Oligomeric cadmium-phthalocyanine complexes: Novel supramolecular free radical structures

Article abstract of DOI:10.1002/chem.200700745

Certain cadmium-metallated phthalocyanines give rise to EPR active triple-decker sandwich complexes containing two Cd ions and three phthalocyanine (Pc) ligands. These have been shown to form when the ligands bear either eight nonperipheral alkyl or alken

Full text of DOI:10.1002/chem.200700745

DOI: 10.1002/chem.200700745
Oligomeric Cadmium–Phthalocyanine Complexes: Novel Supramolecular
Free Radical Structures
[
a]
Isabelle Chambrier, Gaye F. White, and Michael J. Cook*
Abstract: Certain cadmium-metallated
phthalocyanines give rise to EPR
active triple-decker sandwich com-
plexes containing two Cd ions and
three phthalocyanine (Pc) ligands.
These have been shown to form when
the ligands bear either eight non-
peripheral alkyl or alkenyl substituents
or eight peripheral 2-ethylhexyl groups.
They can be derived either from three
equivalents ofa cadmium phthalo-
series of“cross” experiments. The re-
either the added CdPc or, and more
sults indicate that the triple-decker
structures are formed by a self-assem-
bly process. This is deduced from re-
sults that show that they can disassem-
ble and reassemble with incorporation
of differently substituted ligands de-
surprisingly, the H Pc compound. Mass
2
spectrometry has also established that
higher order oligomers can be formed
when steric requirements between the
alkyl substituents on adjacent rings in
the stack are reduced. Thus an isotopic
cluster for a Cd Pc complex has been
rived from either an H Pc or CdPc.
2
5
6
The reassembled structures in these
cross experiments can contain more
than one ligand that originated from
observed when the eight peripheral
substituents are hexyl chains and tetra-
meric complexes are formed when two
different ligands are incorporated
within a stack, with one carrying sub-
stituents at the peripheral sites and the
other bearing substituents at the non-
peripheral sites.
A
C
H
T
R
E
U
N
G
cyanine precursor or from a 2:1 mix-
ture ofa cadmium phthalocyanine
CdPc) and metal-free phthalo-
cyanine (H Pc). The mode oftheir of r-
Keywords: cadmium · mass spec-
(
a
trometry
·
phthalocyanines
·
ACHTREUNG
2
sandwich complexes
mation has been investigated by a
2À
Introduction
the central cavity ofthe macrocyclic ligand (Pc ) and
through introduction ofsubstituents onto the ring system.
Phthalocyanines (Pcs) are man-made macrocycles that play
Common metallated derivatives include dilithium phthalo-
[1]
II
a major role as commercial dyes and pigments. However,
they are also recognised, by virtue oftheir interesting photo-
physical and electrical properties, to be highly significant
cyanine [Li Pc] and 1:1 complexes ofthe ligand with M
2
II
metals to form [M Pc] complexes. Incorporation oftrivalent
III
III
metal ions, for example, Al or In , leads to metallated Pcs
III
“functional materials”. Thus specific derivatives are used as
bearing an axial ligand. Valencies ofM ions are also satis-
charge carriers in photocopiers, as dyes in laser/LED print-
ing and as laser-light absorbers for optical data storage sys-
fied within triple-decker sandwich complexes, [M Pc ], but
2
3
[11]
examples, which include derivatives oflanthanide (III),
A
C
H
T
R
E
U
N
G
[2–5]
[12]
[13]
tems such as CD-Rs.
Others show promise for exploita-
indium
A
H
R
U
G
A
H
R
U
G
[6]
[7]
tion in areas, such as solar cells and gas sensors, while
photoexcited state properties render them suitable for opti-
An important but small sub-group ofPcs are those in
which the formal charge(s) of the ligand do not match those
ofthe normal valencies ofthe metal ions, resulting in rf ee-
radical derivatives. These compounds are especially promis-
ing in the fields of molecular electronics and sensor develop-
[
8]
cal-limiting
and as singlet-oxygen photosensitizers for
[9]
[10]
water purification and photodynamic therapy (PDT).
This range ofapplications arises rf om the ease ofintro-
ducing or tuning specific properties through the incorpora-
tion of up to 70 different metallic or metalloid elements into
[14]
ment. Structure-wise, the simplest example is monolithi-
[
15,16]
um phthalocyanine [LiPc], formed by chemical,
electro-
[15,17]
[15]
chemical or photochemical oxidation of[Li Pc]. It is
an unexpectedly stable radical and one ofthe rare intrinsic
organic semiconductors. The second class ofradical
phthalocyanine are bisphthalocyanines, [M Pc ], formed
2
[
a] Dr. I. Chambrier, Dr. G. F. White, Prof. M. J. Cook
School ofChemical Sciences and Pharmacy
University ofEast Anglia, Norwich NR4 7TJ (UK)
Fax : (+44)1603-592-011
[18]
III
2
[19–22]
[23]
most notably from the lanthanides
and indium. Un-
E-mail: m.cook@uea.ac.uk
doped single crystals of[LuPc ] show significant room-tem-
2
7608
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7608 – 7618

FULL PAPER
[18]
[26]
perature conductivity
and, along with other lanthanide
the literature
although the unsubstituted 1:1 cadmium
bisphthalocyanines, exhibit electrochromism arising from
one-electron transfers between readily accessible redox
phthalocyanine complex is available from a commercial
source. The first indications of the formation of the triple-
decker sandwich complex 1a arose from a reaction expected
to convert the metal-free derivative 2a, a compound substi-
tuted with eight hexyl chains at the non-peripheral
[24]
states.
In a recent communication we isolated the first example
[25]
ofa new class ofphthalocyanine derivative, 1a, Figure 1.
1
,4,8,11,15,18,22,25-sites on the ring system, into the corre-
sponding substituted cadmium phthalocyanine derivative 3a
Scheme 1). Though the isolation ofpure 3a was satisfacto-
(
rily achieved (see next section), it was not without initial dif-
ficulties because of frequent contamination with either the
metal-free precursor 2a or a dark blue material that ulti-
mately proved to be 1a.
[
25]
Figure 1. X-Ray structure of 1a.
The compound combines the triple-decker architecture re-
ferred to above but with a metal(II) ion, in this case cadmi-
um. The complex, in which there is a formal charge imbal-
2
À
2+
ance between the three Pc ligands and two Cd ions, ex-
hibits a remarkable set ofredox states; it is also EPR active
with signal intensity in CH Cl indicating the presence of
2
2
less than one molar equivalent ofa radical species. The
cyclic voltammogram ofthe material in CH Cl provides evi-
Scheme 1. Synthesis of 1a–1d. Reagents: i) Cd
crystallisation CH Cl /methanol.
A
H
R
U
G
2
(OAc) , pentanol; ii) re-
2
2
2
2
dence ofsix one-electron oxidation and twelve one-electron
reduction processes, which, in view ofthe redox inactivity of
cadmium, are attributed to ring-based processes. The finding
that 1a in the rest state can undergo six ring-based one-elec-
Thus an attempt to separate 3a from the crude reaction
product by column chromatography over silica gel led to
both 3a and a fraction that proved to be the starting materi-
al 2a. The latter arose apparently through de-metalation of
3a on the column. When neutral alumina was used as the
stationary phase a minor dark blue fraction was eluted prior
to 3a. The blue material also appeared as a contaminant of
3a during preliminary recrystallisation attempts. Further-
more, a green solution of 3a in chloroform very slowly
changed colour to dark blue on standing. The dark blue
compound was subsequently obtained pure when it crystal-
lised out unexpectedly during attempted slow recrystallisa-
tions of 3a from either THF/MeOH or CH Cl /MeOH; its
2À
tron oxidations implies each ligand is indeed in the Pc
state. Thus the species in the neutral state can be regarded
as having the general structure [H] [Cd Pc ], in which
AHCTREUNG
2
2
3
“
[H] ” refers to protons serving as counterions. The first
2
o
one-electron oxidation from this spin 0 state occurs at E ’=
.162 V, much lower than is typical ofsimple Pc derivatives,
0
and could account for the presence of amounts of an oxi-
dised, spin 1/2 species, leading to the EPR signal.
The present paper reports conditions that favour the for-
mation ofthis new class ofmaterial, the development ofthis
novel chemistry through synthesis ofanalogues, the investi-
gation ofthe sel -f assembling processes involved in generat-
ing these triple-decker complexes and the detection of
higher oligomers.
2
2
structure, Figure 1, was identified as 1a by X-ray crystallog-
[25]
raphy.
The UV/Vis spectra of 3a in hexane and 1a in various sol-
vents are shown as Figure 2. The former gives rise to the
strong Q-band absorption at 708 nm that is characteristic of
a metallated phthalocyanine substituted with alkyl chains.
However, the spectrum of 1a is more complex with elec-
tronic absorptions at 867, 718, 646, 599 and 483 nm (in
hexane); the relative intensities ofsome ofthese bands vary
Results and Discussion
First isolation of compound 1a and characterisation data:
There are very few reports on cadmium phthalocyanines in
Chem. Eur. J. 2007, 13, 7608 – 7618
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7609

M. J. Cook et al.
Optimised conditions for the synthesis of 1a: A reproduci-
ble protocol for obtaining 1a was first sought based on
Scheme 1, with the twin objectives ofoptimising the isola-
tion ofpure 3a and establishing conditions for its conversion
into the triple-decker sandwich complex. The first was ach-
ieved by reacting the metal-free compound, 2a, in refluxing
pentanol with a twofold excess of cadmium acetate, with
reflux maintained for 45 minutes. The hot solution was
added to cold methanol and the mixture left at 48C over-
night. The green precipitate was collected and washed with
MeOH, and the cadmium-metallated compound, 3a, was
separated from excess cadmium acetate by dissolving it into
THF. Rapid addition ofexcess MeOH to the solution re-
precipitated 3a as pure material, uncontaminated by 1a,
and was characterised by elemental analysis, MALDI-TOF
mass spectrometry and NMR spectroscopy.
A protocol for converting 3a into 1a was developed that
involved dissolving 3a in dichloromethane containing a few
drops ofMeOH. The solution slowly turned blue over a
period of15 h at 4 8C. UV/Vis spectroscopic monitoring
over this period showed the gradual formation of 1a. Com-
pound 1a crystallised out upon addition of fu rther methanol
in 73% yield. Parallel experiments demonstrated the impor-
tance of the addition of the first few drops of MeOH to the
solution of 3a in dichloromethane. Excess methanol merely
precipitated 3a from solution; with no methanol added,
peaks corresponding to the formation of the triple-decker
1a were not detected by UV/Vis spectroscopy over 24 h.
The fate of the displaced cadmium ion during the conver-
sion of 3a into 1a is uncertain. We saw no evidence for pre-
cipitated material that could point to the formation of cad-
mium metal or cadmium oxide and we postulate that it may
be taken up as cadmium methoxide.
Figure 2. a) UV/Vis spectrum of 3a in hexane; b) UV/Vis spectra of 1a
in THF (dark blue), toluene (light blue), CH Cl (red) and hexane
green) at the same concentration (15mm).
2
2
(
somewhat in other solvents, for example, THF and toluene,
Figure 2b. The EPR activity of 1a, which was referred to
above, and the intensity ofthe radical signal also shows
some solvent dependency. Measurement ofsignal intensity
relative to that from a known concentration of a standard
In a variant ofthe protocol of r obtaining 1a, the same
procedure was applied to a solution (in CH Cl /MeOH) con-
2
2
spin
label
[(1-oxyl-2,2,5,5-tetramethyl-[D]-pyrroline-3-
taining the two formal components of the triple-decker
methyl)methanethiosulfonate; MTSL], showed the radical
content, that is, mole fraction of radical, to vary as follows:
THF 0.18 (g=1.9934), toluene 0.45 (g=1.9952) and hexane
sandwich compound, that is, two equivalents of 3a and one
equivalent ofthe metal -r fe e analogue,
2a. This mixture
indeed provided an alternative access to 1a, which was iso-
lated in 76% yield, comparable to that obtained from 3a
alone. This result establishes that a metal-free Pc can partic-
ipate in the formation of the triple-decker sandwich com-
plex, a result which is explored further later in this paper.
In a separate experiment, an attempt was made to obtain
1a directly from the metal-free compound 2a by using a fi-
vefold excess of cadmium acetate. The reaction yielded a
blue powder, but this proved not to be 1a. The UV/Vis
0.36 (g=1.9938). We surmise that both the variation in ab-
sorption intensities ofspeci fc UV/Vis bands and radical
content arise from solvent-induced perturbation of the equi-
librium between spin 0 and spin 1/2 species referred to earli-
er.
The presence ofa paramagnetic species accounts of r line
1
broadening that is observed in the H NMR spectrum of 1a.
The spectrum measured ofa solution of 1a in [D ]benzene
6
showed very broad and/or unresolvable NMR signals; that
spectrum showed a broad absorption peak at 721 nm and
1
measured in [D ]chloroform was somewhat better resolved
the product gave a H NMR spectrum in [D ]benzene that
3
6
and showed a broad signal for the aromatic protons. Howev-
er, there are no detectable signals in the “benzylic region”
ofthe spectrum; the most downi fe ld ofthe non-aromatic
proton signals appear at 2.65 ppm. The MALDI-TOF mass
spectrum of 1a shows peaks that can be assigned to frag-
mentation ofthe molecular ion, which itselfis not detected
did not show peak broadening ofthe kind observed in the
spectrum of 1a. The NMR spectrum is complex but well re-
solved, Figure 3. The signals are tentatively assigned to a
triple-decker sandwich complex, but capped with cadmium
monoacetate at both ends (structure 4). Two aromatic sin-
glets at 7.96 and 8.65 ppm integrate for 16H and 8H, re-
spectively, and there are signals at 4.31, 3.54 and 3.30 ppm
each integrating for 16H and correspond to the benzylic
(
isotopic clusters at 2598.7 [Cd Pc ]; 2484.8, [CdPc ], 1299.1-
2 2 2
A
C
H
T
R
E
U
N
G
[CdPc] and 1187.1 [H Pc]).
2
7610
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7608 – 7618

Phthalocyanine Complexes
FULL PAPER
corresponding non-peripherally substituted cadmium-phtha-
locyanine complexes, 3b, 3c and 3d. However, attempts to
metallate the corresponding hexyloxy and hexyloxymethyl
derivatives, 2e and 2 f, with cadmium acetate gave red deriv-
atives that were insoluble in common organic solvents. Cad-
mium is known to have an affinity for oxygenated com-
pounds and at this time we have not attempted to identify
the chemistry involved in these reactions.
The three new cadmium-phthalocyanine complexes 3b–d
were successfully converted into triple-decker sandwich
compounds, 1b–d, respectively, in good yields either by al-
lowing them to stand alone or to self-assemble with an ap-
propriate amount ofthe metal- rf ee analogue under the two
sets of conditions reported above for the formation of 1a.
Evidence for their formation followed from their UV/Vis
spectra, which were superimposable on that for 1a, satisfac-
tory elemental analysis data and MALDI-TOF spectral
data, which showed fragmentation peaks that corresponded
to those observed for the fragmentation of the molecular
1
Figure 3. Aromatic and benzylic regions ofthe H NMR spectrum of 4
measured in [D ]benzene.
6
1
ion of 1a. Though the resolution ofthe H NMR signals for
protons. The protons giving rise to
the signals at 4.31 and 3.30 ppm are
coupled (COSY) and are assigned
to the two sets ofdiastereotopic
protons that arise from the non-
equivalence ofthe two af ces ofthe
outer Pc ligands. The signal at
the aromatic and benzylic protons suffered from the pres-
ence ofa paramagnetic species, the spectrum of 1d clearly
showed two sets ofsignals (ratio 2:1) arising rf om the pro-
tons attached to the terminal alkene units. These protons
are evidently too distant from the core of the molecule to
experience the paramagnetism ofthe fr ee radical.
3
.54 ppm is then assigned to the
Peripherally octasubstituted derivatives: Phthalocyanine de-
rivatives with substituents at the 2,3,9,10,16,17,23,24 (or pe-
ripheral) sites differ from those bearing comparable sub-
stituents at the non-peripheral sites in that they generally
show lower solubility and a greater tendency to aggregate in
solution. This presumably reflects the fact that the substitu-
ent chains are better accommodated broadly within the
plane containing the Pc nucleus: X-ray structural determina-
tions ofnon-peripherally octaalkyl-substituted compounds
show that at least two alkyl chains are forced out of the
benzylic protons ofthe inner Pc
ring. In [D ]chloroform, a signal at
3
À0.82 ppm, integrating for 6H, was
also apparent and is tentatively as-
signed to the acetate groups ofthe
CdOAc-capping moieties. The clos-
est reported structural analogue to
that proposed above which we are
aware ofis a polymeric Pc complex
in which the rings are linked by
[28]
plane and lie approximately orthogonal to the ring plane.
2
+
Hg ions with the terminal rings
capped by Hg
units.
Accordingly, the location ofthe substituents at the peripher-
al sites could experience lower steric strain within sandwich-
type structures.
II
monoacetate
[27]
Unfortunately, it proved impossible to isolate the new oli-
gomeric complex, 4. All attempts to remove the excess cad-
mium acetate used in the reaction led to decomposition of
the product. When a solution ofthe product was le tf in di-
chloromethane for a few hours, the absorption spectrum
slowly developed the characteristics ofthe UV/Vis spectrum
of 1a. However, we did not seek to develop this finding as
an alternative protocol for the synthesis of 1a.
Two Pc ligands bearing eight alkyl chains at the peripheral
sites were selected for investigation, one with 2-ethylhexyl
substituents and the other with less bulky n-hexyl chains.
Thus the metal-free derivatives 5a and 5b were converted
into the corresponding cadmium phthalocyanines, 6a and
6b, Scheme 2. These were satisfactorily characterised and
showed Q-bands at 686 and 688 nm, respectively, that is, to
the blue ofthe corresponding band exhibited by compounds
3
a–d, and to be expected as a consequence of the different
[29]
Syntheses and attempted syntheses of homologues and ana-
logues of 1a
locations ofthe substituents.
As was found for the cadmium phthalocyanine derivatives
3
a–3d, a solution of 6a in dichloromethane turned blue on
Non-peripherally octasubstituted derivatives: Cadmium inser-
addition ofmethanol and a new compound crystallised out
as microcrystals. The UV/Vis spectrum ofthis compound,
Figure 4a, showed a complex band structure a little different
tion into metal-free 1,4,8,11,15,18,22,25-octakis
octakis(decyl)- and octakis(octenyl)phthalocyanines (2b, 2c
and 2d, respectively) was achieved satisfactorily to give the
A
H
R
U
G
A
C
H
T
R
E
U
N
G
ACHTREUNG
1
from that exhibited by compounds 1a–d. Its H NMR spec-
Chem. Eur. J. 2007, 13, 7608 – 7618
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7611

M. J. Cook et al.
[
LuH(Pc) ] or H[Lu(Pc) ], has also proved difficult to
2
2
[11]
detect. The spectrum of 7a included fragmentation peaks
corresponding to [Cd Pc ] and [CdPc].
2
2
A solution ofthe second peripherally substituted cadmi-
um phthalocyanine 6b in dichloromethane also gave a dark
blue precipitate when treated with methanol, as above.
However, the UV/Vis spectrum differed from that exhibited
by 7a, Figure 4b, implying the formation of an alternative
type ofsandwich compound. This was coni fr med by the
MALDI-TOF mass spectrum, which showed a remarkably
complex series ofsignals, Figure 5. The highest molecular
Scheme 2. Synthesis ofperipherally substituted derivatives. Reagents: i)
Cd(OAc) , pentanol; ii) recrystallisation CH Cl /methanol.
A
C
H
T
R
E
U
N
G
2
2
2
Figure 5. MALDI mass spectrum ofthe product resulting rf om the re-
crystallisation of 6b from dichloromethane/MeOH. The 7676.7 cluster
corresponds to a Cd
the 5081.3 cluster to a Cd
and the 3783.5 cluster to a Cd
5
Pc
6
complex, the 6377.0 cluster to a Cd
Pc complex, the 3895.4 to a Cd
Pc complex.
4
Pc
Pc
5
complex,
3
complex
3
4
3
2
3
weight species, m/z 7676.7, corresponds to a structure with
six ligands and five cadmium ions; there are also peaks that
correspond to lower oligomers. Whether these signals corre-
spond to fragmentation peaks or to more than one species
in the recovered product is uncertain. Elemental analysis
did not clarify the situation other than to show that a
[
Cd Pc ] structure alone would not account for the experi-
5 6
mentally determined C, H, and N content. A solution ofthe
product in n-hexane gave rise to an EPR signal.
The formation of structures more complex than a triple-
decker sandwich when the Pc ligand bears less spatially de-
manding groups, that is, hexyl chains, implies that the extent
ofsteric interactions between adjacent ligands in a stack
may be a controlling factor in the oligomerisation process.
Figure 4. a) UV/Vis spectrum of 7a; b) UV/Vis spectrum ofthe product
resulting from the recrystallisation of 6b using the dichloromethane/
MeOH procedure.
trum was poorly resolved; the EPR spectrum (hexane as sol-
vent) showed qualitatively a signal at g=1.9985. The com-
pound was assigned structure 7a on the basis ofelemental
analysis and the MALDI-TOF mass spectrum. Unlike the
spectra of 1a–d, compound 7a gave an isotopic cluster
Other analogues: Metallated tetra-tert-butylphthalocyanines
have been much studied as readily accessible derivatives
soluble in organic solvent and accordingly we investigated
the hitherto unknown cadmium derivative 8. This was pre-
pared from the metal-free analogue by the usual procedure,
but it proved to have unexpectedly low solubility in chloro-
form and methylene chloride, so hindering attempts to con-
vert this into a triple-decker sandwich compound.
(
100%) corresponding to a [Cd Pc ] structure. There was no
2 3
evidence, however, for additional protons in the mass spec-
trum; we note that the “balancing proton” in the blue form
oflutetium bis-phthalocyanine, usually denoted as either
7612
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7608 – 7618

Phthalocyanine Complexes
FULL PAPER
of the latter offering a probe for readily establishing its in-
1
corporation into a triple-decker structure. The H NMR
spectrum ofthe triple-decker complex(es) in the product
mixture indeed showed signals for terminal alkenyl protons;
moreover it showed signals for alkene groups in two differ-
ent environments in a ratio of2:1. This demonstrated that
the ligand ofthe metal- rf ee complex 2d had not only been
incorporated into the triple-decker complex, but also shows
that it is incorporated at both the outer and inner sites. This
result is significant with regard to the formation of triple-
We also undertook a briefinvestigation into the potential
for converting a cadmium porphyrin derivative into a sand-
wich complex, focussing on the tetrakis(3,5-di-tert-butylphe-
nyl)porphyrin ligand. The metal-free analogue was convert-
ed into the cadmium derivative 9, which proved to have sat-
isfactory solubility in chloroform and dichloromethane to
apply the conditions that promoted formation of phthalo-
cyanine triple-decker complexes. However, these conditions
failed to induce a change in the UV/Vis spectrum of 9 over
several days.
decker complexes from 2:1 mixtures of a CdPc and an H Pc.
2
In particular, it precludes a mechanism by which the metal-
free complex merely serves to “cap” an intermediate dimer-
ic structure ofthe type Cd Pc . As expected from the previ-
2
2
ous results for non-peripherally substituted complexes, the
MALDI-TOF spectrum did not show molecular ions, but
showed peaks assignable to the fragment ions containing
both Pc-Cd-Pc and Pc-Cd-Pc’ moieties.
Experiment 2, utilising 3a as the CdPc component and 5a
In the light ofthese investigations it appears that the of r-
mation ofoligomeric cadmium phthalocyanines may be re-
stricted to certain specifically substituted phthalocyanine li-
gands.
as the H Pc’, provided further insight. The MALDI-TOF
2
spectrum ofthe precipitated complexes showed at higher
mass numbers evidence for structure(s) containing four li-
gands: two non-peripherally substituted ones from the CdPc
compound and two peripherally substituted ones arising
Mixed precursor experiments—probes for understanding
oligomer formation: A variety ofexperiments were under-
taken involving two differently substituted cadmium
from the H Pc’ compound. The sequence ofthe ligands in
2
this “2:2” tetrameric stack cannot, ofcourse, be surmised
from the mass spectrum. However, the formation of a tetra-
meric structure containing ligands that, alone, give rise only
to triple-decker complexes (1a and 7a) can be rationalised
in terms ofreduced steric interactions between adjacent li-
gands when one is non-peripherally substituted and the
other peripherally substituted. The second significant isotop-
ic cluster in the mass spectrum corresponds, remarkably, to
a triple-decker complex (7a) containing exclusively the li-
phthalocyanines, referred to as CdPc and CdPc’, and unsur-
A
C
H
T
R
E
U
N
G
prisingly the triple-decker complexes formed from these
gave mass spectral data consistent with the incorporation of
both precursors. Typical evidence from experiments with
two different non-peripherally substituted compounds were
fragmentation peaks corresponding to Pc-Cd-Pc’ species.
In light ofthe of rmation of 1a from a solution containing
two equivalents ofthe substituted cadmium phthalocyanine
gands from the H Pc’ compound.
2
3
a and one equivalent ofthe metal- rf ee analogue 2a, vide
These results, showing the formation of multi-ligand com-
plexes derived either in part or in whole from a metal-free
compound, require one or more processes whereby at some
stage a Cd ion from a CdPc molecule becomes available to
supra, we undertook further mixed precursor experiments to
investigate the mode o fof rmation oftriple-decker com-
plexes by this protocol. These experiments, exemplified by
entries 1 and 2 in Table 1, were undertaken by dissolving
two equivalents ofa substituted CdPc and one equivalent of
a differently substituted metal-free phthalocyanine, denoted
link ligands from H Pc’ molecules. A simple process would
2
involve Cd ion transfer between CdPc and H Pc’ molecules
2
via an intermediate ofthe of rm [Pc-Cd-Pc ’·2H] leading ulti-
mately to a scrambling ofthe Cd ion between the two li-
gands Pc and Pc’. However, although this process cannot be
ruled out, the results ofexperiments 3–6 suggest an alterna-
tive mechanism may prevail that involves triple-decker com-
plexes themselves.
as H Pc’, in dichloromethane as solvent, to which were
2
added two drops ofmethanol. Further methanol was added
after 15 h to precipitate out oligomeric complexes that had
been formed in the solution.
Experiment 1 was undertaken with 3a as the CdPc com-
ponent and 2d as the H Pc’ component, the alkenyl protons
2
Table 1. Results from mixed experiments.
Expt No
Component 1 (2 equiv)
Component 2 (1 equiv)
Selected observations re components ofthe product mixture ofoligomers
1
1
2
3
4
5
6
CdPc (3a)
CdPc (3a)
H
H
H
2
2
2
Pc’ (2d)
Pc’ (5a)
Pc’ (5a)
scrambling ofligand fr om 2d into a triple-decker structure ( H NMR)
formation of ’2:2 tetramers’ with two ligands derived from both 3a and 5a
incorporation ofligand fr om 5a into oligomer
Cd
Cd
Cd
Cd
2
2
2
2
Pc
Pc
Pc
Pc
3
3
3
3
(1a)
(1a)
(7a)
(7a)
CdPc’ (6a)
Pc’ (2a)
CdPc’ (3a)
incorporation ofligand fr om 6a into oligomer
evidence of fo rmation of ’2:2 tetramers’
fragmentation peak corresponding to Pc-Cd-Pc’
H
2
Chem. Eur. J. 2007, 13, 7608 – 7618
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7613

M. J. Cook et al.
These experiments explored the effect of mixing a pre-
formed triple-decker complex, Cd Pc , with a differently
triple-decker sandwich complex containing two Cd ions and
three phthalocyanine ligands. These have been shown to
form when the ligands bear either eight non-peripheral alkyl
or alkenyl substituents or eight peripheral 2-ethylhexyl
groups. The novel triple-decker structures have been fully
characterised by elemental analysis and supported by
MALDI-TOF mass spectrometry data. The complexes are
EPR active and as such add an interesting new sub-class of
compound to the already substantial range ofknown phtha-
locyanine-based derivatives.
2
3
substituted H Pc’ or CdPc’ in CH Cl /MeOH. Experiments 3
2
2
2
and 4 involved the triple-decker 1a admixed with, separate-
ly, 5a and 6a and precipitating the components ofthe mix-
ture by addition ofexcess MeOH a ft er 24 h. The mass spec-
trum ofthe mixture fr om experiment 3 is shown in Figure 6.
Mass spectrometry has also established that higher order
oligomers can be formed when steric requirements between
the alkyl substituents on adjacent rings in the stack are re-
duced. Thus an isotopic cluster for a Cd Pc complex has
5
6
been observed when the eight peripheral substituents are
hexyl chains and tetrameric complexes are formed when
two different ligands are incorporated within a stack, with
one carrying substituents at the peripheral sites and the
other bearing substituents at the non-peripheral sites.
Investigations ofthe optimised synthesis ofthe complexes
have focused on the preparation of the triple-decker materi-
als. These can be derived either from three equivalents of a
simple cadmium phthalocyanine derivative or from a 2:1 mix-
ture ofa cadmium phthalocyanine and a metal- rf ee phthalo-
cyanine. The mode oftheir of rmation has been investigated
by a series of“cross” experiments. The results indicate that
the triple-decker structures are formed by a self-assembly
process. This is deduced from results that show that they can
disassemble and reassemble with incorporation ofdi ef rently
Figure 6. MALDI mass spectrum ofthe mixture rf om experiment 3. The
5
750.9 cluster corresponds to a Cd
Cd Pc’ Pc complex, the 4453.9 cluster to a Cd
cluster to a Cd Pc’Pc complex, the 4006.4 cluster to a Cd
3
Pc’
3
Pc complex, the 5528.6 cluster to a
Pc’ complex, the 4229.7
Pc Pc’ complex,
2
complex and the 2708.7 cluster to a
3
2
2
2
3
3
2
2
2
the 2932.9 cluster to a CdPc’
CdPcPc’ complex (see Table 1).
It is very similar to that obtained from experiment 2. Thus it
shows the presence ofthe triple-decker containing all three
ligands from the metal-free compound. In addition there are
isotopic clusters corresponding to the 2:2 tetramers, referred
to above, and to 1:3 tetramers containing one ligand from
the starting Cd Pc complex 1a and three from the added
metal-free compound. The mass spectrum of the product
from experiment 4 was also complex and included a cluster
corresponding to a mixed triple-decker.
substituted ligands derived from either an H Pc or CdPc. The
reassembled structures in these cross experiments can contain
more than one ligand that originated from either the added
2
CdPc or, and more surprisingly, the H Pc compound.
2
Ongoing work is now focusing on the investigation of the
redox properties ofthe complexes described herein and the
potential applications ofthe materials in device develop-
ment. In particular further work is in progress to expand on
preliminary findings that 1a exhibits remarkable tempera-
2
3
[30]
Experiments 5 and 6 were undertaken for comparative
ture- and voltage-dependent semiconductivity.
purposes from the triple-decker 7a admixed with the H Pc’
2
or CdPc’ compounds, 2a and 3a, respectively. The mass
spectrum ofthe precipitate rf om experiment 5 included a
cluster for the 2:2 tetramers, while that for experiment 6 in-
cluded a cluster for a mixed dimer, Pc-Cd-Pc’, presumably
as a fragmentation peak.
The results ofthis series ofexperiments provide strong
evidence that in the solution phase, the triple-decker com-
plexes can disassemble and the components reassemble with
Experimental Section
General: All solvents were SLR or analytical grade and were used as
supplied. Commercially available materials were used without further pu-
rification. Infrared spectra were recorded as powders on a Perkin–Elmer
Spectrum BX equipped with a DuraSamplIR II ATR attachment. NMR
1
13
spectra were recorded at 400 MHz for H and 75.4 MHz for C on a
Varian Unity plus spectrometer. The residual solvent peak was used as
6 3
reference (7.15 ppm for [D ]benzene and 7.27 ppm for [D ]chloroform).
UV/Vis spectra were recorded on a Hitachi U-3000 spectrophotometer in
the solvent stated. Melting points and phase transition temperatures
inclusion ofthe added CdPc ’ or H Pc’ to generate new
2
mixed triple-deckers.
(
crystal, K; discotic mesophase, D; isotropic liquid, I) were recorded
using an Olympus BH-2 polarising microscope fitted with a Linkham
THM 600 hotstage. Microanalysis data were obtained using a Carlo Erda
Conclusion
1
106 Elemental Analyzer and are quoted to the nearest 0.01%. MALDI-
TOF mass spectra were recorded using an Applied Biosystems Voyager-
DE-STR at the EPSRC National Mass Spectrometry Service Centre at
the University ofWales in Swansea (UK), with DCTB (trans-2-[3-(4- tert-
butylphenyl)-2-methylprop-2-enylidene] malononitrile) as matrix. EPR
Our research has uncovered some novel and unexpected
chemistry ofcadmium phthalocyanines. Spec ii fc cadmium-
metallated derivatives give rise to a remarkable type of
7614
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7608 – 7618

Phthalocyanine Complexes
FULL PAPER
measurements were carried out at 295 K on a Bruker E500 continuous
wave X-band EPR spectrometer fitted with a Bruker ER4123D dielectric
room temperature resonator. (1-Oxyl-2,2,5,5-tetramethyl-[D]-pyrroline-3-
methyl)methanethiosulfonate (MTSL) was used as standard.
Method 2: Compounds 3a (45 mg, 34.7 mmol) and 2a (20.6 mg,
17.3 mmol) were dissolved in dichloromethane (4.5 mL) and five drops of
methanol added. The solution was placed at 48C for 20 h. After that time
more methanol was added to precipitate the dark blue powder (50.2 mg,
7
7%).
Materials: Metal-free phthalocyanine precursors 2a, 2b, 2c, 2e and 2 f
were available from previous reported work on these compounds.
[
31–33]
Tris[1,4,8,11,15,18,22,25-octakis(octyl)phthalocyaninate]dicadmium (1b):
A
H
R
U
G
The precursor to compound 8 was available from Aldrich. The free-base
precursor to compound 9 was prepared by condensation of3,5-di- tert-bu-
tylbenzaldehyde and pyrrole. Compound 2d was prepared from furan
The product was prepared as for 1a (method 1) from 3b. A pure dark
blue powder was obtained after one recrystallisation of the crude product
from CH Cl /MeOH (80%) M.p. 1518C (decomp); IR (neat, ATR): n˜ =
2 2
[
34]
using a route first described for the synthesis of a lower saturated homo-
2915.51, 2846.28, 1604.47, 1365.32, 1299.84, 1271.82, 1189.43,
[
31]
À1
logue.
,4,8,11,15,18,22,25-Octakis
typical experiment, 1,4,8,11,15,18,22,25-octakis
500 mg, 0.42 mmol) in pentan-1-ol (20 mL) was brought to reflux. Cad-
1097.93 cm ; UV/Vis (n-hexane): lmax =872, 719, 644, 599, 547, 484 nm;
+
MALDI-MS: isotopic clusters at m/z (%): 3043.1 (2) [MÀPc] , 2933.2
1
A
H
R
U
G
a
+
+
+
(
12) [MÀCdPc] , 1523.1 (100) [MÀCdPc
2
] , 1411.2 (45) [MÀCd
2 2
Pc ] ;
AHCTREUNG
elemental analysis calcd (%) for C288
432 2
H N24Cd : C 77.64, H 9.77, N 7.54;
(
found: C 77.34, H 9.75, N 7.31].
Tris[1,4,8,11,15,18,22,25-octakis(decyl)phthalocyaninate]dicadmium (1c):
The product was prepared as for 1a (method 1) from 3c. The crude prod-
uct was dissolved in hot CH Cl . Upon cooling, compound 3c crystallised
mium acetate hydrate (99.99+ %; 2 equiv, excess) was added and reflux
was continued for 45 min. The hot mixture was added to excess cold
MeOH (150 mL) and the flask left in the fridge overnight. The green
solid was filtered and washed with MeOH. The green solid was redis-
solved in THF and the solution filtered to remove the insoluble cadmium
salt residue. The solvent was then removed under reduced pressure and
AHCTREUNG
2
2
and was removed by filtration. The blue mother liquor was concentrated
and methanol was added. A dark blue powder crystallised (15%). M.p.
1098C (decomp); IR (neat, ATR): n˜ =2915.73, 2846.2, 1604.07, 1465.67,
the solid reprecipitated from excess MeOH to yield a green powder
À1
À1
(
350 mg, 64%). M.p. (DSC transition energies): K–D 131.18C (0.75 Jg ),
1366.71, 1264.2, 1189.6, 1096.9 cm ; UV/Vis (n-hexane): lmax =718, 647,
À1
D–D 1668C (1.34 Jg ), D–D 192.88C (undetected), D–I 244.48C
599, 550, 483 nm; MALDI-MS: isotopic clusters at m/z (%): 3380.6 (5)
À1
1
+
+
+
(
0.19 Jg ); H NMR (400 MHz, C
6
D
6
containing a drop of[D
5
]pyridine):
[MÀCdPc] , 1747.3 (100) [MÀCdPc
2
] , 1635.4 (32) [MÀCd
2 2
Pc ] ; ele-
d=7.75 (s, 8H), 4.63 (t, J=7.2 Hz, 16H), 2.2 (m, 16H), 1.6 (m, 16H),
mental analysis calcd (%) for C336
found: C 78.56, H 10.47, N 6.71].
528 2
H N24Cd : C 78.69, H 10.38, N 6.55;
1
3
1
C
.1–1.3 (m, 32H), 0.75 ppm (t, J=7.2 Hz, 24H); C NMR (75.4 MHz,
containing a drop of[D ]pyridine): d=155.38, 138.55, 136.68,
6
D
6
5
Synthesis of metal-free 1,4,8,11,15,18,22,25-octakis(oct-7-enyl)phthalo-
cyanine (2d)
-(Oct-7-enyl)furan: In a typical experiment, nBuLi (2.5m solution in
hexanes, 17.6 mL, 44 mmol) was added to a solution o fu fran (5 g,
3.5 mmol) in dry THF (20 mL) with stirring under N
1
2
30.21, 33.02, 32.43, 31.3, 29.34, 22.85, 14.05 ppm; IR (neat, ATR): n˜ =
952.61, 2920.92, 2851.56, 1313.26, 1151.81, 1084.52, 883.15 cm ; UV/Vis
AHCTREUNG
À1
2
(
n-hexane): lmax =709, 643 nm; MALDI-MS: isotopic cluster at m/z (%):
+
1
298 (100) [M] ; elemental analysis calcd (%) for C80
H 8.69, N 8.63; found: C 74.15, H 8.72, N 8.47.
,4,8,11,15,18,22,25-Octakis(octyl)phthalocyaninatocadmium (3b): The
product was prepared as for 3a from 1,4,8,11,15,18,22,25-octakis-
(octyl)phthalocyanine 2b (93%). M.p. K–D 81.88C, D–I 212.38C;
112 8
H N Cd: C 74.01,
7
2
at À788C. The so-
lution was allowed to warm to RT and stirred for 4 h before being cooled
to À788C. 8-Bromooct-1-ene (12 g, 63 mmol) in dry THF (5 mL) was
added dropwise and the solution allowed to warm to RT and stirred over-
night. The solution was poured into ice/water (100 mL), the organic
1
ACHTREUNG
ACHTREUNG
1
H NMR (400 MHz, C
H), 4.83 (t, J=6.8 Hz, 16H), 2.35 (quint, J=7.6 Hz, 16H), 1.77 (quint,
J=7.6 Hz, 16H), 1.41 (quint, J=7.6 Hz, 16H), 1.26 (quint, J=6.8 Hz,
6 6 5
D containing a drop of[D ]pyridine): d=7.87 (s,
4
phase was extracted with ethyl acetate (3”50 mL), dried (MgSO ), and
filtered; the solvents were then removed under reduced pressure to
8
afford a pale orange oil that was used without further purification
1
3
1
C
6H), 1.18 (m, 32H), 0.79 ppm (t, J=7.2 Hz, 24H); C NMR (75.4 MHz,
containing a drop of[D ]pyridine): d=155.32, 138.55, 136.65,
1
(
12.9 g, 100%). H NMR (60 MHz, CDCl
3
): d=7.3 (m, 1H), 6.3 (m, 1H),
6
D
6
5
5
1
.5–6.2 (m, 2H), 4.8–5.2 (m, 2H), 2.6 (t, J=8 Hz, 2H), 1.2–2.3 ppm (m,
0H).
1
30.20, 33.0, 32.04, 31.36, 30.2, 29.67, 29.54, 22.72, 13.98 ppm; IR (neat,
À1
ATR): n˜ =2952.32, 2918.76, 2849.34, 1313.65, 1147.39, 1083.31 cm ; UV/
2
,5-Bis(oct-7-enyl)furan: In a typical experiment, nBuLi (2.5m solution in
Vis (THF): lmax =707, 642 nm; MALDI-MS: isotopic clusters at m/z (%):
hexanes, 25 mL, 63 mmol) was added to a solution of2-(oct-7-enyl) uf ran
12.9 g, 63 mmol) in dry THF (20 mL) with stirring under N
+
1
523.0 (100) [M] ; elemental analysis calcd (%) for C96
H
144
N
8
Cd: C
(
2
at À788C.
7
5.73, H 9.53, N 7.36; found: C 76.0, H 9.5, N 7.22.
The solution was allowed to warm to RT and stirred for 4 h before being
cooled to À788C. 8-Bromooct-1-ene (13 g, 68 mmol) in dry THF (10 mL)
was added dropwise and the solution allowed to warm to RT and stirred
overnight. The solution was poured into ice/water (100 mL), the organic
1
,4,8,11,15,18,22,25-Octakis(decyl)phthalocyaninatocadmium (3c): The
A
H
R
U
G
product was prepared as for 3a from 1,4,8,11,15,18,22,25-octakis-
(decyl)phthalocyanine 2c (94%). M.p. K–D 81.88C, D–I 177.98C;
ACHTREUNG
1
H NMR (400 MHz, C
H), 4.83 (t, J=6.4 Hz, 16H), 2.35 (quint, J=6.4 Hz, 16H), 1.78 (quint,
J=6.2 Hz, 16H), 1.43 (quint, J=6.8 Hz, 16H), 1.17–1.28 (m, 80H),
6 6 5
D containing a drop of[D ]pyridine): d=7.87 (s,
4
phase was extracted with ethyl acetate (3”50 mL), dried (MgSO ), and
filtered; the solvents were then removed under reduced pressure to
afford a pale orange oil. The unreacted 2-(oct-7-enyl)furan was removed
8
1
3
0
.83 ppm (t, J=6.8 Hz, 24H); C NMR (75.4 MHz, C
6 6
D containing a
by distillation leaving the product, which was used without further purifi-
drop of[D ]pyridine): d=155.39, 138.58, 136.69, 130.22, 33.06, 32.03,
5
1
cation (19.2 g, 100%). H NMR (60 MHz, CDCl
3
): d=5.5–6.2 (m, 2H),
3
1.38, 30.31, 29.95, 29.88, 29.74, 29.51, 22.79, 14.04 ppm; IR (neat, ATR):
5
(
.8 (s, 2H), 4.8–5.2 (m, 4H), 2.6 (t, J=8 Hz, 4H), 2.0 (m, 4H), 1.7 ppm
brs, 16H).
,6-Bis(oct-7-enyl)phthalonitrile: In a typical experiment, 2,5-bis(oct-7-
À1
n˜ =2916.87, 2848.31, 1465.90, 1314.86, 1147.33, 1084.2 cm
; UV/Vis
(
1
THF): lmax =708, 641 nm; MALDI-MS: isotopic cluster at m/z (%):
3
+
748.3 (100) [M] ; elemental analysis calcd (%) for C112
H
176
N
8
Cd: C
enyl)furan (19.2 g, 63 mmol) and fumaronitrile (7 g, 90 mmol) were dis-
solved in the minimum amount ofdry THF. The lf ask was sealed under
7
6.99, H 10.15, N 6.41; found: C 77.21, H 10.3, N 6.15.
Tris[1,4,8,11,15,18,22,25-octakis
Method 1: In a typical experiment, 3a (350 mg, 0.27 mmol) was recrystal-
lised thrice from CH Cl /MeOH to afford a dark blue crystalline material
150 mg, 44%). M.p. 2208C (decomp); IR (neat, ATR): n˜ =2950.75,
A
H
R
U
G
N
2
and placed in the fridge for a minimum of 14 days. The cold solution
. Lithium trimethyl-
was added to dry THF (20 mL) at À788C under N
2
2
2
silyl bisamide (1m in THF, 90 mL, 90 mmol) was added. The solution
(
2
turns dark brown and was allowed to warm to RT and stirred for 2 h. The
À1
918.0, 2849.64, 1604.2, 1365.16, 1287.22, 1266.26, 1189.81, 1093.03 cm
;
solution was poured into ice/aq. NH
was extracted with ethyl acetate (3”50 mL), dried (MgSO
4
Cl (100 mL) and the organic phase
) and filtered;
5
UV/Vis
1
(n-hexane):
l
max(e)=718
(1.3”10 ),
599
(0.6”
4
5
À1
3
À1
0 mol dm cm ), 550, 484 nm; MALDI-MS: isotopic clusters at m/z
the solvents were then removed under reduced pressure. The brown resi-
due was purified by column chromatography on silica (eluent: petroleum
ether (bp. 40–608C)/ethyl acetate 10:1). A first fraction was collected
which contained 2,5-bis(oct-7-enyl)furan (1.8 g). The second fraction con-
+
+
(
%): 2598.7 (10) [MÀPc] , 2484.8 (55) [MÀCdPc] , 1299.1 (100)
+
+
[
MÀCdPc
2
] , 1187.1 (55) [MÀCd
2 2
Pc ] ; elemental analysis calcd (%) for
H N24Cd : C 76.17, H 8.95, N 8.88; found: C 76.06, H 8.95, N 8.79.
240 336 2
C
Chem. Eur. J. 2007, 13, 7608 – 7618
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7615

M. J. Cook et al.
À1
+
tained the desired compound 3,6-bis(oct-7-enyl)phthalonitrile as a colour-
2227.8 cm
(CN); EI-MS: m/z 352.3 [M] ; CI-MS: m/z 370.3
1
+
] .
less oil (3.58 g, 16%). M.p. 34.48C; H NMR (60 MHz, CDCl
3
): d=7.4 (s,
A
H
R
U
G
3
2
H), 5.4–6.0 (m, 2H), 4.7–5.2 (m, 4H), 2.8 (t, J=8 Hz, 4H), 1.2–2.2 ppm
2,3,9,10,16,17,23,24-octakis(2-ethyl-1-hexyl)phthalocyanine
5a): The product was prepared as for 2d using 4,5-bis(2-ethylhexyl)-
1
3
(
m, 20H); C NMR (75.4 MHz, CDCl
3
): d=146.25, 138.97, 133.46,
(
1
15.76, 115.14, 114.39, 34.27, 33.54, 30.51, 28.86, 28.6, 28.58 ppm; IR
A
H
R
U
G
À1
(
neat, ATR): n˜ =2228.06 (CN), 1638.15 cm (C=C); EI-MS: m/z (%):
1
3
(
5%). M.p. 280.38C; H NMR (400 MHz, C
6
D
6
): d=9.69 (s, 8H), 3.17
+
3
32 2
48 (20) [M] ; elemental analysis calcd (%) for C24H N : C 82.71, H
m, 16H), 2.06 (m, 8H), 1.33–1.63 (m, 64H), 0.93–1.02 (m, 48H),
9
.25, N 8.04; found: C 82.42, H 9.38, N 7.84].
13
À0.24 ppm (s, 2H); C NMR (75.4 MHz, C
6 6
D ): d=143.18, 135.7, 41.17,
Metal-free 1,4,8,11,15,18,22,25-octakis(oct-7-enyl)phthalocyanine (2d): In
a typical experiment 3,6-bis(oct-7-enyl)phthalonitrile (1.38 g, 4 mmol) in
pentanol (10 mL) was brought to reflux. Lithium metal (excess) was
added in portions to the refluxing solution and reflux was continued for
38.6, 32.79, 28.98, 26.03, 23.16, 14.01, 10.94 ppm; IR (neat, ATR): n˜ =
2387.6 (NH), 2955.03, 2922.04, 2855.84, 1456.35, 1104.11, 1010.29,
À1
894.98 cm ; UV/Vis (THF): lmax =706, 669 nm; MALDI MS: isotopic
+
cluster at m/z (%) 1412.1 (100) [M] ; elemental analysis calcd (%) for
6
h. The solution was then left to cool and acetic acid (3 mL) was added.
C
96
H
146
N
8
: C 81.64, H 10.42, N 7.93; found: C 81.48, H 10.43, N 7.63].
Synthesis of metal-free 2,3,9,10,16,17,23,24-octakis(hexyl)phthalocyanine
5b)
,5-Bis
1.25 g, 5 mmol), [NiCl
0 mmol) were stirred in dry THF (50 mL) under N
Stirring was continued for 30 min. Excess methanol was added and the
mixture placed in the fridge overnight. The precipitate was collected by
filtration and washed with methanol. The product was recrystallised from
THF/methanol to afford dark green fine needles (520 mg, 38%). M.p. K–
D 80.78C, D–D 130.58C, D–I 151.78C; H NMR (270 MHz, C
AHCTREUNG
(
4
(
7
ACHTREUNG
2
A
T
E
N
3 2
) ]
1
6
D
6
): d=
2
7
2
.78 (s, 8H), 5.6–5.8 (m, 8H), 4.9–5.1 (m, 16H), 4.59 (t, J=7.6 Hz, 16H),
.22 (m, 16H), 1.93 (m, 16H), 1.69 (m, 16H), 1.3–1.5 (m, 32H),
a blue solution. A solution of2.5 m nBuLi in hexanes (2 mL, 5 mmol) was
added through a syringe and the solution turned deep red. 4,5-Dichlor-
ophthalonitrile
changed colour to light brown. This was left to stir for a few minutes, and
then the solution was cooled down to À788C. A solution ofhexylzincbro-
mide in THF (0.5m, 100 mL, 50 mmol) was added dropwise. The mixture
was left to warm to room temperature and was stirred overnight. The so-
lution was poured into 5% aqueous HCl (100 mL) and the organic phase
was extracted with ethyl acetate (2”50 mL). It was further washed with
1
3
À0.39 ppm (s, 2H); C NMR (75.4 MHz,
C
6
D
6
): d=139.07, 139.02,
[35]
(5 g, 25 mmol) was added at once and the solution
1
31.11, 114.43, 33.95, 33.07, 30.94, 29.63, 29.57, 29.13 ppm. IR (neat,
À1
ATR): n˜ =3292.42 (NH), 1640.35 cm
thane): lmax (e)=727 (0.63”10 ), 700 nm (0.52”10 mol dm cm ); FAB
MS: m/z (%): 1396.2 (30) [M] ; elemental analysis calcd (%) for
C
(C=C); UV/Vis (dichlorome-
5
5
À1
3
À1
+
96
H
130
N
8
: C 82.59, H 9.39, N 8.03; found: C 82.6, H 9.36, N 7.84.
1
,4,8,11,15,18,22,25-Octakis(oct-7-enyl)phthalocyaninatocadmium (3d):
In a typical experiment, 2d (100 mg, 72 mmol) in pentanol (5 mL) was
brought to reflux. Cadmium acetate (4 equiv, excess) was added and
reflux was continued for 45 min. The hot solution was poured into metha-
nol (50 mL) and the flask placed in the fridge. The green precipitate was
filtered then dissolved in THF (20 mL) through the filter paper, leaving
the excess cadmium acetate. The green solution was concentrated and
methanol (20 mL) was added. The flask was placed in the fridge. The
green precipitate was filtered and washed with methanol to afford the
5
% aq. HCl (30 mL), 5% aq. NaOH (30 mL) and brine (30 mL), dried
(
MgSO ) and filtered; the solvents were then removed under reduced
4
pressure. TLC analysis (eluent: petroleum ether (b.p. 40–608C)/dichloro-
methane 1:1) indicated three main products that were separated by
column chromatography on silica. A first separation was performed
(
eluent: petroleum ether (b.p. 40–608C)/dichloromethane 1:1) and two
fractions were obtained: the first containing triphenylphosphine, the
second a mixture ofproducts. A second separation was per of rmed on
this latter fraction (eluent: petroleum ether (b.p. 40–608C)/ethylacetate
product (85 mg, 78%). M.p. K–D 70.78C, D–D 133.08C, D–I 209.48C;
1
H NMR (400 MHz, [D
8
]THF): d=7.8 (s, 8H), 5.66–5.75 (m, 8H), 4.78–
1
5
:1). Three products were obtained and identified by H NMR spectros-
4
1
.90 (m, 16H), 4.57 (t, J=9.2 Hz, 16H), 2.19 (m, 16H), 1.96–1.99 (m,
1
copy as 1) 4-hexylphthalonitrile: H NMR (400 MHz, CDCl
J=10 Hz, 1H), 7.59 (s, 1H), 7.52 (d, J=10 Hz, 1H), 2.7 (t, J=7.6 Hz,
2
3
3
): d=7.7 (d,
1
3
6H), 1.64 (m, 16H), 1.37 ppm (m, 32H); C NMR (75.4 MHz, C
]pyridine): d=155.36, 139.18, 138.58, 136.71,
30.29, 114.33, 33.96, 32.99, 31.32, 29.69, 29.47, 29.13 ppm; IR (neat,
6 6
D
containing a drop of[D
1
5
H), 1.6 (quint, J=7.2 Hz, 2H), 1.29 (m, 6H), 0.81 ppm (t, J=7.6 Hz,
1
H) ppm; 2) 4-chloro-5-hexylphthalonitrile: H NMR (400 MHz, CDCl
3
):
ATR): n˜ =3073.88, 2921.09, 2850.24, 1639.96 (C=C), 1316.11, 1082.11,
d=7.78 (s, 1H), 7.68 (s, 1H), 2.79 (t, J=7.6 Hz, 2H), 1.62 (quint, J=
.6 Hz, 2H), 1.22–1.38 (m, 6H), 0.88 ppm (t, J=7.2 Hz, 3H); and 3) the
À1
9
89.85, 906.25, 808.7 cm ; UV/Vis (THF): lmax =707 nm; UV/Vis (di-
7
+
chloromethane): lmax =712 nm; FAB MS: m/z (%): 1507.9 (100) [M] ; el-
1
desired 4,5-bis
CDCl ): d=7.57 (s, 2H), 2.67 (t, J=7.6 Hz, 4H), 1.55 (quint, J=7.6 Hz,
H), 1.23–1.4 (m, 12H), 0.85 ppm (t, J=7.2 Hz, 6H).
Metal-free 2,3,9,10,16,17,23,24-octakis(hexyl)phthalocyanine (5b): The
product was prepared as for 2d using 4,5-bis(hexyl)phthalonitrile (1.12 g,
.75 mmol). This yielded metal-free 5b (500 mg, 45%). M.p. K–D
A
H
R
U
G
emental analysis calcd (%) for C96
found: C 76.15, H 8.34, N 7.25].
128 8
H N Cd: C 76.54, H 8.56, N 7.44;
3
[36]
4
Tris[1,4,8,11,15,18,22,25-octakis(oct-7-enyl)phthalocyanine]dicadmium
1d): In a typical experiment, compound 3d (50 mg, 33 mmol) was recrys-
AHCTREUNG
(
AHCTREUNG
tallised slowly from dichloromethane/methanol to afford the product as a
3
2
2
dark blue/black powder (23 mg, 56%). M.p. D–I 152.38C; IR (neat,
50.48C, D-I >3008C (decomp); IR (neat, ATR): n˜ =3290.21 (NH),
À1
ATR): n˜ =2918.08, 2847.58, 1603.81, 1365.58, 1274.01, 904.98 cm ; UV/
À1
918.72, 2850.86, 1463.39, 1321.15, 1102.91, 1009.48, 752.24, 705.63 cm
;
Vis (n-hexane): lmax =719, 600, 483 nm; MALDI-MS: isotopic clusters at
UV/Vis (THF): lmax =704.5, 666.5 nm; MALDI-MS: isotopic cluster at
m/z (%) 1186.9 (100) [M] ; elemental analysis calcd (%) for C80
C 80.89, H 9.67, N 9.43; found: C 80.46, H 9.48, N 9.43.
+
+
m/z (%): 3012.6 (3) [MÀPc] , 2899.7 (10) [MÀPcCd] , 1506.8 (100)
+
114 8
H N :
+
[
MÀPc
2
Cd] ; elemental analysis calcd (%) for C288
H
384
N
24Cd
2
: C 78.49,
H 8.78, N 7.63; found: C 78.49, H 8.87, N 7.56].
Synthesis of metal-free 2,3,9,10,16,17,23,24-octakis(2-ethyl-1-hexyl)-
phthalocyanine (5a)
,5-Bis(2-ethylhexyl)phthalonitrile: The product was obtained from 4,5-
dichlorophthalonitrile (2 g, 0.01 mol) and 2-ethylhexyl zinc bromide
50 mL of0.5 m solution in THF) following the method described for the
2
(
,3,9,10,16,17,23,24-Octakis(2-ethyl-1-hexyl)phthalocyaninatocadmium
6a): The product was prepared as for 3a from 5a (100 mg, 71 mmol).
This yielded 6a as a green powder (100 mg, 94%). M.p. 248.78C;
ACHTREUNG
1
4
H NMR(400 MHz, [D
8
]THF): d=9.14 (s, 8H), 3.18 (m, 16H), 2.04 (m,
8H), 1.28–1.75 (m, 64H), 1.05 (t, J=7.2 Hz, 24H), 0.95 ppm (t, J=
7.2 Hz, 24H); C NMR (75.4 MHz, C D + a drop of[D ]pyridine): d=
6 6 5
1
3
(
synthesis of 5b below. The compound was separated and purified by
155.93, 141.95, 138.03, 124.71, 41.41, 38.73, 32.99, 29.18, 26.15, 23.35,
column chromatography, which afforded two products, 4-(2-ethylhex-
yl)phthalonitrile (0.3 g) and the desired product (1.2 g, 34%). H NMR
14.21, 11.12 ppm; IR (neat, ATR): n˜ =2954.64, 2920.0, 2855.09, 1454.29,
1305.94, 1080.48, 751.84, 720.67 cm ; UV/Vis (THF): lmax =688, 621 nm;
1
À1
(
400 MHz, CDCl
J=6.2 Hz, 2H), 1.16–1.33 (m, 16H), 0.86–0.89 ppm (m, 12H); C NMR
100.55 MHz, CDCl ): d=147.08, 134.85, 115.81, 112.42, 40.37, 37.05,
2.26, 28.6, 25.44, 22.86, 13.97, 10.69 ppm. IR (neat, ATR): n˜ =
3
): d=7.52 (s, 2H), 2.6 (d, J=7.2 Hz, 4H), 1.55 (quint,
UV/Vis (CH
z (%): 1523.0 (100) [M] , 2932.1 (10) [Pc
emental analysis calcd (%) for C96 Cd·2CH
3
2 2
Cl ): lmax =697, 605 nm; MALDI-MS: isotopic clusters at m/
1
3
+
+
+
2
Cd] , 3044.9 (60) [Pc
2 2
Cd ] ; el-
(
3
3
H
144
N
8
OH: C 74.23, H 9.66, N
7.07; found: C 74.25, H 9.67, N 6.67].
7616
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7608 – 7618

Phthalocyanine Complexes
FULL PAPER
2
,3,9,10,16,17,23,24-Octakis
A
H
R
U
(hexyl)phthalocyaninatocadmium (6b): The
dark blue powder was obtained (15 mg). MALDI-MS: isotopic clusters at
m/z (%): 1186.8 (60), 1298.7 (100), 2484.5 (47), 2708.8 (5), 4007.6 (7),
4455 (1).
product was prepared as for 3a from 5b (30 mg, 25 mmol). The hot solu-
tion was poured into excess acetone and the product filtered. This yield-
ed 6b as
1
a
blue powder (22 mg, 67%). M.p. >3008C; H NMR
Experiment 5: This was achieved following the conditions described in
Experiment 1 with 7a (17.8 mg, 4 mmol) and 2a (2.7 mg, 2.2 mmol). A
dark blue powder was obtained (12 mg). MALDI-MS: isotopic clusters at
m/z (%): 1186.8 (73), 1523 (25), 2710.8 (5), 2933.1 (7), 3044.9 (22), 4455
(100), 5526.5 (2).
(
400 MHz, C
6
D
6
+ a drop of[D
5
]pyridine): d=9.67 (s, 8H), 3.08 (t, J=
7
0
.6 Hz, 16H), 1.87 (quint, J=7.6 Hz, 16H), 1.48 (m, 16H), 1.3 (m, 32H),
1
3
.91 ppm (t, J=7.2 Hz, 24H); C NMR (75.4 MHz, C
6 6
D + a drop of
[
D
5
]pyridine): d=155.92, 142.45, 139.29, 126.84, 34.09, 32.08, 31.96, 29.89,
2
1
6
1
2.82, 14.11 ppm; IR (neat, ATR): n˜ =2955.67, 2921.58, 2855.14, 1555.77,
Experiment 6: This was achieved following the conditions described in
Experiment 1 with 7a (16 mg, 3.6 mmol) and 3a (2 mg, 1.6 mmol). A dark
blue powder was obtained (14 mg). MALDI-MS: isotopic clusters at m/z
À1
403.3, 1095.36, 1021.38, 748.67, 665.96 cm ; UV/Vis (THF): lmax =686,
+
19 nm; MALDI-MS: isotopic clusters at m/z (%): 1298.8 (30) [M] ,
+
186.9 (10) [MÀCd] ; elemental analysis calcd (%) for
(
(
%): 1298.7 (19), 1411.1 (32), 1523 (37), 2707.9 (20), 2933.1 (30), 4455.1
100).
80 112 8
C H N Cd·2MeOH: C 72.29, H 8.87, N 8.22; found: C 72.33, H 8.46, N
7
.8.
Tris[2,3,9,10,16,17,23,24-octakis(2-ethyl-1-hexyl)phthalocyaninate]dicad-
mium (7a): The product was prepared as for 1a from 6a (50 mg,
3
6 mmol). This yielded 7a as a dark powder (53 mg, 73%). M.p. 2748C;
À1
IR (neat, ATR): n˜ =2920.0, 1454.23, 1304.0, 1103.65, 747.76, 718.36 cm
;
UV/Vis (hexane): lmax =693, 615, 495, 334 nm; MALDI-MS: isotopic
Acknowledgements
+
+
clusters at m/z (%): 4455.2 (100) [M] , 3045.1 (65) [MÀPc] , 2933.3 (10)
+
+
[
MÀPcCd] , 1523.1 (60) [MÀPc
2
Cd] ; elemental analysis calcd (%) for
We thank the EPSRC and the Iceni Seedcorn Fund LLP for financial
support and the former for access to the MS Service at Swansea (UK).
We also thank Dr. A. N. Cammidge and Dr. M. R. Cheesman for invalua-
ble discussions and advice during the course ofthe program.
288 432 2
C H N24Cd : C 77.64, H 9.77, N 7.54; found: C 77.59, H 9.67, N 7.56.
Recrystallisation of 6b: Complex 6b (22 mg, 17 mmol) was recrystallised
from dichloromethane and methanol to afford a dark blue powder
(
12 mg). M.p. >3008C; IR (neat, ATR): n˜ =2915.23, 2848.07, 1499.16,
À1
1
l
7
456.65, 1361.95, 1314.5, 1103.25, 749.88, 718.34 cm ; UV/Vis (hexane):
max =678, 642, 480, 329 nm; MALDI-MS: isotopic clusters at m/z (%):
676.7 (20), 6377.0 (70), 5081.3 (100), 3895.4 (25), 2596.9 (20), 1296.9
[
1] a) N. B. McKeown, Phthalocyanine Materials: Synthesis, Structure
and Function, (Eds.: B. Dunn, J. W. Goodby, A. R. West), Cam-
bridge University Press, Cambridge (UK), 1998; b) The Porphyrin
Handbook, Vols. 15–19, (Eds.: K. M. Kadish, K. M. Smith, R. Gui-
lard), Academic Press (USA), 2003; c) Phthalocyanines: Properties
and Applications, Vols. 1–4, (Eds.: C. C. Leznoff, A. B. P. Lever),
VCH, Weinheim (Germany), 1996.
(
15).
Synthesis of 2(3),9(10),16(17),23(24)-tetra-tert-butylphthalocyaninato
cadmium (8): 2(3),9(10),16(17),23(24)-Tetra-tert-butylphthalocyanine
(
120 mg, 0.16 mmol) in pentanol (10 mL) was brought to reflux. Cadmi-
um acetate (4 equiv, excess) was added and reflux continued for 45 min.
The hot solution was poured into methanol (50 mL) and the flask placed
in the fridge. The insoluble blue precipitate was filtered and washed with
methanol to afford the product (45 mg, 33%). UV/Vis (dichlorome-
thane): lmax =683, 660, 635 nm.
[
2] P. Gregory, High-Technology Applications of Organic Colorants,
Plenum, New York, 1991.
[
[
3] P. Gregory, J. Porphyrins Phthalocyanines 2000, 4, 432–437.
4] T. Shima, Y. Yamakawa, T. Nakano, J. Kim, J. Tominaga, Jpn. J.
Appl. Phys. Part 2 2006, 45, L1007–L1009.
Synthesis of 5,10,15,20-tetra(3,5-di-tert-butylphenyl)porphyrinato cadmi-
um (9): 5,10,15,20-Tetra(3,5-di-tert-butylphenyl)porphyrin (100 mg,
[
5] H. Mustroph, M. Stollenwerk, V. Bressau, Angew. Chem. 2006, 118,
2
068–2087; Angew. Chem. Int. Ed. 2006, 45, 2016–2035.
9
4 mmol) in pentanol (10 mL) was brought to reflux. Cadmium acetate
[
[
6] Y. Sato, K. Yamagishi, M. Yamashita, Opt. Rev. 2005, 12, 324–327.
7] R. Zhou, F. Josse, W. Gopel, Z. Z. Ozturk, O. Bekaroglu, Appl. Or-
ganomet. Chem. 1996, 10, 557–577.
(
4 equiv, excess) was added and reflux continued for 45 min. The hot so-
lution was poured into methanol (50 mL) and the flask placed in the
fridge. The purple precipitate was filtered and washed with methanol to
afford the product (73 mg, 66%). UV/Vis (dichloromethane): lmax =432,
[
8] a) S. R. Flom in The Porphyrin Handbook, Vol. 19 (Eds.: K. M.
Kadish, K. M. Smith, R. Guilard), Academic Press, Boston, 2003,
pp. 179–189; b) G. de la Torre, P. Vazquez, F. Agullo-Lopez, T.
Torres, Chem. Rev. 2004, 104, 3723–3750.
5
64, 604 nm; MALDI-MS: isotopic clusters at m/z (%): 1174.7 (100)
+
+
[
M] , 1063.8 (10) [MÀCd] .
Experiment 1: Compounds 3a (20.4 mg, 15.7 mmol) and 2d (10.9 mg,
.84 mmol) were dissolved in CH Cl (2 mL). Two drops ofmethanol
[9] D. Wçhrle, O. Suvorova, R. Gerdes, O. Bartels, L. Lapok, N. Bazia-
kina, S. Makarov, A. Slodek, J. Porphyrins Phthalocyanines 2004, 8,
1020–1041.
7
2
2
were added and the solution cooled to 48C. After 24 h excess methanol
was added and the product left to precipitate overnight. This was filtered
[10] a) E. A. Lukꢁyanets, J. Porphyrins Phthalocyanines 1999, 3, 424–
432; b) H. Ali, J. E. van Lier, Chem. Rev. 1999, 99, 2379–2450.
[11] R. Weiss, J. Fischer in The Porphyrin Handbook, Vol. 16 (Eds.:
K. M. Kadish, K. M. Smith, R. Guilard), Academic Press, Elsevier
Science (USA), 2003, pp. 171–246.
1
to afford a dark blue powder (21 mg). H NMR (400 MHz, CDCl
3
) (se-
), 5.6–5.72 (m, HÀC(R)=
), 4.78–4.9 ppm (m, H-C(R)=CH );
MALDI-MS: isotopic clusters at m/z (%): 1298.8 (100), 1506.9 (40),
483.6 (60), 2691.7 (25), 2898.9 (5).
lected data): d=6.04–6.15 (m, HÀC(R)=CH
2
CH
2
), 5.14–5.3 (m, HÀC(R)=CH
2
2
[
[
[
12] J. Janczak, R. J. Kubiak, J. Chem. Soc. Dalton Trans. 1993, 3809–
812.
13] J. Janczak, R. J. Kubiak, J. Richter, H. Fuess, Polyhedron 1999, 18,
775–2780.
14] J. A. de Saja, M. L. Rodriguez-Mendez, Adv. Colloid Interface Sci.
005, 116, 1–11.
2
3
Experiment 2: This was achieved following the conditions described in
Experiment 1 with 3a (26.9 mg, 20.8 mmol) and 5a (14 mg, 10 mmol). A
dark blue powder was obtained (16 mg). MALDI-MS: isotopic clusters at
m/z (%): 1186.9 (100), 1298.8 (35), 1522.1 (7), 2484.8 (6), 2709 (30),
2
2
4
005.9 (10), 4232.3 (3), 4456.5 (5), 5528.2 (4).
[15] H. Homborg, W. Kalz, Z. Naturforsch. B 1978, 33, 968–975.
[16] H. Homborg, C. L. Teske, Z. Anorg. Allg. Chem. 1985, 527, 45–61.
[17] a) H. Sugimoto, T. Higashi, M. Mori, J. Chem. Soc. Chem. Commun.
Experiment 3: This was achieved following the conditions described in
Experiment 1 with 1a (20.6 mg, 5.5 mmol) and 5a (4 mg, 2.8 mmol). A
dark blue powder was obtained (18 mg). MALDI-MS: isotopic clusters at
m/z (%): 1186.8 (100), 1298.7 (10), 1412 (18), 1522.9 (6), 2708.7 (20),
1
983, 622–623; b) H. Sugimoto, M. Mori, H. Masuda, T. Taga, J.
Chem. Soc. Chem. Commun. 1986, 962–963; c) V. B. Zhukhovitskii,
N. N. Ezhova, V. I. Krinichnyi, Y. A. Letuchii, A. M. Tonshin, K. M.
Dyumaev, M. L. Khidekel, Bull. Acad. Sci. USSR Div. Chem. Sci.
1985, 34, 2493–2497.
2
931.9 (3), 4006.4 (5), 4229.7 (3), 4453.9 (5), 5528.6 (3), 5750.8 (2).
Experiment 4: This was achieved following the conditions described in
Experiment 1 with 1a (15.4 mg, 4.1 mmol) and 6a (3.4 mg, 2.2 mmol). A
Chem. Eur. J. 2007, 13, 7608 – 7618
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7617

M. J. Cook et al.
[
18] P. Turek, P. Petit, J. J. Andre, J. Simon, R. Even, B. Boudjema, G.
Guillaud, M. Maitrot, Mol. Cryst. Liq. Cryst. 1988, 161, 323–327.
19] L. G. Tomilova, E. V. Chernykh, N. T. Ioffe, E. A. Luk’yanets, Zh.
Obshch. Khim. 1983, 53, 2594–2601.
[29] N. Kobayashi, H. Ogata, N. Nonaka, E. A. Luk’yanets, Chem. Eur. J.
2003, 9, 5123–5134.
[30] M. J. Cook, I. Chambrier, B. Isare, WO 2005/105820, 2005.
[31] N. B. McKeown, I. Chambrier, M. J. Cook, J. Chem. Soc. Perkin
Trans. 1 1990, 1169–1177.
[32] M. J. Cook, A. J. Dunn, S. D. Howe, A. J. Thomson, K. J. Harrison,
J. Chem. Soc. Perkin Trans. 1 1988, 2453–2458.
[33] A. N. Cammidge, M. J. Cook, K. J. Harrison, N. B. McKeown, J.
Chem. Soc. Perkin Trans. 1 1991, 3053–3058.
[
[
[
[
[
20] H. Sugimoto, T. Higashi, M. Mori, Chem. Lett. 1983, 1167–1170.
21] C. Clarisse, M. T. Riou, Inorg. Chim. Acta 1987, 130, 139–144.
22] N. Kobayashi, Coord. Chem. Rev. 2002, 227, 129–152.
23] J. Janczak, R. Kubiak, A. Jezierski, Inorg. Chem. 1995, 34, 3505–
3
508.
[
[
[
[
[
24] M. T. Riou, M. Auregan, C. Clarisse, J. Electroanal. Chem. Interfa-
cial Electrochem. 1985, 187, 349–354.
25] I. Chambrier, D. L. Hughes, J. C. Swarts, B. Isare, M. J. Cook, Chem.
Commun. 2006, 3504–3506.
[34] T. K. Miyamoto, N. Sugita, Y. Matsumoto, Y. Sasaki, M. Konno,
Chem. Lett. 1983, 1695–1698.
[35] D. Wçhrle, M. Eskes, K. Shigehara, A. Yamada, Synthesis 1993,
194–196.
[36] M. Hanack, P. Haisch, H. Lehmann, L. R. Subramanian, Synthesis
1993, 387–390.
26] W. Chidawanyika, A. Ogunsipe, T. Nyokong, New J. Chem. 2007, 31,
3
77–384.
27] R. D. George, A. W. Snow, P. F. McMillan, V. A. Burrows, J. Am.
Chem. Soc. 1992, 114, 8286–8287.
28] I. Chambrier, M. J. Cook, M. Helliwell, A. K. Powell, J. Chem. Soc.
Chem. Commun. 1992, 444–446.
Received: May 16, 2007
Published online: August 2, 2007
7618
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7608 – 7618