Atropisomeric chiral probes to study the supramolecular organization in porphyrin self-assemblies

Article abstract of DOI:10.1002/ejoc.201200985

A noncatalysed nucleophilic aromatic substitution was used to prepare the (bacterio)chlorophyll mimics 2-Zn, 3-Zn, 4-Zn and 5-Zn. The localisation of the two recognition groups on the same side of the porphyrin macrocycles in 2-Zn, 4-Zn and 5-Zn produces, through metal-ligand interactions, oligomers that display both H-and J-type aggregates. These supramolecular architectures were confirmed by the introduction of optically pure atropisomers in 5-Zn, which produces chiral aggregates. The UV/Vis and electronic circular dichroism spectra of all aggregates are used to propose an excitonic splitting mode that would correspond to this dual type of chromophoric self-assembly.

Full text of DOI:10.1002/ejoc.201200985

FULL PAPER
DOI: 10.1002/ejoc.201200985
Atropisomeric Chiral Probes to Study the Supramolecular Organization in
Porphyrin Self-Assemblies
Cyril Chappaz-Gillot,[a] Gabriel Canard,*[a] Federico Andreoli,[a] Nicolas Vanthuyne,[a]
Michel Giorgi,[b] Jean-Valère Naubron,[b] Valérie Monnier,[b] Roselyne Rosas,[b]
Christian Roussel,[a] and Teodor Silviu Balaban[a]
Keywords: Porphyrinoids / Self-assembly / Chirality / Circular dichroism / Aggregation
A noncatalysed nucleophilic aromatic substitution was used
to prepare the (bacterio)chlorophyll mimics 2-Zn, 3-Zn, 4-Zn
and 5-Zn. The localisation of the two recognition groups on
the same side of the porphyrin macrocycles in 2-Zn, 4-Zn and
5-Zn produces, through metal–ligand interactions, oligomers
that display both H- and J-type aggregates. These supra-
molecular architectures were confirmed by the introduction
of optically pure atropisomers in 5-Zn, which produces chiral
aggregates. The UV/Vis and electronic circular di-
chroism spectra of all aggregates are used to propose an ex-
citonic splitting mode that would correspond to this dual type
of chromophoric self-assembly.
these assemblies, one has to use and interpret complex spec-
troscopic features to ascribe a proposed model of organi-
sation. Consequently, there is still a need for complemen-
tary tools to determine the spatial arrangement of chromo-
phores and thereby explain the resulting electronic and
spectroscopic properties.
Introduction
Chromophoric self-assemblies are of particular interest
in the design of materials for excitation energy transfer and
electron-hole transport systems. The supramolecular organ-
isation of monomers allows the π-electronic systems to
overlap and interact to produce assemblies showing new
functions that are not observed in the corresponding mono-
mers. Usually the UV/Vis spectra of such aggregates feature
a regular arrangement of chromophores through excitonic
couplings that expand the resulting light absorption, which
is crucial for constructing efficient light-harvesting systems.
These properties are directly linked to the supramolecular
organisation of the chromophores, which controls their ori-
entation and orbital overlap. When no single-crystal X-ray
diffraction data are available to accurately characterise
Porphyrins are one of the most studied tectons that have
been incorporated into large assemblies, because their
chemical structure resembles that of (bacterio)chlorophylls,
which build well-defined superstructures in natural light-
harvesting systems such as light-harvesting complexes
(LH1, LH2) or chlorosomes.[1] Different types of organis-
ation can describe the porphyrin self-assemblies, which pro-
duce H- (face-to-face), J- (head-to-tail) or both H- and J-
(slipped cofacial) type aggregates.[2] When two porphyrins
interact, the first set of transition dipoles Bx and By pro-
duces excitonic couplings, resulting in a blueshift (H-di-
mer), redshift (J-dimer) or splitting (slipped cofacial dimer)
of the former degenerate monomer Soret band. Conse-
quently, the UV/Vis spectra of large assemblies containing
mixed aggregate types are generally difficult to interpret due
to the overlap of multiple absorption bands and due to the
combination of multiple excitonic couplings.
[a] Aix-Marseille Université, CNRS-UMR 7313 – Institut des
Sciences Moléculaires de Marseille (ISM2) – Chirosciences,
Faculté des Sciences de Saint-Jérôme Case A62,
Avenue Escadrille Normandie Niemen, 13397 Marseille, Cedex
20, France
Fax: +33-4-91289146
We have studied zinc porphyrin 1-Zn as a (bacterio)-
chlorophyll mimic that produces organised light-harvesting
E-mail: gabriel.canard@univ-amu.fr
Homepage: www.ism2.univ-cezanne.fr
[b] Aix-Marseille Université, FR 1739 - Fédération des Sciences
Chimiques de Marseille – Spectropole, Faculté des Sciences de
Saint-Jérôme Case A62,
self-assemblies (Scheme 1).[3] This compound bears
a
hydroxyethyl and a carbonyl group that coordinate the zinc
ion in a 5½ fashion that is responsible for the organisation
of the (bacterio)chlorophyll aggregates.[4] The localisation
of these two recognition groups on opposite sides of the
Avenue Escadrille Normandie Niemen, 13397 Marseille, Cedex
20, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200985.
6526
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6526–6536

Atropisomeric Chiral Probes
porphyrin macrocycle allows the supramolecular organisa- mimics (Scheme 2). For example, when an excess of propyl-
tion of well-defined J-aggregates in which each zinc ion of amine was added to the meso-bromoporphyrin 6, porphyrin
one porphyrin interacts with the hydroxyethyl group of an- 3-2H was produced in less than 24 hours in good yield. A
other porphyrin and the carbonyl group of a third. In the small volume of tetrahydrofuran was introduced as a cosol-
search for new kinds of supramolecular organisations that vent because compound 6 was not soluble in propylamine.
display original spectroscopic features, we designed zinc Porphyrin 3-2H bears only the NH functional group and its
porphyrin 2-Zn (Scheme 1). Compared with 1-Zn, 2-Zn has structure was ascertained by X-ray single-crystal diffraction
its two recognition groups, the nitrogen NH on the por- (Figure S1 in the Supporting Information). When propyl-
phyrin and the oxygen of an urea, located on the same side amine was replaced by ethylenediamine, the same procedure
of the porphyrin macrocycle.
produced 4-2H, which brings an additional amino NH2 re-
cognition group. This functional group offers numerous
possibilities of further functionalisations. The condensation
of phenyl isocyanate on 4-2H afforded porphyrin 2-2H
quantitatively, in which the two recognition groups are the
meso-NH on the porphyrin and the carbonyl group of a
urea. Furthermore, the NH moieties of the urea can be in-
volved in additional hydrogen-bonding interactions. The in-
troduction of a chiral element was achieved through re-
placement of the phenyl ring of the urea of 2-Zn by an
optically pure N-phenyl-substituted thiazolinethione atrop-
isomer (Scheme 2).[6] The racemic mixture was also prepared
to assess the impact of the optical purity on the supra-
molecular organisation. The isocyanate derivatives (rac)-8,
(aR)-8 and (aS)-8 were prepared in situ by reacting N-(2-
aminophenyl)-4-methylthiazoline-2-thiones (rac)-7, (aR)-7
and (aS)-7 with triphosgene. These isocyanates were added
to a solution of 4-2H and afforded quantitatively the free
base porphyrins (rac)-5-2H, (aR)-5-2H and (aS)-5-2H. Be-
cause of the strong absorbance of porphyrins in the UV/Vis
domain, it was not possible to record rotatory power on
conventional polarimeters. Consequently, the optical puri-
ties of (aS)-5-2H and (aR)-5-2H were analyzed by analytical
chiral HPLC after their synthesis to confirm that no racem-
isation had occurred. First, the racemic compound (rac)-5-
2H was subjected to chiral HPLC analysis to establish the
retention times of the two atropisomers. After screening of
a range of chiral stationary phases and solvents, a Chi-
ralpak IC column was chosen together with a mixture of
hexane and ethanol (80:20) as the solvent system. At 25 °C,
the two atropisomers were well-separated with retention
times of 16.9 and 21.9 min (Figure 1a). The same procedure
was applied to analyse the optical purities of samples of
(aR)-5-2H and (aR)-5-2H, which revealed no detectable ra-
cemisation (Figure 1, b and c). The absolute configurations
of 5-2H were those of optically pure 7 attributed pre-
viously.[6,7] Finally, all the free-base porphyrins were met-
alated by zinc acetate to generate the (bacterio)chlorophyll
mimics 2-Zn, 3-Zn, 4-Zn and 5-Zn. The optical purities of
the zinc complexes (aR)-5-Zn and (aS)-5-Zn could not be
determined because of their strong interactions with the
chiral phases. Because no racemisation occurred during the
preparation of optically pure (aR)-5-2H and (aS)-5-2H and
taking into account the high values of the racemisation bar-
riers measured for thiazolinethione atropisomers
Scheme 1. Chemical groups involved in supramolecular interac-
tions are printed in bold.
We describe here the synthesis and self-assemblies of four
simple (bacterio)chlorophyll mimics bearing one or two re-
cognition groups introduced on the same side of the por-
phyrin core. The incorporation of an atropisomer as an
original chiral element induces a supramolecular chirality
that allows the use of electronic circular dichroism (ECD)
spectra to deconvolute complex UV/Vis spectra of multi-
porphyrinic assemblies. The spectroscopic data were used to
propose a new supramolecular packing that displays both J-
and H-type aggregation.
Results and Discussion
Synthesis
We described recently the introduction of various amines (ca. 130 kJmol–1), we are confident that the conditions used
on the porphyrin core by an noncatalysed nucleophilic aro- for their metalation (zinc acetate, dichloromethane, meth-
matic substitution.[5] This procedure was here used to intro- anol, room temperature) are mild enough to avoid any iso-
duce the recognition groups of the (bacterio)chlorophyll merisation.[6]
Eur. J. Org. Chem. 2012, 6526–6536
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6527

G. Canard et al.
FULL PAPER
Scheme 2. (i) Propylamine, THF; (ii) Zn(OAc)2·2H2O, CH2Cl2, MeOH; (iii) ethylenediamine, THF; (iv) phenyl isocyanate, CH2Cl2; (v) tri-
phosgene, (iPr)2NEt, CH2Cl2; (vi) CH2Cl2.
Coordination to the Zinc Porphyrin by Various Model
Ligands
amine 10, N-methylaniline 11, N,NЈ-dimethylurea 12 and
thiazolinethione 13 and studied by UV/Vis spectroscopy
their ability to coordinate zinc porphyrin 9-Zn (Scheme 3).
The model ligand (ca. 10–15 equiv.) was dissolved in
The aggregation of zinc porphyrins 2-Zn, 3-Zn, 4-Zn and 5-
Zn in nonpolar solvents depends on various intermolecular 3 mL of a solution of zinc-porphyrin 9-Zn (1.4 mg) in
interactions that needed to be identified. For this purpose, dichloromethane (100 mL). The resulting solution was
we chose to analyse the interaction between the zinc met- transferred to a UV quartz cuvette with a 1 cm pathlength
allic centre of the simple zinc porphyrin 9-Zn and model and the UV/Vis spectrum was recorded from 400 to
ligands that mimic the functional groups borne by com- 800 nm. The maxima of the Soret and the Q bands and
pounds 2-Zn, 3-Zn, 4-Zn and 5-Zn. We selected propyl- their redshift values were compared to the maxima of the
6528
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6526–6536

Atropisomeric Chiral Probes
study shows that the aggregation of porphyrins 2-Zn, 3-Zn,
4-Zn and 5-Zn are based on the ligation of the zinc ion by
three different ligands: the meso-NH group of the por-
phyrin, the terminal amino group or the carbonyl of the
urea. The chiral thiazolinethione fragment introduced in 5-
Zn is inert and does not afford a supplementary zinc–ligand
interaction compared to 2-Zn.
Table 1. Maxima of the UV/Vis spectra of the complexes 9-Zn/L.
Ligand L
Ratio L/9-Zn UV/Vis, λmax (Δλ)[a] [nm]
Soret
Q
None
Propylamine 10
N-Methylaniline 11
N,NЈ-Dimethylurea 12 150
Thiazolinethione 13 150
0
15
12
409 (0)
417 (8)
417 (8)
414 (5)
409 (0)
538 (0), 571 (0)
551 (13), 588 (17)
547 (9), 584 (13)
544 (6), 581 (10)
538 (0), 571 (0)
[a] Δλ is the redshift observed (nm) during the coordination pro-
cess.
Self-Assembly of 2-Zn
2-Zn self-assembles in nonpolar solutions at very dilute
concentrations so that only the use of strongly coordinating
solvents such as dimethyl sulfoxide (DMSO) produces
sharp and interpretable signals in the corresponding 1H
NMR spectra. Figure 2 (a) shows the UV/Vis spectrum of
2-Zn aggregates recorded when a small volume (50 μL) of
a concentrated solution of 2-Zn in dichloromethane is di-
luted into larger volume of dry n-heptane (3 mL). The
monomer 2-Zn can be recovered by adding a few drops of
methanol, which causes the gradual disassembly of aggre-
gates due to its competition for metal ligation, forming 2-
Zn/MeOH adducts.
Figure 1. Chromatograms (UV/Vis detection) of the chiral HPLC
analysis of the two atropisomers of (a) (rac)-5-2H, (b) (aR)-5-2H
and (c) (aS)-5-2H.
The self-assembly process causes a strong broadening of
the Soret and Q bands. A small redshift is observed for the
Q bands (566Ǟ573 nm, 612Ǟ616 nm), whereas the Soret
band splits into three bands, one with a maximum at
429 nm and two flanking discernible shoulders, one blue-
shifted to 410 nm, and one redshifted to 450 nm. This result
clearly shows that self-assembly occurs between single 2-Zn
molecules through intermolecular interactions but does not
give any direct information on the supramolecular organi-
zation, which is also illustrated by the blueshifted quenched
fluorescence (Figure S2 in the Supporting Information).
Even if excitonic dipole–dipole theory could explain the
evolution of the UV/Vis spectrum and thereby be a strong
validation requirement of any proposed model,[8] other ex-
Scheme 3. Coordination of the model ligands 10, 11, 12 and 13 by
zinc porphyrin 9-Zn.
reference zinc porphyrin 9-Zn (Table 1). If no changes in perimental evidence is necessary to evaluate the intermo-
the wavelength of the maxima were observed, the L/9-Zn lecular organisation.
ratio was raised to reach 150. Propylamine 10 and N-meth-
We were not able to obtain suitably diffracting single
ylaniline 11 interact strongly with 9-Zn and produced sig- crystals of the 2-Zn aggregates, so we resorted to preparing
nificant redshifts of the Soret and Q bands. The coordina- several related zinc porphyrins to study the different key
tion of N,NЈ-dimethylurea 12 was much weaker and could roles of the functional groups in the construction of the
be visualised only by adding a large excess of the ligand; no assembly and their signatures in the UV/Vis spectra of the
interaction occurred between the thiazolinethione 13 and 9- aggregates. Consequently, the simpler complexes 3-Zn and
Zn despite the application of a very large L/9-Zn ratio. This 4-Zn were prepared (Scheme 2).
Eur. J. Org. Chem. 2012, 6526–6536
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6529

G. Canard et al.
FULL PAPER
dipole–dipole coupling occurring in these oligomers affects
the Bx and By transition dipoles producing an overall blue-
shift of the Soret band that is still relatively small. This
blueshift is also due to the metal–ligand interaction of the
3-Zn/MeOH adduct, which is replaced by a Zn/NH interac-
tion. This finding shows that the NH recognition group on
the porphyrin skeleton is able to interact with the zinc ion
of a second porphyrin to form H-dimers, which are proba-
bly responsible for the blueshift contribution of the split
Soret band observed in 2-Zn aggregates.
Scheme 4. Excitonic dipole–dipole coupling occurring within 2-Zn,
4-Zn or 5-Zn oligomers (forbidden transitions are not shown).
Self-Assembly of 4-Zn
We also prepared porphyrin 4-Zn, which brings an ad-
ditional amino NH2 recognition group. As mono-ortho-an-
iline-substituted zinc porphyrins,[9] 4-Zn spontaneously
forms at a micromolar scale J-dimers through two Zn–NH2
intermolecular interactions even in a strong coordinating
solvent such as DMSO. When the 1H NMR spectrum of 4-
Zn was recorded at a micromolar scale in [D6]DMSO, the
signals of the terminal methylene and amino groups CH2
and NH2 were observed to be strongly shielded (δ = –0.59
and –1.32 ppm, Figure 3). We suspected that this could be
the result of the strong dimerisation of 4-Zn, which places
a part of the alkyl chain above the porphyrin core. It should
be noted that the use of a less polar solvent produces very
broad signals due to aggregation of these dimers. Despite
Figure 2. (a) Solid trace: dilution of a dry, concentrated dichloro-
methane solution of 2-Zn in n-heptane. Dotted trace: same spec-
trum muliplied by 10. Dashed trace: the same solution after ad-
dition of methanol. (b) Solid trace: dilution of a dry, concentrated
dichloromethane solution of 3-Zn in n-heptane. Dashed trace: the
same solution after addition of methanol. (c) Solid trace: dilution
of a dry, concentrated dichloromethane solution of 4-Zn in n-hept-
ane. Dashed trace: the same solution after addition of methanol.
Self-Assembly of 3-Zn
Just as for 2-Zn, 3-Zn self-assembles in nonpolar solvents this strong evidence, one could think that this well-resolved
under very dilute conditions so that, again, only the use 1H NMR spectrum could be assigned to a more complex
of strongly coordinating solvents such as DMSO produced structure such as a cyclic trimer or tetramer. Therefore, a
sharp and interpretable signals in the corresponding 1H detailed mass spectrometric analysis was performed to
NMR spectra. 3-Zn forms H-type oligomers that are disso- study the formation of this dimer (Figure 4). The ESI-MS
ciated by the addition of an excess of a strong coordinating spectrum of 4-Zn recorded in a solution of sodium chloride
ligand such as MeOH or DMSO. The introduction of a in methanol displays the most abundant ion of the proton-
concentrated dichloromethane solution of 3-Zn into n-hept- ated monomer (m/z 659) and its adduct with sodium (m/z
ane produced aggregates that were revealed by a 6–10 nm 681). The two corresponding expected adducts of the dimer
blueshift of the Soret band (430Ǟ424 nm) and the Q bands (m/z 1321 and 1343) can also be seen, revealing the remark-
(565Ǟ555 nm, 612Ǟ601 nm) due to the formation of H- able stability of the dimer even during the ionisation pro-
dimers or H-trimers built on Zn–NH intermolecular inter- cess. With almost the same intensities, ions corresponding
actions (Figure 2, b). As depicted in Scheme 4, the excitonic to the trimer were detected (m/z 1981 and 2003). This trimer
6530
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6526–6536

Atropisomeric Chiral Probes
1
Figure 3. H NMR (400 MHz) spectrum of 4-Zn in [D6]DMSO. Asterisks denote residual solvent peaks.
Figure 4. (a) Positive ESI-MS spectrum of 4-Zn recorded in a solution of sodium chloride in methanol [(4-Zn) is the monomer]. (b) Expan-
sion of the region in which multiple oligomers are detected.
probably results from the self-assembly of one monomer to Na]+; found 1343.3518). DOSY NMR experiments con-
one dimer in the gas phase. The strong self-assembly of 4- firmed the formation of this dimer (see the Supporting In-
Zn is also illustrated by less intense ions corresponding to formation).
higher oligomers. A high-resolution mass spectrometry
The excitonic dipole–dipole coupling occurring in this J-
analysis was performed on the isotopic maximum of the dimer and the strong Zn–NH2 interactions produce a small
sodium-dimer adduct and confirmed its elementary compo- redshift of all bands compared to the spectrum of the
sition (calcd. for C80H60N12Zn2Na+ 1343.3521 [2(4-Zn) + monomer 3-Zn (Figure 2, b and c). The addition of n-hept-
Eur. J. Org. Chem. 2012, 6526–6536
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6531

G. Canard et al.
FULL PAPER
ane causes a strong broadening and splitting of the Soret resolve the 2-Zn supramolecular organization by inducing
band with a maximum at 428 nm and two discernible shoul- chirality to the aggregates. For this purpose, we prepared
ders at 415 and 443 nm, whereas the Q bands are slightly (rac)-5-Zn, (aR)-5-Zn and (aS)-5-Zn in which the phenyl
blueshifted (575Ǟ572 nm, 622Ǟ619 nm) (Figure 2, c). A ring of the urea of 2-Zn is replaced either by a racemic
similar change of the absorption spectrum was observed or an optically pure N-phenyl-substituted-thiazolinethione
during the aggregation of 2-Zn and reveals a supramolec- atropisomer (Scheme 2).[6] A UV/Vis survey of all possible
ular organisation that can only be explained by a new Zn– intermolecular interactions between 2-Zn, 3-Zn, 4-Zn or 5-
NH inter-dimer interaction producing new excitonic cou- Zn monomers showed that zinc ligation by a N-substituted-
plings. In the solid state, the organisation of the chromo- thiazolinethione is impossible, revealing the inertness of this
phore can then be described as a H-aggregate of J-dimers supplementary introduced fragment.
or as a J-aggregate of H-dimers (Scheme 5). When two H-
The self-assembly of 5-Zn produces spectral features that
dimers interact, the angles between the two transition di- are similar to those observed for 2-Zn. A small redshift is
poles BXBX or BYBY are probably small and the splitting observed for the Q bands (561Ǟ572 nm, 613Ǟ618 nm),
energies are weak. The splitting of the Soret band is then whereas the broad Soret band appears split with a maxi-
small and results from four permitted electronic transitions mum at 424 nm and two discernible shoulders at 410 and
ABYZ of similar energies (Scheme 4). An identical splitting 445 nm (Figure 5, a). Another excitonic coupling affects the
mode was used to describe hexameric macrorings of absorption of the thiazolinethione fragment, which is
slipped-cofacial dimers.[10] The same model can be applied slightly modified with the appearance of a shoulder at
for the self-assembly of 2-Zn that is based on Zn–NH and 270 nm, although the maximum is still located at 320 nm.
Zn–O=C intermolecular interactions (Scheme 5). When 2- As for 2-Zn, the aggregation of 5-Zn is illustrated by a blue-
Zn or 4-Zn are dissolved in a nonpolar solvent, the strong- shifted quenched fluorescence (Figure S3 in the Supporting
est metal–ligand interaction Zn–NH or Zn–NH2 produces Information). The identical behaviour of 2-Zn and 5-Zn
H- or J-dimers that assemble into higher oligomers due to during aggregation makes the latter a good chiral model of
secondary and weaker metal ligation Zn–O=C or Zn–NH. the former. It should be noted that the optical purity of 5-
To confirm these hypotheses and the corresponding split- Zn has no impact on the global supramolecular organi-
ting mode, we studied the aggregation of 5-Zn.
sation because the self-assemblies of (rac)-5-Zn, (aR)-5-Zn
and (aS)-5-Zn produce the same results.
Scheme 5. Self-assemblies of zinc porphyrins 2-Zn or 5-Zn (meso-
phenyl groups are omitted for clarity). The same organisation oc-
curs during the aggregation of 4-Zn in which the Zn–O=C intermo-
lecular interactions are replaced by strong Zn–NH2 interactions.
Figure 5. (a) UV/Vis spectra. Solid trace: dilution of a dry, concen-
trated dichloromethane solution of (aR)-5-Zn in n-heptane multi-
plied by 10. Dashed red trace: the same solution after addition of
methanol. (b) ECD spectra. Red trace: dilution of a dry, concen-
trated dichloromethane solution of (aS)-5-Zn in n-heptane. Blue
trace: dilution of a dry, concentrated dichloromethane solution of
Self-Assembly of 5-Zn
ECD is a well-known technique that is used to describe
the interactions of porphyrins or chromophores through
space[11] and has been successfully applied to study numer-
ous chiral assemblies.[12] This technique helped us to fully (aR)-5-Zn in n-heptane.
6532
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6526–6536

Atropisomeric Chiral Probes
Monitoring of the aggregation of (aR)-5-Zn or (aS)-5- The trisignate observed in the ECD spectra during the self-
Zn by ECD spectroscopy revealed intense trisignate Cotton assembly of optically pure (aR)-5-Zn confirms this hypothe-
effects in the Soret band domain (Figure 5, b). This strong sis and results from the mutual addition of the iso-energetic
signature can only be attributed to the chiral packing of transitions B and Y that correspond to clockwise inter-di-
optically pure 5-Zn because it disappears when the ECD mer couplings of dipoles BXBX and BYBY (Scheme 4). The
spectra of (aR)-5-Zn and (aS)-5-Zn monomers are recorded equal energies of transitions B and Y are illustrated by the
(Figure S4 in the Supporting Information). When the corre- intensity of the central maximum, which is roughly the sum
sponding free bases (rac)-5-2H, (aR)-5-2H and (aS)-5-2H of the intensities of the two other maxima of opposite sign.
were subjected to the same assembling procedure, the re-
sulting UV/Vis spectra were only strongly broadened due
to the precipitation of the monomers; the ECD spectra did
Conclusions
The noncatalysed nucleophilic aromatic substitution is a
powerful tool that can be used to introduce functional
groups on the porphyrin moiety.[5,16] We have used this pro-
cedure to prepare (bacterio)chlorophyll mimics in which the
two recognition groups of the zinc porphyrin are placed on
the same side of the porphyrin macrocycle. The stable and
favoured 5½ coordination of the zinc ion produces a novel
packing of the chromophores. The stronger metal–ligand
interaction produces H- or J-dimers that are assembled,
thanks to the weaker metal–ligand bond, into J- or H-type
higher oligomers. A similar model was used to describe the
solid H-packing of chiral bisporphyrins leading to tetra-
mers with complex ECD spectra.[11b] The aggregation of
the dimers is due to weak secondary interactions that are
very labile and less probable. Therefore, the largest compo-
nent of the species in solution seems to be tetramers, which
can be formed with only a single weak interdimer interac-
tion. The chirality of chromophore aggregates can be in-
duced by, or transferred through, strong intermolecular in-
teractions such as a chiral metal–ligation, hydrogen bonds
or electrostatic interactions.[17] We report here how an op-
tically pure atropisomer induces, through weak secondary
supramolecular interactions, a helicity and is used as a chi-
ral probe to reveal the complex spectroscopic features of a
novel supramolecular organisation. This spatial arrange-
ment causes a concomitant blueshift and redshift of the
UV/Vis absorption, which is crucial for constructing ef-
ficient light-harvesting systems. We are now working on the
preparation of derivatives bearing two identical recognition
groups to produce larger assemblies that should display
both J- and H-type architectures.
not reveal any chiral signature in the Soret band domain
(Figure S5 in the Supporting Information). These results
confirm that the metal ligation process is crucial for supra-
molecular organisation of 5-Zn and that the induction
source of the observed supramolecular chirality is brought
about by the chiral thiazolinethione fragment. If the chiral
element is not directly linked to the metal centre, the helic-
ity of 5-Zn aggregates is probably driven by a steric repul-
sion between the thiazoline fragment of one monomer and
the meso-phenyl substituent of the linked monomer produc-
ing a chiral J-dimer (Scheme 5). Another explanation would
be the participation of the chiral thiazoline fragments,
which have a strong dipolar moment of ca. 7 D[13] in sec-
ondary supramolecular interactions such as π-π interac-
tions and H-bonding through the NH moieties of the urea,
which would stabilise the overall packing of the chromo-
phores. Theoretical calculations could shed light on these
interactions and explain the translation of chirality from
the molecular to the supramolecular level.
The ECD spectrum of (aS)-5-Zn aggregates displays a
positive band at 443 nm, a strong negative band at 424 nm
and a third positive band at 408 nm, whereas the (aR)-5-Zn
assembly gives, as expected, the mirror-image induced ECD
spectrum. The ECD signals of the thiazolinethione are af-
fected by the packing of optically pure 5-Zn, which trans-
forms the same sign signals of the monomer into an alter-
nate positive–negative bisignate ECD band that probably
results from excitonic couplings between thiazolinethione
fragments. The trisignate profile of the ECD spectra of 5-
Zn aggregates coincides exactly with the Soret maxima ob-
served in the UV/Vis spectrum, which are fully assigned
thanks to the alternating signs of the ECD bands. We chose
to use the so-called “circular oscillator model” approach to
describe the shapes of the ECD spectra. In this model, both
Soret components (Bx and By) are equally weighted to pre-
dict Soret exciton chirality.[14] The interaction of two 5-Zn
monomers through Zn–NH intermolecular interactions
produces two effective transition dipoles BXBX and BYBY
(Scheme 4). When two H-dimers are assembled in a chiral
oblique J-tetramer, the transition dipoles BXBX and BYBY
interact and afford four permitted UV/Vis transitions A, B,
Y and Z. According to the exciton chirality method, the
ECD spectrum would then display two consecutive couplets
of opposite chirality.[15] The negative twist between the in-
teracting effective transition dipoles will produce a negative
exciton couplet in the low energy domain whereas a positive
Experimental Section
General Remarks: All reagents were used as received. Tetra-
hydrofuran (THF) was distilled from sodium and benzophenone.
Dichloromethane (CH2Cl2), n-hexane and n-heptane were distilled
over calcium hydride. Flash column chromatography was per-
formed on silica gel 60 (230–400 mesh).
1H NMR spectra were recorded with a Bruker Avance 300, a
Bruker Avance 200 or a Bruker Avance 400 Ultrashield NMR spec-
trometer. Chemical shifts are given in ppm relative to residual
peaks of chloroform (δ = 7.26 ppm) or dimethyl sulfoxide (δ =
2.50 ppm). Diffusion experiments were recorded with a Bruker Av-
ance 400 Ultrashield NMR spectrometer. The sequence corre-
sponds to Bruker pulse program ledbpgp2s using stimulated echo,
bipolar gradients and longitudinal eddy current delay as z filter.
couplet will be observed in the high-energy spectral region. The four 2 ms gradients pulses have sine-bell shapes and ampli-
Eur. J. Org. Chem. 2012, 6526–6536
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6533

G. Canard et al.
FULL PAPER
tudes ranging linearly from 2 to 33 G m–1 cm–1 in 16 steps. The dif-
fusion delay was 250 ms and the number of scans 32 (P1 = 9 μs,
PL1 = 0 db). The processing was done using a line broadening of
1 Hz and the diffusion rates Dx calculated using the Bruker pro-
cessing package, by analysing the decay of the natural logarithm of
the normalised signal intensity I/I0 of selected NMR signals as a
function of the square of the pulse gradient strength according to
Equation (1) [I is the observed intensity, I0 is the intensity without
gradients, γ is the gyromagnetic ratio of the observed nucleus, δ is
the length of the gradient pulse, G is the gradient strength, D (dif-
fusion delay) is the delay between the midpoints of the gradients].
at 0 °C. N-Bromosuccinimide (348 mg, 1.96 mmol, 1.4 equiv.) was
added and the disappearance of the starting porphyrin was moni-
tored by TLC. After 20 min, acetone (15 mL) was added and the
mixture was washed with water (3ϫ 300 mL), dried with Na2SO4
and the solvents evaporated. After recrystallisation from CH2Cl2/
n-heptane, porphyrin 6 (856 mg, 1.38 mmol, 99%) was obtained as
a dark-violet solid. Rf = 0.41 (CH2Cl2/n-hexane, 2:3). 1H NMR
3
(200 MHz, CDCl3, 25 °C): δ = 9.66 [d, JH,H = 5.1 Hz, 2 H, H-β],
3
8.89 [d, JH,H = 5.1 Hz, 2 H, H-β], 8.80 (s, 4 H, H-β), 8.15–8.21
(m, 6 H, Ar-H), 7.72–7.79 (m, 9 H, Ar-H), –2.75 (s, 2 H, NH) ppm.
UV/Vis (CHCl3): λmax (logεmax) = 420 (5.59), 518 (4.26), 553 (3.97),
595 (3.73), 651 nm (3.64).
ln(I/I0) = –(γδ)2(Δ – δ/3)DG2
(1)
5-Propylamino-10,15,20-triphenylporphyrin
(3-2H):
5-Bromo-
The NMR tubes were prepared by dissolving the selected com-
pound (17 μmol, ca. 10 mg) in DMSO (0.6 mL). The sample tem-
perature was controlled and stabilised at 30 °C.
10,15,20-triphenylporphyrin (6) (180 mg, 291 μmol) and propyl-
amine (4.8 mL, 58 mmol, 200 equiv.) were dissolved in THF
(3 mL). The mixture was heated to reflux for 20 h. After evapora-
tion, the resulting solid was purified by chromatography on silica
gel (CH2Cl2/n-hexane, 7:3). After evaporation, porphyrin 3-2H
(154 mg, 259 μmol, 89%) was obtained as a dark-violet solid. Rf =
UV/Vis spectra were measured with a Shimadzu UV-2401 (PC) in-
strument. Fluorescence emission spectra were recorded with a Var-
ian Cary Eclipse spectrofluorimeter. ECD spectra were measured
with a JASCO J-815 spectrometer equipped with a JASCO Peltier
cell holder PTC-423 to maintain the temperature at 20.0Ϯ0.2 °C.
Hellma quartz cells of 1 cm optical pathlength were used. Solutions
were prepared by dilution of solid samples in CH2Cl2 (HPLC
grade). The ECD spectrometer was purged with nitrogen before
recording each spectrum, which was baseline subtracted. The base-
line was always measured for the same solvent and in the same cell
as the samples. The spectra are presented without smoothing or
further data processing. High-resolution mass spectrometry
(HRMS-ESI) analyses were performed with a QStar Elite (Applied
Biosystems SCIEX) spectrometer or with a SYNAPT G2 HDMS
(Waters) spectrometer. These two instruments were equipped with
an electrospray ionisation source and were used in the positive ion
mode and with a capillary voltage set at +5500 V. The preliminary
1
0.27 (CH2Cl2/n-hexane, 7:3). H NMR (200 MHz, CDCl3, 25 °C):
3
3
δ = 9.16 [d, JH,H = 4.7 Hz, 2 H, H-β], 8.62 [d, JH,H = 4.8 Hz, 2
H, H-β], 8.58 [d, 3JH,H = 4.5 Hz, 2 H, H-β], 8.57 [d, 3JH,H = 4.7 Hz,
2 H, H-β], 8.15–8.17 (m, 4 H, Ar-H), 8.12–8.13 (m, 2 H, Ar-H),
3
7.70–7.76 (m, 9 H, Ar-H), 6.12 (s, 1 H, NHCH2), 4.36 [t, JH,H
=
3
6.2 Hz, 2 H, CH2NH], 2.06 [sext, JH,H = 7.3 Hz, 2 H, CH2CH3],
1.18 [t, 3JH,H = 7.4 Hz, 3 H, CH3], –1.40 (s, 2 H, NHpyr) ppm. UV/
Vis (CHCl3): λmax (logεmax) = 427 (5.52), 531 (3.89), 583 (4.17),
+
681 nm (4.01). HRMS (ESI): calcd. for C41H34N5 596.2809 [M +
H]+; found 596.2810. Single crystals of porphyrin 3-2H were grown
by slow diffusion of n-heptane into a concentrated solution of por-
phyrin 3-2H in dichloromethane.
5-(2-Aminoethylamino)-10,15,20-triphenylporphyrin (4-2H): 5-
screening of chiral stationary phases was performed with a Lach- Bromo-10,15,20-triphenylporphyrin (6) (315 mg, 0.51 mmol) and
rom-Elite unit consisting of an L-2130 pump, an L-2200 autosam-
pler, an L-2350 oven and an L-2455 DAD detector. The analytical
columns (250ϫ4.6 mm) tested were Chiralpak IA, IB and IC col-
umns from Chiral Technology Europa (Illkirch, France), Whelk-
O1 (S,S) and Ulmo (S,S) from Regis Technologies (Morton Grove,
USA). n-Hexane and ethanol, HPLC grade, were degassed and fil-
tered through a 0.45 μm millipore membrane before use. Retention
times (Rt) in minutes, retention factors ki = (Rti – Rt0)/Rt0 and
enantioselectivity factor α = k2/k1 are given. Rt0 was determined
by injection of tri-tert-butylbenzene.
ethylenediamine (6.8 mL, 102 mmol, 200 equiv.) were dissolved in
THF (6 mL). The mixture was heated to reflux for 24 h. After evap-
oration, the crude residue was dissolved in CH2Cl2 (80 mL),
washed with distilled water (5ϫ 100 mL) and dried with MgSO4.
The resulting solid was purified by chromatography on silica gel
(CH2Cl2/MeOH/Et3N, 98.5:0.5:1). After evaporation, porphyrin 4-
2H (260 mg, 0.433 mmol, 85%) was obtained as a dark-violet solid.
Rf = 0.24 (CH2Cl2/MeOH, 9:1). 1H NMR (200 MHz, CDCl3,
3
3
25 °C): δ = 9.27 [d, JH,H = 4.6 Hz, 2 H, H-β], 8.62 [d, JH,H
=
4.8 Hz, 2 H, H-β], 8.58 [d, 3JH,H = 4.8 Hz, 2 H, H-β], 8.56 [d, 3JH,H
= 4.8 Hz, 2 H, H-β], 8.10–8.19 (m, 6 H, Ar-H), 7.70–7.75 (m, 9 H,
CCDC-876244 (for 3-2H) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
3
3
Ar-H), 4.38 [t, JH,H = 5.6 Hz, 2 H, CH2NH], 3.23 [t, JH,H
=
5.6 Hz, 2 H, CH2NH2], –1.42 (s, 2 H, NHpyrrole) ppm. UV/Vis
(CHCl3): λmax (logεmax) = 426 (5.48), 531 (3.89), 581 (4.13), 679 nm
(3.97). HRMS (ESI): calcd. for C40H33N6 597.2761 [M + H]+;
+
The N-(2-aminophenyl)-4-methylthiazoline-2-thiones (rac)-7, (aR)-
7, (aS)-7 and thiazolinethione 13 were prepared and purified ac-
cording to literature procedures.[6,7]
found 597.2774.
(5-Propylamino-10,15,20-triphenylporphyrinato)zinc (3-Zn): 5-Prop-
ylamino-10,15,20-triphenylporphyrin (3-2H) (100 mg, 168 μmol,
1 equiv.) and zinc acetate dihydrate (55 mg, 252 μmol, 1.5 equiv.)
were dissolved in a mixture of CH2Cl2/MeOH (12 mL, 4:1). The
mixture was stirred at room temperature for 3 h. The mixture was
washed successively with saturated aqueous NaHCO3 (30 mL),
water (2ϫ 30 mL) and dried with MgSO4. After evaporation, the
resulting solid was purified by chromatography on silica gel
(CH2Cl2/MeOH, 99.5:0.5). After recrystallisation from CH2Cl2/n-
heptane, 3-Zn (110 mg, 166 μmol, 99%) was obtained as a dark-
violet solid. Rf = 0.63 (CH2Cl2/MeOH, 99.5:0.5). 1H NMR
(5,15-Diphenylporphyrinato)zinc (9-Zn): Prepared by metalation of
5,15-diphenylporphyrin[18] according to a literature procedure.[19]
1
Rf = 0.68 (MeOH/CH2Cl2, 0.5:99.5). H NMR (200 MHz, CDCl3,
3
25 °C): δ = 10.21 (s, 2 H, H-meso), 9.35 [d, JH,H = 4.5 Hz, 4 H,
3
H-β], 9.05 [d, JH,H = 4.5 Hz, 4 H, H-β], 8.21–8.25 (m, 4 H, Ar-
H), 7.73–7.76 (m, 4 H, Ar-H) ppm. UV/Vis (CHCl3): λmax (logεmax
)
= 413 (5.62), 541 nm (4.22).
5-Bromo-10,15,20-triphenylporphyrin (6): This compound was pre-
pared according to a literature procedure.[20] 5,10,15-Triphenylpor-
phyrine[21] (753 mg, 1.4 mmol, 1 equiv.) and pyridine (450 μL,
5.6 mmol, 4 equiv.) were dissolved in CHCl3 (300 mL) under argon
3
(400 MHz, [D6]DMSO, 25 °C): δ = 9.40 [d, JH,H = 4.8 Hz, 2 H,
3
3
H-β], 8.45 [d, JH,H = 4.4 Hz, 2 H, H-β], 8.38 [d, JH,H = 4.4 Hz,
6534
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6526–6536

Atropisomeric Chiral Probes
4 H, H-β], 8.06–8.09 (m, 4 H, Ar-H), 8.02–8.04 (m, 2 H, Ar-H), (aS)-5-(2-{2-[4-Methyl-2-thioxothiazol-3(2H)-yl]phenylureido}ethyl-
7.74–7.76 (m, 6 H, Ar-H), 7.70–7.72 (m, 3 H, Ar-H), 7.58 [t, 3JH,H
amino)-10,15,20-triphenylporphyrin [(aS)-5-2H]: Prepared as for
porphyrin (rac)-5-2H starting from triphosgene (31 mg, 107 μmol,
3
= 7.2 Hz, 1 H, NHCH2], 4.30 [q, JH,H = 7.2 Hz, 2 H, CH2NH],
3
3
2.01 [sext, JH,H = 7.2 Hz, 2 H, CH2CH3], 1.02 [t, JH,H = 7.2 Hz, 0.53 equiv.), pure (aS)-N-(2-aminophenyl)-4-methylthiazoline-2-
3 H, CH3] ppm. UV/Vis (CHCl3): λmax (logεmax) = 431 (5.54), 564 thione (aS)-7 (71 mg, 320 μmol, 1.6 equiv.), diisopropylethylamine
(4.00), 611 nm (4.09). HRMS (ESI): calcd. for C41H32N5Zn+ (114 μL, 641 μmol, 3.2 equiv.) and porphyrin 4-2H (120 mg,
658.1944 [M + H]+; found 658.1951.
201 μmol, 1 equiv.). Porphyrin (aS)-5-2H (143 mg, 169 μmol, 84%)
was obtained as a dark-violet solid. Chiral HPLC (Chiralpak IC;
n-hexane/ethanol, 4:1; 1 mL/min; UV 420 nm): Rt = 21.95 min.
[5-(2-Aminoethylamino)-10,15,20-triphenylporphyrinato]zinc (4-Zn):
5-(2-Aminoethylamino)-10,15,20-triphenylporphyrin
(4-2H)
(80 mg, 134 μmol, 1 equiv.) and zinc acetate dihydrate (44 mg,
201 μmol, 1.5 equiv.) were dissolved in a mixture of CH2Cl2/MeOH
(25 mL, 4:1). The mixture was stirred at room temperature for 3 h.
The mixture was washed successively with saturated aqueous
NaHCO3 (40 mL), water (2ϫ 40 mL) and dried with MgSO4. After
evaporation, the resulting solid was purified by chromatography on
silica gel (CH2Cl2/MeOH/Et3N, 98.5:0.5:1, then CH2Cl2/MeOH,
99.5:0.5). After recrystallisation from CH2Cl2/n-heptane, zinc por-
phyrin 4-Zn (75 mg, 114 μmol, 85%) was obtained as a dark-violet
solid. Rf = 0.12 (CH2Cl2/MeOH, 99.5:0.5). 1H NMR (400 MHz,
[D6]DMSO, 25 °C): δ = 8.62 [d, 3JH,H = 4.1 Hz, 2 H, H-β], 8.54 [d,
3JH,H = 4.6 Hz, 2 H, H-β], 8.48 [d, 3JH,H = 4.6 Hz, 2 H, H-β], 8.23
(aR)-5-(2-{2-[4-Methyl-2-thioxothiazol-3-(2H)-yl]phenylureido}-
ethylamino)-10,15,20-triphenylporphyrin [(aR)-5-2H]: Prepared as
for the porphyrin (rac)-5-2H starting from triphosgene (31 mg,
107 μmol, 0.53 equiv.), pure (aR)-N-(2-aminophenyl)-4-methylthi-
azoline-2-thione (aR)-7 (71 mg, 320 μmol, 1.6 equiv.), diisopropyl-
ethylamine (114 μL, 641 μmol, 3.2 equiv.) and porphyrin 4-2H
(120 mg, 201 μmol, 1 equiv.). Porphyrin (aR)-5-2H (134 mg,
159 μmol, 79%) was obtained as a dark-violet solid. Chiral HPLC
(Chiralpak IC; n-hexane/ethanol, 4:1; 1 mL/min; UV 420 nm): Rt
= 16.83 min.
5-[2-(Phenylureido)ethylamino]-10,15,20-triphenylporphyrin (2-2H):
5-(2-Aminoethylamino)-10,15,20-triphenylporphyrin (4-2H)
(100 mg, 0.168 mmol) and phenyl isocyanate (24 μL, 0.218 mmol,
1.3 equiv.) were dissolved in freshly distilled CH2Cl2 (10 mL). The
mixture was stirred at room temperature for 30 min. After evapora-
tion, the resulting solid was purified by chromatography on silica
gel (CH2Cl2/AcOEt/Et3N, 80:19:1). After recrystallisation from
CH2Cl2/n-heptane, porphyrin 2-2H (104 mg, 0.145 mmol, 86 %)
was obtained as a dark-violet solid. Rf = 0.09 (CH2Cl2/AcOEt/
3
[d, JH,H = 4.4 Hz, 2 H, H-β], 8.15–8.17 (m, 4 H, Ar-H), 8.10–8.12
(m, 2 H, Ar-H), 7.77–7.78 (m, 6 H, Ar-H), 7.74–7.75 (m, 3 H, Ar-
H), 6.38 (s, 1 H, NHCH2), 2.91 (s, 2 H, CH2NH), –0.59 (s, 2 H,
CH2NH2), –1.32 (s, 2 H, NH2) ppm. UV/Vis (CHCl3): λmax
(logεmax) = 432 (5.45), 575 (3.98), 622 nm (4.16). HRMS (ESI):
calcd. for C40H31N6Zn+ 659.1896 [A + H]+; found 659.1902; calcd.
for C80H60N12Zn2Na+ 1343.3521 (Isotopic maximum) [2A + Na]+;
found 1343.3518 (isotopic maximum).
1
Et3N, 80:19:1). H NMR (400 MHz, [D6]DMSO, 25 °C): δ = 9.36
5-(2-{2-[4-Methyl-2-thioxothiazol-3(2H)-yl]phenylureido}ethyl-
amino)-10,15,20-triphenylporphyrin [(rac)-5-2H]: Triphosgene
(61 mg, 208 μmol, 0.53 equiv.) was dissolved in freshly distilled
CH2Cl2 (7 mL) under argon at 0 °C before a solution of (rac)-N-
(2-aminophenyl)-4-methylthiazoline-2-thione (rac)-7 (138 mg,
623 μmol, 1.6 equiv.) and diisopropylethylamine (222 μL,
1247 μmol, 3.2 equiv.) in freshly distilled CH2Cl2 (8 mL) was
added. The mixture was stirred at room temperature for 90 min,
and the formation of isocyanate (rac)-8 was monitored by TLC [Rf
= 0.80 (CH2Cl2/AcOEt, 4:1)]. A solution of 5-(2-aminoethyl-
amino)-10,15,20-triphenylporphyrin (4-2H) (233 mg, 391 μmol,
1 equiv.) in freshly distilled CH2Cl2 (9 mL) was added and the mix-
ture was stirred at room temperature for 30 min. The mixture was
washed successively with saturated aqueous NaHCO3 (20 mL),
water (20 mL) and dried with MgSO4. After evaporation, the re-
sulting solid was purified by chromatography on silica gel (CH2Cl2/
AcOEt/Et3N, 80:19:1). After recrystallisation from CH2Cl2/n-hept-
ane, porphyrin (rac)-5-2H (287 mg, 340 μmol, 87%) was obtained
as a dark-violet solid. Rf = 0.14 (CH2Cl2/AcOEt/Et3N, 80:19:1). 1H
3
[d, JH,H = 4.7 Hz, 2 H, H-β], 8.61 (s, 1 H, Ar-NH-CO), 8.59 [t,
3
3JH,H = 5.4 Hz, 1 H, CH2NH-CO], 8.36 [d, JH,H = 4.9 Hz, 2 H,
3
3
H-β], 8.28 [d, JH,H = 4.9 Hz, 2 H, H-β], 8.23 [d, JH,H = 4.7 Hz,
2 H, H-β], 8.04–8.06 (m, 4 H, Ar-H), 8.00–8.03 (m, 2 H, Ar-H),
3
7.75–7.78 (m, 6 H, Ar-H), 7.71–7.74 (m, 3 H, Ar-H), 7.35 [t, JH,H
= 7.6 Hz, 2 H, Ar-H], 7.18 [t, JH,H = 7.9 Hz, 2 H, Ar-H], 6.88 [d,
3
3JH,H = 7.3 Hz, 1 H, Ar-H], 6.59 [t, 3JH,H = 5.8 Hz, 1 H, CH2NH],
4.25 [q, 3JH,H = 5.7 Hz, 2 H, CH2-NHCO], 3.57 [q, 3JH,H = 5.8 Hz,
2 H, CH2-NH], –0.38 (s, 2 H, NHpyrrole) ppm. UV/Vis (CHCl3):
λmax (log εmax) = 428 (5.56), 587 (4.22), 682 nm (4.17). HRMS
(ESI): calcd. for C47H38N7O+ 716.3132 [M + H]+; found 716.3139.
{5-[2-(Phenylureido)ethylamino]-10,15,20-triphenylporphyrinato}-
zinc (2-Zn): 5-[2-(Phenylureido)ethylamino]-10,15,20-triphenylpor-
phyrin (2-2H) (90 mg, 126 μmol, 1 equiv.) and zinc acetate dihy-
drate (30 mg, 138 μmol, 1.1 equiv.) were dissolved in a mixture of
CH2Cl2/MeOH (30 mL, 4:1). The mixture was stirred at room tem-
perature for 3 h then washed successively with saturated aqueous
NaHCO3 (50 mL), water (2ϫ 30 mL) and dried with MgSO4. After
evaporation, the resulting solid was purified by chromatography on
3
NMR (400 MHz, CDCl3, 25 °C): δ = 9.00 [d, JH,H = 4.4 Hz, 2 H,
3
3
H-β], 8.52 [d, JH,H = 4.8 Hz, 2 H, H-β], 8.45 [d, JH,H = 4.8 Hz, silica gel (CH2Cl2/MeOH, 98:2). After recrystallisation from
3
2 H, H-β], 8.38 [d, JH,H = 4.4 Hz, 2 H, H-β], 8.07–8.08 (m, 6 H, CH2Cl2/n-heptane, zinc porphyrin 2-Zn (75 mg, 97 μmol, 77%) was
Ar-H), 7.92 [d, JH,H = 8.0 Hz, 1 H, Ar-H], 7.67–7.71 (m, 10 H,
3
obtained as a dark-green solid. Rf = 0.34 (CH2Cl2/MeOH, 9:1). 1H
3
3
3
Ar-H, Ar-NH), 7.36 [t, JH,H = 7.6 Hz, 1 H, Ar-H], 7.11 [t, JH,H
NMR (400 MHz, [D6]DMSO, 25 °C): δ = 9.46 [d, JH,H = 4.6 Hz,
= 7.6 Hz, 1 H, Ar-H], 6.88 (br. s, 1 H, CH2NH-CO), 6.80 [d, 3JH,H
2 H, H-β], 8.63 (s, 1 H, Ar-NH-CO), 8.45 [d, JH,H = 4.6 Hz, 2 H,
3
3
3
3
= 7.6 Hz, 1 H, Ar-H], 5.95 [t, JH,H = 5.8 Hz, 1 H, CH2NH], 5.09 H-β], 8.39 [d, JH,H = 4.6 Hz, 2 H, H-β], 8.38 [d, JH,H = 4.5 Hz,
(s, 1 H, S-CH), 4.25 (m, 2 H, CH2-NHCO), 3.57 (m, 2 H, CH2-
2 H, H-β], 8.05–8.08 (m, 4 H, Ar-H), 8.02–8.04 (m, 2 H, Ar-H),
NH), 1.24 (s, 3 H, CH3), –1.08 (s, 2 H, NHpyrrole) ppm. UV/Vis 7.73–7.77 (m, 7 H, Ar-H, CH2-NH-CO), 7.70–7.72 (m, 3 H, Ar-
3
3
(CHCl3): λmax (logεmax) = 324 (4.32), 426 (5.51), 522 (3.69), 587 H), 7.38 [d, JH,H = 7.6 Hz, 2 H, Ar-H], 7.20 [t, JH,H = 7.8 Hz, 2
+
3
3
(4.02), 682 nm (3.91). HRMS (ESI): calcd. for C51H41N8OS2
H, Ar-H], 6.88 [t, JH,H = 7.4 Hz, 2 H, Ar-H], 6.61 [t, JH,H =
5.8 Hz, 1 H, CH2NH], 4.40 [q, JH,H = 6.0 Hz, 2 H, CH2-NHCO],
3
845.2839 [M + H]+; found 845.2844. Chiral HPLC (Chiralpak IC;
n-hexane/ethanol, 4:1; 1 mL/min; UV 420 nm) Rt1 (aR) = 3.77 [q, 3JH,H = 6.0 Hz, 2 H, CH2-NH] ppm. UV/Vis (CHCl3): λmax
16.97 min, Rt2 (aS) = 21.93 min, k1 (aR) = 4.66, k2 (aS) = 6.31, α (logεmax) = 431 (5.41), 565 (3.97), 613 nm (4.05). HRMS (ESI):
= 1.35, Rs = 2.71.
calcd. for C47H36N7OZn+ 778.2267 [M + H]+; found 778.2249.
Eur. J. Org. Chem. 2012, 6526–6536
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6535

G. Canard et al.
FULL PAPER
[6] N. Vanthuyne, F. Andreoli, S. Fernandez, M. Roman, C. Rous-
sel, Lett. Org. Chem. 2005, 2, 433–443.
[7] a) C. Roussel, C. Popescu, Chirality 1994, 6, 251–260; b) C.
Roussel, M. Roman, F. Andreoli, A. Del Rio, R. Faure, N.
Vanthuyne, Chirality 2006, 18, 762–771.
[8] M. Kasha, H. R. Rawls, M. A. El-Bayoumi, Pure Appl. Chem.
1965, 11, 371–392.
[9] M. Gardner, A. J. Guerin, C. A. Hunter, U. Michelsen, C.
Rotger, New J. Chem. 1999, 23, 309–316.
[5-(2-{2-[4-Methyl-2-thioxothiazol-3(2H)-yl]phenylureido}-
ethylamino)-10,15,20-triphenylporphyrinato]zinc [(rac)-5-Zn]: The
free-base porphyrin (rac)-5-2H (200 mg, 237 μmol, 1 equiv.) and
zinc acetate dihydrate (52 mg, 237 μmol, 1 equiv.) were dissolved in
a mixture of CH2Cl2/MeOH (60 mL, 4:1). The mixture was stirred
at room temperature for 3 h then washed successively with satu-
rated aqueous NaHCO3 (80 mL), water (2ϫ 80 mL) and dried with
MgSO4. After evaporation, the resulting solid was purified by
chromatography on silica gel (CH2Cl2/MeOH, 98:2). After
recrystallisation from CH2Cl2/n-heptane, the zinc porphyrin (rac)-
5-Zn (177 mg, 195 μmol, 82%) was obtained as a dark-green solid.
Rf = 0.13 (CH2Cl2/MeOH, 98:2). 1H NMR (400 MHz, [D6]DMSO,
[10] a) F. Hajjaj, Z. S. Yoon, M.-C. Yoon, J. Park, A. Satake, D.
Kim, Y. Kobuke, J. Am. Chem. Soc. 2006, 128, 4612–4623; b)
R. Takahashi, Y. Kobuke, J. Org. Chem. 2005, 70, 2745–2753.
[11] a) X. Huang, K. Nakanishi, N. Berova, Chirality 2000, 12, 237–
255; b) V. V. Borovkov, T. Harada, Y. Inoue, R. Kuroda, Angew.
Chem. 2002, 114, 1436–1439; Angew. Chem. Int. Ed. 2002, 41,
1378–1381; c) V. V. Borovkov, T. Harada, G. A. Hembury, Y.
Inoue, R. Kuroda, Angew. Chem. 2003, 115, 1788–1791; Angew.
Chem. Int. Ed. 2003, 42, 1746–1749; d) N. Yoshida, T. Ishizuka,
A. Osuka, D. H. Jeong, H. S. Cho, D. Kim, Y. Matsuzaki, A.
Nogami, K. Tanaka, Chem. Eur. J. 2003, 9, 58–75; e) G.
Bringmann, D. C. G. Götz, T. A. M. Gulder, T. H. Gehrke, T.
Bruhn, T. Kupfer, K. Radacki, H. Braunschweig, A. Heck-
mann, C. Lambert, J. Am. Chem. Soc. 2008, 130, 17812–17825;
f) A. Mammana, G. Pescitelli, T. Asakawa, S. Jockusch, A. G.
Petrovic, R. R. Monaco, R. Purrello, N. J. Turro, K. Nakani-
shi, G. A. Ellestad, M. Balaz, N. Berova, Chem. Eur. J. 2009,
15, 11853–11866; g) I. Cohen-Ofri, M. van Gastel, J. Grzyb, A.
Brandis, I. Pinkas, W. Lubitz, D. Noy, J. Am. Chem. Soc. 2011,
133, 9526–9535.
[12] a) H.-Y. Liu, J.-W. Huang, X. Tian, X.-D. Jiao, G.-T. Luo, L.-
N. Ji, Chem. Commun. 1997, 1575–1576; b) T. S. Balaban, M.
Linke-Schaetzel, A. D. Bhise, N. Vanthuyne, C. Roussel, Eur.
J. Org. Chem. 2004, 3919–3930; c) Y. Matano, T. Shinokura,
K. Matsumoto, H. Imahori, H. Nakano, Chem. Asian J. 2007,
2, 1417–1429; d) G. Gottarelli, S. Lena, S. Masiero, S. Pierac-
cini, G. P. Spada, Chirality 2008, 20, 471–485; e) L. Zeng, Y.
He, Z. Dai, J. Wang, Q. Cao, Y. Zhang, ChemPhysChem 2009,
10, 954–962; f) P. Iavicoli, H. Xu, L. N. Feldborg, M. Linares,
M. Paradinas, S. Stafström, C. Ocal, B. Nieto-Ortega, J. Cas-
ado, J. T. López Navarrete, R. Lazzaroni, S. D. Feyter, D. B.
Amabilino, J. Am. Chem. Soc. 2010, 132, 9350–9362.
[13] HyperChem, v. 8.0, Hypercube, Inc., Gainsville, Florida.
[14] G. Pescitelli, S. Gabriel, Y. Wang, J. Fleischhauer, R. W.
Woody, N. Berova, J. Am. Chem. Soc. 2003, 125, 7613–7628.
[15] a) N. Harada, K. Nakanishi, Circular Dichroic Spectroscopy.
Exciton Coupling in Organic Stereochemistry, University Sci-
ence Books, Mill Valley, CA, 1983; b) N. Berova, K. Nakanishi,
in: Circular Dichroism. Principles and Applications, 2nd ed.
(Eds.: N. Berova, K. Nakanishi, R. W. Woody), Wiley-VCH,
New York, 2000, pp. 337–382; c) N. Harada, K. Nakanishi, N.
Berova, in: Comprehensive Chiroptical Spectroscopy, vol. 2
(Eds.: N. Berova, P. L. Polavarapu, K. Nakanishi, R. W.
Woody), Wiley-VCH, New York, 2012, pp. 115–166.
[16] K.-i. Yamashita, K. Kataoka, M. S. Asano, K.-i. Sugiura, Org.
Lett. 2012, 14, 190–193.
3
3
25 °C): δ = 9.46 [d, JH,H = 4.8 Hz, 2 H, H-β], 8.48 [d, JH,H
=
4.8 Hz, 2 H, H-β], 8.42 [d, 3JH,H = 4.8 Hz, 2 H, H-β], 8.41 [d, 3JH,H
= 4.8 Hz, 2 H, H-β], 8.18 [d, 3JH,H = 8.4 Hz, 1 H, Ar-H], 8.07–8.09
(m, 4 H, Ar-H), 8.04–8.06 (m, 2 H, Ar-H), 7.75–7.77 (m, 6 H, Ar-
H), 7.71–7.73 (m, 3 H, Ar-H), 7.64 (s, 2 H, NH-CO), 7.41 (ddd,
3JH,H = 8.8, 3JH,H = 6.8, 4JH,H = 2.0 Hz, 1 H, Ar-H), 7.35 (t, 3JH,H
3
= 5.8 Hz, 1 H, CH2NH), 7.08–7.14 (m, 2 H, Ar-H), 6.73 (d, JH,H
3
= 1 Hz, 1 H, S-CH), 4.37 (br. s, 2 H, CH2-NHCO), 3.74 (q, JH,H
4
= 6.0 Hz, 2 H, CH2-NH), 1.24 (d, JH,H = 1 Hz, 3 H, CH3) ppm.
UV/Vis (CHCl3): λmax (logεmax) = 319 (4.47), 430 (5.49), 561 (4.03),
613 nm (4.02). HRMS (ESI): calcd. for C51H39N8OS2Zn+ 907.1974
[M + H]+; found 907.1961.
[(aS)-5-(2-{2-[4-Methyl-2-thioxothiazol-3(2H)-yl]phenylureido}-
ethylamino)-10,15,20-triphenylporphyrinato]zinc [(aS)-5-Zn]: Pre-
pared as for porphyrin (rac)-5-Zn starting from porphyrin (aS)-5-
2H (75 mg, 88 μmol, 1 equiv.) and zinc acetate dihydrate (19 mg,
88 μmol, 1 equiv.). Zinc porphyrin (aS)-5-Zn (73 mg, 80 μmol,
91%) was obtained as a dark-green solid.
[(aR)-5-(2-{2-[4-Methyl-2-thioxothiazol-3(2H)-yl]phenylureido}-
ethylamino)-10,15,20-triphenylporphyrinato]zinc [(aR)-5-Zn]: Pre-
pared as for porphyrin (rac)-5-Zn starting from porphyrin (aR)-5-
2H (75 mg, 88 μmol, 1 equiv.) and zinc acetate dihydrate (19 mg,
88 μmol, 1 equiv.). Zinc porphyrin (aR)-5-Zn (73 mg, 80 μmol,
91%) was obtained as a dark-green solid.
Supporting Information (see footnote on the first page of this arti-
cle): Structural analysis of porphyrin 3-2H, supplementary figures,
1
DOSY NMR characterization of 4-Zn J-dimer. H NMR spectra
of porphyrins 2-Zn, 3-Zn and 5-Zn.
Acknowledgments
The autors thank Melanie Decostanzi for her assistance in the syn-
thesis of precursors.
[1] T. S. Balaban, in: Handbook of Porphyrin Science with Applica-
tions to Chemistry, Physics, Materials Science Engineering, Bio-
logy and Medicine, vol. 1 (Eds.: K. M. Kadish, K. M. Smith,
R. Guilard), World Scientific, Singapore, 2010, pp. 221–306.
[2] A. Satake, Y. Kobuke, Org. Biomol. Chem. 2007, 5, 1679–1691.
[3] a) T. S. Balaban, Acc. Chem. Res. 2005, 38, 612–623; b) T. S.
Balaban, A. D. Bhise, M. Fischer, M. Linke-Schaetzel, C.
Roussel, N. Vanthuyne, Angew. Chem. 2003, 115, 2190–2194;
Angew. Chem. Int. Ed. 2003, 42, 2140–2144.
[4] T. Jochum, C. M. Reddy, A. Eichhöfer, G. Buth, J. Szmytkow-
ski, H. Kalt, D. Moss, T. S. Balaban, Proc. Natl. Acad. Sci.
USA 2008, 105, 12736–12741.
[5] M. C. Balaban, C. Chappaz-Gillot, G. Canard, O. Führ, C.
Roussel, T. S. Balaban, Tetrahedron 2009, 65, 3733–3739.
[17] G. A. Hembury, V. V. Borovkov, Y. Inoue, Chem. Rev. 2008,
108, 1–73.
[18] C. Brückner, J. J. Posakony, C. K. Johnson, R. W. Boyle, B. R.
James, D. Dolphin, J. Porphyrins Phthalocyanines 1998, 2, 455–
465.
[19] P. J. Angiolillo, V. S. Y. Lin, J. M. Vanderkooi, M. J. Therien, J.
Am. Chem. Soc. 1995, 117, 12514–12527.
[20] R. D. Hartnell, A. J. Edwards, D. P. Arnold, J. Porphyrins
Phthalocyanines 2002, 6, 695–707.
[21] M. O. Senge, X. Feng, J. Chem. Soc., Perkin Trans. 1 2000,
3615–3621.
Received: July 11, 2012
Published Online: October 9, 2012
6536
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6526–6536