Coordination chemistry of 1,3-bis(2-pyridylimino)- and 1,3-bis-(2- thiazolylimino)isoindole copper complexes: Investigation of their catalytic behavior in oxidation reactions

Article abstract of DOI:10.1002/ejic.200400012

The copper complexes [Cu(4-MeBPI)(OAc)] (4), [Cu(4-Me-10-tBuBPI)(OAc)] (5) and [Cu(BTI)(OAc)] (6) [BPI = 1,3-bis(2-pyridylimino)isoindole, BTI = 1,3-bis(2-thiazolylimino)isoindole] were prepared by reaction of the protio ligands with copper(II) acetate. Compounds 4 and 6 were characterized by X-ray diffraction, establishing distorted square-planar coordination geometries of the copper ions. Two monoclinic modifications of 6 (6a and 6b) were found, both crystallizing in the space group P2₁/c, but possessing different cell parameters. In contrast to 6a, which is monomeric in the crystal, the second monoclinic modification 6b has a more complicated crystal structure, which is composed of both monomeric complex units such as those found in 6a and infinite chains of coordination polymers. The copper atoms in the polymeric chains of 6b display fivefold coordination and a ligand polyhedron that is an intermediate form between a trigonal-bi-pyramidal and a square-pyramidal geometry. The allylic peroxylation of cyclohexene with tBuOOH (70% aqueous solution) catalyzed by 4 and 6 (0.17 mol %) gave tert-butylperoxy-3-cyclohexene with selectivities of 86% and 80% (based on cyclohexene) and turnover frequencies of 63 h ^-1 and 18 h^-1, respectively. The peroxylation reaction is thought to proceed according to a Haber-Weiss radical chain mechanism. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200400012

FULL PAPER
Coordination Chemistry of 1,3-Bis(2-pyridylimino)- and 1,3-Bis-
(2-thiazolylimino)isoindole Copper Complexes: Investigation of Their Catalytic
Behavior in Oxidation Reactions
Markus B. Meder[a] and Lutz H. Gade*[a,b]
Keywords: Copper / Allylic peroxylation / Coordination polymers / Radical reactions / Oxidation
The copper complexes [Cu(4-MeBPI)(OAc)] (4), [Cu(4-Me-
10-tBuBPI)(OAc)] (5) and [Cu(BTI)(OAc)] (6) [BPI = 1,3-bis(2-
pyridylimino)isoindole, BTI = 1,3-bis(2-thiazolylimino)isoin-
of coordination polymers. The copper atoms in the polymeric
chains of 6b display fivefold coordination and a ligand poly-
hedron that is an intermediate form between a trigonal-bi-
dole] were prepared by reaction of the protio ligands with pyramidal and a square-pyramidal geometry. The allylic per-
copper(II) acetate. Compounds 4 and 6 were characterized by oxylation of cyclohexene with tBuOOH (70% aqueous solu-
X-ray diffraction, establishing distorted square-planar coor- tion) catalyzed by 4 and 6 (0.17 mol %) gave tert-butylper-
dination geometries of the copper ions. Two monoclinic mo- oxy-3-cyclohexene with selectivities of 86% and 80% (based
difications of 6 (6a and 6b) were found, both crystallizing in
the space group P21/c, but possessing different cell para-
meters. In contrast to 6a, which is monomeric in the crystal,
the second monoclinic modification 6b has a more complic-
ated crystal structure, which is composed of both monomeric
complex units such as those found in 6a and infinite chains
on cyclohexene) and turnover frequencies of 63 h−1 and 18
h−1, respectively. The peroxylation reaction is thought to pro-
ceed according to a Haber−Weiss radical chain mechanism.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
(1)
(2)
The direct oxidation of CϪH bonds is an important
chemical transformation in organic synthesis. A prepara-
tively useful example of such a conversion of hydrocarbons
is the copper-catalyzed KharaschϪSosnovsky reaction, i.e.
the allylic acyloxylation of olefins [Equation (1)].[1] In the
presence of chiral oxazoline-based ligands it has been pos-
sible to render this reaction enantioselective,[2] although a
truly efficient asymmetric catalyst has not as yet been
found. A much less studied variant of this oxidation is the
allylic peroxylation, which is also catalyzed by Cu salts
[Equation (2)]. The synthesis of dialkyl peroxides by the
oxidation of cycloolefins with tert-butyl hydroperoxide, first
reported by Kharasch,[3] has recently attracted renewed at-
tention due to the high diastereoselectivity of the sub-
sequent epoxidation or dihydroxylation of the CϭC bond.[4]
This gives rise to highly functionalized cyclohexane deriva-
tives.
In the standard reaction protocols for allylic peroxyla-
tions, a CuI or CuII salt and, occasionally, an N-donor li-
gand are added to the reaction mixture which assemble to
give the active catalyst, the nature of which is generally not
unambiguously established.
In this paper we report the synthesis and structural
characterization of several CuII complexes containing 1,3-
bis(2-pyridylimino)isoindole (BPI) and 1,3-bis(2-thiazolyl-
imino)isoindole (BTI) ligands, and their use as selective
catalysts in the allylic peroxylation of cyclohexene. First re-
ported in the 1950s,[5] such tridentate ligands have more re-
cently been employed by Siegl and co-workers in the tran-
[a]
´
Laboratoire de Chimie Organometallique et de Catalyse
´
(CNRS UMR 7513), Institut Le Bel, Universite Louis Pasteur,
4, rue Blaise Pascal, 67000 Strasbourg, France
Anorganisch-Chemisches Institut, Universität Heidelberg,
Im Neuenheimer Feld 270, 69120 Heidelberg, Germany
E-mail: lutz.gade@uni-hd.de
[b]
2716
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400012
Eur. J. Inorg. Chem. 2004, 2716Ϫ2722

1,3-Bis(2-pyridylimino)- and 1,3-Bis(2-thiazolylimino)isoindole Copper Complexes
FULL PAPER
sition-metal (mainly cobalt) catalyzed oxidation of satu-
rated hydrocarbons.[6,7] In a systematic study of the cobalt-
catalyzed oxidation of cyclohexane and cyclohexene, Mim-
oun et al. employed tert-butyl hydroperoxide as the oxidiz-
ing reagent;[8] furthermore, the photoactivation of this kind
of BPI-metal peroxo complex for the oxidation of hydro-
carbons has also been reported.[9] Given the facile accessi-
bility of well-defined BPI and BTI transition metal com-
plexes and the previous results on copper-catalyzed allylic
peroxylations, copper complexes of isoindole-derived tri-
dentate ligands appear to be promising candidates for this
type of transformations.
Scheme 2. Synthesis of the (acetato)(BPI)copper complexes 4 and 5
Results and Discussion
Synthesis of the Bis(2-pyridylimino)isoindole (BPI)- and
Bis(2-thiazolylimino)isoindole (BTI)-Copper Complexes
For the synthesis of the bis(2-pyridylimino)isoindole and
bis(2-thiazolylimino)isoindole ligand precursors, the
method previously reported by Siegl was employed.[6,7,10]
The phthalodinitrile starting materials were treated with an
excess of the corresponding 2-aminopyridines or 2-amino-
thiazole in the presence of a catalytic amount of CaCl2
(Scheme 1). The use of 1-hexanol as solvent instead of 1-
butanol, as is reported in the literature for this type of con-
version, gave significantly improved yields of the ligand pre-
cursors 4-MeBPI (1), 4-Me-10-tBuBPI (2) and BTI (3).
Scheme 3. Synthesis of the (acetato)(BTI)copper complex 6
Structural Characterization of 4, 6a and 6b
Whereas a considerable number of X-ray diffraction
studies have been carried out for transition-metal complexes
of BPI, there are only a few previously determined struc-
´
tures for copper complexes. Gagne reported the dimeric
structure of a (carbonato)(BPI)copper complex and the mo-
lecular structure of a µ-oxo-copper pentamer,[11] while Bere-
man, in a study aimed at modeling the coordination sphere
of the CuII site in galactose oxidase, reported the crystal
structure analysis of a pentacoordinate (acetato)(BPI)-
copper complex.[13]
X-ray diffraction studies of both BPI copper compounds
4 and 5 were carried out. However, severe disorder of the
tert-butyl groups in 5, as well as disordered solvent in the
unit cell, precluded a complete structure analysis of the lat-
ter (although the gross structural features displayed in
Scheme 2 were confirmed). A single-crystal X-ray structure
analysis of complex 4 established the structural details of
this class of compounds. Its molecular structure is displayed
in Figure 1, along with the principal bond lengths and
bond angles.
Scheme 1. Synthesis of the BPI ligands 1 and 2
As is apparent from Figure 1, the molecular structure of
4 is distorted from an ideal square-planar geometry. For
The BPI- and BTI-copper complexes [(4-MeBPI)- instance, the N(1)ϪCuϪN(5) angle was found to be
Cu(OAc)] (4), [(4-Me-10-tBuBPI)Cu(OAc)] (5), and [(BTI)- 158.8(1)°, which is significantly more acute than the corre-
Cu(OAc)] (6) were prepared as previously published for sponding bond angle found by Bereman et al. in the penta-
similar complexes (Scheme 2 and Scheme 3);[11,12] com- coordinate
complex
[(10-MeBPI)Cu(OAc)(H2O)]
pound 6 is the first BTI copper complex to be reported. To (169.4°).[13] The acetato ligand is monodentate and the
[14]
˚
avoid the formation of the symmetric complexes [Cu(BPI)2] CuϪO(1) distance of 1.956(2) A is in the expected range.
or [Cu(BTI)2], the exact metalϪligand stoichiometry was The carboxyl group is arranged orthogonally to the metal
chosen, and the reaction temperature kept below 300 K. complex plane due to steric repulsion with the pyridyl rings.
The formulation and the identity of 4Ϫ6 were confirmed This leads to an axial disposition of the O(2) atom of the
by IR spectroscopy and elemental analysis.
carboxyl unit relative to the copper atom. However, the
Eur. J. Inorg. Chem. 2004, 2716Ϫ2722
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2717

M. B. Meder, L. H. Gade
FULL PAPER
Figure 2. Molecular structure of 6a (top); selected bond lengths
and angles are given in Table 1; view of the molecular packing of
6a in the unit cell (bottom)
˚
Table 1. Comparison of selected bond lengths (A) and angles (°)
of 6a and 6b
Figure 1. Molecular structure of complex 4 (top); selected bond
˚
lengths (A) and angles (°): CuϪN(1) 2.023(3), CuϪN(3) 1.884(3),
6a
6b
CuϪN(5) 2.014(3), CuϪO(2) 2.601(3), CuϪO(1) 1.956(2),
C(20)ϪO(1) 1.249(5), C(20)ϪO(2) 1.245(5), N(1)ϪCuϪN(3)
90.6(1), N1ϪCuϪN5 158.2(1), N(1)ϪCuϪO(1) 93.3(1),
CuϪN1
CuϪN3
CuϪN5
CuϪO1
1.991(3)
1.931(3)
1.989(3)
1.948(3)
2.595
1.94(1)
1.98(1)
1.95(1)
2.08(1)
2.54
2.23(1)
89.1(5)
173.0(5)
154.2(5)
120.6(5)
85.2
N(3)ϪCuϪN(5)
89.3
(1),
N(3)ϪCuϪO(1)
166.1(1),
N(5)ϪCuϪO(1) 91.9(1), O(1)ϪC(20)ϪO(2) 122.9(4); view of the
molecular packing of 4 in the unit cell (bottom)
CuϪO2
CuϪO2Ј
Ϫ
˚
CuϪO(2) distance of 2.601 A does not support a significant
N1ϪCuϪN3
N1ϪCuϪN5
N3ϪCuϪO1
N3ϪCuϪO2
O1ϪCuϪO2Ј
CuϪO1ϪC15
O1ϪC15ϪO2
89.7(1)
172.7(1)
175.3(1)
Ϫ
bonding interaction and should be compared with the
˚
CuϪO bond length of 2.466 A for the weakly axially coor-
dinated water molecule in Bereman’s complex.[13] A view of
the packing of complex 4 in the unit cell clearly indicates
that there are no bridging interactions of the acetato groups
between the molecular units that stack in a head-to-tail fa-
shion (Figure 1, bottom).
Ϫ
104.8(2)
122.8(4)
101(1)
123(1)
Since there is no previous report of a fully structurally
characterized BTI complex, attempts were made to crys-
tallize compound 6 in order to carry out an X-ray diffrac-
The replacement of the pyridyl units by thiazolyl groups
tion study. Two monoclinic crystalline modifications, the leads to a different orientation of the neutral N-donor func-
habits of which differed only marginally and which pos- tions and leaves a more open space for the acetate ligand
sessed identical analytical properties, were obtained in vary- than in 4. These two factors lead to a complex geometry,
ing ratios on slow diffusion of n-hexane into a toluene solu- which deviates less from the idealized planar structure than
tion of 6. Mechanical separation of one type of crystal habit that found for 4. The angles N(1)ϪCuϪN(5) and
and subsequent recrystallization again gave both crystalline N(3)ϪCuϪO(1) are 172.7(1) and 175.3(1)°, respectively.
forms. Single-crystal X-ray structure analysis of both crys- Whereas the metalϪN distance of the formally anionic do-
˚
talline modifications 6a and 6b were carried out in order to nor atom is 0.05 A greater than in 4 [6a: 1.931(3), 4:
˚
establish the details of the molecular structure of 6 and its 1.884(3) A], the copperϪthiazolyl-N bond lengths are
˚
arrangement in both solid forms. The molecular structure about 0.03 A shorter than the corresponding bond lengths
of 6a and its molecular packing pattern is displayed in Fig- in the BPI complex. The view of the molecular packing in
ure 2, while the principal bond lengths and bond angles are 6a, which is displayed in Figure 2, indicates a molecular
listed in Table 1.
headϪtail stacking with interlayer distances of about 3.7
2718
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2716Ϫ2722

1,3-Bis(2-pyridylimino)- and 1,3-Bis(2-thiazolylimino)isoindole Copper Complexes
FULL PAPER
˚
A. There is no indication of bonding interactions through
bridging acetato ligands in this structure.
It should be pointed out that upon redissolution of either
of the two crystalline modifications of compound 6 and
In contrast to 6a, the second monoclinic modification 6b subsequent recrystallization of the material, crystals of both
has a more complicated crystal structure composed of both monoclinic forms reprecipitated. The sole species in solu-
monomeric complex units such as those found in 6a and tion is therefore thought to be the monomer, as observed
infinite chains of coordination polymers. A segment of such in 6a. Exclusive formation of either 6a or 6b was not
a chain structure is displayed in Figure 3, while the param- achieved by variation of either the solvent system or the
eters of the complex within the polymeric structure are temperature of crystallization.
listed in Table 1 (those of the monomers are essentially
identical to 6a).
Allylic Peroxylation of Cyclohexane Catalyzed by
Complexes 4؊6
In order to assess the catalytic activity of the copper com-
plexes in the allylic peroxylation of alkenes, the CuII-cata-
lyzed reaction of tBuOOH with cyclohexene was studied. In
contrast to a previous study on the analogous Co-catalyzed
reaction by Mimoun et al., who employed the oxidant in
very high concentrations (96%),[8] we chose the commercial
aqueous solution of 70% tBuOOH. Remarkably, all three
complexes gave the dialkyl peroxide, tert-butylperoxy-3-
cyclohexene, very selectively (Scheme 4). Conversions to
this major product (based on cyclohexene) were in the
range of 80Ϫ90%.
Figure 3. A segment of the chain structure in 6b (top); selected
bond lengths and angles are given in Table 1; polymeric chain
structure of the (acetato)(BTI)copper complex in 6b (bottom)
In contrast to the almost ideally planar four-coordinate
structure of 6a, the copper atoms in the polymeric chains
of 6b display fivefold coordination and a ligand polyhedron
which is an intermediate form between a trigonal-bipyrami-
dal and a square-pyramidal geometry. While the arrange-
ment of the three heterocycles of the BTI ligand remains
Scheme 4. Allylic peroxylation of cyclohexene catalyzed by com-
plexes 4 and 6 giving the dialkyl peroxide in over 80% yield
The principal difference between the three catalysts was
almost planar [N(1)ϪCuϪN(5), 173.0(5)°], the acetato li- the activity observed: the two BPI derivatives display the
gand is tilted out of this plane [N(3)ϪCuϪO(1), 154.2(5)°], same catalytic activity and are significantly more active
therefore making space for the coordination of the second than the BTI complex as is shown in Figure 4 for catalysts
acetato oxygen atom of a neighboring (acetato)(BTI)copper 4 and 6. Both reactions were carried out with a catalyst
coordination unit. Significantly, the CuϪO(1) bond length loading of 0.17 mol %. Considering the nearly linear part of
˚
of 2.08(1) A is longer than in the square-planar monomer, the conversion curves between 10% and 75%, the turnover
and the CuϪO(2Ј) distance of 2.23(1) A is indicative of sub- frequencies amount to 63 hϪ1 for 4 and 18 hϪ1 for the BTI
˚
stantial intermolecular bonding, leading to the polymeric complex 6. The maximum conversions of 86% observed for
complex. Similar CuϪO bond lengths were reported for complex 4 and 80% for complex 6 correspond to turnover
two BPI copper complexes: a dimeric carbonato- and an numbers of 505 and 470, respectively. The reaction product
oligomeric hydroxo-bridged complex.[12]
tert-butylperoxy-3-cyclohexene was isolated by distillation
Eur. J. Inorg. Chem. 2004, 2716Ϫ2722
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2719

M. B. Meder, L. H. Gade
FULL PAPER
and fully characterized. In order to exclude the possibility
The catalytic reaction most probably proceeds according
that a simple copper salt acts as a catalytic species after to a HaberϪWeiss chain mechanism, as has been suggested
initial degradation of the ligands, analogous catalytic runs previously, and which depicted in Scheme 5.[15] The active
using the identical reaction parameters were carried out species is the copper() tert-butyl peroxide complex which
with copper acetate. Very slow oxidation of cyclohexene thermally decomposes to the CuI complex and a peroxyl
was observed with low chemoselectivity; the conversion of radical. This radical intermediate abstracts a hydrogen atom
the starting material was variable, but in all cases well below from the allylic position of cyclohexene, and the cyclohex-
10% after 1500 minutes (the time frame of the experiments enyl radical then recombines with a second peroxyl radical
represented in Figure 4). Thus, for the results presented in generated in the same way. The CuI product is then re-oxid-
this work, the background activity of nonligated copper ized by the alkyl hydroperoxide in several steps generating
salts is insignificant.
tert-butanol and water as reduced products.
When the reaction product tert-butylperoxy-3-cyclohex-
ene was exposed to the copper complexes over a longer per-
iod of time (Ͼ 20 h for 4 and 5 and Ͼ 35 h for 6), slow
conversion of the dialkyl peroxide to cyclohexenone and
tert-butanol was observed (Scheme 6). This slow secondary
reaction limits the yields of the dialkyl peroxide which can
be isolated from the copper-catalyzed peroxylation de-
scribed in this work. It was also observed by the addition
of the Cu complexes to pure, isolated dialkyl peroxide.
Figure 4. Peroxidation of cyclohexene with the catalysts 4 and 6
Scheme 6. Copper-catalyzed slow conversion of tert-butylperoxy-3-
cyclohexene to cyclohexenone and tert-butanol
Conclusion
In this work we have presented a detailed study of the
coordination chemistry of three BPI and BTI copper com-
plexes and their activity as catalysts for the allylic peroxyl-
ation of cyclohexene as substrate of reference. Using the
hydroperoxide oxidant as an aqueous solution with a con-
centration of 70% gave the tert-butyl cyclohexenyl peroxide
with high selectivity. Other previously described reaction
products such as 3-cyclohexenol, 3,3Ј-dicyclohexenyl were
not observed, even in traces, while exposure of the dialkyl
peroxide to the copper catalysts for an extended period of
time led to a very slow degradation to cyclohexenone as a
secondary product. The high selectivity of the peroxylation
on use of the BPI copper compounds and the relatively high
turnover numbers obtained render these systems competi-
tive catalysts for peroxylation reactions. We are currently
studying the regioselectivity of the peroxide formation for
nonsymmetrical olefins in order to assess the general appli-
cability of this method.
Experimental Section
Scheme 5. Postulated HaberϪWeiss mechanism for the Cu-cata-
lyzed peroxylation: generation of the alkylperoxycopper() com-
plexes as the active species and their degradation to the CuI species
with concomitant transfer of the peroxyl group to the allylic posi-
tion of cyclohexene
All manipulations were performed under a nitrogen atmosphere
in standard (Schlenk) glassware. Solvents were dried according to
standard procedures and saturated with nitrogen. The deuterated
2720
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2716Ϫ2722

1,3-Bis(2-pyridylimino)- and 1,3-Bis(2-thiazolylimino)isoindole Copper Complexes
FULL PAPER
solvents used for NMR spectroscopic measurements were degassed
by three successive ‘‘freeze-pump-thaw’’ cycles and dried using
standard techniques.
brown solid, which was recrystallized from toluene/n-hexane. Yield:
82.0 mg (79%). M.p. 124 °C. IR (KBr): ν˜ ϭ 2965 (w), 1671 (m),
1573 (s), 1507 (m), 1478 (m), 1403 (m), 1360 (m), 1326 (w), 1248
(w), 1182 (m), 1143 (m), 1112 (m), 1082 (w), 946 (w), 918 (w), 885
(w), 828 (w), 729 (w) cmϪ1. C29H33CuN5O2 (547.15 g molϪ1):
calcd. C 63.7, H 6.1, N 12.8; found C 63.9, H 6.2, N 12.4%.
1
The H and 13C NMR spectra were recorded using an AMX 400
FT NMR spectrometer and infrared spectra on a PerkinϪElmer
1600 FT-IR spectrometer. Elemental analyses were carried out in
the microanalytical laboratories of the chemistry departments at
Würzburg (Germany) and Strasbourg (France). The compounds
2-amino-4-tert-butylpyridine[16] and 4-methylphthalonitrile[17] were
prepared according to published methods. All other chemicals used
as starting materials were obtained commercially and used without
further purification.
[Cu(BTI)(OAc)] (6a) and [Cu(BTI)(OAc)]ϱ (6b): A suspension of
BTI (3) (0.12 g, 0.38 mmol) in methanol was added to a suspension
of Cu(OAc)2·H2O (0.15 g, 0.76 mmol) in methanol (10 mL). After
stirring for 12 h at room temperature, the reaction mixture was
dried in vacuo and the residue was redissolved in toluene. After
filtration with Celite, the solution was concentrated and the crude
product was recrystallized from toluene/n-hexane. Orange (6a) and
yellow (6b) crystals suitable for X-ray diffraction were grown by
slow diffusion of n-hexane into a saturated solution of 6 in toluene
at room temperature. Yield: 0.13 g (81%). M.p. 121 °C. IR (KBr):
ν˜ ϭ 3107 (w), 2965 (vw), 1583 (m), 1542 (s), 1494 (m), 1469 (w),
1383 (m), 1308 (w), 1298 (w), 1240 (m), 1190 (m), 1102 (s), 1071
Preparation of the Ligands: The synthesis of the ligand precursors
was carried out by a modified protocol based on the method which
was originally published by Siegl.[7,10] This modified reaction proto-
col will be exemplified for the synthesis of the novel derivative 4-
Me-10-tBuBPI (2).
(m), 916 (vw), 877 (m), 813 (vw), 715 (m), 682 (vw) cmϪ1
.
4-Me-10-tBuBPI (2):
(1.50 g, 10.6 mmol),
A
suspension of 4-methylphthalonitrile
2-amino-4-tert-butylpyridine (3.96 g,
C16H11CuN5O2S2 (432.97 g molϪ1): calcd. C 44.4%; H 2.6%; N
16.2; found C 44.7%; H 2.7%; N 15.9%.
26.1 mmol) and CaCl2 (0.16 g, 1.36 mmol) in 1-hexanol (50 cm3)
was refluxed for 18 h. After cooling to room temperature, the yel-
low compound which precipitated was filtered, washed with water
(500 cm3) and dried in vacuo. Yield: 2.53 g (56%). M.p. 204 °C. 1H
NMR (400.16 MHz, C6D6, 295 K): δ ϭ 1.1, 1.2 (2 ϫ s, 2 ϫ 9 H,
10-tBu), 2.05 (s, 3 H, 4-CH3), 6.84 (m, 2 H, 11-H), 6.94 (br. d,
3JH,H ϭ 7.9 Hz, 1 H, 5-H), 7.72 (m, 2 H, 9-H), 8.02 (br. s, 1 H, 3-
Catalytic Peroxylation of Cyclohexene: The catalytic peroxylations
were carried out in neat cyclohexene with catalyst concentrations
of 0.17 mol % and a twofold excess of tert-butyl hydroperoxide.
Samples of ca. 10 µL were taken at the intervals indicated in Fig-
ure 4 and dissolved in 2 mL of toluene. The solution was then in-
jected into a Shimadzu GC-17A/GCMS-QP5050A instrument. The
measured ratio of the substrate and products was calibrated by
comparative measurements with known ratios of pure substances.
The conversion (%) generally refers to the generation of tert-butyl
cyclohex-2-en-1-yl peroxide based on the substrate cyclohexene.
The data displayed are average values of two runs.
3
3
H), 8.12 (d, JH,H ϭ 7.9 Hz, 1 H, 6-H), 8.60 (d, JH,H ϭ 5.3 Hz, 2
H, 12-H), 11.91 (br. s, 1 H, NϪH). {1H}13C NMR (100.6 MHz,
C6D6, 295 K): δ ϭ 21.6 (4-CH3), 30.4 (10-C(CH3)3], 34.5 (10-
C(CH3)3], 117.7 (C-11), 121.1 (C-9), 122.9 (C-5), 123.6 (C-6), 132.8
(C-3), 137.2 (C-1, C-2), 141.9 (C-4), 148.1 (C-12), 154.1 (C-7),
161.8 (C-8, C-10) ppm. IR (KBr): ν˜ ϭ 3237 (m, br), 2962 (m), 2867
(w), 1633 (m), 1590 (s), 1534 (w), 1477 (w), 1401 (w), 1365 (w),
1354 (w), 1310 (vw), 1285 (w), 1265 (w), 1219 (w), 1201 (vw), 1180
(vw), 1108 (vw), 1037 (vw), 928 (m), 890 (m), 826 (m), 716 (m)
cmϪ1. C27H31N5 (425.57 g molϪ1): calcd. C 76.2, H 7.3, N 16.5;
found C 76.5, H 7.4, N 16.3.
The reaction product tert-butyl cyclohex-2-en-1-yl peroxide was
isolated after filtration of the product mixture through a pad of
Celite and removal of the solvent in vacuo. The crude product was
1
subjected to a Kugelrohr distillation at 1 mbar (b.p. ca. 60 °C). H
NMR (400.16 MHz, CDCl3, 295 K): δ ϭ 1.25 (s, 9 H), 1.56 (m, 1
H), 1.73 (m, 2 H), 1.94 (m, 2 H), 2.06 (m, 12 H), 4.40 (m, 1 H),
5.74 (m, 1 H), 5.95 (m, 1 H). {1H}13C NMR (100.6 MHz, CDCl3,
295 K): δ ϭ 18.8, 25.7, 26.8, 27.3, 76.7, 80.0, 124.5, 133.7 ppm. MS
(EI): 170.1 [Mϩ].
Preparation of the Copper Complexes
[Cu(4-MeBPI)(OAc)] (4): A suspension of 1 (87.0 mg, 0.28 mmol)
was added to a stirred suspension of Cu(OAc)2·H2O (0.11 g,
0.56 mmol) in 10 mL of methanol at room temperature. After stir-
ring for 16 h at room temperature, the reaction mixture was filtered
and the resulting deep brown solution was concentrated in vacuo.
The crude product was suspended in toluene and filtered with Ce-
lite. Evaporation of the solvent afforded a green-brown solid, which
was recrystallized from toluene/n-hexane. Yield: 87.0 mg (71%).
Green-brown crystals suitable for X-ray diffraction were grown by
the slow diffusion of n-hexane into a saturated solution of 4 in
toluene at room temperature. M.p. 116 °C. IR (KBr): ν˜ ϭ 2925
(w), 1628 (m), 1597 (m), 1573 (s), 1531 (m), 1467 (m), 1460 (m),
1431 (m), 1398 (vw), 1365 (w), 1280 (w), 1184 (w), 1135 (m), 1077
(w), 1012 (w), 786 (w), 755 (w) cmϪ1. C21H17CuN5O2 (434.95 g
molϪ1): calcd. C 58.0, H 3.5, N 16.1; found C 58.1, H 3.5, N 15.9%.
X-ray Diffraction Study of 4, 6a and 6b: The crystal data were col-
lected on a NoniusϪKappa CCD diffractometer at Ϫ100 °C and
transferred to a DEC Alpha workstation; for all subsequent calcu-
lations the Nonius OpenMoleN package was used.[18] The struc-
tures were solved by direct methods, and absorption corrections
were part of the scaling procedure of the data reductions. After
refinement of the heavy atoms, difference Fourier maps revealed
the maxima of residual electron density close to the positions ex-
pected for the hydrogen atoms; they were introduced as fixed con-
tributors in the structure factor calculations with fixed coordinates
˚
(CϪH: 0.95 A) and isotropic temperature factors [B(H) ϭ 1.3
2
2
˚
Beqv(C) A ], but not refined. Full least-square refinements on F .
A final difference map revealed no significant maxima of electron
density. The scattering factor coefficients and the anomalous dis-
persion coefficients were taken from the literature.[19] Crystal data
and experimental details for the crystals of 4, 6a and 6b are given
in Table 2.
[Cu(4-Me-10-tBuBPI)(OAc)] (5): A suspension of 2 (80.0 mg,
0.19 mmol) was added to a stirred suspension of Cu(OAc)2·H2O
(75.0 mg, 0.38 mmol) in 10 mL of methanol at room temperature.
After stirring for 16 h at room temperature, the reaction mixture
was filtered and the resulting deep brown solution was concen-
trated in vacuo. The crude product was suspended in toluene and
filtered with Celite. Evaporation of the solvent afforded a green-
CCDC-228180 to -228182 contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge at www.ccdc.cam.ac.uk/conts/retrieving.html [or from the
Eur. J. Inorg. Chem. 2004, 2716Ϫ2722
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2721

M. B. Meder, L. H. Gade
FULL PAPER
Table 2. X-ray data collection and processing parameters for 4, 6a and 6b
4
6a
6b
Empirical formula
M
2(C21H17CuN5O2)C7H8
962.03
21(1)
C16H11CuN5O2S2
432.97
Ϫ100(1)
0.71073
P21/c
10.7706(6)
7.6486(4)
21.383(1)
Ϫ
101.577(5)
Ϫ
C16H11CuN5O2S2
432.97
Ϫ100(1)
0.71073
P21/c
5.4021(6)
8.4923(2)
25.679(1)
Ϫ
91.743(5)
Ϫ
Temp. [°C]
˚
Wavelength. [A]
0.71073
¯
Space group
P1
˚
a [A]
8.3873(5)
9.2129(5)
14.8935(9)
92.069(5)
91.730(5)
103.686(5)
1116.5(2)
1
˚
b [A]
˚
c [A]
α [°]
β [°]
γ [°]
3
˚
V [A ]
1725.7(2)
4
1.67
3357.2(3)
8
1.71
Z
Dcalcd. [g cmϪ3
]
1.43
1.010
Abs. coeff./[mmϪ1
]
1.529
1.572
R indices
R1 ϭ 0.046
Rw ϭ 0.065
R1 ϭ 0.035
Rw ϭ 0.041
R1 ϭ 0.092
Rw ϭ 0.141
[6] [6a]
[6b]
W. O. Siegl, J. Org. Chem. 1977, 42, 1872Ϫ1878.
Siegl, Inorg. Nucl. Chem. Letters 1974, 10, 825Ϫ829.
W. O.
Cambridge Crystallographic Data Centre, 12 Union Road, Cam-
bridge CB2 1EZ, UK; Fax: (internat.) ϩ44-1223-336-033; E-mail:
deposit@ccdc.cam.ac.uk].
[7]
[8]
[9]
´
R. R. Gagne, W. A. Marritt, D. N. Marks, W. O. Siegl, Inorg.
Chem. 1981, 20, 3260Ϫ3267.
L. Saussine, E. Brazi, A. Robine, H. Mimoun, J. Fischer, R.
Weiss, J. Am. Chem. Soc. 1985, 107, 3534Ϫ3540.
Acknowledgments
We thank the Deutsche Forschungsgemeinschaft, the CNRS
(France) and the Institut Universitaire de France for financial
[9a] E. T. Farinas, C. V. Nguyen, P. K. Mascharak, Inorg. Chim.
[9b]
Acta 1997, 263, 17Ϫ21.
F. A. Chavez, P. K. Mascharak,
Acc. Chem. Res. 2000, 33, 539Ϫ 545.
[10]
[11]
W. O. Siegl, J. Heterocyclic Chem. 1981, 18, 1613Ϫ1618.
´
support of this work. We are grateful to Dr. Andre De Cian and
´
R. R. Gagne, R. S. Gall, G. C. Lisensky, R. E. Marsh, L. M.
N. Gruber for the X-ray diffraction studies.
Speltz, Inorg. Chem. 1979, 18, 771Ϫ781.
[12] [12a] A. W. Addison, P. J. Burke, K. Henrick, Inorg. Chem. 1982,
[1]
[12b]
[1a] M. S. Kharasch, G. Sosnovsky, J. Am. Chem. Soc. 1958, 80,
21, 60Ϫ63.
E. Balogh-Hergovich, J. Kaiser, G. Speier, G.
[12c]
756Ϫ757. [1b] M. S. Kharasch, G. Sosnovsky, J. Am. Chem. Soc.
1959, 81, 5819Ϫ5824.
Huttner, A. Jacobi, Inorg. Chem. 2000, 39, 4224Ϫ4229.
P. Anderson, A. la Cour, A. Dodd, A. D. Garrett, M. Wicholas,
O.
[2]
Selected early reports of asymmetric KharaschϪSosnovsky re-
Inorg. Chem. 2003, 42, 4513Ϫ4515.
[2a]
[13] [13a]
actions:
A. S. Gokhale, A. B. E. Minidis, A. Pfaltz, Tetra-
R. D. Bereman, G. D. Shields, J. R. Dorfman, J. Bordner,
[2b]
[13b]
hedron Lett. 1995, 36, 1831Ϫ1834.
Argade, X. Chen, M. G. Pamment, Tetrahedron Lett. 1995, 36,
2945Ϫ2948.
1995, 1245Ϫ1246.
Asymmetry 1995, 6, 147Ϫ156. Recent reviews:
Andrus, J. C. Lashley, Tetrahedron 2002, 58, 845Ϫ866.
M. B. Andrus, A. B.
J. Inorg. Biochem. 1983, 19, 75Ϫ93.
Maehlmann, R. D. Bereman, P. Singh, J. Coord. Chem. 1997,
D. M. Baird, W. P.
[2c]
K. Kawasaki, S. Tsumura, T. Katsuki, Synlett
42, 107Ϫ126.
[2d]
[14]
A. Levina, J. Muzart, Tetrahedron:
R. K. Hocking, T. W. Hambley, Inorg. Chem. 2003, 22,
[2e]
M. B.
2833Ϫ2835.
[2f]
[15] [15a]
[15b]
J.
F. Haber, R. Willstädter, Chem. Ber. 1931, 64, 2844Ϫ2856.
[15c]
Eames, M. Watkinson, Angew. Chem. Int. Ed. 2001, 40,
3567Ϫ3571.
F. Haber, J. Weiss, Naturwiss. 1932, 51, 948Ϫ950.
F.
Haber, J. Weiss, Proc. R. Soc. 1934, 147, 332Ϫ351.
P. J. Domaille, R. L. Harlow, S. D. Ittel, W. G. Peet, Inorg.
Chem. 1983, 22, 3944Ϫ3952.
W. O. Siegl, F. C. Ferris, P. A. Mucci, J. Org. Chem. 1977,
42, 3442Ϫ3443.
Nonius OpenMoleN, Interactive Structure Solution, Delft, 1997.
D. T. Cromer, J. T. Waber, International Tables for X-ray Crys-
tallography, The Kynoch Press, Birmingham, 1974.
Received January 8, 2004
[3] [3a]
[16]
[17]
M. S. Kharasch, P. Pauson, W. Nudenberg, J. Org. Chem.
[3b]
1953, 18, 322Ϫ327.
W. Treibs, G. Pellmann, Chem. Ber.
M. S. Kharasch, A. Fono, J. Org.
[3c]
1954, 87, 1201Ϫ1205.
Chem. 1958, 23, 324Ϫ325. [3d] M. S. Kharasch, A. Fono, J. Org.
Chem. 1959, 24, 72Ϫ78.
[18]
[19]
[4] [4a]
M. Schulz, R. Kluge, F. G. Gelalcha, Tetrahedron: Asym-
metry 1998, 9, 4341Ϫ4360. [4b] M. Schulz, R. Kluge, S. Liebsch,
J. Lessig, M. Halik, F. G. Gelalcha, Tetrahedron 1996, 52,
13151Ϫ13166.
Early View Article
Published Online May 5, 2004
[5]
J. A. Elvidge, R. P. Linsted, J. Chem. Soc. 1952, 5008Ϫ5012.
2722
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2716Ϫ2722