Synthesis and electrochemical characterisation of some long chain 1,4,8,11,15,18,22,25-octa-alkylated metal-free and zinc phthalocyanines possessing discotic liquid crystalline properties

Article abstract of DOI:10.1039/b006123i

A series of long chain (C₁₀, C₁₂, C₁₅ and C₁₈) 3,6-dialkylated phthalonitriles, prepared in 4 steps from thiophene, have been converted into the title phthalocyanines as metal-free and zinc derivatives. The symmetric macrocycles were characterised using ¹H NMR, MALDI-TOF ms and solution phase electronic (UV-VIS) spectroscopy. Four ring-based redox processes for the metal-free macrocycles possessing shorter alkyl substituents could be identified in the potential range - 1.8 to 1.2 V vs. Ag/Ag⁺ in dichloromethane or 1.2-dichloroethane. Reversibility became poorer with an increase in chain length of the eight alkyl substituents. The redox processes of the zinc complexes were not as well defined as those of the metal-free phthalocyanines. All compounds exhibited discotic thermotropic liquid crystal behaviour, which was studied using polarised optical microscopy and differential scanning calorimetry (DSC). Up to three different mesophases were detected. The zinc derivatives showed liquid crystalline behaviour at much higher temperatures than the metal-free compounds. The metal-free and zinc phthalocyanines with C₁₂ chains had the lowest crystalline to discotic liquid crystalline mesophase transition temperatures. Variable temperature UV-VIS spectra of thin films (ca. 1000 A thick) of the title compounds cast on glass were obtained for the compounds in the crystalline, meso- and isotropic liquid phase. The spectrum of a thin film of the isotropic liquid resembled the spectrum obtained from a solution of the same compound.

Full text of DOI:10.1039/b006123i

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and electrochemical characterisation of some long chain
1,4,8,11,15,18,22,25-octa-alkylated metal-free and zinc
phthalocyanines possessing discotic liquid crystalline properties
Jannie C. Swarts,*a Ernie H. G. Langner,a Nina Krokeide-Hoveb and Michael J. Cook*b
aDepartment of Chemistry, University of the Orange Free State, Bloemfontein 9300, Republic
of South Africa. E-mail: swartsjc@cem.nw.uovs.ac.za
bSchool of Chemical Sciences, University of East Anglia, Norwich NR4 7TJ, UK
Received 28th July 2000, Accepted 11th October 2000
First published as an Advance Article on the web 14th December 2000
A series of long chain (C10, C12, C15 and C18) 3,6-dialkylated phthalonitriles, prepared in 4 steps from
thiophene, have been converted into the title phthalocyanines as metal-free and zinc derivatives. The symmetric
1
macrocycles were characterised using H NMR, MALDI-TOF ms and solution phase electronic (UV-VIS)
spectroscopy. Four ring-based redox processes for the metal-free macrocycles possessing shorter alkyl
substituents could be identi®ed in the potential range 21.8 to 1.2 V vs. Ag/Agz in dichloromethane or 1,2-
dichloroethane. Reversibility became poorer with an increase in chain length of the eight alkyl substituents. The
redox processes of the zinc complexes were not as well de®ned as those of the metal-free phthalocyanines. All
compounds exhibited discotic thermotropic liquid crystal behaviour, which was studied using polarised optical
microscopy and differential scanning calorimetry (DSC). Up to three different mesophases were detected. The
zinc derivatives showed liquid crystalline behaviour at much higher temperatures than the metal-free
compounds. The metal-free and zinc phthalocyanines with C12 chains had the lowest crystalline to discotic
liquid crystalline mesophase transition temperatures. Variable temperature UV-VIS spectra of thin ®lms (ca.
Ê
1000 A thick) of the title compounds cast on glass were obtained for the compounds in the crystalline, meso-
and isotropic liquid phase. The spectrum of a thin ®lm of the isotropic liquid resembled the spectrum obtained
from a solution of the same compound.
liquid crystalline between 89.6 and 225.1 ³C. In contrast the
temperature range in which the corresponding nickel derivative
Introduction
is liquid crystalline is 64.2±135.9 ³C.11
Phthalocyanines have been investigated inter alia for applica-
tions such as dyestuffs,1 blue and green pigments,2 chemical
sensors,1,3 photodynamic cancer drugs,4 non-linear optical
materials,5 converters of visible light energy (from solar
radiation) into various other forms of energy including
chemical6 and electrical energy,7 catalysts in a variety of
chemical reactions including hydrogenation of multiple bonds8
and oxidation of thiols,9 and as liquid crystals.10
Investigations into the solution phase behaviour of the zinc
series showed that solubility and aggregation in cyclohexane is
critically affected by chain length. ZnPc±C10 (Pc~phthalocya-
nine) shows departure16 from the Beer±Lambert law at
1.561024 mol dm23 in contrast to the ZnPc±C5 through
ZnPc±C9 derivatives which showed departures between ca.
161025 and 461025 mol dm23. Non-aggregating phthalo-
cyanines are of interest in luminescence studies because
quenching is minimised and have been identi®ed as having
potential importance in photodynamic therapy.
The present paper describes the synthesis of novel longer
chain homologues of the zinc series and of the metal-free series
from which they were obtained. These have enabled us to
investigate the effect of the longer chains on mesophase
behaviour, thin ®lm properties and solution phase properties.
The potential for encapsulating the phthalocyanine core within
the compounds' own side chains becomes a real possibility and
this has been explored in the solution phase by an electro-
chemical investigation, the ®rst of its kind to be applied to any
of the octa-alkyl phthalocyanine derivatives investigated
hitherto.
One of our groups has reported wide ranging investigations
of non-peripherally substituted 1,4,8,11,15,18,22,25-octakis-
alkylphthalocyanines. These have been obtained as metal-free,
copper, cobalt, nickel11 and zinc12 derivatives and homologous
series have been synthesized in which the length of the alkyl
chains has been varied from methyl through to decyl. The
compounds have proved of interest for a number of reasons.
Examples form unusually well ordered spin coated ®lms,11,13
those with longer chains form liquid crystal phases,11
a
compound within its mesophase has shown rapid response
and recovery times in gas sensing studies14 and zinc derivatives
show bene®cial properties in the photodynamic therapy of
cancer.15 Structure±behaviour relationships have been estab-
lished particularly in the area of columnar mesophase
formation and the role of chain length and metal ion on
phase transition temperatures and mesophase range has been
established.11 In each series liquid crystallinity is exhibited by
compounds with C6±C10 chains. The series of zinc derivatives is
unique in that the C5 chain proved to be the limiting chain
length which supports mesophase behaviour. Zinc also has a
bene®cial effect in stabilising the mesophases, enhancing the
temperature range over which liquid crystalline properties
could be detected. The Zn±C10H21 derivative, for example, is
Results and discussion
Synthesis and characterisation
Phthalocyanines were prepared according to Scheme 1, and
results are summarised in Tables 1 and 2. We report here
optimised conditions for techniques described in earlier work12
as well as a new synthetic approach to the oxidation of long
434
J. Mater. Chem., 2001, 11, 434±443
This journal is # The Royal Society of Chemistry 2001
DOI: 10.1039/b006123i

View Article Online
Scheme 1 Syntheses of phthalocyanines 6 and 7. a: R~(CH2)9CH3, b: R~(CH2)11CH3, c: R~(CH2)14CH3, d: R~(CH2)17CH3. In route 3, oxone
was reacted in situ with acetone to yield the effective oxidant, dimethyldioxirane. Conversion of 4 to 5 is accompanied by SO2 elimination and
dehydrogenation. MCPBA~m-chloroperoxybenzoic acid.
chain dialkylated thiophenes to the corresponding dioxides.
Lithiation of thiophene, 1, to give the dilithiated product, 2,
followed by alkylation of the intermediate to give 3 were high
yielding (w80%) reactions, even for long linear alkyl chains
containing up to 18 carbon atoms. Shorter reaction times than
y24 h lowered the yields for long chain dialkylations.
Oxidation of the 2,5-dialkylthiophene derivatives 3 to the
corresponding 2,5-dialkylthiophene 1,1-dioxide 4 became more
dif®cult on increasing the length of the alkyl chains. Potassium
permanganate failed to oxidise the derivatised thiophenes to
the corresponding oxides. Oxidation with sodium perborate17
proved to be only effective for linear alkyl substituents
containing up to 10 carbon atoms.12 For R~(CH2)11CH3,
yields with this oxidant fell from y46% (for the C10H21
derivative) to 17%. No thiophene dioxide was isolated for the
oxidation of 3 with R~(CH2)17CH3. The poor solubility of the
dialkylated thiophenes, 3, with the longer alkyl chains in the
solvent of choice for sodium perborate-induced oxidations,
glacial acetic acid, probably contributes much to the observed
decrease in yields. When oxidation of series 3 was performed
with m-chloroperoxybenzoic acid, MCPBA, in dichloro-
methane18 the yield of 2,5-bis(dodecyl)thiophene 1,1-dioxide
was increased to 41% after 18 hours, but yields for the C18H37
derivative remained unsatisfactory. However, when oxidation
of series 3 was performed with dimethyldioxirane,19 near
quantitative oxidation was observed for all dialkylated
thiophenes provided stirring of the heterogeneous reaction
mixture was ef®cient. We found that dimethyldioxirane need
not be isolated by distillation prior to thiophene oxidation as
described by Miyahara and co-workers.20 In our modi®cation
of Miyahara's method,20 dimethyldioxirane was generated in
situ in the reaction mixture already containing the thiophene
that is to be oxidised by reacting acetone with Oxone1
(2KHSO5?KHSO4?K2SO4) in the presence of potassium
hydrogen carbonate. The concentration of the generated
dimethyldioxirane varies, but good results in our one-pot
procedure were obtained by assuming a concentration of
0.08 mol dm23. This concentration is the average of concen-
trations found when the generated dimethyldioxirane is
distilled from sul®de-free generation vessels as were reported
by various other researchers.20
The Diels±Alder condensation between fumaronitrile and
the thiophene 1,1-dioxides, 4, followed by subsequent in situ
SO2 extrusion and dehydrogenation of the intermediate to
give the phthalonitrile derivatives, 5, proceeded simulta-
neously in a sealed glass tube over 16 hours at 160 ³C in the
presence of a minimal quantity of chloroform (1.0±1.5 cm3
for 6.5 mmol fumaronitrile, the volume needed to wash the
reactants through the nozzle of the tube). Larger quantities
of chloroform reduced the yield of the desired phthalonitrile
dramatically. Cyclisation of the phthalonitriles in re¯uxing
pentanol in the presence of lithium metal to give the
Table 1 Preparation and characterisation of 2,5-dialkylthiophenes (3), 2,5-dialkylthiophene 1,1-dioxides (4), via dimethyldioxirane oxidation, and
3,6-dialkylphthalonitriles (5). See Scheme 1
% Found(requires)
Compound Yield(%) Mp/³C Formula
d(1H), 270 MHz, CDCl3
C
H
S
3a
3b
3c
3d
4a
4b
4c
4d
5a
5b
5c
5d
70
42
80
94
48b
98
95
85
25
27
26
27
166a
25(217a) C28H52
C24H44
S
S
S
S
79.2(79.0) 12.3(12.2) 8.7(8.8) 0.90 (t, 6 H), 1.3 (m, 28 H), 1.64 (m, 4 H), 2.71 (t, 4 H), 6.48 (s, 2 H)
79.9(79.9) 12.4(12.5) 7.6(7.6) 0.90 (t, 6 H), 1.3 (m, 38 H), 1.65 (m, 4 H), 2.73 (t, 4 H), 6.53 (s, 2 H)
80.4(80.8) 12.7(12.8) 6.1(6.4) 0.90 (t, 6 H), 1.3 (m, 48 H), 1.68 (m, 4 H), 2.78 (t, 4 H), 6.58 (s, 2 H)
81.6(81.6) 13.0(13.0) 5.5(5.4) 0.88 (t, 6 H), 1.3 (m, 60 H), 1.63 (m, 4 H), 2.76 (t, 4 H), 6.54 (s, 4 H)
43
56
C34H64
C40H76
61±62 C24H44SO2 72.6(72.7) 10.9(11.2) 7.8(8.1) 0.90 (t, 6 H), 1.3 (m, 28 H), 1.64 (m, 4 H), 2.50 (t, 4 H), 6.22 (s, 2 H)
73
79
88
70
75
79
82
C
28H52SO2 73.8(74.2) 11.5(11.6) 6.8(7.1) 0.90 (t, 6 H), 1.3 (m, 36 H), 1.64 (m, 4 H), 2.49 (t, 4 H), 6.25 (s, 2 H)
C34H64SO2 75.9(76.0) 12.0(12.0) 6.0(6.0) 0.90 (t, 6 H), 1.3 (m, 48 H), 1.62 (m, 4 H), 2.48 (t, 4 H), 6.25 (s, 2 H)
40H76SO2 77.4(77.4) 12.4(12.3) 4.7(5.1) 0.90 (t, 6 H), 1.28 (m, 60 H), 1.62 (m, 4 H), 2.47 (t, 4 H), 6.23 (s, 2 H)
C28H44N2 82.1(82.3) 11.2(10.9) 6.8(6.9)c 0.90 (t, 6 H), 1.28 (m, 28 H), 1.64 (m, 4 H), 2.86 (t, 4 H), 7.50 (s, 2 H)
C
C
32H52N2 82.8(82.7) 11.1(11.2) 6.1(6.0)c 0.90 (t, 6 H), 1.3 (m, 38 H), 1.63 (m, 4 H), 2.85 (t, 4 H), 7.48 (s, 2 H)
C38H64N2 83.4(83.1) 11.7(11.8) 4.9(5.1)c 0.90 (t, 6 H), 1.3 (m, 48 H), 1.63 (m, 4 H), 2.86 (t, 4 H), 7.48 (s, 2 H)
C44H76N2 83.6(83.5) 12.0(12.1) 4.2(4.4)c 0.89 (t, 6 H), 1.3 (m, 60 H), 1.63 (m, 4 H), 2.83 (t, 4 H), 7.47 (s, 2 H)
c
aBp at 0.1 torr (~13.33 Pa). bOxidised with MCPBA. Nitrogen content.
J. Mater. Chem., 2001, 11, 434±443
435

View Article Online
436
J. Mater. Chem., 2001, 11, 434±443

View Article Online
followed the same trends as described before23 for the
lower homologues; data are summarised in Tables 1 and 2.
The solubility properties of the C18 compound prepared in
this study stand in contrast to what was found for the C4±
C10 range of octa-alkylated phthalocyanines. It was
previously documented that the solubility of octa-alkylated
phthalocyanines in the C4±C10 alkyl chain length range
generally increased.12 This result is in accordance with the
rule of thumb that the solubility of phthalocyanines is
enhanced by bulky substituents.24 Our results with the C18
derivative, therefore, indicate that there is a limit to which
the solubility of phthalocyanines can be enhanced by simply
increasing the size of the substituent.
Fig. 1 UV/VIS solution (THF) spectra at 25 ³C of 6d (---) and 7d (Ð)
showing the Q and Soret bands at ca. 700±730 and 350±360 nm
UV/VIS spectroscopy
Ê
respectively. Left insert: The Q-band of a ca. 1000 A thick ®lm of 7b on
All metal-free phthalocyanines are turquoise green while the
zinc derivatives are bright blue in the crystalline state. In
solution, however, both the metal-free and zinc compounds
appear to have a turquoise green colour. The characteristic UV
(Soret band) and intense visible region (Q-band) absorptions of
the phthalocyanines are documented in Table 2. Metal-free and
metal-containing phthalocyanines differ in having D2 h and
D4 h symmetry respectively and this is manifested in differences
especially in the Q-band region. The metallated phthalocya-
nines all showed a single intense maximum at about 700 nm.
However, degeneracy of the lowest energy singlet state of the
metallated derivatives is lifted in the metal-free derivatives.
This leads to a splitting of the Q-band into a Qx and Qy
component at about 729 and 698 nm (Fig. 1). The low intensity
bands at shorter wavelengths than the Q-band, ca. 665 and
630 nm, are vibronic in origin. It is known that aggregation of
phthalocyanines in solution signi®cantly lowers the extinction
coef®cients of solutions of these compounds.12,16 The extinc-
tion coef®cients, e, reported in Table 2 were obtained from
measurements on solutions in the concentration range 1±
25 mmol dm23 for series 6 and 1±10 mmol dm23 for series 7. No
variation of e was detected which implied phthalocyanine
aggregation is absent in this concentration range (Fig. 1).
glass at 30 ((a), crystalline solid phase K), 80 ((b), mesophase D3), 130
((c), mesophase D1), 180 ((d), also mesophase D1) and 230 ³C ((e),
isotropic liquid I). Right insert: The Beer±Lambert law, A~ecl, [shown
at 700 nm for 7a ($) and 730 nm for 6a (&)] is valid for all compounds
6 and 7, indicating aggregation is not a factor in the investigated
concentration ranges.
corresponding lithium phthalocyanines was more ef®cient
over 18 hours than over shorter (y5 h) reaction times. The
lithiated phthalocyanines were converted into the metal-free
derivatives, 6, by acid work-up. The complexation of Zn2z
with series 6 to give 7 were very ef®cient reactions, with
yields w90% over 2 hours. It was notable that attempts to
obtain zinc phthalocyanines directly by reacting
5
(R~C10H21) with ZnBr2 in the presence of dimethylami-
noethanol (DMAE) according to the procedure of Torres
and co-workers,21 failed. Only the metal free phthalocyanine
was obtained following work-up with acetic acid. Puri®ca-
tion of the phthalocyanines 6 and 7 became more dif®cult
with increased chain lengths. They suffered co-precipitation
of impurities or incorporation of solvent molecules upon
crystallisation. In addition, the C18 substituted phthalocya-
nines 6 and 7 were only marginally soluble in THF but
insoluble in all other common organic solvents at room
temperature. They do, however, dissolve well in warm
chloroform or THF. Even after rigorous puri®cation
procedures, the elemental analysis of the ZnPc±C15 and
ZnPc±C18 derivatives were still marginal. Apart from
explaining this by solvent co-precipitation (see Table 2), it
should be noted that many phthalocyanines are notoriously
dif®cult to combust.22 However, these two compounds are
deemed to be pure by chromatography, spectroscopy and
matrix-assisted laser de-
Mesophase behaviour
The phase transformations from crystalline solid to discotic
mesophase (KADi) and from discotic mesophase to isotropic
liquid (DiAI), i~1, 2 or 3, were investigated using polarised
optical microscopy. In many cases the colour and/or texture
sorption/ionisation time-of-¯ight mass spectrophotometry
(MALDI-TOF ms). 1H NMR data for all synthesised
compounds con®rmed their purity and structure and
Fig. 3 Left: Photograph showing simultaneously the fan (D1, majority
of graph) and needle (D2, next to the yellow spot, needles are bundled
together in the form of ®bres) texture of 6c. The mixed texture was
obtained by cooling 6c at 10 ³C min21 from 95 to 90 ³C and then slowly
heating (2 ³C min21) to 92 ³C. Right: Photograph showing simulta-
neously the blue crystalline phase (K) and green D2 mesophase with a
mosaic pattern. The mixed texture was obtained by cooling 6c at a rate
of 5 ³C min21 and the photograph was taken at 66±67 ³C.
Fig. 2 The temperature range in which the zinc phthalocyanines, ZnPc,
right, are liquid crystalline is much larger than for the metal-free
derivatives, 2HPc, left. The temperatures at which the transitions took
place were determined on a polarised optical microscope. DH for each
phase change as well as limitations to the data for the ZnPc±C15 and
ZnPc±C18 derivatives are as per Table 2.
J. Mater. Chem., 2001, 11, 434±443
437

View Article Online
changes from one mesophase to another were extremely subtle
and thus dif®cult to detect with an optical microscope. The
results reported in Table 2 and Fig. 2, therefore, represent the
combined conclusions obtained from differential scanning
calorimetry (DSC) as well. However, a detailed discussion on
the DSC results, together with those from an electrochemical
investigation of a series of different metallated phthalocya-
nines, will be given in a subsequent paper.25 At least two
mesophases denoted as D1 and D2, where D1 is the highest
temperature mesophase, were detected for all compounds
except the C10 and C18 metal-free phthalocyanines. The
absence of an observable second mesophase for the C10 and
mesophase transitions observed in the present study may arise
from the length of the side chains of the title compounds. It is
quite conceivable that the long chain length of the substituents
on phthalocyanines 6 and 7 interfere with the ease by which any
semi-ordered liquid crystalline packing may be achieved. This
interference will result in a slower formation of any order in the
packing of the liquid crystalline phase and it may lead to
pronounced distortions in molecular packing.
As can be seen in Fig. 2, the zinc derivatives showed liquid
crystalline behaviour over a much larger temperature range than
the metal-free derivatives, indicating that zinc has the capability
of stabilising mesophases at higher temperatures compared to
hydrogen. The C12 macrocycle proved to be a turning point. It
showed the lowest KAD transition temperature (60.0³C) of all
phthalocyanines investigated thus far in this series and it showed
mesophase behaviour over a 145.6 ³C temperature range. Only
the C8 derivative had a larger mesophase behaviour temperature
range (153.2³C). The minimum KAD transition temperature of
60 ³C indicates that C12 alkyl chains approach the limiting length
where the waxy (amorphous) properties of long chain alkyl
substituents do not substantially override the liquid crystalline
properties of the molecule as a whole. The slow increase in KAD
transition temperatures at higher chain lengths is typical of the
increase in melting points for waxes with an increase in chain
length as illustrated by the melting points of tetrakis(decyl)am-
monium bromide (89 ³C) and tetrakis(octadecyl)ammonium
bromide (103 ³C).
The insert in Fig. 1 shows the Q-band absorbtion of the
ZnPc±C12 compound, 7b, cast as a thin ®lm on glass at a
temperature within each of the crystalline, discotic mesophases
and isotropic liquid phases. The spectrum at 30 ³C (i.e. of the
compound in crystalline form, the line labelled (a)) shows two
peaks at 630 and 765 nm respectively. The spectrum at 80 ³C
(line (b)) is of the compound in the low temperature
mesophase, D3. The peak at 630 nm was shifted to 642 nm,
while the 765 nm peak was not moved. The appearance of a
new peak at 714 nm is, however, faintly apparent. At the top
end of the second, high-temperature (D1), mesophase tem-
perature range, at 180 ³C, the spectrum (line (d)) shows a
distinct new peak at 714 nm while the peak at 765 nm has
disappeared. A weaker peak is also observed at 658 nm. A
spectrum of the compound at 230 ³C (i.e. in the isotropic liquid
state, line (e)) shows an intense Q-band maximum at 714 nm
together with a shoulder at 666 nm. The appearance of the
spectrum of the isotropic liquid is very similar to that of the
compound in solution, which showed Qmax at l~700 nm and a
shoulder at 665 nm. Thin ®lm variable temperature visible
spectra of the C15 and C18 zinc phthalocyanines had the same
features and followed the same trend. The spectrum of spin
coated ®lms of the metal free derivatives also showed two peaks
(at l~640 and 771 nm at 35 ³C for the C12 compound) in the
crystalline state which collapsed to Qx and Qy bands with
C
18 metal-free derivative does not preclude its existence, but it
was not observable by the measuring methods at our
disposable. The C15 and C18 zinc phthalocyanines are the
®rst examples of the series 7 class of phthalocyanines to show
three observable mesophases, D1, D2 and D3, between the
crystalline and isotropic liquid states. The only other
compound within the series of complexes thus far studied
(Zn, Cu, Ni, Co and 2H complexes having substituents with
C4±C10 chainlengths)11 that showed three observable meso-
phases (discotic and monotropic), was the 2HPc±C8 species. In
previous reports on the C4 to C10 derivatives,11 different
mesophases were tentatively assigned by comparing the optical
textures observed under the polarising microscope with those
whose packing symmetries were previously determined from X-
ray diffraction studies. These designations were Dhd, a discotic
disordered mesophase phase showing a fan like texture and
comprised of columns of cofacially stacked molecules arranged
within a two dimensional hexagonal lattice symmetry, a second
disordered hexagonal mesophase, Dhd, with a needle-like
texture and a third disordered mesophase possessing rectan-
gular symmetry, Drd, which is characterised by a mosaic
texture. In the cases reported hitherto, the fan texture of the
Dhd mesophase was observed at the ®rst transition upon
cooling from the isotropic liquid. The needle texture was
infrequently observed upon further cooling and the mosaic
texture was associated with the mesophase just prior to
conversion to a solid crystal. In these previous reports, texture
appearances could be generated reversibly and unambiguously.
In this study, we found that the texture observed under the
microscope was very much a function of the rate of cooling or
heating. In the case of the C15 metal-free derivative it was found
that a fast cooling rate (10 ³C min21) resulted in a needle-like
texture upon cooling from the isotropic liquid at 94 ³C
(IAD1transition). A slow cooling rate (2 ³C min21) resulted
in a texture which resembled, but was not clearly, the fan for
the same transition. However on heating the needle-like texture
from 90 to 92 ³C, it slowly converted in appearance from the
needle texture to the fan-like texture. No colour change was
observed for this texture change. The photograph in Fig. 3
(left) shows both the needle-like bundles stacked together to
create the appearance of ®bres and the `fan-like' texture
observed during this transition. Further cooling produces a
clear colour change from greenish to bluish when crystal-
lisation (DAK transformation) sets in. In many heating and
cooling cycles subtle changes from an apparent `mosaic' to a
fan-like texture was also observed as shown in Fig. 3 (right). It
appears that although the long-chain alkylated phthalocya-
nines of the present study do follow the same general trends as
described before,11 the amount of disorder within the cofacially
arranged two dimensional molecular columns are much more
pronounced than was experienced with the shorter chain
derivatives.
lmax~734 and 704 nm upon heating to a temperature above
the transition temperature for isotropic liquid formation. With
one exception, all peak maxima for the C15 and C18 compounds
(metal-free and zinc-containing) were within 5 nm of those of
the C12 derivative. Only the lower wavelength peak of the
metal-free C18 compound differed in that it was broadened with
an additional shoulder at 674 nm.
Cyclic voltammetry
Cyclic voltammetry results, obtained from ca. 1 mM solutions
of the compounds in either dichloromethane or 1,2-dichloro-
ethane, are summarised in Table 3. Although the four observed
redox signals in the potential range 21.8 to 1.2 V vs. Ag/Agz
were more ideal for the metal free phthalocyanines (2HPc), 6,
than for the zinc derivatives, ZnPc, neither exhibited ideal
reversible behaviour. Electrochemical reversibility for one-
electron processes at 25 ³C is characterised by peak potential
Accordingly, although by analogy to previous studies the
observed phases D1, D2 and D3 are expected to conform to the
Drd, fan; Dhd, needle, and Dhd, mosaic, two dimensional
symmetry or packing modes, in the absence of X-ray
diffraction studies, no de®nite assignment to this effect can
be made. A possible explanation for the less than distinct
438
J. Mater. Chem., 2001, 11, 434±443

View Article Online
J. Mater. Chem., 2001, 11, 434±443
439

View Article Online
Fig. 5 Cyclic voltammograms of (in order from bottom to top) 6d, 6a,
7a and 7d, showing the four ring centred redox processes I, II, III and
IV in 1,2-dichloroethane (1 mmol dm23) containing 0.1 mol dm23
tetrabutylammonium hexa¯uorophosphate at 70 ³C at a scan rate of
50 mV s21 on a Pt working electrode at 75 ³C. Conditions for 6a only:
in dichloromethane at 25 ³C. Limitations to the data for the ZnPc±C18
derivative are as per Table 2.
Fig. 4 Cyclic voltammograms showing the four ring centred redox
processes I, II, III and IV of 6b in 1,2-dichloroethane (1 mmol dm23
)
containing 0.1 mol dm-3 tetrabutylammonium hexa¯uorophosphate at
scan rates 50 (smallest ip values), 100, 150, 200 and 250 mV s21 on a Pt
working electrode at 25 ³C. The peak labelled Fc is that of ferrocene
that has been added as an internal marker to the solution containing 6b.
differences of 59 mV and is independent of scan rate.26 The
2HPc±C10 and 2HPc±C12 derivatives showed good, well
de®ned, CV's in dichloromethane or 1,2-dichloroethane at
25 ³C (Fig. 4). For all the other compounds, peak currents
tended to approach zero at room temperature and better results
were obtained at 70 ³C in 1,2-dichloroethane. However, even at
70 ³C in 1,2-dichloroethane, only one well identi®ed quasi-
reversible peak at 0.292 mV was observed (Fig. 5) for the
ZnPc±C18 derivative.
reported versus SCE but in this work, to allow direct
comparison with our results, were adjusted to be versus a
Ag/Agz (from 0.1 mol dm23 AgNO3 in acetonitrile) reference
electrode, which has a potential of z0.31 V versus SCE as
reported elsewhere.29 Our octa-substituted metal free com-
plexes showed one more oxidation process at 0.733 V but the
two more negative reduction processes could not be detected
under our experimental conditions.
Compared to the unsubstituted phthalocyanine, the effect of
octa-alkyl substitution makes the ®rst oxidation process (wave
III) ca. 0.03 V more positive, and both the ®rst (wave II) and
second (wave I) reduction processes about 0.20 V more
negative. Likewise, in DMF, formal reduction potentials for
the one observed oxidation and four observed reduction waves
for unsubstituted zinc phthalocyanines28,30 are 0.36, 21.17,
21.61, 22.16 and 22.56 V versus Ag/Agz respectively. At
150 ³C in non-coordinating 1-methylnaphthalene one oxida-
tion wave at 0.45 and two reduction processes at 21.09 and
21.45 V are reported.27 Under our experimental conditions
(70 ³C, 1,2-dichloroethane), the octa-alkylated ZnPc's showed
a second oxidation peak, but the two most negative reduction
processes reported for ZnPc itself could not be detected. As can
be seen from eqn. (1), the ®rst oxidation wave for our
complexes was observed at 0.219 V, some 0.14 V less positive
than for the unsubstituted compound in DMF, while the ®rst
reduction wave was observed at ca. 21.335 mV, ca. 0.17 V
more negative in character. Surprisingly, the second reduction
wave of our compounds coincided almost exactly with that of
unsubstituted ZnPc in DMF at 21.557 V versus Ag/Agz.
The two reduction waves, waves I and II (eqn. (1)) were very
poorly de®ned for all our Zn complexes and became worse as
the chain length of the alkyl substituents increased. For the
ZnPc±C18 derivative, the only readily identi®able peak was that
of wave III. When compared to the ferrocene/ferricenium (Fc/
Fcz) couple, wave III for the 2HPc±C10 and 2HPc±C12
compounds were essentially reversible in either dichloro-
methane or 1,2-dichloroethane. When free ferrocene is used
as an internal marker in the presence of 2HPc's (Fig. 4)
according to the method described by Gagne,31 a slightly better
DEp value of 83 mV for the ferrocene signals is observed, but
Eo/ for the Fc/Fcz couple shifted to 201 mV vs. Ag/Agz
compared to 220 mV (Table 3) in the absence of any
phthalocyanines. The drift in Eo/ for ferrocene in the presence
of phthalocyanines is probably a consequence of the closeness
of Eo/ of wave III, see eqn. (1) above and Fig. 4, to that of the
Fc/Fcz couple. For the Zn derivatives, these two potentials are
All four observed redox processes are associated with the
ring-based electron transfer couples [MPc]22/[MPc]2 (wave I,
?
z
Fig. 4 and 5), [MPc]2 /[MPc] (wave II), [MPc]/[MPc] (wave
?
?
2z
III) and [MPc]z /[MPc]
(wave IV) with M~2H or Zn as
?
demonstrated in eqn. (1). Regarding 2HPc's, except for wave
IV, the chain length of the substituted alkyl groups did not
result in changes in the formal reduction potentials, Eo/~
(EpazEpc)/2 where Epa and Epc are the peak anodic and peak
cathodic potentials respectively, of more than 30 mV (Table 3).
However, the 2HPc±C18 derivative showed anomalous beha-
viour in that wave I could not be observed and wave IV was
poorly de®ned under our experimental conditions. Also, Eo/ of
wave III was 50 mV more positive than that of the C10
compound. Eo/ for wave IV became 80 and 160 mV more
positive when moving from the 2HPc-C10 to the C12 and C15
compounds respectively. Eo/ values, at a scan rate of 50 mV s21
vs. Ag/Agz, shown in eqn. (1), refer to the C10 compound.
(1)
Poor solubility of unsubstituted phthalocyanines precludes
solution phase electrochemistry of these compounds in non-
coordinating solvents such as CH2Cl2 or 1,2-dichloroethane.
To overcome this, Campbell27 studied the unsubstituted metal-
free phthalocyanine at 150 ³C in 1-methylnaphthalene. Only
two reduction waves at 20.92 and 21.37 V versus Ag/Agz
were reported. However, ®ve formal reduction potentials at
0.33, 20.97, 21.37, 22.24 and 22.51 V versus a Ag/Agz
reference electrode are on record28 for the unsubstituted 2HPc
species in DMF. Literature cited potentials were originally
440
J. Mater. Chem., 2001, 11, 434±443

View Article Online
overlapping so much that peak separation was not possible.
For this reason we chose to give our experimentally determined
potentials versus Ag/Agz as reference electrode. For the 2HPc±
have established that ®lms formed under these conditions are
33
Ê
ca. 1000 A thick.
C
15 compound, DEp values for wave III were 120 mV or larger
Spectroscopy
depending on the scan rate; for the 2HPc±C18 derivative DE
values were 168 mV or larger. The increase in DEp values with
increase in alkyl chain length (and also scan rate) indicates that
the rate of electron transfer between the electrode and the
phthalocyanines are severely hampered by the alkyl side chains
and it becomes progressively worse as the side chain length
increases. This must be a function of the availability of a non-
hindered, ef®cient route that will allow fast electron ¯ow
between the electrode and the phthalocyanine core and can be
explained by the length of the side chains on the phthalocya-
nines. It can be envisaged that the longer the chain length of the
eight alkyl substituents, the more ef®ciently they will coil and
wrap around the phthalocyanine core, thereby partially
isolating the active redox centre from the electrode.
The peak cathodic and peak anodic current ratio, ipc/ipa, for
wave III (2HPc's), approached unity which indicated this
electrochemical process was not accompanied by any other
physical or chemical process. All the other 2H- and ZnPc's
waves investigated here deviated substantially from unity.
However, since no additional wave apart from the four
belonging to the phthalocyanine system could be detected,
we conclude that the deviation from unity in waves I, II and IV
is probably not the consequence of a chemical process that
accompanies these processes either. Rather, we are of the
Proton NMR-spectra at 298 K were recorded at 270 MHz on a
JEOL EX 270 instrument and chemical shifts are referenced to
SiMe4 at 0.00 ppm. IR spectra (thin ®lms between NaCl plates
of either pure liquids or Nujol mulls of solids) were recorded on
a Perkin Elmer 297 instrument. Conventional electronic
spectra were recorded in THF on a Varian Cary 50 UV/VIS
dual beam spectrophotometer at 25 ³C. Thin ®lm variable
temperature visible spectra were recorded on a Hitachi U-3000
spectrophotometer ®tted with a near-infrared sensitive detector
and polaroid polarizers. The ®lm-bearing substrate was located
in a Mettler FP82 hotstage, orientated orthogonally to the
incident beam and controlled by an FP80 Processor. MALDI-
TOF mass spectra were obtained using a Kratos Kompact
MALDI 3 with a dithranol matrix at the University of
Manchester.
Differential scanning calorimetry and microscopy
Transitions between crystal, liquid crystal and isotropic liquid
states were monitored by DSC (for enthalpy changes, ca. 7 mg
samples at heating and cooling rates of 10 ³C min21 between
270 ³C and a convenient maximum temperature at least 30 ³C
higher than the melting point of the compounds) using a TA
Instruments DSC 10 thermal analyser ®tted with a DuPont
Instruments mechanical cooling accessory and a TA Instru-
ments Thermal Analyst 2000 data processing unit. Visual
measurements (to obtain transition temperatures) were per-
formed on an Olympus BH-2 polarising microscope in
conjunction with a Linkam TMS 92 thermal analyser with a
Linkam THM 600 cell. The heating and cooling rates used were
opinion that the lack of suitable solubility of these in situ
generated [MPc]22, [MPc]2 and [MPc] species in the non-
2z
?
polar solvents dichloromethane or 1,2-dichloroethane disal-
lows the generation of a CV half wave of suf®cient intensity
during the reverse sweep to maintain ipc/ipa ratios approaching
unity. The possibility of these intermediates forming thin
coatings by adsorption on the electrode, as described else-
where,32 is to a large extent limited in view of the protective
layer, or globule, that the eight long alkyl substituents
apparently form around the phthalocyanine core. It appears
that the consequence of this wrapping of side chains around the
phthalocyanine core to effectively encapsulate it, in certain
circumstances proves to be bene®cial, while in other cases not.
either 5 ³C min21 or 2 ³C min21
.
Cyclic voltammetry
Measurements utilizing ca. 1 mmol dm23 phthalocyanine
solutions at 25 ³C in dichloromethane, for 2H±C10 and 2H±
C12 derivatives, or at 70 ³C in 1,2-dichloroethane for all
compounds except the 2H±C10 derivative, with 0.100 mol dm23
tetra-n-butylammonium hexa¯uorophosphate (Fluka, electro-
chemical grade) as supporting electrolyte, were conducted
under a blanket of puri®ed argon utilising a BAS model CV-27
voltammograph interfaced with a personal computer. A three-
electrode cell, which utilized a Pt-auxiliary electrode, a Pt
working electrode with surface area 0.0707 cm2, and a Ag/Agz
(0.100 mol dm23, AgNO3 in acetonitrile) reference electrode34
mounted on a Luggin capillary35 ®lled with a 0.100 mol dm23
supporting electrolyte solution in acetonitrile, was employed.
Data, uncorrected for junction potentials, were collected with
an Adalab-PCTM and AdaptTM data acquisition kit (Inter-
active Microwave, Inc.) with locally developed software, and
analysed with Hyperplot (JHM International, Inc.). All
temperatures were kept constant to within 0.2 ³C.
Experimental
Materials
Reagents (Lancaster or Aldrich) were used without further
puri®cation. Dry, air-free THF was obtained by re¯uxing
under nitrogen over sodium metal and distillation just prior to
use. Water was puri®ed through a standard Millipore system to
a resistivity of at least 18 MV cm. Fluka silica gel 60 grade
60741 was used in chromatographic separations. TLC was
performed using Fluka grade 60778 silica gel on aluminium
sheets.
Thin ®lm preparation
Cleaning of glass slides for thin ®lm deposition: Microscope
glass slides (BDH) were wiped with Kimwipe paper, washed in
a stream of methanol and then sonicated for 5 minutes in water
containing 2±4% Decon 24 (Fisher), cleaned with a stream of
water, sonicated in pure water, and dried in a stream of propan-
2-ol.
Synthesis
Only one representative example is provided for each of the
general methods.
2,5-Dialkylthiophenes
3. 2,5-Bis(octadecyl)thiophene
Films: Spin-coated ®lms of all phthalocyanines were
prepared at room temperature (50 ³C for the C18 derivatives)
from THF solutions (2.5 mg in 0.1 ml) dropped on a previously
cleaned microscope glass slide rotating at 2000 rpm using a
spinner (Headway Research, Inc.). Spinning was continued for
20 s by which time the solvent had evaporated. Previous
measurements of ®lm thickness (mechanical stylus equipment)
(Table 1, 3d). To a stirred solution of thiophene (8.4g,
0.1 mol) in THF (dried over sodium; 50 cm3), precooled to
275 ³C in a CO2(s)±acetone bath, was added under nitrogen
over 30 min n-butyllithium in hexane (100 cm3 of
a
2.5 mol dm23 solution, 0.25 mol, 2.5 equivalents). After the
addition was completed the reaction mixture was allowed to
spontaneously heat up to room temperature, while stirring
J. Mater. Chem., 2001, 11, 434±443
441

View Article Online
continued for 24 h at room temperature. The dilithiated species
2a that formed during this time was not isolated but treated in
situ at 275 ³C under nitrogen over 30 min with 1-bromoocta-
decane (83.35 g, 0.25 mol). The stirred reaction mixture was
then allowed to warm to room temperature. After 24 hours it
was poured onto ice, the product extracted with diethyl ether
(3675 cm3) and the organic extracts dried (MgSO4) and
®ltered. Removal of solvents followed by recrystallisation of
the residue from diethyl ether or warm ethanol gave 2,5-
bis(octadecyl)thiophene 3d (55.4 g, 94%) as an off-white solid,
mp 56 ³C (Found: C, 81.6; H, 13.0; S, 5.5. C40H76S requires C,
81.6; H, 13.0; S, 5.4%); dH(270 MHz, CDCl3) 0.88 (6 H, t,
26CH3), 1.3 (60 H, m, 306CH2), 1.64 (4 H, m, 26CH2), 2.71
(4 H, t, 26CH2) and 6.48 (2 H, s, C4H2S). Note: The decyl and
dodecyl derivatives 3a and 3b were fractionally distilled at
0.1 mmHg for puri®cation.
removal and recrystallisation of the residue from ethanol gave
pure off-white 2,5-bis(octadecyl)thiophene 1,1-dioxide 4d
(11.6 g, 94%), mp 88 ³C (Found: C, 77.4; H, 12.4; S, 4.7.
C40H76SO2 requires C, 77.4; H, 12.3; S, 5.1%); dH(270 MHz,
CDCl3) 0.90 (6 H, t, 26CH3), 1.28 (60 H, m, 306CH2), 1.62 (4
H, m, 26CH2), 2.47 (4 H, t, 26CH2) and 6.23 (2 H, s,
C4H2SO2). Note: this method gave all long chain dialkylated
thiophene 1,1-dioxides in high yield.
3,6-Dialkylphthalonitriles 5. 3,6-Bis(octadecyl)phthalonitrile
(Table 1, 5d). In a typical experiment 2,5-bis(octadecyl)thio-
phene 1,1-dioxide 4d (5 g, 8.06 mmol) and fumaronitrile
(0.629 g, 8.06 mmol) were introduced into a glass tube using
a minimum volume of chloroform (ca. 1.5 cm3). The tube was
sealed and heated at 160 ³C for 18 h. The contents of the tube
were dissolved in chloroform and the solvent evaporated under
reduced pressure at 90 ³C. The oily residue was kept under
reduced pressure until it no longer liberated gas (presumed to
be SO2) any more (ca. 20±30 min). The remaining residue was
chromatographed over silica with toluene as eluent. The second
eluted band afforded after solvent removal and recrystallisa-
tion from ethanol 3,6-bis(octadecyl)phthalonitrile 5d (1.4 g,
27%) as an off-white solid, mp 82 ³C (Found: C, 83.6; H, 12.0;
N, 4.2. C44H76N2 requires C, 83.5; H, 12.1; N, 4.4%);
2,5-Dialkylthiophene 1,1-dioxides 4. 2,5-Bis(dodecyl)thiophene
1,1-dioxide (Table 1, 4b), sodium perborate method. To a stirred
mixture of 2,5-bis(dodecyl)thiophene (11.776 g, 20 mmol) in
glacial acetic acid (85 cm3) was added solid NaBO3?4H2O
(23.079 g, 150 mmol) over ca. 5 min with care being taken to
maintain the internal temperature between 48 and 52 ³C. The
reaction mixture was then stirred at 50 ³C for another 5 h before it
was left at room temperature for 16 h. The acetic acid was
removed under reduced pressure and the solid residue stirred with
80 cm3 H2O and ®ltered. The remaining solid was extracted with
diethyl ether (100 cm3) and the organic solution washed with H2O
(100 cm3) and brine (100 cm3) before it was dried (MgSO4).
Removal of the solvent and recrystallisation of the waxy residue
from hexane gave 2,5-bis(dodecyl)thiophene 1,1-dioxide 4b (1.55g,
17%) as a white solid, mp 73 ³C (Found: C, 73.8; H, 11.5; S, 6.8.
C28H52SO2 requires C, 74.2; H, 11.6; S, 7.1%); dH(270 MHz,
CDCl3) 0.90 (6 H, t, 26CH3), 1.3 (36 H, m, 186CH2), 1.64 (4 H,
m, 26CH2), 2.49 (4 H, t, 26CH2) and 6.25 (2 H, s, C4H2SO2).
Note: This method was not successful for the syntheses of the
pentadecyl and octadecyl derivatives 4c and 4d respectively.
n
max(Nujol) 2 230 cm21 (CMN); dH(270 MHz, CDCl3) 0.89 (6
H, t, 26CH3), 1.3 (60 H, m, 306CH2), 1.63 (4 H, m, 26CH2),
2.83 (4 H, t, 26CH2) and 7.47 (2 H, s, C6H2). Note: shorter
chain alkylated phthalonitriles12 were obtained in yields
approaching 50%.
Metal-free phthalocyanines 6. 1,4,8,11,15,18,22,25-Octakis
(pentadecyl)phthalocyanine (Table 2, 6c). In a typical synth-
esis 3,6-bis(pentadecyl)phthalonitrile 5c (0.800 mg, 1.46 mmol)
was dissolved in warm (80 ³C) pentanol (8 cm3). An excess of
clean lithium metal (0.3±0.4 g, 43±58 mmol) was added in small
portions and the mixture heated for 16 h at 110 ³C. The cooled,
deep green coloured suspension was stirred with acetone
(50 cm3), the solution ®ltered and the solids washed with
acetone (50 cm3) before the combined acetone solutions were
concentrated to ca. 25 cm3. Acetic acid (50 cm3) was added, the
heterogeneous mixture stirred for 30 minutes and the pre-
cipitate collected to afford after recrystallisation from THF±
methanol or CHCl3±methanol 1,4,8,11,15,18,22,25-octakis-
(pentadecyl)phthalocyanine 6c (170 mg, 21%), phase transi-
tions as per Table 2 (Found: C, 82.3; H, 11.2; N, 5.6.
2,5-Bis(pentadecyl)thiophene 1,1-dioxide (Table 1, 4c), m
-chloroperoxybenzoic acid method. A solution of 2,5-bis(pen-
tadecyl)thiophene (10.08 g, 20 mmol) in 250 cm3 dichloro-
methane was stirred with 50±60% m-chloroperoxybenzoic acid
(12.5 g, y30 mmol) at 0 ³C in the presence of an excess of
NaHCO3 (50 g, 77 mmol). The heterogeneous mixture was left
standing for 16 hours at 5 ³C before all solids were ®ltered and
washed with dichloromethane (26100 cm3). The combined
organic phases were washed with 20% NaOH (26100 cm3),
H2O (26100 cm3) and dried (MgSO4) to give, after solvent
removal and recrystallisation of the waxy residue from warm
ethanol, 2,5-bis(pentadecyl)thiophene 1,1-dioxide 4c (2.0 g,
19%) as a white solid, mp 79 ³C (Found: C, 75.9; H, 12.0; S, 6.0.
C34H64SO2 requires C, 76.0; H, 12.0; S, 6.0%); dH(270 MHz,
CDCl3) 0.90 (6 H, t, 26CH3), 1.3 (48 H, m, 246CH2), 1.62 (4
H, m, 26CH2), 2.48 (4 H, t, 26CH2) and 6.25 (2 H, s,
C4H2SO2). Note: This method gave C12 and shorter chain
dialkylated thiophene 1,1-dioxides in 41% yield or better.
C128H210N8?2CHCl3 requires C, 82.6; H, 11.4; N, 6.0%);
dH(270 MHz, C6D6) 0.90 (24 H, t, 86CH3), 1.25 (112 H, m,
566CH2), 1.5 (16 H, m, 86CH2), 1.85 (16 H, m, 86CH2),
2.38 (16 H, m, 86CH2), 4.77 (16 H, m, 86CH2) and 7.95 (8 H,
s, 46C6H2). Note: in some experiments it was bene®cial to
chromatograph the crude product on silica with petroleum
ether : THF (99 : 1) prior to recrystallisation. The C18 derivative
was not soluble enough to allow chromatography.
Zinc-containing phthalocyanines 7. 1,4,8,11,15,18,22,25-
Octakis(octadecyl)phthalocyaninatozinc(II) (Table 2, 7d). A
stirred mixture of metal-free 1,4,8,11,15,18,22,25-octakis(octa-
decyl)phthalocyanine (6d, 240 mg, 0.099 mmol), zinc acetate
dihydrate (90 mg, 0.41 mmol) and pentanol (10 cm3) was
heated at 130 ³C for 5 h. Methanol (20 cm3) was added to
the cooled reaction mixture and the precipitate ®ltered and
washed with excess methanol. A ®ltered, concentrated, warm
(60 ³C) chloroform solution of the collected blue precipitate
was submerged in a water bath at 60 ³C (to retard the cooling
rate) and allowed to cool slowly to room temperature. The
collected blue solid was washed thoroughly with cold THF and
then with chloroform to yield 1,4,8,11,15,18,22,25-octakis-
(octadecyl)phthalocyaninatozinc(II), 7d, (105 mg, 41%), phase
2,5-Bis(octadecyl)thiophene 1,1-dioxide (Table 1, 4d),
dimethyldioxirane method. To a 2 l 2-neck ¯ask, equipped
with an ef®cient stirrer, a large CO2±acetone condenser and
charged with H2O (440 cm3), acetone (320 cm3) and NaHCO3
(240 g) was added a solution of 2,5-bis(octadecyl)thiophene
(11.76 g, 20 mmol) in dichloromethane (360 cm3). The resulting
heterogeneous mixture was cooled in an ice bath before solid
oxone (400 g) was carefully added over 30 min under ef®cient
stirring. Stirring continued at room temperature for 16 hours
before water (2 l) was added to dissolve most inorganics. The
decanted aqueous layer, and all remaining solids were extracted
with dichloromethane (1 l), the combined organic phases
washed with water (1±2 l) and dried (MgSO4) before solvent
442
J. Mater. Chem., 2001, 11, 434±443

View Article Online
K. J. Harrison, Mol. Cryst. Liq. Cryst., 1991, 196, 103; M. J. Cook,
J. Mater. Sci., Mater. Electron., 1994, 5, 117; M. J. Cook,
S. J. Cracknell and K. J. Harrison, J. Mater. Chem., 1991, 1, 703
and references therein..
transitions as per Table 2 (Found: C, 78.8; H, 11.4; N, 3.9.
176H304N8Zn?CHCl3 requires C, 78.4; H, 11.3; N, 4.1%);
dH(270 MHz, CDCl3, 40 ³C) 0.92 (24 H, t, 86CH3), 1.1±1.7
(240 H, m, 1206CH2), 2.18 (16 H, m, 86CH2), 4.55 (16 H, m,
86CH2) and 7.85 (8 H, s, 46C6H2).
C
12 N. B. McKeown, I. Chambrier and M. J. Cook, J. Chem. Soc.,
Perkin Trans. 1, 1990, 1169; I. Chambrier, M. J. Cook,
S. J. Cracknell and J. McMurdo, J. Mater. Chem., 1993, 3, 841.
13 M. J. Cook, D. A. Mayes and R. H. Poynter, J. Mater. Chem.,
1995, 5, 2233.
Acknowledgement
14 J. D. Wright, P. Roissin, G. P. Rigby, R. J. M. Nolte, M. J. Cook
and S. C. Thorpe, Sens. Actuators B, 1993, 13±14, 276.
15 C. Ometto, C. Fabris, C. Milanesi, G. Jori, M. J. Cook and
D. A. Russell, Br. J. Cancer, 1996, 74, 1891.
16 M. J. Cook, I. Chambrier, S. J. Cracknell, D. A. Mayes and
D. A. Russell, Photochem. Photobiol., 1995, 62, 542.
17 A. McKillop and J. A. Tarbin, Tetrahedron, 1990, 31, 5955.
18 R. Curci, A. Giovine and G. Modena, Tetrahedron, 1966, 22, 1235;
D. J. Brown and P. W. Ford, J. Chem. Soc. (C), 1969, 2720;
G. A. Russel and L. A. Ochrymowycz, J. Org. Chem., 1970, 35,
2106.
19 R. W. Murray and R. Jeyaraman, J. Org. Chem., 1985, 50, 2847;
P. F. Corey and F. E. Ward, J. Org. Chem., 1986, 51, 1925;
R. W. Murray, R. Jeyaraman and M. K. Pillay, J. Org. Chem.,
1987, 52, 746; W. Adam, Y.-Y. Chan, D. Cremer, J. Gauss,
D. Scheutzow and M. Schindler, J. Org. Chem., 1987, 52, 2800;
W. Adam, W. Haas and G. Sieker, J. Am. Chem. Soc., 1984, 106,
5020.
20 Y. Miyahara and T. Inazu, Tetrahedron Lett., 1990, 31, 5955.
21 E. M. Maya, P. Haisch, P. Vazquez and T. Torres, Tetrahedron,
1998, 54, 4397.
The authors gratefully acknowledge ®nancial support from
NRF (South Africa, JCS), EPSRC (MJC) and the Central
Research Funds of the University of the Orange Free State
(JCS). JCS also thanks the University of the Orange Free State
for sabbatical leave and the Association of Commonwealth
Universities for a fellowship. MJC thank the British Council
and NAVF, Norway, for support of NK-H of the University of
Bergen during her secondment to UEA. We thank Ms V. Boote
at Manchester University for running MALDI-TOF spectra
and Dr I. Fernandes is gratefully acknowledged for invaluable
support and discussions. This work was undertaken in part as a
contribution to projects run under the auspices of CEC Cost
programme 514 and CEC RTN contract RT N1-1999-00314.
References
1
D. Birkett, Chem. Ind., 2000, 178; R. Ao, L. Kilmmert and
D. Haarer, Adv. Mater., 1995, 7, 495; Phthalocyanines: Properties
and Applications, eds. C. C. Leznoff and A. B. P. Lever, VCH, New
York, 1989, vol. 1; Phthalocyanines: Properties and Applications,
eds. C. C. Leznoff and A. B. P. Lever, VCH, New York, 1992, vol.
2; Phthalocyanines: Properties and Applications, eds. C. C. Leznoff
and A. B. P. Lever, VCH, New York, 1993, vol. 3; Phthalocya-
nines: Properties and Applications, eds. C. C. Leznoff and
A. B. P. Lever, VCH, New York , 1996, vol. 4.
22 K. Poon, Y. Yan, X. Li and D. P. Ng, Organometallics, 1999, 18,
3528; M. Hu, N. Brasseur, S. Z. Yildiz and J. E. van Lier, J. Med.
Chem., 1998, 41, 1789.
23 M. J. Cook, S. J. Cracknell, G. R. Moore, M. J. Osborne and
D. J. Williamson, Magn. Reson. Chem., 1991, 29, 1053.
È
24 E. Musluoglu, V. Ahsen, A. GuÈl and O. BekaÃroglu, Chem. Ber.,
1991, 124, 2531.
25 M. J. Cook, W. L. Davis and J. C. Swarts, unpublished work.
26 D. H. Evans, K. M. O' Connell, R. A. Peterson and M. J. Kelly,
J. Chem. Educ., 1983, 60, 291; P. T. Kissinger and W. R. Heineman,
J. Chem. Educ., 1983, 60, 702; J. J. van Benschoten, J. Y. Lewis
and W. R. Heineman, J. Chem. Educ., 1983, 60, 772; T. Sawyer
and J. L. Roberts Jr., Experimental Electrochemistry for Chemists,
Wiley, New York, 1974, p. 118.
27 A. B. P. Lever, E. R. Milaeva and G. Speier in Phthalocyanines:
Properties and Applications, eds. C. C. Leznoff and A. B. P. Lever,
VCH, New York, 1993, vol. 3, pp. 1±69; R. H. Campbell,
G. A. Heath, G. T. Hefter and R. C. S. McQueen, J. Chem
Soc., Chem. Commun., 1983, 1123.
28 D. W. Clack, N. S. Hush and I. S. Woolsey, Inorg. Chim. Acta,
1976, 19, 129.
29 Y. L. Wong, J. F. Ma, W. F. Law, Y. Yan, W. T. Wong,
Z. Y. Zhang, T. C. Mak and D. K. P. Ng, Eur. J. Inorg. Chem.,
1999, 313.
2
3
K. Kasuga and M. Tsutsui, Coord. Chem. Rev., 1980, 32, 67;
Organic coating Technology, Wiley, New York, 1961, vol. 2,
pp 896±899; C. R. Martens in Riegel's Handbook of Industrial
Chemistry, ed. J. A. Kent, Van Nostrand-Reinhold, New York,
1974, ch. 22.
J. D. Wright, Prog. Surf. Sci., 1989, 31, 1; A. W. Snow and
W. R. Barger in Phthalocyanines: Properties and Applications, eds.
C. C. Leznoff and A. B. P. Lever, VCH, New York, 1989, vol. 1,
p. 341; A. W. Snow, W. R. Barger, M. Klusty, H. WoÈhltjen and
N. L. Jarvis, Langmuir, 1986, 2, 513.
H. Ali and J. E. van Lier, Chem. Rev., 1999, 99, 2379; L. Milgrom
and S. Macrobert, Chem. Br., 1998, 45; R. W. Boyle, C. C. Leznoff
and J. E. van Lier, Br. J. Cancer, 1993, 67, 1177.
G. de la Torre, P. VaÂzquez, F. AgulloÂ-LoÂpez and T. Torres,
J. Mater. Chem., 1998, 8, 1671; A. Grund, A. Kaltbeitzel,
A. Mathy, R. Schwarz, C. Bubeck, P. Vernmehren and M. Hanack,
J. Phys. Chem., 1992, 96, 7450.
4
5
30 A. Giraudeau, A. Louati, M. Gross, J. J. Andre, J. Simon, C. H. Su
and K. M. Kadish, J. Am. Chem. Soc., 1983, 105, 2917.
31 R. R. Gagne, C. A. Koval and G. C. Licensky, Inorg. Chem., 1980,
19, 2854.
32 Z. Jin, K. Nolan, C. R. McArthur, A. B. P. Lever and C. C. Leznof,
J. Organomet. Chem., 1994, 205, 468; E. S. Schmidt,
T. S. Calderwood and T. C. Bruice, Inorg. Chem., 1986, 25, 3718.
33 G. C. Bryant, M. J. Cook, C. Ruggiero, T. J. Ryan, A. J. Thorne,
S. D. Haslam and R. M. Richardson, Thin Solid Films, 1994, 243,
316.
6
7
8
9
D. WoÈhrle, D. Schlettwein, M. Kirschenmann, M. Kaneko and
A. Yamada, J. Macromol. Sci. Chem., 1990, A27, 1239.
H. Eichhorn, J. Porphyrins Phthalocyanines, 2000, 4, 88; D. WoÈhrle
and D. Meissner, Adv. Mater., 1991, 3, 129.
H. Eckert and Y. Kiesel, Angew. Chem., Int. Ed. Engl, 1981, 20,
473.
T. Buck, D. WoÈhrle, G. Schulz-Ekloff and A. Andreef, J. Mol.
Catal., 1991, 70, 259.
10 M. J. Cook, G. Cooke and A. Jafari-Fini, Chem. Commun., 1996,
1925; C. Piechocki, J. Simon, A. Skoulious, D. Guillon and
P. J. Weber, J. Am. Chem. Soc., 1982, 104, 5245.
11 A. S. Cherodian, A. N. Davies, R. M. Richardson, M. J. Cook,
N. B. McKeown, A. J. Thomson, J. Feijoo, G. Ungar and
34 D. T. Sawyer and J. L. Roberts Jr., Experimental Electrochemistry
for Chemists, Wiley, New York, 1974, p. 54.
35 G. A. Mabbott, J. Chem. Ed., 1983, 60, 697.
J. Mater. Chem., 2001, 11, 434±443
443