Three-spin system with a twist: A bis(semiquinonato)copper complex with a nonplanar configuration at the copper(II) center

Article abstract of DOI:10.1002/anie.200462339

(Chemical Equation Presented) A new twist: Structural twisting by about 32° between the semiquinoneiminecopper(II) entities in 2 caused by weak Cu-S interaction results in a different spin arrangement (↑,↓,↑) than for planar 1 (↑,↑,↓). Most revealingly, 2

Full text of DOI:10.1002/anie.200462339

Angewandte
Chemie
Copper Complexes
stituent in the o-position of the N-aryl group to form QyCꢀ, can
considerably change the structure and the properties of this
Three-Spin System with a Twist:
A Bis(semiquinonato)copper Complex with a
Nonplanar Configuration at the
Copper(ii) Center**
three-spin system.
The new ligand system was synthesized in the reduced
form H2Qy.[8] Its reaction with CuCl produced an air-stable
complex [(Qy)Cu(Qy)] (2) which could be crystallographically
characterized (Figure 1).[9] A significantly twisted arrange-
Shengfa Ye, Biprajit Sarkar, Falk Lissner,
Thomas Schleid, Joris van Slageren, Jan Fiedler, and
Wolfgang Kaim*
Complexes with
a significantly nonplanar environment
around tetracoordinate copper(ii) centers are comparatively
rare and were often investigated in connection with copper
enzyme modeling[1] or with the highly distorted type 1 copper
centers in blue copper proteins.[2] These centers exhibit partial
sulfur coordination and act as electron-transfer sites, shuttling
between the CuII and CuI states. The CuII,I couple can also be
active in the electron-transfer interaction with one[3,4] or
two[5,6] non-innocent quinone ligands (Q). While a biochem-
ically relevant valence tautomer equilibrium CuII(Q2ꢀ)QCuI
(QCꢀ) has been observed within[7] and outside[4] oxidase
enzymes, the focus with Cu(Q)2 compounds has been on the
study of spin–spin interaction in stable three-spin systems
(QCꢀ)CuII(QCꢀ).[5,6] Using sterically protected o-benzosemi-
quinonemonoimines, such as QxCꢀ, Chaudhuri et al. have
described a planar such complex [(QxCꢀ)CuII(QxCꢀ)] (1), with a
(›,›,fl) ground state containing two antiferromagnetically
coupled radical ligands.[6]
Figure 1. Molecular structure of complex 2. Selected bond lengths [ꢀ]
and angles [8]: Cu-O1 1.9173(15), Cu-O2 1.9290(16), Cu-N1
1.9372(19), Cu-N2 1.9368(18), O1-C1 1.296(3), O2-C22 1.292(3), N1-
C6 1.355(3), N2-C27 1.352(3), C1-C2 1.430(3), C22-C23 1.430(3), C2-
C3 1.380(3), C23-C24 1.373(3), C3-C4 1.433(3), C24-C25 1.433(3), C4-
C5 1.371(3), C25-C26 1.368(3), C5-C6 1.419(3), C26-C27 1.421(3), C6-
C1 1.453(3), C27-C22 1.455(3); O1-Cu-O2 149.81(8), O1-Cu-N2
97.00(7), O2-Cu-N1 98.30(7), N2-Cu-N1 171.95(8), O1-Cu-N1 84.51(7),
O2-Cu-N2 84.40(7).
Herein we describe how a minor modification, the
introduction of a potentially coordinating methylthio sub-
ment is found about the copper center which has a N1CuO1/
N2CuO2 dihedral angle of 32.20(9)8 and is coordinated by
two ligands through the N and O donors. However, there is
also a weak additional interaction from S1 to the copper atom
(Cu–S1 3.198(1) ꢀ; Cu–S2 3.475(1) ꢀ). Very long CuII–
S(thioether) separations have also been reported for certain
type 1 copper centers in proteins, especially azurin.[2] The
structural data show unambiguously that the ligands are in the
semiquinone state,[10] which implies the + ii oxidation state
for the copper center in its highly twisted coordination
environment. The sum of bond angles at the copper is 685.88,
about halfway between the values for square planar (7208)
[*] S. Ye, B. Sarkar, Dr. F. Lissner, Prof. Dr. T. Schleid, Prof. Dr. W. Kaim
Institut fꢁr Anorganische Chemie
ꢀ
and tetrahedral geometry (ca. 6578). Although the Cu O and
Universitꢂt Stuttgart
Pfaffenwaldring 55, 70550 Stuttgart (Germany)
Fax: (+49)711-685-4165
ꢀ
Cu N bonds are comparable, the N1-Cu-N2 angle is closer to
linearity at 171.858 than the O1-Cu-O2 angle (149.708)—
possibly a result of steric repulsion between the aryl
substituents.
E-mail: kaim@iac.uni-stuttgart.de
Dr. J. van Slageren
1. Physikalisches Institut, Universitꢂt Stuttgart
Pfaffenwaldring 57, 70550 Stuttgart (Germany)
Figure 2 shows the dependence of cT on T for the three-
spin compound [(QyCꢀ)CuII(QyCꢀ)].[11a] The room temperature
cT value of 0.46 cm3 Kmolꢀ1 is much lower than expected
(1.125 cm3 Kmolꢀ1) for three uncoupled S = 1/2 spins with g
ꢁ 2, which indicates predominantly antiferromagnetic
exchange interactions in the system. At lower temperatures,
the cT value reaches a plateau at cT= 0.365 cm3 Kmolꢀ1, in
agreement with an S = 1/2 ground state. At the lowest
temperatures the cT value decreases again. This system of
three exchange-coupled spins can be described by the
Dr. J. Fiedler
´
J. Heyrovsky Institute of Physical Chemistry
Academy of Sciences of the Czech Republic
ˇ
Dolejskova 3, 18223 Prague (Czech Republic)
[**] Support from the Deutsche Forschungsgemeinschaft, the Fonds der
Chemischen Industrie, the Grant Agency of the Czech Republic
(Grant No. 203/03/0821), and the Ministry of Education of the
Czech Republic (Grant COSTOCD15.10) is gratefully acknowl-
edged.
Angew. Chem. Int. Ed. 2005, 44, 2103 –2106
DOI: 10.1002/anie.200462339
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2103

Communications
weaker. For more definitive conclusions to be drawn the
excited S = 1/2 (›,›,fl) state needs to be determined exactly.
The EPR results for 2 differ significantly from those of 1
(with planar configuration at the metal):[6] Instead of a signal
for copper(ii) with g > 2[6] we observe a partially resolved
radical-type spectrum at g = 2.0012 in CH2Cl2 solution
(Figure 3). In the frozen state at 110 K or 4 K the g compo-
Figure 2. Magnetic susceptibility–temperature product for 2 as a func-
&
tion of temperature ( ) at 1 T and fit (c) using the parameters
g=2.00, J=ꢀ414 cmꢀ1, and J’=ꢀ114 cmꢀ1
.
isotropic exchange Hamiltonian [Eq. (1)] where spins S1 and
S3 refer to the two QyCꢀ radicals and spin S2 is the CuII center.
h ¼ ꢀJ ðS1 ꢂ S2 þ S2 ꢂ S3ÞꢀJ0 S1 ꢂ S3
ð1Þ
The exchange coupling in such a system leads to three spin
states, two with S = 1/2 and one with S = 3/2, the energies of
which depend on the sign and magnitude of the two exchange
interactions.[6] The two S = 1/2 states differ in the relative
orientation of the magnetic moments of spin S1 and S3 which
can be described by S* = S1 + S3. The spin functions are fully
described by the total and intermediate spins as j ST S*i =
1
1
3
j = 0i, j = 1i, and j = 1i, or in symbolic fashion (fl,›,fl),
2
2
2
(›,›,fl), and (›,›,›). Their energies are given by E(›,›,fl) =
Figure 3. EPR spectrum of complex 2 in dichloromethane at
210 K (top) with computer simulation (bottom): a(63,65Cu)=0.22 mT,
a(14N)=0.70 mT, a(1H)=0.36 mT (1H,H3).
3
ꢀJ + J’, E(fl,›,fl) = 0, and E(›,›,›) = ꢀ = J, taking the (fl,›,fl)
2
state as the energy origin. The cT product as a function of
temperature is then given by Equation (2).
g2 þ g2 exp½ðJꢀJ0Þ=k Tꢃ þ 10 exp½3 J=2 k Tꢃ
12,0
nents were found at g1 = 2.050, g2 = 1.9937 (A2 = 2.0 mT), and
g3 = 1.957, gav = 2.0006. The g value and hyperfine data are
fully compatible with a copper-coordinated 3,5-di-tert-butyl-
o-semiquinoneimine radical anion,[3a,4] the slightly lower
g value signifies the presence of low-lying excited states
with non-zero orbital angular momentum.[12]
The EPR result in particular points to a (›,fl,›) ground
state for [(QyCꢀ)CuII(QyCꢀ)] with a dominant antiferromagnetic
metal–ligand (and not ligand–ligand[6]) interaction. Clearly
the interligand spin–spin interaction between the two semi-
quinone ligands is diminished as a consequence of the
nonplanar coordination arrangement.
N
4 k
A m2B
12,1
ð2Þ
cT ¼
1 þ exp½ðJꢀJ0Þ=k Tꢃ þ 2 exp½3 J=2 k Tꢃ
The S = 1/2 ground state can have either (›,›,fl) or (fl,›,fl)
character, depending on whether the antiferromagnetic
interaction between the two radicals or between the radical
and metal are dominating. Electron paramagnetic resonance
(EPR) results indicate that the lowest spin state has predom-
inant radical character (see below), from which it can be
concluded that this lowest spin state is best described as
(fl,›,fl). Thus the antiferromagnetic metal–radical interaction
must be stronger than that between the two radicals (j J j >
j J’ j ). Since EPR shows no evidence for excited spin states
between 110 K and room temperature in the form of extra
signals or shifting signals (in the case of fast spin-state
interconversion), the excited S = 1/2 state must lie at high
energy, meaning that j J j ꢀ j J’ j is large.
Fitting a susceptibility curve using the Equation (2) is
problematic since the parameters are interdependent and
there is no unique solution.[11b] However, the results suggest
that the metal–radical interaction is strongly antiferromag-
netic with a magnitude of J = ꢀ414 cmꢀ1, while the radical–
radical interaction is also antiferromagnetic, but probably
In the absence of detailed structural and magnetic results,
the appearance of a radical-type EPR signal could suggest an
alternative oxidation state combination of (QyCꢀ)CuI(Qy) for 2
which would also be in agreement with a nonplanar metal
configuration. However, the susceptibility behavior and the
bond lengths within the two very similar ligands clearly
demonstrate the presence of [(QyCꢀ)CuII(QyCꢀ)] with a (›,fl,›)
ground state.
Like the planar analogue 1[6] with QxCꢀ, the compound 2
described herein undergoes four reversible one-electron
steps, two oxidations at + 0.37 and ꢀ0.38 V, and two
2104
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 2103 –2106

Angewandte
Chemie
reductions at ꢀ1.07 and ꢀ1.49 V (versus Fc+/0 (Fc = ferro-
cene) in CH2Cl2/0.1m Bu4NPF6). The results of spectroelec-
trochemistry investigations in the UV/Vis-NIR regions are
summarized in Figure 4, Table 1, [Eq. (3)]: On oxidation to 2+
the long-wavelength bands attributed to intraligand (IL) and
ligand-to-metal charge transfer (LMCT) transitions are
diminished and a strong band emerges at 515 nm, possibly
associated with the formation of a quinone ligand in
[(Qy)CuII(QyCꢀ)]+. The second oxidation to [(Qy)CuII(Qy)]2+
leaves only the band arising from quinone, shifted to 528 nm.
One-electron reduction could lead to either [(QyCꢀ)CuII-
(Qy2ꢀ)]ꢀ or [(QyCꢀ)CuI(QyCꢀ)]ꢀ. Spectroelectrochemical reduc-
tion gives rise to a strong, broad (Dn1/2 = 2600 cmꢀ1) band at
1940 nm (5150 cmꢀ1, Figure 4, Table 1) in addition to absorp-
Table 1: Absorption data of complexes.[a]
Compound lmax [nm] (103 e [mꢀ1 cmꢀ1])
2
316(15.7), 345(16.5), 446 sh(5.3), 790(5.4), 1062(3.0)
259(16.5), 242(15.8), 515(11.9), 930(2.7)
238(14.2), 270(14.7), 300 sh(13.1), 424(10.5), 528(11.6)
245 sh(13.2), 273(14.0), 358(14.1), 526(3.8), 695(3.6),
1940(4.5)
2+
22+
2ꢀ
22ꢀ
247 sh(11.6), 270(12.3), 307(12.2), 364(12.4)
[a] In CH2Cl2 solution, charged species were electrochemically generated
in CH2Cl2/0.1m Bu4NPF6.
tions at 526 and 695 nm. The intense band (oscillator strength
f = 5.4 ꢁ 10ꢀ2) in the near infrared is more compatible with the
ligand-to-ligand intervalence charge-transfer transition[13] for
a [(QyCꢀ)CuII(Qy2ꢀ)]ꢀ formulation, however, the structure of
this species is not known. The second reduction to either
[(QyCꢀ)CuI(Qy2ꢀ)]2ꢀ or [(Qy2ꢀ)CuII(Qy2ꢀ)]2ꢀ does not produce
intense absorptions above 450 nm. More detailed assignments
of oxidation state combinations and electronic transitions will
require high-quality excited-state TD-DFT calculations.
Summarizing, we have found an additional factor in
determining the oxidation and spin state for the already
complicated interaction[3–6] between the copper(ii,i) couple
and quinonoid ligands, that is, the possible structural dis-
tortion caused by secondary coordination. While radical
anion ligands[12] already yield unusual complexes with tran-
sition metals in “normal” configurations,[3b,6,14,15] the geo-
metrical distortion can add, literally, another twist to this
remarkable class of compounds, allowing for a striking switch
0, ꢀ,2ꢀ
of spin combinations. The behavior of the Qy C
redox
system as a ligand towards other transition metals is currently
being investigated.
Received: October 18, 2004
Published online: February 21, 2005
Keywords: copper · electrochemistry · EPR spectroscopy ·
.
radical ligands · spin–spin interactions
[1] Y. Wang, J. L. DuBois, B. Hedman, K. O. Hodgson, T. D. P.
Stack, Science 1998, 279, 537.
[2] a) W. Kaim, J. Rall, Angew. Chem. 1996, 108, 47; Angew. Chem.
Int. Ed. Engl. 1996, 35, 43; b) A. Messerschmidt, Struct. Bonding
(Berlin) 1998, 90, 37; c) A. J. Vila, C. O. Fernandez in Handbook
Figure 4. Spectroelectrochemistry study of complex 2 in CH2Cl2/0.1m
Bu4NPF6.
Angew. Chem. Int. Ed. 2005, 44, 2103 –2106
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2105

Communications
on Metalloproteins (Eds.: A. Bertini, A. Sigel, H. Sigel), Marcel
Dekker, New York, 2001, p. 813; d) M. D. Harrison, C. Denni-
son, ChemBioChem 2004, 5, 1579.
methane solution of 2 layered with n-hexane. Crystal data:
C42H54CuN2O2S2 (746.53); monoclinic P21/n; a = 16.3439(2), b =
14.5226(2), c = 16.9358(3) ꢀ; b = 90.140(1)8; V= 4019.8(1) ꢀ3;
[3] a) J. Rall, W. Kaim, J. Chem. Soc. Faraday Trans. 1994, 90, 2905;
b) C. G. Pierpont, C. W. Lange, Prog. Inorg. Chem. 1994, 41, 331;
c) C. G. Pierpont, Coord. Chem. Rev. 2001, 99, 216; d) D. A.
Shultz, Comments Inorg. Chem. 2002, 23, 1.
Z = 4; 1calcd = 1.234 gcmꢀ3
;
m = 0.683 mmꢀ1
;
T= 100(2) K;
number of unique reflections 9142, number of parameters 498;
R1 = 0.0475, wR2 = 0.1143 (I > 2s(I)); R1 = 0.0570, wR2 =
0.1203 (all data); GOOF = 1.044. The structure was solved by
direct methods and refined using the SHELXS-97/SHELXL-97
program package (G. M. Sheldrick, SHELXS-97, Program for
the Solution of Crystal Structures, University of Gꢂttingen,
Germany, 1997, G. M. Sheldrick, SHELXL-97, Program for the
Refinement of Crystal Structures, University of Gꢂttingen,
Germany, 1997). The scattering factors for the neutral atoms
were those incorporated in the programs. CCDC-251450 con-
tains the supplementary crystallographic data for this paper.
These data can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[4] a) J. Rall, M. Wanner, M. Albrecht, F. M. Hornung, W. Kaim,
Chem. Eur. J. 1999, 5, 2802; b) W. Kaim, M. Wanner, A. Knꢂdler,
S. Zalis, Inorg. Chim. Acta 2002, 337, 163; c) G. A. Abakumov,
V. K. Cherkasov, V. I. Nevodchikov, V. A. Kuropatov, G. T. Yee,
C. G. Pierpont, Inorg. Chem. 2001, 40, 2434; d) G. Speier, Z.
Tyeklꢃr, P. Tꢄth, E. Speier, S. Tisza, A. Rockenbauer, A. M.
Whalen, N. Alkire, C. G. Pierpont, Inorg. Chem. 2001, 40, 5653.
[5] a) G. A. Abakumov, V. K. Cherkasov, M. P. Bubnov, O. G.
Ellert, Yu. V. Rakitin, L. N. Zakharov, Yu. T. Struchkov, Yu. N.
Safꢅyanov, Izv. Akad. Nauk Ser. Khim. 1992, 10, 2315; b) A. Dei,
D. Gatteschi, C. Sangregorio, L. Sorace, Acc. Chem. Res. 2004,
37, 827.
[10] a) S. Bhattacharya, P. Gupta, F. Basuli, C. G. Pierpont, Inorg.
Chem. 2002, 41, 5810; b) K. S. Min, T. Weyhermꢆller, E. Bothe,
K. Wieghardt, Inorg. Chem. 2004, 43, 2922.
[6] P. Chaudhuri, C. N. Verani, E. Bill, E. Bothe, T. Weyhermꢆller,
K. Wieghardt, J. Am. Chem. Soc. 2001, 123, 2213.
[7] a) D. M. Dooley, M. A. McGuirl, D. E. Brown, P. N. Turowski,
W. S. McIntire, P. F. Knowles, Nature 1991, 349, 262; b) M. A.
McGuirl, D. E. Brown, D. M. Dooley, J. Biol. Inorg. Chem. 1997,
2, 336.
[11] a) The magnetic susceptibility measurements were performed
on a powder sample of 2 using a Quantum Design MPMSXL7
SQUID magnetometer. The data were corrected for the
diamagnetic contributions to the molar magnetic susceptibility
using Pascalꢅs constants and for temperature-independent para-
magnetism (TIP = 120 ꢁ 10ꢀ6 cm3 molꢀ1); b) The following con-
straints were used: The g values of the radicals and the CuII
center were taken to be 2.00 (see EPR discussion). Furthermore,
the difference between the two exchange interaction parameters
was not allowed to decrease below 300 cmꢀ1 in order to keep the
excited S = 1/2 state at high energies. From the fit (T> 50 K), the
following parameters are obtained: J = ꢀ414 cmꢀ1, and J’ =
ꢀ114 cmꢀ1 (ꢄ 6 cmꢀ1). If not constrained, the radical–radical
exchange interaction tends to increase, leading to similar values
[8] Preparation of Qy: A solution of 2-(methylthio)aniline (1.44 mL,
11.5 mmol) in n-heptane (4 mL) was added dropwise to a
suspension of 3,5-di-tert-butylcatechol (2.56 g, 11.5 mmol) and
triethylamine (0.15 mL) in n-heptane (10 mL). The solution was
heated to reflux for 3 h, the solvent then evaporated to dryness,
and the residue washed with cold n-heptane to yield a colorless
1
solid. Yield 2.37 g (60%). H NMR (250 MHz, 300 K, CDCl3):
d = 6.35–7.48 (m, 6H, arom), 2.45 (s, 3H, SCH3), 1.44 (s, 9H,
C(CH3)3), 1.26 ppm (s, 9H, C(CH3)3). Elemental analysis (%)
calcd for C21H29NOS: C 73.42, H 8.51, N 4.08; found: C 73.49, H
8.42, N 4.07.
for both exchange parameters, J = ꢀ367 cmꢀ1
ꢀ357 cmꢀ1, without significantly improving the fit.
[12] W. Kaim, Coord. Chem. Rev. 1987, 76, 187.
, and J’ =
[9] Synthesis of 2: A solution of Qy (736 mg 2.14 mmol), NEt3
(0.5 mL), and CuCl (106 mg, 1.07 mmol) was heated to reflux
in acetonitrile (25 mL) for 4 h in the presence of air. The dark
precipitate was collected by filtration, then washed with cold
[13] a) D. Herebian, E. Bothe, F. Neese, T. Weyhermꢆller, K.
Wieghardt, J. Am. Chem. Soc. 2003, 125, 9116; b) A. Dogan, B.
Sarkar, A. Klein, F. Lissner, T. Schleid, J. Fiedler, S. Zalis, V. K.
Jain, W. Kaim, Inorg. Chem. 2004, 43, 5973.
[14] W. Kaim, Dalton Trans. 2003, 761.
[15] P. Chaudhuri, K. Wieghardt, Prog. Inorg. Chem. 2001, 50, 151.
acetonitrile.
A dark green microcrystalline material was
obtained. Yield 520 mg (65%). Elemental analysis (%) calcd
for C42H54CuN2O2S2: C 67.57, H 7.29 N 3.75; found: C 67.30, H
7.31, N 3.73. Single crystals were obtained from a dichloro-
2106
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 2103 –2106