Novel EDTA-ligands containing an integral perylene bisimide (PBI) core as an optical reporter unit

Article abstract of DOI:10.1039/c4ob01007h

The synthesis, characterization and metal complexation of a new class of perylene bisimides (PBIs) as an integral part of ethylenediaminetetraacetic acid (EDTA) are reported. The simplest representative, namely derivative 1a, was synthesized both by a convergent as well as a direct approach while the elongated derivatives, 1b and 1c, were obtained only via a convergent synthetic pathway. All these new prototypes of water-soluble perylenes are bolaamphiphiles and were fully characterized by ¹H- and ¹³C-NMR spectroscopy, matrix assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry and IR spectroscopy. In order to acquaint ourselves with the behaviour in solution of our PBIs bearing dendritic wedges, the simplest derivative, 1a, was chosen and tested by means of UV/Vis and fluorescence spectroscopy as well as by zeta-potential measurements. A photoexcitation induced intramolecular photo-electron transfer (PET) can be observed in these molecules. Therefore their potential applications as sensors can be imagined. Model compound 1a efficiently coordinates trivalent metal cations both in water and dimethyl sulfoxide (DMSO). Significantly, the effects of the complexation strongly depend on the aggregation state of the PBI molecules in solution. As a matter of fact, in water, the presence of M ³⁺ ions triggers the formation of light emitting supramolecular aggregates (excimers). On the other hand, in DMSO-rich solutions metal complexation leads to the suppression of the PET and leads to a strong fluorescence enhancement. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ob01007h

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Novel EDTA-ligands containing an integral
perylene bisimide (PBI) core as an optical
reporter unit†
Cite this: Org. Biomol. Chem., 2014,
12, 7045
Mario Marcia,a Prabhpreet Singh,b Frank Hauke,a Michele Magginic and
Andreas Hirsch*a
The synthesis, characterization and metal complexation of a new class of perylene bisimides (PBIs) as
an integral part of ethylenediaminetetraacetic acid (EDTA) are reported. The simplest representative,
namely derivative 1a, was synthesized both by a convergent as well as a direct approach while the
elongated derivatives, 1b and 1c, were obtained only via a convergent synthetic pathway. All these new
prototypes of water-soluble perylenes are bolaamphiphiles and were fully characterized by 1H- and
13C-NMR spectroscopy, matrix assisted laser desorption ionization-time of flight (MALDI-TOF) mass
spectrometry and IR spectroscopy. In order to acquaint ourselves with the behaviour in solution of our
PBIs bearing dendritic wedges, the simplest derivative, 1a, was chosen and tested by means of UV/Vis and
fluorescence spectroscopy as well as by zeta-potential measurements. A photoexcitation induced intra-
molecular photo-electron transfer (PET) can be observed in these molecules. Therefore their potential
applications as sensors can be imagined. Model compound 1a efficiently coordinates trivalent metal
cations both in water and dimethyl sulfoxide (DMSO). Significantly, the effects of the complexation
strongly depend on the aggregation state of the PBI molecules in solution. As a matter of fact, in water,
the presence of M3+ ions triggers the formation of light emitting supramolecular aggregates (excimers).
On the other hand, in DMSO-rich solutions metal complexation leads to the suppression of the PET and
leads to a strong fluorescence enhancement.
Received 15th May 2014,
Accepted 10th July 2014
DOI: 10.1039/c4ob01007h
www.rsc.org/obc
PBIs are usually insoluble in water. Therefore, several
attempts have been performed in order to increase their solubi-
Introduction
Perylene bisimides (PBIs) represent a class of aromatic lity in aqueous media, either with functionalization in the bay
chromophores with a series of interesting properties. From an region12–16 or at the side imide groups.17–25 Recently, we have
industrial point of view they find widespread applications as introduced a series of highly water-soluble PBIs containing
high-performance pigments1 due to their fascinating opto- anionic (R1) or cationic (R2) Newkome dendronized substitu-
electronic and electrochemical properties.2 Therefore, they rep- ents – representative examples are given in Fig. 1. Their aggre-
resent integral components in photovoltaic devices,3 sensors,4 gation behavior dependent on the pH value, ionic strength
OLEDs,5 OFETs6 as well as dye-lasers7 and have been used as and concentration26–28 has been investigated and based on
building blocks for liquid crystals,8,9 for membrane labelling10 their pronounced solubility, their ability to exfoliate and stabil-
and in photodynamic therapy.11
ize single walled carbon nanotubes (SWCNTs)29 and graphene
in water30,31 has been demonstrated.
Here, we report the synthesis, characterization and pro-
perties of a new family of PBI-based amphiphiles. These
derivatives belong to a novel class of PBI-based surfactants,
where di- and polyamine spacers have been used as building
blocks for the construction of arrays of oligocarboxylic acid
molecules. They can be regarded as elongated ethylenediamine-
tetraacetic acid (EDTA) ligands containing an integral aromatic
perylene bisimide (PBI) core as an optical reporter unit.
In addition, these EDTA-PBIs (Fig. 2) are characterized by a
very small periphery in comparison with the Newkome dendro-
aDepartment of Chemistry and Pharmacy and Institute of Advanced Materials and
Processes (ZMP), Friedrich-Alexander University Erlangen – Nürnberg, Henkestrasse
42, 91054 Erlangen, Germany. E-mail: andreas.hirsch@fau.de;
Fax: (+49)9131-8526864
bDepartment of Chemistry, UGC Centre for Advanced Studies, Guru Nanak Dev
University, Amritsar 143005, India
cDepartment of Chemical Sciences, University of Padua, Via Marzolo 1, 35126
Padua, Italy
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4ob01007h
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7045–7058 | 7045

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 Structure of 1st generation Newkome dendronized PBIs.
Fig. 2 Structure of EDTA-PBIs, 1a–c.
nized ones (Fig. 1). The presence of big bulky substituents at allotropes where strong aggregation of the exfoliating agent is
both termini of the PBI is expected to strongly determine their generally pursued32 in order to optimise the production of
aggregation behavior in solution, leading to the formation of individualized graphene layers/carbon nanotubes in solution.
monomers even at high concentrations. Although this feature Therefore, smaller side groups emphasize the importance of
is generally desirable, it could be a hurdle in certain specific the molecular core, still ensuring good water solubility and
applications, such as the dispersion/exfoliation of carbon offering supplementary chelation properties. Moreover, the
7046 | Org. Biomol. Chem., 2014, 12, 7045–7058
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
terminal units of PBIs 1a–c provide an additional feature The solubility in organic solvents (such as DMSO; N-methyl
as they enable facile metal-chelation ability in both pyrrolidone, methanol; N,N-dimethyl formamide and diethyl
a
aqueous and organic solutions. Due to the presence of the ether) can be drastically increased by the addition of acids like
PBI core as an optical reporter unit, the complexation of TFA, formic acid or hydrochloric acid (HCl(aq)).
various metal cations can be studied easily by means of optical
In the latter case, though, the introduction of water leads to
analytical techniques, such as UV/Vis and fluorescence a strong aggregation of the EDTA-PBI derivative by intermole-
spectroscopy.
cular π–π stacking interactions. These induced aggregation
phenomena were investigated by H-NMR spectroscopy as well
1
as absorption and emission spectroscopy (ESI†).
Two aqueous (diluted acid and basic medium) and two
organic (DMSO and DMSO with addition of TFA) solvents have
Results and discussion
For the synthesis of 1a, two different chemical strategies were been chosen to study the aggregation behavior of the
pursued. The first route (Scheme 1, top) is based on a classical EDTA-PBI derivative 1a. For each system the most relevant
convergent synthetic approach, where the selectively functiona- optical properties have been determined and are listed in
lized protected amine 2 was condensed with perylene-3,4,9,10- Table 1.
tetracarboxylic dianhydride (PTCDA), yielding the tert-butyl
Normally, PBIs without functional moieties attached to the
protected version 3, which was subsequently deprotected in tri- aromatic core can exist as aggregated as well as monomeric
fluoroacetic acid (TFA) – for further details regarding the syn- species in solution, depending on the solvent used.
thesis of compound 2 see ESI.† The free acid 1a is obtained
In the case of monomeric PBIs, the distinct absorption
with an overall yield of 21.6% (based on the starting material band relative to the electronic transition S0→S1 is generally
putrescine). In addition, this approach is quite time consum- found in the range 400–600 nm and is characterized by three
ing and employs several selective protection and deprotection well resolved vibronic peaks. The predominant two features,
steps, which impede
synthesis.
a
straightforward scale-up of the located at ≈530 nm (0,0) and ≈490 nm (0,1), respectively,
exhibit a ratio greater than 1.6.34 Upon aggregation, pro-
The second strategy (Scheme 1, bottom) involves the direct nounced modifications appear in the spectra. The intensity of
alkylation of a PBI-based precursor amine and is performed in the (0,0) peak is reduced, while that of the (0,1) peak increases
only three steps: (a) the condensation of PTCDA with accompanied by a pronounced reduction of the absorption
putrescine, which yields the perylene bisamine 4; (b) the alkyl- coefficient.17 The ratio between the (0,0) and (0,1) peaks
ation of 4 to the corresponding tetraester 3 and (c) the quanti- reaches a value of ≈0.7 or lower in the case of strong aggrega-
tative removal of the tert-butyl protection group releasing the tion.29,35 As a result, the colour of the solution changes from
free acid 1a with an overall yield of 11.6%.
orange to red.
The direct synthesis is primarily based on low cost chemi-
As depicted in Fig. 3, 1a is significantly aggregated, at room
cals and does not require the use of protected amine precur- temperature, both under diluted basic (NaOH(aq), c = 1 ×
sors such as amine 2. Therefore, the outlined synthetic route 10−3 M) and acid (HCl(aq), c = 1 × 10−3 M) aqueous conditions,
provides the basis for a cost efficient large scale synthesis of due to the strong π–π stacking interactions of the perylene
these novel EDTA-PBI derivatives.
aromatic cores.
Nevertheless, the drawback of this approach is the alkyl-
The behavior of 1a can be explained as follows. Under basic
ation of the derivative 4 due to its low solubility in the aqueous conditions, pH ≈ 11, the EDTA-PBI is completely
common suitable solvents for SN2 reactions (e.g. CH3CN), deprotonated (EDTA is completely deprotonated at pH > 10.6).
which is responsible for the relatively moderate overall yield.
Therefore, the aggregation proceeds most certainly via the for-
With this knowledge in hand, the highly branched EDTA- mation of PBI stacked systems where each molecule is rotated
PBIs 1b and 1c were synthesized solely according to a conver- with respect to its nearest neighbors. In such a way, the
gent approach (Schemes 2 and 3). The synthesis of the respect- electrostatic repulsion of the residual charges situated at the
ive precursor amines 2, 11 and 17 is reported in the ESI.† All periphery of each molecule would be minimized.36
the derivatives were fully characterized by 1H- and 13C-NMR
In acid aqueous solution the aggregation is even more pro-
and IR spectroscopy as well as by mass spectrometry and nounced. As a matter of fact, at the pH under investigation
elemental analysis. (pH ≈ 3) it is impossible to have a complete protonation of 1a
For the detailed investigation of the aggregation behavior of due to the fact that the isoelectric point of 1a is located at a
these novel compounds and for the direct comparison of their value of pI = 2 or below.37 We expect thus that 1a exists in an
results with other commonly available chelating agents, the equilibrium between several partly protonated species
simplest derivative 1a has been chosen and solution based (HxEDTAy+, where x < 4 and y < 2) in solution. Therefore, no
optical measurements were carried out.
sufficient electrostatic repulsion is believed to hinder the
aggregation by π–π stacking successfully. Moreover, the absorp-
Compound 1a exhibits a good solubility in basic (NaOH(aq)
,
c = 1 × 10−3 M) as well as acidic (HCl(aq), c = 1 × 10−3 M) tion spectrum of 1a in diluted HCl(aq) shows a great broaden-
aqueous media. Among the common organic solvents, 1a ing of the (0,1) peak in comparison with solutions of 1a in
shows very good solubility only in dimethyl sulfoxide (DMSO). diluted NaOH(aq) at the same concentration of the surfactant.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7045–7058 | 7047

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Convergent (top) and direct (bottom) synthesis of EDTA-PBI 1a. (i) Imidazole, Zn(OAc)2, 110 °C, 4 h, yield: 64%; (ii) TFA–CH2Cl2 (2 : 1),
rt, 5 days, yield: 84%; (iii) toluene, reflux, 4 h, yield: 89%; (iv) acetonitrile, DIPEA, tert-butyl bromoacetate, 60 °C, 24 h, yield: 13%; (v) formic acid, RT,
2 days, yield: 100%.
The enlargement of the (0,1) peak accounts generally for an
Furthermore, H-aggregates38 are formed in aqueous solu-
extended aggregation in solution17 and this confirms con- tion both in diluted NaOH(aq) and HCl(aq). In both cases, the
sequently that in diluted HCl(aq) the aggregation takes place presence of PBI-dimers39 is excluded.
more easily. In addition, further aggregation might derive
In order to gain more insight into these aggregation
from intermolecular hydrogen bonds between the protonated phenomena, we decided to perform additional zeta potential
carboxylic moieties.
measurements. The results are given in Table 2.
7048 | Org. Biomol. Chem., 2014, 12, 7045–7058
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 2 Convergent synthesis of EDTA-PBI 1b. (i) Imidazole, Zn(OAc)2, 110 °C, 4 h, yield: 53%; (ii) TFA–CH2Cl2 (2 : 1), RT, 5 days, yield: 82%.
Scheme 3 Convergent synthesis of EDTA-PBI 1c. (i) Imidazole, Zn(OAc)2, 110 °C, 4 h, yield: 56%; (ii) TFA–CH2Cl2 (2 : 1), RT, 5 days, yield 74%.
Zeta potential (ζ) measurements are used in order to
predict the stability of dispersions. According to Greenwood
et al.40 a zeta potential value higher than 30 mV indicates
Table 1 Optical properties of 1a in aqueous and organic solvents
that a colloidal suspension is rather stable; moderate instabil-
Property
NaOH(aq) HCl(aq)
)
4.38b 4.25b
543, 500 549, 481
587, 546 587, 546
DMSO
4.53c
H+-DMSOd
4.76c
ity is characterized by a ζ value between 10 mV and 30 mV
while extensive coagulation/flocculation occurs for a ζ value
around 0 mV.
From the results given in Table 2, it is possible to conclude
that 1a is more stable in diluted NaOH(aq) than in diluted
HCl(aq). However, in both cases the suspensions are rather
unstable (ζ < 15 mV).
log εmax (M−1 cm−1
λabs, max (nm)
λfluo, max (nm)
Φ (%)a
528, 494, 460 528, 494, 460
580, 543
580, 543
63.9 10.6
2.3 0.2 0.27 0.05 9.1 1.1
a The fluorescence quantum yield (Φ) is calculated taking fluorescein as
the reference in NaOH 0.1 M.33 b ε Calculated at 500 nm. c ε Calculated at
494 nm. d DMSO with addition of TFA, 3 × 10−4 M.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7045–7058 | 7049

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 Absorption and fluorescence profiles of 1a in NaOH(aq) (red
curve) and in HCl(aq) (black curve), c = 5 × 10−6 M.
Fig. 4 Absorption and fluorescence profiles of 1a in DMSO (red curve)
and in acid–DMSO (black curve), c = 5 × 10−6 M.
that the addition of diluted HCl(aq) would add water to the
system, which leads to a more pronounced aggregation of 1a.
First of all, in pure DMSO the behavior of 1a is quite
different (Fig. 4) from that in aqueous solutions at room temp-
erature. As a matter of fact, the (0,0) peak is higher in intensity
with respect to the (0,1) peak and their relative ratio is about
1.2. This value is lower than the threshold for completely
Table 2 Zeta potential measurements of 1a under aqueous conditions
(c = 1 × 10−6 M)
Property
NaOH(aq), c = 1 × 10−3
M
HCl(aq), c = 1 × 10−3
M
Zeta potential (mV) −13.3 2.6
0.7 0.7
The values of the UV and zeta potential measurements also monomeric PBIs, (0,0)/(0,1) ratio ≈1.6, and therefore we
corroborate with the fluorescence quantum yield data. conclude that 1a is present as a mixture of the monomeric and
Emission spectroscopy provides further insights into the aggregated species in solution.
equilibrium between monomeric and aggregated species. The
Due to the residual aggregation and because of the PET
dominant fluorescence can always be traced back to the mono- process, the fluorescence quantum yield of 1a in DMSO
meric species. In this case the emission profile is also the remains pretty low (Φ < 10%) but five times higher with
mirror image of the absorption spectrum. As aggregation takes respect to the value measured in NaOH(aq) solutions, where
place (Fig. 3) the absorption spectrum changes and therefore the EDTA-PBI is strongly aggregated. Upon addition of a small
the emission does not appear any longer as the mirror image amount of TFA to a DMSO solution of 1a the fluorescence
of the absorption spectrum.
increases (Fig. 4) as a result of the protonation of the two ter-
As evidenced from Table 1, the fluorescence quantum yield tiary amines, which lowers the energy of their nonbonding
(Φ) is approximately ten times lower for 1a in diluted HCl(aq) orbital below that of the HOMO of the PBI, allowing to switch
than for 1a in diluted NaOH(aq). Such drastically decreased Φ on the fluorescence of the chromophore.43 Under such con-
values in diluted HCl(aq) solutions might, however, appear ditions the fluorescence quantum yield increases drastically
counterintuitive. As a matter of fact, the tertiary amine func- (Table 1). Concomitantly, due to protonation of the tertiary
tionalities, situated at both termini of the PBI acceptor aro- amines, the aggregation via π–π stacking is lowered and the
matic core, determine a pronounced fluorescence quenching solubility of 1a is increased.
as a result of intra-molecular photo-induced electron transfer
(PET).
The possibility of switching on/off the fluorescence is
remarkable and offers access to a broad range of applications
Similar PBI dyes, which also bear bisamine functionalities for 1a and analogue PBIs.44–47 Moreover, as reported in the
linked to the perylene bisimide aromatic core, are generally literature48,49 it might be possible to correlate the fluorescence
characterized by an increase of the fluorescence either upon intensity with the pH and therefore obtain information about
acid addition19 or by chemical derivatization41 and also in the the acid/base equilibrium in solution. In our case, though,
presence of metal cations.42 However, in diluted HCl(aq) it is after the strong aggregation of 1a took place, titration with
observed that the aggregation process prevails over the amine either NaOH(aq) or HCl(aq) led to very little modifications of the
protonation. The aggregation by π–π stacking contributes fluorescence intensity and therefore hindered reliable determi-
therefore to an overall fluorescence quenching even if the PET nation of the six pKas values of 1a. However, due to the pres-
process might be partly suppressed.
ence of a ligand periphery attached to a PBI reporter unit,
When the aggregation in solution is prevented it is possible derivative 1a provides the possibility to investigate the coordi-
to examine the PET process in detail and to understand how it nation of metal ions in solution by means of spectroscopic
influences the fluorescence of PBI 1a. For this study, we measurements.
decided to investigate a solution of 1a in DMSO, in the pres-
ence or absence of an acid. TFA was chosen due to the fact agents have been reported, but to the best of our knowledge
Both naphthalene50,51 and perylene52–58 based chelating
7050 | Org. Biomol. Chem., 2014, 12, 7045–7058
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
study was accomplished in distilled water, where 1a is soluble
only at concentrations below 10−5 M. As mentioned before,
under aqueous conditions aggregation takes place easily. This
has the following drawbacks. Above all, the fluorescence of 1a
is greatly quenched and therefore it is more difficult to discern
the origin of the PET process. Moreover, the introduction of
ion species in a solution where 1a is already aggregated might
also favor the formation of extended aggregates, which could
be defined as supra-molecular polymers in solution.59–67
Fig. 5 Structures of EDTA (top) and IDA (bottom).
In a typical experiment, to a solution of 1a in distilled water
(c = 5 × 10−6 M), 10 equivalents of the chosen metal cations
were added at room temperature. The solution was agitated
mechanically for five minutes by means of a platform shaker
and then the fluorescence spectrum was recorded and com-
none of them with an elongated EDTA-like structure contain-
ing an integral PBI unit as an optical receptor unit.
As a matter of fact, the tailor-made periphery of 1a was pared with a reference solution of 1a in distilled water. The
designed to efficiently chelate metal cations. The lock-and-key
integrated fluorescence intensity in the presence (F) and
interaction of the metal cations with the tertiary amine func- absence (F0) of the metal cation was determined subsequently
tionalities at both sides of the EDTA-PBI will result in energy/ and the results were compared in terms of their ratio (F/F0). A
electron transfer, which will be transmitted to the central PBI
value of F/F0 higher than 1 indicates an enhancement of the
core via the PET process. By monitoring changes in the fluore- fluorescence of the PBI (EF effect), while a value lower than 1
scence of 1a it can be defined as a set of parameters in order indicates a fluorescence quenching (QF effect).
to determine the influence of different metals on the
The influence of the metal counter-ions has also been
EDTA-PBI chromophore.
tested by addition of 10 equivalents of different sodium salts
Of course, the aggregation by π–π stacking of 1a in water to a standard solution of 1a in distilled water. The results
and organic solvents will strongly influence the complexation
show that different anions contribute on average to the same
of metal cations in solution. Such hydrophobic interactions EF/QF in the presence of the same metal cations and therefore
could be responsible for the formation of a specific environ- a counter-ion effect is generally excluded. The results of the
ment in solution which allows easier coordination of certain
metal cations and not those expected for similar chelating have been divided according to the charge of the respective
complexation experiments for 1a in an aqueous environment
agents, such as EDTA or iminodiacetic acid, IDA (Fig. 5).
cations (see ESI†). First of all, it can be noted that the addition
The structure of 1a resembles that of EDTA with the inser-
of alkaline metal cations leads to an enhancement of the fluo-
tion of an integral PBI core between the two chelating moi- rescence of 1a in water solutions only in the presence of K+
eties. Upon insertion, the separation between the two tertiary and Cs+, while this does not happen for Li+ and Na+. Alkaline
amine substituents is enhanced and an intramolecular EDTA-
like coordination geometry (hexadentate ligand) is lost. decrease in the fluorescence intensity of 1a.
earth metal ions have a negligible effect or induce a little
However, an octahedral chelating geometry may be involved
upon intermolecular interactions in solution, e.g. due to
These preliminary results suggest that most of the alkaline
and alkaline earth cations are too small to interact efficiently
aggregation of the PBI molecules. Nevertheless, if in the aggre- with the chelating part of the EDTA-PBI. This might be
gates the peripheral groups remain remote enough, the coordi- explained considering the ionic radius of alkaline and alkaline
earth metal cations. Actually K+ and Cs+ possess the highest
nation geometry of 1a should resemble more that of IDA
(tridentate ligand).
Due to this complexity, we assume that the coordination environment created by 1a micelles in water solution only big
geometry of 1a in solution not only would be that of IDA or
EDTA, but also would be more reasonably a mixture of both. side groups and slightly affect the PET process.
values. Therefore, it might be argued that in a particular
cations can efficiently bind the nitrogen atom of the ligand
The complexation of metal cations was investigated both in
aqueous and organic media. Working in water based systems,
The addition of transition metals and lanthanides pro-
motes, instead, a drastic fluorescence quenching (F/F0 < 0.16),
on the one hand, is generally preferred for biological appli- with the exception of Au3+ (F/F0 ≈ 0.3) and Ag+ (F/F0 ≈ 0.7). In
cations although in our case this means a stronger aggregation particular the most pronounced QF values were obtained for
addition of copper ions (both Cu+ and Cu2+, QF = 0.0066) and
of the PBI aromatic moieties. In organic solvents, on the other
hand, the π–π stacking is generally hindered but only niche trivalent cations (M3+ QF ≈ 0.01). Such a drastic fluorescence
applications can be addressed. Moreover, in non-aqueous quenching is to be attributed to electron/energy transfer
systems it is more difficult to account for acid/base equilibria
which may affect the complexation.
between the d electrons of the transition metals and the
EDTA-PBI accepting molecules. As a matter of fact, the pres-
In order to gain the most comprehensive view of the affinity ence of transition metals or lanthanide cations leads to loca-
of 1a for metal cations in aqueous solutions at room tempera-
ture, thirty common salts have been employed (ESI†). The first centers and the PBI moieties which result in the dramatic
lized redox reactions (electron transfer) between the metal
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7045–7058 | 7051

View Article Online
Paper
Organic & Biomolecular Chemistry
quenching of the fluorescence of 1a. Additionally, the metal 680 nm, QF ≈ 0.01). The formation of this new band in the
cations might bridge vicinal PBI-micelles causing the formation emission spectrum of PBIs has been extensively discussed in
of extended aggregates. This highly impressive fluorescence the literature. According to Wang et al.35,76–78 who examined
quenching is observed mainly for bivalent metal cations with the self-organization of a PEG functionalized PBI in chloro-
the exception of Pb2+, whose QF is not as pronounced (F/F0 ≈ form solution, such a red shifted band in the emission spec-
0.4) and also for Cu+. Actually, Pb2+ is not a transition metal trum of PBIs may be attributed to the fluorescence properties
and a lower influence on the emission of the EDTA-PBI should of molecular assemblies of increased size. The same con-
be expected. Moreover, it is important to underline that a clusions have also been outlined by Neuteboom and co-
remarkable quenching of the fluorescence of 1a can also be workers79 when studying the optical features of polytetrahydro-
observed in the presence of metals with a d10 electronic con- furan substituted PBI polymers in ODCB. Arnaud et al.,71
figuration which cannot profit from ligand field effects, such Würthner et al.,70,80,81 Datar et al.,82 and Yagai et al.75 reported
as Zn2+, Cd2+ and Hg2+. Such metals do have a completely as well the formation of this red-shifted band due to self-
filled d level and cannot take part in any direct electron transfer assembled/polymeric PBI in solutions.
process. However, it has been reported that depending on the
To probe the stability of these metal-induced aggregated
geometry of the complex in solution other types of electron emitting species in solution, temperature dependent fluo-
transfer processes may contribute to the quenching of the rescence measurements have been performed. As depicted in
fluorescence of the chromophore.68 In addition, De Santis Fig. 7, an increase in temperature (from 25 °C to 75 °C) leads
et al.69 showed that Zn2+ has a specific affinity towards carb- normally to an increase of the F/F0 value and this correlates
oxylic acid groups which may lead to a strong fluorescence with a lower aggregation in solution. This observation corro-
quenching of the chromophore.
borated with a decrease in intensity and eventually with the
The addition of trivalent metal cations contributes, as well, disappearance of the band situated between 660 and 680 nm
to global quenching of the fluorescence of 1a. In particular, in the fluorescence spectrum. The only exception to this
the QF values are comparable to those recorded for copper pattern is found for In3+.
cations (F/F0 ≪ 0.1). Nevertheless, the quenching of fluo-
Considering that for the other metal cations such a band
rescence is now accompanied by a strong modification of the vanishes around 45 °C, we assume a higher stability for
emission spectrum of the EDTA-PBI molecule. As a matter of 1a-In3+ complexes and therefore presumably a higher binding
fact, the presence of trivalent metal cations induced the for- constant. For this cation in fact, the F/F0 ratio remains practi-
mation of a new broad band in the emission spectrum of 1a cally constant even at higher temperatures and the band due
(Fig. 6), which may account for the formation of emitting to PBI excimers persists till 65 °C.
aggregated species (excimers) in solution.70–75
Additionally, with the help of Van’t Hoff plots (see ESI†), it
Furthermore, after a decantation of 2–3 days, the solutions is possible to evaluate the thermodynamic parameters associ-
of 1a titrated with 10 equivalents of M3+ become transparent ated with the denaturation process. As a matter of fact, as the
and a solid precipitate was formed. The new broadened feature temperature increases the supramolecular structures based on
present in the emission spectrum of 1a, which is not present PBI agglomerates coordinated to metal ions tend to disaggre-
in the normal emission spectrum of 1a under aqueous con- gate in solution. In detail, the F/F0 ratio can be considered pro-
ditions, is located between 660 and 680 nm and the λmax of the portional to the equilibrium constant of the process (Keq) and
peak shifts depending on the metal cation added (Sm3+ λmax
≈
a plot of ln(F/F0) vs. T −1 can thus provide information about
660 nm, QF ≈ 0.05; In3+, Gd3+ and La3+ λmax ≈ 670 nm, QF ≈ the standard enthalpy (ΔH°) and entropy (ΔS°) of the de-
0.01; Fe3+ and Sc3+ λmax ≈ 675 nm, QF < 0.01; Ce3+ λmax
≈
naturation process for each ion. As shown in the ESI,† the plots
ln(F/F0) vs. T −1 are linear only at temperatures above 45 °C.
Fig. 6 Fluorescence spectra of 1a in water after addition of 10 eq. of
M3+ (Sm3+ is not shown for clarity, due to its higher QF value).
Fig. 7 Temperature dependent fluorescence spectra of 1a in water
after addition of 10 eq. of M3+
.
7052 | Org. Biomol. Chem., 2014, 12, 7045–7058
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Therefore, we assumed that both parameters are temperature
dependent and decided to extrapolate their values only in the
temperature interval 45 °C ≤ T ≤ 75 °C. With the exception of
Ce3+ and Sm3+ ΔH° can be determined to be 54.1
1.5 kJ
mol−1 and ΔS° to be 15.9 2.7 J mol−1. For the former ions,
ΔH° can be calculated as 16.2 1.3 kJ mol−1 and 31.9 2.1 kJ
mol−1 and ΔS° as 2.2 0.3 J mol−1 and 9.2 0.4 J mol−1
,
respectively. In the case of In3+ no calculation was carried out
due to the fact that the denaturation process appears not to
take place in the temperature range studied. The positive
enthalpy and entropy values indicate that the denaturation of
the metal-induced aggregates of 1a in solution is an endother-
mic process and therefore favored at high temperatures.
So far, we have described the complexation of 1a under
aqueous conditions, where 1a is aggregated and the introduc-
tion of trivalent metal ions most likely triggers the formation
of PBI emitting aggregates in solution.
Fig. 8 Absorption spectra of 1a in the DMSO–water mixture (9 : 1) after
addition of 10 eq. of Mx+
.
In order to gain more insights into the interaction of the
EDTA-PBI with metal cations in solution, it has been decided
to add the metal ions to 1a in the water–DMSO mixture. As
mentioned above, 1a is predominantly present as a monomer
in DMSO and these conditions should provide a better visual-
ization of the true interaction of a single PBI molecule with
the metal cation in solution. DMSO–water systems have been
very well studied in the literature.83–86 Due to the complex
interactions of water and DMSO molecules, which lead for
example to a pronounced melting point depression for a
DMSO molar fraction (xDMSO) of about 0.3387 it would be
difficult to attempt a detailed description of the effect of metal
ions on the PET of 1a over the entire molar fraction range.
Therefore, we preferred to focus our attention on the region of
0.8 ≤ xDMSO ≤ 1 (the so-called DMSO-rich region). Moreover, in
order to ensure sufficient solubility of the metal salts and to
avoid as much as possible the hydrophobic effect, it was
decided to work with mixtures where xDMSO = 0.9.
Five salts of transition metals were chosen to be investi-
gated. Namely, three divalent (Ni2+, Co2+ and Cu2+) and two tri-
valent (Al3+ and Fe3+) metal ions have been tested. These five
metal cations were selected among those who should form the
most stable complexes with IDA and EDTA moieties. First of
all, absorption measurements have been performed to investi-
gate the influence of these metal cations on the spectroscopic
properties of 1a. As the EDTA-PBI is mostly present as a
monomer in solution, absorption spectroscopy can provide
precious information about the metal complexation (Fig. 8).
The addition of bivalent and trivalent metal cations to the
EDTA-PBI solution results generally in a global decrease of
Fig. 9 Comparison of the F/F0 ratios for complexes of 1a with M2+ and
M3+ in a DMSO–water (9 : 1) mixture.
Upon addition of divalent metal cations, quenching of the
PBI fluorescence is observed, most likely due to the electron
transfer processes from the metal center to the aromatic core.
The affinity for trivalent metal cations remains high as well.
In DMSO-rich solutions, however, the addition of trivalent
cations is now correlated with an enhanced fluorescence of 1a
(Fig. 9). The presence of predominant monomeric species in
solution allows a successful interaction of the EDTA-PBI with
the metal cations which lead to a drastic quenching of the PET
process (Al3+ EF ≈ 1.18; Fe3+ EF ≈ 1.83).
Conclusion
the intensity of the absorption bands and in a lower value of In this contribution the first representatives of a new class of
the I529/493 ratio. As described before, a decrease in the latter perylene bisimide-based surfactants are reported. In particular,
parameter accounts for the formation of aggregates in solu- the synthesis, optical characterization and metal complexation
tion. In particular, these effects are strictly dependent on the ability of the PBI derivative 1a were investigated in detail. Two
metal ion and were recorded to be maximal for Co2+ and Al3+. reasonable synthetic pathways, leading to the target structure
Further proof of the interaction between 1a and the metal 1a, have been elaborated and discussed. The full optical
cations in DMSO-rich solutions is given by fluorescence characterization (UV/Vis and emission spectroscopy) has been
spectroscopy measurements. In DMSO-rich solution 1a is still reported both under aqueous and organic conditions. In the
mainly present as a monomer in solution.
former solvent a detailed study of the aggregation behavior has
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7045–7058 | 7053

View Article Online
Paper
Organic & Biomolecular Chemistry
been presented on the basis of absorption as well as fluore- sensitive reactions were performed under a N2 atmosphere.
scence data and zeta potential measurements, both under CH2Cl2 was distilled over CaH2, THF over Na/benzophe-
basic and acidic conditions.
none and DMF dried over 4 Å molecular sieves. Chromato-
Moreover, the influence of the intra-molecular PET process graphic purifications were performed with silica gel from
on the spectroscopic properties of dye 1a in DMSO solution Merck (Kieselgel 60, 40–60 µm, 230–400 mesh ASTM) in stan-
has been described. It has also been shown that upon addition dard glass columns. TLC was performed on aluminium sheets
of an acid the PET process can be suppressed and the coated with silica gel from Merck (F254). 1H and 13C NMR
1
peculiarity of the phenomenon has been discussed with spectra were recorded with a Brucker AV500 (500 MHz for H
regard to feasible applications. For example, it could be ima- and 125 MHz for 13C) spectrometer, a Jeol JNM EX 400
gined to fully exploit this pH-driven fluorescence sensitivity to (400 MHz for 1H and 100 MHz for 13C) and a Jeol Brucker
develop EDTA-PBI water soluble sensors. Furthermore, a Avance 300 (300 MHz for 1H and 75 MHz for 13C) spectro-
detailed investigation of the complexation of metal cations meter. Chemical shifts are reported in ppm at room tempera-
both in water and DMSO–water mixtures has been offered. ture (RT) using CDCl3 as the solvent and the internal standard,
This study revealed the importance of the aggregation state of unless otherwise indicated. Abbreviations used for splitting
the EDTA-PBI in the complexation properties of 1a. Under patterns are s = singlet, d = doublet, t = triplet, m = multiplet,
aqueous conditions, where bulky aggregates prevail, only big and dd = double doublet. IR spectra were recorded with a
alkaline metal cations lead to the suppression of the PET FT-IR Nicolet 5700, a Brucker Tensor 27 (ATR) and an ASI
process. Bi- and trivalent cations were responsible for a pro- React IRTM 1000 spectrometer. For UV/Vis spectra a Perkin
nounced fluorescence quenching (QF). Additionally, the com- Elmer Lambda 1050 was used. Fluorescence was measured
plexation of trivalent metal ions is characterized by the with a Horiba Scientific Fluorolog-3 spectrometer with a PMT
formation of emitting aggregate species (excimers), which detector. MALDI-TOF mass spectrometry was carried out on a
modify profoundly the emission spectrum of the EDTA-PBI Shimadzu AXIMA Confidence, N2 UV laser (337 nm), 50 Hz
itself. The origin and stability of these species were discussed (reflectron). The matrices used were 2′,4′,6′-trihydroxyaceto-
by means of temperature dependent fluorescence spectroscopy phenone monohydrate (THAP), 2,5-dihydoxybenzoic acid (DHB),
and In3+ has been observed to form the strongest complexes, 3,5-dimethoxy-4-hydroxycinnamic acid (SIN) and 2-[(2E)-3-(4-
among the metal cations tested. A strong affinity of 1a for di- tert-butylphenyl)-2-methylprop-2-enylidene]
malonitrile
and trivalent metal ions has been demonstrated in DMSO-rich (DCTB). ESI mass spectrometry was performed with an Agilent
solutions as well. The former metal cations are involved in Technologies 1100 Series LC/MSD Trap-SL spectrometer
energy/electron transfer also with predominantly monomeric equipped with an ESI source, a hexapole filter and an ionic
PBIs. The addition of the latter results, instead, in a strong trap and Brucker maXis 4G. Zeta-potential measurements were
fluorescence enhancement (EF).
carried out on a Malvern Zetasizer Nano system with
The intriguing structure of derivatives 1a–c which combines irradiation from a 633 nm He–Ne laser. The solution of 1a
namely an electron deficient central aromatic structure (PBI with the metal cations were prepared by mechanically stirring
core), a polycarboxylic backbone and a chelating periphery the solutions with a Heidolph Unimax 1010 platform shaker.
renders them suitable for many applications. Among the most For elemental analyses, a CE instrument EA 1110 CHNS was
appealing ideas, it would be interesting to exploit the potential used.
of 1a for the aqueous exfoliation of carbon allotropes or other
In the following the syntheses of the different PBI deriva-
2D-layered inorganic materials, such as molybdenum disul- tives are presented. The synthesis of the amine precursors 2,
phide (MoS2) or tungsten disulphide (WS2). In the latter case, 11 and 17 is presented in the ESI.† Compound 4 was prepared
in fact, it is known that specific ions (like Ni2+ or Al3+) are used according to a slightly modified procedure adapted from Xue
to create inclusion during the exfoliation process and help to et al.89 For the putrescine based derivative 1a, the direct syn-
hinder the re-stacking of the dispersed material.88
Finally, the results of the complexation study, collected both defined as route ii.
thesis is defined as route i, while the convergent synthesis is
under aqueous and organic conditions, underline the striking
N-Bis-(4-aminobutane)-3,4,9,10-PBI (4). [route i]: A mixture
affinity of 1a for heavy trivalent metal ions. The identification of of PTCDA (10 g, 2.6 × 10−2 mol) and 1,4-diaminobutane (4 eq.)
such a successful interaction of transition metals and lantha- in toluene (250 mL) was stirred at reflux (110 °C) for 4 hours.
nide cations with EDTA-PBIs opens up the way to challenging After cooling down to room temperature, the mixture was fil-
application in water decontamination, e.g. the detection of hard tered under vacuum and washed with toluene. The crude solid
metals which is of extreme environmental importance.
was then re-suspended in KOH 5 M (200 mL) and stirred for
15 hours at ambient temperature. Subsequently, the suspen-
sion was filtered and 4 was collected as a red-brownish solid,
which was dried under vacuum (16.4 g, yield = 89%).
Experimental section
Materials and methods
1H NMR (500 MHz, DMSO-d6 + TFA): δ = 1.65–1.76 (8H,
+
2 m, 4CH2), 2.89 (4H, m, 2CH2CH2NH3 ), 4.07 (4H, t,
+
Reagents and solvents were purchased from Acros Organics, 2NCH2CH2), 7.74 (6H, t, 2CH2NH3 ), 8.30 (4H, d, ArH), 8.55
Sigma Aldrich and used without further purification. Moisture (4H, d, 4ArH) ppm.
7054 | Org. Biomol. Chem., 2014, 12, 7045–7058
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
ESI-MS: m/z 533.3 (M + H)+, 267.2 (M + 2H)2+
.
13C NMR (75 MHz, CDCl3, 25 °C): δ = 22.080 (2 C, CH2),
IR (KBr disc): ν = 3350, 3300 (primary amine, –NH2 stretch- 24.631 (2 C, CH2), 40.479 (2 C, CH2), 55.547 (2 C, CH2), 58.388
ing); 2927, 2849 (stretching –CH2); 1690, 1653 (stretching CvO (4 C, CH2), 122.430 (2 C, Ar-C), 124.888 (4 C, Ar-CH), 126.837
bisimide) cm−1
.
(2 C, Ar-C), 129.781 (4 C, Ar-C), 133.566 (4 C, Ar-CH), 136.483
N-Bis-(tert-butyl-(2,2′-aminobutylazanediyl)-diacetate)-3,4,9,10 (4 C, Ar-C), 166.146 (4 C, CON), 169.575 (4 C, COO) ppm.
PBI (3). [route i]: A mixture of 4 (280 mg, 5.3 × 10−4 mol), MALDI-TOF (DHB): m/z 649 (MH − 2CH2CO2)+, 708 (MH −
acetonitrile (15 mL), DIPEA (10 eq.) and tert-butyl bromo- CH2CO2)+, 765 (M + H)+, 787 (M + Na)+.
acetate (8 eq.) was stirred at 60 °C for 24 hours. Once
IR (ATR): ν = 3468.87, 3016.23, 2969.14, 2547.61, 1735.46,
cooled down to room temperature, it was vacuum filtered and 1687.25, 1645.08, 1593.03, 1576.71, 1441.85, 1402.08, 1381.67,
the crude solid was washed with acetonitrile and water. Sub- 1341.39, 1246.11, 1169.21, 1137.12, 1088.06, 809.08, 794.31,
sequently the solid residue was dissolved in chloroform (5 mL) 744.42, 719.63 cm−1
.
and hexane was added (100 mL). The mixture was stirred for EA for C46H39F9N4O18 (765) × 3CF3COOH: calcd C 49.92, H
10 minutes at room temperature and then allowed to stand for 3.55, N 5.06; found C 49.37; H 4.10; N 4.96.
one night. The precipitate was filtered and dried under
vacuum. 3 is isolated as a brown solid (70 mg, yield = 13%).
Tetra-tert-butyl 2,2′,2″,2′′′((1,3,8,10-tetraoxoanthra[2,1,9-
def:6,5,10-d′e′f′]di-isoquinoline-2,9(1H,3H,8H,10H)-diyl)bis-
[route ii]: Precursor amine 2 (404 mg, 1.3 × 10−3 mol), (propane-3,1diyl)-bis(((2-(tert-butoxy)-2-oxoethyl)azanediyl)bis-
PTCDA (250 mg, 6.4 × 10−4 mol), imidazole (868 mg) and zinc (butane-4,1-diyl))bis(azanetriyl))-tetraacetate (5). Precursor
acetate (35 mg) were mixed and heated up to 110 °C for 4 h. amine 11 (0.62 g, 1.3 mmol), PTCDA (0.25 g, 0.64 mmol),
Afterwards, dichloromethane was added to the solid residue imidazole (0.87 g, 13.0 mmol) and zinc acetate (0.035 g,
and column chromatography in CH2Cl2–EtOH (98 : 2) was per- 0.2 mmol) were heated at 110 °C for 4 h. Afterwards, dichloro-
formed to isolate a red solid, 3 (404 mg, yield = 64%).
methane was added and the solid residue was purified by
1H NMR (300 MHz, CDCl3, 25 °C): δ = 1.44 (s, 36H, 12 × column chromatography (SiO2, dichloromethane–ethanol,
CH3), 1.63 (quintuplet, J = 7.0 Hz, 4H, 2 × CH2), 1.79 (quintu- 95 : 5). 5 is isolated as a red solid (gummy) (1.9 g, yield =
plet, J = 7.5 Hz, 4H, 2 × CH2), 2.77 (t, J = 7.6 Hz, 4H, 2 × CH2), 53.3%).
3.44 (s, 8H, 4 × NCH2), 4.21 (t, J = 7.4 Hz, 4H, 2 × CH2), 8.33 (d,
J = 8.0 Hz, 4H, ArH), 8.48 (d, J = 8.0 Hz, 4H, ArH) ppm.
1H NMR (300 MHz, CDCl3): δ = 1.43 (s, 18H, 6 × CH3), 1.44
(bs, 36H, 12 × CH3) 1.47–1.48 (m, 4H, 2 × CH2), 1.74 (broad
13C NMR (75 MHz, CDCl3, 25 °C): δ = 25.536 (2 C, CH2), quintuplet, 4H, 2 × CH2), 1.92 (quintuplet, J = 7.2 Hz, 4H, 2 ×
25.641 (2 C, CH2), 28.049 (12 C, CH3), 40.267 (2 C, CH2), CH2), 2.63 (t, J = 6.8 Hz, 4H, 2 × CH2), 2.68 (t, J = 7.0 Hz, 4H, 2
t
53.864 (2 C, CH2), 55.732 (4 C, CH2), 80.732 (4 C, quat. C Bu), × CH2), 2.78 (t, J = 7.0 Hz, 4H, 2 × CH2), 3.30 (s, 4H, 2 × NCH2),
122.469 (4 C, Ar-CH), 122.850 (2 C, Ar-C), 125.388 (2 C, Ar-C), 3.41 (s, 8H, 4 × NCH2), 4.24 (t, J = 7.6 Hz, 4H, 2 × CH2), 8.51 (d,
128.552 (4 C, Ar-C), 130.625 (4 C, Ar-CH), 133.581 (4 C, Ar-C), J = 8.0 Hz, 4H, ArH), 8.61 (d, J = 8.0 Hz, 4H, ArH) ppm.
162.762 (4 C, CON), 170.773 (4 C, COO) ppm.
13C NMR (75 MHz, CDCl3): δ = 25.18 (2 C, CH2), 25.66 (2 C,
CH2), 25.87 (2 C, CH2), 28.04 (6 C, CH3), 28.07 (12 C, CH3),
MALDI-TOF (THAP): m/z 989 (M + H)+, 1011 (M + Na)+.
IR (ATR): ν = 2976.44, 2932.02, 1731.53, 1692.97, 1654.17, 38.91 (2 C, CH2), 51.76 (2 C, CH2), 53.73 (2 C, CH2), 54.06 (2 C,
1594.27, 1340.47, 1251.21, 1215.23, 1142.65, 988.15, 809.48, CH2), 55.14 (2 C, CH2), 55.72 (4 C, CH2), 80.56 (2 C, quat. C
t
745.88 cm−1
.
tBu), 80.70 (4 C, quat. C Bu), 122.71 (4 C, Ar-CH), 123.07 (2 C,
EA for C56H68N4O12: calcd C 68.00, H 6.93, N 5.66; found C Ar-C), 125.83 (2 C, Ar-C), 128.90 (4 C, Ar-C), 130.91 (4 C, Ar-
67.66; H 6.90; N 5.65.
CH), 133.99 (4 C, Ar-C), 162.99 (4 C, CON), 170.79 (4 C, COO),
N-Bis-(tert-butyl-(2,2′-aminobutylazanediyl)-diacetic
acid)- 170.95 (2 C, COO) ppm.
3,4,9,10-PBI (1a). [route i]: A solution of 3 (520 mg, 5.3 × 10−4
MS-ESI(+): m/z = 1332 [M+ + H].
mol) in formic acid (20 mL) was stirred at room temperature
IR (ATR): ν = 2975.77, 2933.03, 2865.49, 1729.00, 1694.37,
for 2 days. Acetonitrile (20 mL) was added and a solid precipi- 1655.12, 1594.05, 1440.45, 1402.63, 1365.62, 1345.19, 1247.93,
tated. The solvent was evaporated in vacuo. The crude solid 1216.78, 1145.75, 1069.42, 848.50, 809.39, 744.38 cm−1
was washed two times with acetonitrile and once with diethyl EA for C74H102N6O16: calcd C 66.74, H 7.72, N 6.31; found C
ether. 1a is isolated as a reddish solid (400 mg, quantitative 66.30; H 7.80; N 6.35.
yield). 2,2′,2″,2′′′-((1,3,8,10-Tetraoxoanthra[2,1,9-def:6,5,10-d′e′f′]-
.
[route ii]: A solution of 3 (250 mg, 2.5 × 10−4 mol) in TFA– diisoquinoline-2,9(1H,3H,8H,10H)-diyl)bis(propane-3,1diyl)bis-
CH2Cl2 (1 : 1) was stirred at room temperature for 5 days. After (((carboxymethyl)azanediyl)-bis(butane-4,1-diyl))bis(azanetriyl))-
evaporation of the solvent, the product was precipitated by tetraacetic acid (1b). 5 (0.5 g, 0.37 mmol) was dissolved in
addition of diethylether. The solid was filtered and dried 18 mL of TFA. The reaction mixture was stirred for 3 days at
in vacuo. A red-brown solid was obtained (160 mg, yield = room temperature. After evaporation of the solvent, the
84%).
product was precipitated on addition of diethyl ether. After fil-
1H NMR (300 MHz, CDCl3, 25 °C): δ = 1.99–2.06 (m, 8H, 4 × tration, the product was dried under vacuum. 1b is isolated as
CH2), 3.69 (t, J = 7.4 Hz, 4H, 2 × CH2), 4.39 (t, J = 6.6 Hz, 4H, a dark red solid (0.30 g, yield = 81.9%).
2 × CH2), 4.45 (s, 8H, 4 × NCH2), 8.80–8.86 (m, 8H, perylene
ArH) ppm.
1H NMR (300 MHz, TFA–CDCl3; (1 : 1)): δ = 1.85 (bq, 8H, 4 ×
CH2), 2.25 (bq, J = 6.8 Hz, 4H, 2 × CH2), 3.33–3.46 (m, 12H, 6 ×
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7045–7058 | 7055

View Article Online
Paper
Organic & Biomolecular Chemistry
CH2), 4.02–4.30 (m, 12H of 6 × NCH2 superimposed with 4H 2.50 (broad quintuplet, 4H, 2 × CH2), 3.49 (bt, 8H, 4 × CH2),
protons of 2 × CH2), 8.61 (d, J = 7.6 Hz, 4H, perylene ArH), 8.67 3.57 (bt, 8H, 4 × CH2), 3.68 (bt, 8H, 4 × CH2), 4.21 (s, 4H, 2 ×
(d, J = 8.4 Hz, 4H, perylene ArH) ppm.
NCH2), 4.39 (s, 12H, 6 × NCH2), 8.78 (d, J = 6.0 Hz, 4H, ArH),
8.84 (d, J = 6.0 Hz, 4H, ArH) ppm.
MS-ESI(+): m/z = 996 [M+ + 2H].
IR (ATR): ν = 3016.80, 2973.38, 2546.03, 1735.01, 1691.09,
1649.09, 1593.05, 1577.36, 1401.75, 1343.23, 1172.33, 1129.89,
MS-ESI(+): m/z = 1226 [M+ + 2H].
IR (ATR): ν = 2976.98, 2935.25, 1727.43, 1695.03, 1655.53,
1594.39, 1365.73, 1247.92, 1216.67, 1144.99, 848.11, 809.43,
809.29, 794.60, 719.90 cm−1
.
EA for C50H54N6O16 × 5CF3COOH: calcd C 46.04, H 3.80, N 744.48 cm−1
5.37; found C 45.74; H 4.29; N 5.07.
.
EA for C60H72N8O20 × 5CF3COOH: calcd C 46.83, H 4.32, N
Tetra-tert-butyl 2,2′,2″,2′′′-(11,11′-((1,3,8,10-tetraoxoanthra- 6.24; found C 45.66; H 4.38; N 6.12.
[2,1,9-def:6,5,10-d′e′f′]diisoquinoline-2,9(1H,3H,8H,10H)-diyl)-
bis(propane-3,1-diyl-bis(2,2,15,15-tetramethyl-4,13-dioxo-3,14-
dioxa-6,11-diazahexadecane-11,6-diyl-bis(propane-3,1-diyl)))bis-
(azanetriyl))tetraacetate (6). Precursor amine 17 (0.25 g,
Acknowledgements
0.38 mmol), PTCDA (0.07 g, 0.17 mmol), imidazole (0.24 g, MM, FH and AH thank the Deutsche Forschungsgemeinschaft
3.46 mmol) and zinc acetate (0.01 g, 0.05 mmol) were heated (DFG – SFB 953, Project A1 “Synthetic Carbon Allotropes”) and
at 110 °C for 4 h. Afterwards, dichloromethane was added to the Interdisciplinary Center for Molecular Materials (ICMM)
the solid residue and the residue was purified by column for the financial support. The research leading to these results
chromatography (SiO2, dichloromethane–ethanol, 95 : 5). 6 is has partially received funding from the European Union
isolated as a dark red solid (0.22 g, yield = 74.3%).
Seventh Framework Programme under grant agreement no.
1H NMR (300 MHz, CDCl3): δ = 1.41–1.42 (s, 72H, 24 × CH3. 604391 Graphene Flagship. PS is thankful to the Alexander
This multiplet superimposes on the signals of 8H protons of 4 von Humboldt foundation for granting the AvH research
× CH2), 1.61 (quintuplet, J = 6.9 Hz, 4H, 2 × CH2), 1.91 (quintu- Fellowship.
plet, J = 7.0 Hz, 4H, 2 × CH2), 2.57–2.60 (m, 12H, 6 × CH2), 2.68
(t, J = 7.4 Hz, 4H, 2 × CH2), 2.77 (t, J = 7.0 Hz, 4H, 2 × CH2),
3.20 (s, 4H, 2 × NCH2), 3.28 (s, 4H, 2 × NCH2), 3.40 (s, 8H, 4 ×
NCH2), 4.22 (t, J = 7.4 Hz, 4H, 2 × CH2), 8.43 (d, J = 8.0 Hz, 4H,
Notes and references
ArH), 8.55 (d, J = 8.0 Hz, 4H, ArH) ppm.
1 F. Wurthner, Chem. Commun., 2004, 1564.
2 T. Weil, T. Vosch, J. Hofkens, K. Peneva and K. Mullen,
Angew. Chem., Int. Ed., 2010, 49, 9068.
3 C. Li and H. Wonneberger, Adv. Mater., 2012, 24, 613.
4 M. Ramesh, H. C. Lin and C. W. Chu, Biosens. Bioelectron.,
2013, 42, 76.
5 Y. Li, T. Liu, H. Liu, M. Z. Tian and Y. Li, Acc. Chem. Res.,
2014, 47, 1186.
6 A. Ringk, X. R. Li, F. Gholamrezaie, E. C. P. Smits,
A. Neuhold, A. Moser, C. Van der Marel, G. H. Gelinck,
R. Resel, D. M. de Leeuw and P. Strohriegl, Adv. Funct.
Mater., 2013, 23, 2016.
13C NMR (75 MHz, CDCl3): δ = 25.219 (2 C, CH2), 25.372 (2
C, CH2), 25.884 (2 C, CH2), 26.056 (2 C, CH2), 28.023 (12 C,
CH3), 28.040 (6 C, CH3), 28.064 (6 C, CH3), 38.906 (2 C, CH2),
51.789 (2 C, CH2), 51.885 (2 C, CH2), 52.113 (2 C, CH2), 53.805
(2 C, CH2), 54.220 (2 C, CH2), 55.191 (2 C, CH2), 55.519 (2 C,
t
CH2), 55.704 (4 C, CH2), 80.482 (4 C, quat. C Bu), 80.523 (2 C,
t
t
quat. C Bu), 80.684 (2 C, quat. C Bu), 122.776 (4 C, Ar-CH),
123.101 (2 C, Ar-C), 125.908 (2 C, Ar-C), 128.952 (4 C, Ar-C),
130.956 (4 C, Ar-CH), 134.072 (4 C, Ar-C), 163.032 (4 C, CON),
170.728 (4 C, COO), 170.923 (4 C, COO) ppm.
MS-ESI(+): m/z = 1675 [M+ + 2H].
IR (ATR): ν = 2976.98, 2935.25, 1727.43, 1695.03, 1655.53,
1594.39, 1365.73, 1247.92, 1216.67, 1144.99, 848.11, 809.43,
7 H. Langhals, Helv. Chim. Acta, 2005, 88, 1309.
8 M. A. Muth, G. Gupta, A. Wicklein, M. Carrasco-Orozco,
T. Thurn-Albrecht and M. Thelakkat, J. Phys. Chem. C, 2014,
118, 92.
9 T. J. Zhang, D. M. Sun, X. K. Ren, L. L. Liu, G. Y. Wen,
Z. J. Ren, H. H. Li and S. K. Yan, Soft Matter, 2013, 9,
10739.
744.48 cm−1
.
EA for C92H136N8O20: calcd C 66.00, H 8.19, N 6.69; found C
64.87; H 8.65; N 6.56.
2,2′,2″,2′′′-((1,3,8,10-Tetraoxoanthra[2,1,9-def:6,5,10-d′e′f′]-
diisoquinoline-2,9(1H,3H,8H,10H)-diyl)bis(propane-3,1diyl)-
bis(((carboxymethyl)azanediyl)bis(butane-4,1-diyl))-bis(((carboxy- 10 L. Wang, L. Xu, K. G. Neoh and E.-T. Kang, J. Mater. Chem.,
methyl)azanediyl)bis(propane-3,1diyl))bis-(azanetriyl))tetra- 2011, 21, 6502.
acetic acid (1c). (0.23 g, 0.14 mmol) was dissolved in 18 mL 11 F. Yukruk, A. L. Dogan, H. Canpinar, D. Guc and
TFA. The reaction mixture was stirred for 6 days at RT. After E. U. Akkaya, Org. Lett., 2005, 7, 2885.
evaporation of the solvent, the product was precipitated on 12 M. Z. Yin, J. Shen, R. Gropeanu, G. O. Pflugfelder, T. Weil
addition of diethyl ether. After filtration, the product was dried and K. Mullen, Small, 2008, 4, 894.
under vacuum. 1c is isolated as a dark red solid (9.4 mg, 13 M. Yin, J. Shen, G. O. Pflugfelder and K. Mullen, J. Am.
yield = 56.2%). Chem. Soc., 2008, 130, 7806.
1H NMR (300 MHz, TFA–CDCl3; (1 : 1)): δ = 1.99 (broad 14 C. Kohl, T. Weil, J. Q. Qu and K. Mullen, Chem. – Eur. J.,
quintuplet, 8H, 4 × CH2), 2.41 (broad quintuplet, 4H, 2 × CH2),
2004, 10, 5297.
7056 | Org. Biomol. Chem., 2014, 12, 7045–7058
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
15 S. K. Yang, X. H. Shi, S. Park, S. Doganay, T. Ha and 42 G. Turkmen, S. Erten-Ela and S. Icli, Dyes Pigm., 2009, 83,
S. C. Zimmerman, J. Am. Chem. Soc., 2011, 133, 13206. 297.
16 Z. G. Zhang, C. L. Zhan, X. Zhang, S. L. Zhang, J. H. Huang, 43 L. Zang, R. C. Liu, M. W. Holman, K. T. Nguyen and
A. D. Q. Li and J. N. Yao, Chem. – Eur. J., 2012, 18, 12305. D. M. Adams, J. Am. Chem. Soc., 2002, 124, 10640.
17 S. Rehm, V. Stepanenko, X. Zhang, T. H. Rehm and 44 A. P. de Silva, T. S. Moody and G. D. Wright, Analyst, 2009,
F. Wurthner, Chem. – Eur. J., 2010, 16, 3372. 134, 2385.
18 H. Langhals, W. Jona, F. Einsiedl and S. Wohnlich, Adv. 45 R. A. Bissell, A. P. De Silva, H. Q. N. Gunaratne,
Mater., 1998, 10, 1022.
P.
L.
M.
Lynch,
G.
E.
M.
Maguire
and
19 Y. Liu, K. R. Wang, D. S. Guo and B. P. Jiang, Adv. Funct.
Mater., 2009, 19, 2230.
K. R. A. S. Sandanayake, Chem. Soc. Rev., 1992, 21, 187.
46 A. P. de Silva, J. Phys. Chem. Lett., 2011, 2, 2865.
20 T. Heek, C. Fasting, C. Rest, X. Zhang, F. Wurthner and 47 R. A. Bissell, A. P. de Silva, H. Q. N. Gunaratne,
R. Haag, Chem. Commun., 2010, 46, 1884.
21 B. X. Gao, H. X. Li, H. M. Liu, L. C. Zhang, Q. Q. Bai and
X. W. Ba, Chem. Commun., 2011, 47, 3894.
22 U. Hahn, S. Engmann, C. Oelsner, C. Ehli, D. M. Guldi and
T. Torres, J. Am. Chem. Soc., 2010, 132, 6392.
P. L. M. Lynch, G. E. M. Maguire, C. P. McCoy and
K. R. A. S. Sandanayake, Top. Curr. Chem., 1993, 168, 223.
48 L. M. Daffy, A. P. de Silva, H. Q. N. Gunaratne, C. Huber,
P. L. M. Lynch, T. Werner and O. S. Wolfbeis, Chem. – Eur.
J., 1998, 4, 1810.
23 K.-R. Wang, H.-W. An, F. Qian, Y.-Q. Wang, J.-C. Zhang and 49 N. I. Georgiev, A. R. Sakr and V. B. Bojinov, Dyes Pigm.,
X.-L. Li, RSC Adv., 2013, 3, 23190. 2011, 91, 332.
24 M. A. Abdalla, J. Bayer, J. O. Radler and K. Mullen, Angew. 50 H. N. Lee, Z. C. Xu, S. K. Kim, K. M. K. Swamy, Y. Kim,
Chem., Int. Ed., 2004, 43, 3967. S. J. Kim and J. Yoon, J. Am. Chem. Soc., 2007, 129, 3828.
25 J. K. Gallaher, E. J. Aitken, R. A. Keyzers and 51 H. Tian, J. Gan, K. C. Chen, J. He, Q. L. Song and
J. M. Hodgkiss, Chem. Commun., 2012, 48, 7961. X. Y. Hou, J. Mater. Chem., 2002, 12, 1262.
26 J. Schonamsgruber, B. Schade, R. Kirschbaum, J. Li, 52 X. R. He, H. B. Liu, Y. L. Li, S. Wang, Y. J. Li, N. Wang,
W. Bauer, C. Bottcher, T. Drewello and A. Hirsch,
J. C. Xiao, X. H. Xu and D. B. Zhu, Adv. Mater., 2005, 17,
Eur. J. Org. Chem., 2012, 6179.
2811.
27 C. D. Schmidt, C. Boettcher and A. Hirsch, Eur. J. Org. 53 X. J. Liu, N. Zhang, J. Zhou, T. J. Chang, C. L. Fang and
Chem., 2007, 5497. D. H. Shangguan, Analyst, 2013, 138, 901.
28 C. D. Schmidt, C. Bottcher and A. Hirsch, Eur. J. Org. 54 K. Peneva, G. Mihov, A. Herrmann, N. Zarrabi, M. Borsch,
Chem., 2009, 5337.
T. M. Duncan and K. Mullen, J. Am. Chem. Soc., 2008, 130,
5398.
55 K. Qvortrup, A. D. Bond, A. Nielsen, C. J. McKenzie,
K. Kilsa and M. B. Nielsen, Chem. Commun., 2008, 1986.
29 C. Backes, C. D. Schmidt, K. Rosenlehner, F. Hauke,
J. N. Coleman and A. Hirsch, Adv. Mater., 2010, 22, 788.
30 J. M. Englert, J. Rohrl, C. D. Schmidt, R. Graupner,
M. Hundhausen, F. Hauke and A. Hirsch, Adv. Mater., 56 H. X. Wang, D. L. Wang, Q. Wang, X. Y. Li and
2009, 21, 4265. C. A. Schalley, Org. Biomol. Chem., 2010, 8, 1017.
31 N. V. Kozhemyakina, J. M. Englert, G. A. Yang, E. Spiecker, 57 M. T. Vagnini, A. L. Smeigh, J. D. Blakemore, S. W. Eaton,
C. D. Schmidt, F. Hauke and A. Hirsch, Adv. Mater., 2010,
22, 5483.
32 H. Yang, Y. Hernandez, A. Schlierf, A. Felten, A. Eckmann,
N. D. Schley, F. D’Souza, R. H. Crabtree, G. W. Brudvig,
D. T. Co and M. R. Wasielewski, Proc. Natl. Acad. Sci.
U. S. A., 2012, 109, 15651.
S. Johal, P. Louette, J. J. Pireaux, X. Feng, K. Mullen, 58 B. Roy, T. Noguchi, D. Yoshihara, Y. Tsuchiya, A. Dawn and
V. Palermo and C. Casiraghi, Carbon, 2013, 53, 357. S. Shinkai, Org. Biomol. Chem., 2013, 12, 561.
33 D. Magde, R. Wong and P. G. Seybold, Photochem. Photo- 59 Y. Che, X. M. Yang and L. Zang, Chem. Commun., 2008,
biol., 2002, 75, 327. 1413.
34 C. Huang, S. Barlow and S. R. Marder, J. Org. Chem., 2011, 60 V. Stepanenko, M. Stocker, P. Muller, M. Buchner and
76, 2386. F. Wurthner, J. Mater. Chem., 2009, 19, 6816.
35 A. D. Q. Li, W. Wang and L. Q. Wang, Chem. – Eur. J., 2003, 61 V. Stepanenko and F. Wurthner, Small, 2008, 4, 2158.
9, 4594.
62 F. Wurthner and A. Sautter, Org. Biomol. Chem., 2003, 1,
36 X. W. Yu, C. L. Zhan, X. L. Ding, S. L. Zhang, X. Zhang,
240.
H. Y. Liu, L. L. Chen, Y. S. Wu, H. B. Fu, S. G. He, Y. Huang 63 F. Wurthner, C. C. You and C. R. Saha-Moller, Chem. Soc.
and J. N. Yao, Phys. Chem. Chem. Phys., 2013, 15, 11960.
37 A. Bossi and P. G. Righetti, Electrophoresis, 1997, 18, 2012.
38 D. Gorl, X. Zhang and F. Wurthner, Angew. Chem., Int. Ed.,
2012, 51, 6328.
39 W. E. Ford, J. Photochem., 1987, 37, 189.
40 R. Greenwood and K. Kendall, J. Eur. Ceram. Soc., 1999, 19,
479.
Rev., 2004, 33, 133.
64 C. C. You and F. Wurthner, J. Am. Chem. Soc., 2003, 125,
9716.
65 P. J. Stang and B. Olenyuk, Acc. Chem. Res., 1997, 30, 502.
66 C. Piguet, M. Borkovec, J. Hamacek and K. Zeckert, Coord.
Chem. Rev., 2005, 249, 705.
67 L. Brunsveld, B. J. B. Folmer, E. W. Meijer and
R. P. Sijbesma, Chem. Rev., 2001, 101, 4071.
41 H. Langhals and W. Jona, Chem. – Eur. J., 1998, 4, 2110.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 7045–7058 | 7057

View Article Online
Paper
Organic & Biomolecular Chemistry
68 L. Fabbrizzi, M. Licchelli, P. Pallavicini, D. Sacchi and 79 E. E. Neuteboom, S. C. J. Meskers, E. W. Meijer and
A. Taglietti, Analyst, 1996, 121, 1763. R. A. J. Janssen, Macromol. Chem. Phys., 2004, 205, 217.
69 G. DeSantis, L. Fabbrizzi, M. Licchelli, A. Poggi and 80 Z. J. Chen, V. Stepanenko, V. Dehm, P. Prins,
A. Taglietti, Angew. Chem., Int. Ed., 1996, 35, 202.
70 F. Würthner, C. Thalacker, S. Diele and C. Tschierske,
Chem. – Eur. J., 2001, 7, 2245.
71 A. Arnaud, J. Belleney, F. Boue, L. Bouteiller, G. Carrot and
W. Wintgens, Angew. Chem., Int. Ed., 2004, 43, 1718.
72 K. Balakrishnan, A. Datar, T. Naddo, J. L. Huang, R. Oitker,
L. D. A. Siebbeles, J. Seibt, P. Marquetand, V. Engel and
F. Würthner, Chem. – Eur. J., 2007, 13, 436.
81 T. Heek, F. Würthner and R. Haag, Chem. – Eur. J., 2013,
19, 10911.
82 A. Datar, K. Balakrishnan and L. Zang, Chem. Commun.,
2013, 49, 6894.
M. Yen, J. C. Zhao and L. Zang, J. Am. Chem. Soc., 2006, 83 M. R. Harpham, N. E. Levinger and B. M. Ladanyi, J. Phys.
128, 7390. Chem. B, 2008, 112, 283.
73 F. Wurthner, Z. Chen, V. Dehm and V. Stepanenko, Chem. 84 D. B. Wong, K. P. Sokolowsky, M. I. El-Barghouthi,
Commun., 2006, 1188.
74 B. Bodenant, F. Fages and M. H. Delville, J. Am. Chem. Soc.,
1998, 120, 7511.
75 S. Yagai, Y. Monma, N. Kawauchi, T. Karatsu and
A. Kitamura, Org. Lett., 2007, 9, 1137.
E. E. Fenn, C. H. Giammanco, A. L. Sturlaugson and
M. D. Fayer, J. Phys. Chem. B, 2012, 116, 5479.
85 J. Catalan, C. Diaz and F. Garcia-Blanco, J. Org. Chem.,
2001, 66, 5846.
86 A. Luzar, J. Mol. Liq., 1990, 46, 221.
76 W. Wang, J. J. Han, L. Q. Wang, L. S. Li, W. J. Shaw and 87 B. Kirchner and M. Reiher, J. Am. Chem. Soc., 2002, 124,
A. D. Q. Li, Nano Lett., 2003, 3, 455. 6206.
77 W. Wang, L. S. Li, G. Helms, H. H. Zhou and A. D. Q. Li, 88 B. K. Miremadi and S. R. Morrison, J. Appl. Phys., 1988, 63,
J. Am. Chem. Soc., 2003, 125, 1120. 4970.
78 W. Wang, W. Wan, H. H. Zhou, S. Q. Niu and A. D. Q. Li, 89 L. A. Xue, N. Ranjan and D. P. Arya, Biochem., 2011, 50,
J. Am. Chem. Soc., 2003, 125, 5248.
2838.
7058 | Org. Biomol. Chem., 2014, 12, 7045–7058
This journal is © The Royal Society of Chemistry 2014