An unprecedented bridging phenoxyl radical in dicopper(II) complexes: Evidence for an S = 3/2 spin state

Article abstract of DOI:10.1002/anie.200461462

(Chemical Equation Presented) Cooperativity between metal centers and a bridging radical ligand, reflected by ferromagnetic exchange coupling, was studied with the dicopper complex shown. The stability, and thus the reactivity of the complex, is finely tu

Full text of DOI:10.1002/anie.200461462

Communications
Metal–Radical Complexes
An Unprecedented Bridging Phenoxyl Radical in
Dicopper(ii) Complexes: Evidence for an S = 23
Spin State**
Fabien Michel, Stꢀphane Torelli, Fabrice Thomas,*
Carole Duboc, Christian Philouze, Catherine Belle,
Sylvain Hamman, Eric Saint-Aman, and Jean-
Louis Pierre
The cooperativity between a radical and a transition-metal
ion in a single molecular entity is demonstrated in highly
sophisticated biocatalysts.[1] Galactose oxidase, which has an
[*] F. Michel, Dr. S. Torelli, Dr. F. Thomas, Dr. C. Philouze, Dr. C. Belle,
Dr. S. Hamman, Pr. J.-L. Pierre
Laboratoire d’Etudes Dynamiques et Structurales de la Sꢀlectivitꢀ
Equipe de Chimie Biomimꢀtique
UMR CNRS 5616, ICMG FR CNRS 2607
Universitꢀ J. Fourier
BP 53, 38041 Grenoble Cedex 9 (France)
Fax: (+33)4-7651-4836
E-mail: fabrice.thomas@ujf-grenoble.fr
Dr. C. Duboc
High Magnetic Field Laboratory
CNRS-MPI UPR 5021
BP 166, 38042 Grenoble Cedex 9 (France)
Pr. E. Saint-Aman
LEOPR, UMR CNRS 5630, ICMG FR CNRS 2607
Universitꢀ J. Fourier
BP 53, 38041 Grenoble Cedex 9 (France)
[**] The authors thank the Institute of Metals in Biology of Grenoble,
which provided facilities for the EPR studies, and Dr. Stephane
Mꢀnage for technical assistance with these experiments and fruitful
discussions.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
438
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200461462
Angew. Chem. Int. Ed. 2005, 44, 438 –441

Angewandte
Chemie
active site composed of a tyrosyl radical magnetically coupled
to a cupric ion, nicely exemplifies this synergy. In this example
the two redox centers work together to promote the two-
electron oxidation of primary alcohols to give aldehydes.[2]
For a better understanding of the metal–radical interaction
(and its biological relevance), transition-metal complexes
involving one or two coordinating radicals have been
characterized recently.[3] So far no complexes with a bridging
radical have been reported. We describe herein an unprece-
dented metal–radical system, a dicopper(ii) m-phenoxyl com-
plex. This was achieved using the dicopper(ii) complexes of
the dinucleating ligand HL. Its oxidative chemistry is
compared to that of the copper(ii) complex of the mono-
nucleating analogue HL’.
Figure 1. X-ray crystal structures of 1–3 (ORTEP diagrams, thermal
ellipsoids at the 30% probability level). Selected bond lengths [ꢁ] and
angles [8]: [L(CH3CN)2Cu2]3+ (1): Cu1-O1 2.170(2), Cu2-O1 2.150(2),
Cu1-Cu2 4.059(5), Cu1-O1-Cu2 140.0(1); [L(mOH)Cu2]2+ (2):[5] Cu1-O1
1.997(3), Cu2-O1 2.039(3), Cu1-Cu2 2.980(9), Cu1-O-Cu2 95.2(2);
[L’Cu]2 (3): Cu1-O1 2.077(2), Cu1-O2 1.923(2), Cu2-O1 1.917(2),
Cu2-O2 2.303(2), Cu1-Cu2 3.123(5), O1-Cu1-O2 82.72(7), O1-Cu2-O2
77.05(7).
When two equivalents of Cu(ClO4)2·6H2O and HL were
mixed in acetonitrile in the presence of one equivalent of
NEt3, complex [L(CH3CN)2Cu2]3+ (1) was obtained as its
2+
ClO4 salt.[4] Its crystal structure (Figure 1) shows that the
ꢀ
ligand binds two copper(ii) atoms having square-pyramidal
geometry through coordination to three nitrogen atoms from
the ligand (two pyridyl N atoms and one tertiary amino
group) and one from an exogenous acetonitrile molecule. The
coordination sphere is completed by an oxygen atom of a m-
phenolato group, which bridges the two metal centers having
a separation of 4.06 ꢀ. The use of HL and two equivalents of
both Cu(ClO4)2·6H2O and NEt3 affords complex
ꢀ
[L(mOH)Cu2]2+ (2) as its ClO4 salt.[5,6] Its crystal structure
(Figure 1) has been previously reported:[5] the coordination
sphere of the copper atoms contains three nitrogen atoms
from the ligand (two pyridyl and one tertiary amino group)
completed by two oxygen atoms (from one phenolato and one
hydroxo groups) which bridge the metal centers. The two
copper ions have trigonal-bipyramidal coordination geome-
ꢀ
try, with a Cu Cu distance of 2.98 ꢀ. Addition of one
equivalent of Cu(CF3SO3)2 or Cu(ClO4)2·6H2O to HL’ in
acetonitrile in the presence of one equivalent of NEt3 affords
Figure 2. Electronic spectra of 0.1 mm solutions of 1 (solid lines) and
electrochemically generated 1C+ (dotted lines) recorded in CH3CN
(0.01m TBAP) at 233 K (l=1 cm). Inset: Cyclic voltamogram (under Ar
atmosphere) of 1 (1 mm in CH3CN, 0.1m TBAP) at 298 K; scan rate:
2+
ꢀ
ꢀ
the dimer [L’Cu]2
(3) as its CF3SO3 or ClO4 salt
(Figure 1).[4] In 3 each copper atom is coordinated by two
pyridyl nitrogens, one amino nitrogen, and two m-phenolato
oxygens (one from each ligand), and the intermetallic
distance is 3.12 ꢀ. This dimer is relatively stable, as reflected
0.1 Vsꢀ1
.
by its Kpy value of 0.297mꢀ1 [7]
.
All the ligands thus afford
higher wavelengths attributed to d–d transitions.[8] The X-
band EPR spectrum of 1 in CH3CN exhibits a DMS = ꢁ 2
transition at g = 4.3, associated with spin triplet resonances at
g = 2.24 and 1.99, indicating that the two copper(ii) atoms are
weakly ferromagnetically exchange-coupled. The zero-field
splitting (zfs) parameters (D = 0.036 cmꢀ1, E = 0 cmꢀ1) and
copper(ii) complexes involving at least one coordinating m-
phenolato unit.
The UV/Vis spectra of 1 (Figure 2), 2, and 3 exhibit a m-
phenolato-to-copper charge-transfer (CT) transition in the
region between 450 and 550 nm, and a broad absorption at
Angew. Chem. Int. Ed. 2005, 44, 438 –441
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
439

Communications
gk = 2.30, g? = 2.07 were obtained from simulation. Using the
point dipole approximation,[9] we calculated a 4.5-ꢀ separa-
tion between the two copper(ii) atoms, a value in reasonable
agreement with the X-ray structure analysis. Complex 2 is
EPR-silent, indicating that the two copper(ii) atoms are
strongly antiferromagnetically exchange-coupled, due to
optimal magnetic orbital overlap.[5] Complex 3 exhibits a
ferromagnetic coupling between the two copper atoms similar
to that in 1 but with larger zfs parameters (D = 0.080 cmꢀ1,
E = 0.027 cmꢀ1, g1 = 2.04, g2 = 2.06, g3 = 2.12), as is expected
for a complex with a shorter intermetallic separation. (The
1
2
complex [L’(py)Cu]+ exhibits a SCu
mononuclear copper(ii) complex).
=
signal typical for a
The cyclic voltammograms of 1 (Figure 2 inset) and 2 in
CH3CN (0.1m tetrabutylammonium perchlorate (TBAP))
display a reversible one-electron wave in the positive region
of potentials (E1/2 = 0.36 V vs Fc+/Fc, DEp = 0.09 V and E1/2
0.45 V, DEp = 0.09 V for 1 and 2, respectively), attributed to
=
+
Figure 3. a) X-Band EPR spectrum of electrochemically generated 1C
(1 mm in CH3CN, 0.1m TBAP): T=4 K, modulation 0.393 mT/
the oxidation of the m-phenolato moiety into a m-phenoxyl
100 kHz, frequency 9.44 GHz, power 1 mW. Solid lines: experimental,
dotted lines: simulation. b) 115-GHz EPR spectrum 1C+ (5 mm in
CH3CN, 0.1m TBAP) at T=5 K. Solid lines: experimental, dotted lines:
simulation (parameters are given in the text).
radical.[10] This signal is irreversible both for 3 (at Ep
=
a
0.45 V) and [L’(py)Cu]+ (at Epa = 0.14 V), showing that the
radical evolves on the experiment timescale. Complexes 1–3
are thus oxidized in the same potential range, highlighting
that the structure of the complex affects weakly the oxidation
potential of the m-phenolato ligand. However, [L’(py)Cu]+ is
oxidized at a potential 0.3 V lower, which can be explained by
the lower electron density at the oxygen atom of a m-
phenolato ligand compared to that of a phenolate group. In
addition, we have previously shown that species such as 3 do
not retain their dimeric structure upon oxidation: if they are
stable enough to be generated, the phenoxyl radical species
evolve towards the corresponding monomers.[11] Thus, only 1
and 2 could be oxidized to give complexes with bridging
phenoxyl radicals.
Although the DMS = ꢁ 3 transition is weak (its intensity
decreases as the temperature increases), these results dem-
+
onstrate that 2C also exhibits a quartet ground state.
In summary, m-phenoxyl dicopper(ii) complexes in which
the two copper(ii) and the radical spins are ferromagnetically
exchange-coupled could be obtained. Their stability, and thus
their reactivity, is finely tuned by both the nuclearity of the
complex and the nature of coordinating solvent. Since tyrosyl
residues are ubiquitous in metalloenzymes, such species could
be biologically relevant. On the other hand, the chemical
reactivity of these m-phenoxyl dicopper(ii) species which
formally contain “three oxidizing equivalents” constitute a
promising and fascinating area for the studies of new chemical
properties.
+
+
The one-electron-oxidized complexes 1C and 2C exhibit
similar UV/Vis features: p–p* transitions typical of phenoxyl
radicals[12] are observed at 440 nm (3330mꢀ1 cmꢀ1) and 600 nm
+
(560mꢀ1 cmꢀ1) for 1C
(Figure 2), and at 445 nm
(4730mꢀ1 cmꢀ1)[13] and 600 nm (1000mꢀ1 cmꢀ1)[13] for 2C .
+
Received: July 28, 2004
From the decay of the former band, a half-life of 22.3 min
+
+
was obtained for 1C at 298 K, and of less than 20 s for 2C at
290 K.
Keywords: bridging ligands · copper · EPR spectroscopy ·
.
magnetic properties · radicals
The 9.4-GHz EPR spectrum of the electrochemically
+
generated 1C recorded at 4 K (Figure 3) exhibits a DMS = ꢁ 3
transition at g = 8, typical of an S = 32 spin state, whose
intensity is proportional to the reciprocal of the absolute
temperature (Tꢀ1). This quartet thus corresponds to the
ground state.[14] The zfs parameters D = ꢀ0.056 cmꢀ1 and E =
0.015 cmꢀ1 (gk = 2.142, g? = 2.039) were obtained from simu-
lation of both the 9.4-GHz and 115-GHz EPR spectra
(Figure 3). The D value is close to that reported for an
excited S = 32 spin state in a triangular tricopper(ii) complex
(ꢀ53.5 mT),[15] but significantly different from those reported
for organic triradicals (j 0.003–0.008 j cmꢀ1)[16] and mononu-
clear bis(phenoxyl) radical copper(ii) complexes (j 0.4 j
cmꢀ1).[17] The 9.4-GHz EPR spectrum of 2C+ exhibits two
[1] J. Stubbe, W. A. van der Donk, Chem. Rev. 1998, 98, 705; W.
Kaim, Dalton Trans. 2003, 761.
[2] N. Ito, S. E. V. Phillips, C. Stevens, Z. B. Ogel, M. J. McPherson,
J. N. Keen, K. D. S. Yadav, P. J. Knowles, Nature 1991, 350, 87;
J. W. Whittaker in Advances in Protein Chemistry, Vol. 60 (Eds.:
F. M. Richards, D. S. Eisenberg, J. Kuriyan), Academic Press,
Elsevier, 2002, p. 1 – 49.
[3] For reviews see: B. A. Jazdzewski, W. B. Tolman, Coord. Chem.
Rev. 2000, 200–202, 633; S. Itoh, M. Taki, S. Fukuzumi, Coord.
Chem. Rev. 2000, 198, 3; P. Chaudhuri, K. Wieghardt, Prog.
Inorg. Chem. 2001, 50, 151.
[4] Crystal
data:
for
1(ClO4ꢀ)3·0.33H2O·0.66EtOH:
C
40.33H46.67Cl3Cu2N9O15, purple–blue prisms, monoclinic, space
sets of signals, a weak DMS = ꢁ 3 transition at g = 8 and a
group P21/c, a = 21.322(2), b = 10.920(3), c = 23.131(5) ꢀ, a =
1
2
90, b = 115.41(1), g = 908, V= 4864.5(6) ꢀ3, Z = 4, D =
dominating SCu
=
signal (attributed to a degraded complex
containing noninteracting copper(ii) nuclei) at g = 2.
1.544 g.cmꢀ3
,
T= 294 K, F(000) = 2322.60, m = 1.114 mmꢀ1
,
440
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 438 –441

Angewandte
Chemie
l(MoKa) = 0.71073 ꢀ, 55223 reflections collected, 13287 unique,
8321 reflections (I > 2s), Rint = 0.07971. The final agreement
factors are R1 = 0.0609 for data with F0 > 2s(F0) and wR2 =
0.1006 for all data. CCDC 244865 contains the supplementary
crystallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or
from the Cambridge Crystallographic Data Centre, 12, Union
Road, Cambridge CB21EZ, UK; fax: (+ 44)1223-336-033; or
deposit@ccdc.cam.ac.uk); for 31.5(ClO4ꢀ)3: C63H66Cl3Cu3N9O18,
brown needles, triclinic, space group Pꢀ1, a = 12.605(1), b =
13.213(1), c = 21.692(2) ꢀ, a = 74.5(3), b = 86.8(3), g =
67.6(3)8, V= 3214(6) ꢀ3, Z = 2, D = 1.585 g.cmꢀ3, T= 150 K,
[16] E. Beer, J. Daub, C. Palivan, G. Gescheidt, J. Chem. Soc. Perkin
Trans. 2 2002, 1605; D. A. Schultz, R. K. Kumar, J. Am. Chem.
Soc. 2001, 123, 6431.
[17] E. Bill, J. Mꢂller, T. Weyhermꢂller, K. Wieghardt, Inorg. Chem.
1999, 121, 5795; F. Thomas, O. Jarjayes, C. Duboc, C. Philouze, E.
Saint-Aman, J.-L. Pierre, Dalton Trans. 2004, 2662.
F(000) = 1578, m = 1.188 mmꢀ1
reflections collected, 21243 unique, 16811 reflections (I > 2s),
int = 0.06431. The final agreement factors are R1 = 0.0547 for
, l(MoKa) = 0.71073 ꢀ, 43303
R
data with F0 > 2s(F0) and wR2 = 0.0763 for all data. CCDC
245087 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cam-
bridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB21EZ, UK; fax: (+ 44)1223-336-033; or deposit@
ccdc.cam.ac.uk).
[5] C. Belle, C. Bꢁguin, I. Gautier-Luneau, S. Hamman, C. Philouze,
J.-L. Pierre, F. Thomas, S. Torelli, E. Saint-Aman, M. Bonin,
Inorg. Chem. 2002, 41, 479.
[6] Mixing only one equivalent of Cu(ClO4)2·6H2O and HL in
acetonitrile in the presence of one equivalent of NEt3 did not
afford the expected mononuclear copper(ii) complex of HL.
Instead, 2 was isolated in 40% yield.
[7] Addition of pyridine (py) to 3 induces a dimmer-to-monomer
conversion yielding [L’(py)Cu]+. The exogenous pyridine coor-
dinates to the copper instead of one m-phenolato oxygen. Kpy
=
([L’(py)Cu]+2/([pyridine]2 [[L’Cu]22+]) was determined by UV/
Vis titration from the equilibrium: [L’Cu]22++2pyridineQ
2[L’(py)Cu]+.
[8] UV/Vis data (lmax (nm), e (mꢀ1 cmꢀ1) in CH3CN: 1: 547 (350), d–d
transitions appear as a broad shoulder; 2: 447 (330), 805 (190); 3:
463 (340), 880 (250); [L’(py)Cu]+: 531 (330), 880 (210).
[9] R. Srinivasan, I. Sougandi, K. Velavan, R. Venkatesan, B.
Verghese, P. Sambasiva Rao, Polyhedron 2004, 23, 1115.
[10] A linear dependence of E1/2 as a function of the m-phenolate s+
Hammet constant is observed for related compounds (Figure S8
in the Supporting Information). Moreover, the UV/Vis feature
of the oxidized species matches well that reported for methoxy-
phenoxyl–copper(ii) complexes (D. Zurita, I. Gautier-Luneau, S.
Mꢁnage, J.-L. Pierre, J. Biol. Inorg. Chem. 1997, 2, 46; Y.
Shimazaki, S. Huth, S. Hirota, O. Yamauchi, Inorg. Chim. Acta
2002, 331, 168). This further confirms that the one-electron
electrochemical process is ligand-based rather than metal-based.
[11] This is evident in the CV curve from an irreversible anodic signal
(corresponding to the oxidation of the bridging phenolato
moiety) associated, in the reverse scan, with a cathodic peak
which is observed ca. 0.3 V lower, that is, at a potential value
similar to that of the nonbridging phenoxyl/phenolato couple:
See F. Michel, F. Thomas, S. Hamman, E. Saint-Aman, C.
Bucher, J.-L. Pierre, Chem. Eur. J. 2004, 10, 4115.
[12] J. G. Radziszewski, M. Gil, A. Gorski, J. Spanget-Larsen, J.
Waluk, B. J. Mroz, J. Chem. Phys. 2001, 115, 9733.
[13] The e values are probably under estimated for 2C+ due to its low
stability.
[14] The 9.4-GHz EPR spectrum of 1C+ measured at 215 K lacks most
of the transitions associated with the quartet. It still exhibits a
signal at g = 2.06 that could be attributed to an excited (S = 12)
state, indicating that the exchange constant J is weak.
[15] B. Cage, F. A. Cotton, N. S. Dalal, E. A. Hillard, B. Rakvin,
C. M. Ramsey, C. R. Chim. 2003, 6, 39.
Angew. Chem. Int. Ed. 2005, 44, 438 –441
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
441