First example of a metal(0) carbonyl complex with a paramagnetic nitronyl nitroxide-substituted triphenylphosphine ligand

Article abstract of DOI:10.1039/a808113a

A stable metal(0) carbonyl complex with a paramagnetic nitronyl nitroxide-substituted triphenylphosphine ligand has been prepared and gives rise in the crystal to a 1D-array of ferromagnetically coupled S = 1/2 spins.

Full text of DOI:10.1039/a808113a

View Article Online / Journal Homepage / Table of Contents for this issue
First example of a metal(0) carbonyl complex with a paramagnetic
nitronyl nitroxide-substituted triphenylphosphine ligand
Corinne Rancurel,a Jean-Pascal Sutter,*a Thierry Le Hoer†,b Lahcene Ouahabb and
Olivier Kahn*a
L e t t e r
a L aboratoire des Sciences Moleculaires, Institut de Chimie de la Matiere Condensee de
Bordeaux, UPR CNRS n¡9048, Avenue du Dr. A. Schweitzer, F-33608 Pessac, France.
E-mail: jpsutter=icmcb.u-bordeaux.fr
b Groupe Materiaux Moleculaires, L aboratoire de Chimie du Solide et Inorganique Moleculaire,
UMR CNRS 6511, Universite Rennes I, 35042 Rennes, France
Recei¿ed 19th October 1998
A stable metal(0) carbonyl complex with a paramagnetic nitro-
nyl nitroxide-substituted triphenylphosphine ligand has been
prepared and gives rise in the crystal to a 1D-array of ferro-
magnetically coupled S = 1/2 spins.
1,3-dihydroxy-2-[4@-(diphenylphosphine)phenyl]-4,4,5,5-tetra-
methyl-imidazoline 1 and [Mo(CO) (thf)] to give the radical
5
precursor complex [Mo(CO) (1)] 3. The latter was subse-
5
quently oxidized by NaIO to a†ord 4. Route B is very conve-
4
nient since during the oxidation step to generate the radical,
the phosphorus atom is e†ectively protected from oxidation
by its bonding to the Mo centre.
In the course of our research on the tailoring of organic mol-
ecules likely to be building blocks for organic1 and metallo-
organic2 magnetic materials we recently synthesized a series of
novel open-shell phosphine derivatives.3 These compounds
comprise a triphenylphosphine core and a nitronyl nitroxide
moiety. Phosphines have been chosen in view of their ability
to be very good ligands for a large variety of metals in low
oxidation states. Moreover, depending on the metal involved
in the complex formation process, a selectivity in the coordi-
nation site of the phosphine derivative, i.e. P vs. nitroxyl O
atoms, is anticipated due to the characteristics of the hetero-
atoms likely to interact with the metal centre. We now report
that (4-nitronyl nitroxide-phenyl)diphenylphosphine 2, can be
An X-ray crystal structure¤ established that 4 is indeed a
Mo(CO) moiety substituted by phosphine 2. An ORTEP
5
view is presented in Fig. 1 with selected bond lengths and
angles. Interestingly it appears that in the solid state the nitro-
nyl nitroxide moiety tends to be coplanar with the substituted
phenyl. This should ensure spin polarization over the whole
molecule and especially on the adjacent phenyl group.4 A view
of the crystal packing shows that the nitronyl nitroxide groups
are aligned in a head to tail fashion along the b axis to form
pseudo chains (Fig. 2). It is noticeable that within these align-
ments the shortest intermolecular distances involving the
nitroxyl O atoms are found with H in the ortho position of the
neighbouring nitronyl nitroxide substituted phenyl group
[O(6)É É ÉH(10) 2.66 Ó and O(7)É É ÉH(8) 2.81 Ó] and the H atom
in the meta position [O(6)É É ÉH(11) 2.86 Ó and O(7)É É ÉH(7)
2.821 Ó]. The separation between two neighbouring nitroxide
O-atoms O(6) (x, y, z) and O(7) ([x ] 1, y [ 1, [z) is 3.892
reacted with a metal carbonyl compound, [Mo(CO) (thf)], to
5
a†ord a stable Mo(0) complex, 4, displaying a phosphine
ligand with an unpaired electron. The syntheses and magnetic
properties of both 2 and 4 as well as the X-ray structure of 4
are presented.
The synthesis of phosphine 2 has already been brieÑy
reported.3 However, considerable variations in the efficiency
of the Ðnal oxidation step to produce the radical were
observed. We found that the use of Ag O instead of PbO as
2
2
Ó. The shortest interchain distances involving O(6) and O(7)
are to H(21) (2.752 Ó) and to H(27) (3.085 Ó), respectively, in
the meta position of non-substituted phenyl groups. However,
the interchain separation of the spin carriers is large, and the
2
2
oxidant gave reproducibly 2 from 1 in yields up to 70% [eqn.
(1)]. The complex [Mo(CO) (2)], 4, was synthesized according
5
to two di†erent routes (routes A and B, Scheme 1). Following
route A, the reaction of phosphine 2 with the adduct
[Mo(CO) (thf)] in thf gave 4 as a stable deep-blue compound
5
which could be puriÐed by column chromatography. The
alternative synthesis (route B) consists of the reaction of
Scheme 1
New J. Chem., 1999, 1333È1335
1333

View Article Online
Fig. 3 Temperature dependence of s T for the phosphine 2 and the
M
complex 4. The solid line represents the analytical Ðt based on a chain
Fig. 1 View of molecule 4 showing the atom labelling scheme. Selec-
ted bond lengths (Ó): N(1)ÈO(7) 1.266(16), N(2)ÈO(6) 1.304(17), N(1)È
C(12) 1.37(2), N(2)ÈC(12) 1.32(2), C(12)ÈC(9) 1.479(9), C(6)ÈP 1.835(6),
PÈMo 2.546(2), C(3)ÈMo 2.005(8), MoÈ(other CO) 2.031 (mean). Angle
between mean planes N(1)ÈC(12)ÈN(2) and C(9)ÈC(10)ÈC(8) 3.9¡.
model of S \ 1/2 spins in ferromagnetic interaction.
have been analysed in terms of a one-dimensional Heisenberg
chain model for ferromagnetically coupled S \ 1/2 spins.7
This model leads to an intrachain interaction parameter equal
to J \ 0.26 cm~1 (with H \ [JS É S ).
Compound 4, obtained in one step from phosphine 2, illus-
trates the possibility of switching the solid state magnetic
behaviour, from antiferromagnetic in 2 to ferromagnetic in 4,
by small chemical alterations of a molecular brick. An inter-
esting aspect of complex 4 consists of its overall tetrahedral
geometry. However, the spin carrier is only on one phenyl
group, i.e. in one direction. In order to increase the magnetic
dimensionality of the network, polyradical species have to be
considered.
magnetic interaction is assumed to occur mainly within the
molecules aligned along the b axis.
The temperature (T ) dependence of the molar magnetic sus-
i
i`1
ceptibility (s ) of compounds 2 and 4 was measured in the
M
temperature range 2È300 K using a SQUID susceptometer,
Fig. 3 shows the s T vs. T plots. At high temperature (50È
M
300 K), for both 2 and 4, s T is ca. 0.37 cm3 K mol~1, the
M
expected value for non-correlated S \ 1/2 spins. As the tem-
perature is lowered from ca. 50 K down to 2 K the com-
pounds do not behave in the same way. For phosphine 2,
s T decreases very rapidly as T is lowered whereas for
M
complex 4, s T increases more and more rapidly to reach
M
0.44 cm3 K mol~1 at 2 K, the lowest temperature investigated.
These behaviours are characteristic for intermolecular anti-
ferromagnetic and ferromagnetic interactions in 2 and 4,
respectively. The intermolecular O(6)É É ÉC(10) and O(7)É É ÉC(8)
proximities are anticipated to play an important role in the
ferromagnetic interactions observed for complex 4 at low tem-
perature. Taking into account a McConnell-type spin polar-
ization mechanism,5 the through space magnetic interaction
between the nitroxide moiety and the ortho position of the
neighbouring phenyl group should indeed lead the spins to
align ferromagnetically.6 The magnetic data of compound 4
Experimental
Reagent grade chemicals were used as received. THF was dis-
tilled from sodium benzophenone ketyl under nitrogen, prior
to use. Silica gel (70È230 mesh) was used for chromatographic
separations.
Synthesis of 2
Ag O (0.714 mmol, 165 mg) was added to a suspension of
2
13 (0.476 mmol, 200 mg) in CH Cl (30 mL). After about 30
2
2
min the solution turned blue. The reaction was monitored by
TLC (silica, CH Cl ) and stopped before the formation of a
phosphine oxide derivative, indicated by the presence of a
blue spot staying at the reference line. The reaction mixture
was Ðltered through Celite} and the solvent removed in vacuo
2
2
to give 2 as a blue oil which crystallised by addition of Et O.
2
Anal. Calc. (found) for C
H
N P: C 71.93 (72.04), H 6.28
25 26
2
(6.36), N 6.71 (6.60%). IR (KBr, cm~1): 1434 m, 1419 s, 1388
m, 1363 vs, 751 m, 697 vs, 508 m. UVÈVis (CH Cl ): j/nm (e,
2
2
dm3 mol~1 cm~1), 632(360), 595(360), 375(10 300), 300(14 400),
272(14 600).
Synthesis of 4 (route B)
A solution of [Mo(CO) (thf)]8 in thf, prepared from
5
Mo(CO) (0.583 mmol, 154 mg), was added to a solution of 1
6
Fig. 2 View of the crystal lattice along the b direction showing the
(0.697 mmol, 293 mg) in thf. After 30 min stirring, the solvent
spin carriers alignment [the Mo(CO) moieties along the c axis have
was removed in vacuo to give 3 as a white powder which was
5
been omitted]. Selected intermolecular distances (Ó): O(6)É É ÉO(7)
dissolved in CH Cl , and NaIO (0.777 mmol, 166 mg) in
3.892(8), O(6)É É ÉC(10) 3.33(2), O(6)É É ÉH(10) 2.66, O(6)É É ÉC(11) 3.40(2),
O(6)É É ÉH(11) 2.86, O(7)É É ÉC(7) 3.472(19), O(7)É É ÉH(7) 2.821,
O(7)É É ÉC(8) 3.44(2), O(7)É É ÉH(8) 2.81. Mean angle between two adja-
cent nitronyl nitroxide groups 0¡.
2
2
4
H O added, leading to a blue solution. Compound 4 was
2
extracted with CH Cl
and chromatographed (Silica
2
2
gel/CH Cl ) to give 4, as a blue crystalline solid, after removal
2
2
1334
New J. Chem., 1999, 1333È1335

View Article Online
2 D. Lezno†, C. Rancurel, J. P. Sutter, S. Golhen, L. Ouahab, S. J.
Rettig and O. Kahn, Mol. Cryst., L iq. Cryst. (in press); J. P. Sutter,
M. L. Kahn, S. Golhen, L. Ouahab and O. Kahn Chem. Eur. J.,
1998, 4, 571; J. P. Sutter, M. Fettouhi, L. Li, C. Michaut, L.
Ouahab and O. Kahn, Angew. Chem., Int. Ed. Engl., 1996, 35,
2113; Y. Pei, A. Lang, P. Bergerat, O. Kahn, M. Fettouhi and L.
Ouahab, Inorg. Chem. 1996, 35, 193.
of the solvent [90 mg; 27% based on Mo(CO) ]. Anal. Calc.
6
(found) for C
H
N PMo: C 55.14 (54.35), H 4.01 (4.37), N
30 26
2
4.29 (4.07%). IR (CH Cl , cm~1): 2073 m, 1990 w, 1946 vs.
2
2
UVÈVis (CH Cl ): j/nm (e/dm3 mol~1 cm~1), 597(350),
2
2
377(12 100), 360(8200).
3 C. Rancurel, J. P. Sutter, O. Kahn, P. Guionneau, G. Bravic and
D. Chasseau, New J. Chem., 1997, 21, 275.
Notes and references
¤ Crystal data for 4: C
H N O PMo, (M \ 653.4, monoclinic,
4 T. Akita, Y. Mazahiro, K. Kobayashi, N. Koga and H. Iwamura,
J. Org. Chem., 1995, 60, 2092.
30 26
2 7
space group C2, a \ 14.860(10), b \ 13.641(2), c \ 15.399(7) Ó,
b \ 103.81(2)¡, V \ 3031(3) Ó3, Z \ 4, k \ 0.533 mm~1. 3937 ReÑec-
tions were measured (3794 unique), Ðnal R1 [I [ 2p(I)] \ 0.0455,
wR2 \ 0.0938. The intensities were collected with an Enraf-Nonius
CAD4 di†ractometer using monochromated Mo-Ka radiation. The
structure was solved by direct methods (SHELXS-86)9 and reÐned by
full-matrix least squares methods (SHELXL-97).10 CCDC reference
number 440/071. See http://www.rsc.org/suppdata/njc/1998/1333/ for
crystallographic Ðles in .cif format.
5 H. M. McConnel, J. Chem. Phys., 1963, 39, 1910; H. Iwamura,
Adv. Phys. Org. Chem., 1990, 26, 179253.
6 M. Okumura, W. Mori and K. Yamaguchi, Chem. Phys. L ett.,
1994, 219, 36; Mol. Cryst. L iq. Cryst., 1993, 232, 35.
7 G. A. Baker, Jr., G. S. Rushbrooke and H. E. Gilbert, Phys. Rev.
A, 1964, 135, 1272.
8 W. Strohmeier and K. Gerlach, Chem. Ber., 1961, 94, 398.
9 G. M. Sheldrick, SHELXS-86. Program for the Solution of Crystal
Structures, University of Gottingen, Germany, 1990.
1 J. P. Sutter, A. Lang, O. Kahn, C. Paulsen, L.Ouahab and Y. Pei,
J. Magn. Magn. Mater., 1997, 171, 147; A. Lang, Y. Pei, L.
Ouahab and O. Kahn, Adv. Mater., 1996, 8, 60; J. P. Sutter, N.
Daro, S. Golhen, L. Ouahab and O. Kahn, Mol. Cryst. L iq. Cryst.,
in press.
10 G. M. Sheldrick, SHELXL-97. Program for the ReÐnement of
Crystal Structures, University of Gottingen, Germany, 1997.
L etter 8/08113A
New J. Chem., 1999, 1333È1335
1335