Selective copper-mediated halogenation of aromatic rings under mild conditions

Article abstract of DOI:10.1002/ejic.201100489

A mild copper-mediated halogenation reaction of phenolic rings is reported. The reaction of bis(2-hydroxybenzyl)-1,3-diaminopropane (H₂bhbd) with copper(II) chloride in acetonitrile generates linear trinuclear [Cu ₃(bhcbd)₂Cl₂](CH₃CN), containing modified ligands, whose phenol moieties are selectively chlorinated at the 5-position. Under comparable experimental conditions with copper(II) bromide, bromination of the ligand is observed, albeit at a slower reaction rate. In the presence of trialkylorthoformates, which are used as dehydrating agents, a ring closure of the ligand is observed after removal of copper to yield a product with a six-membered ring. This product has been isolated and characterized by NMR spectroscopy and MS. Similarly, bromination of bis(2-hydroxybenzyl)-1,3- diiminopropane (H₂bhbdi) occurs in the presence of copper(II) bromide, with concomitant formation of a linear trinuclear complex. Surprisingly, an asymmetric dinuclear copper(II) coordination compound without chlorination of the ligand is obtained when the related ligand bis(2-hydroxybenzyl)-1,3-diminopropane (H₂bhbdi) reacts with copper(II) chloride. Copyright

Full text of DOI:10.1002/ejic.201100489

FULL PAPER
DOI: 10.1002/ejic.201100489
Selective Copper-Mediated Halogenation of Aromatic Rings Under Mild
Conditions
Ilaria Gamba,[a,b,c] Patrick Gamez,[d] Enrico Monzani,[a] Luigi Casella,[a] Ilpo Mutikainen,[e]
and Jan Reedijk*[b,f]
Keywords: Oxidation / Copper / Ring Closure / Halogenation
A mild copper-mediated halogenation reaction of phenolic
rings is reported. The reaction of bis(2-hydroxybenzyl)-1,3-
diaminopropane (H2bhbd) with copper(II) chloride in aceto-
nitrile generates linear trinuclear [Cu3(bhcbd)2Cl2](CH3CN),
containing modified ligands, whose phenol moieties are se-
lectively chlorinated at the 5-position. Under comparable ex-
perimental conditions with copper(II) bromide, bromination
of the ligand is observed, albeit at a slower reaction rate.
In the presence of trialkylorthoformates, which are used as
dehydrating agents, a ring closure of the ligand is observed
after removal of copper to yield a product with a six-mem-
bered ring. This product has been isolated and characterized
by NMR spectroscopy and MS. Similarly, bromination of
bis(2-hydroxybenzyl)-1,3-diiminopropane (H2bhbdi) occurs
in the presence of copper(II) bromide, with concomitant for-
mation of a linear trinuclear complex. Surprisingly, an asym-
metric dinuclear copper(II) coordination compound without
chlorination of the ligand is obtained when the related ligand
bis(2-hydroxybenzyl)-1,3-diminopropane (H2bhbdi) reacts
with copper(II) chloride.
mechanistic understanding of their (bio)chemical function-
ing. In fact, enzymatic systems are used nowadays to per-
form such selective reactions. For example, FAD-dependent
halogenases[5] have been recently characterized and em-
ployed to achieve selective chlorination of aromatic rings,
and nonheme halogenases have been used to chlorinate
saturated hydrocarbons.[6] The use of enzymes to perform
halogenation of small molecules is, however, expensive due
to the purification procedure needed to recover the biocata-
lyst.
Coordination chemistry may help to improve the efficacy
of such biomimetic processes to generate carbon–halogen
bonds, as demonstrated by recent reports on the halogena-
tion of alkanes and arenes[7–9] and on the halofunctionali-
zation of olefins.[10] Carbon–halogen bond formation by re-
ductive elimination has been extensively studied for many
late transition metal centres, such as Pd,[11,12] Pt[13] and
Rh,[14] and has been reported for Ni as well.[15,16] In con-
trast, Cu-mediated C–X (X = halogen) bond formation has
received little attention, despite the obvious advantages of
substituting expensive and/or toxic metals with a more envi-
ronmentally friendly alternative in catalytic processes.
This study reports unique aerobic arene halogenations
that have been carried out at room temperature using CuCl2
or CuBr2 as the source of chlorine or bromine and with
dioxygen as the sole oxidant. The presence of the common
dehydrating agent triethylorthoformate appears to be cru-
cial to avoid side-reactions, i.e. hydroxylation, instead of
halogenation.
Introduction
The halogenation of aromatic rings under mild condi-
tions is hard to perform and the synthetic methods cur-
rently available are poorly selective.[1] Nature has developed
a variety of elegant processes for halogenating specific sub-
strates with regio- and stereoselectivity. Hence, an improved
understanding of the biological routes to the halogenation
of substrates could lead to the development of new efficient
procedures. The discovery and characterization of naturally
occurring enzymatic halogenation mechanisms has become
an active area of research, increasing the number of known
halogenating enzymes.[2,3]
Halogenases are divided into five different classes[4]
whose structural characterization has provided a better
[a] Laboratory of Bioinorganic Chemistry, Department of
Chemistry,
University of Pavia,
via Taramelli 12, 27100 Pavia, Italy
[b] Leiden Institute of Chemistry, Leiden University,
P. O. Box 9502, 2300 RA Leiden, The Netherlands
E-mail: reedijk@chem.leidenuniv.nl
[c] Present address: Universidad Autónoma de Madrid,
Departamento de Química Inorgánica, Madrid, Spain
[d] ICREA, Universitat de Barcelona,
Departament de Química Inorgànica QBI,
Martí i Franquès 1-11, 08028 Barcelona, Spain
[e] Laboratory of Inorganic Chemistry, Department of Chemistry,
University of Helsinki,
P. O. Box 55, A. I. Virtasenaukio 1, 00014 Finland
[f] Department of Chemistry, King Saud University,
P. O. Box 2455, Riyadh 11451, Saudi Arabia
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201100489.
4360
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 4360–4368

Selective Cu-Mediated Halogenation of Aromatic Rings
The structural formulae of the starting and modified li-
gands are depicted in Figure 1.
Surprisingly, the reaction of H2bhbdi with CuCl2 in ace-
tonitrile yields a dinuclear copper(II) complex which does
not exhibit ligand modification, even in the presence of alk-
ylorthoformate. In fact, with this ligand, the same coordi-
nation compounds are obtained in methanol and acetoni-
trile, as illustrated by single-crystal X-ray diffraction stud-
ies.
Results and Discussion
Figure 1. Structures of the starting ligand bis(2-hydroxybenzyl)-
1,3-diaminopropane (H2bhbd) and the final chlorinated or bromin-
ated ligand (H2bhcbd or H2bhbrbd, respectively); and bis(2-hy-
droxybenzyl)-1,3-diminopropane (H2bhbdi) and the resulting bro-
minated ligand (H2bhbrbdi), with the ring numbering used.
General
The reactions reported have been carried out in air, as
dioxygen is required for the reaction to proceed; hence, no
chlorination is observed under anaerobic conditions. The
following Equation (1) describes the overall process, in the
case of chlorination of H2bhbd [H4LH2 in Equation (1)]:
The chlorination of a pyrazole ligand in the presence of
copper(II) ions has been reported and the corresponding
dinuclear copper(II) complex of the modified ligand was
characterized by X-ray diffraction.[17] Previous studies of
copper(II) coordination compounds from the ligand
H2bhbd have revealed the generation of linear trinuclear
metal clusters.[18] These complexes were synthesized in
methanol, and their characterization did not show any
modification of the ligand structure.
3 CuCl2 + 3 O2 + 2 H4LH2 Ǟ Cu3(LCl2)2Cl2 + 6 H2O
(1)
Note that the H atoms listed in front of L denote acidic
ionisable hydrogen atoms, whereas those listed behind L
indicate aromatic H atoms that may be replaced by chloride
ions.
The 5-position on the aromatic ring is most prominently
activated through copper-mediated oxidation of the ligand;
no reaction is found when Cu2+ is absent.
Bromination of H2bhbd occurs under comparable reac-
tion conditions with copper(II) bromide, but the process
takes place at a slower rate. Contrary to the case of the
chlorination, after 24 h, the reaction is not complete and
the original ligand is still observed, together with mono-
and dibrominated products, as detected by MS. The forma-
tion of polybrominated products clearly occurs before full
conversion of the starting ligand.
A dinuclear copper(II) species [Cu2(bhbdi)Cl2] is ob-
tained from methanol or acetonitrile reaction mixtures con-
taining H2bhbdi and copper(II) chloride. No modification
of H2bhbdi is observed in acetonitrile, under the reaction
conditions used with H2bhbd [H2L2H2 in Equations (2)
and (3)].
Subsequent studies using the ligand H2bhbdp [H2bhbdp
=
1,7-bis(2-hydroxyphenyl)-2,6-diaza-4-hydroxyheptane]
with copper(II) chloride in acetonitrile yielded a trinuclear
compound and a tetranuclear copper(II) cluster that also
have been structurally characterized.[19] In this case, partial
chlorination and hydroxylation of the ligand aryl groups at
the 5-position have been observed. In contrast, the trinu-
clear copper(II) clusters prepared from H2bhbdp and CuCl2
in methanol were found not to exhibit such halogenation of
the aromatic rings.[20]
As the solvent (acetonitrile) appears to be crucial for li-
gand modification, coordination reactions between CuX2
and H2bhbd in acetonitrile have now been carried out. In
this study, it is shown that the selective and complete chlori-
nation or bromination of the aromatic rings of the H2bhbd
ligand occurs in acetonitrile at room temperature, simply
through reaction of the ligand with CuCl2 or CuBr2 in air
The following Equation (2) describes the syntheses of
such dinuclear copper(II) complexes:
and both in the presence or absence of a dehydrating agent, 2 CuCl2 + H2L2H2 Ǟ Cu2(L2H2)Cl2 + 2 HCl
(2)
i.e. trimethyl- or triethylorthoformate.
The reaction in presence of copper(II) bromide [Equa-
tion (3)] in acetonitrile leads to the formation of a trinuclear
copper(II) cluster, and shows (partial) bromination of the
5-positions of the ligand aromatic rings, as deduced from
X-ray diffraction and elemental analysis.
When triethylorthoformate (eof) or trimethylorthofor-
mate (mof) is present in the reaction mixture (used to avoid
hydroxylation due to water traces), the main isolated prod-
uct after copper removal, is a six-membered ring, appar-
ently obtained by a condensation reaction involving the li-
gand and the alkylorthoformate.
3 CuBr2 + 2 O2 + 2 H2L2H2 Ǟ Cu3(L2Br2)2Br2 + 4 H2O
(3)
The synthesis of copper(II) complexes with the related
ligand H2bhbdi [bis(2-hydroxybenzyl)-1,3-diiminopropane]
has also been studied. Bromination of H2bhbdi occurs in
the presence of copper(II) bromide in acetonitrile, through
the formation of a linear trinuclear copper(II) complex.
UV/Vis studies allow clarification of the mechanistic fea-
tures of the reaction.
Description of the structures of [Cu3(bhcbd)2Cl2](CH3CN)
(1), [Cu2(bhbdi)Cl2] (2) and [Cu3(bhbrbdi)2Br2] (3)
The single-crystal X-ray analysis of 1, depicted in Fig-
ure 2, shows that 1 is a trinuclear centrosymmetric complex,
Eur. J. Inorg. Chem. 2011, 4360–4368
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4361

J. Reedijk et al.
FULL PAPER
consisting of three Cu2+ atoms in a linear arrangement, from two bridging bhbd2– ligands. A packing diagram is
comparable to copper(II) clusters previously obtained from given in Figure S2.
the same ligand in methanol.[18] The terminal metal centres,
The molecular structure of 2 is shown in Figure 3. Com-
Cu(1) and Cu(1a) are in a hardly distorted square-pyrami- pound 2 consists of a dinuclear asymmetric copper(II) unit,
dal geometry. More geometric details are given in Table 1, whose metal centres are tetracoordinated. The Cu(1) centre
and the crystallographic data are given in Table 2. The has a N2O2 square-planar coordination environment,
square base of the pyramid is formed by the N2O2 donor whereas Cu(2) exhibits a distorted tetrahedral Cl2O2 geom-
set from a (bcbd)2– ligand. At the apical position, a weakly etry. The structure is characterized by torsion angles O(21)–
coordinated Cl– disordered over two positions is present Cu(2)–O(1)–Cu(1) of 9.14(7)° and N(9)–Cu(1)–O(1)–Cu(2)
(ratio 30:70). Details are given in Figure S1 and Table S1. of 165.32(7)°. Details and packing are depicted in Figure 3,
The phenolate group acts as a bridging ligand, connecting S3 and S4 and listed in Tables 1, 2 and S2.
the terminal copper to the central Cu(2) ion. The central
The plane including the copper ions and the chloride li-
copper atom Cu(2) is tetracoordinate in a square-planar en- gands and the plane containing the copper centres and the
vironment with torsion angles O(1A)–Cu(2)–O(21)–Cu(1) ligand oxygen donors form a dihedral angle of 67.71°, and
of 159.3(2)° and O(21)–Cu(2)–O(1)–Cu(1) of 21.3(2)°, and the distance between the copper(II) atoms is 3.0884(8) Å.
Cl– has a very weak interaction with Cu(2); see Figure S1. The solid-state EPR spectrum, recorded at room tempera-
The Cu1–Cu2 distance is 2.9644(12) Å. The coordination ture, is silent, indicative of a strong interaction between the
environment of Cu(2) is formed by four phenolato O atoms two metal centres.
Table 1. Selected bond lengths [Å] and (dihedral) angles [°] for 1, 2 and 3.
1
2
3
Cu1–O1
Cu1–N9
Cu1–Cl1B
Cu1–Cl1A
Cu1–N13
Cu1–O21
Cu2–O1
Cu2–O21
Cu1–Cu2
Cu2–Cl1B
N9–Cu1–N13
N9–Cu1–Cl1B
O1–Cu1–N9
O1–Cu1–O21
O1–Cu2–O21
Cu2–O1–Cu1
N9–Cu1–O21
N13–Cu2–O1
1.948(7)
2.007(7)
2.570(12)
2.540(3)
2.018(8)
1.987(5)
1.916(5)
1.921(6)
2.964(1)
3.325(5)
94.9(3)
Cu1–O1
Cu1–N9
Cu1–Cu2
Cu1–O1
Cu1–N13
Cu1–O21
Cu2–O1
Cu2–O21
1.951(1)
1.954(1)
Cu1–O1
Cu1–N9
Cu1–Br1A
Cu1–Br1B
Cu1–N13
Cu1–O21
Cu2–O1
Cu2–O21
Cu1–Cu2
N13–Cu1–N9
N9–Cu1–Br1A
N9–Cu1–Br1B
O1–Cu1–N9
O1–Cu1–O21
O21–Cu2–O1
Cu1–O1–Cu2
N13–Cu1–O1
N9–Cu2–O21
1.995(2)
1.970(3)
2.7181(6)
2.7480(12)
1.949(3)
1.968(2)
1.956(2)
1.942(2)
2.8648(4)
97.44(11)
112.32(8)
103.6(3)
92.24(10)
77.79(9)
78.41(9)
94.20(9)
168.5(1)
154.4(1)
3.0884(8)
1.9710(17)
1.9336(14)
2.0181(13)
1.9553(14)
2.1980(7)
2.2045(6)
99.33(7)
76.62(6)
93.47(6)
126.58(4)
135.41(5)
102.16(6)
165.32(7)
168.2(1)
Cu2–Cl1
Cu2–Cl2
N9–Cu1–N13
O21–Cu1–O1
O1–Cu1–N9
O1–Cu2–Cl1
O21–Cu2–Cl2
Cu1–O1–Cu2
N9–Cu1–O21
N13–Cu1–O1
100.6(4)
93.0(3)
75.6(2)
77.9(3)
100.2(3)
163.8(2)
164.9(2)
166.7(5)
Table 2. Crystallographic data for 1, 2 and 3.
1
2
3
Formula
C38H38Cl4Cu3N6O4
975.16
monoclinic
173(2)
C17H16Cl2Cu2N2O2
478.30
monoclinic
173(2)
C34H21.9Br2.10Cu3N4O4
918.96
monoclinic
173(2)
Molecular weight
Crystals system
T [K]
Space group
P21/c
P21/c
P21/n
Cell length [Å] (a, b, c)
Cell angle [°] (β) (α, γ = 90°)
Cell volume [Å3]
Z
13.767(2), 14.789(3), 10.758(2)
101.08(3)
2149.5
2
1.507
990
1.764
11.270(2), 8.391(2), 18.291(4) 8.9610(10), 10.7000(10), 17.321(2)
97.64(3)
1714.4
4
1.853
960
2.806
100.880(10)
1630.9
4
1.871
913
Dcalcd. [mg/m3]
F(000)
M [mm–1]
4.557
Crystals size [mm]
θmin, θmax [°]
0.20ϫ0.20ϫ0.05
2.75–25.05
35418
3788 [R(int) = 0.0688]
0.0694, 0.1494, 1.309
0.30ϫ0.27ϫ0.10
2.70–27.48
26155
3898 [R(int) = 0.0393]
0.0228, 0.0630, 1.104
0.4865, 0.7667
0.10ϫ0.10ϫ0.10
3.00–27.50
28581
3691 [R(int) = 0.0497]
0.0292, 0.0749, 1.174
–0.472, 0.593
Reflections collected
Independent reflections
R1[a], wR2[b], S[c]
Min/max residual density [eÅ–3] –0.636, 0.732
2 2
2 2
[a] R1 = Σ|Fo| – |Fc|/Σ|Fo|. [b] wR2 = [Σ{w(Fo2–Fc ) }/Σ{w(Fo2)2}]1/2. [c] Goodness-of-fit S = [Σw(Fo2–Fc ) / (n – p)]1/2, where n is the
number of reflections and p the number of parameters.
4362
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 4360–4368

Selective Cu-Mediated Halogenation of Aromatic Rings
Figure 2. ORTEP representation of the molecular structure of the
centrosymmetric trinuclear compound [Cu3(bhcbd)2Cl2](CH3CN)
(1), as determined by X-ray diffraction (only one of the disordered
coordinated Cl ions is shown).
Figure 4. ORTEP representation of the molecular structure of the
trinuclear compound [Cu3(bhbrbdi)2Br2] (3), as determined by X-
ray diffraction. The ring bromination was estimated to be around
2–3%. H atoms were omitted for clarity, except at C5/17. Coordi-
nated Br1 is disordered.
C34H21.9Br2.10Cu3N4O4 as specified in Table 2. Actually, the
elemental analysis of the crystalline material (see Exp. Sect.)
confirms that about 2.5% bromination of the ligand was
achieved. As in 1, the coordinated bromide is disordered
over two positions (ratio 96:4) and also bridges the central
CuII ion with Br–Cu = 3.06–3.17 Å.
Two more single crystal structures of reaction products
were obtained with compositions close to that of 3, i.e. 3a
and 3b, and they are represented in Figures S7–S10, with
geometric details given in Tables S4 and S5. Compounds 3
and 3a differ slightly in the degree of ring bromination, but
are basically the same and are 3–5% brominated in the ring.
In 3b the degree of monobromination observed is higher
and corresponds to 35%; in this case acetonitrile is present
as a sixth ligand, semicoordinated at 3.05 Å, with an un-
usually small Cu–N–C angle of 109°.
Figure 3. ORTEP representation of the molecular structure of the
dinuclear compound [Cu2(bhbdi)Cl2] (2), as determined by X-ray
diffraction. The compound was isolated from a 1:1 (or a 1:2) mix-
ture of the ligand H2bhbdi and copper(II) chloride in acetonitrile,
with or without the use of a dehydrating agent.
The coordination environment of Cu2 in all three cases
is formed by four phenolato O atoms in the equatorial
Partial bromination of H2bhbdi is observed when copper plane. The bromides axially coordinated to Cu1 are sem-
bromide reacts with the ligand in acetonitrile (with a L/M icoordinated to the central Cu2 ion (Cu2–Br = on average
ratio of 1:3). Single crystals of 3 have been obtained from 3.06 Å).
several reaction mixtures, and X-ray diffraction studies re-
The Cu–Cu separation distances, Cu1–O1–O21–Cu2 tor-
vealed the partial bromination of the ligand 5-positions of sions angles and τ parameters for 1, 2 and 3 are presented in
the trinuclear linear copper(II) cluster obtained (see Fig- Table 3. Corresponding values for closely related alkoxido-
ure 4 and S5). Packing details are given in Figure S6, and bridged copper(II) complexes are included for comparison.
geometric details are given in Tables 1, 2 and S3. The geom-
The τ values for 1 and 3 (listed in Table 3) are indicative
etry for CuI is distorted five-coordinate, as often seen for of a slight distortion from square-pyramidal geometry in
CuII;[21] in fact τ = 0.17, where τ is 0 and 1 for ideal square- case of 3. Other structural data for 1, 2 and 3 were com-
pyramidal and trigonal-bipyramidal geometries, respec- pared with those of other trinuclear linear alkoxido-bridged
tively.[22]
clusters with the same type of ligand, reported in our earlier
The 100% bromination of the 5-position would be given work (Table 3).[20–22] It is seen that the Cu1–O1–O21–Cu2
by the molecular formula C34H24Br6Cu3N4O4 (six Br torsion angle in the case of 3 is significantly smaller than
atoms: two coordinated anions and four in the aromatic 5- that observed for the other compounds and for the related
positions). The 2–3% Br population at the 5-position found copper clusters obtained from bhbdp (6, 7 and 8). This
by XRD analysis is characterized by the formula smaller torsion angle might be related to the size of the
Eur. J. Inorg. Chem. 2011, 4360–4368
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4363

J. Reedijk et al.
FULL PAPER
Table 3. Selected bond lengths [Å] and (dihedral) angles [°] for a series of related alkoxido-bridged copper(II) compounds.
[b]
Complex
Cu–Cu
Cu1–O1–Cu2
O1–Cu2–O21 Cu1–O1–O21–Cu2 τ(Cu1)
Ref.
[d]
[Cu3(bhcbd)2Cl2](CH3CN) (1)
[Cu2(bhbdi)Cl2 (2)][a]
[Cu3(bhbrbdi)2Br2] (3)
2.9644(12)
3.0884(8)
2.8648(4)
2.9852(7)
2.9374(7)
3.0855(9)
2.9869(5)
100.2(1)
102.2(1)
94.2(1)
100.9(2)
98.4(1)
103.3(1)
101.1(1)
97.5(1)
77.9(3)
74.6(1)
78.4(1)
79.1(1)
76.4(1)
78.6(1)
78.8(1)
81.3(1)
153.4(3)
0.02
–
[d]
168.55(1)
141.2 (1)
166.8(2)
154.58(18)
179.6(2)
[d]
0.23
0.03
0.03
0.22
0.05
0.02
[20]
[20]
[21]
[22]
[21]
[Cu3(bhbd)2(CH3OH) (ClO4)2] (4)
2
[Cu3(bhbd)2Cl2](CH3OH)4 (5)
[Cu4(bhcbdp)2(μ-Cl)2Cl2](CH3CN)(6[c])
[Cu3(bhbdp)2(CH3OH)2(ClO4)2] (7)
165.18(15)
156.30(10)
[Cu3(bhbdp)2Cl2(CH3CN)2](CH3CN)2 (8) 2.9722(6)
[a] Dinuclear copper(II) compound. [b] The τ parameter is used to characterize the geometry of five-coordinate compounds, between the
trigonal bipyramid and the square-based pyramid. τ is calculated as (β-α)/60, where α and β are opposite angles in the xy plane. The τ
value ranges from 0 to 1. A value of zero characterizes a compound with a perfect square-pyramidal geometry, and a value of one
symbolizes a perfect trigonal-bipyramidal geometry.[22] [c] Tetranuclear copper(II) compound. [d] This work.
axially coordinated bromide ion and may reflect a higher of eof, complete conversion of the original ligand into
tension in the alkoxido bridge.
mono- and dichlorinated species is observed (see below and
the Supporting Information Figures S11–S18 for ESI-MS
analyses). At longer reaction times, the formation of poly-
chlorinated species is also detected by ESI-MS. The modifi-
cation in the multiplicity of the signals in the aromatic re-
gion of the H NMR spectra (after copper removal) agrees
with the substitution of a proton at the 5-position of the π
rings (Figure 5).
Furthermore, the isotope patterns of the signals detected
by ESI-MS are in agreement with the presence of two chlo-
ride atoms in the molecule (vide infra). With a M/L ratio
of 1:1, the principal product obtained after 24 h reaction
time is the monochlorinated ligand, as evidenced by ESI-
MS measurements.
With CuBr2, the formation of polybrominated species is
already observed while the original ligand is still present,
and a mixture of various products is obtained. Hence, the
reaction carried out with copper(II) chloride is more selec-
tive and only the mono- and dichlorinated products are
produced.
Characterization of the Modified Ligand After Copper(II)
Removal
1
The chlorination of H2bhbd has been studied for dif-
ferent metal/ligand ratios and as a function of time. For a
proper characterization of the reaction mixture, the cop-
per(II) is removed at the end of the reaction by addition of
excess (4 equiv.) Na2S, which produces a precipitate of cop-
1
per sulfide. The copper-free products were analyzed by H
NMR and ESI-MS spectroscopy to detect any ligand modi-
fication. A detailed procedure is presented in the Support-
ing Information.
With a copper(II) chloride/H2bhbd ratio of 3:1, after
24 h of reaction in air at room temperature in the presence
Ring-Closure Reaction Involving the Drying Agent eof (or
mof)
An additional modification of H2bhbd that takes place
in the presence of eof (or mof) has been detected by ESI-
MS spectroscopy, after the removal of copper. The expected
mass peaks for the reaction mixture, after Cu2+ removal,
are m/z = 355 (dichlorinated product + H+), 321 (mono-
chlorinated ligand + H+) and 287 (original ligand). How-
ever, ESI-MS peaks are observed at m/z = 365, 331 and
297. These three values are all 10 Dalton units larger than
those expected for all the species anticipated. Details of the
mass spectrum for the dichlorinated product are depicted
in Figure 6.
The same feature is observed for the bromination reac-
tion in the presence of dehydrating agent. The higher mass
values are ascribed to a condensation reaction between the
free ligand and (m)ethylformate obtained by hydrolysis of
eof or mof. The resulting six-membered ring can evolve to
form a cationic stabilized species through the elimination of
an (m)ethoxide molecule (Figure 7).
1
Figure 5. Aromatic regions of the H NMR spectra of (a) starting
ligand bis(2-hydroxybenzyl)-1,3-diaminopropane (H2bhbd) and (b)
the final chlorinated ligand H2bhcbd.
4364
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 4360–4368

Selective Cu-Mediated Halogenation of Aromatic Rings
pearance of the absorption bands at 350 and 640 nm, and
the formation of the dibrominated ligand is characterized
by the appearance of a new band at 320 nm (Figure 8 and
S19).
Figure 6. Enlargement of part of the ESI mass spectrum (after cop-
per removal) obtained after reaction of copper(II) chloride with
H2bhbd at room temperature for 24 h. The isotopic pattern is in
agreement with the presence of two chlorine atoms in the molecule.
Figure 7. Proposed mechanism for the formation of the six-mem-
bered, ring-closed cationic product.
Figure 8. UV/Vis spectra for a solution of H2bhbdi in acetonitrile
(black line) monitored as a function of time after addition of
CuBr2. The insets show enlargements of the regions 290–400 and
400–800 nm.
The NMR signal at 8.5 ppm found for the isolated
bischlorinated ligand, i.e. after copper(II) removal and puri-
fication by chromatography, is assigned to the C–H of the
ring-closed product (as has been reported also in previous
studies).[23] The positive charge is delocalized between the
two nitrogen atoms; hence the symmetry of the molecule is
conserved, as evidenced by the NMR spectrum (see Sup-
porting Information).
To further prove the formation of the ring-closed prod-
uct, H2bhbd has been dissolved in an acetonitrile solution
containing an excess of eof (or mof). The resulting mixture
has been analyzed by ESI-MS, and a weak peak is detected
at m/z = 297, which corresponds to the cyclic compound.
In the mass spectrum, the dominant signal belongs to the
protonated ligand (286+1). When H2bhbd reacts with an
excess of ethylformate (instead of eof, triethyl orthofor-
mate), the ESI mass spectrum shows a single peak at m/z =
297; the molecular weight peak corresponding to the origi-
nal ligand is not detected. Therefore, ethylformate also al-
lows a complete conversion of the ligand to the cyclized
compound. The reaction of H2bhbd with an excess of form-
aldehyde has been studied as well and a mass peak at m/z
= 299 is detected. The ligand reacts with CH2O to generate
a saturated six-membered ring, as is well known in the case
of 1,3-diamine in the presence of formates.[24]
The reaction rate, k1, for the monobromination step can
be determined by following the decrease in the absorbance
at 340 nm (at this wavelength, i.e. the isosbestic point for
the mono- to the dibrominated ligand reaction step, the ab-
sorbance changes are solely due to the formation of the
copper complex of the monobrominated ligand). The data
have been fitted with the classical equation for an ex-
ponential decay, Equation (4), which can be used in the case
of an irreversible reaction.
Abs(340 nm) = Abs340(ϱ) + B ϫexp(–k1 ϫ t)
(4)
where Abs340(ϱ) is the residual value of absorbance at
340 nm, B is the variation of absorbance observed during
the process, k1 is the rate constant for the monosubstitution
process and t is the time expressed in seconds.
The data obtained following the increase in the new band
at 320 nm can be interpolated in Equation (5), which can
be used to determine the value of the rate constant for the
formation of the coordination compound with the dibromin-
ated ligand.
Abs(320 nm) = Abs320(ϱ) + C ϫexp(–k1 ϫ t) + D ϫexp(–k2 ϫ t)
(5)
As expected, with H2bhbdi, this ring closure reaction is
not observed under the above conditions as the imine
groups cannot react with eof (or mof), in contrast to the
N–H groups of H2bhbd.
where Abs320(ϱ) is the residual value of absorbance at
320 nm, C is the variation of absorbance for the first bromi-
nation step and D is that for the second bromination occur-
ring with the rate constant k2.
The fitting of the data for the reaction carried out with
a M/L ratio of 3:2 (which corresponds to the ratio observed
by XRD) gives values of 1.42(Ϯ0.03) ϫ10–3 for k1 and
Bromination of H2bhbdi Investigated by UV/Vis
Spectroscopy
The modification of H2bhbdi in the presence of copper 2.9(Ϯ0.2) ϫ10–4 for k2 (Figure 9). Details for the variation
bromide has been studied by UV/Vis spectroscopy. The for- of k1 and k2 are given in Figure S20 as a function of the
mation of the monobrominated ligand leads to the disap- concentration of the reagents.
Eur. J. Inorg. Chem. 2011, 4360–4368
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4365

J. Reedijk et al.
FULL PAPER
Figure 10. Proposed mechanism for the oxidative chlorination of
H2bhbd mediated by CuII and Cl– (shown for the monochlorin-
ation step only). L can be a solvent molecule, such as CH3CN,
which is known to stabilize copper(I) species.
the rate-determining step for the reaction. The copper(I)
centres are then reoxidized by dioxygen (clearly no reaction
occurs in the absence of molecular oxygen, so the halogena-
tion of the ring should occur after the reoxidation of the
metal centres). Finally, through deprotonation and re-
arrangement, the aromatic ring is regenerated (IV; Fig-
ure 10).
The proposed mechanism explains the high selectivity of
the nucleophilic attack of the phenolate functions on the
para (5) carbon atom, which can be achieved through the
formation of a phenoxonium cation as the key active spe-
cies, as observed in related studies, i.e. in the case of the
oxidative polymerization of 2,6-dimethylphenol.[25]
The formation of polyhalogenated products or C–C cou-
Figure 9. Representation of the fitting of the variation of (top) the
absorbance at 340 nm with Equation (4) and (bottom) the ab-
sorbance at 320 nm with Equation (5).
The reaction (in the absence of eof) has been investigated
by varying the ligand and CuBr2 concentrations. The rate
constant for the monobromination step does not change
with the concentration of the ligand or CuBr2.
Conclusions
The halogenation of the aromatic rings of H2bhbd and
H2bhbdi, mediated by copper(II), occurs under extremely pling products, which are usually observed for radical-based
mild conditions, i.e. in air and at room temperature. The processes mediated by copper ions, is negligible. Moreover,
reaction observed in acetonitrile does not take place in no radical signal was detected by EPR spectroscopy under
methanol or dichloromethane. The precise role of the sol- the experimental conditions of the present study. Neverthe-
vent has yet to be elucidated; the fact that acetonitrile is less, a radical pathway cannot be excluded.
known to stabilize copper(I) species may play a role here.
Phenoxonium cationic species are also known to be stabi-
The oxidative halogenation occurs with a high selectivity lized by coordination to metal ions,[26] especially in presence
for the 5-position on the aromatic ring, which is also not of dinuclear copper(II) centres.
yet understood. A possible mechanism for the selective acti-
Theoretical mechanistic studies carried out in case of the
vation of the 5-position on the aromatic ring is proposed in oxidative polymerization of phenols have invoked a two-
Figure 10. The oxidative halogenation of the phenol groups electron transfer from a phenolate ligand to the metal (di-
may occur through the formation of a stabilized quinonic nuclear CuII) to form an oxido-coordinated singlet phen-
intermediate associated with a reduction of both bridged oxonium cation (i.e. with a positively charged ring) and a
copper(II) ions. Next, a chloride anion could selectively at- reduced metal centre.[25b,27]
tack the “activated” 5-position of the aromatic ring. Subse-
quent rearrangement of the quinone generates the chlori- time has allowed us to underline that the process takes place
nated arene. through monohalogenation steps with a high selectivity for
The investigation of the distribution of products in real
Thus, after formation of the trinuclear copper(II) com- aromatic substitution in the 5-position (more activated
pound and deprotonation of the phenolic functional groups compared to the 3-position, which is also known to be acti-
of H2bhbd (I; Figure 10), the phenolato unit is oxidized to vated).[28]
the corresponding phenoxonium cation, through the re-
In the presence of eof (or mof), ring-closed products are
duction of two copper(II) ions to a copper(I) species (II; generated. The original (H2bhbd) and halogenated (mono-
Figure 10). The phenoxonium cation could then be attacked or bischlorinated) ligands present in solution react quanti-
by a chloride anion (or bromide) (III; Figure 10), which is tatively with the hydrolyzed drying agent (eof or mof) to
4366
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 4360–4368

Selective Cu-Mediated Halogenation of Aromatic Rings
tion of CuCl2 (273 mg, 2 mmol) in acetonitrile (15 mL). In both
solutions, triethyl orthoformate was added (trimethyl orthoformate
can also be used) to remove traces of water present in the solvent
and therefore avoid the formation of partially hydroxylated prod-
ucts. The resulting reaction mixture was stirred in air for 24 h and
subsequently filtered. Slow evaporation of the solvent led to the
generate six-membered ring compounds that have been
identified by ESI-MS measurements.
The bromination of H2bhbdi is observed in the trinuclear
copper(II) cluster obtained in acetonitrile. Single-crystal X-
ray diffraction studies show that the brominated copper
compound exhibits the same structure as that observed for
the chlorinated one.
formation of brown crystals after 4–5 d. Yield: 13%. IR: ν =
˜
3139.5, 2924.4, 2866.2, 1683.2, 1596.0, 1575.7, 1479.8, 1456.5,
1418.8, 1322.9, 1273.5, 1195.0, 1078.8, 1043.9, 1011.9, 956.7, 866.6,
753.3, 730.0, 671.9, 401.7 cm–1. C34H32Cl6Cu3N4O4 (964.01): calcd.
UV/Vis spectroscopic studies have allowed us to deter-
mine the rate constants for the mono- and bisbromination
steps. The rate constant determined for the monobromin- C 42.36, H 3.35, N 5.81; found C 38.72, H 3.49, N 5.24. The C, H,
N analyses of the powders obtained fit with the molecular formula
C34H32N4; however, the samples contained impurities, probably
copper chloride used in excess to perform the reaction.
ation step, k1, is four times higher than that observed for
the formation of the bisbrominated product, k2. The ob-
served rate constants, k1 and k2, do not vary with the rea-
gent concentrations. Hydroxylated products have not been
observed, at least when dry solvents were used or in the
presence of a dehydrating agent (eof or mof).
Under the experimental conditions used for H2bhbd, the
ligand H2bhbdi is not chlorinated and a dinuclear cop-
per(II) compound is obtained; the same coordination com-
pound is obtained in methanol and acetonitrile, which has
been structurally characterized by X-ray diffraction.
Further studies are necessary to elucidate the intimate
mechanism of the selective activation of the 5-position of
the aromatic ring, which is still poorly understood. The
comprehension of the process, through the determination
of the active species involved, can lead to the development
of mild, efficient and selective catalysts for halogenation re-
actions of aromatic functions.
Synthesis of [Cu2(bhbdi)Cl2] (2): A solution of H2bhbdi (364 mg,
1.29 mmol) in acetonitrile (15 mL) was added to a solution of
CuCl2 (257 mg, 1.92 mmol) in acetonitrile (15 mL). In both solu-
tions triethyl orthoformate was added as a drying agent (trimethyl
orthoformate can also be used). The resulting reaction mixture was
stirred in air for 24 h and subsequently filtered. Slow evaporation
of the solvent led to the formation of dark crystals after one week.
Yield: 37%. C17H18Cl2Cu2N2O2: C, 42.51 H 3.78, N 5.83; found C
42.98, H 3.29, N 6.00. IR: ν = 354, 380, 418, 452, 538, 598, 604,
˜
668, 686, 740, 736, 76, 856, 898, 912, 962, 960, 1038, 1070, 1074,
1104, 1154, 1198, 1220, 1286, 1346, 1404, 1436, 1476, 1548, 1594,
1616, 2820, 2898, 3012 cm–1.
Synthesis of [Cu3(bhbrbdi)Br2] (3): A solution of H2bhbdi (368 mg,
1.30 mmol) in acetonitrile (15 mL) was added to a solution of
CuBr2 (574 mg, 2.58 mmol) in acetonitrile (15 mL). In both solu-
tions, triethyl orthoformate was added as a drying agent (trimethyl
orthoformate can also be used). The resulting reaction mixture was
stirred in air for 3 h and subsequently filtered. Slow evaporation
of the solvent led to the formation of dark crystals after 3–4 d.
C34H24Br8Cu3N4O4 (bisbrominated compound): calcd. C 29.54, H
1.75, N 4.05, Br 46.24 found C 44.30, H 3.53, N 6.30, Br 18.60.
Accordingly, the% of Br detected corresponds to the 2.5% of bro-
mination. This synthesis was repeated several times, with the aim
to increase and understand the degree of bromination. However,
observed differences were quite small. In two cases singles crystals
were obtained and analyzed as compounds 3a and 3b. Their com-
position is slightly different from 3, and the structures are presented
in the SI.
Experimental Section
General Remarks: All reagents and solvents were purchased from
commercial sources and used as received. C, H, N determinations
were performed with a Series II CHNS/O 2400 analyzer. 1H NMR
spectra were recorded with a DPX300 Bruker spectrometer. The
ESI-MS measurements were carried out with a Thermo Finnigan
Aqa mass detector.
Synthesis of H2bhbdi: Bis(2-hydroxybenzyl)-1,3-diminopropane
(H2bhbdi) was prepared by condensation of salicylaldehyde and
1,3-diamine in methanol. Yield: 87%. C17H20N2O2 (284.36): calcd.
C 71.8, H 7.09, N 9.85; found C 70.5, H 6.9, N 9.9. ESI-MS: m/z
Chlorination and Bromination Reaction of H2bhbd and H2bhbdi: A
solution of H2bhbd (360 mg, 1.25 mmol) or H2bhbdi (360 mg,
1.28 mmol) in acetonitrile (15 mL) was added to a solution of
CuCl2 (410 mg, 3 mmol) or CuBr2 (682 mg, 3 mmol) in acetonitrile
(15 mL). In both solutions, triethyl orthoformate was added as de-
hydrating agent (also trimethyl orthoformate can be used). The fi-
nal solution was stirred in air for 24 h, subsequently filtered and
analyzed by ESI-MS.
= 282.9. IR: ν = 3058.1, 2947.6, 2872.0, 1636.7, 1616.4, 1580.8,
˜
1557.7, 1496.1, 1417.8, 1383.9, 1363.5, 1340.3, 1314.1, 1281.3,
1212.4, 1145.0, 1122.3, 1104.9, 1084.6, 1051.8, 1026.4, 1007.1,
974.1, 880.0, 855.0, 779.3, 749.9, 736.0, 667.8, 640.1, 571.5, 552.8,
520.3, 464.2, 404.6, 355.2, 329.0 cm–1. 1H NMR (300 MHz, CDCl3;
Me4Si): δ = 2.11 (m, 2 H, CH2), 3.74 (t, 4 H, CH2), 6.81–7.32 (m,
8 H, Ph), 8.46 (s, 2 H, CH) ppm.
Extraction of the Ligand:
A solution of H2bhbd (360 mg,
Synthesis of H2bhbd: Bis(2-hydroxybenzyl)-1,3-diaminopropane
(H2bhbd) was prepared by condensation of salicylaldehyde and 1,3-
diamine, followed by reduction of the imine bond in methanol with
NaBH4. Yield: 94%. C17H22N2O2 (286.37): calcd. C 71.3, H 7.7, N
1.25 mmol) in acetonitrile (15 mL) was added to a solution of
CuCl2 (410 mg, 3 mmol) in acetonitrile (15 mL). In both solutions,
triethyl orthoformate was added as dehydrating agent (also tri-
methyl orthoformate can be used). After 24 h of reaction, a 10 mL
sample of the reaction mixture was treated with excess aqueous
Na2S (4 equiv.), producing a black precipitate after partial solvent
removal from the CH3CN phase. The solid material was removed
by filtration, and the filtrate was extracted into dichloromethane.
9.8; found C 70.6, H 7.6, N 9.9. ESI-MS: m/z = 286.95. IR: ν =
˜
3291.8, 1595.8, 1456.0, 1386.1, 1256.0, 1100.0, 1027.4, 994.4, 837.2,
1
751.8, 723.3 cm–1. H NMR (300 MHz, CDCl3; Me4Si): δ = 1.79
(m, 2 H, CH2), 2.75 (t, 4 H, CH2), 3.99 (s, 4 H, CH2), 6.75–6.83
(m, 4 H, Ph), 6.98 (d, 2 H, Ph), 7.16 (t, 2 H, Ph) ppm.
1
The dried organic phase was characterized by H NMR and ESI-
Synthesis of [Cu3(bhcbd)2Cl2](CH3CN) (1): A solution of H2bhbd
(360 mg, 1.25 mmol) in acetonitrile (15 mL) was added to a solu-
MS spectroscopy. C17H20Cl2N2O2 (355.26): calcd. C 57.47, H 5.67,
N 7.89, Cl 19.96; found C 48.76, H 4.81, N 6.26, Cl 18.4. The C,
Eur. J. Inorg. Chem. 2011, 4360–4368
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
4367

J. Reedijk et al.
FULL PAPER
[7] D. Kalyani, A. R. Dick, W. Q. Anani, M. S. Sanford, Tetrahe-
dron 2006, 62, 11483–11498.
[8] T. S. Mei, R. Giri, N. Maugel, J. Q. Yu, Angew. Chem. Int. Ed.
2008, 47, 5215–5219.
[9] F. Kakiuchi, T. Kochi, H. Mutsutani, N. Kobayashi, S. Urano,
M. Sato, S. Nishiyama, T. Tanabe, J. Am. Chem. Soc. 2009,
131, 11310–11311.
H, N analyses fit with the formula C18H21Cl2N2O2, ascribed to the
ring-closed product. Moreover, these analyses suggest the presence
of remaining Na2S salts. ESI-MS: 365.01 (in the presence of dehy-
drating agent, 356.04 under dry solvent conditions) IR: ν = 3291.5,
˜
2924.4, 2866.2, 1683.2, 1596.0, 1575.7, 1479.8, 1456.5, 1322.9,
1273.5, 1241.5, 1195.0, 1151.4, 1078.8, 1043.9, 1011.9, 956.7, 866.6,
753.3, 730.0, 671.9, 581.8, 401.7 cm–1. 1H NMR (300 MHz, CDCl3;
Me4Si): δ = 1.92 (m, 2 H, CH2), 3.27 (t, 3 H, CH2), 4.59 (s, 4 H,
CH2), 6.82 (d, 2 H, Ph), 7.20 (d, 2 H, Ph), 7.26 (s, 1 H, Ph) ppm.
[10] D. Kalyani, M. S. Sanford, J. Am. Chem. Soc. 2008, 130, 2150–
2151.
[11] N. D. Ball, M. S. Sanford, J. Am. Chem. Soc. 2009, 131, 3796–
3797.
[12] D. C. Powers, T. Ritter, Nat. Chem. 2009, 1, 302–309.
[13] A. Yahav-Levi, I. Goldberg, A. Vigalok, A. N. Vedernikov, J.
Am. Chem. Soc. 2008, 130, 724–731.
Bromination of H2bhbdi Investigated by UV/Vis Spectroscopy: A
typical experiment is as follows: CuBr2 (0.180 μmol) was added to
a solution of H2bhbdi (2 mL CH3CN, 0.128 μmol, 0.064 mm). The
reaction mixture was monitored by UV/Vis spectroscopy for 2.5 h,
recording one spectrum every 30 s (between 180–1000 nm). This
investigation was carried out with different CuBr2 concentrations
(in the range 0.022–0.112 mm), as well as various concentrations of
H2bhbdi (between 0.016 and 0.096 mm).
[14] C. M. Frech, D. Milstein, J. Am. Chem. Soc. 2006, 128, 12434–
12435.
[15] B. L. Lin, C. R. Clough, G. L. Hillhouse, J. Am. Chem. Soc.
2002, 124, 2890–2891.
[16] A. T. Higgs, P. J. Zinn, S. J. Simmons, M. S. Sanford, Organo-
metallics 2009, 28, 6142–6144.
X-ray Crystallographic Analysis and Data Collection: Crystallo-
graphic data and refinement details are given in Table 1. A single
crystal was selected for the X-ray measurements and mounted to
the glass fiber using the oil drop method,[29] and data were collected
at 173(2) K with a Nonius Kappa CCD diffractometer (Mo-Kα ra-
diation, graphite monochromator, λ = 0.71073 Å). The intensity
data were corrected for Lorentz and polarization effects and for
absorption. The programs COLLECT,[30] SHELXS-97,[31] and
SHELXL-97[32] were used for data reduction, structure solution,
and structure refinement, respectively. The non-hydrogen atoms
were refined anisotropically. The H atoms were geometrically fixed
and allowed to ride on the parent atoms. CCDC-815974 (for 1),
-815975 (for 2), -815976 (for 3), -815972 (for 3a) and -815973
(for 3b) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[17] J. Pons, A. Chadghan, A. Alvarez-Larena, J. F. Piniella, J. Ros,
Inorg. Chem. Commun. 2001, 4, 610–612.
[18]
Y. F. Song, P. Gamez, O. Roubeau, P. Mutikainen, U. Turp-
einen, J. Reedijk, Inorg. Chim. Acta 2005, 358, 109–115.
Y. F. Song, G. A. van Albada, J. Tang, I. Mutikainen, U. Turp-
einen, C. Massera, O. Roubeau, J. S. Costa, P. Gamez, J. Reed-
ijk, Inorg. Chem. 2007, 46, 4944–4950.
[19]
[20] Y. F. Song, P. Gamez, O. Roubeau, M. Lutz, A. L. Spek, J.
Reedijk, Eur. J. Inorg. Chem. 2003, 2924–2928.
[21] E. L. Muetterties, L. J. Guggenberger, J. Am. Chem. Soc. 1974,
96, 1748–1756.
[22] A. W. Addison, T. N. Rao, J. Reedijk, J. Van Rijn, G. C. Ver-
schoor, J. Chem. Soc., Dalton Trans. 1984, 1349–1356.
[23] a) D. Rix, S. Labat, L. Toupet, C. Crevisy, M. Mauduit, Eur.
J. Inorg. Chem. 2009, 1989–1999; b) A. Paczal, A. Kotschy,
Lett. Org. Chem. 2007, 4, 563–566; c) A. V. Malkov, A. J. P.
Stewart-Liddon, F. Teply, L. Kobr, K. W. Muir, D. Haigh, P.
Kocovsky, Tetrahedron 2008, 64, 4011–4025.
[24] a) M. Senkus, J. Am. Chem. Soc. 1946, 68, 1611–1613; b) M.
Nasr, I. Nabih, J. H. Burckhalter, J. Med. Chem. 1978, 21, 295–
298.
[25] a) S. Kobayashi, H. Higashimura, Prog. Polym. Sci. 2003, 28,
1015–1048; b) P. J. Baesjou, W. L. Driessen, G. Challa, J. Reed-
ijk, J. Mol. Catal. A 1999, 140, 241–253; c) S. Gupta, J. A. P. P.
van Dijk, P. Gamez, G. Challa, J. Reedijk, Appl. Catal A: Gene-
ral 2007, 319, 163–170.
Supporting Information (see footnote on the first page of this arti-
cle): Details of the X-ray structures of the compounds and NMR,
UV/Vis and ESI mass spectra.
Acknowledgments
[26] A. Vigalok, B. Rybtchinski, Y. Gozin, T. S. Koblenz, Y. Ben-
David, H. Rozenberg, D. Milstein, J. Am. Chem. Soc. 2003,
125, 15692–15693.
The authors are indebted to the Italian Ministry of Scientific Re-
search for a grant to I. G. to perform this research at the Leiden
Institute of Chemistry. P. G. acknowledges the support of ICREA
(Institució Catalana de Recerca i Estudis Avançats).
[27] a) P. J. Baesjou, W. L. Driessen, G. Challa, J. Reedijk, J. Am.
Chem. Soc. 1997, 119, 12590–12594); b) J. Gao, J. H. Reibensp-
ies, A. E. Martell, Inorg. Chim. Acta 2002, 338, 157–164; c)
W. L. Driessen, P. J. Baesjou, J. E. Bol, H. Kooijman, A. L.
Spek, J. Reedijk, Inorg. Chim. Acta 2001, 324, 16–20; d) F. J.
Viersen, G. Challa, J. Reedijk, Polymer 1990, 31, 1361–1367.
[28] a) G. A. Olah, Acc. Chem. Res. 1971, 4, 240–248; b) G. A.
Olah, S. Kobayash, J. Am. Chem. Soc. 1971, 93, 6964–6967; c)
J. M. Gnaim, R. A. Sheldon, Tetrahedron Lett. 1995, 36, 3893–
3896.
[1] a) P. Kovacic, N. O. Brace, J. Am. Chem. Soc. 1954, 76, 5491–
5494; b) R. Giger, R. Rubinstein, R. Ginsburg, D. Ginsburg,
Tetrahedron 1973, 29, 2393–2393; c) D. I. Robinson, D. C.
Sherrington, C. J. Suckling, Tetrahedron 1984, 40, 785–791.
[2] H. Deng, S. L. Cobb, A. D. Gee, A. Lockhart, L. Martarello,
R. P. McGlinchey, D. O’Hagan, M. Onega, Chem. Commun.
2006, 652–654.
[3] H. Deng, S. L. Cobb, A. R. McEwan, R. P. McGlinchey, J. H.
Naismith, D. O’Hagan, D. A. Robinson, J. B. Spencer, Angew.
Chem. Int. Ed. 2006, 45, 759–762.
[29] T. Kottke, D. Stalke, J. Appl. Crystallogr. 1993, 26, 615–619.
[30] Nonius BV, COLLECT, The Netherlands, Delft, 2002.
[31] G. M. Sheldrick, SHELXS-97 Program for Crystal Structure
Determination, University of Göttingen, Göttingen, Germany,
1997.
[4] L. C. Blasiak, C. L. Drennan, Acc. Chem. Res. 2009, 42, 147–
155.
[5] S. G. Van Lanen, S. Lin, G. P. Horsman, B. Shen, FEMS
Microbiol. Lett. 2009, 300, 237–241.
[6] S. Pandian, M. A. Vincent, I. H. Hillier, N. A. Burton, Dalton
Trans. 2009, 6201–6207.
[32] G. M. Sheldrick, University of Göttingen, Göttingen, Ger-
many, 1997.
Received: May 12, 2011
Published Online: August 23, 2011
4368
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 4360–4368