Reactions of copper(II)-H2O2 adducts supported by tridentate bis(2-pyridylmethyl)amine ligands: Sensitivity to solvent and variations in ligand substitution

Article abstract of DOI:10.1021/ic800845h

The copper(II) complexes 1^H and 1^Ar(X), supported by the N,N-di(2-pyridylmethyl)benzylamine tridentate ligand (L^H) or its derivatives having m-substituted phenyl group at the 6-position of pyridine donor groups (L^Ar(X)), have been prepared, and their reactivity toward H₂O₂ has been examined in detail at low temperature. Both copper(II) complexes exhibited a novel reactivity in acetone, giving 2-hydroxy-2-hydroperoxypropane (HHPP) adducts 2^H and 2 ^Ar(X), respectively. From 2^Ar(X), an efficient aromatic ligand hydroxylation took place to give phenolate-copper(II) complexes 4 ^Ar(X). Detailed spectroscopic and kinetic analyses have revealed that the reaction proceeds via an electrophilic aromatic substitution mechanism involving copper(II)-carbocation intermediates 3^Ar(X). Theoretical studies at the density functional theory (DFT) level have strongly implicated conjugate acid/base catalysis in the O-O bond cleavage and C-O bond formation steps that take the peroxo intermediate 2^Ar(X) to the carbocation intermediate 3^Ar(X). In contrast to the 2^Ar(X) cases, the HHPP-adduct 2^H reacted to give a copper(II)-acetate complex [Cu ^II(L^H)(OAc)](ClO₄) (6^H), in which one of the oxygen atoms of the acetate co-ligand originated from H ₂O₂. In this case, a mechanism involving a Baeyer-Villiger type 1,2-methyl shift from the HHPP-adduct and subsequent ester hydrolysis has been proposed on the basis of DFT calculations; conjugate acid/base catalysis is implicated in the 1,2-methyl shift process as well. In propionitrile, both 1^H and 1^Ar(X) afforded simple copper(II)-hydroperoxo complexes LCu^II-OOH in the reaction with H₂O₂, demonstrating the significant solvent effect on the reaction between copper(II) complexes and H₂O₂.

Full text of DOI:10.1021/ic800845h

Inorg. Chem. 2008, 47, 8222-8232
Reactions of Copper(II)-H2O2 Adducts Supported by Tridentate
Bis(2-pyridylmethyl)amine Ligands: Sensitivity to Solvent and Variations
in Ligand Substitution
Atsushi Kunishita,† Joseph D. Scanlon,‡ Hirohito Ishimaru,§ Kaoru Honda,| Takashi Ogura,§
Masatatsu Suzuki, Christopher J. Cramer,*,‡ and Shinobu Itoh*,†
Department of Chemistry, Graduate School of Science, Osaka City UniVersity, 3-3-138 Sugimoto,
Sumiyoshi-ku, Osaka 558-8585, Japan, Department of Chemistry and Research Computing Center,
UniVersity of Minnesota, 207 Pleasant Street SE, Minneapolis, Minnesota 55455, Graduate School
of Life Science, UniVersity of Hyogo, 3-2-1 Kouto, Kamigori-cho, Ako-gun, Hyogo 678-1297,
Japan, and DiVision of Material Sciences, Graduate School of Natural Science and Technology,
Kanazawa UniVersity, Kakuma-machi, Kanazawa, 920-1192, Japan
Received May 9, 2008
The copper(II) complexes 1H and 1Ar(X), supported by the N,N-di(2-pyridylmethyl)benzylamine tridentate ligand (LH)
or its derivatives having m-substituted phenyl group at the 6-position of pyridine donor groups (LAr(X)), have been
prepared, and their reactivity toward H2O2 has been examined in detail at low temperature. Both copper(II) complexes
exhibited a novel reactivity in acetone, giving 2-hydroxy-2-hydroperoxypropane (HHPP) adducts 2H and 2Ar(X)
respectively. From 2Ar(X), an efficient aromatic ligand hydroxylation took place to give phenolate-copper(II) complexes
Ar(X). Detailed spectroscopic and kinetic analyses have revealed that the reaction proceeds via an electrophilic
,
4
aromatic substitution mechanism involving copper(II)-carbocation intermediates 3Ar(X). Theoretical studies at the
density functional theory (DFT) level have strongly implicated conjugate acid/base catalysis in the O-O bond
cleavage and C-O bond formation steps that take the peroxo intermediate 2Ar(X) to the carbocation intermediate
3
Ar(X). In contrast to the 2Ar(X) cases, the HHPP-adduct 2H reacted to give a copper(II)-acetate complex
[CuII(LH)(OAc)](ClO4) (6H), in which one of the oxygen atoms of the acetate co-ligand originated from H2O2. In this
case, a mechanism involving a Baeyer-Villiger type 1,2-methyl shift from the HHPP-adduct and subsequent ester
hydrolysis has been proposed on the basis of DFT calculations; conjugate acid/base catalysis is implicated in the
1,2-methyl shift process as well. In propionitrile, both 1H and 1Ar(X) afforded simple copper(II)-hydroperoxo complexes
LCuII-OOH in the reaction with H2O2, demonstrating the significant solvent effect on the reaction between copper(II)
complexes and H2O2.
Introduction
copper catalyzed oxidation reactions.1-5 In many cases, the
reaction yields a mononuclear copper(II)-hydroperoxo com-
plex LCuII-OOH (A in Chart 1), exhibiting an intense
absorption band around 325-387 nm assigned to a
HOO-fCuII ligand-to-metal charge transfer (LMCT) transi-
The reactions of copper(II) complexes and hydrogen
peroxide (H2O2) have been studied extensively to gain insight
into the structure, physicochemical properties, and reactivity
of the various reactive intermediates involved in copper
monooxygenases and copper oxidases, as well as numerous
(1) (a) Wada, A.; Harata, M.; Hasegawa, K.; Jitsukawa, K.; Masuda, H.;
Mukai, M.; Kitagawa, T.; Einaga, H. Angew. Chem., Int. Ed. 1998,
37, 798–799. (b) Yamaguchi, S.; Wada, A.; Nagatomo, S.; Kitagawa,
T.; Jitsukawa, K.; Masuda, H. Chem. Lett. 2004, 1556–1557. (c)
Mizuno, M.; Honda, K.; Cho, J.; Furutachi, H.; Tosha, T.; Matsumoto,
T.; Fujinami, S.; Kitagawa, T.; Suzuki, M. Angew. Chem., Int. Ed.
2006, 45, 6911–6914. (d) Maiti, D.; Sarjeant, A. A. N.; Karlin, K. D.
J. Am. Chem. Soc. 2007, 129, 6720–6721. (e) Maiti, D.; Lucas, H. R.;
Sarjeant, A. A. N.; Karlin, K. D. J. Am. Chem. Soc. 2007, 129, 6998–
6999.
* To whom correspondence should be addressed. E-mail: shinobu@
sci.osaka-cu.ac.jp (S.I.), cramer@umn.edu (C.J.C.).
†
Osaka City University.
University of Minnesota.
University of Hyogo.
Kanazawa University.
‡
§
|
8222 Inorganic Chemistry, Vol. 47, No. 18, 2008
10.1021/ic800845h CCC: $40.75  2008 American Chemical Society
Published on Web 08/13/2008

Reactions of Copper(II)-H2O2 Adducts
Chart 1
although compounds C and D are more typically generated
by the reaction of copper(I) complexes with molecular
oxygen.7,8
For the preparation of mononuclear copper(II)-hydroper-
oxo complex A, tetradentate tripodal ligands such as tris(2-
pyridylmethyl)amine (TPA) and its derivatives have been
frequently employed. In this ligand system, metastable end-
on hydroperoxo complexes with a trigonal bipyramidal
geometry are obtained, since such tetradentate ligands afford
only one vacant coordination site for the external ligands.1
The most prominent example is the first X-ray structure
determination of LCuII-OOH (A) by Masuda and co-workers
using a TPA derivative involving two 6-pivalamide-2-
pyridylmethyl groups.1a In this complex, the end-on hydro-
peroxo complex is further stabilized by an intramolecular
hydrogen bonding interaction between the pivalamide NH
groups and the proximal oxygen atom of the -OOH group.1a
Likely because of this, Masuda’s hydroperoxo complex
exhibits little redox reactivity.9
tion and a resonance Raman band at 822-900 cm-1
attributable to the O-O bond stretching vibration of the
-OOH group.1-3 The mononuclear copper(II)-hydroperoxo
complex A has been suggested as one of the important
intermediates involved in the copper monooxygenases such
as peptidylglycine R-amidating monooxygense (PAM) and
dopamine ꢀ-monooxygenase (DꢀM), as well as in copper
oxidases such as copper-containing amine oxidases (AO) and
galactose oxidase (GAO).6
With respect to the oxygenation reactivity of cop-
per(II)-H2O2 adducts, Karlin and co-workers recently re-
ported oxidative N-dealkylation and aromatic hydroxylation
reactions when supported copper(II) complexes were treated
with H2O2 in acetone in the presence of triethylamine as a
base.1d,e In these cases, a dimethylamino group (-NMe2)
or a p-tert-butylphenyl group (-C6H4-p-tBu), respectively,
was attached at the 6-position of one of the pyridine rings
of TPA.1d,e The authors suggested that a LCuII-OOH
intermediate (A) was the reactive species in both reactions.1d,e
By contrast, Suzuki and co-workers reported an auto-
oxidation type ligand modification (formation of a ligand-
based alkylperoxo-copper(II) complex) from the reaction of
H2O2 and a copper(I) complex supported by bis(6-methyl-
2-pyridylmethyl)(2-pyridylmethyl)amine (Me2-TPA) in
acetonitrile.1c They suggested that a Cu(III)dO type species
(also formulated as Cu(II)-O•-) was a possible active
intermediate for the hydrogen atom abstraction from the
6-methyl group of ligand that was inferred as the initial step
of the auto-oxidation type chain reaction.1c
Recently, we have found that copper(II) complexes sup-
ported by tridentate bis(2-pyridylmethyl)amine ligands con-
taining a series of m-substituted phenyl groups at the
6-position of pyridines (LAr(X), X ) OMe, Me, H, Cl, and
NO2, see Chart 2), when treated with H2O2 in acetone in the
presence of triethylamine (i.e., nearly the same experimental
conditions as those for Karlin’s reactions mentioned above),
gave a new type of CuII-H2O2 adduct E (Scheme 1).10 In
our case, an acetone molecule is incorporated into the
complex to make a 2-hydroxy-2-hydroperoxypropane (HHPP)
adduct, which then undergoes an efficient aromatic ligand
In some cases, the reactions of copper(II) complexes with
H2O2 have given dinuclear copper(II) complexes B (R )
Ar or H) containing a µ-η1:η1-hydroperoxo bridge; these
exhibit similar spectroscopic features to those of complex
A.4 The (µ-η2:η2-peroxo)dicopper(II) and bis(µ-oxo)dicop-
per(III) complexes, C and D, have also been generated in
rare cases.5 The physicochemical properties and reactivity
of such dinuclear copper complexes have been studied
extensively to shed light on the mechanistic details of copper
proteins (enzymes) incorporating a dinuclear copper moiety
in their active sites such as hemocyanin (dioxygen carrier
protein), catechol oxidase, and tyrosinase (monooxygenase),
(2) (a) Ohtsu, H.; Itoh, S.; Nagatomo, S.; Kitagawa, T.; Ogo, S.; Watanabe,
Y.; Fukuzumi, S. Chem. Commun. 2000, 1051–1052. (b) Kodera, M.;
Kita, T.; Miura, I.; Nakayama, N.; Kawata, T.; Kano, K.; Hirota, S.
J. Am. Chem. Soc. 2001, 123, 7715–7716. (c) Ohtsu, H.; Itoh, S.;
Nagatomo, S.; Kitagawa, T.; Ogo, S.; Watanabe, Y.; Fukuzumi, S.
Inorg. Chem. 2001, 40, 3200–3207. (d) Fujii, T.; Naito, A.; Yamagu-
chi, S.; Wada, A.; Funahashi, Y.; Jitsukawa, K.; Nagatomo, S.;
Kitagawa, T.; Masuda, H. Chem. Commun. 2003, 2700–2701. (e)
Yamaguchi, S.; Nagatomo, S.; Kitagawa, T.; Funahashi, Y.; Ozawa,
T.; Jitsukawa, K.; Masuda, H. Inorg. Chem. 2003, 42, 6968–6970. (f)
Osako, T.; Nagatomo, S.; Kitagawa, T.; Cramer, C. J.; Itoh, S. J. Biol.
Inorg. Chem. 2005, 10, 581–590. (g) Yamaguchi, S.; Kumagai, A.;
Nagatomo, S.; Kitagawa, T.; Funahashi, Y.; Ozawa, T.; Jitsukawa,
K.; Masuda, H. Bull. Chem. Soc. Jpn. 2005, 78, 116–124.
(3) (a) Capdevielle, P.; Maumy, M. Tetrahedron Lett. 1990, 31, 3891–
3892. (b) Chen, P.; Fujisawa, K.; Solomon, E. I. J. Am. Chem. Soc.
2000, 122, 10177–10193. (c) Ohta, T.; Tachiyama, T.; Yoshizawa,
K.; Yamabe, T.; Uchida, T.; Kitagawa, T. Inorg. Chem. 2000, 39,
4358–4369. (d) Cheruzel, L. E.; Cecil, M. R.; Edison, S. E.; Mashuta,
M. S.; Baldwin, M. J.; Buchanan, R. M. Inorg. Chem. 2006, 45, 3191–
3202.
(4) (a) Karlin, K. D.; Ghosh, P.; Cruse, R. W.; Farooq, A.; Gultneh, Y.;
Jacobson, R. R.; Blackburn, N. J.; Strange, R. W.; Zubieta, J. J. Am.
Chem. Soc. 1988, 110, 6769–6780. (b) Kodera, M.; Tachi, Y.; Hirota,
S.; Katayama, K.; Shimakoshi, H.; Kano, K.; Fujisawa, K.; Moro-
oka, Y.; Naruta, Y.; Kitagawa, T. Chem. Lett. 1998, 389–390. (c)
Battaini, G.; Monzani, E.; Perotti, A.; Para, C.; Casella, L.; Santa-
gostini, L.; Gullotti, M.; Dillinger, R.; Na¨ther, C.; Tuczek, F. J. Am.
Chem. Soc. 2003, 125, 4185–4198. (d) Itoh, K.; Hayashi, H.; Furutachi,
H.; Matsumoto, T.; Nagatomo, S.; Tosha, T.; Terada, S.; Fujinami,
S.; Suzuki, M.; Kitagawa, T. J. Am. Chem. Soc. 2005, 127, 5212–
5223.
(7) Itoh, S. In ComprehensiVe Coordination Chemistry II; Que, L., Jr.,
Tolman, W. B., Eds.; Elsevier: Amsterdam, 2004; Vol. 8, pp 369-393.
(8) (a) Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Chem. ReV. 2004,
104, 1013–1045. (b) Lewis, E. A.; Tolman, W. B. Chem. ReV. 2004,
104, 1047–1076.
(9) Masuda and coworkers have also demonstrated that the substituents
at the 6-position of the pyridine nucleus of TPA largely affect the
stability and reactivity of the copper(II)-hydroperoxo complexes A1b.
(10) Kunishita, A.; Teraoka, J.; Scanlon, J. D.; Matsumoto, T.; Suzuki,
M.; Cramer, C. J.; Itoh, S. J. Am. Chem. Soc. 2007, 129, 7248–7249.
(5) (a) Osako, T.; Nagatomo, S.; Tachi, Y.; Kitagawa, T.; Itoh, S. Angew.
Chem., Int. Ed. 2002, 41, 4325–4328. (b) Shimokawa, C.; Teraoka,
J.; Tachi, Y.; Itoh, S. J. Inorg. Biochem. 2006, 100, 1118–1127.
(6) Itoh, S. Curr. Opin. Chem. Biol. 2006, 10, 115–122.
Inorganic Chemistry, Vol. 47, No. 18, 2008 8223

Kunishita et al.
[CuII(LH)(OAc)](ClO4) (6H). An acetone solution (100 mL) of
1H (126.6 mg, 2.0 mM) was cooled down to -80 °C using a dry
ice-AcOEt bath. Then, 30% H2O2 aqueous solution (10 equiv) and
Et3N (2 equiv) were added to the solution. The resulting mixture
was stirred for 2 h at -80 °C and then gradually warmed up to
room temperature. After stirring for additional 30 min at room
temperature, the solvent was reduced under reduced pressure to
give a blue residue, to which ether (100 mL) was added. Standing
the mixture for several minutes in a glovebox resulted in the
precipitation of a blue powder. The supernatant was then removed
by decantation, and the remaining green-brown solid was washed
with ether three times and dried to give 6H in an 85% yield. FT-IR
(KBr) 1605, 1561 (OAc-), 1088, and 625 cm-1 (ClO4-); HRMS
(FAB, pos) m/z ) 411.1008 (M+); calcd for C21H22CuN3O2
411.1008; ESI-MS m/z ) 411.12 (C21H22CuN3O2); Anal. Calcd for
[CuII(LH)(OAc)](ClO4) (6H) (C21H29ClCuN3O9.5): C, 43.91; H, 5.09;
N, 7.31. Found: C, 43.71; H, 4.89; N, 7.77. Isotope labeling
experiments were performed using H218O2 instead of H216O2 [ESI-
MS m/z ) 413.09 (C21H22CuN316O18O)] and using acetone-d6
instead of acetone-h6 [ESI-MS m/z ) 414.19 (C21H19D3CuN3O2)]
by the same method as described above.
Physical Methods. FT-IR spectra were recorded on a Jasco
FTIR-4100, and UV-visible spectra were taken on a Jasco V-570
or a Hewlett-Packard 8453 photo diode array spectrophotometer.
1H NMR spectra were recorded on a JEOL FT-NMR Lambda
300WB or a JEOL FT-NMR GX-400 spectrometer. Electron spin
resonance (ESR) spectra were recorded on BRUKER E-500
spectrometer at -150 °C and were simulated by using Bruker
SimFonia (ver. 1.25) on a HP personal computer (Windows XP).
Mass spectra were recorded on a JEOL JMS-700T Tandem MS-
station mass spectrometer. ESI-MS (electrospray ionization mass
spectra) measurements were performed on a PE SCIEX API 150EX
or a Micromass LCT spectrometer. Elemental analyses were
recorded with a Perkin-Elmer or a Fisons instruments EA1108
Elemental Analyzer.
Resonance Raman scattering was excited at 406.7 nm from Kr+
laser (Spectra Physics, BeamLok 2060). Resonance Raman scat-
tering was dispersed by a single polychromator (Ritsu Oyo Kogaku,
MC-100) and was detected by a liquid nitrogen cooled CCD
detector (Roper Scientific, LNCCD-1100-PB). The resonance
Raman measurements were carried out using a rotated cylindrical
cell thermostatted from -70 to -90 °C by flashing cold liquid
nitrogen gas.
Cyclic voltammetric measurements were performed on an ALS-
630A electrochemical analyzer in deaerated CH3CN containing 0.10
M NBu4ClO4 as a supporting electrolyte. A Pt working electrode
(BAS) was polished with BAS polishing alumina suspension and
rinsed with acetone before use. The counter electrode was a
platinum wire. The measured potentials were recorded with respect
to a Ag/AgNO3 (0.01 M) reference electrode. All electrochemical
measurements were carried out in a glovebox filled with Ar gas at
25 °C.
Chart 2
Scheme 1
hydroxylation reaction via an electrophilic aromatic substitu-
tion mechanism.10 All these results suggest that more detailed
studies of ligand effects, as well as solvent effects, on the
reactivity of copper complexes toward H2O2 are necessary.
In this work, we have examined in more detail the reaction
of copper(II) complexes and H2O2 by using the bis(2-
pyridylmethyl)amine tridentate ligands LH and LAr(X) shown
in Chart 2 in both acetone and propionitrile. The results
demonstrate that the structure and reactivity of the active-
oxygen species generated are significantly affected by both
the supporting ligand and the solvent. Theoretical studies at
the density functional theory (DFT) level have been em-
ployed to provide important insights into the structures and
reactivity of the active oxygen species.
Experimental Section
Syntheses. The reagents and the solvents used in this study,
except the ligands and the copper complexes, were commercial
products of the highest available purity and were further purified
by standard methods, as necessary.11 The copper(II) complexes of
ligand LAr(X), 1Ar(X), were obtained as reported previously.10 Ligand
LH was synthesized according the reported procedures.12
Caution! The perchlorate salts in this study are all potentially
explosiVe and should be handled with care.
[CuII(LH)(CH3CN)2](ClO4)2 (1H). CuII(ClO4)2 ·6H2O (148 mg,
0.40 mmol) was added to an acetonitrile solution (20 mL) of ligand
LH (116 mg, 0.40 mmol). After stirring for 5 min at room
temperature, insoluble material was removed by filtration. Addition
of ether (100 mL) to the filtrate gave a purple powder that was
precipitated by standing the mixture for several minutes. The
supernatant was then removed by decantation, and the remained
purple solid was washed with ether three times and dried to give
complex 1H in 88%. Micro crystals of 1H were obtained by vapor
diffusion of ether into an acetonitrile solution of the complex. FT-
IR (KBr) 1094 and 625 cm-1 (ClO4-); HRMS (FAB+) m/z )
451.0348, calcd for C19H19ClCuN3O4 ) 451.0360; Anal. Calcd for
[CuII(L3H)(CH3CN)2](ClO4)2(C23H25Cl2CuN5O8): C, 43.58; H, 3.97;
N, 11.05. Found: C, 43.74; H, 3.93; N, 11.12.
Crystallographic Studies. The single crystal was mounted on
a glass-fiber. X-ray diffraction data were collected by a Rigaku
RAXIS-RAPID imaging plate two-dimensional area detector using
graphite-monochromated Mo KR radiation (λ ) 0.71069 Å) to 2θmax
of 55°. All the crystallographic calculations were performed by
using the Crystal Structure software package of the Molecular
Structure Corporation [Crystal Structure: Crystal Structure Analysis
Package version 3.8.1, Molecular Structure Corp. and Rigaku Corp.
(2005)]. The structures were solved with SIR92 and refined with
CRYSTALS. All non-hydrogen atoms and hydrogen atoms were
(11) Armarego, W. L. F.; Perrin, D. D. In Purification of Laborator
Chemicals, 4th ed.; Butterworth-Heinemann: Oxford, 1996; pp 176
and 215.
(12) Kawahara, S.; Uchimaru, T. Chem.sEur. J., 2001, 2437–2442.
8224 Inorganic Chemistry, Vol. 47, No. 18, 2008

Reactions of Copper(II)-H2O2 Adducts
refined anisotropically and isotropically, respectively. Atomic
coordinates, thermal parameters, and intramolecular bond distances
and angles are deposited in the Supporting Information (CIF file
format).
Kinetic Measurements. Kinetic measurements for the oxygen-
ation reaction of copper(II) complexes 1Ar(X) (X ) NO2, Cl, H,
Me, OMe) and 1H were performed using a Hewlett-Packard 8453
photo diode array spectrophotometer with a Unisoku thermostatted
cell holder designed for low temperature measurements (USP-203,
a desired temperature can be fixed within (0.5 °C) in acetone and
propionitrile at -70 °C and a multiscan double mixing stopped-
flow spectrophotometer designed for low-temperature measurements
(RSP-1000, Unisoku Co., Ltd.) in acetone at 0 °C. The pseudo-
first-order rate constants for the formation and decomposition
processes of the intermediate were determined from the plots of ln
∆A versus time based on the time course of the absorption change
at λmax due to the intermediate and/or the product.
Figure 1. ORTEP drawing of 1H showing 50% probability thermal
ellipsoids. The counteranion and the hydrogen atoms are omitted for clarity.
Computational Methods. All molecular structures were fully
optimized using the generalized gradient approximation (GGA)
density functional (mPWPW91) that combines the exchange
functional of Perdew,13 as modified by Adamo and Barone,14 with
the correlation functional of Perdew and Wang.15 For efficiency,
the benzyl group substituting the tertiary nitrogen of the experi-
mental ligands was truncated to a methyl group in the computational
models. In geometry optimizations, atomic orbital basis functions
were taken for Cu from the Stuttgart effective core potential and
basis set,16 for N and O from the 6-311G(d) basis,17 and for C
and H from the 6-31G(d) basis.17 Analytic vibrational frequencies
were computed for all stationary points to confirm their natures as
minima or transition-state (TS) structures and also to compute
thermal contributions to enthalpy and free energy using the ideal-
gas, rigid-rotator, harmonic-oscillator approximations.18 Single-point
calculations on gas-phase structures using the implicit polarized
continuum solvation model (PCM)19 were performed to quantify
the influence of solvation on the free energy. Single-point gas-phase
calculations were also performed using the M06L functional20 to
examine the sensitivity to the density functional in certain instances.
Table 1. Summary of the X-ray Crystallographic Data of Compound 1H
compound
1H
formula
C23H25N5CuCl2O8
633.93
formula weight
crystal system
space group
a, Å
b, Å
c, Å
monoclinic
C2/c(#15)
25.19(3)
10.717(11)
21.45(2)
90
99.5(5)
90
5709.5(113)
8
R, deg
ꢀ, deg
γ, deg
V, Å3
Z
F(000)
2600.00
1.475
153
Dcalcd, g/cm-3
T, K
crystal size, mm
µ(Mo KR), cm-1
2θmax, deg
no. of reflns measd.
no. of reflns obsd.
no. of variables
0.20 × 0.20 × 0.20
10.049
54.9
26036
4021([I > 1.00σ(I)])
374
0.0550
0.0778
Results and Discussion
a
R
b
Rw
Characterization of Copper(II) Starting Materials. The
copper(II) starting materials 1H and 1Ar(X) supported by LH
and LAr(X) (see Chart 2), respectively, were efficiently
prepared by mixing the ligand and CuII(ClO4)2 · 6H2O in
acetonitrile. The crystal structure of 1H is shown in Figure 1
together with the crystallographic data and the selected bond
lengths and angles presented in Tables 1 and 2, respectively.
Crystal structures of 1Ar(X) (X ) Me, H, Cl, and NO2) have
already been communicated.10
GOF
1.000
2
a R ) ∑||Fo| - |Fc||/∑|Fo|. b Rw ) [∑w(|Fo| - |Fc|)2/∑wFo ]1/2.
Table 2. Selected Bond Lengths (Å) and Angles (deg) of Compound
1H
Compound 1H
Cu(1)-N(1)
Cu(1)-N(3)
Cu(1)-N(5)
1.980(4)
2.035(4)
2.439(5)
Cu(1)-N(2)
Cu(1)-N(4)
1.978(4)
1.974(4)
N(1)-Cu(1)-N(2)
N(1)-Cu(1)-N(4)
N(2)-Cu(1)-N(4)
N(5)-Cu(1)-N(1)
N(5)-Cu(1)-N(3)
Copper(II) complex [CuII(LH)(CH3CN)2](ClO4)2 (1H) ex-
hibits a square pyramidal geometry (τ ) 0.07)21 with the
nitrogen atoms, N(1), N(2), and N(3) of LH and N(4) of one
of the acetonitrile co-ligands, occupying the basal plane, and
another acetonitrile nitrogen N(5) at the axial position.
165.25(18)
97.64(18)
96.91(18)
87.82(17)
102.91(15)
N(1)-Cu(1)-N(3)
N(2)-Cu(1)-N(3)
N(3)-Cu(1)-N(4)
N(5)-Cu(1)-N(2)
N(5)-Cu(1)-N(4)
82.76(16)
82.50(17)
169.26(18)
95.12(17)
87.82(17)
(13) Perdew, J. P. In Electronic Structure of Solids ′91; Ziesche, P., Eschrig,
H., Eds.; Akademie Verlag: Berlin, 1991, pp 11-20.
(14) Adamo, C.; Barone, V. J. Chem. Phys. 1998, 108, 664–675.
(15) Perdew, J.; Wang, Y. Phys. ReV. B 1992, 45, 13244–13249.
(16) Dolg, M.; Wedig, U.; Stoll, H.; Preuss, H. J. Chem. Phys. 1987, 86,
866–872.
(17) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
(18) Cramer, C. J. Essentials of Computational Chemistry: Theories and
Models, 2nd ed.; John Wiley & Sons: Chichester, 2004.
(19) (a) Tomasi, J.; Mennucci, B.; Cances, E. J. Mol. Struct. (Theochem)
1999, 464, 1211–226. (b) Tomasi, J.; Mennucci, B.; Cammi, R. Chem.
ReV. 2005, 105, 2999–3093.
(21) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor, G. C.
(20) Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.
J. Chem. Soc., Dalton Trans. 1984, 1349–1356.
Inorganic Chemistry, Vol. 47, No. 18, 2008 8225

Kunishita et al.
pyramidal geometry in the solution.23 Thus, it can be said that
the structures of 1Ar(Me), 1Ar(Cl), and 1Ar(NO2) in the solution are
similar to those in the crystals (τ ) 0.45-0.56), whereas the
square pyramidal structure of 1Ar(H) found in the crystal (τ )
0.08) may be significantly distorted toward trigonal bipyramidal
in the solution.
Table 3. Redox Potential, UV-vis Data (d-d Band), and ESR
Parameters of 1H and 1Ar(X)
E1/2
∆Ep
λmax (ε,
complex
(V)a
(V)b
M-1 cm-1
)
ESR parametersc
1H
-0.14 0.10 645 nm (150) g1 ) 2.260 A1 ) 170 G
g2 ) 2.065 A2 ) 25 G
g3 ) 2.047 A3 ) 18 G
1Ar(OMe)
1Ar(Me)
1Ar(H)
0.18 0.10 715 nm (150) g1 ) 2.315 A1 ) 139 G
845 nm (120) g2 ) 2.100 A2 ) 20 G
g3 ) 2.040 A3 ) 24 G
The UV-vis spectra of the copper(II) complexes presented
in Table 3 support the notion of their solution structures.
Namely, complex 1H exhibits a d-d band at 645 nm, which is
a typical one for copper(II) complexes with square pyramidal
geometry, whereas 1Ar(X) complexes show two d-d bands at
∼710 nm and ∼860 nm, which are characteristics for the
distortion of copper(II) toward trigonal bipyramidal geometry.23
All copper(II) complexes gave a reversible redox couple (∆Ep
∼ 100 mV) in CH3CN (Table 3). The cyclic voltammograms
are presented in Supporting Information, Figures S7-S12.
Apparently, the redox potential of 1Ar(X) is more than 300 mV
higher than that of 1H. This can be attributed to the weaker
electron-donor ability of LAr(X) as discussed above. With respect
to the order of E1/2 in the series of 1Ar(X), the electron-
withdrawing m-substituents such as NO2 and Cl further weaken
the electron-donor ability of pyridine as compared with the
electron-donating m-substituents such as Me and OMe.
Reaction of Copper(II) Complexes and H2O2 in
Acetone (CH3C(O)CH3). As has already been reported,10
1Ar(X) readily reacts with H2O2 in a 1:1 ratio in acetone at
-70 °C in the presence of triethylamine (1 equiv) to produce
2-hydroxy-2-hydroperoxypropane (HHPP) adducts 2Ar(X) (E
in Scheme 1), which exhibit a relatively intense LMCT band
at ∼420 nm (ε ) ∼1000 M-1 cm-1) together with a weak
d-d band at ∼650 nm (ε ) ∼200 M-1 cm-1) (eq 1).10
0.19 0.10 717 nm (170) g1 ) 2.315 A1 ) 139 G
860 nm (150) g2 ) 2.103 A2 ) 20 G
g3 ) 2.045 A3 ) 24 G
0.20 0.10 717 nm (135) g1 ) 2.313 A1 ) 138 G
855 nm (90)
g2 ) 2.100 A2 ) 19 G
g3 ) 2.050 A3 ) 25 G
1Ar(Cl)
0.24 0.12 720 nm (150) g1 ) 2.315 A1 ) 138 G
865 nm (130) g2 ) 2.100 A2 ) 23 G
g3 ) 2.040 A3 ) 24 G
0.30 0.10 710 nm (100) g1 ) 2.310 A1 ) 140 G
860 nm (100) g2 ) 2.100 A2 ) 22 G
g3 ) 2.030 A3 ) 24 G
1Ar(NO2)
a In CH3CN containing 0.1 M tetrabutylammonium perchlorate (TBAP)
at 25 °C; working electrode Pt, counter electrode Pt, reference electrode
Ag/0.01 M AgNO3, sweep rate 50 mV s-1 b Peak separation. c In CH3CN
.
at 123 K, microwave frequency 9.40-9.41 GHz, modulation frequency 100
kHz, modulation amplitude 5 G, microwave power 0.20-0.33 mW.
[CuII(LAr(H))(CH3CN)(ClO4)](ClO4) (1Ar(H)) showed a similar
square pyramidal structure (τ ) 0.08) with an acetonitrile
molecule occupying the equatorial position and ClO4- as the
axial co-ligand.10 It should be noted that the distances
between Cu(1) and the pyridine nitrogen atoms N(1) and
N(2) in 1H, 1.980 and 1.978 Å (see Table 2), are much shorter
than those in 1Ar(H), 2,031 and 2.010 Å, whereas the
Cu(1)-N(3) and Cu(1)-N(4) distances of 1H, 2.035 and
1.974 Å, are longer than those of 1Ar(H), 2,006 and 1.950
Å.10 Similar structural features [elongation of Cu(1)-N(1)py
and Cu(1)-N(2)py distances and shortening of Cu(1)-
N(3)amine and Cu(1)-N(4)solv distances] are observed in
1
Ar(Me), 1Ar(Cl), and 1Ar(NO2), although their structural distortion
from square pyramidal geometry becomes larger (τ )
0.45-0.56).10 The elongation of the Cu(1)-Npy distances
in 1Ar(X) as compared with those in 1H can be attributed to
the steric repulsion between the cupric ion and the aromatic
substituents at the 6-position of pyridine donor groups and/
or the repulsion between the aromatic substituents and the
co-ligands of 1Ar(X)
substituents makes the electron-donor ability of pyridine in
Ar(X) weaker than that in 1H, which is reflected in the redox
.
22 Such the steric effects of the aromatic
1
potentials described below.
The ESR data of the copper(II) starting materials are
summarized in Table 3. The spectra themselves are presented
in Supporting Information, Figures S1-S6. The A1 value (170
G) of complex 1H is a normal one for copper(II) complexes
with square pyramidal geometry.23 Thus, the square pyramidal
geometry (τ ) 0.07) of 1H in the crystal may be maintained in
the solution (frozen CH3CN). On the other hand, all 1Ar(X)
complexes exhibit relatively smaller A1 values (135-139 G),
suggesting that they have a significantly distorted square
Intermediate 2Ar(NO2) (X ) NO2), for example, gave isotope
sensitive resonance Raman bands at 855, 823, 792, and 545
cm-1 in perdeuterated acetone (acetone-d6), which shifted
to 825, 803, 785, and 525 cm-1 when H218O2 was employed
instead of H216O2.10 Appearance of the multiple Raman bands
in the 800 cm-1 region and their associated isotope shifts
(∆ν ) 30, 20, and 7 cm-1), as well as their peak intensity
(22) Nagao, H.; Komeda, N.; Mukaida, M.; Suzuki, M.; Tanaka, K. Inorg.
Chem. 1996, 35, 6809–6815.
(23) Mukherjee, R. In ComprehensiVe Coordination Chemistry II; Mc-
Cleverty, J. A., Meyer, T. J., Eds.; Elsevier: Amsterdam, 2004; Vol.
6, pp 747-910, and references cited therein.
8226 Inorganic Chemistry, Vol. 47, No. 18, 2008

Reactions of Copper(II)-H2O2 Adducts
Chart 3
of Figure 2, whereas Masuda’s hydroperoxo copper(II)
complex exhibits only a single Raman band at 834 cm-1
assigned to the 16O-16O stretching vibration.2d These results
strongly suggest the generation of an HHPP-adduct 2H (E)
in analogy to the 2Ar(X) examples (eq 2). Incorporation of an
acetone molecule into 2H was confirmed by noting that the
multiple Raman bands at 850, 833, and 820 cm-1 of 2H,
generated by the reaction with H218O2 in acetone-h6, were
shifted to 829, 814, and 793 cm-1 when the sample was
generated in acetone-d6 (Supporting Information, Figure
S13).27 The reason for the difference in reactivity between
our system and that of the Masuda’s system is not obvious
at present.
Figure 2. Spectral change for the reaction of 1H (0.2 mM) with H2O2 (2.0
mM) in the presence of NEt3 (2.0 mM) in acetone at -70 °C. Inset:
Resonance Raman spectra of 2H (5 mM) generated by using H216O2 (solid
line) and H218O2 (dashed line) obtained with λex ) 406.7 nm in acetone at
-80 °C; s denotes the solvent bands.
patterns, are very close to the resonance Raman data of the
copper(II)-alkylperoxo complexes LCuII-OOR [L ) hydro-
trispyrazolylborate and R ) tert-butyl (But) or cumyl (Cm)]
reported by Solomon and co-workers and also to those of
the iron(III)-alkylperoxo complex LFeIII-OOBut (L ) TPA
derivatives) reported by Que and co-workers.3b,24 By analogy
to those resonance Raman data of the related alkylperoxo
complexes,3b,24 the multiple Raman bands around 800 cm-1
of 2Ar(NO2) can be assigned as mixed O-O/C-O/C-C
vibrations. Then, the Raman band at 545 cm-1 can be
attributed to the Cu-O stretching vibration. ESR and ESI-
MS studies unambiguously support the formation of
10
2Ar(NO2)
.
A similar HHPP-adduct of iron(III) has been
reported by Que and co-workers in the reaction of FeII(TPA)
and H2O2 in acetone.25,26
The reaction of 1H without the 6-phenyl substituent and
H2O2 has also been examined under the same experimental
conditions. The spectral change for the reaction is shown in
Figure 2, where a relatively intense absorption band at 375
nm (ε ) 2500 M-1 cm-1) appears together with a weak d-d
band at 650 nm (220 M-1 cm-1). The spectrum is quite
different from that of Masuda’s copper(II)-hydroperoxo
complex LCuII-OOH (A) generated by using a similar bis(2-
pyridylmethyl)amine ligand LH(Bu) having a tert-butyl N-
substituent (see Chart 3) in acetone; λmax ) 350 nm (ε )
3400 M-1 cm-1), 564 (150), and 790 (55).2d
The resonance Raman features of our intermediate are also
different from those of Masuda’s LCuII-OOH complex of
LH(Bu). Namely, our intermediate showed multiple Raman
bands at 879, 864, and 847 cm-1, which shifted to 850, 833,
and 819 cm-1 upon 18O-substitution as shown in the inset
The ESR spectrum of 2H shown in Figure 3, which is
different from that of the starting material 1H (see Table 3),
indicates that intermediate 2H also has a tetragonal geom-
etry.23 Moreover, double integration of the ESR spectrum
of 2H indicated that 98% of the spin remained, confirming
10
the mononuclearity of 2H as in the case of 2Ar(X)
.
Reactivity of HHPP-Adducts 2. The HHPP-adducts 2Ar(X)
have previously been shown to undergo an aromatic ligand
hydroxylation reaction, giving phenolate-copper(II) product
complexes 4Ar(X) (eq 3).10 Isotope labeling experiments using
H218O2 unambiguously indicated that the origin of the
phenolate oxygen in the products was hydrogen peroxide,
and X-ray analysis on a single crystal of 4Ar(X) confirmed
the position of the hydroxylation reaction (para to the
substituent X, see eq 3).10 Demetalation of 4Ar(X) by an
(24) (a) Zang, Y.; Elgren, T. E.; Dong, Y.; Que, L., Jr J. Am. Chem. Soc.
1993, 115, 811–813. (b) Zang, Y.; Kim, J.; Dong, Y.; Wilkinson, E. C.;
Appelman, E. H.; Que, L., Jr J. Am. Chem. Soc. 1997, 119, 4197–
4205.
(27) The low temperature ESI-MS measurement gave a set of peaks at
m/z ) 443.1 with H216O2 that shifts to 447.1 upon H218O2 substitution
(Supporting Information, Figure S14). The mass distribution patterns
about these peaks, as well as their isotope shifts, are fully consistent
with the HHPP-adduct structure of copper(II) complex 2H. However,
the mass data also fit the structures of suggested intermediates 5H and
5aH shown in Figure 7.
(25) Payeras, A. M.; Ho, R. Y. N.; Fujita, M.; Que, L., Jr Chem.sEur. J.
2004, 10, 4944>4953.
(26) The reversible addition of H2O2 to acetone to form 2-hydroxy-2-
hydroperoxypropane (HHPP) was first studied in detail by (a) Sauer,
M. C. V.; Edwards, J. O. J. Phys. Chem. 1971, 75, 3004–3011.
Inorganic Chemistry, Vol. 47, No. 18, 2008 8227

Kunishita et al.
isotope effect (KIE) (KIE ) 0.9 ( 0.02; kH ) 5.0 × 10-3
s-1 and kD ) 5.5 × 10-3 s-1) was obtained using the
perdeuterated ligand LH-d10 (replacing all protons of the
6-phenyl groups), suggesting that the rate-limiting step does
not involve proton migration from the phenyl substituent of
3Ar(X) but instead involves a conversion of sp2 aromatic
carbon to sp3 carbon.31 These kinetic results suggest that the
reaction of 2Ar(X) to 3Ar(X) involves an electrophilic attack
by the peroxo moiety to the phenyl substituent at the
6-position of pyridine, leading the C-O bond formation.
At the mPWPW91 level of theory, we located a TS structure
for an electrophilic aromatic substitution whereby 2Ar(H) reacts
to form the intermediate complex 3Ar(H) (Figure 4). The TS
structure has a forming C-O bond of 1.800 Å and a breaking
O-O bond of 1.504 Å. The interaction between the Cu atom
and the O atom being transferred remains strong, as evidenced
by a Cu-O bond length of 1.935 Å. The other two O atoms of
the HHPP fragment are 2.540 and 2.799 Å distant from the Cu
atom, suggesting only weak interactions with the metal. The
enthalpy of activation (including solvation free energy) is
computed to be 98.0 kJ/mol, which is in quite poor agreement
with the experimental enthalpy of activation of 24.9 kJ/mol (vide
supra). However, the product immediately deriving from the
Figure 3. (A) ESR spectrum of 2H and (B) the computer simulation
spectrum with the parameters g1 ) 2.240, g2 ) 2.052, g3 ) 2.060, A1 )
168.0 G, A2 ) 32.0 G, and A3 ) 3.0 G.
ordinary workup treatment using NH4OH(aq) gave the
hydroxylated ligands LAr(X)-OH in yields of 99% for X )
OMe, 98% for X ) Me, 85% for X ) H, 99% for X ) Cl,
and 53% for X ) NO2.10
aromatic hydroxylation is predicted to be the intermediate 3Ar(H)
,
which, relative to 2Ar(H), has shifted a proton from the original
hydroxyl group of the HHPP moiety to the now acetal oxygen
atom derived from breaking the O-O bond. This suggests that
the ring hydroxylation reaction should be catalyzed by general
acid or base catalysis to either remove the proton from the
original hydroxyl group or donate a proton to the distal oxygen
of the peroxy fragment (making the acetal a better leaving group
in what may be viewed as a nucleophilic attack by the aromatic
ring on the other proximal peroxy oxygen atom), or both as
illustrated in Scheme 2.
Indeed, when the same reaction coordinate is computed
with an ammonium ion (R ) H in Scheme 2) as a proton
donor to the distal oxygen of the peroxy fragment, the
enthalpy of activation is predicted to decrease to 56.6 kJ/
mol, and, when an ammonia molecule (R ) H in Scheme
2) is added to act as a base and accept the hydroxyl proton,
the activation enthalpy is reduced to 9.0 kJ/mol. In the TS
structure for this final case, the hydroxyl proton is not in
flight to the ammonia acceptor but there is a strong hydrogen
bond facilitating the reaction. In the experimental reaction
system, the conjugate acid/base pair is likely to be triethy-
lammonium/triethylamine (R ) Et in Scheme 2). The greater
steric bulk of these species may rationalize the greater
experimental enthalpy of activation (24.9 kJ/mol) compared
to that obtained by computation, but theory makes clear that
protonation of the peroxy species can significantly accelerate
the ring hydroxylation, in a fashion analogous to the
mechanism proposed for the formation of the active iron-
During the course of the reaction from 2Ar(X) to 4Ar(X)
,
another intermediate 3Ar(X) having a rather featureless
UV-vis spectrum was detected (eq 3).10 The conversion of
2Ar(X) to 3Ar(X) obeyed first-order kinetics, and from the
temperature dependence of the decay rate we obtained the
activation parameters ∆Hq ) 24.9 ( 1.2 kJ mol-1 and ∆Sq
) -162.9 ( 5.5 J K-1 mol-1.10 Furthermore, Hammett
analysis (plot of log kobs vs σ+) gave F ) -2.2 (r2 ) 0.99),
which is very close to other F values reported for aromatic
hydroxylation reactions of dicopper(II)-peroxo complexes
(-1.8 to -2.2).28-30 In addition, a negative deuterium kinetic
(30) Matsumoto, T.; Furutachi, H.; Kobino, M.; Tomii, M.; Nagatomo, S.;
Tosha, T.; Osako, T.; Fujinami, S.; Itoh, S.; Kitagawa, T.; Suzuki, M.
J. Am. Chem. Soc. 2006, 128, 3874–3875.
(28) Itoh, S.; Kumei, H.; Taki, M.; Nagatomo, S.; Kitagawa, T.; Fukuzumi,
S. J. Am. Chem. Soc. 2001, 123, 6708–6709.
(29) Palavicini, S.; Granata, A.; Monzani, E.; Casella, L. J. Am. Chem.
Soc. 2005, 127, 18031–18036.
(31) In our previous communication,10 KIE was reported as 1.0. However,
multiple re-examinations gave a negative KIE as 0.9 ( 0.02.
8228 Inorganic Chemistry, Vol. 47, No. 18, 2008

Reactions of Copper(II)-H2O2 Adducts
Figure 4. The mPWPW91 computed TS structure is shown at left with selected bond lengths (Å); the atoms of the 6-pyridyl phenyl ring that is not reacting
have been deleted for clarity. The product derived from the reaction, assigned as 3Ar(H), is shown at right. In the ball-and-stick structure, the Cu atom is
bronze, the O atoms red, the N atoms blue, the C atoms gray, and the H atoms white.
Scheme 2
oxo species in horseradish peroxidase, for instance.32 M06L
single point calculations on the relevant structures for the
uncatalyzed, protonated, and proton-shuttle facilitated reac-
tions predict enthalpies of activation (including solvation)
of 105.7, 59.1, and 8.7 kJ/mol, respectively, suggesting little
sensitivity to the functional for the modeling of ring
oxidation. Finally, the large, negative entropy of activation
observed experimentally (∆Sq ) -162.9 ( 5.5 J K-1 mol-1)
is consistent with our proposal of general acid/base catalysis.
When the ammonia/ammonium ion pair is included in the
computation of the aromatic hydroxylation, the product deriving
from the TS structure is observed to have undergone a
spontaneous 1,2-hydride shift from the oxygenated position to
the adjacent carbocation (see Scheme 2). Thus, the UV spectrum
observed for the 3Ar(X) species may be for cyclohexadienone
intermediates rather than carbocationic intermediates. In either
case, however, rearomatization of intermediate 3Ar(H) can be
accomplished by proton transfer to either amine base or the
dimethoxyacetal anion (whether intra- or intermolecularly),
The HHPP-adduct 2H reported here, in contrast to 2Ar(X)
,
cannot undergo aromatic hydroxylation. Instead it decom-
posed gradually at -70 °C (kdec ) 8.2 × 10-3 s-1) to give
the copper(II)-acetate complex [CuII(LH)(OAc)](ClO4) (6H).
Product analysis on a scale-up reaction (2.0 mM) gave 6H
in a 85% yield. Isotope labeling experiments using H218O2
demonstrated that one of the oxygen atoms of the acetate
ion originated from hydrogen peroxide as shown in the inset
of Figure 5. Formation of acetate from the HHPP-adduct of
FeIII(TPA) has also been reported by Que et al.25
In this case, mPWPW91 calculations suggest that a
plausible mechanism (Figure 6) involves a tautomerization
leading, upon hydrolysis of the acetal, to the final product 4Ar(H)
,
acetone, and water (Scheme 2). The net driving force for this
conversion is very large, with a computed exergonicity of about
-344 kJ/mol at 298 K. In support of this mechanism, KIE )
1.5 (kH ) 3.8 × 10-2 s-1 and kD ) 2.6 × 10-2 s-1) was
experimentally obtained in the transformation process from
3
Ar(H) to 4Ar(H), when perdeuterated ligand LH-d10 was employed
instead of LH.
Figure 5. Spectral change for the decomposition of 2H (0.2 mM) in acetone
at -70 °C. Inset: ESI-MS spectra of 6H generated (A) by using H216O2 and
(B) by using H218O2.
(32) Newmyer, S. L.; Ortiz de Montellano, P. R. J. Biol. Chem. 1995, 270,
19430.
Inorganic Chemistry, Vol. 47, No. 18, 2008 8229

Kunishita et al.
Figure 6. Reaction mechanism for the conversion of 2H to 6H. Optimized geometries for the two TS structures are shown with selected bond lengths (Å).
See Figure 4 for atom color codes.
of the HHPP ligand to protonate the hydroperoxy moiety
instead of the hydroxide. This tautomer taut-2H is predicted
to be higher in free energy (including solvation) by about
80.5 kJ/mol at 203 K. From this tautomer, a TS structure
for a Baeyer-Villiger type 1,2-methyl shift was found in
which the breaking of the O-O bond is advanced (1.775
Å) relative to the breaking C-C bond (1.618 Å), and the
incipient hydroxide interacts strongly with the copper atom,
with a Cu-O bond length of 1.967 Å. The free energy of
activation for this process was computed to be 11.3 kJ/mol
relative to the tautomerized intermediate, or 91.8 kJ/mol
relative to 2H. The interpretation of the experimentally
observed decomposition rate constant (kdec ) 8.2 × 10-3 s-1)
using TS theory for a unimolecular process suggests a free
energy of activation of 57.2 kJ/mol at 203 K, which is lower
than the value predicted from the mPWPW91 calculations.
As noted above for ring oxidation, however, this reaction
would likely be significantly facilitated by employing an
external base/conjugate acid pair to accomplish the proton
shuttling rather than enforcing tautomerization as an initial
step; we did not, however, search for such structures
specifically in this case. In addition, this reaction does show
some sensitivity to the functional employed: at the M06L
level the 203 K free energy of activation including solvation
is predicted to be 77.0 kJ/mol, which is considerably closer
to the experimental value (57.2 kJ/mol).
to generate a copper-coordinated hemiorthoacetate. Theory
predicts the energies of the TS and orthoester product
structures to be -143.0 and -181.2 kJ/mol relative to 2H
(the free energy of activation for the ester hydrolysis is thus
70.2 kJ/mol). In the TS structure for nucleophilic addition,
the breaking and making Cu-O bond lengths are similar,
2.088 and 2.128 Å, respectively. We did not attempt to
determine a precise pathway for breakdown of the orthoester
to coordinated acetate and methanol (many different alterna-
tives involving intra- and intermolecular proton transfer are
obviously possible and would be expected to occur with very
low barrier). We did compute the net exergonicity for the
entire reaction of 2H to 6H, and this is -295.5 kJ/mol at the
solvated mPWPW91 level.
Another potential pathway for decomposition was considered,
namely a 1,3-methyl shift to generate methoxide and acetic acid
ligands directly. However, no reasonable TS structures could
be found for this process starting from 2Hsonly very high-
energy species with the methyl group migrating essentially as
a free methyl radical could be located.33 Even after deproto-
nation of the hydroxyl group by triethylamine to make the
acetate a better leaving group, free energies of activation of
about 200 kJ/mol were computed. Transfer of the proton from
the HHPP hydroxyl group to the peroxyl oxygen that is not
bonded to copper leads to spontaneous C-O bond cleavage to
regenerate the original hydroperoxycopper acetone complex.
On that basis, we consider the Baeyer-Villiger type 1,2-methyl
shift to be more likely than the 1,3-methyl shift to explain the
observed reactivity of 2H. However, we cannot rule out the
The coordinated hydroxyl group in intermediate 5H, which
is -213.2 kJ/mol more stable than the initial reactant 2H at
the mPWPW91 level, can subsequently nucleophilically
attack the carbonyl carbon of the coordinated methyl acetate
8230 Inorganic Chemistry, Vol. 47, No. 18, 2008

Reactions of Copper(II)-H2O2 Adducts
possibility that a proton shuttle like that implicated in ring
hydroxylation might not make the 1,3-methy shift pathway
accessible.
Reaction of Copper(II) Complexes and H2O2 in
Propionitrile (CH3CH2CN). The reaction of 1Ar(H) and H2O2
in propionitrile under otherwise the same experimental
conditions (at -70 °C in the presence of 1 equiv of Et3N)
gave a spectrum exhibiting an intense absorption band at
360 nm (ε ) 3150 M-1 cm-1, Figure 7), which is quite
different from that of 2Ar(H) (λmax ) 420 nm, ε ) 1100 M-1
cm-1).10 Although the instability of this intermediate 7Ar(H)
did not permit detailed characterization by ESR, resonance
Raman, and ESI-mass spectra (decomposition rate kdec ) 2.4
× 10-2 s-1 at -70 °C), we assume that 7Ar(H) is a simple
hydrogen peroxide adduct LCuII-OOH (Type A) as in the
reaction of 1H and H2O2 in propionitrile shown below.
Interestingly, decomposition of 7Ar(H) resulted in a color
change of the solution from green to pale yellow; the d-d bands
due to the copper(II) complex in the visible region were
completely bleached (see Figure 7), and a copper(I) complex
of LAr(H) (8Ar(H)) was obtained as the final product (Supporting
Information, Figure S15). In this case, the original ligand was
recovered nearly quantitatively (no ligand modification). These
results clearly indicate that LCuII-OOH species supported by
LAr(H) (7Ar(H)) have neither arene hydroxylation ability nor
aliphatic C-H bond activation ability but instead undergo
CuII-O bond homolysis to give the copper(I) product 8Ar(H)
(eq 4). This result is in sharp contrast to the reactions of Karlin’s
TPA system, where aromatic hydroxylation as well as oxidative
N-dealkylation reactions took place in the reactions of copper(II)
complexes and H2O2 in acetone.1d,e
Figure 7. Spectral change for the reaction of 1Ar(H) (0.6 mM) with H2O2
(0.6 mM) in the presence of Et3N (0.6 mM) in CH3CH2CN at -70 °C.
Inset: the time course of the absorption change at 360 nm.
Figure 8. Spectral change for the reaction of 1H (0.2 mM) with H2O2 (2.0
mM) in the presence of Et3N (2.0 mM) in propionitrile at -70 °C. Inset:
Resonance Raman spectrum of 7H (5 mM) generated by using H216O2
obtained with λex ) 406.7 nm in propionitrile at -80 °C, s denotes the
solvent peaks.
The reaction of 1H and H2O2 in propionitrile also gave an
intense absorption band at 384 nm (ε ) 2850 M-1 cm-1,
Figure 8), which is also different from that of the HHPP-
adduct 2H (Figure 2). In this case, the hydroperoxo copper(II)
LCuII-OOH (7H) could be detected by its resonance Raman
spectrum, which exhibited a peak at 878 cm-1 attributable
to the O-O stretching vibration, although the isotope shift
could not be detected because of the overlap of the 18O-18O
stretching vibration peak with the large solvent peak around
840 cm-1 (see inset of Figure 8).
Conclusions
The reaction of copper(II) complexes 1H and 1Ar(X)
supported by the bis(2-pyridylmethyl)amine derivatives (LH
and LAr(X), respectively, see Chart 2) and H2O2 have been
examined in detail. Both complexes provided 2-hydroxy-2-
hydroperoxypropane (HHPP) adduct complexes 2H and 2Ar(X)
in acetone, whose formation has been confirmed by UV-vis,
resonance Raman, ESR, and ESI-MS spectra. In the case of
2
Ar(X), aromatic ligand hydroxylation took place to give the
phenolate-copper(II) product complexes 4Ar(X) in good yields,
Inorganic Chemistry, Vol. 47, No. 18, 2008 8231

Kunishita et al.
during the course of which another intermediate 3Ar(X) was
also detected. Kinetic studies including deuterium KIE and
Hammett analysis of the reaction have indicated that the
reaction process from 2Ar(X) to 3Ar(X) involves C-O bond
formation between the peroxo moiety of 2Ar(X) and one of
the phenyl substituents on the pyridine donor groups through
an electrophilic aromatic substitution mechanism. Our DFT
calculations suggest that this C-O bond formation is
significantly facilitated by ammonium/amine acid/base ca-
talysis. Such an acid/base catalysis not only enhances the
heterolytic O-O bond cleavage necessary for the electro-
philic attack by the peroxo moiety on the aromatic ring of
the ligand but also makes the acetal a better leaving group.
The relative importance of the ultimate deprotonation from
the hydroxyl group of the HHPP moiety in 2Ar(X) for the
aromatic ligand hydroxylation is evident, since no such ligand
hydroxylation occurred in a simple alkylperoxo complex
LCuII-OOCm (Cm ) cumyl) supported by the same
tridentate ligand.34
the HHPP moiety from the hydroxyl group to the peroxo
group is important and is facilitated by amine/ammonium
acid/base catalysis.
In propionitrile as the solvent under otherwise identical
reaction conditions with H2O2, simple copper(II)-hydroper-
oxo complexes 7H and 7Ar(X) were produced. From these
complexes no ligand modification took place, which is an
interesting contrast with the TPA systems of Karlin, where
otherwise analogous LCuII-OOH species have been reported
to undergo both aromatic hydroxylation and oxidative
N-dealkylation of the supporting ligands.1d,e
Our results thus highlight the significant sensitivities to
solvent and ligand variation that are exhibited by reactions
between the copper(II) complexes and H2O2. In addition, our
computational work suggests that acid/base catalysis can play
a significant role in the O-O bond activation of reactive
LCuII-OOR species.
Acknowledgment. This work was financially supported
in part by Grants-in-Aid for Scientific Research on Priority
Area (Nos. 18033045, 19020058, 19027048, and 19028055
for S.I.) and by Global Center of Excellence (GCOE)
Program (“Picobiology: Life Science at Atomic Level” to
T.O.) from MEXT, Japan. The U.S. National Science
Foundation also provided funding (CHE-0610183).
In contrast to adducts having the opportunity to oxidize
aromatic ring substituents, another HHPP-adduct 2H gave a
copper(II)-acetate complex 6H, in which the acetate co-ligand
was derived from the HHPP moiety via a Baeyer-Villiger-
like 1,2-methyl shift and subsequent hydrolysis of the methyl
acetate so generated by the LCuII-OH species. In this case,
DFT calculations again suggested that proton migration in
Supporting Information Available: The ERS spectra (Figures
S1-S6) and the cyclic voltammograms (Figures S7-S12) of the
copper(II) starting materials, the resonance Raman spectrum of 2H
generated by using H218O2 in acetone-d6 (Figure S13), the ESI-MS
of 2H (Figure S14) and 8Ar(H) (Figure S15), and atomic coordinates
for the computed structures (Table S1). This material is available
free of charge via the Internet at http://pubs.acs.org.
(33) In the case of the HHPP-adduct of FeIII(TPA), the formation of acetate
product was suggested to involve O-O bond homolytic cleavage of
the HHPP moiety and subsequent methyl radical release from the
generated (Me)2C(OH)O· intermediate.25
(34) Kinishita, A.; Ishimaru, H.; Nakashima, S.; Ogura, T.; Itoh, S. J. Am.
Chem. Soc. 2008, 130, 4244–4245.
IC800845H
8232 Inorganic Chemistry, Vol. 47, No. 18, 2008