Selective, catalytic aerobic oxidation of alcohols using CuBr2 and bifunctional triazine-based ligands containing both a bipyridine and a TEMPO group

Article abstract of DOI:10.1039/b820554j

Three novel, bifunctional triazine-based ligands, namely 4-bpyT, 5-bpyT and 6-bpyT, containing both a TEMPO and a bipyridine moiety have been synthesized. These bpy/TEMPO-based molecules have been used as catalyst precursors for the copper-catalyzed aerob

Full text of DOI:10.1039/b820554j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Selective, catalytic aerobic oxidation of alcohols using CuBr2 and bifunctional
triazine-based ligands containing both a bipyridine and a TEMPO group†
Zhengliang Lu,a Tim Ladrak,a Olivier Roubeau,b John van der Toorn,a Simon J. Teat,c Chiara Massera,d
Patrick Gamez*a and Jan Reedijk*a
Received 17th November 2008, Accepted 24th February 2009
First published as an Advance Article on the web 24th March 2009
DOI: 10.1039/b820554j
Three novel, bifunctional triazine-based ligands, namely 4-bpyT, 5-bpyT and 6-bpyT, containing both a
TEMPO and a bipyridine moiety have been synthesized. These bpy/TEMPO-based molecules have
been used as catalyst precursors for the copper-catalyzed aerobic oxidation of alcohols to aldehydes and
ketones, in the presence of tert-BuOK as co-catalyst. The complexes obtained in situ from ligands
4-bpyT and 5-bpyT with copper(II) bromide in a 2:1 acetonitrile/water mixture, selectively catalyze the
aerobic oxidation of primary benzylic, allylic and aliphatic alcohols and secondary benzylic alcohols.
The rate of oxidation achieved using compound 4-bpyT is slightly lower than that of compound 5-bpyT.
Surprisingly, the [copper/6-bpyT] system is not an efficient catalyst. The distinct catalytic behaviour of
the three complexes is most likely due to the different position of the anchoring point of the bipyridine
moiety on the triazine core, thereby inducing dissimilar steric effects. The effect of the substitution
position of the bipyridine unit is reflected by the Vis-NIR spectra of the corresponding copper(II)
complexes, which show similar LMCT and d-d transitions for 4-bpyT and 5-bpyT, while these
absorption bands are significantly red-shifted in the case of the [Cu(II)/6-bpyT] complex. These
differences are indicative of different coordination environments around the CuII centres in those
compounds.
Single-crystal X-ray diffraction studies reveal that [Cu2(4-bpyT)2Br4](CH3CN)7 (6) and
[Cu2(5-bpyT)2Br4](CH3CN)2 (7) are comparable dinuclear compounds with pentacoordinated copper
ions, in a distorted square-pyramidal geometry in 6 and in a distorted trigonal-bipyramidal geometry in
7. These two coordination geometries are also reflected by their slightly different Vis-NIR results.
Cu(6-bpyT)Br2 (8) is mononuclear, with the CuII ion in a distorted tetrahedral geometry, suggesting a
relationship with its catalytic inactivity.
these oxidants are often expensive and tend to produce copious
Introduction
amounts of wastes; therefore they do not satisfy the requirements
The selective oxidation of primary and secondary alcohols to,
of no pollution and good economy. From the point of view of green
respectively, aldehydes and ketones, without over-oxidation, is
chemistry, it is stringent to search for new environmentally friendly,
one of the most significant synthetic transformations in research
low-cost and clean redox systems for this type of transformation.
laboratories and industry.1–4 Many well-established methods to
Moreover, these “green” catalysts and/or “green” methodologies
carry out this reaction still involve the use of stoichiometric
should be applicable to a wide range of alcohols and be highly
amounts of inorganic oxidants, such as chromium(VI) oxide,1
active and selective. Systems based on organic nitroxyl radi-
KMnO4, SeO2 or hypervalent iodine compounds,5,6 or organic
cals, such as 2,2,6,6-tetramethyl-piperidyl-1-oxy (TEMPO) have
oxidants (like in the Moffat–Swern oxidation7–9). Unfortunately,
shown high efficiencies and selectivities in many cases of alcohol
oxidation.10,11 Thus, the Montanari–Anelli oxidation reported
aLeiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University,
P. O. Box 9502, 2300 RA Leiden, The Netherlands
in 198712 is a versatile and efficient catalytic procedure, which
proceeds under mild conditions to convert both primary and
secondary alcohols to their corresponding carbonyl compounds in
high yields, even in large-scale operations. However, this catalytic
system is ecologically not attractive due to the use of buffered
bleach solutions as the terminal oxidant and the use of bromide
ions as co-catalyst. From an economical and environmental point
of view, the use of dioxygen would be ideal, because it is cheap,
readily available and “green”; in addition, in this case water is the
only by-product.
bUniversite´ Bordeaux 1, CNRS-CRPP, 115, avenue du Dr Schweitzer, 33600
Pessac, France
cALS, Berkeley Lab, 1 Cyclotron Rd, MS2-400, Berkeley, CA 94720, USA
dDipartimento di Chimica Generale ed Inorganica, Chimica Analitica,
Chimica Fisica, Universita` degli Studi di Parma, Viale G. Usberti 17/A,
43100 Parma, Italy
† Electronic supplementary information (ESI) available: Crystal data and
structure refinement for compounds 5a and 5b and the ligands 4-bpyT
and 6-bpyT, as well as the ORTEP views of the molecular structures
of 5a, 5b, 4-bpyT and 6-bpyT. X-Ray crystallographic analysis and data
collection for 5a, 5b, 4-bpyT and 6-bpyT. Details of visible spectroscopic
study of the Cu systems. CCDC reference numbers 702598–702604. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/b820554j
Since this pioneering work, a great deal of attention has
been paid to the development of TEMPO-containing catalytic
systems. The low toxicity of this radical reagent and the good
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3559–3570 | 3559
©

View Article Online
chemoselectivities achievable have led to the development of a
variety of homogeneous and heterogeneous catalytic, TEMPO-
based systems for the oxidation of alcohols using O2 or air
as a terminal co-oxidant.11–19 Most of the catalytic systems
so far reported, including a number of immobilized TEMPO
variants on both organic and inorganic supports, as well as
ionic liquids, use molecular oxygen as the terminal oxidant
and transition metal ions as co-catalysts (mainly copper, iron,
palladium and ruthenium).19–37 For example, a three-component
TEMPO/FeCl3/NaNO2 catalytic system38 was recently reported,
which is capable of catalyzing the aerobic oxidation of various “ac-
tivated” primary and secondary alcohols in acetonitrile under mild
conditions. In fact the catalyst is not very efficient with less reactive
alkyl alcohols, such as 2-octanol. A four-component system27
consisting of acetamido-TEMPO/Cu(ClO4)2/TMDP/DABCO
dine/pyrazole ligand and a TEMPO unit, CuBr2 and potassium
tert-butoxide for the aerobic oxidation of benzylic and aliphatic
primary alcohols to the corresponding aldehydes under mild
conditions at room temperature in acetonitrile/water.42 Because
earlier studies showed that bpy ligands lead to efficient copper/bpy
catalysts,16 three bifunctional triazine molecules holding bipyri-
dine ligands anchored at different positions were designed and
synthesized from 2,4,6-trichloro-1,3,5-triazine (1). The chlorine
atoms of 1 can be easily and selectively substituted by any
nucleophilic reagent, through the control of the substitution
temperature.44 Therefore, the synthesis of triazine derivatives
bearing three different substituents is possible. The first step
toward the preparation of triazine derivatives holding a bpy unit is
to design and synthesize bipyridines bearing a side arm containing
a functional group, which will be used to connect the N,N-ligand
to the triazine core.
(TMDP
=
4,4¢trimethylenedipyridine; DABCO
=
1,4-
diazabicyclo[2.2.2]octane) was also described for the room-
temperature aerobic oxidation of alcohols. Most systems
reported function well in organic solvents, such as DMSO
(dimethyl sulfoxide) or toluene, probably as the result of a
better stabilisation of the transition-metal catalyst which is
often water-sensitive. Nevertheless, water is an economically
and ecologically more desirable solvent and studies devoted to
developing water-tolerant procedures deserve more attention. Up
until now, only a few systems in pure water or mixtures of various
organic solvents and water have been published.15,16,39,40
Thus, 4-methyl-4¢-hydroxyethyl-2,2¢-bipyridine 3a can be sim-
ply synthesized from commercially available 4,4¢-dimethyl-2,2¢-
bipyridine 2a in a yield of 49% (Scheme 1A), applying a literature
procedure45 with slight modifications (see Experimental section).
5-Methyl-5¢-hydroxyethyl-2,2¢-bipyridine 3b is analogously pre-
pared in 47% yield following the reported synthetic method,
from commercially available 5,5¢-dimethyl-2,2¢-bipyridine 2b. The
hydroxyethyl derivatives 3a and 3b are obtained by lithiation of,
respectively, 2a and 2b using freshly prepared LDA (lithium diiso-
propylamide obtained in situ from n-butyllithium and dry amine at
-78 ◦C), followed by reaction with an excess of paraformaldehyde.
6-Hydroxyethyl-2,2¢-bipyridine 3c is produced in 49% yield from
6-methyl-2,2¢-bipyridine 2c which can be obtained by methylation
of◦commercially available 2,2¢-bipyridine with methyllithium at
A few years ago, some of us reported a [copper(2,2¢-
bipyridine)/TEMPO]-based catalytic system15,16 capable of selec-
tively converting primary alcohols to the corresponding aldehydes
at room temperature in acetonitrile/water. Recently, this efficient
catalytic system with bipyridine ligands has been extended to the
use of mixed pyrazole/pyridine ligands.41 Lately, we reported a
new catalytic system42 based on 1,3,5-triazine derivatives hold-
ing both a pyrazole/pyridine ligand (to coordinate copper(II)
ions) and a TEMPO moiety. These bifunctional molecules con-
taining simultaneously the copper(II) catalyst and the radical-
containing unit were found to be able to perform the mild
aerobic oxidation of primary alcohols to the corresponding
aldehyde-based products at room temperature. In the present
study, this synthetic strategy has been further developed and
different bipyridine (bpy) ligands (in fact, bipyridine ligands
have proven to be prime ligands for this reaction16) have been
anchored to a triazine ring containing a TEMPO unit. Thus,
the effect on the catalytic activity of the anchoring point of the
bipyridine ligand (4-, 5- and 6-positions of the pyridine ring) to
the triazine core has been investigated. Three copper(II) complexes
from the different bpy/TEMPO ligands have been obtained and
structurally characterized. Their significant structural disparities
are correlated with their drastically distinct catalytic behaviour.
Another interesting question deals with whether or not the
TEMPO group interacts with Cu(II), as has been reported for
semiquinone radicals.43
0
C.46 Under these circumstances the second 6-position of the
bipyridine ligand is not methylated, therefore preventing potential
steric interactions upon coordination of the N2-donor to copper.
The first chloride of 1 was “neutralized” by substitution reaction
with diphenylamine, producing compound 4 (Scheme 1A). In fact,
only two positions of the triazine ring are required in the present
study, to attach two functional groups (a bipyridine unit and a
TEMPO moiety). The third position can be used at a later stage
to introduce an additional functional group, or to anchor the
Cu/TEMPO catalyst on a solid support.
Next, the hydroxyethyl-bipyridines 3a-c have been deprotonated
using NaH in dry THF and subsequently connected to compound
4 by substitution of a second chloride, generating compounds 5a–c
in good to excellent yields (see Experimental section). Finally,
commercially available 4-amino-2,2,6,6-tetramethylpiperidine-1-
oxyl (4-amino TEMPO) has been connected to the intermediates
5a–c by substitution of the last chloride, in the presence of N,N-
diisopropylethylamine (DiPEA) under reflux in THF (Scheme 1B).
The final TEMPO-containing bipyridine ligands 4-bpyT, 5-bpyT
and 6-bpyT have been obtained in reasonable to excellent yields
(see Experimental section).
Structural characterization of copper(II) coordination compounds
Results and discussion
The reactions of CuBr2 with the ligands 4-bpyT, 5-bpyT and
6-bpyT in acetonitrile produce compounds 6, 7 and 8, respec-
tively, whose solid-state structures have been determined by
X-ray diffraction studies. Crystallographic data collection and
structure refinement details for compounds 6–8 are summarized in
Synthesis of the 1,3,5-triazine-based bipyridine/TEMPO ligands
Recently, some of us reported on a catalytic system which
comprised a bifunctional triazine derivative bearing a mixed pyri-
3560 | Dalton Trans., 2009, 3559–3570
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Scheme 1 Synthetic pathway for the preparation of the TEMPO/bipyridine ligands 4-bpyT, 5-bpyT and 6-bpyT.
˚
atom O1 (Cu1 ◊ ◊ ◊ O1 = 2.811(2) A) from the ethoxy bipyridine
Table 1, and selected interatomic distances and angles are listed in
Table 2.
arm belonging to the triazine-coordinated 4-bpyT ligand. The
intramolecular Cu ◊ ◊ ◊ Cu separation is 7.2470(8) A, which is close
to the shortest intermolecular Cu ◊ ◊ ◊ Cu distance of 7.9360(8)
A. In the crystal lattice, the dicopper units are linked through
intermolecular hydrogen bonding interactions, generating a two-
dimensional supramolecular framework.
X-Ray diffraction studies for [Cu2(5-bpyT)2Br4](CH3CN)2 (7)
show that its solid-state structure resembles that of coordination
compound 6 (see Fig. 1 and 2). However, the divalent Cu1 ions are
in a distorted trigonal-bipyramidal coordination geometry, while
the copper(II) ions in 6 exhibit a square pyramidal coordination
geometry (see above). The trigonal bipyramid (t5 = 0.802)47 is
composed of two bromide ions (Br1 and Br2), one nitrogen
atom (N1) from a triazine ring and two nitrogen atoms (N12A
and N17A) from a bipyridine moiety belonging to an adjacent
ligand molecule. The equatorial plane is defined by the atoms
Br1, N1 and N12A which are bonded to the Cu1 centre with
˚
Crystallographic structure analysis of [Cu2(4-bpyT)2Br4]-
(CH3CN)7 (6) indicates that the centrosymmetric, neutral din-
uclear compound presents two crystallographically equivalent
bridged copper(II) centres and two ligand molecules (Fig. 1).
The asymmetric unit of 6 thus contains one Cu ion and one
4-bpyT ligand. The CuN3Br2 chromophore is formed by one
triazine N atom (N1), two bromide anions (Br1 and Br2) and
two N atoms (N5A and N6A) from a bipyridine moiety belonging
to an adjacent ligand molecule, resulting in a distorted penta-
coordinated environment around the Cu1 centre. Cu1 exhibits a
distorted square-pyramidal geometry characterized by a t5 value47
˚
˚
of 0.022. The axial Cu1–Br1 bond length of 2.7152(4) A is much
˚
longer than that of the in-plane Cu1–Br2 bond of 2.4363(5) A.
The plane of the square pyramid is completed by the nitrogen
atoms N1, N5A and N6A with Cu–N bond distances of 2.041(2),
˚
2.021(2) and 2.018(2) A, respectively. All Cu–N and Cu–Br bond
˚
distances of 2.5548(5), 2.136(2), and 2.064(2) A, respectively.
The axial ligands are Br2 (Cu1–Br2 = 2.3870(5) A) and N17A
lengths are in the normal range for this type of coordination
environment.48 In addition, Cu1 is semi-coordinated by the oxygen
˚
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3559–3570 | 3561
©

View Article Online
Table 1 Crystal data and structure refinement for compounds 6, 7 and 8
Compound 6
Compound 7
Compound 8
Formula
C88H103Br4Cu2N23O4
1993.65
0.45 ¥ 0.05 ¥ 0.02
150(2)
Triclinic, P-1
Green
9.5728(8)
10.7767(9)
23.002(2)
87.355(2)
84.576(2)
83.800(2)
2347.0(3)
1
1.411
1022
2.745
2.79–33.70
34487(0.0437)
14014/568
1.040
0.0454/0.1206
0.0676/0.1308
1.138/-0.576
C78H88Br4Cu2N18O4
1788.38
0.35 ¥ 0.15 ¥ 0.10
150(2)
Triclinic, P-1
Orange
10.2403(4)
13.9158(5)
16.3583(6)
69.404(2)
71.830(2)
75.030(3)
2044.14(13)
1
1.453
912
3.139
2.60–34.34
28135(0.0513)
12116/479
1.079
0.0627/0.1869
0.0705/0.1938
1.627/-2.225
C36H39Br2CuN8O2
839.11
0.14 ¥ 0.09 ¥ 0.01
150(2)
Triclinic, P-1
Brown
9.7023(11)
10.1421(11)
20.239(2)
98.255(2)
95.249(2)
113.578(2)
1781.9(3)
1
1.564
852
3.594
2.44–33.64
37239(0.0468)
10609/447
1.024
0.0368/0.0788
0.0510/0.0842
0.821/-0.501
Fw (g mol-1)
Crystal size (mm3)
Temperature (K)
Crystal system, space group
Crystal color
˚
a (A)
˚
b (A)
˚
c (A)
a (deg)
b (deg)
g (deg)
3
˚
Volume (A )
Z
Calcd density (g cm-3)
F(000)
Absorption coefficient (mm-1)
q for data collection (deg)
Reflections collected (Rint
Data/params
)
Goodness of fit on F2
R1 (wR2) [I > 2s(I)]
R1 (wR2) (all data)
3
˚
Largest diff. peak/hole (e A )
◦
˚
Table 2 Selected bond lengths (A) and bond angles ( ) for compounds 6,
7 and 8
Compound 6
Bond lengths
Cu1–N1
2.041(2)
2.021(2)
2.018(2)
Cu1–Br1
Cu1–Br2
2.7152(4)
2.4363(5)
Cu1–N5A
Cu1–N6A
Bond angles
N1–Cu1–N5A
N1–Cu1–N6A
N5A–Cu1–N6A
N1–Cu1–Br1
N1–Cu1–Br2
Compound 7
Bond lengths
Cu1–N1
Cu1–N12A
Cu1–N17A
Bond angles
N1–Cu1–N12A
N1–Cu1–N17A
N12A–Cu1–N17A
N1–Cu1–Br1
N1–Cu1–Br2
Compound 8
Bond lengths
Cu1–N5
163.21(9)
93.80(9)
79.94(9)
105.59(6)
86.88(7)
N5A–Cu1–Br1
N5A–Cu1–Br2
N6A–Cu1–Br1
N6A–Cu1–Br2
Br1–Cu1–Br2
90.06(6)
95.06(7)
89.88(7)
164.51(7)
104.871(16)
2.136(2)
2.064(2)
2.013(2)
Cu1–Br1
Cu1–Br2
2.5548(5)
2.3870(5)
Fig. 1 ORTEP drawing of compound 6. Thermal ellipsoids are drawn at
the 50% probability level. All H atoms and lattice solvent molecules are
omitted for clarity.
125.56(8)
88.94(8)
79.67(8)
107.60(5)
91.24(6)
N12A–Cu1–Br1
N12A–Cu1–Br2
N17A–Cu1–Br1
N17A–Cu1 Br2
Br1–Cu1–Br2
125.47(6)
95.13(6)
90.97(6)
173.70(6)
94.98(2)
2.015(2)
2.005(2)
Cu1–Br1
Cu1–Br2
2.3791(5)
2.3521(5)
Cu1–N6
Bond angles
N5–Cu1–N6
N5–Cu1–Br1
N5–Cu1–Br2
81.93(9)
95.65(6)
160.89(7)
N6–Cu1–Br1
N6–Cu1–Br2
Br1–Cu1–Br2
138.00(7)
99.59(7)
95.713(17)
Fig. 2 ORTEP drawing of compound 7. Thermal ellipsoids are drawn at
the 50% probability level. All H atoms and lattice solvent molecules are
omitted for clarity.
˚
(Cu1–N17A = 2.013(2) A). The distortion of the coordination
geometry is confirmed by the in-plane angles ranging from
79.67(8) to 125.47(6)◦. The average Cu–N and Cu–Br bond
distances are comparable with those reported in the literature.49
˚
The intramolecular Cu ◊ ◊ ◊ Cu separation is 8.2281(6) A, while
˚
The O7 ◊ ◊ ◊ Cu1 distance of 2.913(2) A is indicative of a weak,
the shortest intermolecular Cu ◊ ◊ ◊ Cu distance is even shorter, i.e.
semi-coordinating interaction between the Cu and O atoms.50
7.044(6) A. As for 6, an intermolecular hydrogen bond network
˚
3562 | Dalton Trans., 2009, 3559–3570
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
links the dinuclear molecules, resulting in a two-dimensional
supramolecular framework.
spectra of the copper complexes in the solid state and in solution,
which moreover show drastically distinct catalytic activities (see
section on Catalytic activities).
As shown in Fig. 3 [Cu(6-bpyT)Br2] (8) is a mononuclear com-
pound, in contrast to compounds 6 and 7, which are dinuclear. The
Cu1 centre is coordinated to two bromide atoms (Br1 and Br2) and
two nitrogen atoms from a chelating bipyridine moiety (N5 and
N6), generating a distorted tetrahedral coordination environment
with a dihedral angle of 69.7◦ and further characterized by a
t4 value of 0.433.51 The Cu–Br bond lengths of 2.3791(5) and
Ligand field, EPR and mass spectrometry studies
Comparative UV-visible-NIR spectroscopy studies of the cop-
per(II) complexes obtained in situ from the ligands 4-bpyT,
5-bpyT and 6-bpyT and CuBr2 in acetonitrile solution and in
the solid state (diffuse reflectance) have been carried out (see the
Electronic Supplementary Information (ESI) for the experimental
procedures). A selection of the spectra recorded is depicted in
Fig. 4. The spectra of species obtained from 1:1 mixtures of CuII:4-
bpyT and CuII:5-bpyT in acetonitrile show similar features and
both differ from those of CuII:6-bpyT. In the solid state the d-
d bands differ from the solutions, which is due to the fact that
different chromophores are usually present in the solution. The
solid state LF spectra do agree with the 3D structures, with
compound 6 (maximum at about 690 nm) indicative of square
pyramidal, compound 7 (maxima at 790 nm, with shoulders at
lower energy) indicative of trigonal bipyramidal geometry, and
compound 8 (maxima at 810, 920 and 1020 nm) indicative of
distorted tetrahedral geometry.50,53 The solid state EPR spectra
have been recorded, but do not provide detailed information, due
to exchange narrowing in the solid state, where the Cu(II) ions
are too close to resolve hyperfine or even g aniosotropy. So in
most cases only signals at g = 2.05 are observed, with a shoulder
near g = 2.10–2.15.
˚
2.3521(5) A are slightly longer than those found in related crystal
structures reported earlier.52 However, the average Cu–N bond
˚
length of 2.010 A is comparable with the distances found in
Cu compounds containing chelating bipyridine molecules. The
˚
intermolecular Cu ◊ ◊ ◊ Cu separation is 4.297(3) A, which is much
shorter than the distances observed in compounds 6 and 7 (see
above). In fact two adjacent mononuclear species are weakly
connected through quite long, semi-coordinating Cu ◊ ◊ ◊ Br bonds
˚
(Cu1–Br2 d = 3.4510(7) A; symmetry operation, d = 1 - x,
1 - y, -z), producing a dimer. Similar to compound 7, intra
and intermolecular hydrogen bonding interactions are present
in the crystal lattice, which link the pseudo-dimers to generate
two-dimensional supramolecular sheets. Those adjacent layers are
stacked by intermolecular p ◊ ◊ ◊ p interactions between bipyridine
˚
moieties (with a separation distance of 3.6296(17) A) to produce
a three-dimensional supramolecular framework.
Fig. 3 ORTEP drawing of compound 8. Thermal ellipsoids are drawn at
the 50% probability level. All H atoms and lattice solvent molecules are
omitted for clarity.
As mentioned above, the crystal structure of compound 8
is significantly different from those of compounds 6 and 7. It
appears that the different position of the anchoring point on
the bipyridine ligand to the ethyl spacer (connecting the bipyri-
dine unit to the triazine ring) plays an important, most likely steric,
role in the resulting solid-state structure of the corresponding
coordination compound. Thus, compound 6 exhibits a square
pyramidal coordination geometry, whereas the environment of 7
is trigonal bipyramidal. In 8, copper(II) is tetracoordinated, most
likely because the coordination of a fifth ligand is too sterically
demanding, even though a bromide ion is located at a distance
Fig. 4 Vis-NIR spectra for 1:1 acetonitrile solutions of Cu:ligand (black
curves) and for 1:1:1 acetonitrile solutions of Cu:ligand:added base (red
curves) (A: 4-bpyT; B:5-bpyT; C: 6-bpyT). For C (compound 8), no effect
of base is observed.
More useful information could be obtained from the solution
visible-UV spectra (see Fig. 4). The absorption band centred near
500 nm in solution is characteristic of a ligand-to-metal charge
transfer (LMCT) from a bromide to a CuII centre.48,53–55 With the
ligand 4-bpyT, this band is only seen as a shoulder at 490 nm;
the d-d absorption band at 655 nm is in agreement with d-d
transitions of square-pyramidal CuII complexes. With the ligand
5-bpyT, the LMCT band is resolved at 500 nm, and slightly shifted
to a longer wavelength (by comparison with the ligand 4-bpyT),
˚
of 3.4510(7) A from the metal centre; see above. Interestingly,
these steric effects (characterizing the ligands 4-bpyT, 5-bpyT and
6-bpyT) induce clear differences in the corresponding Vis-NIR
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3559–3570 | 3563
©

View Article Online
Table 3 [Copper/5-bpyT]-catalyzed aerobic oxidation of benzyl alcohol
the d-d band agrees with 5-coordinated Cu, most likely closest to
square pyramidal. With the ligand 6-bpyT, the strong absorption
band at 535 nm is assigned as the LMCT band, and the bands
near 785 nm and 1000 nm are assigned to the distorted tetrahedral
CuII centre, just like in the solid state. As is evidenced in Fig. 4,
the LMCT band is red-shifted (from 4-bpyT to 6-bpyT), as a result
of the change in coordination geometry around the CuII ion, but
perhaps also due to the change in the Cu–Br bond lengths. As
the Cu–Br bond distance becomes shorter, the electrons are more
delocalized around the coordinated bromide ion; therefore, a lower
energy is required for the LMCT transition, giving rise to a shifting
of the corresponding band to a longer wavelength. The average
Cu–Br bond lengths are 2.5757 for 4-bpyT (compound 6), 2.4709
to benzaldehydea
Entry Benzyl alcohol
5-bpyT
tert-BuOK
CuBr2
Conv (%)b
1
2
3
4
5
6
7
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
Yes
No
No
Yes
Yes
No
Yes
No
Yes
No
Yes
No
Yes
Yes
No
No
Yes
No
Yes
Yes
Yes
—
—
—
—
15%
—
62%
a For the reaction procedure, see Experimental section. b After 2 h reaction
time.
˚
for 5-bpyT (compound 7) and 2.3656 A for 6-bpyT (compound 8).
This sequence is most likely influenced by the position (namely the
4-, 5- and 6-positions on the aromatic ring) of the anchoring point
(leading to distinct steric hindrance) on the bipyridine moieties.
Upon addition of one equivalent of potassium tert-butoxide
(tert-BuOK), the d-d transition is not altered for compound 8,
suggesting no change in the chromophore. However, the initial
absorption band at 655 nm for both compounds 6 and 7 is shifted
to a lower energy, resulting in a broad absorption centred around
730 nm. This shifting indicates that those transitions occur more
easily than without base present, which is consistent with the
fact that the catalytic activities of the copper complexes with
the ligands 4-bpyT and 5-bpyT are accelerated with addition
of base while no difference is observed for the ligand 6-bpyT
(see section on Catalytic activities below). These solution studies
suggest that the initial species formed in situ from CuBr2 and
the corresponding ligand, and whose catalytic properties are
studied, exhibit coordination geometries comparable to those of
the crystallographically characterized compounds. Actually, mass
spectrometry studies of the copper(II) coordination compounds
6, 7 and 8 dissolved in acetonitrile reveal that mononuclear
species are present in solution for all three samples (see Figures
S5–S7, ESI). It thus appears that the triazine-bridged dinuclear
compounds 6 and 7 dissociate in solution (in fact, triazine N
donors are known to be very weak ligands for CuII) to form a
mononuclear species, similar to that of 8. Therefore, it can be
reasonably considered that mononuclear species are involved as
catalyst precursors in the copper-mediated oxidation of alcoholic
substrates.
Scheme 2 [Cu/bpyT/base]-catalysed aerobic oxidation of alcohols.
Only 15% of benzaldehyde is produced in 2 h without base (Table 3,
entry 5). When one equivalent of base is added, the conversion
significantly increases to 62% after 2 h (Table 2, entry 7). The role
of the base is probably to deprotonate the alcohol, and thus, favour
its coordination (as an alcoholate) to the copper species, resulting
in enhancement of the activity.15 The changes in the LF spectra,
vide supra, support this. At room temperature, benzyl alcohol is
converted to 42% benzaldehyde, whereas 100% conversion can
be reached with the system CuBr2/bpy/TEMPO reported earlier
after a reaction time of 2.5 h16 (Table 4, entries 1 and 2). The present
lower activity is probably due to both the lack of flexibility of the
TEMPO unit (anchored on the 1,3,5-triazine ring) and the short
spacer linking the bipyridine moiety to the triazine core. Most
likely the copper/bipyridine complex cannot interact through
space with an intramolecular TEMPO group. Therefore, the design
and preparation of bifunctional triazine derivatives with longer
spacers connecting the bipyridine ligand and the TEMPO unit
on the triazine ring is currently being investigated. Such strategy
should favour the intramolecular [Cu/bpy] ◊ ◊ ◊ TEMPO interac-
tion (Scheme 3) and lessen the intermolecular one. At 50 ◦C, the
improvement of the oxidation of benzyl alcohol is not significant
(Table 4, entries 1 and 3). The use of 1.67 mol% of catalyst instead
of 5 mol% results in a significant lessening of catalytic activity.
Indeed, after 3 hours at 50 ◦C, a conversion of 26% is reached,
whereas 69% of benzaldehyde is obtained in the presence of
5 mol% of catalyst. Besides benzyl alcohol (an activated alcohol),
1-octanol (a non-activated alcohol) has also been used as a second
model substrate to examine the efficiency of the catalytic system.
[Copper/5-bpyT] smoothly and selectively oxidizes 1-octanol into
1-octanal with a conversion of 61% after 6 h at 50 ◦C (Table 4,
entry 5). In the case of 1-octanol, the increase of the reaction
temperature results in an increased conversion; indeed, 50% of 1-
octanal is produced at 50 ◦C, while only 25% of product is detected
at 25 ◦C, after a reaction time of 3 h (Table 4, entries 4 and 5).
Next, the catalytic system CuBr2/5-bpyT/tert-BuOK has been
evaluated in the oxidation of a series of primary and secondary
alcohols, under optimized conditions (Table 4). As is evidenced
Room temperature EPR spectra of the solutions were not very
informative, although in some cases a clear signal of the TEMPO
radical could be observed (g = 2.004, A = 15 Gauss).
Catalytic activities
As described in earlier reports, the use of CuBr2 in combination
with TEMPO can be efficiently applied for the catalytic oxidation
of various alcohols.16,41 The ligand 5-bpyT has been primarily
tested in the copper-catalyzed aerobic oxidation of two model sub-
strates, i.e. benzyl alcohol and 1-octanol. The catalytic reactions
were performed applying the optimized experimental conditions
described earlier,15 namely using 5 mol% of CuBr2 together with
5 mol% of the ligand and 5 mol% of tert-BuOK (as basic co-
catalyst) in acetonitrile/water (2:1) (Scheme 2). The results are
listed in Table 3. In the absence of 5-bpyT or CuBr2, no reaction oc-
curs (Table 3, entries 1–4 and 6) using benzyl alcohol as a substrate.
3564 | Dalton Trans., 2009, 3559–3570
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 4 Oxidation of selected alcohols to corresponding aldehydes and
ketones with the copper/5-bpyT/base system
Entry Substrate
Temp (◦C) Time (h) Conversion (%)a
1
Benzyl alcohol
25
1.5
10
2.5
1
42
100
100
36
2
3
Benzyl alcohol15
Benzyl alcoholb
25
50
2
3
55
69
5
10
3
24
3
88
100
25
59
50
4
5
1-Octanol
1-Octanol
25
50
6
61
6
7
8
Geraniol
Crotyl alcohol
1-Phenylethanol
50
50
50
4
3
1
2
94
100
23
40
Scheme 3 Expected copper active species using the bifunctional tri-
azine-based ligands.
3
7
57
94
9
10
11
12
Diphenyl methanol
2-Octanol
Cyclohexanol
Benzyl alcohol 1/
1-Phenylethanol 1
50
50
50
50
10
5
5
<1
—
—
85/4
products could be detected by gas chromatography. Surprisingly,
the CuBr2/5-bpyT mixture can efficiently and selectively catalyze
even the oxidation of the secondary alcohol, 1-phenylethanol to
94% acetophenone after a reaction time of 7 hours (entry 8).
In earlier reports where TEMPO and bipyridine molecules were
used as separate catalytic precursors,15,33 the resulting catalytic
mixture (i.e. CuBr2/bpy/TEMPO) was not able to mediate such a
1-phenylethanol to acetophenone transformation. This difference
in chemoselectivity between a priori analogous systems suggests
that different mechanisms of action are involved. Further studies
(including theoretical investigations) are required to explain this
unexpected distinction. The oxidation of diphenyl methanol,
an activated alcohol, cannot be achieved (entry 9). The steric
hindrance due to the two phenyl groups apparently hinders the
coordination of the substrate to the copper centre, which is crucial
for the C–H abstraction from the alcoholate by the coordinated
TEMPO molecule. Octan-2-ol is not converted to octan-2-one
by the present system (entry 10). This result may be justified by
the very low reactivity of octan-2-ol compared to benzyl alcohol
or octan-1-ol. This hypothesis is corroborated by the use of
cyclohexanol (another non-activated secondary alcohol as well)
which is also not converted to cyclohexanone (entry 11). These
results clearly indicate that, for secondary alcohols, benzylic or
allylic activation of the substrate is necessary to observe an efficient
catalytic oxidation. Longer reaction times as well as elevated
temperature are required to oxidize aliphatic primary alcohols,
which has also been observed with related systems.56
1
2
3
1
2
3
1
100/13
100/24
2/–
13
14
Benzyl alcohol 1/
1-Phenylethanol 0.1
50
50
2.5/–
8.5/–
52
Benzyl alcohol 1/
acetophenone 0.1
2
3
81
94
a Selectivity always >99% based on GC. b The oxidation of benzyl alcohol
performed with 1.67 mol% of catalyst produces only 26% of benzaldehyde
after 3 hours.
The oxidation of secondary alcohols is usually easier than
that of primary alcohols, owing to an increased electron density
on the former. However, the catalytic system herein described
displays comparable conversion rates for benzyl alcohol and 1-
phenylethanol (the conversion rate is actually slightly lower for
1-phenylethanol; see Table 4, entries 3 and 8). A competitive
oxidation experiment with a 1:1 mixture of benzyl alcohol and
1-phenylethanol has then been carried out to investigate the
selectivity of the catalyst towards primary and secondary alcohols.
After a reaction time of 1 h, 85% of benzyl alcohol had been
converted to benzaldehyde, while only 4% of acetophenone was
found to be produced (entry 12). This experiment illustrates the
good selectivity of the catalytic system towards primary alcohols.
Fig. 5 Oxidation of model substrates in the presence of the ligands
4-bpyT and 5-bpyT with CuBr2 and base in CH3CN/H2O 2:1 (blue lines:
4-bpyT; red lines: 5-bpyT; full squares: benzyl alcohol at RT; full triangles:
1-octanol at RT; full circles: 1-phenylethanol at 50 ◦C; green asterisks:
benzyl alcohol at 50 ◦C).
in Table 4, various primary benzylic, allylic and aliphatic alcohols
have been successfully oxidized. Allylic alcohols, such as crotyl
alcohol and geraniol (entries 6 and 7) were selectively converted
to the corresponding aldehydes in excellent yields and no by-
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3559–3570 | 3565
©

View Article Online
Surprisingly, the conversion of benzyl alcohol is significantly faster
in the presence of 1-phenylethanol (compare entries 3 and 12
in Table 4). In contrast, the conversion of 1-phenylethanol is
drastically reduced (compare entries 8 and 12 in Table 4). This
rate enhancement may be explained by the involvement of 1-
phenylethanol and/or tiny amounts of acetophenone to generate
active species capable of mediating the aerobic oxidation of benzyl
alcohol. Thus, after a reaction time of 2 h, 13% conversion
of 1-phenylethanol is detected, while benzyl alcohol is already
fully oxidized. Even after 3 h, only 24% of acetophenone is
produced, whereas 57% of 1-phenylethanol is converted to the
carbonyl product in the absence of benzyl alcohol (compare
entries 8 and 12). To investigate the influence of 1-phenylethanol
or acetophenone on the catalytic oxidation of benzyl alcohol,
experiments have been carried out in the presence of these two
compounds (see entries 13 and 14 in Table 4). Remarkably,
the oxidation of benzyl alcohol with a catalytic amount of 1-
phenyl ethanol is drastically hampered (entry 13). Actually, after
a reaction time of 3 hours, only 8.5% of benzaldehyde is formed,
while the conversion is 69% when pure benzyl alcohol is used (entry
3). In contrast, the presence of 10% of acetophenone results in
an enhancement of production of benzaldehyde (94% conversion
instead of 69%; compare entries 3 and 14). It thus appears that the
presence of acetophenone leads to the formation of catalytically
active species. On the contrary, 10% of 1-phenylethanol generates
a poorly active catalyst. Apparently, a significant quantity of
1-phenylethanol is required to produce a sufficient amount of
acetophenone that gives rise to an effective catalyst (compare
entries 12 and 13). Further investigations (beyond the scope of
the present study) are necessary to appraise the nature of these
active species.
as well as secondary benzylic alcohols to the corresponding
aldehydes and ketones with good to excellent yields under mild
conditions in acetonitrile/water (1:1). However, the copper(II)
complex obtained from the ligand 6-bpyT and CuBr2 under the
same conditions used for 4-bpyT and 5-bpyT, does not catalyze
the aerobic oxidation of any of the common alcohols. This drastic
difference is likely to be related to the steric effects originating from
the different position of the anchoring point of the bpy ligand on
the triazine ring; we cannot exclude that it may perhaps also simply
be due to complete precipitation of any active catalyst in the case
of 6-bpyT. The influence of this anchoring position is also reflected
by the Vis-NIR spectra of the three copper complexes in solution
and in the solid state. Indeed, the copper coordination compounds
6 and 7 containing 4-bpyT and 5-bpyT respectively, exhibit similar
LMCT and d-d transition absorption bands, thus indicating
that both complexes have comparable coordination environments
around the corresponding copper centres. In contrast, compound
8, obtained from the ligand 6-bpyT, shows a significantly different
Vis-NIR spectrum (both in the solid state and in solution)
suggesting a drastically distinct environment around the metal
centre. This result is confirmed by the X-ray crystal structures
of the three solid compounds, which reveal that 6 and 7 are
pentacoordinated while 8 is tetracoordinated in the solid state. The
addition of base (i.e. tert-BuOK, used for efficient catalysis) leads
to a red-shift of the d-d transition for compounds 6 and 7, whereas
the ligand-field spectrum of 8 remains unaffected. The mechanism
of the catalytic CuBr2/TEMPO is currently (also computationally)
being investigated.
Experimental
In order to determine the efficiency of the catalysts obtained
from the different bpy/TEMPO ligands, a number of substrates
have been tested. Similar conversions have been achieved with
the ligands 4-bpyT (Table S1) and 5-bpyT; however, the reaction
rate observed with 4-bpyT is slightly lower than that of 5-bpyT.
Surprisingly, the use of 6-bpyT as a bpy/TEMPO ligand does not
lead to an effective catalyst. Actually, no oxidation is observed at
all, not even with the “easily oxidizable” benzyl alcohol. In fact,
a green precipitate immediately appears when one equivalent of
tert-BuOK is added to this system, which is not observed with
the ligands 4-bpyT and 5-bpyT. Most likely, the precipitation of
copper catalytic species causes a quenching of the reaction (as
no homogeneous species, able to catalyze the oxidation reaction,
are present in solution). Further investigations are required to
appraise the mechanism of these reactions.
General
All chemicals were used as commercially obtained without further
purification. 1H and 13C NMR spectra were recorded using a
DPX 300 Bruker (300 MHz) instrument. Chemical shifts are
reported in ppm (parts per million) relative to the solvent peak. X-
Band electron paramagnetic resonance (EPR) measurements were
performed on a Bruker EMX spectrometer. Elemental analyses (C,
H, N) were carried out on a Perkin-Elmer 2400 series II analyzer.
FTIR spectra were recorded with a Perkin-Elmer Paragon 1000
FTIR spectrophotometer, equipped with a Golden Gate ATR
device, using the reflectance technique (4000-300 cm-1). ESI mass
analyses were carried out on a Voyager Elite from PerSeptive
Biosystems. UV-Vis spectra were recorded as solid and in solution
using a Perkin-Elmer Lambda 900 spectrophotometer by using
MgO as a reference. 2,4-Dichloro-6-diphenylamino-1,3,5-triazine
457 was prepared from commercial product 2,4,6-trichloro-1,3,5-
triazine 1 and diphenylamine applying a synthetic procedure
described in detail elsewhere.58 6-Methyl-2,2¢-bpyridine 2c was
synthesized according to a reported procedure.46
Concluding remarks
Three novel bifunctional triazine-based ligands holding both a
TEMPO unit and a bipyridine ligand have been designed and
synthesized. The three bpy/TEMPO ligands are differentiated by
the different position of the anchoring point of the bipyridine
moiety on the triazine core. These bifunctional molecules have
been investigated as catalysts with CuBr2 for the aerobic oxidation
of alcohols in the presence of tert-BuOK as basic co-catalyst.
The copper(II) complexes obtained from CuBr2 in situ from
the ligands 4-bpyT and 5-bpyT selectively catalyze the aerobic
oxidation of primary benzylic, allylic, and aliphatic alcohols,
4-Methyl-4¢-hydroxyethyl-2,2¢-bipyridine (3a)
The preparation of compound 3a was carried out according
to an experimental procedure described in the literature.45 To
a solution of diisopropylamine (1.75 mL, 12.5 mmol) in THF
(20 mL) at -78 ◦C, n-butyllithium (2.5 M solution in hexanes,
4.41 mL, 11 mmol) was added. The reaction mixture was stirred
3566 | Dalton Trans., 2009, 3559–3570
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
at -78 ◦C for 20 min. A solution of commercially available (Acros
organics) 4,4-dimethyl-2,2¢-bipyridine 2a (1.84 g, 10 mmol) in dry
THF (20 mL) was transferred using a cannula to the previous
solution. The reaction mixture was stirred at -78 ◦C for 2 h.
Dry paraformaldehyde (3.0 g, 99 mmol) was added quickly to
the above mixture. The resulting reaction mixture was stirred
for another hour at -78 ◦C before the low-temperature bath was
removed and the reaction was stirred overnight. The reaction was
quenched with 1 mL of water and the mixture was extracted with
CH2Cl2. The combined organic layers were dried over anhydrous
Na2SO4 and the solvent was removed under reduced pressure to
yield a brown oil, which was purified by column chromatography.
The pure compound 3a was obtained as a light yellow oil. Yield:
1.0 g (47%). 1H MNR and 13C NMR data were in agreement with
those in the literature.45
by column chromatography. Pure compound 3c was obtained
as a light yellow oil. Yield: 2.22 g (49%). H NMR (300 MHz,
1
CDCl3) d = 8.60 (d, J = 4.0, 1H), 8.25 (d, J = 8.0, 1 H), 8.18
(d, J = 7.5, 1 H), 7.73 (dd, J = 1.6, 7.7, 1 H), 7.67 (dt, J =
2.8, 7.4, 1 H), 7.21 (ddd, J = 1.0, 4.9, 7.4, 1 H), 7.11 (d, J =
7.6, 1 H), 4.08 (t, J = 5.8, 2 H), 3.04 (t, J = 5.8, 2 H) ppm; 13C
NMR (75 MHz, CDCl3) d = 159.2, 155.4, 154.7, 148.6, 137.0,
136.6, 123.3, 123.2, 120.6, 118.5, 61.2, 39.0 ppm. MS-EI+ (m/z)
1
[M + H]+ 200.99 (calcd 200.09). See ESI for H/13C and mass
spectra.
4-(2-(5-(4-Methylpyridin-2-yl)pyridin-3-yl)ethoxy)-6-chloro-N,N-
diphenyl-1,3,5-triazin-2-amine (5a)
A solution of compound 3a (1.0 g, 4.7 mmol) in dry THF (5 mL)
was added dropwise using a syringe to a stirred suspension of
NaH (0.6 g, 60% dispersion in mineral oil, in 40 mL of dry
THF, 15 mmol). Argon was bubbled through the solution for
30 minutes. Next, this solution was added slowly to a stirred
solution of 4 (1.5 g, 4.7 mmol) in 40 mL of THF under argon.
After completion of the addition, the reaction mixture was
stirred overnight at room temperature. The reaction mixture was
subsequently quenched with the addition of 1 mL of distilled
water. Then, the solvent was removed under reduced pressure.
The resulting product was extracted three times with ethyl acetate.
The combined organic layers were washed with saturated NaCl
and dried over anhydrous MgSO4. The organic phase was filtered
and the solvent was evaporated under reduced pressure. The title
compound was purified by column chromatography. Pure 5a was
obtained as a colourless solid. Yield: 0.988 g (45%). Anal. Calcd
for C28H23ClN6O: C 67.94, H 4.68, N, 16.98; found: C 67.72, H,
4.86, N, 16.29%.1H NMR (300 MHz, CDCl3) d = 8.55 (d, J =
5.0, 2 H), 8.22 (s, 1 H), 8.19 (s, 1 H), 7.41–7.26 (m, 10 H), 7.15
(d, J = 4.9, 1 H), 7.03 (dd, J = 1.4, 5.0, 1 H), 4.43 (t, J = 7.0,
2 H), 3.04 (t, J = 7.0, 2 H), 2.45 (s, 3 H) ppm; 13C NMR (75 MHz,
CDCl3) d 171.2, 170.2, 166.8, 156.3, 155.6, 149.1, 148.8, 148.1,
147.2, 142.2, 129.1, 127.4, 127.0, 124.7, 124.2, 121.9, 121.5, 67.2,
34.3, 21.1 ppm. MS-EI+ (m/z) [M]+ 494.98 (calcd 494.16). See
ESI for 1H/13C and mass spectra.
5-Methyl-5¢-hydroxyethyl-2,2¢-bipyridine (3b)
To a solution of◦diisopropylamine (1.75 mL, 12.5 mmol) in THF
(16 mL) at -78 C was added n-butyllithium (2.5 M solution in
hexane, 4.41 mL, 11 mmol). The reaction mixture was stirred
at -78 ◦C for 20 min. Commercially available (Acros organics)
5,5¢-dimethyl-2,2¢-bipyridine 2b (1.84 g, 10 mmol) was dissolved
in 22 mL of dry THF and transferred using a cannula to the
previous solution. The reaction mixture was stirred at -78 ◦C for
2 h. Dry paraformaldehyde (3.0 g, 99 mmol) was added to the
reaction mixture. After complete addition, the reaction mixture
was stirred for another hour at -78 ◦C. The low-temperature bath
was subsequently removed and the reaction mixture was stirred
overnight. 1 mL of water was added to quench the reaction and the
mixture was extracted with CH2Cl2. The combined organic layers
were dried over anhydrous Na2SO4 and the solvent was removed
under reduced pressure to yield a brown oil, which was purified
by column chromatography. Pure compound 3b was obtained as a
light yellow oil. Yield: 1.05 g (49%). 1H NMR (300 MHz, CDCl3)
d = 8.46 (d, J = 1.5, 2 H), 8.21 (dd, J = 5.0, 8.0, 2 H), 7.66
(dd, J = 2.2, 8.1, 1H), 7.61 (dd, J = 1.5, 8.1, 1 H), 3.87 (t, J =
6.5, 2 H), 2.87 (t, J = 6.5, 2 H), 2.37 (s, 3 H) ppm; 13C NMR
(75 MHz, CDCl3) d = 154.4, 153.4, 149.6, 149.5, 137.5, 137.5,
134.5, 134.3, 133.3, 120.6, 120.5, 62.9, 36.1, 18.3 ppm. MS-EI+
(m/z) [M + H]+ 215.01 (calcd 214.11). See ESI for 1H/13C and mass
spectra.
The solid-state structure of compound 5a has been determined
by single-crystal X-ray diffraction and is depicted in Figure S1
(ESI). Details of the crystals data are collected in Table S2
(ESI).
6-Hydroxyethyl-2,2¢-bipyridine (3c)
4-(2-(6-(5-Methylpyridin-2-yl)pyridin-3-yl)ethoxy)-6-chloro-N,N-
To a solution of diisopropylamine (3.5 mL, 25 mmol) in THF
(30 mL) at -78 ◦C, was added n-butyllithium (2.5 M solution
in hexanes, 10 mL, 25 mmol). The reaction mixture was stirred
at -78 ◦C for 20 min. A solution of 6-methyl-2,2¢-bipyridine 2c
(3.86 g, 22.7 mmol) in dry THF (50 mL) was transferred using
a cannula to the previous solution. The reaction mixture was
stirred at -78 ◦C for 2 h. To this reaction mixture was added dry
paraformaldehyde (3.0 g, 99 mmol) and the resulting solution was
stirred for another hour at -78 ◦C. The low-temperature bath was
then removed and the reaction mixture was stirred overnight. The
reaction was quenched by the addition of 1 mL of water and the
mixture was extracted with CH2Cl2. The combined organic layers
were dried over anhydrous Na2SO4 and the solvent was removed
under reduced pressure to yield a brown oil, which was purified
diphenyl-1,3,5-triazin-2-amine (5b)
Argon was bubbled for 30 minutes through a mixture of 3b (1 g,
4.7 mmol) and NaH (0.6 g, 60% dispersion in mineral oil, 15 mmol)
in 30 mL of THF. This solution was added slowly to a stirred
solution of 4 (1.5 g, 4.7 mmol) in 40 mL of THF. After completion
of the addition, the reaction mixture was stirred overnight at room
temperature. The mixture was diluted with sat. NaCl solution and
extracted with ethyl acetate. The combined organic layers were
dried over anhydrous MgSO4. The title compound was purified
by column chromatography. Pure 5b was obtained as a colourless
solid. Yield: 1.08 g (46%). Anal. Calcd for C28H23ClN6O: C 67.94,
1
H 4.68, N, 16.98; found: C 67.48, H, 4.28, N, 16.63%. H NMR
(300 MHz, CDCl3) d = 8.50 (s, 1 H), 8.40 (d, J = 1.7, 1 H), 8.25
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3559–3570 | 3567
©

View Article Online
(dd, J = 3.3, 8.1, 2 H), 7.62 (dd, J = 1.5, 8.1, 1 H), 7.46 (dd,
J = 2.2, 8.2, 1 H), 7.42–7.26 (m, 10 H), 4.37 (t, J = 7.1, 2 H),
2.99 (t, J = 7.1, 2 H), 2.40 (s, 3 H) ppm; 13C NMR (75 MHz,
CDCl3) d = 171.3, 170.3, 166.9, 154.8, 153.4, 149.5, 149.4, 142.2,
137.4, 137.4, 133.3, 132.5, 129.2, 127. 5, 127.1, 120.4, 67.9, 31.9,
18.3 ppm. MS-EI+ (m/z) [M]+ 494.84 (calcd 494.16). See ESI for
1H/13C and mass spectra.
The solid-state structure of 5b has been determined by single-
crystal X-ray diffraction and is depicted in Figure S2 (ESI). Details
of the crystals data are collected in Table S2 (ESI).
4-(2-(5¢-Methyl-2,2¢-bipyridin-5-yl)ethoxy)-6-(4-amino-2,2,6,6-
tetramethylpiperidine-1-oxyl)-N,N-diphenyl-1,3,5-triazin-2-amine
(5-bpyT)
DiPEA (4 mmol, 0.660 mL) was added to a solution of compound
5b (1.07 g, 2.17 mmol) in THF (20 mL). To this mixture was
added a solution of 4-amino TEMPO (2.50 mmol, 0.427 g) in
THF (4 mL). The resulting solution was refluxed for two days
and then cooled to room temperature. The mixture was diluted
with saturated NaCl solution and extracted three times with ethyl
acetate. The combined organic layers were dried over anhydrous
MgSO4 and the solvent was evaporated under reduced pressure.
The title compound was purified by column chromatography. Pure
5-bpyT was obtained as a pink solid. Yield: 0.51 g (37%). Anal.
Calcd for C37H41N8O2: C 70.56, H 6.56, N, 17.79; found: C 71.02,
H, 6.69, N, 17.87%. MS-EI+ (m/z) [M]+ 629.12 (calcd 629.34). See
ESI for the mass spectrum.
4-(2-(6-(Pyridin-2-yl)pyridin-2-yl)ethoxy)-6-chloro-N,N-diphenyl-
1,3,5-triazin-2-amine (5c)
Argon was bubbled for 30 minutes through the reaction mixture
obtained from a solution of compound 3c (1.0 g, 5 mmol) in
15 mL of THF and a suspension of NaH (0.6 g, 60% dispersion in
mineral oil, 15 mmol) in 40 mL of THF. Next, the suspension was
added slowly to a stirred solution of 4 (1.6 g, 5 mmol) in 40 mL of
THF. After completion of the addition, the reaction mixture was
stirred overnight at room temperature. The reaction was quenched
by addition of 1 mL of water and extracted with ethyl acetate.
The combined organic layers were dried over anhydrous MgSO4
and the solvent was evaporated under reduced pressure. The title
compound was purified by column chromatography. Pure 5c was
obtained as a colourless solid. Yield: 0.890 g (40%). Anal. Calcd
for C27H21ClN6O: C 67.43, H 4.40, N, 17.47; found: C 67.30, H,
4.22, N, 17.05%. 1H NMR (300 MHz, CDCl3) d = 8.66 (ddd, J =
0.9, 1.8, 4.8, 1 H), 8.35 (dt, J = 1.0, 8.0, 1 H), 8.23 (d, J = 7.7, 1 H),
7.79 (td, J = 1.8, 7.7, 1 H), 7.69 (t, J = 7.8, 1 H), 7.35 (dt, J = 1.8,
7.8, 1 H), 7.31–7.21 (m, 10 H), 7.08 (dd, J = 0.7, 7.6, 1 H), 4.65
(t, J = 6.9, 2 H), 3.20 (t, J = 6.8, 2 H) ppm; 13C NMR (75 MHz,
CDCl3) d = 171.2, 170. 5, 166.8, 156.9, 156.2, 155.6, 149.0, 142.3,
137.2, 136.7, 129.1, 127.4, 126.9, 123.5, 123.5, 121.1, 118.8, 67.3,
36.9 ppm. MS-EI+ (m/z) [M + H]+ 480.96 (calcd 480.15). See ESI
for 1H/13C and mass spectra.
4-(2-(2,2¢-Bipyridin-6-yl)ethoxy)-6-(4-amino-2,2,6,6-tetramethyl-
piperidine-1-oxyl)-N,N-diphenyl-1,3,5-triazin-2-amine (6-bpyT)
To a solution of compound 5c (0.89 g, 1.86 mmol) in THF (40 ml)
was added DiPEA (4 mmol, 0.660 mL). A solution of 4-amino
TEMPO (2.2 mmol, 0.36 g) in dry THF (4 ml) was added to
this mixture. The resulting solution was refluxed for two days
after which it was cooled to room temperature. The mixture was
diluted with saturated NaCl solution and extracted three times
with ethyl acetate. The solvent was dried over anhydrous MgSO4
and evaporated under reduced pressure. The orange compound
was purified by column chromatography yielding pure 6-bpyT as
a pink solid. Yield: 0.83 g (72%). Anal. Calcd for C36H39N8O2:
C 70.22, H 6.38, N, 18.20; found: C 70.14, H, 6.58, N, 18.13%.
MS-EI+ (m/z) [M + H]+ 616.01 (calcd 615.32). See ESI for the
mass spectrum.
The solid-state structure of the ligand 6-bpyT has been deter-
mined by single-crystal X-ray diffraction and is depicted in Fig. S4
(ESI). Details of the crystal data are listed in Table S3 (ESI).
4-(2-(4¢-Methyl-2,2¢-bipyridin-4-yl)ethoxy)-6-(4-amino-2,2,6,6-
tetramethylpiperidine-1-oxyl)-N,N-diphenyl-1,3,5-triazin-2-amine
(4-bpyT)
[Cu2(4-bpyT)2Br4](CH3CN)7 (6)
A solution of CuBr2 (16.8 mg, 0.075 mmol) in 4 mL of acetonitrile
was added to a stirred solution of 4-bpyT (31 mg, 0.05 mmol)
in 6 mL of acetonitrile. The solution immediately became green.
After 10 min of stirring, the reaction mixture was filtered. The
green filtrate was left unperturbed at room temperature in air
for the slow evaporation of the solvent. Green crystals of [Cu2(4-
bpyT)2Br4](CH3CN)7 (6) were collected by filtration after two days
and characterized by X-ray diffraction analysis. Yield: 15 mg (30%,
based on the ligand). Anal. Calcd for C74H82Br4Cu2N16O4 without
solvent molecules (%): C, 52.09; H, 4.84; N, 13.13. Found: C, 51.68;
H, 4.72; N, 12.82. MS-EI+ (m/z) [6 - Br]+ 772.66 (calcd 773.18
(100.0%)). See ESI for mass spectrum (Figure S5).
To a solution of compound 5a (0.984 g 2 mmol) in THF (20 mL)
was added DiPEA (4 mmol, 0.660 mL). Next, a solution of
4-amino-2,2,6,6-tetramethylpiperidine-1-oxyl (4-amino TEMPO)
(2.50 mmol, 0.427 g) in THF (4 mL) was added to the previous
solution. The resulting reaction mixture was refluxed for four days
before it was cooled to room temperature. The mixture was diluted
with saturated NaCl solution and extracted with ethyl acetate.
The combined organic layers were dried over anhydrous MgSO4
and the solvent was evaporated under reduced pressure. The title
compound was purified by column chromatography. Pure 4-bpyT
was obtained as a pink solid. Yield: 1.10 g (87%). Anal. Calcd for
C37H41N8O2: C 70.56, H, 6.56, N, 17.79; found: C 70.64, H 6.61,
N, 17.10%. MS-EI+ (m/z) [M + H]+ 630.08 (calcd 629.34). See
ESI for the mass spectrum.
[Cu2(5-bpyT)2Br4](CH3CN)2 (7)
A solution of CuBr2 (16.8 mg, 0.075 mmol) in 4 mL of acetonitrile
was added into a stirred solution of 5-bpyT (31 mg, 0.05 mmol) in
6 mL of acetonitrile. The solution immediately turned dark-green.
After 10 min of stirring, the reaction mixture was filtered and the
dark-green filtrate was left unperturbed for the slow evaporation
The solid-state structure of the ligand 4-bpyT has been deter-
mined by single-crystal X-ray diffraction and is depicted in Figure
S3 (ESI). Details of the crystals data are collected in Table S3
(ESI).
3568 | Dalton Trans., 2009, 3559–3570
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
of the solvent at room temperature in air. Dark-green crystals of
[Cu2(5-bpyT)2Br4](CH3CN)2 (7) were collected by filtration after
two days and characterized by X-ray diffraction. Yield: 12 mg
(27%, based on the ligand). Anal. Calcd for C78H88Br4Cu2N18O4
(%): C, 52.39; H, 4.96; N, 14.10. Found: C, 52.05; H, 4.91; N,
13.98. MS-EI+ (m/z) [7 - Br]+ 772.60 (calcd 773.18 (100.0%)). See
ESI for mass spectrum (Figure S6).
5 J. March, Advanced Organic Chemistry: Reactions, Mechanisms, and
Structure, 4th edn, John Wiley & Sons, New York, 1992.
6 D. B. Dess and J. C. Martin, J. Am. Chem. Soc., 1991, 113, 7277–7287.
7 A. J. Mancuso and D. Swern, Synthesis, 1981, 165–185.
8 K. E. Pfitzner and J. G. Moffatt, J. Am. Chem. Soc., 1963, 85, 3027–
3027.
9 K. E. Pfitzner and J. G. Moffatt, J. Am. Chem. Soc., 1965, 87, 5661–
5670.
10 R. A. Sheldon, I. W. C. E. Arends, G. J. Ten Brink and A. Dijksman,
Acc. Chem. Res., 2002, 35, 774–781.
11 W. Adam, C. R. Saha-Moller and P. A. Ganeshpure, Chem. Rev., 2001,
101, 3499–3548.
Cu(6-bpyT)Br2 (8)
12 P. L. Anelli, C. Biffi, F. Montanari and S. Quici, J. Org. Chem., 1987,
52, 2559–2562.
A solution of CuBr2 (16.8 mg, 0.075 mmol) in 4 mL of acetonitrile
was added to a stirred solution of 6-bpyT (31 mg, 0.05 mmol). The
solution immediately turned dark-red. After 10 min of stirring,
the reaction mixture was filtered and the resulting dark-red filtrate
was left unperturbed for the slow evaporation of the solvent at
room temperature in air. Dark-red crystals of 8 were collected by
filtration after two days and characterized by X-ray diffraction.
Yield: 16 mg (38%, based on the ligand). Anal. Calcd for the
monomer C36H39Br2CuN8O2 (%): C, 51.53; H, 4.68; N, 13.35.
Found: C, 51.42; H, 4.45; N, 13.27. MS-EI+ (m/z) [8 - Br]+ 758.58
(calcd 759.17 (100.0%)). See ESI for mass spectrum (Figure S7).
13 P. J. Figiel, M. Leskela and T. Repo, Adv. Synth. Catal., 2007, 349,
1173–1179.
14 R. A. Sheldon and I. W. C. E. Arends, J. Mol. Catal. A: Chem., 2006,
251, 200–214.
15 P. Gamez, I. W. C. E. Arends, J. Reedijk and R. A. Sheldon, Chem.
Commun., 2003, 2414–2415.
16 P. Gamez, I. Arends, R. A. Sheldon and J. Reedijk, Adv. Synth. Catal.,
2004, 346, 805–811.
17 R. H. Liu, X. M. Liang, C. Y. Dong and X. Q. Hu, J. Am. Chem. Soc.,
2004, 126, 4112–4113.
18 F. Minisci, C. Punta and F. Recupero, J. Mol. Catal. A: Chem., 2006,
251, 129–149.
19 I. E. Marko, P. R. Giles, M. Tsukazaki, S. M. Brown and C. J. Urch,
Science, 1996, 274, 2044–2046.
20 T. Iwahama, Y. Yoshino, T. Keitoku, S. Sakaguchi and Y. Ishii, J. Org.
Chem., 2000, 65, 6502–6507.
Description of a typical catalytic experiment
The oxidation reactions were carried out in air in a 10 mL round-
bottom flask equipped with a magnetic stirrer. Typically, the
alcohol (2.0 mmol) was dissolved in 3 mL of a CH3CN/H2O (2:1)
solvent mixture. 11.2 mg (0.1 mmol) of tert-BuOK were added,
followed by 22.3 mg (0.1 mmol) of CuBr2, resulting in a blue-
green suspension. The ligand (60.4 mg, 0.1 mmol) was added and
the reaction suspension slowly turned dark-green. Samples of the
reaction mixture were taken out regularly to monitor the reaction
by GC.
The products of the reaction were determined by comparison
with the commercially available carbonyl compounds. The yields
of aldehydes or ketones were estimated by GC analyses with a
Hewlett-Packard 5890 Series GC equipped with a CP-Sil-5 CB
WCOT Fused Silica capillary column (50 m ¥ 0.25) using dodecane
as external standard (the samples were diluted in a diethyl ether
solution of dodecane).
21 S. R. Reddy, S. Das and T. Punniyamurthy, Tetrahedron Lett., 2004, 45,
3561–3564.
22 K. P. Peterson and R. C. Larock, J. Org. Chem., 1998, 63, 3185–3189.
23 D. R. Jensen, J. S. Pugsley and M. S. Sigman, J. Am. Chem. Soc., 2001,
123, 7475–7476.
24 S. S. Stahl, Science, 2005, 309, 1824–1826.
25 H. Shimizu, S. Onitsuka, H. Egami and T. Katsuki, J. Am. Chem. Soc.,
2005, 127, 5396–5413.
26 S. Velusamy, M. Ahamed and T. Punniyamurthy, Org. Lett., 2004, 6,
4821–4824.
27 N. Jiang and A. J. Ragauskas, J. Org. Chem., 2006, 71, 7087–7090.
28 Y. Kashiwagi, H. Ikezoe and T. Ono, Synlett., 2006, 69–72.
29 C. Bolm and T. Fey, Chem. Commun., 1999, 1795–1796.
30 A. E. Dijksman, I. W. C. E. Arends and R. A. Sheldon, Chem. Commun.,
2000, 271–272.
31 G. Pozzi, M. Cavazzini, S. Quici, M. Benaglia and Dell’G. Anna, Org.
Lett., 2004, 6, 441–443.
32 N. Jiang and A. J. Ragauskas, Tetrahedron Lett., 2005, 46, 3323–
3326.
33 N. Jiang and A. J. Ragauskas, Org. Lett., 2005, 7, 3689–3692.
34 A. P. Dobbs, M. J. Penny and P. Jones, Tetrahedron Lett., 2008, 49,
6955–6958.
35 O. Holczknecht, M. Cavazzini, S. Quici, I. Shepperson and G. Pozzi,
Adv. Synth. Catal., 2005, 347, 677–688.
Acknowledgements
36 A. Gheorghe, E. Cuevas-Yanez, J. Horn, W. Bannwarth, B. Narsaiah
and O. Reiser, Synlett., 2006, 2767–2770.
37 I. A. Ansari and R. Gree, Org. Lett., 2002, 4, 1507–1509.
38 N. W. Wang, R. H. Liu, J. P. Chen and X. M. Liang, Chem. Commun.,
2005, 5322–5324.
The authors are grateful for support from the Graduate Research
School Combination “Catalysis”, a joint activity of the graduate
research schools NIOK, HRSMC, and PTN, and the COST
program Action D35/0011. Coordination of some of our research
by the FP6 Network of Excellence “Magmanet” (contract number
515767) is also kindly acknowledged. The Advanced Light Source
is supported by the Director, Office of Science, Office of Basic
Energy Sciences, of the U.S. Department of Energy under Contract
No. DE-AC02–05CH11231.
39 G. J. ten Brink, I. W. C. E. Arends and R. A. Sheldon, Adv. Synth.
Catal., 2002, 344, 355–369.
40 S. Striegler, Tetrahedron, 2006, 62, 9109–9114.
41 J. S. Uber, Y. Vogels, D. van den Helder, I. Mutikainen, U. Turpeinen,
W. T. Fu, O. Roubeau, P. Gamez and J. Reedijk, Eur. J. Inorg. Chem.,
2007, 4197–4206.
42 Z. L. Lu, J. S. Costa, O. Roubeau, I. Mutikainen, U. Turpeinen, S. J.
Teat, P. Gamez and J. Reedijk, Dalton Trans., 2008, 3567–3573.
43 O. Kahn, R. Prins, J. Reedijk and J. S. Thompson, Inorg. Chem., 1987,
26, 3557–3561.
References
44 P. de Hoog, P. Gamez, W. L. Driessen and J. Reedijk, Tetrahedron Lett.,
2002, 43, 6783–6786.
45 R. H. Terbrueggen, T. W. Johann and J. K. Barton, Inorg. Chem., 1998,
37, 6874–6883.
46 T. Garber, S. Vanwallendael, D. P. Rillema, M. Kirk, W. E. Hatfield,
J. H. Welch and P. Singh, Inorg. Chem., 1990, 29, 2863–2868.
1 M. Hudlicky, Oxidation in Organic Chemistry, ACS Monograph Series,
American Chemical Scociety, Washington, DC, 1990.
2 R. A. Sheldon and J. K. Kochi, Metal Catalyzed Oxidation of Organic
Compounds, Academic Press, New York, 1984.
3 T. Mallat and A. Baiker, Chem. Rev., 2004, 104, 3037–3058.
4 T. Vogler and A. Studer, Synthesis, 2008, 1979–1993.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3559–3570 | 3569
©

View Article Online
47 A. W. Addison, T. N. Rao, J. Reedijk, J. Vanrijn and G. C. Verschoor,
J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
48 T. Pintauer, J. Qiu, G. Kickelbick and K. Matyjaszewski, Inorg. Chem.,
2001, 40, 2818–2824.
49 R. P. Hammond, M. Cavaluzzi, R. C. Haushalter and J. A. Zubieta,
Inorg. Chem., 1999, 38, 1288–1292.
53 A. B. P. Lever, Inorganic Electronic Spectroscopy, 2nd. edn, Elsevier,
Amsterdam, 1984.
54 N. Ray, S. Tyagi and B. Hathaway, J. Chem. Soc., Dalton Trans., 1982,
143–146.
55 E. I. Solomon, K. W. Penfield and D. E. Wilcox, Struct. Bond., 1983,
53, 2–57.
50 B. J. Hathaway and D. E. Billing, Coord. Chem. Rev., 1970, 5, 143–
207.
56 S. Mannam, S. K. Alamsetti and G. Sekar, Adv. Synth. Catal., 2007,
349, 2253–2258.
57 Z. L. Lu, P. Gamez, I. Mutikainen, U. Turpeinen and J. Reedijk, Cryst.
Growth Des., 2007, 7, 1669–1671.
51 L. Yang, D. R. Powell and R. P. Houser, Dalton Trans., 2007, 955–
964.
52 R. D. Willett, G. Pon and C. Nagy, Inorg. Chem., 2001, 40, 4342–
4352.
58 P. de Hoog, P. Gamez, H. Mutikainen, U. Turpeinen and J. Reedijk,
Angew. Chem., Int. Ed., 2004, 43, 5815–5817.
3570 | Dalton Trans., 2009, 3559–3570
This journal is
The Royal Society of Chemistry 2009
©