Synthesis of unsymmetrical alkyloxy/aryloxy-azaphthalocyanines based on a transetherification reaction

Article abstract of DOI:10.1002/ejoc.201100689

An approach leading to unsymmetrical azaphthalocyanines (AzaPcs) bearing one aryloxy and seven butoxy substituents on the periphery is described. Several cyclotetramerization methods were tested, including the template effect of a metal salt in a melt/sol

Full text of DOI:10.1002/ejoc.201100689

FULL PAPER
DOI: 10.1002/ejoc.201100689
Synthesis of Unsymmetrical Alkyloxy/Aryloxy-azaphthalocyanines Based on a
Transetherification Reaction
[a]
[b]
[b]
[b]
ˇ
Veronika Novakova, Miroslav Miletin, Kamil Kopecky, Sárka Franzová, and
Petr Zimcik*[b]
Keywords: Phthalocyanines / Fluorescence / Singlet oxygen / Template synthesis / Ethers
An approach leading to unsymmetrical azaphthalocyanines
(AzaPcs) bearing one aryloxy and seven butoxy substituents
on the periphery is described. Several cyclotetramerization
methods were tested, including the template effect of a metal
salt in a melt/solution and the Linstead method, but only one
was found to be suitable for the isolation of unsymmetrical
alkyloxy/aryloxy-AzaPcs. Undesired transetherification with
alkoxides, which is common with these types of compounds,
was considered to be an advantage in this approach, and an
unsymmetrical AzaPc was isolated in 10% yield. An analysis
of the transetherification process is also described. Unsym-
metrical zinc(II) and magnesium(II) AzaPcs showed excellent
spectral properties (λmax = 620 nm; ε = 1.5–2.2ϫ105 dm3
mol–1 cm–1) as well as promising singlet oxygen (ΦΔ = 0.44
and 0.26, respectively) and fluorescence quantum yields (ΦF
= 0.50 and 0.65, respectively). The metal-free derivative (λmax
= 603 and 640 nm; ε = 0.80 and 1.2ϫ105 dm3 mol–1 cm–1) dis-
sipated energy in a different way, and the ΦΔ and ΦF values
were one order of magnitude smaller (0.030 and 0.053,
respectively). All the parameters were comparable to the cor-
responding symmetrical AzaPcs containing eight butoxy sub-
stituents, which suggests that an unsymmetrical composition
of AzaPc on the periphery has no negative effect on the pho-
tophysical or photochemical properties.
ΦF and ΦΔ because of effective intramolecular charge trans-
fer.[13] Such AzaPcs have recently found application as dark
quenchers in DNA hybridization probes.[14] Unsymmetrical
compounds, with only one or two modifiable groups on the
periphery of the AzaPcs for binding to biomolecules, are
highly desirable in the synthesis of conjugates in PDT, fluo-
rescent probes, and dark quenchers, because this prevents
unwanted polymerization reactions. Several unsymmetrical
alkylsulfanyl-[13,15] as well as alkylamino-substituted[13,14,16]
AzaPcs have already been prepared by using the Linstead
method,[17,18] employing alkoxides as initiators of cyclo-
tetramerization. However, carbon atoms 5 and 6 of the pyr-
azine moiety in alkyloxy- or aryloxy-5,6-disubstituted pyr-
azine-2,3-dicarbonitriles, which are precursors of alkyloxy-
and aryloxy-AzaPcs, are highly electron-deficient. That is
why application of the Linstead method for cyclotetra-
merization was reported to always yield transetherification
of peripheral substituents by the alkoxide used for initia-
tion.[11,19–21] This problem has only recently been solved for
symmetrical AzaPcs bearing peripheral chains connected
through oxygen atoms.[11,21,22] However, the synthesis of un-
symmetrical alkyloxy- or aryloxy-AzaPcs still represents a
challenge. On the other hand, alkyloxy- or aryloxy-substi-
tuted Pcs are easily synthesized without any problems with
stability of the peripheral substituents.[23,24]
Introduction
It is not surprising that phthalocyanines (Pcs), which are
structural analogues of porphyrins, have become the target
of many research groups, because their promising photo-
physical and photochemical properties mean that Pcs play
an important role in fields such as photodynamic therapy
(PDT),[1,2] chemical sensors,[3] catalytic chemistry,[4] photo-
voltaics,[5–7] and nonlinear optics.[8,9] Their aza analogues,
azaphthalocyanines (AzaPcs) from the group of tetrapyraz-
inoporphyrazines, have comparable properties with several
advantages, for example, better solubility and a diverse
range of peripheral substituents. Peripheral substitution
plays a crucial role in defining the properties of AzaPc and
Pc derivatives. It has been shown that alkylsulfanyl-[10] and
alkyloxy-substituted[11,12] AzaPcs are characterized by high
singlet oxygen (ΦΔ) and fluorescence quantum yields (ΦF),
which makes them ideal candidates for cancer treatment by
PDT, or fluorescence detection. On the other hand, alk-
ylamino substitution leads to values of almost zero for both
[a] Department of Biophysics and Physical Chemistry, Charles
University in Prague,
Heyrovskeho 1203, 50005 Hradec Kralove, Czech Republic
[b] Department of Pharmaceutical Chemistry and Drug Control,
Charles University in Prague,
Heyrovskeho 1203, 50005 Hradec Kralove, Czech Republic
Fax: +420-495067167
This work mainly concentrates on new synthetic ap-
proaches to unsymmetrical aryloxy- and alkyloxy-substi-
tuted AzaPcs. Several strategies are discussed for the forma-
E-mail: petr.zimcik@faf.cuni.cz
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201100689.
Eur. J. Org. Chem. 2011, 5879–5886
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5879

P. Zimcik et al.
FULL PAPER
tion of the AzaPc core. Subsequently, the photophysical
and photochemical properties of the synthesized unsym-
metrical AzaPcs were determined so that their potential in
PDT and photodetection applications could be assessed.
Results and Discussion
Synthesis
Cyclotetramerization of substituted pyrazine-2,3-dicarbo-
nitriles based on a template effect of a metal salt or a
stepwise condensation of precursors after initiation by alk-
oxide anions (Linstead method) are two general procedures
that have been used to generate the AzaPc core. Several
approaches allowing selective formation of unsymmetrical
Pcs and related compounds have been described.[25,26] How-
ever, a statistical condensation starting from two different
precursors (A and B) followed by separation of the required
congener from a statistical mixture by column chromatog-
raphy seems to be the simplest technique leading to unsym-
metrical AzaPcs of the AAAB type.[27] Furthermore, the
use of subazaphthalocyanines (subAzaPcs), which is one of
the selective approaches,[28] has not yet been reported, and
our own preliminary results indicated that the synthesis of
subAzaPcs might be another synthetic challenge.
Scheme 1. Unsymmetrical condensation: template effect of a metal
salt. Reagents and conditions: (i) Zn(quinoline)2Cl2, quinoline (two
drops), 240 °C, 10 min; (ii) Zn(CH3COO)2, quinoline, 160 °C, 24 h.
1777.7). However, only a negligible amount of 4 was de-
The structure of the desired unsymmetrical aryloxy- tected on TLC (silica; chloroform/acetone, 20:1; Rf = 0.11)
AzaPc of the AAAB type was designed according to the when compared with 3. No improvement was observed
following requirements. Precursor 1, with two hydroxy upon changing the ratio of starting materials to 3:1 (excess
groups intended for binding AzaPcs to biomolecules, of 1) or stepwise addition of 2 to the reaction mixture.
should form one quarter of the final AzaPc (Scheme 1). The Hence, although the synthesis of unsymmetrical aryloxy-
rest of the molecule should be built from precursor 2, the substituted AzaPc by using the template effect was clearly
bulky 2,6-diisopropylphenoxy groups of which preclude un- successful, the yields were not sufficiently satisfactory to
desired aggregation behavior of the final unsymmetrical scale-up this reaction.
AzaPc. As mentioned above, the Linstead method of cyclo-
For this reason, the concept of introducing one func-
tetramerization leads to transetherification of peripheral tional group to the periphery of the macrocycle was
alkyloxy- or aryloxy-substituted AzaPcs. For this reason, changed. The novel idea was to take advantage of the trans-
the template effect was considered first as the cyclotetra- etherification reaction, instead of considering it to be a dis-
merization method so that the peripheral substituents in the advantage. In this new approach, compound 1 reacted with
precursors 1 and 2 could be retained.
magnesium butoxide in butanol to initiate AzaPc core for-
Compounds 1 and 2 (Scheme 1) were mixed in a 1:1 mo- mation (Scheme 2). At the same time, the butoxide anion,
lar ratio with 2 equiv. of Zn(quinoline)2Cl2 and heated ac- behaving as a strong nucleophile, attacked the electron-de-
cording to the method developed by Mørkved et al.[29] The ficient carbon atoms at positions 5 and 6 of the pyrazine
reaction mixture melted and immediately turned green after ring of precursor 1, as anticipated. The reaction was
applying temperatures higher than 220 °C, indicating the stopped after 2 h, and the crude mixture was analyzed by
formation of the AzaPc core. However, TLC investigation mass spectrometry. The mass fragments corresponding to
and MS analysis of the crude reaction mixture showed the AzaPcs 6 (eight butoxy; m/z = 1120.5), 7 (seven butoxy, one
presence of only two AzaPcs, 3 and 5, with no trace of the 4-(hydroxymethyl)phenoxy; m/z = 1170.4) and the com-
unsymmetrical congeners. The likely explanation was that pound bearing six butoxy and two 4-(hydroxymethyl)phen-
the reaction did not take place in solution but in a mixture oxy peripheral substituents (m/z = 1220.5) were detected in
of solids that melted at high temperatures.
the mass spectrum, together with clear signals of the corre-
Makhseed et al. published several papers on the synthesis sponding dimers and trimers (see Figure S1 in the Support-
of symmetrical aryloxy-substituted Pcs and AzaPcs in solu- ing Information). The mixture was also analyzed by TLC
tion.[21,30–32] Similar conditions (quinoline, zinc acetate, (silica; toluene/pyridine/MeOH, 10:1:1). Selected blue frac-
160 °C) were applied for the precursors 1 and 2 in subse- tions were scraped from the TLC plate and analyzed by
quent exploratory reactions (Scheme 1). Mass spectra taken mass spectrometry (MS). The two most intense fractions
directly from the crude reaction mixture showed mass frag- (with approximately the same intensity) corresponded to 6
ments corresponding to the AzaPc 3 (m/z = 1992.9), AzaPc and 7 (Rf = 0.73 and 0.62, respectively) and were used for
4 (m/z = 1885.8), and traces of the A2B2 congener (m/z = subsequent kinetic studies.
5880
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5879–5886

Unsymmetrical Alkyloxy/Aryloxy-azaphthalocyanines
Scheme 2. Unsymmetrical condensation, a transetherification ap-
proach (8 and 10 were prepared previously).
To shed some light on the transetherification process, the
progress of the cyclotetramerization of 1 in excess magne-
sium butoxide was monitored during the reaction. The
quantitative isolation of 6 and 7 from the two-dimensional
TLC plate was monitored by UV/Vis spectroscopy (for de-
tails see the Experimental Section and the Supporting In-
formation, Figures S2 and S3). The other congeners were
not monitored, because their amounts seemed to be low
according to MS and TLC analysis. Each experiment was
Figure 1. Kinetic study of the transetherification: (a) increase of 6
(filled triangles) and 7 (empty squares) in time. (b) Change of the
6/7 ratio with reaction time; (dots) cyclotetramerization of 1 with
magnesium butoxide in butanol, (empty triangles) reflux of 7 with
magnesium butoxide in butanol. The amounts of 6 and 7 in the
reaction mixture used to calculate the 6/7 ratio were derived from
absorbance at the Q-band and the extinction coefficient in THF.
These results point to the following plausible sequence
performed in triplicate, and the kinetic behavior was similar of reactions (Scheme 3). The transetherification runs very
for all cases. The results of one representative experiment rapidly on the pyrazine-2,3-dicarbonitrile 1, giving rise to
are depicted in Figure 1.
compound 12 and, subsequently, to 13. The latter two com-
A slow initial phase of the cyclotetramerization within pounds undergo statistical condensation to form 6 and 7.
the first hour was indicated by a light blue-green color of As the transetherification proceeds further, the amount of
the solution and low absorbance of both products in the Q- 13 in the solution increases, and the ratio 6/7 also increases
band (Figure 1a). Subsequently, the amounts of both due to the changed ratio of the starting materials. The
AzaPcs were significantly elevated, indicating a rapid cyclo- transetherification process of any remaining 12 to give 13 is
tetramerization reaction. The amount of 7 in the solution most likely fully complete after 2 h; thus, compound 7
stopped increasing after approximately 2 h, whereas the reaches a plateau (Figure 1a) due to the lack of one precur-
amount of 6 continued to increase for a further hour under sor. The rest of 13 in solution is transformed into AzaPc 6
reflux. The amount of 7 slightly decreased between 2 and within the next hour, and the cyclotetramerization of all
5 h reaction time. Interestingly, initially the amount of 7 pyrazine-2,3-dicarbonitriles is finished after approximately
slightly exceeded the amount of 6 (Figure 1b). However, the 3 h from the start of the reaction. This is in good agreement
ratio of 6/7 increased with further heating and reached a with the previously published direct cyclotetramerization of
plateau after 3 h, with only a moderate increase thereafter. 13 to 6 in which the maximum yield was obtained after the
For comparison, when the same reaction was performed same reaction time of 3 h.[10] The transetherification of the
with two precursors (1 and 13) instead of only compound final AzaPc core likely takes place simultaneously, as de-
1, very similar results were obtained with slightly higher duced from the decreasing amount of 7 in the reaction mix-
ratios of 6/7. To examine the transetherification of the ture after 2 h. However, this rate is several times lower than
AzaPc core in more detail, isolated 7 was heated to reflux the transetherification of the precursors, as shown for the
with magnesium butoxide in butanol under the same condi- reaction of 7 with magnesium butoxide. As a consequence,
tions used for the previous cyclotetramerization studies, and this process contributes only slightly to changes in the ratio
the kinetic results were compared (Figure 1b). In this case, of 6/7.
the ratio of 6/7 changed very slowly and achieved a value
Despite the fact that compounds 6 and 7 were substi-
of only 0.25 after 4 h of reaction.
tuted by non-branched aliphatic substituents, no aggrega-
Eur. J. Org. Chem. 2011, 5879–5886
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5881

P. Zimcik et al.
FULL PAPER
The ether linkage between the peripheral substituents and
the AzaPc macrocycle caused a small hypsochromic shift
of the Q-band in comparison to unsubstituted zinc-AzaPc
(ZnAzaPc; λmax = 636 nm in pyridine[33]). Similar observa-
tions were previously published.[11] It was also reported that
aryloxy substituents caused less pronounced hypsochromic
shifts than alkyloxy groups, because aryl moieties contrib-
uted to a conjugation of the macrocyclic system.[11] How-
ever, taking into account that 7, 9, and 11 contain only one
aryloxy and seven alkyloxy substituents, it is not surprising
that the Q-band positions differed only by 1 nm (Table 1)
from the symmetrical derivatives 6, 8, and 10, respectively.
All Q-bands were sharp, indicating no aggregation.
Scheme 3. Suggested reactions occurring during transetherification.
tion was observed in the UV/Vis spectra at concentrations
used for photophysical and photochemical measurements
(see below). Thus, although the steric bulk of the 2,6-diiso-
propylphenoxy groups considered in the AzaPc template
approach may be advantageous, the butoxy groups also
seem to be a suitable peripheral substituent that also en-
sures beneficial properties of the final AzaPcs.
Results of the analysis described above indicated an opti-
mal reaction time of 2 h for the scale-up synthesis of 7 from
1 with a yield of 10%. This yield is typical of AAAB-type
congeners from a statistical condensation and is compar-
able with other reported reactions.[14,15] To investigate the
photophysical and photochemical properties of the zinc(II)
complex, which is generally more suitable for PDT, it was
prepared from 7. The central, weakly chelated magne-
sium(II) ion was removed under acidic conditions (p-tolu-
enesulfonic acid) to form metal-free 9 in 32% yield. The
lower yield of this reaction may be attributed to a loss dur-
ing purification (strong silica binding of 9) rather than to
low conversion. Heating of 9 in N,N-dimethylformamide
(DMF) with zinc acetate gave zinc complex 11 almost
quantitatively (95% yield).
Figure 2. Normalized absorption spectra of 7 (dotted), 9 (dashed),
and 11 (full) in THF.
The strength of absorption is a very important parameter
in many applications connected with light absorption.
Whereas porphyrins have extinction coefficients in the Q-
band of only 1000–2000 dm3 mol–1 cm–1 and chlorins
20000–30000 dm3 mol–1 cm–1, Pcs and AzaPcs achieved val-
ues that were generally several times higher.[34,35] This was
also valid for compounds studied in this work (see
Table 1);
ε values were found to be approximately
100000 dm3 mol–1 cm–1 for the metal-free derivative and
over 150000 dm3 mol–1 cm–1 for the metal complexes.
UV/Vis Absorption
Fluorescence Emission and Singlet Oxygen Production
The absorption spectra of the prepared AzaPcs in tetra-
Quantum yields of singlet oxygen as well as fluorescence
hydrofuran (THF) showed shapes that were typical for Pcs were determined in THF by using a comparative method
and related compounds, and were composed of a low-en- with ZnPc as reference (ΦΔ = 0.53 in THF,[36] ΦF = 0.30 in
ergy Q-band and a high-energy B-band at around 620 and chloronaphthalene[37]) (Table 1). Singlet oxygen production
360 nm, respectively (Figure 2, Table 1). The Q-band of was calculated from the decomposition of the chemical trap
metal-free 9 was split due to the loss of symmetry (D2h). 1,3-diphenylisobenzofuran (DPBF). No changes in the ab-
Table 1. Spectroscopic, photophysical, and photochemical properties of studied compounds in THF.
Absorbance: λmax [nm] (ε [105 dm3 mol–1 cm–1])
Fluorescence: λmax [nm]
ΦF
ΦΔ
[a]
[a]
M
6
Mg
Mg
2H
2H
Zn
620 (2.45)
621 (2.22)
643 (0.98), 603 (0.67)
643 (1.12), 603 (0.80)
618 (2.42)
625
627
646
647
623
625
0.96
0.65
0.056
0.053
0.51
0.50
0.26
0.26
0.029
0.030
0.47
0.44
7
8
9
10
11
Zn
619 (1.53)
[a] Mean of three independent measurements. Estimated error Ϯ20% and Ϯ10% for ΦF and ΦΔ, respectively. No changes in absorption
spectra were observed during the measurements.
5882
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5879–5886

Unsymmetrical Alkyloxy/Aryloxy-azaphthalocyanines
sorption spectra appeared during measurements, indicating
that neither photodegradation nor aggregation of AzaPcs
occurred. The ΦΔ values decreased in the order 11 Ͼ 7 Ͼ
9, corresponding to 0.44, 0.26 and 0.030, respectively. Due
to their similar peripheral substitution patterns, they did
not differ much from the symmetrical compounds, which
have ΦΔ values of 0.47, 0.26 and 0.029 for 10 Ͼ 6 Ͼ 8,
respectively. It is clear that neither the introduction of a free
hydroxy group nor the unsymmetrical composition of the
macrocycle negatively influenced the singlet oxygen quan-
tum yields. It is known that zinc(II) chelated in the center
of the macrocycle increases the probability of transition of
the Pc excited S1 state to the long-lived triplet T1 state, as
well as extend the T1 state lifetime.[38] As a consequence,
the singlet oxygen quantum yields of ZnPcs are high. This
was also valid in this case, and ΦΔ values of zinc(II) com-
plexes 10 and 11 significantly exceeded the others in the
series.
An opposite dependence was found in fluorescence emis-
sion. Centrally chelated magnesium(II) increased the fluo-
rescence quantum yields of AzaPcs. The ΦF of both magne-
sium complexes 6 and 7 were extremely high (0.96 and 0.65,
respectively). On the other hand, both zinc(II) complexes
10 and 11 had lower, but still very strong, ΦF values of
approximately 0.50. Both ΦF and ΦΔ values determined for
metal-free derivatives were about one order of magnitude
lower than for the metal complexes, indicating that they dis-
sipated energy through different pathways. The sum of the
quantum yields of 6 was apparently higher than 1, which is
not generally possible; however, it was still within the esti-
mated experimental error of 20% and 10% for ΦF and ΦΔ,
respectively. Fluorescence emission spectra had typical
shapes, with only small Stokes shifts (4–6 nm) (Figure 3).
Superposing the excitation spectra and the absorption spec-
tra (see Figure S4 in the Supporting Information) lead to
the conclusion that the observed fluorescence arose from
the studied compounds. The acquired data also suggests
that the compounds were exclusively in monomeric forms
because the aggregates have different absorption spectra
and, besides a few exceptions of J-dimers,[39,40] usually do
not fluoresce.
Conclusions
A synthetic method that can be used to generate unsym-
metrical AzaPcs bearing peripheral substituents connected
through oxygen was developed for the first time. Cyclotetra-
merization methods based on the template effect did not
lead to the desired unsymmetrical product either at all or
only in very low yield. A detailed study of the Linstead
method showed that, under specific conditions, transetheri-
fication can be a suitable method that gives unsymmetrical
alkyloxy/aryloxy-AzaPcs in reasonable yields. Analysis of
the reaction kinetics data showed that the transetherifi-
cation took place rapidly on pyrazine-2,3-dicarbonitriles,
giving new precursors that cyclotetramerized to unsymmet-
rical AzaPcs. The final AzaPcs undergo the transetherifi-
cation at much lower rate. The described approach can
serve to introduce one modifiable group onto the AzaPc
core, which can be used for binding to a target. According
to photophysical and photochemical measurements, intro-
duction of the free hydroxy group on the periphery did not
influence the good quantum yields of fluorescence, the gen-
eration of singlet oxygen, or the absorption spectra. The
studied AzaPcs may therefore be suitable substrates for fur-
ther investigations on photodynamic therapy or fluorescent
probes due to the high singlet oxygen and fluorescence
quantum yields.
Experimental Section
General: All organic solvents used were of analytical grade. Anhy-
drous butanol was stored over magnesium and distilled prior to
use. All chemicals for synthesis were obtained from established sup-
pliers (Aldrich, Acros, Merck) and used as received. Zinc–phthalo-
cyanine (ZnPc) was purchased from Aldrich. TLC was performed
on Merck aluminum sheets with silica gel 60 F254. Merck Kieselgel
60 (0.040–0.063 mm) was used for column chromatography. Melt-
ing points were measured with an Electrothermal IA9200 Series
digital melting point apparatus (Electrothermal Engineeering Ltd.,
Southend-on-Sea, Essex, Great Britain). Infrared spectra were
measured with a Nicolet 6700 in ATR mode. 1H and 13C NMR
spectra were recorded with a Varian Mercury Vx BB 300 spectrom-
eter. Chemical shifts are reported relative to Si(CH3)4 and were
locked to the signal of the solvent. The UV/Vis spectra were re-
corded with a Shimadzu UV-2401PC spectrophotometer. The fluo-
rescence spectra were obtained with an AMINCO-Bowman Series
2 luminescence spectrometer. MALDI-TOF mass spectra were re-
corded in the positive reflectron mode with a Voyager-DE STR
mass spectrometer (Applied Biosystems, Framingham, MA, USA)
in trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]malo-
nonitrile as the matrix. The instrument was calibrated externally
with a five-point calibration using Peptide Calibration Mix1
(LaserBio Labs, Sophia-Antipolis, France). High-resolution mass
spectra were obtained by using the same instrument, but each spot
was further calibrated internally after addition of metal-free aza-
phthalocyanine with peripheral camphorquinone units[41] to the
mass of its monomer (m/z = 954.5030 [M]+) and dimer, which also
appeared in the mass spectrum (m/z = 1909.0061 [2 M]+). Com-
pounds 1–3, 5, 10[11] and 6, 8, 13[10] were prepared as reported.
Template Effect; Cyclotetramerization in a Melt: Compounds 1
(83 mg, 0.22 mmol), 2 (107 mg, 0.22 mmol) and Zn(quinoline)2Cl2
Figure 3. Normalized emission spectra of 7 (dotted), 9 (dashed),
and 11 (full) in THF.
Eur. J. Org. Chem. 2011, 5879–5886
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5883

P. Zimcik et al.
FULL PAPER
(185 mg, 0.44 mmol, prepared according to literature[29]) were thor-
oughly mixed, two drops of freshly distilled quinoline were added,
and the mixture was heated at 240 °C for 10 min. The solid was
washed with water/methanol (1:1) and analyzed by TLC (chloro-
form/pyridine, 20:1) and by mass spectrometry.
Kinetic Study: Magnesium turnings (136 mg, 10.2 mmol) and a
crystal of iodine were heated to reflux in anhydrous butanol
(10 mL) for 3 h. Precursor 1 (300 mg, 0.8 mmol) or 1 (110 mg,
0.4 mmol) and 13 (150 mg, 0.4 mmol) were added. The samples
(50 μL) were taken from the reaction at time intervals and diluted
with chloroform (1 mL) to ensure complete dissolution. The
chloroform solution of the sample (20 μL) was spotted on a TLC
plate and two-dimensional TLC (see the Supporting Information)
was used to separate 6 and 7. First, TLC elution was performed
with toluene/pyridine/MeOH (10:1:1), then the TLC plate was
thoroughly dried, rotated by 90° and the developed with toluene/
pyridine/acetone (15:1:1) as mobile phase. Spots corresponding to
AzaPcs 6 and 7 were scraped from the TLC plate and quantitatively
extracted with THF. The solvent was evaporated to dryness, the
sample was then dissolved in THF (2 mL), and the absorption
spectrum was recorded. The amounts of 6 and 7 were monitored
by observing the intensity of absorption at the Q-band maximum
(620 and 621 nm, respectively). In another experiment, magnesium
turnings (13.6 mg, 1.0 mmol) and a crystal of iodine were heated
to reflux in anhydrous butanol (2 mL) for 3 h, and 7 (94 mg,
0.08 mmol) was added. The reaction was monitored as described
above.
Template Effect; Cyclotetramerization in Solution: Compounds 1
(83 mg, 0.22 mmol), 2 (321 mg, 0.66 mmol) and anhydrous zinc
acetate (40 mg, 0.22 mmol) were thoroughly dried in a drying pistol
(5 mbar, 78 °C, 5 h), transferred into a round-bottomed flask and
put under argon. Freshly distilled quinoline (3 mL) was added, and
the mixture was immersed in an oil bath that was preheated to
160 °C and heated for 24 h. Water/methanol (1:1, 10 mL) was
added, and the precipitate was collected. The crude mixture was
dissolved in pyridine (3 mL) and mixed again with water/methanol
(1:1, 20 mL) to remove residual quinoline. The precipitate was col-
lected and dried. The composition of the mixture of products was
analyzed by TLC (chloroform/acetone, 20:1) and by mass spec-
trometry.
Linstead Method of Cyclotetramerization
From Precursor 1: Magnesium turnings (182 mg, 7.49 mmol) and a
crystal of iodine were heated to reflux in anhydrous butanol
(10 mL) for 3 h. Compound 1 (400 mg, 1.06 mmol) was added, and
the mixture was heated to reflux for 2 h. Water/methanol/acetic
acid (5:5:1, 20 mL) was poured into the reaction mixture, and the
precipitate was collected, washed with water/methanol (1:1) and
dried. The desired congener 7 was separated by column chromatog-
raphy on silica with gradient elution (toluene/pyridine/acetone
15:1:1 and then 8:1:1). The isolated fractions were purified by col-
umn chromatography again (toluene/pyridine/acetone 15:1:1) to
yield blue solids 6 (74 mg, 25%) and 7 (32 mg, 10%).
2,3,9,10,16,17,23,24-Octabutoxy-1,4,8,11,15,18,22,25-octaaza-
phthalocyanine (8): UV/Vis (THF): λmax (ε) = 643 (98100), 603
( 6 66 0 0 ) , 5 9 3 ( 5 45 0 0 ) , 5 5 2 ( 1 91 0 0 ) , 4 1 6 ( 3 65 0 0 ) , 3 46
(96700 dm3 mol–1 cm–1) nm. Other analytical data were published
elsewhere.[10]
3,9,10,16,17,23,24-Heptabutoxy-2-[4-(hydroxymethyl)phenoxy]-
1,4,8,11,15,18,22,25-octaazaphthalocyanine (9): p-Toluenesulfonic
acid (62 mg, 0.32 mmol) in THF (3 mL) was added to a solution
of 7 (38 mg, 0.032 mmol) in THF (10 mL), and the reaction mix-
ture was stirred at room temp. for 2 h. The solution was concen-
trated under reduced pressure, and water (15 mL) was added. The
precipitate was collected and washed thoroughly with water and
briefly with methanol. The crude product was purified by column
chromatography on silica (chloroform/acetone, 10:1) to yield a
(2,3,9,10,16,17,23,24-Octabutoxy-1,4,8,11,15,18,22,25-octaaza-
phthalocyanininato)magnesium(II) (6): UV/Vis (THF): λmax (ε) =
620 (244700), 596 (br.), 566 (30400), 426 (15400), 365
(138400 dm3 mol–1 cm–1) nm. Other analytical data were published
elsewhere.[10]
{3,9,10,16,17,23,24-Heptabutoxy-2-[4-(hydroxymethyl)phenoxy]-
1,4,8,11,15,18,22,25-octaazaphthalocyaninato}magnesium(II) (7):
1H NMR [300 MHz, CDCl3/[D5]pyridine (2:1)]: δ = 0.66–0.88 (m,
2 1 H , O C H 2 C H 2 C H 2 C H 3 ) , 1 . 1 1 – 1 . 4 2 ( m , 1 4 H ,
OCH2CH2CH2CH3), 1.51–1.79 (m, 14 H, OCH2CH2CH2CH3),
4.21–4.66 (m, 17 H, OCH2CH2CH2CH3, CH2OH and CH2OH),
7.30 (d, J = 8 Hz, 2 H, ArH), 7.41 (d, J = 7 Hz, 2 H, ArH) ppm.
13C NMR [75 MHz, CDCl3/[D5]pyridine (2:1)]: δ = 13.26, 18.59,
18.62, 18.71, 30.12, 30.16, 30.19, 30.24, 30.63, 63.41, 67.14, 67.22,
67.26, 67.36, 120.53, 127.51, 138.52, 139.75, 140.14, 140.54, 140.61,
1
green solid (12 mg, 32 %). H NMR [300 MHz, CDCl3/[D5]pyr-
idine, (2:1)]: δ = 0.61–1.01 (m, 21 H, OCH2CH2CH2CH3), 1.25–
1.58 (m, 14 H, OCH2CH2CH2CH3), 1.61–1.87 (m, 14 H,
OCH2CH2CH2CH3), 4.05–4.76 (m, 17 H, OCH2CH2CH2CH3,
CH2OH and CH2OH) ppm; signals of aromatic hydrogen atoms
were not detected. 13C NMR [75 MHz, CDCl3/[D5]pyridine, (2:1)]:
δ = 13.36, 13.42, 13.48, 18.58, 18.65, 18.78, 18.89, 30.04, 30.26,
30.35, 63.64, 67.15, 67.23, 67.35, 67.57, 120.55, 127.46, 151.57 and
151.65 ppm. IR (ATR): ν = 3297, 2959, 2933, 2873, 1638, 1538,
˜
1505, 1479, 1448, 1379, 1316, 1250, 1191, 1148, 1061, 1018, 949,
925, 808, 741 cm–1. MALDI-TOF: m/z = 1148 [M]+, 1171 [M +
Na]+, 1187 [M + K]+, 2297 [2 M]+, 2320 [2 M + Na]+, 2336
[2 M+ K]+. HRMS (MALDI-TOF): calcd. for [M]+ 1148.5668;
found 1148.5652. UV/Vis (THF): λmax (ε) = 643 (112500),
604 (79900), 594 (60700), 555 (20600), 418 (39800), 347
(107300 dm3 mol–1 cm–1) nm.
140.75, 150.37, 151.52, 151.75, 151.95, 152.05 ppm. IR (ATR): ν =
˜
2959, 2928, 2873, 1726, 1710, 1692, 1678, 1666, 1659, 1641, 1631,
1620, 1612, 1599, 1502, 1493, 1479, 1462, 1443, 1378, 1305, 1253,
1156, 1116, 1061, 1019, 952, 935, 851, 750 cm–1. MALDI-TOF: m/z
= 1170 [M]+, 1193 [M + Na]+, 1209 [M + K]+, 2341 [2 M]+, 2364
[2 M + Na]+, 2380 [2 M + K]+, 3511 [3 M]+, 3534 [3 M + Na]+.
HRMS (MALDI-TOF): calcd. for [M]+ 1170.5362; found
1170.5391. UV/Vis (THF): λmax (ε) = 621 (222100), 596 (28800),
566 (28400), 367 (131600 dm3 mol–1 cm–1) nm.
(2,3,9,10,16,17,23,24-Octabutoxy-1,4,8,11,15,18,22,25-octaaza-
phthalocyaninato)zinc(II) (10): UV/Vis (THF): λmax (ε) = 618
( 2 4 2 2 0 0 ) , 5 9 3 ( 3 2 4 0 0 ) , 5 6 3 ( 3 1 1 0 0 ) , 4 0 4 ( b r. ) , 3 5 8
(143500 dm3 mol–1 cm–1) nm. Other analytical data were published
elsewhere.[10]
From Precursors 1 and 13: Magnesium turnings (182 mg,
7.49 mmol) and a crystal of iodine were heated to reflux in anhy-
drous butanol (10 mL) for 3 h. Compounds 1 (200 mg, 0.53 mmol)
and 13 (147 mg, 0.53 mmol) were added at once, and reflux was
continued for 2 h. Purification was achieved as described for 7 to
{3,9,10,16,17,23,24-Heptabutoxy-2-[4-(hydroxymethyl)phenoxy]-
1,4,8,11,15,18,22,25-octaazaphthalocyaninato}zinc(II) (11): Anhy-
yield blue solids 6 (109 mg, 37%) and 7 (19 mg, 6%). The analytical drous zinc acetate (19 mg, 0.104 mmol) in DMF (2 mL) was added
data were identical to those of the products of cyclotetramerization
to 10 (12 mg, 0.010 mmol) in chloroform (5 mL), and the mixture
from precursor 1.
was heated at 160 °C for 3 h. The chloroform was evaporated, water
5884
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5879–5886

Unsymmetrical Alkyloxy/Aryloxy-azaphthalocyanines
(10 mL) was added, and the precipitate was collected and washed
thoroughly with water and briefly with methanol. The crude prod-
uct was purified by column chromatography on silica (chloroform/
acetone, 2:1) to yield a green solid (12 mg, 95%). 1H NMR
[300 MHz, CDCl3/[D5]pyridine (2:1)]: δ = 0.70–0.87 (m, 21 H,
OCH2CH2CH2CH3), 1.23–1.48 (m, 14 H, OCH2CH2CH2CH3),
1.61–1.85 (m, 14 H, OCH2CH2CH2CH3), 4.38–4.70 (m, 17 H,
OCH2CH2CH2CH3, CH2OH and CH2OH), 7.39 (d, J = 8 Hz, 2
H, ArH), 7.49 (d, J = 8 Hz, 2 H, ArH) ppm. 13C NMR [75 MHz,
CDCl3/[D5]pyridine (2:1)]: δ = 13.31, 18.68, 18.71, 18.76, 18.78,
30.21, 30.31, 63.55, 67.20, 67.24, 67.26, 120.64, 127.72, 138.63,
139.67, 140.17, 140.36, 140.43, 140.55, 141.98, 150.34, 151.76,
to give an absorption at the Q-band maximum of ca. 0.05. Absorp-
tion, emission, and excitation spectra were collected. Both reference
and sample were excited at 590 nm. ΦF was calculated by using
Equation (3):
(3)
where F is the integrated area under the emission spectrum, A is
the absorbance at the excitation wavelength (590 nm), and n is the
refractive index of the solvent. Superscripts R and S correspond
to the reference and sample, respectively. Excitation spectra were
collected with emission wavelength fixed at 678 nm. All experi-
ments were performed three times, and data presented in the paper
represent a mean of these three experiments (estimated error
Ϯ20%).
152.00 ppm. IR (ATR): ν = 2958, 2927, 2872, 1678, 1650, 1640,
˜
1543, 1502, 1441, 1377, 1305, 1254, 1192, 1119, 1062, 1018, 953,
933, 849, 745 cm–1. MALDI-TOF: m/z = 1210 [M]+, 1233 [M +
Na]+, 1249 [M + K]+, 2421 [2 M]+, 2444 [2 M + Na]+ and 2460 [2
M + K]+. HRMS (MALDI-TOF): calcd. for [M]+ 1210.4803;
found 1210.4841. UV/Vis (THF): λmax (ε) = 618 (153000), 595
(22300), 564 (20200), 360 (92700), 407 (17700 dm3 mol–1 cm–1) nm.
Supporting Information (see footnote on the first page of this arti-
cle): Mass spectrum of the crude mixture from the transetherifi-
cation approach; UV/Vis spectra obtained for the kinetic study;
absorption, emission, and excitation spectra of compounds 7, 9,and
11; 2D TLC images.
Singlet Oxygen Measurements: Quantum yields of singlet oxygen
(ΦΔ) were determined in THF according to a previously published
procedure[42] by using the decomposition of the chemical trap 1,3-
diphenylisobenzofuran (DPBF). Zinc–phthalocyanine (ZnPc) was
used as a reference (ΦΔ = 0.53 in THF[36]). In detail, the procedure
was as follows: A stock solution of DPBF in THF (5ϫ10–5 m,
2.5 mL) was transferred into a 10ϫ10 mm quartz optical cell and
bubbled with oxygen for 1 min. A defined amount of stock solution
of the tested dye in THF (usually 30 μL) was then added. Ab-
sorbance of the final dye solution at the Q-band maximum was
always ca. 0.1. The solution was then stirred and irradiated for
defined times by using a halogen lamp (Tip, 300 W). Incident light
was filtered through a water filter (6 cm) and an orange HOYA G
filter to remove infrared light and light Ͻ 506 nm, respectively. All
experiments were performed three times, and data presented in the
paper represent a mean of these three experiments (estimated error
Ϯ10%). Singlet oxygen quantum yield (ΦΔ) was calculated by using
Equation (1):
Acknowledgments
Financial support from the Charles University of Prague through
Project SVV 263001 and from the Czech Science Foundation
(P207/11/1200) is gratefully acknowledged. The authors would like
to thank Jirˇí Kunesˇ for NMR measurements and Vojteˇch Tambor
for MS measurements.
[1] P. Zimcik, M. Miletin, H. Radilova, V. Novakova, K. Kopecky,
J. Svec, E. Rudolf, Photochem. Photobiol. 2010, 86, 168–175.
[2] A. Erdogmus, T. Nyokong, J. Mol. Struct. 2010, 977, 26–38.
[3] M. Kandaz, M. N. Yarasir, T. Güney, A. Koca, J. Porphyrins
Phthalocyanines 2009, 13, 712–721.
[4] J. K. Joseph, S. L. Jain, B. Sain, Ind. Eng. Chem. Res. 2010, 49,
6674–6677.
[5] M. Garcia-Iglesias, J. J. Cid, J. H. Yum, A. Forneli, P. Vazquez,
M. K. Nazeeruddin, E. Palomares, M. Gratzel, T. Torres, En-
ergy Environ. Sci. 2011, 4, 189–194.
(1)
[6] J. J. Cid, J. H. Yum, S. R. Jang, M. K. Nazeeruddin, E. M. Fer-
rero, E. Palomares, J. Ko, M. Gratzel, T. Torres, Angew. Chem.
Int. Ed. 2007, 46, 8358–8362.
[7] C. G. Claessens, U. Hahn, T. Torres, Chem. Rec. 2008, 8, 75–
97.
where k is the slope of a plot of the dependence of ln(A0/At) on
irradiation time t, with A0 and At being the absorbances of the
DPBF at 414 nm before irradiation and after irradiation time t,
respectively. IaT is the total amount of light absorbed by the dye.
Superscripts R and S indicate reference and sample, respectively.
IaT is calculated as the sum of intensities of the absorbed light Ia
at wavelengths from 506 to 800 nm (step 0.5 nm). Ia at a given
wavelength is calculated by using Beer’s law [Equation (2)]:
[8] D. Dini, M. J. F. Calvete, M. Hanack, W. Z. Chen, W. Ji, Arki-
voc 2006, (iii), 77–96.
[9] D. Dini, M. Hanack, M. Meneghetti, J. Phys. Chem. B 2005,
109, 12691–12696.
[10] P. Zimcik, M. Miletin, M. Kostka, J. Schwarz, Z. Musil, K.
Kopecky, J. Photochem. Photobiol. A: Chem. 2004, 163, 21–28.
ˇ
[11] V. Novakova, P. Zimcik, M. Miletin, P. Vu˚jteˇch, S. Franzová,
Dyes Pigm. 2010, 87, 173–179.
[12] V. Novakova, P. Zimcik, M. Miletin, K. Kopecky, Z. Musil,
Eur. J. Org. Chem. 2010, 732–739.
(2)
where transmittance of the filter at a given wavelength is given by
I0, and absorbance of the dye at this wavelength is given by A.
[13] V. Novakova, P. Zimcik, M. Miletin, L. Vachova, K. Kopecky,
K. Lang, P. Chábera, T. Polívka, Phys. Chem. Chem. Phys.
2010, 12, 2555–2563.
[14] K. Kopecky, V. Novakova, M. Miletin, R. Kucˇera, P. Zimcik,
Bioconjugate Chem. 2010, 21, 1872–1879.
[15] P. Zimcik, M. Miletin, V. Novakova, K. Kopecky, M. Nejedla,
V. Stara, K. Sedlackova, Aust. J. Chem. 2009, 62, 425–433.
[16] K. Kopecky, P. Zimcik, V. Novakova, M. Miletin, Z. Musil, J.
Stribna, Dyes Pigm. 2008, 78, 231–238.
Fluorescence Quantum Yields Measurements: Fluorescence quan-
tum yields (ΦF) were determined by the comparative method using
ZnPc as a reference (ΦF = 0.30 in chloronaphthalene[37]) in a way
similar to that published previously. Thus, THF (2.5 mL) was
transferred into a quartz optical cell (10ϫ10 mm), and a defined
amount of a stock solution of the studied dye in THF was added
Eur. J. Org. Chem. 2011, 5879–5886
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5885

P. Zimcik et al.
FULL PAPER
[17] R. P. Linstead, A. R. Lowe, J. Chem. Soc. 1934, 1022–1027.
[18] G. T. Byrne, R. P. Linstead, A. R. Lowe, J. Chem. Soc. 1934,
1017–1022.
[19] E. H. Mørkved, H. Ossletten, H. Kjøsen, Acta Chem. Scand.
1999, 53, 1117–1121.
[20] R. Z. Uslu Kobak, E. S. Öztürk, A. Koca, A. Gül, Dyes Pigm.
2010, 86, 115–122.
[21] S. Makhseed, J. Samuel, F. Ibrahim, Tetrahedron 2008, 64,
8871–8877.
[32] S. Makhseed, F. Ibrahim, C. G. Bezzu, N. B. McKeown, Tetra-
hedron Lett. 2007, 48, 7358–7361.
[33] E. H. Mørkved, T. Andreassen, V. Novakova, P. Zimcik, Dyes
Pigm. 2009, 82, 276–285.
[34] Y. Rio, M. S. Rodriguez-Morgade, T. Torres, Org. Biomol.
Chem. 2008, 6, 1877–1894.
[35] Handbook of Porphyrin Science, vol. 9 (Eds.: K. M. Kadish,
K. M. Smith, R. Guilard), World Scientific Publishing Co. Pte.
Ltd., Singapore, 2010.
[22] E. H. Mørkved, T. Andreassen, P. Bruheim, Polyhedron 2009,
[36] L. Kaestner, M. Cesson, K. Kassab, T. Christensen, P. D. Ed-
minson, M. J. Cook, I. Chambrier, G. Jori, Photochem. Pho-
tobiol. Sci. 2003, 2, 660–667.
28, 2635–2640.
[23] A. Weitemeyer, H. Kliesch, D. Wöhrle, J. Org. Chem. 1995, 60,
4900–4904.
[37] P. G. Seybold, M. Gouterman, J. Mol. Spectrosc. 1969, 31, 1–
[24] T. E. Youssef, M. Hanack, J. Porphyrins Phthalocyanines 2005,
13.
9, 28–31.
[38] H. Ali, J. E. van Lier, Chem. Rev. 1999, 99, 2379–2450.
[39] K. Kameyama, M. Morisue, A. Satake, Y. Kobuke, Angew.
Chem. Int. Ed. 2005, 44, 4763–4766.
[40] V. Novakova, P. Zimcik, K. Kopecky, M. Miletin, J. Kunesˇ, K.
Lang, Eur. J. Org. Chem. 2008, 3260–3263.
[25] G. De La Torre, C. G. Claessens, T. Torres, Eur. J. Org. Chem.
2000, 2821–2830.
[26] T. Torres, J. Porphyrins Phthalocyanines 2000, 4, 325–330.
ˇ
[27] K. Kopecky, D. Satinský, V. Novakova, M. Miletin, A. Svo-
boda, P. Zimcik, Dyes Pigm. 2011, 91, 112–119.
[28] N. Kobayashi, R. Kondo, S. Nakajima, T. Osa, J. Am. Chem.
Soc. 1990, 112, 9640–9641.
[29] E. H. Mørkved, N. K. Afseth, H. Kjøsen, J. Porphyrins Phthalo-
cyanines 2006, 10, 1301–1308.
[30] S. Makhseed, J. Samuel, Dyes Pigm. 2009, 82, 1–5.
[31] F. Ibrahim, S. Makhseed, Arkivoc 2008, (x), 85–94.
[41] C. K. Jang, S. H. Byun, S. H. Kim, D. K. Lee, J. Y. Jaung, J.
Porphyrins Phthalocyanines 2009, 13, 794–797.
[42] Z. Musil, P. Zimcik, M. Miletin, K. Kopecky, M. Link, P. Pe-
trik, J. Schwarz, J. Porphyrins Phthalocyanines 2006, 10, 122–
131.
Received: May 17, 2011
Published Online: August 16, 2011
5886
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5879–5886