Electrophilic arene hydroxylation and phenol O-H oxidations performed by an unsymmetric μ-I·1:I·1-O 2-peroxo dicopper(II) complex

Article abstract of DOI:10.1002/chem.201102372

Reactions of the unsymmetric dicopper(II) peroxide complex [Cu ^II₂(μ-I·¹: I·¹-O₂)(m-XYL^N3N4)]²⁺ (1O₂, where m-XYL is a heptadentate N-based ligand), with phenolates and phenols are described. Complex 1O₂ reacts with p-X-PhONa (X=MeO, Cl, H, or Me) at -90°C performing tyrosinase-like ortho-hydroxylation of the aromatic ring to afford the corresponding catechol products. Mechanistic studies demonstrate that reactions occur through initial reversible formation of metastable association complexes [Cu^II₂(μ- I·¹:I·¹-O₂)(p-X-PhO) (m-XYL^N3N4)]⁺ (1O₂·X-PhO) that then undergo ortho-hydroxylation of the aromatic ring by the peroxide moiety. Complex 1O₂ also reacts with 4-X-substituted phenols p-X-PhOH (X=MeO, Me, F, H, or Cl) and with 2,4-di-tert-butylphenol at -90°C causing rapid decay of 1O₂ and affording biphenol coupling products, which is indicative that reactions occur through formation of phenoxyl radicals that then undergo radical C-C coupling. Spectroscopic UV/Vis monitoring and kinetic analysis show that reactions take place through reversible formation of ground-state association complexes [Cu^II₂(μ- I·¹:I·¹-O₂)(X-PhOH) (m-XYL^N3N4)]²⁺ (1O₂·X-PhOH) that then evolve through an irreversible rate-determining step. Mechanistic studies indicate that 1O₂ reacts with phenols through initial phenol binding to the Cu₂O₂ core, followed by a proton-coupled electron transfer (PCET) at the rate-determining step. Results disclosed in this work provide experimental evidence that the unsymmetric 1O₂ complex can mediate electrophilic arene hydroxylation and PCET reactions commonly associated with electrophilic Cu₂O₂ cores, and strongly suggest that the ability to form substrate·Cu₂O₂ association complexes may provide paths to overcome the inherent reactivity of the O ₂-binding mode. This work provides experimental evidence that the presence of a H⁺ completely determines the fate of the association complex [Cu^II₂(μ-I·¹: I·¹-O₂)(X-PhO(H))(m-XYL^N3N4)] ⁿ⁺ between a PCET and an arene hydroxylation reaction, and may provide clues to help understand enzymatic reactions at dicopper sites.

Full text of DOI:10.1002/chem.201102372

FULL PAPER
DOI: 10.1002/chem.201102372
ꢀ
Electrophilic Arene Hydroxylation and Phenol O H Oxidations Performed
by an Unsymmetric
AHCTUNGTRENNUNG
Isaac Garcia-Bosch, Xavi Ribas,* and Miquel Costas*[a]
Abstract: Reactions of the unsymmet-
decay of 1O2 and affording biphenol
coupling products, which is indicative
that reactions occur through formation
of phenoxyl radicals that then undergo
(PCET) at the rate-determining step.
ric dicopper(II) peroxide complex
Results disclosed in this work provide
experimental evidence that the unsym-
metric 1O2 complex can mediate elec-
trophilic arene hydroxylation and
PCET reactions commonly associated
with electrophilic Cu2O2 cores, and
strongly suggest that the ability to form
substrate·Cu2O2 association complexes
may provide paths to overcome the in-
herent reactivity of the O2-binding
mode. This work provides experimental
evidence that the presence of a H+
completely determines the fate of the
[CuII (m-h1:h1-O2)(m-XYLN3N4)]2+ (1O2,
2
where m-XYL is a heptadentate N-
based ligand), with phenolates and
phenols are described. Complex 1O2
reacts with p-X-PhONa (X=MeO, Cl,
H, or Me) at ꢀ908C performing tyrosi-
nase-like ortho-hydroxylation of the ar-
omatic ring to afford the corresponding
catechol products. Mechanistic studies
demonstrate that reactions occur
through initial reversible formation of
ꢀ
radical C C coupling. Spectroscopic
UV/Vis monitoring and kinetic analysis
show that reactions take place through
reversible formation of ground-state as-
sociation complexes [CuII (m-h1:h1-O2)-
2
A
(m-XYLN3N4)]2+
(1O2·X-
PhOH) that then evolve through an ir-
reversible rate-determining step. Mech-
anistic studies indicate that 1O2 reacts
with phenols through initial phenol
binding to the Cu2O2 core, followed by
metastable
association
complexes
[CuII (m-h1:h1-O2)
G
association complex [CuII (m-h1:h1-O2)-
2
2
XYLN3N4)]+ (1O2·X-PhO) that then un-
dergo ortho-hydroxylation of the aro-
matic ring by the peroxide moiety.
Complex 1O2 also reacts with 4-X-sub-
stituted phenols p-X-PhOH (X=MeO,
Me, F, H, or Cl) and with 2,4-di-tert-bu-
tylphenol at ꢀ908C causing rapid
a
proton-coupled electron transfer
(X-PhO(H))(m-XYLN3N4)]n+ between a
ACHTUNGTRENNUNG
PCET and an arene hydroxylation re-
action, and may provide clues to help
understand enzymatic reactions at di-
copper sites.
Keywords: coordination modes
·
electron transfer · reaction mecha-
nisms · reactive intermediates · hy-
droxylation
Introduction
histidine residues in the reduced CuI state. The identity of
the species formed in pMMO after reaction with O2 remains
unknown,[11] but in Tyr, CO, and Hc, O2 binding results in
Multicopper proteins that bind and activate O2 play impor-
tant roles in a number of biological processes.[1–5] Among
them, proteins in which O2 binding and activation takes
place at a dicopper site include Hemocyanin (Hc), which
acts as an O2 carrier in arthropods and mollusks,[6–8] tyrosi-
nase (Tyr), which catalyzes 4eꢀ oxidation of tyrosine to dop-
aquinone in the first step of melanin biosynthesis,[9] catechol
oxidase (CO), which carries out the 2eꢀ oxidation of cate-
chols to quinones,[10] and particulate methane monooxyge-
nase (pMMO), which catalyzes methane oxidation to metha-
nol.[11] Irrespective of their function, this group of proteins
shares a nitrogen-rich active site and, except for pMMO, in
the absence of O2, each copper ion is three coordinate to
the formation of a common CuII (m-h2:h2-O2) species.[9,10,12]
2
From this step the chemistry diverges depending on the spe-
cific protein. The chemical reasons that underlay the differ-
ent reactivity from an apparently common Cu2O2 species
remain a matter of intense study in enzymology and synthet-
ic model chemistry.[4,13] It has been shown that one of the
factors that modulate the diverse chemistry is substrate ac-
cessibility to the active site.[14–17] Other factors, such as asym-
metry between the two copper coordination sites, and spe-
cific substrate orientation at the Cu2O2 core, can also be im-
portant.[13,18] For instance, crystallographic analyses show
that the oxygenated form of tyrosinase (oxyTyr) binds a
molecule of the phenolic substrate in one of the copper cen-
ters,[9] triggering a rotation of the peroxo core and orientat-
ing the ortho-phenolic site into a position that is conducive
[a] I. Garcia-Bosch, Dr. X. Ribas, Dr. M. Costas
QBIS Group, Departament de Quꢀmica
Universitat de Girona
for
electrophilic
attack
by
the
Cu2-O2
center
Campus de Montilivi,
(Scheme 1).[9,15,19] Beyond its biological relevance, the topic
of O2 activation at dicopper sites has elements of fundamen-
tal chemical interest and significance. From a very basic
point of view, it involves the use of inexpensive, abundant,
17071 Girona, Catalonia (Spain)
E-mail: miquel.costas@udg.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201102372.
Chem. Eur. J. 2012, 18, 2113 – 2122
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2113

Scheme 3. Selected Cu2O2 complexes that mimic Tyr activity.
molecules were envisioned as interesting substrate probes
because their oxidation has technological relevance,[27–29] and
is a very common reaction in a myriad of enzymatic systems.
Two basic oxidative transformations are commonly found
Scheme 1. Proposed mechanism for the hydroxylation of phenols cata-
lyzed by Tyr.
ꢀ
for this type of substrate; homolytic breakage of the O H
and environmentally benign reagents, O2 and copper, to me-
diate highly selective oxidative transformations.[20] In the
particular case of Tyr, a chemically challenging regioselec-
bond, and oxidation of the aromatic ring. Examples of the
first reaction involve tyrosil radical formation in Type I ribo-
nucleotide reductase, in which a diiron dimer reacts with a
tyrosine residue,[30,31] and cytochrome c oxidase, in which the
ꢀ
tive C H oxidation takes place. Because of this, the topic
ꢀ
has received much interest from researchers in chemical
fields that range from coordination chemistry to synthetic
methods. A number of synthetic complexes have been pre-
pared that can bind O2, and their reactivity has been ex-
plored.[13,21–26] Arising from these studies, three basic Cu2O2
core structures have been widely described (Scheme 2), and
breakage of the O O bond between Fe and Cu centers, on
the way towards the 4eꢀ reduction of O2, is helped by hydro-
gen atom transfer (HAT) from a nearby tyrosine.[32] Oxida-
tion of the arene is exemplified in the hydroxylation taking
place in Tyr,[13] and in the synthesis of the TPQ (2,4,5-trihy-
droxyphenylalaninequinone) cofactor in copper-containing
amine oxidases.[33] Stoichiometric oxidation of phenols and
phenolates by synthetic Cu2O2 compounds has been previ-
ously studied by several groups. Intermolecular hydroxyl-
ation of phenolates has been described for a few Cu2O2 spe-
cies (Scheme 3), which can be regarded as functional Tyr
models. These include CuII -m-h2:h2-O2 complexes described
Scheme 2. Most common Cu2O2 cores.
2
by Casella,[34] Itoh,[35] and Stack.[36,37] In the first two cases,
substrate binding precedes electrophilic attack of the perox-
ide ligand towards the arene. In the latter case, phenolate
it has been shown that each specific core determines the
particular spectroscopic and chemical properties.[13,21,23]
binding causes isomerization of the core towards CuIII
2ACHTUNGTRENNUNG(m-
Whereas end-on trans-CuII (m-h1:h1-O2) species exhibit nu-
O)2 that then effects the attack on the aromatic ring. Bis-
oxo CuIII2(m-XYLMeAN(O)2]2+ complex (Scheme 3) have
also been shown to be capable of effecting phenolate bind-
ing and ortho-hydroxylation, and constitute an alternative
2
cleophilic/basic behavior, side-on CuII (m-h2:h2-O2) and bis-
2
m-oxo dicopperACHTUNGTRENNUNG 2ACHTUTGNREN(NNGU m-O)2) cores show an electrophil-
(III) (CuIII
ic/acid oxidant character.[13,21] The reactivity of the different
Cu2O2 units has therefore been generally understood on the
basis of the core (Scheme 2).[25] Taking into consideration
that most dicopper enzymes contain elements of asymmetry
in their active site that make the two copper sites inequiva-
lent, and speculating that this aspect could introduce addi-
tional factors that mediate the inherent reactivity of the site,
we have been interested in studying O2-activation reactions
mediated by unsymmetric dicopper complexes. Phenolic
model for Tyr.[38] Substrate accessibility to the CuIII
2ACHTUNGTRENNUNG(m-O)2
core has been shown to be necessary for the reaction to
occur.[39] Very recently, we reported an intriguing example
where the unsymmetric [CuII (m-h1:h1-O2)(m-XYLN3N4)]2+
2
(1O2; Scheme 4) reacted with para-substituted sodium phe-
nolates, yielding the corresponding catechols.[40] Complex
1O2 constitutes the only example of electrophilic arene hy-
droxylation mediated by an end-on Cu2(m-h1:h1-O2) species.
2114
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 2113 – 2122

X. Ribas, M. Costas et al.
FULL PAPER
Catalytic ortho-hydroxylation of phenols with a synthetic di-
copper complex has been recently accomplished.[41] Phenol
oxidation by well-defined Cu2O2 complexes has also been
studied.[42–49] Fundamental details for the reaction were pro-
(Scheme 4).[51,52] In these complexes the authors propose
that the O2 molecule adopts a m-h1:h2-O2 binding mode. We,
however, formulate 1O2 as [CuII (m-h1:h1-O2)(m-XYLN3N4)]2+
2
ꢀ
because the energy of the O O stretching frequency in
vided by Itoh and Fukuzumi, who showed that CuII (m-h2:h2-
Cu2O2 complexes is very sensitive to the binding mode,[23]
2
O2) and CuIII
2A
and the n
C
ꢀ
noxyl radicals.[50] Mechanistic analysis showed that the reac-
very similar. Since the binding mode of O2 in 2O2 can be
unambiguously established as m-h1:h1-O2 on the basis of its
spectroscopic data and the bis-tetradentate nature of the
m-XYLN3N4 ligand,[23] we concluded that 1O2 also adopts an
end-on binding mode. Unfortunately, no crystallographic
characterization of any of the unsymmetric Cu2O2 com-
plexes has been accomplished, and a note of caution is war-
ranted.[40,51,52] Complex 1O2 can be used to perform the
ortho-hydroxylation of sodium phenolates to form the corre-
sponding catechol products.[40] Reaction of 1O2 with elec-
tron-rich p-X-C6H5O-Na+ (X=H, Cl, F, MeO) substrates in
acetone at cryogenic temperatures (ꢀ90 to ꢀ608C) results
in rapid bleaching of the spectral features associated with
1O2 (l=478 nm, e=7800mꢀ1 cmꢀ1, and a broad shoulder be-
tween 575 and 700 nm). Acidic workup at the end of the re-
actions and subsequent HPLC-MS analyses show the forma-
tion of p-X-catechols in moderate yields (14–38%) with re-
spect to 1O2 (see the Supporting Information, Table S1). No
catechol product was formed when electron-poor phenolates
(X=CN, NO2, and CO2Me) were used as substrates. The
electron-rich, sterically more demanding 2,4-di-tert-butyl-
phenolate was neither hydroxylated nor oxidized to the cor-
responding diphenol-coupled product. Compound 1O2 is
therefore a functional model for Tyr, but arene hydroxyl-
ation occurs exclusively for electron-rich, sterically unhin-
dered phenolate substrates. Previous DFT calculations pro-
vided computational evidence that the coordination of the
ꢀ
tions entail PCET from the phenol O H bond to the Cu2O2
core. The unsymmetric [CuII (m-h1:h2-O2)(LN3N4)]2+ complex
A
2
(Scheme 4) has also been shown to be competent at media-
Scheme 4. Complexes 1O2, 2O2, and unsymmetric dicopper complex
ACHTUNGTRENNUNG
[CuII(LN3N4)(O2)]2+ described by Itoh and co-workers.[51]
ting the reaction.[51] Finally, a tricopper complex containing
a Cu3
II,II,III(m3-O) core can also oxidize 2,4-di-tert-butylphe-
ACHTUNGTRENNUNG
nol, acting as a 2eꢀ oxidant to yield one equivalent of diphe-
nol-coupled product per tricopper molecule. A substrate
binding event was deduced from kinetic analyses, which in-
dicated saturation kinetics on phenol concentration.[49] In
this work, we explore the reaction of phenolates and phe-
phenolate does not cause isomerization of the CuII O2 core
2
to fundamentally more electrophilic centers such as CuII -m-
2
nols with the unsymmetric trans-m-h1:h1-peroxo-CuII species
h2:h2-O2 or CuIII (m-O)2, and that the CuII -m-h1:h1-O2 species
2ACHTUNGRTNEGNU
2
2
(1O2). We show that, in both cases, reactions involve forma-
tion of spectroscopically detectable [Cu2O2(substrate)] ad-
ducts, (substrate=p-X-phenolate ion or p-X-phenol) ad-
ducts, and that the formation of such adducts is a prerequi-
mediates electrophilic attack at the arene ring through a
¼
readily small activation barrier (DG =61.5 kJmolꢀ1).[40]
Kinetic analysis: Because of the unique character of the in-
termolecular arene hydroxylation reactivity exhibited by
1O2, a detailed kinetic study was performed. Monitoring the
reaction between 1O2 and p-X-phenolates by using UV/Vis
ꢀ
site for both arene hydroxylation and phenol O H oxida-
tion.
spectroscopy revealed a shift in the more intense lmax
=
Results and Discussion
478 nm band (corresponding to 1O2) to lmax =470 nm. Sys-
tematic kinetic experiments were conducted under the same
conditions ([1O2]=0.1 mm, ꢀ908C) varying the concentra-
ortho-Hydroxylation of para-substituted phenolates
tion of the phenolate substrate. Decay in intensity at lmax =
478 nm was monitored by UV/Vis, and decay rates (kobs
)
We have recently communicated the synthesis and charac-
were obtained by fitting the data to a single exponential
decay. At low [p-X-PhONa], kobs was linearly related to [p-
X-PhONa], but a saturation curve was observed at higher
concentrations (Figure 1). This behavior was observed for
all the phenolate substrates studied (4-methoxyphenolate, 4-
methylphenolate, 4-fluorophenolate, phenolate, and 4-chlor-
ophenolate) and is consistent with a two-step mechanistic
scheme in which a substrate-dependent pre-equilibrium pro-
terization of 1O2 and [CuII (m-h1:h1-O2)(m-XYLN4N4)]2+
2
(2O2; Scheme 4), which were formulated as trans-1,2-peroxo
dicopper(II) complexes on the basis of their Raman and
UV/Vis spectra, and DFT calculations.[40] The spectroscopic
features of complex 1O2 also match those corresponding to
Cu2O2 complexes containing unsymmetric N3N4-heptaden-
tate ligands recently described by Itoh and co-workers
Chem. Eur. J. 2012, 18, 2113 – 2122
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2115

Arene Hydroxylation and Phenol Oxidation
were plotted against the corresponding Hammett substituent
constants s+, a linear regression was obtained with 1=ꢀ2.7
(Figure 2), which is indicative of an electrophilic reaction.
Figure 1. Plots of kobs(ONa) versus [4-X-PhONa] for different substrates
performed by 1O2.
Figure 2. Hammet plot for the ortho-hydroxylation of 4-X-PhONa
(k2(ONa)) by 1O2.
cess defined by Keq(ONa) precedes a rate-determining step
(RDS) decay (k2(ONa); see Scheme 5). From the saturation
curves, a Michaelis–Menten type analysis was performed
from which Keq(ONa) and k2(ONa) values were extracted (see
the Supporting Information); these are summarized in
Table 1. These equilibrium constants can be compared to
The 1 value appears to be substantially more negative than
CHTUNGTRENNUNG
values obtained with model complexes [CuII (MeL66)(O2)]2+
(ꢀ1.8), [CuII
N
Table 1. Rate constants for the reaction between 1O2 and different
sodium phenolates.[a]
Substrate
T [K]
Keq[ONa] [Mꢀ1
]
k2[ONa] [sꢀ1
]
s+
4-MeO-PhONa
4-Me-PhONa
4-F-PhONa
PhONa
183
183
183
178
183
188
193
198
203
183
183
5600ꢁ400
1000ꢁ50
1440ꢁ20
4200ꢁ200
3200ꢁ100
2200ꢁ300
1800ꢁ80
1430ꢁ30
1210ꢁ40
3240ꢁ80
3200ꢁ500
13.77ꢁ0.06
4.1ꢁ0.9
ꢀ0.78
ꢀ0.31
ꢀ0.07
0
1.98ꢁ0.03
0.06ꢁ0.01
1.40ꢁ0.08
3.4ꢁ0.4
Temperature-dependence kinetic analysis: The reaction with
sodium phenolate was analyzed in the range of temperatures
178 to 203 K (see the Supporting Information, Figure S2).
The Keq(ONa) and k2(ONa) values (Table 1) calculated from the
saturation curves at different temperatures allowed the con-
struction of van’t Hoff and Eyring plots, respectively, from
which both thermodynamic parameters (Table 2) for the as-
sociative pre-equilibrium process (Keq(ONa)) and the activa-
tion parameters for the RDS (k2(ONa)) could be extracted
(Figure 3). The thermodynamic properties associated with
the equilibrium step show that the process is enthalpically
favorable DH8=ꢀ15.3ꢁ0.7 kJmolꢀ1 and the negative sign
of the entropic change (DS8=ꢀ17ꢁ4 Jmolꢀ1 Kꢀ1) is consis-
tent with the associative nature of the reaction. The coordi-
nation event is energetically favorable (DG8=ꢀ12ꢁ2 kJ
·molꢀ1 at 183 K) because the enthalpic contribution of the
formation of the CuII-phenolate adduct is higher than the
6.7ꢁ0.4
10.0ꢁ0.8
27ꢁ3
[D5]PhONa
4-Cl-PhONa
1.59ꢁ0.06
1.4ꢁ0.2
0
0.11
[a] See Experimental Section and the Supporting Information for further
details.
those described for lithium phenolate binding to CuII -m-
2
h2:h2-O2 and CuIII
2ACHTUNGTRENNUNG
(m-O)2 cores. For example, the binding
constant of lithium p-Cl-phenolate to [CuII (LPy2Bz)2(O2)]2+
2ACHNUTTGRENNUNG
is approximately 1000mꢀ1,[35] which is similar to our system
(ca. 3200mꢀ1). On the other hand, sodium phenolate binds
2ACHTUNGTRENNUNG
to [CuII (MeL66)(O2)]2+ with weaker binding constants
(Keq(ONa) <100mꢀ1), and temper-
ature-dependence kinetic analy-
sis shows that this difference
Table 2. 1 values, thermodynamic, and activation parameters for the ortho-hydroxylation of phenolates per-
formed by selected model systems and Tyr.
has
an
enthalpic
origin
¼[a]
¼[b]
(Table 2).[34] We suggest that
this difference reflects the fact
that that the N3CuIIO site in
1O2 is more electrophilic and/
or less sterically crowded than
Complex
1
DH8[a]
DS8[b]
DH
DS
1O2
ꢀ2.7
ꢀ2.4
ꢀ1.8
ꢀ1.8
ꢀ2.2
ꢀ1.9
ꢀ15.3ꢁ0.7
ꢀ17ꢁ4
42ꢁ2
61ꢁ9
29.1ꢁ3
n.d.
40.2
37.1ꢁ0.5
ꢀ45ꢁ11
Tyrosinase[c]
–
–
ꢀ24/+21ꢁ11
ꢀ115ꢁ15
n.d.
A
(MeL66)(O2)]2+[d]
ꢀ8.1ꢁ1.2
ꢀ8.9ꢁ6.2
N
–
–
–
–
–
–
R
n.d.
ꢀ55ꢁ3
the N3CuO2 site in [CuII -
N
2
ACHTUNGTRENNUNG
(MeL66)(O2)]2+. On the other
[a] kJmolꢀ1. [b] Jmolꢀ1 Kꢀ1. [c] See Granata et al.[19] [d] See Palavicini et al.[34] [e] See Itoh et al.[35] [f] See
hand, when the k2(ONa) values Mirica et al.[37] [g] See Company et al.[38]
2116
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 2113 – 2122

X. Ribas, M. Costas et al.
FULL PAPER
Figure 3. Van’t Hoff plot (top) and Eyring plot (bottom) for the reaction
of complex 1O2 with NaOPh. The van’t Hoff plot was performed for the
Keq(ONa) substrate binding event; the Eyring plot was calculated from the
k2(ONa) values. The range of temperatures employed was 178–203 K.
entropic energetically nonfavorable organization of the
system. On the other hand, activation values for the pheno-
late oxidation process derived from the Eyring plot are
¼
¼
DH =42ꢁ2 kJmolꢀ1 and DS =ꢀ45ꢁ11 Jmolꢀ1 Kꢀ1. The
Scheme 5. Top) Schematic mechanism proposed for the ortho-hydroxyl-
ation of sodium phenolates performed by 1O2. Bottom) Detailed sche-
matic diagram summarizing the DFT calculations, showing the rate-deter-
mining arene oxidation event.[40]
enthaplic activation barrier is similar to values obtained for
¼
dinuclear complexes [CuII
G
¼
40.2 kJmolꢀ1) and [CuIII2(m-XYLMeAN (m-O)2]2+ (DH =
)2AHCUTNGERNNUG
37.1 kJmolꢀ1), but somewhat smaller than in [CuII -
[34]
G
.
The free-energy activation barrier ex-
tracted from the kinetic analysis (DG =50ꢁ4 kJmolꢀ1 at
183 K) is in reasonably close agreement with the DFT-com-
puted value for attack of the peroxide ligand in 1O2 on the
aromatic ring of 4-Me-PhO (61 kJmolꢀ1).[40] On the other
hand, the corresponding activation enthalpy in Tyr is signifi-
electrophilic attack of the peroxide moiety on the aromatic
ring.[18] Computational analysis (Scheme 5, bottom) indicate
that the barrier for peroxide attack on the aromatic ring is
small (14.7 kcalmolꢀ1), and the computed transition state
contains the peroxide moiety bound with a m-h1 bridging
mode (Scheme 5, bottom). This transition state evolves to
cantly larger (DH =61ꢁ4 kJmolꢀ1), and it is proposed to
¼
ꢀ
ꢀ
be associated with the O O cleavage event.
form an intermediate in which the peroxide Caryl bond has
been formed and the aromaticity of the arene has been dis-
rupted. The experimentally determined inverse KIE is
indeed consistent with the disruption of the aromaticity of
the arene substrate predicted from the calculations. It
should be noticed that the structure of the transition state is
very similar to that previously described by Siegbahn in a
DFT-computed mechanism of tyrosinase.[53] In addition, in-
tramolecular arene hydroxylation by a m-1,1-hydroperoxide
Kinetic isotope effects: Kinetic isotopic effects (KIE) were
determined by using sodium [D5]phenolate as a substrate
(see the Supporting Information, Figure S3). The equilibri-
um constant was not altered by substrate deuteration
(Keq(ONa)(H)/Keq(ONa)(D) =0.99), but a kinetic isotopic effect
was found for the RDS (k2(ONa)(H)/k2(ONa)(D) =0.88). The in-
verse KIE is consistent with a peroxide electrophilic attack
on the aromatic ring, involving sp2 to sp3 isomerization in
the RDS.
also finds precedent in the literature.[54] Subsequent O O
ꢀ
bond cleavage and rearomatization should follow to form
the catechol product, but computational details for these
steps are still under study. Most significant, the presence of
available coordination sites that allow for substrate binding
appears to play a key role in eliciting the arene oxidation re-
Mechanistic scenario: The sum of all experimental data is
consistent with a mechanistic scenario (Scheme 5) in which
reversible phenolate binding to the N3CuII center is followed
by electrophilic attack of the trans-peroxo moiety on the ar-
omatic ring. Previous DFT analyses indicate that phenolate
binding to the tridentate copper atom places the p system of
the arene ring adjacent to the peroxide oxygen atom bound
to the tetradentate Cu center. This offers a pathway for s*
action by the CuII O2 center. This is reflected in the inability
2
of the related symmetric trans-peroxo complex [CuII (m-
2
h1:h1-O2)(m-XYLN4N4)]2+
,
2O2
to
form
analogous
Cu2O2·OPh-X adducts, or to perform analogous ortho-hy-
droxylation of phenolates.
Chem. Eur. J. 2012, 18, 2113 – 2122
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2117

Arene Hydroxylation and Phenol Oxidation
Oxidation of phenols
Product analysis: In acetone solution at ꢀ908C, 1O2 reacts
with 2,4-di-tert-butylphenol, 4-methylphenol, and 4-chloro-
ꢀ
phenol to yield the corresponding C C coupling 2,2’-diphe-
nol products in 50% (0.25 mol product/mol 1O2), 30%
(0.15 mol product/mol 1O2), and 28% (0.14 mol product/
mol 1O2), respectively. No additional products were detect-
ed when reactions were analyzed by HPLC or 1H NMR
spectroscopy (see the Supporting Information for experi-
ꢀ
mental details). Formation of C C diphenol coupling prod-
ucts indicates that reactions involve 1eꢀ oxidation of the
phenol to form a phenoxyl radical, which then evolves
ꢀ
through C C radical coupling. In addition, yields of less
than 50% suggest that 1O2 acts as a 1eꢀ oxidant, as previ-
Figure 4. Reaction between 1O2 (0.1 mm) and 4-MePhOH (100 mm) at
183 K. Inset: decay of the UV/Vis band (l=478 nm) over time, after 4-
Me-PhOH addition.
ously documented for [CuII
2A
-
-
2
(LPy1)(O)2]2+, which contain a CuII -m-h2:h2-O2 and a CuIII
(m-O)2 core, respectively.[50] Therefore, 1O2, [CuII -
2
(LPy2)(O2)]2+, and [CuIII
(LPy1)(O)2]2+, all of which contain a
nols, such as 2,4,6-tri-tert-butylphenol, 2,6-di-tert-butylphe-
nol, or 4-methoxy-2,6-di-tert-butylphenol, did not react with
1O2. Binding at the N3Cu site is further supported by the
lack of an analogous interaction between 2O2 and phenols;
2O2 reaction with phenols causes a bleaching of its UV/Vis
features without formation of oxidation products. With the
exception of electron-poor phenols (X=CN, NO2, or
CO2Me), the new species [1O2·X-PhOH] that forms from
this interaction is not even stable at cryogenic temperatures
(ꢀ908C), and its visible spectroscopic features rapidly
bleach. On the other hand, species [1O2·X-PhOH] in which
X=CN, NO2, or CO2Me are stable at ꢀ908C and no change
to their UV/Vis features was observed over time (hours).
Cu2O2 unit, but in which O2 is bound with a different bind-
ꢀ
ing mode, or in which the O O bond is already cleaved,
behave as 1eꢀ oxidants against phenols. However, the relat-
ed symmetric complex [CuII (m-XYLN4N4)(O2)]2+ was not
2
capable of oxidizing 2,4-di-tert-butylphenol under the same
conditions. The latter observation conforms to the reported
ACTHNUTRGNEUNG
reactivity of [CuII (m-h1:h1-O2)(tpa)2]2+ (tpa=tris(2-pyridil-
2
methyl)amine), the prototypical example of a dicopper per-
oxide complex with a trans-O2-binding mode. This complex
undergoes acid–base reaction with phenols, extruding H2O2
and generating copper–phenolate species.[43] On the other
hand, the reactivity of 1O2 against phenols bears resem-
blance to that of [CuII
(LN3N4)(O2)]2+
.
This reactivity may
[51]
suggest that the formulation of 1O2 as a m-h1:h2-peroxo di-
Kinetic analysis: Kinetic analysis of the reaction of 1O2 with
different phenols was performed to shed light on the reac-
tion mechanism. Reaction rates were obtained by monitor-
ing the decay of the band at lmax =478 nm. Decays could be
satisfactorily fitted as single exponential processes, and
kobs(OH) values were extracted from these plots. At low [4-X-
PhOH], kobs(OH) values increased linearly with substrate con-
centration, but, at higher concentrations, a saturation curve
was evident (Figure 6). This behavior can be understood by
copper(II) complex cannot be completely ruled out.
UV/Vis profile of the reaction of 1O2 with phenols: When
the reaction of 1O2 with phenols in acetone solution was
monitored by UV/Vis spectroscopy at low temperature
(ꢀ908C), it was found to occur in two kinetically distinct
steps. A bathochromic shift of the spectral features of 1O2
(l=478 nm) was initially observed upon addition of the sub-
strate. The magnitude of the shift was dependent on the
nature of the phenol and reflected its electron-releasing
nature. The more electron-releasing the phenol, the larger
the shift towards higher energy. This shift (Figure 4) is there-
fore indicative of a ground state interaction between the
phenol and 1O2, which most likely involves phenol binding
at the unsaturated N3Cu site to form a [1O2·XPhOH] spe-
cies (see the Supporting Information, Figure S4). Characteri-
zation of [1O2·XPhOH] by vibrational spectroscopy (reso-
nance Raman) could not be achieved because the species
proved to be highly photosensitive and because of fluores-
cence interference problems. However, the shape of the
UV/Vis spectra of [1O2·X-PhOH] was almost identical to
that of 1O2 (Figure 4) and, because of this, we infer that the
peroxide ligand is retained. Consistent with this scheme,
which involves a substrate–1O2 binding event, bulky phe-
assuming that a fast pre-equilibrium, expressed by Keq(OH)
,
precedes an irreversible RDS (k2(OH)). By applying a Mi-
chaelis–Menten type of analysis, Keq(OH) and k2(OH) values
corresponding to the reaction with the series of phenols sub-
strates could be determined (Table 3). Binding constants
Keq(OH) obtained from this analysis are small and show little
influence on the nature of X (11.8<Keq <138). No clear ten-
dency of the relative Keq(OH) values arise from a comparison
among the series of phenol substrates employed (Table 3).
For this reason, it is likely that the relative Keq(OH) values re-
flect a subtle combination of electronic, steric, and/or supra-
molecular interaction factors. On the other hand, the magni-
tude of the k2(OH) values increase as the redox potential for
the 1eꢀ oxidation of phenols decrease. When the k2(OH)
values were plotted against the correspondent redox poten-
tials for the 1eꢀ oxidation of the phenol substrates, a linear
2118
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 2113 – 2122

X. Ribas, M. Costas et al.
FULL PAPER
Table 3. Rate constants for the reaction between 1O2 and phenols.
bimolecular nature, but en-
thalpically driven. The low free-
energy value reflects rather
modest binding constants be-
tween phenol substrates and
the Cu2O2 center. In compari-
son, phenol binding constants,
Keq(OH), are two orders of mag-
nitude smaller than those deter-
mined for phenolates, as initial-
ly expected because of the
anionic character of the latter.
Comparison between the re-
spective thermodynamic param-
eters associated with phenol
and phenolate binding to 1O2
show that binding of the former
is both enthalpically and en-
tropically less favored. On the
other hand, activation parame-
ters for the RDS of the reaction
[a]
[b]
Substrate
T [K]
Keq[OH] [Mꢀ1
]
k2[OH] [sꢀ1
]
E0
BDEO-H
ox
4-MeO-PhOH
183
183
183
183
183
183
183
183
183
183
183
188
193
198
203
183
183
183
183
183
41ꢁ1
138ꢁ7
138ꢁ17
27.3ꢁ0.5
11.8ꢁ0.2
15.8ꢁ0.6
–
0.35ꢁ0.09
0.010ꢁ0.003
0.005ꢁ0.001
0.042ꢁ0.002
0.03ꢁ0.01
0.024ꢁ0.01
–
1.43
1.46
–
ꢀ0.07
80.92
81.85
–
–
–
84.58
75.55
79.15
–
81.65
–
2,4-di-tert-butyl-PhOH
2,4-di-tert-butyl-PhOD
4-PhO-PhOH
4-Ph-PhOH
4-Me-PhOH
4-MeO-2,6-di-tert-butylPhOH
2,4,6-tri-tert-butyl-PhOH
4-F-PhOH
2,6-di-tert-butyl-PhOH
PhOH
1.54
–
1.58
1.61
1.62
–
–
–
27ꢁ2
0.0015ꢁ0.0002
–
–
14.9ꢁ0.1
13ꢁ2
12ꢁ1
10.2ꢁ0.8
9ꢁ1
0.00072ꢁ0.00001
0.0017ꢁ0.0003
0.0025ꢁ0.0008
0.009ꢁ0.002
0.0244ꢁ0.0001
0.00047ꢁ0.00001
0.0006ꢁ0.0001
–
[D5]-PhOD
4-Cl-PhOH
4-CO2Me-PhOH
4-CN-PhOH
4-NO2-PhOH
27ꢁ2
13ꢁ1
–
–
–
–
–
1.63
85.65
–
89.25
91.65
–
–
–
–
–
[a] Values extracted from Osako et al.[50] (V vs. SCE). [b] Values (kcalmolꢀ1) extracted from Wright et al.[55,56]
(k2(OH)) could be obtained from
¼
¼
correlation was observed (r2 =0.974) with a slope of approxi-
mately ꢀ0.49 (Figure 5). Significant deviation in this correla-
tion was observed for 2,4-di-tert-butyl phenol. We infer that
the steric bulk at the 4-position of this phenol substrate re-
duces the rate of the RDS, perhaps by introducing space
constrains to the transition state.
an Eyring plot (DH =52ꢁ10 kJmolꢀ1, DS =ꢀ15ꢁ
6 JKꢀ1 molꢀ1, DG =55ꢁ10 kJmolꢀ1 at 183 K). The reaction
¼
is thus characterized by a small activation barrier, and the
sign of the activation entropy is indicative of a more ordered
transition state. Finally, primary KIE values of 1.53 and 2.00
were determined by comparing k2(OH) reaction rates for
phenol against [D6]phenol, and 2,4-di-tert-butylphenol
against [D1]2,4-di-tert-butylphenol (Figure 6), respectively.
Figure 5. Plots of (RT/F)ln(k2(OH)) against the oxidation potentials (E0
)
ox
of phenols for the reaction between 1O2 and the selected phenols. The
data corresponding to 2,4-di-tert-butyl phenol has not been used in the
correlation.
Figure 6. Plot of kobs(OH) reaction rates determined for the decay of
[1O2·X-PhOH] in acetone at ꢀ908C, formed by reaction of substrate
(2,4-di-tert-butylphenol and [D1]2,4-di-tert-butylphenol) and 1O2. Decay
was monitored at lmax =478 nm.
Temperature-dependence kinetic analysis: Further mechanis-
tic details were obtained by following the kinetics of the re-
action with phenol at different temperatures (see the Sup-
porting Information, Figure S5). From the saturation curves,
equilibrium constants Keq(OH) and first-order constants k2(OH)
were extracted. These values were used to build a van’t Hoff
plot (see the Supporting Information, Figure S6) from which
the thermodynamic parameters associated with phenol-bind-
ing reaction could be determined to be DH8=ꢀ8.9ꢁ
0.7 kJmolꢀ1 and DS8=ꢀ26ꢁ4 JKꢀ1 molꢀ1. This analysis indi-
cates that binding is slightly exergonic (DG0 =ꢀ4.1ꢁ
1 kJmolꢀ1 at 183 K), entropically disfavored because of its
These values are small, but significant and are indicative of
ꢀ
a O H bond breakage event in the RDS. The KIE values
are also close to those obtained by Itoh and co-workers for
the PCET of the oxidation of phenols by CuII -m-h2:h2-O2
2
2ACTHNUTRGNEUNG
and a CuIII (m-O)2 cores systems,[50] and it is also consistent
with other experimental values obtained from the oxidation
of phenols through a PCET mechanism.[57]
Chem. Eur. J. 2012, 18, 2113 – 2122
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2119

Arene Hydroxylation and Phenol Oxidation
Mechanistic scenario: Summarizing all the mechanistic infor-
mation obtained, we can draw a scenario whereby sub-
strate–1O2 binding precedes a reaction in which phenol oxi-
dation occurs. This precoordination process has a low ener-
getic profile (ca. 4 kJmolꢀ1) that reflects the poor coordina-
tion ability of neutral phenols, and can be understood as a
weak interaction between the N3Cu center and the phenol
oxygen atom. It is also possible that other kinds of interac-
tion exist, such as p stacking between the substrate and aro-
matic groups of complex 1O2, and/or hydrogen bonding be-
an interesting comparison. As observed for 1O2, these reac-
tions also afford modest KIE values, and similar linear cor-
relations exist between the redox driving force and reaction
rates, but the correlations exhibit more pronounced slopes
(ca. 0.7).[50] These reactions also differ from that of 1O2 be-
cause they can be kinetically described as bimolecular pro-
cesses; no spectroscopic nor kinetic evidence for the inter-
mediacy of a substrate·Cu2O2 association complex was
found, although such intermediates may be inferred because
the analogous reactions with phenolates occur through ini-
tial phenolate binding.[35] Therefore, the measured reaction
rates may also contain a contribution from an initial Cu2O2/
phenol association event. That is not the case for 1O2, be-
cause phenol binding and eꢀ/H+ transfer are well separated,
and the agreement with theory is remarkable. Product quan-
tification established that 1O2 acts as a 1eꢀ oxidant. This is
ꢀ
tween the phenol (O H) hydrogen atom and the peroxide
moiety (Scheme 6). Most significant is the existence of a
also the case for [CuII (m-h2:h2-O2)
G
2ACHTUNGTRENNUNG(m-
2
(LPy1)2]2+
O)2ACHTUGNTERNUNGN .
tion appear to be very short-lived, do not accumulate in so-
lution, and are consequently difficult to characterize. We hy-
ACTHNUTRGNEUNG
pothesize that a [CuII(OOH)CuI] species may be formed in
this reaction, but their exact nature cannot be ascertained.
2ACTHNUTRGNEUNG
We disfavor mixed-valence bis-m-oxo [CuIII (m-O)2CuII]2+
ꢀ
formulations resulting from O O breakage because the ster-
ically demanding m-XYLN3,N4 ligand appears to be unsuita-
ble for supporting the short Cu···Cu distances of such a com-
pact core.[40] Moreover, these new species do not lead to fur-
ther substrate oxidation reactions, and therefore some kind
of rapid decomposition reaction, presumably involving per-
oxide disproportionation, must exist.
The role of substrate binding: From an inspection of the re-
activity exhibited by 1O2 and 2O2 (Scheme 4), substrate
binding is seen to be a crucial step for enabling both arene
hydroxylation and phenol oxidation. Complex 2O2 shares
with 1O2 the same m-h1:h1-O2 binding mode, but it is not ca-
pable of oxidizing phenolates or phenols because its copper
centers are coordinatively saturated (N4CuO). The ability of
1O2 to mediate “electrophilic” PCET and arene hydroxyl-
Scheme 6. Mechanism proposed for the reversible binding of phenols to
1O2 followed by PCET to form a phenoxyl radical.
linear correlation between the redox potential of the phe-
nols and (RT/F)ln(k2(OH)), with a slope value of approxi-
mately ꢀ0.5 (Figure 5). This graphic can be understood as a
ation reactions, previously limited to CuII (m-h2:h2-O2) and
2
¼
DDG8 versus DDG correlation, and, in the present case, in-
CuIII
2ACHTGUNTERNNU(G m-O)2 cores, strongly suggest that a substrate-binding
dicates that an electron transfer is involved in the RDS.[58]
The slope of this correlation has mechanistic significance;
by applying a Markus formalism, a concerted PCET should
yield a slope of ꢀ0.5, provided the reorganization barrier
(l) is larger than the driving force (DG8).[57,59] On the other
hand, hydrogen atom transfer reactions (HAT) yield very
small slopes that indicate a rather small influence of the
redox driving force (EoxꢀEred) in the overall HAT reaction
rate.[50] The experimental value of ꢀ0.49 is in very good
agreement with that expected for a PCET reaction. Further-
more, the existence of a small but significant KIE in the oxi-
rather than an O2-binding mode enhances the electrophilic
reactivity of the Cu2O2 core. Indeed, copper ions in CuII (m-
2
h2:h2-O2) and CuIII
2ACTHNUTRGNE(NUG m-O)2 cores generally have low coordi-
nation numbers, which allow for substrate binding, but
CuII (m-h1:h1-O2) species generally possesses N4CuO units.[23]
2
Nevertheless, because crystallographic characterization of
1O2 could not be accomplished, a cautionary note should be
made at this point. The chameleonic electrophilic/nucleo-
philic reactivity exhibited by 1O2 is reminiscent of that ex-
hibited by [CuII (LN3N4)(O2)]2+ [51] and may also be consistent
CHTUNGTERNNUNG ,
2A
with a m-h1:h2-O2 binding mode. We notice, however, that
2A
1O2 and [CuII (LN3N4)(O2)]2+ also show substantial differen-
ꢀ
dation of phenols indicates that breakage of the O H bond
CHTUNGTRENNUNG
also takes place during the RDS. The experimental data is
therefore indicative of a PCET reaction. Moreover, PCET
ces in their respective reactivities. Most importantly, we be-
lieve that the spectroscopic data is conclusive in the assign-
ment of the binding mode of 1O2 as end-on. However, we
cannot completely discount the possibility that a m-h1:h2-O2
reactions from phenols to [CuII (m-h2:h2-O2)
(LPy2)2]2+ and
2
[CuIII
2ACHTUNGTRENNUNG(m-O)2ACHTUNGTRENNUNG
(LPy1)2]2+ have recently been studied and offer
2120
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 2113 – 2122

X. Ribas, M. Costas et al.
FULL PAPER
¼
mode is close in energy and could contribute to the ob-
served reactivity.
k2(ONa) oxidation rate has a higher entropic barrier (DS
(phenolate)=ꢀ45 Jmolꢀ1 Kꢀ1) than k2(OH) (DS (phenol)=
¼
ꢀ15 Jmolꢀ1 Kꢀ1), which indicates that peroxide attack on the
arene requires a tighter, more ordered transition state than
PCET of the phenol. This ordered transition state may re-
flect the requirement for a suitable orientation of the arene
p orbitals and the s* peroxide orbitals along the so-called
s* mechanism for Tyr-like reactivity.[15]
Phenol versus phenolate binding: The kinetic analysis pro-
vides interesting clues for the substrate binding reaction.
First, the Keq(ONa) are roughly 100 times larger than Keq(OH)
,
likely because of the phenolate anionic character (DG8
(188 K); DG8(phenolate)ꢀDG8
ACHTUNGTRENNUNG
sumably because of this, phenolate binding to Cu2O2 species
has precedent, but evidence for analogous phenol binding is
unknown to the best of our knowledge. Our kinetic analysis
also suggests a slightly higher organization of the phenol
Conclusion
binding
reaction
(DS8
G
vs.
This work provides experimental evidence that 1O2 medi-
ates electrophilic arene hydroxylation and PCET reactions
commonly associated to electrophilic Cu2O2 cores. These re-
sults strongly suggest that the ability to form substrate·-
Cu2O2 association complexes may provide paths to over-
come the inherent reactivity of the O2-binding mode.[59] This
work also provides experimental evidence that the presence
of H+ completely determines the fate of the association
DS8(phenolate)=ꢀ17 Jmolꢀ1 Kꢀ1), which is indicative of a
higher organization in 1O2·X-PhOH with respect to 1O2·X-
PhO species. Because the Cu O(H)Ph bond in phenolates
is expected to be shorter and stronger than in phenols on
the basis of charge considerations, the reverse respective
order measured in the entropic component of the reactions
suggest that some other “ordering factor” contributes in
1O2·X-PhOH. We suggest that a hydrogen-bonding interac-
ACTHNUTRGNEUNG
complex [CuII (m-h1:h1-O2)(X-PhO(H))(m-XYLN3N4)]n+ be-
2
ꢀ
tion between the phenol O H bond and the peroxide
oxygen atoms (Scheme 6), and/or p-staking interactions ac-
tween a PCET and an arene hydroxylation reaction.
counts for this additional entropic contribution.
Phenolate ortho-hydroxylation versus phenol (1H+/1eꢀ) oxi-
dation: Although the mechanistic details of Tyr activity have
been widely studied, some questions arise from its unique
reactivity.[13] The first step in the oxidation of phenols by the
enzyme involves coordination of the phenol substrate to the
Cu2O2 center. However, it is not clear which base in the
enzyme active site is responsible for deprotonating the
phenol substrate.[13] On the other hand, the oxidation of
phenols through PCET reactions is one of the most common
reactions in biological systems and is involved in a number
of diverse oxidation processes. The observation that Cu2O2
cores are capable of reacting with phenols to form phenoxyl
radicals raises the question of what is the chemical basis for
Experimental Section
General Procedures: UV/Vis spectroscopy was performed with an Agi-
lent 50 Scan (Varian) UV/Vis spectrophotometer with 1 cm quartz cells.
Low temperature control was achieved with a cryostat obtained from
Unisoku Scientific Instruments, Japan. NMR spectra were recorded with
a Bruker DPX400 spectrometer under standard conditions. Complex
[CuI2(m-XYLN3N4)](CF3SO3)2 (1) and [CuI2(m-XYLN4N4)]
ACTHNUTRGNENUG CAHUTGNTREN(NUGN ClO4)2 (2) were
synthesized as previously described.[40] Sodium phenolates were synthe-
sized from the correspondent phenols as described elsewhere[39] and were
1
characterized by H NMR analysis.
Product analysis: Analysis of the products obtained from the reaction
with phenolates was previously described.[40] The products obtained from
the reaction of 1O2 with phenols were analyzed by 1H NMR spectrosco-
py in [D6]acetone with acetophenone as internal standard. In a typical ex-
periment, complex 1O2 or 2O2 was generated at ꢀ908C by bubbling mo-
lecular O2 through a solution of a complex 1 previously prepared under
anaerobic conditions ([1O2], [2O2]=0.1 mm). The excess O2 was re-
ꢀ
eliciting electrophilic arene hydroxylation or phenol O H
bond breakage. Our observations,[40] along with those previ-
ously described for CuII (m-h2:h2-O2) and CuIII
2ACTHNUTRGNE(UNG m-O)2
2
moved by several vacuum/N2 cycles and
a solution containing 10–
cores,[34–39,48,49] indicate that, irrespective of the O2 binding
mode, the form by which the phenol/phenolate substrate
binds to the dicopper center has a decisive role in its reactiv-
ity towards electrophilic aromatic oxidation, or phenoxyl
radical generation. The decisive role played by H+ in facili-
tating the phenoxyl radical generation can be easily under-
stood by considering that 1eꢀ oxidation of phenolates
occurs at higher potential than in phenols.[4] Thus, phenolate
binding quenches its 1eꢀ oxidation through electron trans-
fer. Whether phenolate binding also has an effective role in
activating either the arene or the Cu2O2 core, or both, to-
wards aromatic hydroxylation cannot be concluded. Howev-
er, it is interesting to note that k2(ONa) values corresponding
to the hydroxylation are 20 times larger than k2(OH) values
100 equiv of phenol was added. The reaction was quenched by the addi-
tion of HCl (0.5m, 3 mL) and the acetone was removed from the mixture
under reduced pressure. The resulting aqueous phase was extracted with
CH2Cl2 (3ꢂ10 mL), the organic fraction was dried over MgSO4, and the
solvent was removed under reduced pressure. The resulting product was
solved with [D6]acetone and analyzed by 1H NMR spectroscopy using
ꢀ
mesitylene as internal standard. The presence of C C coupled products
was confirmed by HPLC-MS analysis.
Reaction with phenolates and phenols; kinetic analysis: In a typical ex-
periment, an anhydrous acetone solution of complex 1 or 2 ([0.1 mm],
3 mL) was prepared under anaerobic conditions ([O2]<1 ppm, [H2O]<
1 ppm) and was placed in a special UV/Vis cell (1 cm path length) with
the appropriate design for the cryostat-UV/Vis system. The solution was
cooled to ꢀ908C and molecular oxygen was injected using an O2-filled
balloon. After accumulation of 1O2 (or 2O2), 0.2 mL of an anhydrous
acetone solution containing the corresponding equivalents of the desired
sodium phenolate/phenol were injected and the decay of the 1O2 band
was recorded until no significant changes in the UV/Vis spectra were ob-
served.
¼
¼
corresponding to the PCET reaction (DG (188 K); DG
¼
(phenol)ꢀDG (phenolate)=5 kJmolꢀ1. Remarkably, the
Chem. Eur. J. 2012, 18, 2113 – 2122
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2121

Arene Hydroxylation and Phenol Oxidation
[32] E. Kim, E. E. Chufan, K. Kamaraj, K. D. Karlin, Chem. Rev. 2004,
104, 1077–1134.
Acknowledgements
[33] M. Mure, Acc. Chem. Res. 2004, 37, 131–139.
[34] S. Palavicini, A. Granata, E. Monzani, L. Casella, J. Am. Chem. Soc.
2005, 127, 18031–18036.
[35] S. Itoh, H. Kumei, M. Taki, S. Nagatomo, T. Kitagawa, S. Fukuzumi,
J. Am. Chem. Soc. 2001, 123, 6708–6709.
I. G.-B. thanks MICINN for a PhD grant. M. C. and X. R. acknowledge
financial support from ICREA-Academia Awards, Catalan SGR 2009-
SGR637, MICINN-Spain (CTQ2009–08464/BQU), the European Re-
search Foundation for Project ERC-2009-StG-239910, and Consolider In-
genio/CSD2010–00065.
[36] L. M. Mirica, D. J. Rudd, M. A. Vance, E. I. Solomon, K. O. Hodg-
son, B. Hedman, T. D. P. Stack, J. Am. Chem. Soc. 2006, 128, 2654–
2665.
[37] L. M. Mirica, M. Vance, D. J. Rudd, B. Hedman, K. O. Hodgson,
E. I. Solomon, T. D. P. Stack, Science 2005, 308, 1890–1892.
[38] A. Company, S. Palavicini, I. Garcia-Bosch, R. Mas-Ballestꢄ, L.
Que, Jr., E. V. Rybak-Akimova, L. Casella, X. Ribas, M. Costas,
Chem. Eur. J. 2008, 14, 3535–3538.
[39] S. Herres-Pawlis, P. Verma, R. Haase, P. Kang, C. T. Lyons, E. C.
Wasinger, U. Flrke, G. Henkel, T. D. P. Stack, J. Am. Chem. Soc.
2009, 131, 1154–1169.
[40] I. Garcia-Bosch, A. Company, J. R. Frisch, M. Torrent-Sucarrat, M.
Cardellach, I. Gamba, M. Guell, L. Casella, L. Que, X. Ribas, J. M.
Luis, M. Costas, Angew. Chem. 2010, 122, 2456–2459; Angew.
Chem. Int. Ed. 2010, 49, 2406–2409.
[1] E. I. Solomon, P. Chen, M. Metz, S.-K. Lee, A. E. Palmer, Angew.
Chem. 2001, 113, 4702–4724; Angew. Chem. Int. Ed. 2001, 40, 4570–
4590.
[2] E. I. Solomon, U. M. Sundaram, T. E. Machonkin, Chem. Rev. 1996,
96, 2563–2605.
[3] E. I. Solomon, J. W. Ginsbach, D. E. Heppner, M. T. Kieber-
Emmons, C. H. Kjaergaard, P. J. Smeets, L. Tian, J. S. Woertink,
Faraday Discuss. 2011, 148, 11–39.
[4] S. Itoh, S. Fukuzumi, Acc. Chem. Res. 2007, 40, 592–600.
[5] S. Itoh in Dicopper Enzymes, Vol. 8 (Eds.: J. A. McCleverty, T. J.
Meyer), Elsevier, 2003, 369–393.
[6] K. A. Magnus, H. Ton-That, J. E. Carpenter, Chem. Rev. 1994, 94,
727–735.
[41] M. Rolff, J. Schottenheim, G. Peters, F. Tuczek, Angew. Chem. 2010,
122, 6583–6587; Angew. Chem. Int. Ed. 2010, 49, 6438–6442.
[42] N. Kitajima, T. Loda, Y. Iwata, Y. Moro-oka, J. Am. Chem. Soc.
1990, 112, 8833–8839.
[43] P. P. Paul, Z. Tyeklꢅr, R. R. Jacobson, K. D. Karlin, J. Am. Chem.
Soc. 1991, 113, 5322–5332.
[44] H. V. Obias, Y. Lin, N. N. Murthy, E. Pidcock, E. I. Solomon, M.
Ralle, N. J. Blackburn, Y.-M. Neuhold, A. D. Zuberbꢆhler, K. D.
Karlin, J. Am. Chem. Soc. 1998, 120, 12960–12961.
[45] S. Mahapatra, J. A. Halfen, E. C. Wilkinson, L. Que, Jr., W. B.
Tolman, J. Am. Chem. Soc. 1994, 116, 9785–9786.
[46] J. A. Halfen, J. Victor G. Young, W. B. Tolman, Inorg. Chem. 1998,
37, 2102–2103.
[47] V. Mahadevan, J. L. DuBois, B. Hedman, K. O. Hodgson, T. D. P.
Stack, J. Am. Chem. Soc. 1999, 121, 5583–5584.
[48] V. Mahadevan, M. J. Henson, E. I. Solomon, T. D. P. Stack, J. Am.
Chem. Soc. 2000, 122, 10249–10250.
[49] M. Taki, S. Teramae, S. Nagatomo, Y. Tachi, T. Kitagawa, S. Itoh, S.
Fukuzumi, J. Am. Chem. Soc. 2002, 124, 6367–6377.
[50] T. Osako, K. Ohkubo, M. Taki, Y. Tachi, S. Fukuzumi, S. Itoh, J.
Am. Chem. Soc. 2003, 125, 11027–11033.
[51] Y. Tachi, K. Aita, S. Teramae, F. Tani, Y. Naruta, S. Fukuzumi, S.
Itoh, Inorg. Chem. 2004, 43, 4558–4560.
[52] Y. Tachi, Y. Matsukawa, J. Teraoka, S. Itoh, Chem. Lett. 2009, 38,
202–203.
[53] P. E. M. Siegbahn, J. Biol. Inorg. Chem. 2003, 5, 567–576.
[54] G. Battaini, E. Monzani, A. Perotti, C. Para, L. Casella, L. Santagos-
tini, M. Gullotti, R. Dillinger, C. Nꢇther, F. Tuczek, J. Am. Chem.
Soc. 2003, 125, 4185–4198.
[55] J. S. Wright, D. J. Carpenter, D. J. McKay, K. U. Ingold, J. Am.
Chem. Soc. 1997, 119, 4245–4252.
[56] J. S. Wright, E. R. Johnson, G. A. DiLabio, J. Am. Chem. Soc. 2001,
123, 1173–1183.
[57] I. J. Rhile, T. F. Markle, H. Nagao, A. G. DiPasquale, O. P. Lam,
M. A. Lockwood, K. Rotter, J. M. Mayer, J. Am. Chem. Soc. 2006,
128, 6075–6088.
[58] R. A. Marcus, N. Sutin, Biochim. Biophys. Acta 1985, 811, 265–322.
[59] M. S. Graige, M. L. Paddock, J. M. Bruce, G. Feher, M. Y. Okamura,
J. Am. Chem. Soc. 1996, 118, 9005–9016.
[7] K. E. van Holde, K. I. Miller, H. Decker, J. Biol. Chem. 2001, 276,
15563–15566.
[8] K. E. van Holde, K. I. Miller, Adv. Protein Chem. 1995, 47, 1–81.
[9] Y. Matoba, T. Kumagai, A. Yamamoto, H. Yoshitsu, M. Sugiyama, J.
Biol. Chem. 2006, 281, 8981–8990.
[10] C. Gerdemann, C. Eicken, B. Krebs, Acc. Chem. Res. 2002, 35, 183–
191.
[11] R. Balasubramanian, S. M. Smith, S. Rawat, L. Yatsunyk, T. L.
Stemmler, A. C. Rosenzweig, Nature 2010, 465, 115–131.
[12] M. E. Cuff, K. I. Miller, K. E. van Holde, W. A. Hendrickson, J.
Mol. Biol. 1998, 278, 855–870.
[13] M. Rolff, J. Schottenheim, H. Decker, F. Tuczek, Chem. Soc. Rev.
2011, 40, 4077–4098.
[14] C. Morioka, Y. Tachi, S. Suzuki, S. Itoh, J. Am. Chem. Soc. 2006,
128, 6788–6789.
[15] H. Decker, T. Schweikardt, F. Tuczek, Angew. Chem. 2006, 118,
4658–4663; Angew. Chem. Int. Ed. 2006, 45, 4546–4550.
[16] H. Decker, F. Tuczek, Trends Biochem. Sci. 2000, 25, 392–397.
[17] K. Suzuki, C. Shimokawa, C. Morioka, S. Itoh, Biochemistry 2008,
47, 7108–7115.
[18] H. Decker, R. Dillinger, F. Tuczek, Angew. Chem. 2000, 112, 1656–
1660; Angew. Chem. Int. Ed. 2000, 39, 1591–1595.
[19] A. Granata, E. Monzani, L. Bubacco, L. Casella, Chem. Eur. J.
2006, 12, 2504–2514.
[20] L. Que, W. B. Tolman, Nature 2008, 455, 333–340.
[21] E. A. Lewis, W. B. Tolman, Chem. Rev. 2004, 104, 1047–1076.
[22] G. Battaini, A. Granata, E. Monzani, M. Gullotti, L. Casella, Adv.
Inorg. Chem. 2006, 58, 185–233.
[23] L. M. Mirica, X. Ottenwaelder, T. D. P. Stack, Chem. Rev. 2004, 104,
1013–1046.
[24] S. Schindler, Eur. J. Inorg. Chem. 2000, 2311–2326.
[25] L. Q. Hatcher, K. D. Karlin, J. Biol. Inorg. Chem. 2004, 9, 669–683.
[26] N. Kitajima, Y. Moro-oka, Chem. Rev. 1994, 94, 737–757.
[27] A. Prokofieva, S. Dechert, C. Grosse, G. M. Sheldrick, F. Meyer,
Chem. Eur. J. 2009, 15, 4994–4997.
[28] A. Prokofieva, A. I. Prikhodꢃko, S. Dechert, F. Meyer, Chem.
Commun. 2008, 1005–1007.
[29] A. S. Hay, H. S. Blanchard, G. F. Endres, J. W. Eustance, J. Am.
Chem. Soc. 1959, 81, 6335–6336.
[30] J. Stubbe, D. G. Nocera, C. S. Yee, M. C. Y. Chang, Chem. Rev. 2003,
103, 2167–2202.
Received: July 29, 2011
[31] E. Y. Tshuva, S. J. Lippard, Chem. Rev. 2004, 104, 987–1012.
Published online: January 16, 2012
2122
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 2113 – 2122