Higgins, H. C. J. Chem. Phys. 1950, 18, 265–274; Borden, W. T.; Davidson, E. R. J. Am. Chem. Soc. 1977, 99,
4587–4594; Ovchinnikov, A. A. Theor. Chim. Acta 1978, 47, 297–304.
3. Sasaki, S.; Iwamura, H. Chem. Lett. 1992, 1759–1762; Yoshioka, N.; Lahti, P. M.; Kaneko, T.; Kuzumaki, E.; Nishide, H. J.
Org. Chem. 1994, 59, 4272–4276; Nishide, H.; Kaneko, T.; Nii, T.; Katoh, K.; Tsuchida, E.; Lahti, P. M. J. Am. Chem. Soc.
1996, 118, 9695–9704; Nishide, H.; Hozumi, Y.; Nii, T.; Tsuchida, E. Macromolecules 1997, 30, 3986–3991.
4. Wautelet, P.; Bieber, A.; Turek, P.; Le Moigne, J.; André, J.-J. Mol. Cryst. Liq. Cryst. 1997, 305, 55–67; Wautelet, P. Doctor
Thesis of the University Louis Pasteur; Strasbourg, France, 1996.
5. Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975, 50, 4467–4470; Cassar, L. J. Organomet. Chem. 1975, 93,
253–257.
6. Wytko, J. A.; Weiss, J. Tetrahedron Lett. 1991, 32, 7261–7264.
7. Netzke, K.; Snatzke, G. Chem. Ber. 1989, 122, 1365–1370.
8. Sayre, R. J. Am. Chem. Soc. 1955, 77, 6689–6690; Lamchen, M.; Mittag, T. W. J. Chem. Soc. 1966, 2300–2303.
9. Romero, F.; Ziessel, R.; De Cian, A.; Fischer, J.; Turek, P. New J. Chem. 1996, 20, 919–924; Romero, F.; Ziessel, R.
Tetrahedron Lett. 1999, 40, 1895–1898.
10. Ulrich, G.; Ziessel, R. Tetrahedron Lett. 1994, 35, 1215–1218.
11. Selected data for 1: UV–vis (CH2Cl2) λmax, nm (ε, M−1.cm−1), 592 (1130); IR, cm−1 (KBr) 1363 (νN–O); anal. calcd for
C35H43N6O6·CH2Cl2: C, 59.34; H; 6.22; N; 11.53; found: C, 59.11; H, 6.20; N, 11.28; FAB+ m/z 645 [M+H]+; compound
2: UV–vis 595 (700), 450 (712); IR: 1365 (νN–O), 1549 (νC_N); anal. calcd for C35H43N6O5·CH2Cl2: C, 60.67; H; 6.36;
N; 11.79; found: C, 60.55; H, 6.39; N, 11.65; FAB+ m/z 629 [M+H]+; compound 3: UV–vis 602 (560), 445 (1380),
IR 1363 (νN–O), 1544 (νC_N); anal. calcd for C35H43N6O4·CH2Cl2: C, 62.06; H; 6.51; N; 12.06; O, 9.18; found: C,
62.14; H, 6.52; N, 11.99; O, 9.30; FAB+ m/z 613 [M+H]+; X-ray data for 1: C35H43N6O6.CH2Cl2, monoclinic, P21/n,
a=12.254(1), b=18.293(2), c=16.892(2), β=93.877(7), Z=4, R=0.061, Rw=0.083; 3: C35H43N6O4.CH2Cl2, monoclinic,
P21/n, a=12.414(2), b=18.2021(9), c=16.643(1), β=93.712(8), Z=4, R=0.084, Rw=0.094. Atomic coordinates, bond lengths,
angles and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre.