Quantitative Evaluation of d-π Interaction in Copper(I) Complexes and Control of Copper(I)-Dioxygen Reactivity

Article abstract of DOI:10.1002/chem.200305263

Crystal structures of the copper(I) complexes 1^x, 2, and 3 of a series of tridentate ligands L1^x, L2, and L3, respectively (L1 ^x: p-substituted derivatives of N,N-bis[2-(2-pyridyl)ethyl]-2-phenylethylamine; X = H, Me, OMe, Cl, NO ₂; L2: N,N-bis[2-(2-pyridyl)ethyl]-2-methyl-2-phenylethylamine; L3: N,N-bis[2-(2-pyridyl)ethyl]-2,2-diphenylethylamine) were solved to demonstrate that all the copper(I) complexes involve an η² copper-arene interaction with the phenyl ring of the ligand sidearm. The Cu¹ ion in each complex has a distorted tetrahedral geometry consisting of the three nitrogen atoms (one tertiary amine nitrogen atom and two pyridine nitrogen atoms) and C₁-C₂ of the phenyl ring of ligand sidearm, whereby the Cu-C distances of the copper-arene interaction significantly depend on the para substituents. The existence of the copper-arene interaction in a nonpolar organic solvent (CH₂Cl₂) was demonstrated by the observation of an intense MLCT band around 290 nm, and the magnitude of the interaction was evaluated by detailed analysis of the ¹H and ¹³C NMR spectra and the redox potentials E_1/2 of the copper ion, as well as by means of the ligand-exchange reaction between the phenyl ring and CH₃CN as an external ligand. The thermodynamic parameters ΔH° and ΔS° for the ligand-exchange reaction with CH₃CN afforded a quantitative measure for the energy difference of the copper-arene interaction in the series of copper(I) complexes. Density functional studies indicated that the copper(I)-arene interaction mainly consists of the interaction between the d_z2 orbital of Cu ¹ and a π orbital of the phenyl ring. The copper(I) complexes 1^x reacted with O₂ at -80°C in CH₂Cl ₂ to give the corresponding (μ-η²:η ²-peroxo)dicopper(II) complexes 4, the formation rates k _obs of which were significantly retarded by stronger d-π interaction, while complexes 2 and 3, which exhibit the strongest d-π interaction showed significantly lower reactivity toward O₂ under the same experimental conditions. Thus, the d-π interaction has been demonstrated for the first time to affect the copper(I)-dioxygen reactivity, and represents a new aspect of ligand effects in copper(I)-dioxygen chemistry.

Full text of DOI:10.1002/chem.200305263

FULL PAPER
Quantitative Evaluation of d p Interaction in Copper(i) Complexes and
Control of Copper(i) Dioxygen Reactivity
Takao Osako,[a] Yoshimitsu Tachi,[a] Matsumi Doe,[a] Motoo Shiro,[b] Kei Ohkubo,[c]
Shunichi Fukuzumi,*[c] and Shinobu Itoh*[a]
Abstract: Crystal structures of the cop-
per(i) complexes 1X, 2, and 3 of a series
of tridentate ligands L1X, L2, and L3,
respectively (L1X: p-substituted deriva-
tives of N,N-bis[2-(2-pyridyl)ethyl]-2-
phenylethylamine; X=H, Me, OMe,
Cl, NO2; L2: N,N-bis[2-(2-pyridyl)eth-
yl]-2-methyl-2-phenylethylamine; L3:
N,N-bis[2-(2-pyridyl)ethyl]-2,2-diphenyl-
ethylamine) were solved to demon-
strate that all the copper(i) complexes
involve an h2 copper arene interaction
with the phenyl ring of the ligand side-
arm. The CuI ion in each complex has
a distorted tetrahedral geometry con-
sisting of the three nitrogen atoms (one
tertiary amine nitrogen atom and two
pyridine nitrogen atoms) and C1ꢀC2 of
the phenyl ring of ligand sidearm,
whereby the CuꢀC distances of the
copper arene interaction significantly
depend on the para substituents. The
existence of the copper arene interac-
tion in nonpolar organic solvent
complexes. Density functional studies
indicated that the copper(i) arene in-
teraction mainly consists of the interac-
tion between the dz2 orbital of CuI and
a p orbital of the phenyl ring. The cop-
per(i) complexes 1X reacted with O2 at
ꢀ808C in CH2Cl2 to give the corre-
sponding (m-h2:h2-peroxo)dicopper(ii)
complexes 4, the formation rates kobs of
which were significantly retarded by
stronger d p interaction, while com-
plexes 2 and 3, which exhibit the stron-
gest d p interaction showed significant-
ly lower reactivity toward O2 under the
same experimental conditions. Thus,
the d p interaction has been demon-
strated for the first time to affect the
copper(i) dioxygen reactivity, and rep-
resents a new aspect of ligand effects
in copper(i) dioxygen chemistry.
a
(CH2Cl2) was demonstrated by the ob-
servation of an intense MLCT band
around 290 nm, and the magnitude of
the interaction was evaluated by de-
1
tailed analysis of the H and 13C NMR
spectra and the redox potentials E1/2 of
the copper ion, as well as by means of
the ligand-exchange reaction between
the phenyl ring and CH3CN as an ex-
ternal ligand. The thermodynamic pa-
rameters DHo and DSo for the ligand-
exchange reaction with CH3CN afford-
ed
a quantitative measure for the
energy difference of the copper arene
interaction in the series of copper(i)
Keywords: copper ¥ Pi interactions ¥
N
ligands
¥
OꢀO activation
¥
substituent effects
Introduction
Weak interactions such as hydrogen bonding, p p stacking,
cation p, and CH p interactions are recognized as essential
tools in molecular recognition and supramolecular chemis-
[a] T. Osako, Dr. Y. Tachi, M. Doe, Prof. S. Itoh
7]
Department of Chemistry, Graduate School of Science, Osaka City
University
3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585 (Japan)
Tel/Fax: (+81)6-6605-2564
try.[1 Such interactions play versatile roles not only in the
structural regulation of molecular architectures but also in
controlling their physicochemical properties and functions.
Transition metal p complexes of aromatic compounds are
also well known as important intermediates in a wide varie-
E-mail: shinobu@sci.osaka-cu.ac.jp
[b] Dr. M. Shiro
ty of catalytic reactions with industrial and synthetic applica-
X-ray Research Laboratory, Rigaku Agency
3-9-12 Matsubara, Akishima-shi, Tokyo 196-8666 (Japan)
10]
tions.[8
For copper, however, structurally characterized
19]
arene complexes are relatively rare,[11
and very little is
[c] Dr. K. Ohkubo, Prof. S. Fukuzumi
Department of Material and Life Science, Graduate School of Engi-
neering, Osaka University
CREST, Japan Science and Technology Corporation
2-1 Yamada-oka, Suita, Osaka 565-0871 (Japan)
known about the physicochemical aspects of the copper ar-
ene interaction.
Copper complexes of a wide variety of ligands have been
developed, especially in the field of bioinorganic chemistry,
to replicate the structures and functions of the active sites of
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
237
Chem. Eur. J. 2004, 10, 237 246
DOI: 10.1002/chem.200305263
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPER
copper proteins (enzymes).[20] In particular, dioxygen activa-
tion by copper(i) complexes is an important and attractive
research objective, not only in bioinorganic chemistry but
27]
also in catalytic oxidation reactions.[21 In recent years, ex-
tensive efforts have been devoted to clarifying the effects of
ligands on the structure and reactivity of copper dioxygen
intermediates, and demonstrated that nitrogen-donor capa-
bility and denticity (didentate versus tridentate versus tetra-
dentate), as well as steric effects of the supporting ligands,
34]
are crucial in controlling the structure and reactivity.[28
Recently, it was shown that the presence of acidic protons
on the nitrogen donor atoms of the ligand alters the struc-
ture and stability of resulting Cu2/O2 complexes.[35,36] Since
most of the ligands so far been employed contain aromatic
groups, copper(i) arene interactions may also play impor-
tant roles in controlling the structure and reactivity of the
copper(i) complexes toward di-
Table 2, in which the data of 2 are also included. It is note-
worthy that the unit cell of 1H contains ten crystallographi-
cally independent molecules, and is thus an unusually long
rectangular parallelepiped (a=15.695(2), b=78.34(1), c=
17.389(4) ä), while the other complexes form normal unit
oxygen. However, little atten-
tion has been paid to copper(i)
arene interactions in copper(i)
dioxygen chemistry.[37]
We report herein the first
quantitative evaluation of the
d p interaction in copper(i)
complexes of a series of bis[2-
(2-pyridyl)ethyl]amine triden-
tate ligands L1 L3. The crystal
structures, spectroscopic fea-
tures (NMR and UV/Vis), and
redox behavior of the copper(i)
complexes were systematically
investigated to provide pro-
found insights into the structure
and physicochemical features of
the copper(i) arene interaction.
Moreover, it was shown that
the reactivity of the copper(i)
complexes toward dioxygen is
significantly influenced by the
ligand substituents through the
d p interaction in the copper(i)
starting materials.
Results and Discussion
Structural characterization of
copper(i)--arene
interactions:
Crystal structures of [CuI(L1X)]-
ClO4 (1X, X=H, Me, OMe, Cl,
and NO2) and [CuI(L3)]ClO4
(3) were determined (Figure 1),
while that of [CuI(L2)]ClO4 (2)
was already reported in our
previous paper.[19] The crystallo-
graphic data of 1X and 3 are
presented in Table 1, and select-
ed bond lengths around the
Figure 1. ORTEP plots of 1X (X=OMe, Me, H, Cl, NO2) and 3 with 50% probability thermal ellipsoids The
copper ion are summarized in counteranion and hydrogen atoms are omitted for clarity.
238
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 237 246

Copper Arene Interactions
237 246
Table 1. Summary of X-ray crystallographic data.
[CuI(L1H)]ClO4 (1H)
[CuI(L1Me)]ClO4
(1Me
[CuI(L1OMe)]ClO4
(1OMe
[CuI(L1Cl)]ClO4
[CuI(L1NO )]ClO4
[CuI(L3)]ClO4 (3)
2
(1Cl)
(1NO
)
2
)
)
empirical for-
mula
C22H25N3O3ClCu
C23H27N3O4ClCu
C23H27N3O5ClCu
C22H24N3O4Cl2Cu
C22H24N4O6ClCu
C28H29N3O4ClCu
formula weight 494.46
508.49
524.48
528.90
539.46
570.55
crystal system
space group
a [ä]
b [ä]
c [ä]
monoclinic
orthorhombic
Pna21 (no. 33)
33.879(1)
10.9386(4)
11.9705(5)
90
monoclinic
P21/n (no. 14)
9.9597(2)
17.1256(5)
13.3791(3)
102.674(1)
2226.41(8)
4
monoclinic
P21/n (no. 14)
12.6164(8)
12.9552(9)
15.007(1)
111.821(2)
2277.1(3)
4
monoclinic
P21 (no. 4)
12.6536(4)
12.9535(5)
14.8924(6)
112.4000(5)
2256.8(1)
4
monoclinic
P21/n (no. 14)
8.6219(5)
17.4195(9)
17.0987(8)
95.844(2)
2554.7(2)
4
P21/a (no. 14)
15.695(2)
78.34(1)
17.389(4)
90.73(2)
21380(6)
40
b [8]
V [ä3]
Z
4436.1(2)
8
F(000)
10240.00
1.536
2112.00
1.523
1088.00
1.565
1088.00
1.543
1112.00
1.588
1184.00
1.483
1calcd [gcmꢀ3
T [8C]
]
ꢀ180
ꢀ115
ꢀ115
ꢀ115
ꢀ115
ꢀ115
crystal size
[mm]
0.25î0.15î0.07
0.20î0.20î0.10
0.20î0.30î0.10
0.20î0.20î0.10
0.20î0.20î0.10
0.20î0.20î0.10
m(MoKa) [cmꢀ1
]
28.89
11.41
11.43
12.28
11.34
10.00
diffractometer Rigaku RAXIS-RAPID
Rigaku RAXIS-
RAPID
MoKa
Rigaku RAXIS-
RAPID
MoKa
Rigaku RAXIS-
RAPID
MoKa
Rigaku RAXIS-
RAPID
MoKa
Rigaku RAXIS-
RAPID
MoKa
radiation
CuKa (l=1.54186 ä)
(l=0.71069 ä)
55.0
39027
(l=0.71069 ä)
55.0
20743
(l=0.71069 ä)
55.0
20670
(l=0.71069 ä)
55.0
21756
(l=0.71069 ä)
54.9
23207
2qmax [8]
no. of reflns
measd
136.5
231838
no. of reflns
obsd
36884 [I>ꢀ3.00s(I), 2
q<136.518]
5073 [I>0.01s(I)] 4179 [I>1.0s(I)]
3303 [I>0.5s(I)]
5205 [I>0.01s(I)] 4156 [I>1.0s(I)]
no. of variables 2791
632
326
314
662
364
R[a]
0.058
0.141
1.12
0.032
0.076
0.98
0.027
0.039
0.89
0.079
0.103
1.34
0.037
0.049
0.96
0.041
0.076
0.82
[b]
Rw
GOF
[a] R=ꢀjjFo jꢀjFc jj/ꢀjFo j. [b] Rw =[ꢀw(jFo jꢀjFc j)2/ꢀw(Fo2)2]1/2; w=1/s2(jFo j).
Table 2. Selected bond lengths [ä] around the copper ion.[a]
parently, the CuꢀC distances in the copper(i) arene interac-
tion largely depend on the para substituent X of the ligands,
although the CuꢀN distances are rather constant in the
Complex
CuꢀNamine
CuꢀNpy
CuꢀC1
CuꢀC2
1H
2.089
2.114
2.098
2.137
2.115
2.113
2.131
2.006
2.013
1.992
1.972
1.967
2.001
2.002
2.336
2.359
2.514
2.656
2.597
2.309
2.347
2.211
2.207
2.195
2.476
2.388
2.220
2.172
1Me
1OMe
1Cl
series
(dCuꢀN(amine) =2.115ꢁ0.024 ä;
dCuꢀN(py) =1.990ꢁ
0.023 ä; Table 2). The CuꢀC2 bond lengths of 1H, 1Me, 1OMe
,
2, and 3 (2.172 2.211 ä) are shorter than those of 1Cl and
1NO
2
1NO (2.476 and 2.388 ä), while the CuꢀC1 distances in 1OMe
,
2
2[b]
3
1Cl, and 1NO (2.514 2.656 ä) are longer than those of 1H,
2
1
Me, 2, and 3 (2.309 2.359 ä). Thus, the difference in bond
[a] Average values are presented where more than two crystallographical-
ly independent molecules exist in the unit cell. [b] Data are taken from
the literature.[19]
length between CuꢀC1 and CuꢀC2 is significantly larger in
1OMe than in the the other complexes. Crystal packing forces
may influence the CuꢀC bond lengths in the crystal. Howev-
er, such an effect is relatively small, if any, since the differ-
ences in CuꢀC1 and CuꢀC2 bond lengths among the ten
crystallographically independent molecules of 1H in the unit
cell are relatively small (CuꢀC1 2.326ꢁ0.069 ä, CuꢀC2
2.213ꢁ0.035 ä). Thus, it can be concluded that the differen-
ces in the CuꢀC bond lengths are mainly attributable to the
difference in strength of the copper(i) arene interaction.
This issue is further examined below.
The CꢀC bond lengths of the phenyl rings of the copper(i)
arene complexes are listed in Table S7 (Supporting Informa-
tion), in which the corresponding values of another phenyl
group in 3 without a copper(i) arene interaction are also in-
cluded. Overall, the copper(i) arene interaction has little
effect on the structure of the aromatic ring of the com-
plexes.
cells containing one or two crystallographically independent
molecules (Table 1 and Supporting Information: Figures S1
S6 and Tables S1 S6).
All the copper(i) complexes exhibit a similar type of cop-
per(i) arene interaction in the crystal, in which the phenyl
ring of the ligand sidearm is positioned just above the cop-
per(i) ion to undergo a coordinative interaction in a h2 fash-
ion (see Figure 1). Thus, the copper(i) ion in each complex
adopts a distorted tetrahedral geometry consisting of the
three nitrogen atoms (one Namine and two Npy) and the C1ꢀ
C2 moiety of the ligand sidearm. The h2 bonding interaction
is, however, significantly unsymmetrical, and the CuꢀC1
bond is always longer than the CuꢀC2 bond (Table 2).[38] Ap-
239
Chem. Eur. J. 2004, 10, 237 246
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

S. Fukuzumi, S. Itoh et al.
FULL PAPER
Table 3. UV/Vis and thermodynamic data for the titration of the cop-
per(i) complexes with CH3CN in CH2Cl2.
The features of the copper(i) arene interaction found in
the crystal structures are well reproduced by the optimized
structures obtained by DFT calculations (see Experimental
Section).[39] Figure S7 (Supporting Information) shows the
calculated structures of copper(i) arene complexes together
with the CuꢀC1 and CuꢀC2 bond lengths. The calculated
CuꢀC2 bond lengths are always shorter than the CuꢀC1
[a]
Complex
lmax
e
Kas
DHo
DSo
[nm] [mꢀ1 cmꢀ1
]
[mꢀ1
]
[kJmolꢀ1
]
[JKꢀ1 molꢀ1
]
1H
290
294
297
8820
7000
6.4ꢁ0.1
8.6ꢁ0.1
ꢀ12.4ꢁ1.6 ꢀ34ꢁ7
ꢀ13.2ꢁ0.9 ꢀ35ꢁ4
ꢀ15.6ꢁ2.4 ꢀ42ꢁ10
ꢀ18.2ꢁ1.9 ꢀ44ꢁ8
ꢀ20.6ꢁ0.5 ꢀ44ꢁ2
ꢀ9.7ꢁ0.8 ꢀ52ꢁ3
ꢀ7.3ꢁ0.3 ꢀ48ꢁ2
1Me
1OMe
1Cl
8520 12.2ꢁ0.1
288
8190 30.3ꢁ0.2
1NO
2
3
287
91.7ꢁ2.0
[b]
2
bond lengths, and the average CuꢀC lengths [(dCuꢀC1
+
290[c]
289
9700[c] 0.21ꢁ0.01
11000 0.10ꢁ0.02
dCuꢀC2)/2] of 1H and 1Me are shorter than those of 1OMe, 1Cl,
and 1NO
, as was observed in the crystal structures
2
[a] At ꢀ208C. [b] The e value could not be determined accurately due to
overlap with the strong absorption band of the nitro group in the ligand.
[c] The data were taken from the literature.[19]
(Figure 1). The larger difference in bond length between
CuꢀC1 and CuꢀC2 in 1OMe is also reproduced by the DFT
calculation, but the calculated CuꢀC2 bond lengths of 1Cl
and 1NO are shorter than those in the crystal structures
2
(Table 2). The interaction between the dz2 orbital of CuI and
the p orbital of the benzene ring is clearly seen at the
HOMO-5 level of 1H and 1Me, as shown in Figure 2, where
the CuꢀC2 interaction is much more evident than the CuꢀC1
interaction.
transition.[18] This assignment is supported by the 13C NMR
chemical shifts of the aromatic carbon atoms of the ligand
sidearm of 1Me in CD2Cl2, which are shifted to positions sim-
ilar to those of the free ligand on addition of CD3CN
(Table 4). Further studies are required for detailed discus-
sion of the MLCT transitions.
Figure 3 shows the spectral change in the titration of 1Me
by CH3CN in CH2Cl2 at ꢀ208C as a typical example. Disap-
pearance of the absorption band at around 290 nm may be
due to a ligand-exchange reaction between the aromatic
ring and the added CH3CN (Scheme 1), as demonstrated in
the CuI h2-indolyl system.[18] The association constant of
CH3CN to the copper(i) complex Kas =[CuIL¥CH3CN]/
[CuIL][CH3CN] was determined to be 8.6ꢁ0.1mꢀ1 by ana-
lyzing the absorption change (inset of Figure 3), and the Kas
values at ꢀ208C for all other complexes, determined by the
Table 4. 13C NMR data of the aromatic group in the free ligands and the
copper(i) complexes.[a]
C1
C2 (C6)
C3 (C5)
C4
[b]
L1H
dligand
141.37
134.71
ꢀ6.66
138.21
131.75
ꢀ6.46
136.61
133.36
126.29
ꢀ7.07
140.11
134.60
ꢀ5.51
149.67
147.18
ꢀ2.49
146.95
138.81
ꢀ8.14
144.53
138.27
ꢀ6.26
129.18
123.57
ꢀ5.61
129.02
123.43
ꢀ5.59
127.20
130.03
125.18
ꢀ4.85
130.67
126.31
ꢀ4.36
130.09
127.12
ꢀ2.97
127.72
121.24
ꢀ6.48
128.63
124.71
ꢀ3.92
128.57
129.34
0.77
129.24
129.89
0.65
129.62
113.98
114.54
0.56
128.51
129.14
0.63
123.57
123.91
0.34
128.53
129.45
0.92
128.62
129.53
0.91
126.12
127.42
1.30
135.63
137.36
1.73
135.08
158.30
159.33
1.03
131.66
133.12
1.46
146.71
144.49
ꢀ2.22
126.22
127.61
1.39
126.45
127.79
1.34
[c]
dcomplex
Dd[d]
[b]
L1Me
dligand
[c]
dcomplex
Dd[d]
[e]
dcomplex
[b]
L1OMe
dligand
Figure 2. d p orbital interaction of a) 1H and b) 1Me, obtained by DFT cal-
culation at the HOMO-5 level.
[c]
dcomplex
Dd[d]
[b]
L1Cl
dligand
[c]
dcomplex
Dd[d]
Copper(i) arene interaction in solution: In the UV/Vis spec-
trum, the copper(i) complexes have a characteristic absorp-
tion band around 290 nm (e=7000 11000mꢀ1 cmꢀ1)
(Table 3). This absorption band can be assigned to metal-to-
ligand charge transfer (MLCT) involving the phenyl group,
since no such absorption band around 290 nm was observed
for a similar copper(i) complex of a bis[2-(2-pyridyl)ethyl]-
amine tridentate ligand without d p interaction.[19] The ab-
sorption band around 290 nm disappeared when CH3CN
was added to a solution of the complex in CH2Cl2. A similar
absorption band at 308 nm (e=18000mꢀ1 cmꢀ1) in a CuI-h2-
indolyl complex, which disappeared on addition of CH3CN,
has also been assigned to a metal-to-indole charge-transfer
[b]
L1NO
dligand
2
[c]
dcomplex
Dd[d]
[b]
L2
L3
dligand
[c]
dcomplex
Dd[d]
[b]
dligand
[c]
dcomplex
Dd[d]
[a] In CD2Cl2. [b] Chemical shift of the free ligand. [c] Chemical shift of
the copper(i) complex. [d] Dd=dcomplexꢀdligand. [e] Chemical shifts of the
aromatic carbon atoms of the ligand sidearm of [CuIL1Me]ClO4 (0.025m)
measured in CD2Cl2 containing CD3CN (1.43m) at 258C. Under these
conditions, 18% of the copper(i) complex retains the d p interaction.
240
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 237 246

Copper Arene Interactions
237 246
maximum energy difference reaches 13.3 kJmolꢀ1 between 3
and 1NO
.
2
Studies by 1H and 13C NMR spectroscopy provided fur-
ther insight into the copper(i) arene interaction in solution.
Assignment of all signals in the 1H and 13C NMR spectra
was accomplished by employing 2D NMR techniques such
as COSY, NOESY, HMQC, and HMBC, as summarized in
1
the Experimental Section. The aromatic regions of the H
and 13C NMR spectra of 1Me in CD2Cl2 at 258C are shown in
1
Figure 4 as typical examples. Apparently, the two H and 13C
nuclei of the ortho- (2- and 6-) and meta- (3- and 5-) posi-
tions of the phenyl group of the ligand sidearm are magneti-
cally equivalent. If the copper(i)--arene interaction in solu-
tion were fixed at the C1ꢀC2 moiety, as in the crystal struc-
ture, the H and 13C nuclei in the 2- and 3-positions would
1
1
give rise to different peaks from those of H and 13C at the
Figure 3. Spectral change in the titration of 1Me (1.0î10ꢀ4 m) with CH3CN
at ꢀ208C in CH2Cl2. Inset: Plot of (AꢀA0)/(A¥ꢀA) versus [CH3CN]
based on the absorption change at 294 nm.
6- and 5-positions, respectively. Thus, the phenyl group is
possibly swinging above the copper(i) ion as illustrated in
Scheme 2, whereby the ortho (2- and 6-) and meta (3- and 5-)
positions of the phenyl group become equivalent. The 1H
and 13C NMR signals of the aromatic rings did not change at
all, even at ꢀ808C, and this suggests that the energy barrier
for the ring swinging above the copper(i) ion is very low.
Table 4 summarizes the 13C NMR data of the aromatic
groups of the free ligands and the corresponding copper(i)
complexes. The copper(i) arene interaction induces upfield
shifts of the 13C nuclei bound to the copper(i) ion [C1 and C2
(C6)], while 13C signals of C3 (C5) and C4 exhibit downfield
shifts, with the single exception of C4 in L1NO2. The upfield
shifts of C1 and C2 (C6) can be attributed to the electron-do-
nating effect of the copper(i) ion through the d p interac-
same method, are listed in Table 3. In addition, thermody-
namic parameters DH0 and DS0 for the binding of CH3CN
to the copper(i) center were determined from the tempera-
ture dependence of Kas according to the equation lnKas =
ꢀDH0/RT+DS0/R (Table 3 and Supporting Information,
Figure S8).
The strength of CH3CN binding Kas can be evaluated
from the DH0 term, which reflects differences in the binding
energy between CH3CN and the phenyl ring to the copper(i)
ion. Since the Cu MeCN bond strength may be virtually the
same in the CH3CN complex (right species in Scheme 1) ir-
tion. The electron-withdrawing substituents of 1Cl and 1NO
2
resulted in smaller upfield shifts than for the other complex-
ess, consistent with their weaker copper(i) arene interac-
tions.
The copper(i) arene interaction was also examined by
cyclic voltammetry in CH2Cl2. The copper(i) complexes ex-
hibit a reversible or quasireversible redox couple due to
one-electron oxidation/reduction of the copper center. A
typical example of a cyclic voltammogram is shown in
Figure 5, and the redox potentials (E1/2 versus Fc/Fc+) of all
the copper(i) complexes are listed in Table 5. In this case,
the E1/2 values may reflect an inductive effect of the arene
substituents, that is, the more strongly electron donating the
substituent is, the more positive is the E1/2 value in the series
of 1X.[40]
Scheme 1. Ligand exchange between the h2-arene interaction and aceto-
nitrile.
respective of the nature of X, the difference in DH0 may di-
rectly reflect the energy difference in the copper(i) arene in-
teraction in the d p complexes (left-hand side of Scheme 1),
that is, the smaller the ꢀDH0 value, the stronger the cop-
per(i) arene interaction. Apparently, the copper(i) arene in-
teraction is stronger in 1H and 1Me (smaller in Kas), while
para OMe, Cl, and NO2 substituents weaken the copper(i)
arene interaction in the series of 1X. There is a compensat-
ing effect between DH0 and DS0. On the other hand, the
methyl and phenyl groups at the benzylic position in 2 and 3
result in smaller ꢀDH0 and larger ꢀDS0 values. In the case
of 2 and 3, there is probably steric crowding due to the close
proximity of R (Me and Ph) and the coordinated CH3CN,
which leads to the weaker binding. Thus, the Kas values of 2
and 3 are significantly smaller than those of 1X. Then, the
Copper(i) dioxygen reactivity: In our previous study[41] cop-
per(i) complex 1H was shown to react with O2 at a low tem-
perature to give (m-h2:h2-peroxo)dicopper(ii) complex 4, a
model compound for the active oxygen intermediate of he-
mocyanin, tyrosinase, and catechol oxidase (Scheme 3).[42]
Oxygenation reactions of other copper(i) complexes 1X
(X=Me, OMe, Cl, and NO2) were examined under the
same experimental conditions (at ꢀ808C in CH2Cl2). The
copper(i) complexes 1X reacted with O2 to give an oxygenat-
ed intermediate which was ESR-silent and gave essentially
the same absorption spectrum as that of the (m-h2:h2-peroxo)-
dicopper(ii) complex derived from 1H (lmax =364 nm).[41]
241
Chem. Eur. J. 2004, 10, 237 246
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

S. Fukuzumi, S. Itoh et al.
FULL PAPER
These features are strong evi-
dence for the formation of (m-
h2:h2-peroxo)dicopper(ii) com-
plexes 4 in all cases. Interesting-
ly, however, the formation rate
kobs of the peroxo intermediate
depended significantly on the
para substituent X of L1X, and
the copper(i) complexes 2 and 3
exhibited significantly lower re-
activity toward O2 under the
same experimental conditions
(Table 5).[19] Thus, the reactivity
of the copper(i) complexes
toward O2 is significantly al-
tered by the ligand substituents
R and X. More importantly, the
trend in formation rates kobs of
the peroxo intermediates 4 is
very similar to that of Kas, that
is, the stronger the copper(i)
arene interaction, the lower is
the reactivity of 1X towards di-
oxygen. Thus, it is concluded
that the interaction between O2
and CuI is significantly affected
by the para substituents X and
the benzylic substituents R via
the copper(i) arene interaction.
Summary: The d p interaction
in copper(i) complexes of
a
series of bis[2-(2-pyridyl)ethyl]-
amine tridentate ligands has
been systematically investigated
for the first time by X-ray crys-
tallographic analysis, UV/Vis
and 2D-NMR spectroscopy,
cyclic voltammetry, and DFT
calculations. It is suggested that
the interaction involves the dz2
orbital of copper(i) and a p or-
bital of the arene ring, whereby
Figure 4. 1H and 13C NMR (aromatic region) of 1Me in CD2Cl2.
the copper(i) ion acts as an
electron donor. Thus, the inter-
action between the d orbital of CuI and the aromatic p orbi-
tal is stronger when the para substituent X is an electron-do-
nating group such as H or Me, while electron-withdrawing
substituents such as Cl and NO2 weaken the overlap of the
two orbitals. In the case of OMe, the situation is somewhat
complicated, since the methoxyl group has both p-donor ca-
pability and a s-electron-withdrawing effect, especially on
the position meta to X (C2). Overall, the OMe group may
act as a s-electron acceptor, weakening the d p interaction
to some extent. Further studies are required to obtain more
detailed insight into the substituent effects on the copper(i)-
arene interaction.
Scheme 2. Swinging of the phenyl ring in the copper(i) arene complex.
242
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 237 246

Copper Arene Interactions
237 246
The Pt working electrodes were polished with an alumina polishing sus-
pension and rinsed with CH2Cl2 before use. The counterelectrode was a
Pt wire. A silver pseudoreference electrode was used, and the potentials
were determined relative to ferrocene/ferricenium (Fc/Fc+). All electro-
chemical measurements were carried out at 258C under an atmospheric
pressure of Ar in a glove box (DBO-1KP, Miwa Co. Ltd.).
N,N-Bis[2-(2-pyridyl)ethyl]-2-(4-methyoxyphenyl)ethylamine
(L1OMe
)
was prepared by reaction of 2-(4-methoxyphenyl)ethylamine (3.0 g,
20 mmol) and 2-vinylpyridine (10.5 g, 100 mmol) in refluxing methanol
(50 mL) containing acetic acid (6.0 g, 100 mmol) for 10 d, and purified by
column chromatography (SiO2), as for other ligands reported previous-
ly.[41,44] Pale brown oil (54.3% yield); FAB-HRMS(positive ion): m/z:
362.227 [M+1].
L1H: 1H NMR (600 MHz, CD2Cl2, 278C): d=2.71 (dd, J=10.3, 6.8 Hz, 2
H; NCH2CH2Ph), 2.80 (dd, J=10.3, 6.8 Hz, 2H; NCH2CH2Ph), 2.89 (dd,
J=8.1, 5.6 Hz, 4H; NCH2CH2Py), 2.97 (dd, J=8.1, 5.6 Hz, 4H;
NCH2CH2Py), 7.06 (d, J=7.7 Hz, 2H; HPy-3), 7.10 (ddd, J=7.5, 4.9,
0.9 Hz, 2H; HPy-5), 7.14 (d, J=7.0 Hz, 2H; HPh-2, HPh-2’), 7.17 (t, J=
7.3 Hz, 1H; HPh-4), 7.25 (t, J=7.2 Hz, 2H; HPh-3 and HPh-3’), 7.55 (td, J=
7.5, 1.8 Hz, 2H; HPy-4), 8.49 ppm (ddd, J=4.9, 1.8, 0.9 Hz, 2H; HPy-6);
13C NMR (600 MHz, CD2Cl2, 278C): 34.12 (NCH2CH2Ph), 36.30
(NCH2CH2Py), 54.22 (NCH2CH2Py), 56.33 (NCH2CH2Ph), 121.34 (CPy-5),
123.70 (CPy-3), 126.12 (CPh-4), 128.57 (CPh-3) 129.18 (CPh-2), 136.41 (CPy-4),
141.37 (CPh-1), 149.46 (CPy-6), 161.27 ppm (CPy-2).
Figure 5. Cyclic voltammogram of [CuI(L1Me)]ClO4 (1Me, 2.0î10ꢀ3 m) in
CH2Cl2 containing 0.1m TBAP; working electrode Pt, counter electrode
Pt, pseudo-reference electrode Ag wire, scan rate 10 mVsꢀ1
.
Table 5. Electrochemical data from cyclic voltammetry (E1/2 and DE)[a]
and the second-order rate constants kobs for the formation of (m-h2:h2-per-
oxo)dicopper(ii) complexes 4.[b]
L1Me 1H NMR (600 MHz, CD2Cl2, 278C): d=2.30 (s, 3H; CH3), 2.67
:
(dd, J=8.5 and 5.3 Hz, 2H; NCH2CH2Ar), 2.77 (dd, J=8.5, 5.3 Hz, 2H;
NCH2CH2Ar), 2.89 (dd, J=8.1, 5.4 Hz, 4H; NCH2CH2Py), 2.97 (dd, J=
8.1, 5.4 Hz, 4H; NCH2CH2Py), 7.02 (d, J=7.9 Hz, 2H; HAr-2, HAr-2’), 7.05
(d, J=7.5 Hz, 2H; HPy-3), 7.07 (d, J=7.9 Hz, 2H; HAr-3, HAr-3’), 7.09 (ddd,
J=7.7, 4.9, 1.0 Hz, 2H; HPy-5), 7.54 (td, J=7.7, 1.9 Hz, 2H; HPy-4),
8.49 ppm (ddd, J=4.9, 1.9, 1.0 Hz, 2H; HPy-6); 13C NMR (600 MHz,
CD2Cl2, 278C): 21.07 (OCH3), 33.68 (NCH2CH2Ar), 36.46 (NCH2CH2Py),
54.26 (NCH2CH2Py), 56.44 (NCH2CH2Ar), 121.28 (CPy-5), 123.65 (CPy-3),
129.02 (CAr-2), 129.24 (CAr-3), 135.63 (CAr-4), 136.30 (CPy-4), 138.21 (CAr-1),
149.51 (CPy-6), 161.37 ppm (CPy-2).
Complex
E1/2 [V] versus Fc/Fc+
DE [V]
kobs [mꢀ1 sꢀ1
]
1H
0.07
0.05
0.05
ꢀ0.01
ꢀ0.03
0.16
0.53
0.12
0.18
0.24
0.15
0.17
0.34
0.28ꢁ0.01
0.70ꢁ0.05
1.96ꢁ0.10
1.51ꢁ0.06
1Me
1OMe
1Cl
1NO
2
3
5.21ꢁ0.30
2
[c]
[c]
0.15
L1OMe 1H NMR (600 MHz, CD2Cl2, 278C): d=2.65 (dd, J=10.0, 7.1 Hz,
:
2H; NCH2CH2Ar), 2.76 (dd, J=10.0, 7.1 Hz, 2H; NCH2CH2Ar), 2.89
(dd, J=8.0, 5.4 Hz, 4H; NCH2CH2Py), 2.97 (dd, J=8.0, 5.4 Hz, 4H;
[a] The electrochemical measurements were performed in CH2Cl2 con-
taining 0.1m tetrabutylammonium perchlorate (TBAP) at a scan rate of
10 50 mVsꢀ1 at 258C. [b] At ꢀ808C in CH2Cl2. [c] Very slow.
NCH2CH2Py), 3.77 (s, 3H; OCH3), 6.80 (dt, J=8.7, 2.1 Hz, 2H; HAr-3
,
HAr-3’), 7.05 (d, J=7.7, 2.1 Hz, 2H; HAr-2, HAr-2’), 7.06, (d, J=7.7 Hz, 2H;
HPy-3), 7.09 (ddd, J=7.7, 4.9, 1.0 Hz, 2H; HPy-5), 7.55 (td, J=7.7, 1.8 Hz, 2
H; HPy-4), 8.50 ppm (ddd, J=4.9, 1.8, 1.0 Hz, 2H; HPy-6); 13C NMR
(600 MHz, CD2Cl2, 278C): 33.25 (NCH2CH2Ar), 36.47 (NCH2CH2Py),
54.26 (NCH2CH2Py), 55.53 (OCH3), 56.53 (NCH2CH2Ar), 113.98 (CAr-3),
121.27 (CPy-5), 123.63 (CPy-3), 130.03 (CAr-2), 133.36 (CAr-1), 136.28 (CPy-4),
149.50 (CPy-6), 158.30 (CAr-4), 161.37 ppm (CPy-2).
L1Cl 1H NMR (600 MHz, CD2Cl2, 278C): d=2.66 (dd, J=7.8, 5.6 Hz, 2
:
Scheme 3. Reaction of 1X with O2.
H; NCH2CH2Ar), 2.76 (dd, J=7.8, 5.6 Hz, 2H; NCH2CH2Ar), 2.85 (dd,
J=7.9, 6.6 Hz, 4H; NCH2CH2Py), 2.95 (dd, J=7.9, 6.6 Hz, 4H;
NCH2CH2Py), 7.00 (d, J=7.8 Hz, 2H; HPy-3), 7.03 (d, J=8.4 Hz, 2H; HAr-
2, HAr-2’), 7.09 (ddd, J=7.6, 4.9, 1.0 Hz, 2H; HPy-5), 7.20 (d, J=8.4 Hz, 2
H; HAr-3, HAr-3’), 7.53 (td, J=7.6, 1.9 Hz, 2H; HPy-4), 8.49 ppm (ddd, J=
4.9, 1.9, 1.0 Hz, 2H; HPy-6); 13C NMR (600 MHz, CD2Cl2, 278C): 33.55
(NCH2CH2Ar), 36.44 (NCH2CH2Py), 54.20 (NCH2CH2Py), 56.06
(NCH2CH2Ar), 121.30 (CPy-5), 123.66 (CPy-3), 128.51 (CAr-3), 130.67 (CAr-2),
131.66 (CAr-4), 136.29 (CPy-4), 140.11 (CAr-1), 149.52 (CPy-6), 161.29 ppm
(CPy-2).
Experimental Section
General: All chemicals used in this study, except for the ligands and the
complexes, were commercial products of the highest available purity and
were further purified by standard methods, if necessary.[43] Synthetic pro-
cedures for the ligands L1X and L3, except for L1OMe, were reported pre-
viously.[19,41,44] FT-IR spectra were recorded with a Shimadzu FTIR-
8200PC. UV/Vis spectra were measured with a Hewlett Packard HP8453
diode-array spectrophotometer with a Unisoku thermostatically control-
led cell holder designed for low-temperature measurements (USP-203).
Mass spectra were recorded with a JEOL JMS-700T Tandem MS station
or a PE SCIEX API 150EX (for ESI-MS). NMR spectra were recorded
L1NO
:
1H NMR (600 MHz, CD2Cl2, 278C): d=2.76 2.82 (m, 4H;
2
NCH2CH2Ar), 2.84 (dd, J=7.7, 6.8 Hz, 4H; NCH2CH2Py), 2.96 (dd, J=
7.7, 6.8 Hz, 4H; NCH2CH2Py), 7.00 (d, J=7.7 Hz, 2H; HPy-3), 7.09 (ddd,
J=8.7, 4.9, 0.9 Hz, 2H; HPy-5), 7.21 (d, J=8.7 Hz, 2H; HAr-2, HAr-2’), 7.53
(td, J=7.7, 1.8 Hz, 2H; HPy-4), 8.03 (d, J=8.7 Hz, 2H; HAr-3, HAr-3’),
8.48 ppm (ddd, J=4.9, 1.8, 0.9 Hz, 2H; HPy-6); 13C NMR (600 MHz,
CD2Cl2, 278C): 34.13 (NCH2CH2Ar), 36.37 (NCH2CH2Py), 54.07
(NCH2CH2Py), 55.50 (NCH2CH2Ar), 121.36 (CPy-5), 123.57 (CAr-3), 123.64
(CPy-3), 130.09 (CAr-2), 136.32 (CPy-4), 146.71 (CAr-4), 149.67 (CAr-1), 149.54
(CPy-6), 161.15 ppm (CPy-2).
1
on a Bruker Avance 600 spectrometer. H NMR spectra were referenced
to the residual proton resonance of the solvent, and 13C NMR spectra to
the solvent resonance (CD2Cl2: d(1H)=5.32, d(13C)=53.8). Complete
peak assignments in the 1H and 13C NMR spectra of the ligands and the
copper(i) complexes were accomplished by employing 2D NMR techni-
ques (COSY, NOESY, HMQC, and HMBC). Cyclic voltammetry (CV)
was performed on an ALS-630A electrochemical analyzer in anhydrous
CH2Cl2 containing 0.1m NBu4ClO4 (TBAP) as supporting electrolyte.
L2: 1H NMR (600 MHz, CD2Cl2, 278C): d=1.15 (d, J=6.9 Hz, 3H;
NCH2CH(CH3)Ph), 2.61 (dd, J=12.8, 8.0 Hz, 1H; NCH2CH(CH3)Ph),
2.71 (dd, J=12.8, 8.0 Hz, 1H; NCH2CH(CH3)Ph), 2.82 2.98 (m, 7H;
243
Chem. Eur. J. 2004, 10, 237 246
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

S. Fukuzumi, S. Itoh et al.
FULL PAPER
NCH2CH(CH3)Ph, NCH2CH2Py, NCH2CH2Py), 7.00 (d, J=7.8 Hz, 2H;
ion): m/z: 407.9 [M+]; elemental analysis (%) calcd for C23H28O4.5N3-
H
H
Py-3), 7.11 (ddd, J=7.5, 6.0, 0.9 Hz, 2H; HPy-5), 7.16 (d, J=7.5 Hz, 2H;
Ph-2, HPh-2’), 7.19 (t, J=7.5 Hz, 1H; HPh-4), 7.28 (t, J=7.5 Hz, 2H; HPh-3
CuCl: C 53.38, H 5.45, N 8.12; found: C 53.61, H 5.42, N 8.00.
[CuI(L1Cl)]ClO4 (1Cl) was prepared in 59% yield in a similar manner to
the synthesis of 1OMe by treating L1Cl (109.8 mg, 0.3 mmol) with
[CuI(CH3CN)4]ClO4 (96.1 mg, 0.3 mmol). All procedures were performed
in a glove box ([O2]<0.1 ppm). 1H NMR (600 MHz, CD2Cl2, 278C): d=
2.77 (t, J=6.2 Hz, 2H; NCH2CH2Ar), 2.9 3.2 (brm, 10H; NCH2CH2Ar,
NCH2CH2Py, NCH2CH2Py), 7.11 (d, J=8.1 Hz, 2H; HAr-2 and HAr-2’),
7.30 (d, J=8.1 Hz, 2H; HAr-3, HAr-3’), 7.35 7.40 (m, 4H; HPy-3, HPy-5), 7.84
(t, J=7.5 Hz, 2H; HPy-4), 8.22 ppm (br, 2H; HPy-6); 13C NMR (600 MHz,
CD2Cl2, 278C): 32.53 (NCH2CH2Ar), 34.35 (NCH2CH2Py), 54.68
(NCH2CH2Py), 56.27 (NCH2CH2Ar), 123.71 (CPy-5), 126.31 (CAr-2), 126.37
(CPy-3), 129.14 (CAr-3), 133.12 (CAr-4), 134.60 (CAr-1), 139.23 (CPy-4), 150.36
(CPy-6), 160.44 ppm (CPy-2); FT-IR (KBr): n˜ =1121, 1089, 625 cmꢀ1
(ClO4ꢀ); ESI-MS (positive ion.): m/z: 428.0 [M+]; elemental analysis (%)
calcd for C22H24O4N3CuCl2: C 49.96, H 4.57, N 7.94; found: C 49.92, H,
4.77, N, 7.61.
,
HPh-3’), 7.54 (td, J=7.7, 1.9 Hz, 2H; HPy-4), 8.52 ppm (d, J=4.2, Hz, 2H;
H
36.08 (NCH2CH2Py), 38.79 (NCH2CH(CH3)Ph), 54.73 (NCH2CH2Py),
62.68 (NCH2CH(CH3)Ph), 121.33 (CPy-5), 123.82 (CPy-3), 126.22 (CPh-4),
127.72 (CPh-2), 128.53 (CPh-3), 136.46 (CPy-4), 146.95 (CPh-1), 149.26 (CPy-6),
161.28 ppm (CPy-2).
Py-6); 13C NMR (600 MHz, CD2Cl2, 278C): 19.98 (NCH2CH(CH3)Ph),
1
L3: H NMR (600 MHz, CD2Cl2, 278C): d=2.81 (dd, J=7.9, 6.8 Hz, 4H;
NCH2CH2Py), 2.97 (dd, J=7.9, 6.8 Hz, 4H; NCH2CH2Py), 3.20 (d, J=
7.7 Hz, 2H; NCH2CHPh2), 4.15 (t, J=7.7 Hz, 1H; NCH2CHPh2), 6.87 (d,
J=7.8 Hz, 2H; HPy-3), 7.08 (ddd, J=7.4, 4.9, 1.1 Hz, 2H; HPy-5), 7.18 (tt,
J=7.8, 1.3 Hz, 2H; HPh-4), 7.21 (dd, J=7.1, 1.3 Hz, 4H; HPh-2, HPh-2’), 7.26
(td, J=7.3, 1.9 Hz, 4H; HPh-3, HPh-3’), 7.49 (ddd, J=7.8, 7.4, 1.8 Hz, 2H;
HPy-4), 8.50 ppm (ddd, J=4.9, 1.8, 0.8 Hz, 2H; HPy-6); 13C NMR
(600 MHz, CD2Cl2, 278C): 35.95 (NCH2CH2Py), 50.17 (NCH2CHPh2),
54.49 (NCH2CH2Py), 60.19 (NCH2CHPh2), 121.24 (CPy-5), 123.67 (CPy-3),
126.45 (CPh-4), 128.62 (CPh-3), 128.63 (CPh-2), 136.27 (CPy-4), 144.53 (CPh-1),
149.45 (CPy-6), 161.28 ppm (CPy-2).
[CuI(L1NO )]ClO4 (1NO ) was prepared in 68% yield in a similar manner
2
2
to the synthesis of 1OMe by treating L1NO (112.9 mg, 0.3 mmol) with
2
[CuI(CH3CN)4]ClO4 (96.1 mg, 0.3 mmol). All procedures were performed
in a glove box ([O2]<0.1 ppm). 1H NMR (600 MHz, CD2Cl2, 278C): d=
2.86 (t, J=7.0 Hz, 2H; NCH2CH2Ar), 3.05 3.15 (br, 8H; NCH2CH2Py),
3.07 (t, J=7.0 Hz, 2H; NCH2CH2Ar), 7.33 (d, J=8.6 Hz, 2H; HAr-2, HAr-
2’), 7.38 (ddd, J=7.8, 5.4, 1.2 Hz, 2H; HPy-5), 7.41 (d, J=7.8 Hz, 2H; HPy-
3), 7.87 (td, J=7.8, 1.8 Hz, 2H; HPy-4), 8.11 (d, J=8.6 Hz, 2H; HAr-3, HAr-
3’), 8.37 ppm (ddd, J=5.4, 1.8, 0.8 Hz, 2H; HPy-6); 13C NMR (600 MHz,
CD2Cl2, 278C): 32.88 (NCH2CH2Ar), 34.83 (NCH2CH2Py), 55.04
(NCH2CH2Py), 56.83 (NCH2CH2Ar), 123.86 (CPy-5), 123.91 (CAr-3), 126.35
(CPy-3), 127.12 (CAr-2), 139.53 (CPy-4), 144.49 (CAr-4), 147.18 (CAr-1), 150.63
(CPy-6), 160.58 ppm (CPy-2); FT-IR (KBr): n˜ =1514, 1345 (NO2) 1113, 1089,
625 cmꢀ1 (ClO4ꢀ); ESI-MS (positive ion): m/z: 439.0 [M+]; elemental
analysis (%) calcd for C22H25O6.5N4CuCl: C 48.18, H 4.59, N 10.22;
found: C 48.41, H 4.50, N 10.07.
Caution! The perchlorate salts prepared in this study are all potentially
explosive and should be handled with care.
[CuI(L1OMe)]ClO4 (1OMe): L1OMe (108.4 mg, 0.3 mmol) was treated with
[CuI(CH3CN)4]ClO4 (96.1 mg, 0.3 mmol) in CH2Cl2 (5 mL) under an Ar
atmosphere. After stirring for 30 min at room temperature, insoluble ma-
terial was removed by filtration. Addition of diethyl ether (100 mL) to
the filtrate gave a pale yellow powder that precipitated on allowing the
mixture to stand for several minutes. The supernatant was then removed
by decantation, and the remaining pale yellow solid was washed with di-
ethyl ether three times and dried (68% yield). All procedures were per-
formed in a glove box (DBO-1 KP, Miwa Co. Ltd., [O2]<0.1 ppm).
1H NMR (600 MHz, CD2Cl2, 278C): d=2.73 (brm, 2H; NCH2CH2Ar),
2.8 3.3 (brm, 10H; NCH2CH2Ar, NCH2CH2Py, NCH2CH2Py), 3.77 (s, 3
H; OCH3), 6.89 (d, J=8.4 Hz, 2H; HAr-3, HAr-3’), 7.11 (d, J=8.4 Hz, 2H;
[CuI(L2)]ClO4 (2): Synthetic procedure, mass spectrum, and elemental
analysis of this compound were reported previously.[19] The NMR data re-
corded with a Bruker Avance 600 spectrometer in CD2Cl2 are as follows:
1H NMR (600 MHz, CD2Cl2, 278C): d=1.20 (d, J=6.9 Hz, 3H;
NCH2CH(CH3)Ph), 2.63 (dd, J=12.9, 4.2 Hz, 1H; NCH2CH(CH3)Ph),
2.75 (br, 1H; NCH2CH2Py), 2.80 (br, 1H; NCH2CH2Py), 2.85 (br, 1H;
NCH2CH2Py), 2.93 (br, 1H; NCH2CH2Py), 2.95 (br, 1H; NCH2CH2Py),
2.98 (m, 1H; NCH2CH(CH3)Ph), 3.13 (brt, J=12.0 Hz, 1H;
NCH2CH2Py), 3.23 (brt, J=14.6 Hz, 1H; NCH2CH2Py), 3.30 (brt, J=
11.7 Hz, 1H; NCH2CH2Py), 3.44 (t, J=12.9 Hz, 1H; NCH2CH(CH3)Ph),
7.18 (d, J=7.6 Hz, 2H; HPh-2, HPh-2’), 7.21 (br, 1H; HPy-5), 7.31 (br, 1H;
H
Ar-2, HAr-2’), 7.37, (br, 4H; HPy-3, HPy-5), 7.82 (brt, J=7.0 Hz, 2H; HPy-4),
8.19 ppm (br, 2H; HPy-6); 13C NMR (600 MHz, CD2Cl2, 278C): 32.49
(NCH2CH2Ar), 34.04 (NCH2CH2Py), 54.56 (NCH2CH2Py), 55.88
(OCH3), 56.28 (NCH2CH2Ar), 114.54 (CAr-3), 123.59 (CPy-5), 125.18 (CAr-
2), 126.64 (CPy-3), 126.29 (CAr-1), 138.86 (CPy-4), 150.13 (CPy-6), 159.33 (CAr-
4), 160.43 ppm (CPy-2); FT-IR (KBr): n˜ =1247, 1028 (OCH3) 1115, 1089,
625 cmꢀ1 (ClO4ꢀ); ESI-MS (positive ion), m/z: 424.0 [M+]; elemental
analysis (%) calcd for C23H27O5N3CuCl: C 52.67, H 5.19, N 8.01; found:
C 52.38, H 5.23, N 7.95.
[CuI(L1H)]ClO4 (1H): Synthetic procedure, mass spectrum, and elemental
analysis of this compound were reported previously.[41] The NMR data re-
corded on a Bruker Avance 600 spectrometer in CD2Cl2 are as follows:
1H NMR (600 MHz, CD2Cl2, 278C): d=2.82 (t, J=6.1 Hz, 2H;
NCH2CH2Ph), 3.18 (t, J=6.1 Hz, 2H; NCH2CH2Ph), 2.8 3.2 (brm, 8H;
NCH2CH2Py, NCH2CH2Py), 7.18 (d, J=7.5 Hz, 2H; HPh-2 and HPh-2’),
7.30 (m, 2H; HPy-5), 7.31 (t, J=7.5 Hz, 1H; HPh-4), 7.33 (d, J=7.8 Hz, 2
H; HPy-3), 7.38 (t, J=7.5 Hz, 2H; HPh-3, HPh-3’), 7.79 (t, J=7.8, 1.8 Hz, 2H;
H
H
Py-3), 7.33 (t, J=7.6 Hz, 1H; HPh-4), 7.36 (br, 1H; HPy-5), 7.37 (br, 1H;
Py-3), 7.39 (t, J=7.6 Hz, 2H; HPh-3, HPh-3’), 7.76 (br, 1H; HPy-4), 7.80 (br,
1H; HPy-4), 7.94 (br, 1H; HPy-6), 8.04 ppm (br, 1H; HPy-6); 13C NMR
(600 MHz,
CD2Cl2,
278C):
20.93
(NCH2CH(CH3)Ph),
33.63
(NCH2CH2Py), 33.98 (NCH2CH2Py), 38.96 (NCH2CH(CH3)Ph), 55.33
(NCH2CH2Py), 62.45 (NCH2CH(CH3)Ph), 121.24 (CPh-2), 123.37 (CPy-5),
123.66 (CPy-5’), 126.04 (CPy-3), 126.25 (CPy-3’), 127.61 (CPh-4), 129.45 (CPh-3),
138.81 (CPh-1), 138.96 (CPy-4), 149.87 (CPy-6), 150.28 (CPy-6’), 160.27 ppm
(CPy-2).
H
Py-4), 8.08 ppm (d, J=4.1 Hz, 2H; HPy-6); 13C NMR (600 MHz, CD2Cl2,
278C): 31.95 (NCH2CH2Ph), 34.03 (NCH2CH2Py), 54.54 (NCH2CH2Py),
56.02 (NCH2CH2Ph), 123.37 (CPy-3), 123.57 (CPh-2), 127.40 (CPy-5), 127.42
(CPh-4), 129.34 (CPhr-3), 138.90 (CPy-4), 134.71 (CPh-1), 150.23 (CPy-6),
160.30 ppm (CPy-2).
[CuI(L1Me)]ClO4 (1Me) was prepared in 56% yield in a similar manner to
the synthesis of 1OMe by treating L1Me (103.6 mg, 0.3 mmol) with
[CuI(CH3CN)4]ClO4 (96.1 mg, 0.3 mmol). All procedures were performed
[CuI(L3)]ClO4 (3) was prepared in 67% yield in a similar manner to the
synthesis of 1OMe by treating L3 (203.8 mg, 0.5 mmol) with
[CuI(CH3CN)4]ClO4 (160.1 mg, 0.5 mmol). All procedures were per-
formed in a glove box ([O2]<0.1 ppm). 1H NMR (600 MHz, CD2Cl2, 27
8C): d=2.90 (br, 2H; NCH2CH2Py), 3.08 (br, 2H; NCH2CH2Py), 3.15
(br, 4H; NCH2CH2Py, NCH2CH2Py), 3.38 (d, J=8.6 Hz, 2H;
NCH2CHPh2), 4.20 (t, J=8.6 Hz, 1H; NCH2CHPh2), 7.23 (br, 2H; HPh-4),
7.29 (m, 2H; HPy-5), 7.32 (m, 4H; HPh-2, HPh-2’), 7.35 (br, 4H; HPh-3, HPh-3’),
7.37 (m, 2H; HPy-3), 7.81 (td, J=7.8, 1.8 Hz, 2H; HPy-4), 7.95 ppm (ddd,
J=5.3, 0.9, 0.6 Hz, 2H; HPy-6); 13C NMR (600 MHz, CD2Cl2, 278C): 33.80
(NCH2CH2Py), 49.41 (NCH2CHPh2), 54.79 (NCH2CH2Py), 60.44
(NCH2CHPh2), 123.64 (CPy-5), 124.71 (CPh-2), 126.30 (CPy-3), 127.79 (CPh-4),
129.53 (CPh-3), 138.27 (CPh-1), 138.98 (CPy-4), 150.13 (CPy-6), 160.27 ppm
(CPy-2); FT-IR (KBr): n˜ =1115, 1089, 625 cmꢀ1 (ClO4ꢀ); FAB-MS(positive
ion): m/z: 470.1 [M+]; elemental analysis for [CuI(L3)]ClO4, calcd (%)
for C28H29O4N3CuCl: C 58.94, H 5.12, N 7.36; found: C 58.68, H 5.04, N
7.29.
1
in a glove box ([O2]< 0.1 ppm). H NMR (600 MHz, CD2Cl2, 278C): d=
2.32 (s, 3H; CH3), 2.76 (t, J=6.0 Hz, 2H; NCH2CH2Ar), 3.15 (t, J=
6.0 Hz, 2H; NCH2CH2Ar), 2.8 3.2 (brm, 8H; NCH2CH2Py,
NCH2CH2Py), 7.08 (d, J=7.8 Hz, 2H; HAr-2, HAr-2’), 7.18 (d, J=7.8 Hz, 2
H; HAr-3, HAr-3’), 7.32 (td, J=7.8, 1.1 Hz, 2H; HPy-5), 7.33 (d, J=7.8 Hz, 2
H; HPy-3), 7.80 (td, J=7.8, 1.8 Hz, 2H; HPy-4), 8.10 ppm (dt, J=7.8,
1.8 Hz, 2H; HPy-6); 13C NMR (600 MHz, CD2Cl2, 278C): 21.12 (CH3),
32.86 (NCH2CH2Ar), 32.98 (NCH2CH2Py), 54.55 (NCH2CH2Py), 56.38
(NCH2CH2Ar), 123.43 (CAr-2), 123.48 (CPy-5), 126.18 (CPy-3), 129.89 (CAr-3),
131.75 (CAr-1), 137.36 (CAr-4), 138.84 (CPy-4), 150.10 (CPy-6), 160.31 ppm
(CPy-2); FT-IR (KBr): n˜ =1112, 1089, 625 cmꢀ1 (ClO4ꢀ); ESI-MS (positive
244
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 237 246

Copper Arene Interactions
237 246
X-ray structure determination: Single crystals of the copper(i) complexes
for X-ray structural analysis were obtained by vapor diffusion of diethyl
ether into a solution of the complex in CH2Cl2. In the case of [CuI(L1
[1] Chem. Rev. 1997, 97, 1231 1734: Special issue on molecular recogni-
tion.
[2] J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 1995.
[3] T. E. Creighton, Proteins: Structures and Molecular Properties, 2nd
ed., Freeman, New York, 1993.
NO2
)]ClO4, a few drops of methanol were added to a solution of [CuI(L1
NO2
)]ClO4 in CH2Cl2. Single crystals were mounted on
a CryoLoop
(Hamptom Research Co.). X-ray diffraction data were collected by a
Rigaku RAXIS-RAPID imaging plate two-dimensional area detector
[4] W. Saenger, Principles of Nucleic Acid Structure, Springer, Heidel-
berg, 1984.
[5] D. A. Dougherty, Science 1996, 271, 163 168.
[6] M. Nishio, M. Hirota, Y. Umezawa, The CH-p Interactions, Wiley-
VCH, New York, 1998.
[7] O. Yamauchi, A. Odani, M. Takani, J. Chem. Soc. Dalton Trans.
2002, 3411 3421.
[8] E. L. Muetterties, J. R. Bleeke, E. J. Wucherer, T. A. Albright,
Chem. Rev. 1982, 82, 499 525.
[9] E. Maslowsky, Jr., J. Chem. Educ. 1993, 70, 980 984.
[10] G. W. Parshall, S. D. Ittel, Homogeneous Catalysis: The Applications
and Chemistry of Catalysis by Soluble Transition Metal Complexes,
2nd ed., Wiley-Interscience, New York, 1992.
[11] R. W. Turner, E. L. Amma, J. Am. Chem. Soc. 1966, 88, 1877
1882.
[12] M. B. Dines, P. H. Bird, J. Chem. Soc. Chem. Commun. 1973, 12.
[13] P. F. Rodesiler, E. L. Amma, J. Chem. Soc. Chem. Commun. 1974,
599 600.
[14] M. Pasquali, C. Floriani, A. Gaetani-Manfredotti, Inorg. Chem.
1980, 19, 1191 1197.
[15] H. Schmidbaur, W. Bublak, B. Huber, G. Reber, G. M¸ller, Angew.
Chem. 1986, 98, 1108; Angew. Chem. Int. Ed. Engl. 1986, 25, 1089
1090.
[16] M. Niemeyer, Organometallics 1998, 17, 4649 4656.
[17] a) W. S. Striejewske, R. R. Conry, Chem. Commun. 1998, 555 556;
b) R. R. Conry, W. S. Striejewske, A. A. Tipton, Inorg. Chem. 1999,
38, 2833.
[18] Y. Shimazaki, H. Yokoyama, O. Yamauchi, Angew. Chem. 1999, 111,
2561 2563; Angew. Chem. Int. Ed. 1999, 38, 2401 2403.
[19] T. Osako, Y. Tachi, M. Taki, S. Fukuzumi, S. Itoh, Inorg. Chem.
2001, 40, 6604 6609.
[20] Bioinorganic Chemistry of Copper (Eds.: K. D. Karlin, Z. Tyeklµr),
Chapman & Hall, New York, 1993.
using graphite-monochromated MoKa radiation (l=0.71070 ä) to
2
qmax =55.08 or CuKa radiation (l=1.54186 ä) to 2qmax =136.68. All crys-
tallographic calculations, except for [CuI(L1H)]ClO4, were performed by
using the Crystal Structure software package of the Rigaku Corporation
and Molecular Structure Corporation (Crystal Structure: Crystal Struc-
ture Analysis Package version 2.0, Rigaku Corp. and Molecular Structure
Corp., 2001). The crystallographic calculation of [CuI(L1H)]ClO4 was per-
formed by using the teXsan crystallographic software package of Molecu-
lar Structure Corporation (1999). Crystal structures except for
[CuI(L1H)]ClO4 were solved by direct methods and refined by full-matrix
least-squares methods using SIR-92. All non-hydrogen atoms and hydro-
gen atoms were refined anisotropically and isotropically, respectively. In
the case of [CuI(L1H)]ClO4, the crystal structure was solved by direct
methods and refined by full-matrix least squares using SHELX-97. All
non-hydrogen atoms of [CuI(L1H)]ClO4 were refined anisotropically, but
the hydrogen atoms were not refined. Selected bond lengths and angles
are given in Supporting Information (Tables S1 S6), and the X-ray struc-
ture determination and details of the crystallographic data are deposited
as a CIF file.
CCDC-213269 213274 contain 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 Crys-
tallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK;
fax: (+44)1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk).
Kinetic measurements: The reactions of the copper(i) complexes with O2
were carried out in a UV/Vis cell with 1 mm path length that was held in
a Unisoku thermostatically controlled cell holder USP-203 (a desired
temperature can be fixed within ꢁ0.58C). After keeping the deaerated
solution of the copper(i) complex (2.5î10ꢀ3 m) in the cell at a desired
temperature for several minutes, dry dioxygen gas was continuously sup-
plied by gentle bubbling from a thin needle. Formation of the (m-h2:h2-
peroxo)dicopper(ii) complex was monitored by means of the increase in
the absorption at 364 nm. The reactions obeyed second-order kinetics,
and the second-order rate constants kobs were obtained as the slopes of
linear second-order plots of (AꢀA0)/{(A¥ꢀA)[Cu]0} versus time, where
A0 and A¥ are the initial and final absorption at 364 nm and [Cu]0 is the
initial concentration of 1X.
[21] A. G. Blackman, W. B. Tolman in Metal-Oxoand Metal-PeroxoSpe-
cies in Catalytic Oxidations (Ed.: B. Meunier), Springer, Berlin,
2000, pp. 179 211.
[22] L. Que, Jr., W. B. Tolman, Angew. Chem. 2002, 114, 1160 1185;
Angew. Chem. Int. Ed. 2002, 41, 1114 1137.
[23] M.-A. Kopf, K. D. Karlin in Biomimetic Oxidations Catalyzed by
Transition Metal Complexes (Ed.: B. Meunier), Imperial College
Press, London, 1999, pp. 309 362.
[24] V. Mahadevan, R. J. M. K. Gebbink, T. D. P. Stack, Curr. Opin.
Chem. Biol. 2000, 4, 228 234.
[25] S. Itoh, S. Fukuzumi, Bull. Chem. Soc. Jpn. 2002, 75, 2081 2095.
[26] E. I. Solomon, U. M. Sundaram, T. E. Machonkin, Chem. Rev. 1996,
96, 2563 2605.
[27] S. Schindler, Eur. J. Inorg. Chem. 2000, 2311 2326.
[28] W. B. Tolman, Acc. Chem. Res. 1997, 30, 227 237.
[29] A. P. Cole, D. E. Root, P. Mukherjee, E. I. Solomon, T. D. P. Stack,
Science 1996, 273, 1848 1850.
Theoretical calculations: Density functional calculations were performed
on a COMPAQ DS20E computer using the Amsterdam Density Func-
tional (ADF) program version 1999.02 developed by Baerends et al.[45]
The electronic configurations of the molecular systems were described by
an uncontracted triple-z Slater-type orbital basis set (ADF basis set V)
with a single polarization function for each atom. Core orbitals were
frozen through 1s (C, O, N), 2p (Cl), and 3p (Cu). The calculations were
performed by using the local exchange-correlation potential by Vosko
et al.[46] and the nonlocal gradient corrections by Becke[47] and Perdew[48]
during the geometry optimizations. First-order scalar relativistic correla-
tions were added to the total energy. Final geometries and energetics
were optimized by using the algorithm of Versluis and Ziegler[49] provid-
ed in the ADF package and were considered converged when the
changes in bond lengths between subsequent iterations fell below 0.01 ä.
[30] V. Mahadevan, M. J. Henson, E. I. Solomon, T. D. P. Stack, J. Am.
Chem. Soc. 2000, 122, 10249 10250.
[31] M. Taki, S. Teramae, S. Nagatomo, Y. Tachi, T. Kitagawa, S. Itoh, S.
Fukuzumi, J. Am. Chem. Soc. 2002, 124, 6367 6377.
[32] D. J. E. Spencer, N. W. Aboelella, A. M. Reynolds, P. L. Holland,
W. B. Tolman, J. Am. Chem. Soc. 2002, 124, 2108 2109.
[33] N. W. Aboelella, E. A. Lewis, A. M. Anne M. Reynolds, W. W. Bren-
nessel, C. J. Christopher J. Cramer, W. B. Tolman, J. Am. Chem. Soc.
2002, 124, 10660 10661.
[34] H. Hayashi, S. Fujinami, S. Nagatomo, S. Ogo, M. Suzuki, A.
Uehara, Y. Watanabe, T. Kitagawa, J. Am. Chem. Soc. 2000, 122,
2124 2125.
Acknowledgment
This work was financially supported in part by Grants-in-Aid for Scientif-
ic Research (No. 15350105) from the Ministry of Education, Culture,
Sports, Science and Technology, Japan and by Research Fellowships of
Japan Society for the Promotion of Science for Young Scientists.
[35] H.-C. Liang, C. X. Zhang, M. J. Henson, R. D. Sommer, K. R. Hat-
well, S. Kaderli, A. D. Zuberb¸hler, A. L. Rheingold, E. I. Solomon,
K. D. Karlin, J. Am. Chem. Soc. 2002, 124, 4170 4171.
245
Chem. Eur. J. 2004, 10, 237 246
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

S. Fukuzumi, S. Itoh et al.
FULL PAPER
[36] L. M. Mirica, M. Vance, D. J. Rudd, B. Hedman, K. O. Hodgson,
E. I. Solomon, T. D. P. Stack, J. Am. Chem. Soc. 2002, 124, 9332
9333.
[37] A copper(i) arene interaction was proposed to influence the cop-
per(i) dioxygen reactivity in some instances (K. D. Karlin, M. S.
Nasir, B. I. Cohen, R. W. Cruse, S. Kaderli, A. D. Zuberb¸hler, J.
Am. Chem. Soc. 1994, 116, 1324 1336 and ref. [39]). However, it
has never been proved.
[38] The CuꢀC2 distances of the copper(i) complexes are comparable to
the CuꢀC distances in known CuI-h2-benzene complexes (2.09
2.30 ä)[11, 12,15] and in the recently reported CuI-h2-phenyl (2.07
2.30 ä),[16] CuI-h2-naphthyl (2.13, 2.41 ä),[17] and CuI-h2-indolyl
(2.23, 2.27 ä)[18] complexes.
[39] DFT studies on a copper(i) benzene system with various basis sets
have been reported: T. K. Dargel, R. H. Hertwig, W. Koch, Mol.
Phys. 1999, 96, 583 591.
[41] S. Itoh, H. Nakao, L. M. Berreau, T. Kondo, M. Komatsu, S. Fukuzu-
mi, J. Am. Chem. Soc. 1998, 120, 2890 2899.
[42] N. Kitajima, K. Fujisawa, Y. Moro-oka, J. Am. Chem. Soc. 1989,
111, 8975 8976.
[43] D. D. Perrin, W. L. F. Armarego, D. R. Perrin, Purification of Labo-
ratory Chemicals, 4th ed., Pergamon Press, Elmsford, New York,
1996.
[44] T. Osako, S. Nagatomo, Y. Tachi, T. Kitagawa, S. Itoh, Angew.
Chem. 2002, 114, 4501 4504; Angew. Chem. Int. Ed. 2002, 41,
4325 4328.
[45] a) E. J. Baerends, D. E. Ellis, P. Ros, Chem. Phys. 1973, 2, 41 51;
b) B. te Velde, E. J. Baerends, J. Comput. Phys. 1992, 99, 84 98.
[46] S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200 1211.
[47] A. D. Becke, Phys. Rev. A 1988, 38, 3098 3100.
[48] a) J. P. Perdew, Phys. Rev. B 1986, 33, 8822 8824; b) J. P. Perdew,
Phys. Rev. B 1986, 34, 7406.
[40] There is no copper arene interaction in the copper(ii) complex of
1H: S. Itoh, T. Kondo, M. Komatsu, Y. Ohshiro, C. Li, N. Kanehisa,
Y. Kai, S. Fukuzumi, J. Am. Chem. Soc. 1995, 117, 4714 4715.
[49] L. Versluis, T. Ziegler, J. Chem. Phys. 1988, 88, 322 328.
Received: June 23, 2003 [F5263]
246
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 237 246