Self-sorted assembly in a mixture of donor and acceptor chromophores

Article abstract of DOI:10.1002/chem.201000596

A simple and novel supramolecular approach for orthogonal self-assembly of donor and acceptor chromophores has been demonstrated. Suitably designed 1,5-dialkoxynaphthalene (DAN) and naphthalene tetracarboxylic acid diimide (NDI) derivatives were used as t

Full text of DOI:10.1002/chem.201000596

DOI: 10.1002/chem.201000596
Self-Sorted Assembly in a Mixture of Donor and Acceptor Chromophores
Mijanur Rahaman Molla, Anindita Das, and Suhrit Ghosh*[a]
Abstract: A simple and novel supra-
molecular approach for orthogonal
self-assembly of donor and acceptor
chromophores has been demonstrated.
Suitably designed 1,5-dialkoxynaphtha-
lene (DAN) and naphthalene tetracar-
boxylic acid diimide (NDI) derivatives
were used as the donor and acceptor
systems, respectively. The molecular
design for self-sorting relies upon the
precise control over the placement of
the self-complementary hydrogen-
bonding units (amide functionality)
with respect to the individual chromo-
phore. By design, the distances be-
tween the two amide groups in the
donor and acceptor chromophores are
not identical, and consequently the
effect of the hydrogen-bonding interac-
tion cannot be maximised in the case
of alternate donor–acceptor-type p-
to be sacrificed, and segregated assem-
bly among the individual chromo-
phores should be enforced by virtue of
the much stronger effects of hydrogen
bonding and p–p stacking. Detailed
spectroscopic studies were carried out
to probe the mode of self-assembly in
various derivatives of the DAN–NDI
donor–acceptor pairs to establish the
utility of the molecular design as a gen-
eralised one for orthogonal self-assem-
bly.
stacking. Thus
a
relatively weak
charge-transfer interaction is expected
Keywords: chromophores · charge
transfer · donor–acceptor systems ·
self-assembly · self-sorting
Introduction
More recently, Shinkai and co-workers have reported self-
sorting of bis-amide-functionalised perylene bisimide (PBI)
and tetraamide-functionalised oligothiophene (OT) chromo-
phores in solution[12b] by using the inherent difference in the
self-assembly propensities of the two chromophores, which
arises from the difference in the number of peripheral
amide functionalities involved in hydrogen-bonding interac-
tions. The drawback of this approach is that the equilibrium
constant for self-assembly of either the donor or acceptor
chromophore must be significantly weaker than for the
other one. Further, the shape and size of the PBI and OT
chromophores are very different, and thus it is possible,
even in the absence of hydrogen bonding, that they form or-
thogonal self-assembled structures under suitable condition,
as demonstrated by Meijer and co-workers in the case of
functionalised PBI and oligo(p-phenylenevinylene) chromo-
phores.[12c]
Electron-rich (donor) and -deficient (acceptor) aromatic p
systems often form charge-transfer (CT) complexes,[1] which
have been extensively utilised to generate many elegant
supramolecular structures, such as organogels,[2] synthetic
ion channels,[3] rotaxanes,[4] catenanes,[5] folded oligomers[6]
and polymers,[7] liquid crystalline materials[8] and dendritic
assembly.[9] However, little is known regarding the self-sort-
ing[10] of individual chromophores in the donor–acceptor
mixture, which is highly relevant in the context of organic
electronic device applications, such as in bulk heterojunction
solar cell, for efficient charge-separation and charge-carrier
mobility.[11] In the recent past, Venkataraman and co-work-
ers have demonstrated the segregated assembly of donor
and acceptor p systems based on mutually phobic aliphatic
fluorocarbon–hydrocarbon interactions in the solid state.[12a]
Herein, we introduce a novel and versatile strategy for
self-sorted assembly in a mixture of donor and acceptor
chromophores in solution, based on specific directional
supramolecular interactions. The design is based on linking
the individual chromophores with two amide functionalities
in such a manner so that the distances between the two
amide functionality are different in the donor and acceptor
systems. We envisaged that such a molecular design will
ensure that in case of alternate donor–acceptor type p stack-
[a] M. R. Molla, A. Das, Dr. S. Ghosh
Polymer Science Unit
Indian Association for the Cultivation of Science
2A & B Raja S. C. Mullick Road
Kolkata, 700 032 (India)
Fax : (+91)33-2473-2805
E-mail: psusg2@iacs.res.in
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000596.
10084
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10084 – 10093

FULL PAPER
ing, intermolecular hydrogen-bonding interactions between
the amide groups cannot operate simultaneously because of
geometrical constraint. On the other hand it is also expected
that given a choice, the mode of self-assembly will be dictat-
ed by the relatively stronger hydrogen-bonding interaction
rather than weak CT interactions. Thus in the process of
self-assembly the donor and acceptor chromophores would
sacrifice the weak CT interaction to gain the synergistic
effect of p–p stacking and hydrogen bonding, which can be
achieved only in the case of segregated assembly
(Scheme 1).
To test this hypothesis we studied the self-assembly of 1,5-
dialkoxynaphthalene (DAN) derivative D1 and naphtha-
lene-1,4,5,8-tetracarboxylic acid diimide (NDI) derivative
A1 (Scheme 1) as donor and acceptor chromophores, re-
spectively. The peripheral trioctyloxybenzamide groups were
incorporated to improve the solubility of A1 and D1 in non-
polar solvents, in which the effect of the hydrogen-bonding
interaction should be realised to a greater extent. We chose
the DAN–NDI donor–acceptor pair because formation of
their co-assembly aided by a CT interaction is well-estab-
lished[3–6] and if they behave in an opposite manner in our
system it will strongly justify the utility of the present ap-
proach for self-sorting of donor–acceptor systems in solu-
tion.
procedure.[10e] Diamine-functionalised donor intermediate 8
was synthesised from 1,5-dialkoxynaphthalene in a few stan-
dard synthetic steps and was reacted with compound 4 to
give D1 in 68% isolated yield. Compound D2 was synthes-
ised in a single step by alkylation of 1,5-dialkoxynaphtha-
lene with dodecylbromide and isolated in 65% yield. For
the synthesis of A1, acid chloride 4 was reacted with excess
ethylenediamine to produce intermediate compound 9,
which was reacted with naphthalene tetracarboxylic acid di-
AHCTUNGTREGaNNNU nhydride to generate acceptor compound A1 in 70% yield.
For the synthesis of A2, a slightly different procedure was
followed in which 3,4,5-trioctyloxymethylbenzoate was di-
rectly reacted with excess hydrazine hydrate to generate the
intermediate compound 10, which was then coupled with
naphthalene tetracarboxylic acid dianhydride to produce A2
in 81% yield. Compound A3 was synthesised by reaction of
naphthalene tetracarboxylic acid dianhydride with dodecyl-
AHCTUNGTRENNUNG
pendent variations in their UV/Vis absorption spectra
(Figure 1). Chloroform is known to be benign for rigid p
systems,[10e,12c] whereas in the non-polar solvent methylcyclo-
hexane (MCH), both p–p stacking and hydrogen-bonding
interaction should be more favourable. The absorption spec-
trum of A1 in CHCl3 is compared with that in MCH/CHCl3
(95:5; v/v)[13] in Figure 1a. In CHCl3, well-resolved bands
can be seen in the range of l=300 to 400 nm due to a p–p*
transition polarised along the long axis of the monomeric
NDI chromophores.[14] In MCH/CHCl3 (95:5), a decrease in
the absorbance intensity (ꢀ60%) was observed for all the
major bands along with significant broadening of the spec-
trum which clearly suggests p–p stacking among the NDI
chromophores.[15] Similarly, a distinct hypochromic shift was
Results and Discussion
Synthetic routes for various donor and acceptor chromo-
phores are depicted in Scheme 2. 3,4,5-Trioctyloxybenzoyl-
chloride (4) was synthesised from commercially available
3,4,5-trihydroxymethylbenzoate by following a literature
Scheme 1. Top: Schematic representation of the proposed self-sorted assembly of donor and acceptor chromophores; bottom: the structures of the vari-
ous molecules studied.
Chem. Eur. J. 2010, 16, 10084 – 10093
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10085

S. Ghosh et al.
Scheme 2. Synthetic routes for various donor and acceptor chromophores. Reagents and conditions: a) n-octyl bromide, K2CO3, DMF, 808C, 88%;
b) KOH, EtOH/H2O, 1008C, 81%; c) SOCl2, CH2Cl2, RT; d) 1,2-dibromoethane (excess), CH3CN, K2CO3, 26%; e) NaN3, DMF, 1008C, 93%; f) H2
(55 psi), 10% Pd/C, EtOAc, 95%; g) Et3N, CH2Cl2, 08C!RT, 68 %; h) CH2Cl2, 08C!RT, 56%; i) DMF, 1408C, 70%; j) hydrazine hydrate, MeOH/THF,
708C, 96%; k) DMF, 1408C, 81%; l) KOH, TBAB, H2O, 95 8C, 65%; m) DMF, 1408C, 71%.
observed for all the major bands of D1 in the UV/Vis spec-
trum in MCH/CHCl3 (95:5) compared with that in pure
CHCl3 (Figure 1b) due to self-assembly.[16]
To elucidate the role of hydrogen bonding in self-assem-
bly as depicted in Scheme 1, we examined the effect of
MeOH, a protic hydrogen-bonding-competing solvent, on
the spectral properties of A1 and D1.[17] The spectral varia-
tions are shown in Figure 2. It can be seen that with gradual
addition of MeOH to a solution of either A1 (Figure 2a) or
D1 (Figure 2c) in MCH/CHCl3 (95:5) the UV/Vis spectrum
changed drastically for both chromophores, and after the ad-
dition of a certain amount of MeOH the spectra almost re-
sembled that of the respective monomeric chromophores.
Such a dramatic effect of MeOH on disassembly clearly
demonstrates the strong influence of hydrogen-bonding in-
teractions on the self-assembly process. In Figure 2b, the
variation in the absorbance of A1 at l=382 nm is plotted
against amount of MeOH and it can be seen that as an in-
creasing amount of MeOH is added, the absorbance gradu-
ally increases and almost reaches saturation at ꢀ5% MeOH
(v/v). Similarly, for D1 the change in absorbance at l=
312 nm is plotted as a function of MeOH in Figure 2d; in
this case the saturation was achieved after the addition of
only ꢀ1% (v/v) MeOH, probably because of the reduced
Figure 1. UV/Vis spectra of a) A1 and b) D1 in CHCl3 (c) and MCH/
CHCl3 (95:5; a). Concentration of chromophore=0.1 mm and T=
stability of the aggregates[6e] in this case as compared with
258C.
the A1 assembly.
10086
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10084 – 10093

Self-Sorted Chromophore Assembly
FULL PAPER
Figure 2. Variation in the UV/Vis spectra of a) A1 and c) D1 in MCH/CHCl3 (95:5) as a result of gradual addition of MeOH; arrows indicate the spectral
changes with increasing amount of MeOH; initial concentration of chromophore=0.1 mm, volume of the solution=1.8 mL, path length of the cuvette=
1 cm, T=258C. Variation in the absorbance at b) l=382 nm in Figure 2a and d) l=312 nm in Figure 2c as a function of volume of added MeOH.
Then we examined the spectral behaviour of the equimo-
lar mixture of A1 and D1 in MCH/CHCl3 (95:5). It can be
seen in Figure 3a that the spectrum of the mixture of A1
and D1 is almost identical to that generated by mathemati-
cal summation of the spectra of the individual chromo-
phores, which suggests that the presence of D1 or A1 has no
influence on the self-assembly of A1 or D1, respectively.
More significantly, the spectrum of the mixture does not ex-
hibit any CT absorption band (Figure 3b) in the range of l=
450 to 600 nm, which eliminates the possibility of donor–ac-
ceptor-type p-stacking between NDI and DAN chromo-
phores.[3–6] These observations unambiguously support the
self-sorted assembly of individual chromophores as depicted
in Scheme 1.
To gain further insight into the self-assembly process, we
carried out variable-temperature UV/Vis absorption experi-
ments of A1, D1 and an equimolar mixture of A1 and D1 in
MCH/CHCl3 (95:5). The spectral variations of A1 and D1
are shown in Figure 4a and b, respectively. It can be seen
that in both cases the spectral features gradually change
from aggregated to monomeric chromophores with increas-
ing temperature, as expected for reversible self-assembly
process. Further, we carried out variable-temperature ex-
periments with a 1:1 mixture of A1 and D1 and found signif-
icant hyperchromic shifts of all the major bands with in-
creasing temperature (Figure 4c) as expected for the aggre-
gate to monomer conversion. From the temperature-depen-
dent absorption spectra, the stabilities of the p-stacked as-
semblies of A1 and D1 were estimated by calculating the
respective aagg(T) (mole fraction of aggregate at tempera-
ture T) values by using Equation (1):[10e]
AðTÞ ꢁ Amon
aaggðTÞ ꢀ
ð1Þ
Aagg ꢁ Amon
in which Aagg, Amon and A(T) are the absorbance at a partic-
ular wavelength in the UV/Vis spectra for the fully aggre-
gated, monomeric and in-between state at temperature T,
respectively. The calculations were first carried out for the
spectral data obtained from the experiments with individual
chromophores (Figure 4a and b). For both A1 and D1, the
UV/Vis spectra in MCH/CHCl3 (95:5) at 258C and in CHCl3
were considered to reflect the fully aggregated state and the
monomeric state, respectively. The intensity of the bands at
l=382 and 312 nm were considered for the calculation of
the aagg(T) value of A1 and D1, respectively. The aagg values
estimated from these calculations are plotted as a function
of temperature in Figure 2d, and from these plots the values
of a50(T) (the temperature at which aagg =0.5) were found to
Chem. Eur. J. 2010, 16, 10084 – 10093
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10087

S. Ghosh et al.
be 64 and 318C for A1 and D1, respectively. Such a low
a50(T) value for D1 compared with A1 is attributed to the
lower stability of the donor–donor p-stacked assembly due
to electrostatic repulsion, and is consistent with the litera-
ture report.[6e] Independently, the aagg values for A1 and D1
were calculated from the spectral variations of their mixed
solution (Figure 4c) by monitoring the absorbance at l=382
and 312 nm,[19] respectively, and plotted in Figure 2d. Re-
markably, for both A1 and D1 the plot of aagg versus T
based on the data obtained from the individual experiments
and the mixed experiment looks almost identical (Fig-
ure 4d). This striking similarity of the temperature-depen-
dent aggregate-to-monomer conversion profile provides
strong evidence for the proposed self-sorted assembly of in-
dividual chromophores in the donor–acceptor mixed system.
So far we have discussed self-sorting of D1 and A1 from
an equimolar mixture only. We also wanted to examine the
effect of the stoichiometric imbalance of the two chromo-
phores on their self-assembly. Therefore, we studied the
UV/Vis absorption spectra of D1+A1 in MCH/CHCl3
(95:5) at different relative molar ratios of the two chromo-
phores, with the total concentration kept constant (Fig-
ure 5a). The absorbance of the mixed spectra at l=382 nm
is solely due to A1 and is not affected by the absorbance of
the D1, but it was not possible to identify such a separate
band for D1 in the mixed spectra. Therefore, to get the
spectral contribution of D1 alone (inset Figure 5a), the spec-
trum of A1 was subtracted
Figure 3. a) UV/Vis spectrum and b) CT region of the spectra of A1+D1
(1:1) in MCH/CHCl3 (95:5; c) and the mathematical sum of the indi-
vidual spectrum (a). Concentration of chromophore=0.1 mm in each
measurements and T=258C.
from the corresponding mixed
spectrum in each mixture by
using Equation (2):
SD1 ¼ SmixedꢁðaA1 ꢂ SA1
Þ
ð2Þ
in which SD1, Smixed, SA1 and aA1
are the modified donor spec-
trum, original spectrum of the
mixture, spectrum of A1 only
and mole fraction of A1 in the
mixture, respectively. The ab-
sorbance at l=382 nm (for A1)
in the mixed spectra and that at
l=325 nm (for D1) in the sub-
tracted spectra were plotted as
a function of mole fraction of
D1 in Figure 5b. If there were a
CT interaction between the
donor–acceptor chromophores
at higher stoichiometric imbal-
ance, a deviation from linearity
would have been observed.
However, that possibility could
be completely eliminated be-
cause at least in the range of
D1/A1=1:4 to D1/A1=4:1
both the plots behave linearly
(see Figure 5b).
Figure 4. Temperature-dependent UV/Vis absorption spectra of a) A1 (T=25–658C), b) D1 (T=25–508C) and
c) a 1:1 mixture of A1+D1 (T=25–658C) in MCH/CHCl3 (95:5) at a concentration of 0.1 mm (individual
chromophores); arrows indicate spectral changes upon increasing temperature. d) Variation in aagg as a func-
~
~
*
*
tion of temperature; : A1 only, : A1 in mixture, : D1 only, : D1 in mixture.
10088
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10084 – 10093

Self-Sorted Chromophore Assembly
FULL PAPER
the crystal structure of NDI.[21] For the D1 assembly, the
chromophores were packed with significant rotational dis-
placement along the long axis and the interplanar perpen-
dicular distance between the two p planes was found to be
3.88 ꢁ, which suggests weak p–p interactions between the
DAN chromophores. To further support our hypothesis re-
garding the basis of the observed self-sorting, we wanted to
identify suitable alternate derivatives of the same donor–ac-
ceptor pair as controlled molecules in which the relative dis-
tances between the two amide functionalities in donor and
acceptor chromophores are mutually adjusted so that inter-
molecular hydrogen bonding is feasible in the case of alter-
nate donor–acceptor p stacking. To predict the molecular
structure, we carried out energy-minimisation calculations
with different DAN and NDI derivatives, in which the
donor was fixed as D1 and the acceptor was modified from
A1 by changing only the number of methylene groups be-
tween the NDI chromophore and the trioctyloxybenzamide
groups. We found that in case of alternately stacked D1 and
A2 (Scheme 1), the distance among the amide hydrogen and
carbonyl oxygen of the amide groups between the neigh-
bouring chromophores was 2.04 ꢁ (Figure 6c), which is well
below the limit of hydrogen-bonding interaction distance
(2.45 ꢁ). If this prediction is indeed true, one should expect
an alternate donor–acceptor-type assembly in the A2+D1
mixture due to the synergistic effect of CT and hydrogen-
bonding interactions, unlike the self-sorting in the A1+D1
system. To check this, we synthesised A2 (Scheme 1). Self-
assembly of A2 alone was examined by comparing its ab-
sorption spectrum in MCH/CHCl3 (95:5) with that in CHCl3
and p-stacking was evident of A2 from spectral changes
that were similar to those of A1 (Figure 7a). It is worth
noting that in case of A2, the functional group should be
called hydrazide instead of amide, although the NHCO
moiety responsible for hydrogen-bonding remains same. To
confirm that the self-assembly in A2 is also primarily driven
by hydrogen bonding, we examined the effect of MeOH on
the spectrum of the self-assembled chromophore. In Fig-
ure 7b, it can be seen that the aggregated spectrum clearly
changes into the monomeric spectrum on gradual addition
of MeOH, and thus the involvement of hydrogen bonding
on the self-assembly is supported. We then examined the
nature of the co-assembly between A2 and D1. When A2
was mixed with D1 in MCH/CHCl3 (95:5), an orange-col-
oured solution formed sponta-
Figure 5. a) UV/Vis absorption spectra of A1+D1 at various ratios in
MCH/CHCl3 (95:5); inset: the subtracted spectra for D1 only, as ob-
tained by using Equation (2). b) Plot of absorbance as a function of the
mole fraction of D1. Total chromophore concentration was 0.1 mm.
To understand the packing of the chromophores in the p-
stacked assemblies, we carried out molecular modelling
studies for generating energy-minimised structures of A1
and D1 (Figure 6a and b, respectively).[20] The distance be-
tween the amide hydrogen atom of a particular chromo-
phore and the carbonyl oxygen of the immediately neigh-
bouring chromophore (indicated as red dashed lines in
Figure 6) was found to be 2.01 and 2.05 ꢁ for A1 and D1,
respectively, which is well within the limit of the hydrogen-
bonding distance. The p–p stacking distance between two
NDI chromophores in the A1 assembly (Figure 6a) was
found to be 3.41 ꢁ, which exactly matches that observed in
neously
(Figure 8a),
which
clearly suggests an alternate
donor–acceptor-type p-stacking
system. As expected from a cur-
sory observation, a broad CT
band appeared at l=550 nm in
the UV/Vis spectrum (Fig-
ure 8b, inset) owing to the
donor–acceptor
interaction
among DAN and NDI chromo-
phores.[6] It is worth noting that
no orange colouration (Fig-
Figure 6. Packing model of a) A1, b) D1 and c) A2+D1.
Chem. Eur. J. 2010, 16, 10084 – 10093
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10089

S. Ghosh et al.
spectra of A2+D1 with that of A1+D1 at a significantly
higher concentration (0.5 mm), and even there found the ap-
pearance of an intense CT band only in the former case but
not for the latter mixture (Figure 9a). To ascertain that hy-
drogen bonding is indeed involved in the donor–acceptor-
type self-assembly, we carried out the MeOH addition ex-
Figure 7. a) UV/Vis spectra of A2 (concentration=0.1 mm) in CHCl3
(c) and MCH/CHCl3 (95:5; a) at 258C; b) Effect of MeOH addi-
tion (0–3% v/v) on the UV/Vis spectra of A2 in MCH/CHCl3 (95:5);
arrows indicate spectral changes with increasing amounts of MeOH.
ure 8a) and no CT band were visible in the UV/Vis spec-
trum of the self-sorting mixture, that is, A1+D1 (Figure 3b).
Furthermore, the UV/Vis spectrum of A2+D1, unlike A1+
D1, does not match with the calculated spectrum generated
by mathematical summation of the individual spectrum (Fig-
ure 8b). We also compared the CT region of the UV/Vis
Figure 9. a) CT region of the UV/Vis spectra of equimolar mixtures of
A3+D2 (a) A1+D1 (g) and A2+D1 (c) in MCH/CHCl3
(95:5); total chromophore concentration=0.5 mm and T=258C. b) Effect
of MeOH addition (0–5%) on the CT band of A2+D1 (1:1) in MCH/
CHCl3 (95:5); arrows indicate spectral changes with increasing MeOH.
Figure 8. a) Photograph of A1+D1 (1:1; left) and A2+D1 (1:1; right) in MCH/CHCl3 (95:5); total chromophore concentration=0.5 mm. b) UV/Vis
spectrum of A1+D1 (1:1) in MCH/CHCl3 (95:5; c) and the mathematical sum of the individual spectra (a); inset: the CT region of the spectra;
concentration of each chromophore=0.1 mm and T=258C.
10090
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10084 – 10093

Self-Sorted Chromophore Assembly
FULL PAPER
librate at RT for 2 h before spectral measurements. For variable-tempera-
ture experiments, a solution (2 mL) of a particular chromophore in
MCH/CHCl3 (95:5) was heated and allowed to equilibrate for 15 min
equilibrium at the desired temperature before each measurement.
periment and found that the intensity of the CT-band de-
creased rapidly with increasing amounts of MeOH (Fig-
ure 9b). However, the CT interaction is known to be inher-
ently weak in polar media,[1] and thus the observed effect of
MeOH might not be only due to the breaking of hydrogen
bonding. To unambiguously demonstrate the role of hydro-
gen bonding, we synthesised two other controlled molecules
A3 and D2 (Scheme 1) that were devoid of any amide func-
tionality. In the UV/Vis spectra of A3+D2 (1:1), no CT
band (Figure 9a) and no indication of p stacking (Figure S2
in the Supporting Information) was observed, which sug-
gests that hydrogen bonding is essential for self-assembly in
the present system, at least in the concentration range at
which the experiments were performed (0.1–0.5 mm).
Synthesis and characterisation: Synthesis of various compounds has been
achieved by following procedures reported in the literature.[10e,15b,25] New
1
compounds have been characterised by using H NMR and UV/Vis spec-
troscopies, melting point (if solid), MS (ESI) and elemental analysis,
whereas previously reported compounds have been characterised by
1H NMR and UV/Vis spectroscopies and melting point analysis (if solid).
Synthetic procedures for the new and final compounds are described
here and rest are included in the Supporting Information.
1,5-Bis(2-azidoethoxy)naphthalene
(7):
Compound
6
(2.762 g,
7.38 mmol,) and NaN3 (9.599 g, 147.68 mmol) were dissolved in dry DMF
(45 mL), and the reaction mixture was stirred at 1008C for 12 h under an
N2 atmosphere. The reaction was stopped, cooled to RT and poured into
ice-cold H2O (100 mL) to give an off-white solid, which was filtered. The
obtained solid was washed several times with H2O and dried under
vacuum to afford the desired product (2.05 g, 93%). M.p. 117–1208C;
1H NMR (300 MHz, CDCl3, TMS): d=7.907 (d, J=8.7 Hz, 2H), 7.393 (t,
J=8.1 Hz, 2H), 6.851 (d, J=7.8 Hz, 2H), 4.327 (t, J=5.1 Hz, 4H),
3.728 ppm (t, J=5.1 Hz, 4H); FTIR (KBr): n˜ =2104.19 cmꢁ1 for azide
stretching; UV/Vis (CH2Cl2): lmax (e)=326 (5150), 312 (6862), 295
(8976), 285 nm (7074mꢁ1 cmꢁ1); MS (ESI): m/z calcd for C14H17N6O2:
301.14 [M+3H]+; found: 301.02; elemental analysis calcd (%) for
C14H14N6O2: C 56.37, H 4.73, N 28.17; found: C 56.42, H 4.87, N 28.26.
Conclusion
In conclusion, we have demonstrated a simple, novel and
versatile supramolecular approach for the self-sorted assem-
bly of donor and acceptor chromophores in solution. The
design is based on molecular-level control over the place-
ment of self-complementary hydrogen-bonding units (amide
functionality) with respect to the individual chromophore.
The distances between the two amide groups in the donor
and acceptor chromophores are not identical and conse-
quently the number of hydrogen-bonding interactions
cannot be maximised in the case of alternate donor–accept-
or-type p stacking. Thus by virtue of the much stronger hy-
drogen-bonding and p–p-stacking effects, the relatively
weak CT interaction is sacrificed and self-sorting among the
individual chromophores could be enforced. In a controlled
experiment in which the distance between the two amide
bonds was almost identical for two different types of chro-
mophores, alternate donor–acceptor-type p stacking was ob-
served. We believe the findings of this work and its rele-
vance to various other interdisciplinary research fields will
significantly add to the ongoing effort to correlate the supra-
molecular organisation of various functional p systems to
their macroscopic properties.[22] We are now testing the
scope of this approach for specific supramolecular organisa-
tions of n- and p-type semiconductors and studying the
charge-transport properties of self-sorted blends.
2-[1-(2-Aminoethoxy)naphthalen-5-yloxy)ethanamine] (8): 10% Pd/C
catalyst (200 mg) was added to a solution of compound 7 (2.014 g,
6.75 mmol) in EtOAc (45 mL), and the reaction mixture was mechanical-
ly stirred at RT under H2 (55 psi) for 5 h. The reaction was stopped, the
catalyst was filtered through Celite, and the filtrate was concentrated to
afford the crude product as an off-white solid (1.58 g, 95%). M.p. 83–
858C; 1H NMR (300 MHz, CDCl3, TMS): d=7.862 (d, J=8.5 Hz, 2H),
7.365 (t, J=8.0 Hz, 2H), 6.856 (d, J=7.5 Hz, 2H), 4.168 (t, J=5.0 Hz,
4H), 3.221 ppm (t, J=5.0 Hz, 4H); FTIR (KBr): n˜ =2104.19 cmꢁ1 peak
for azide is absent; UV/Vis (THF/CH2Cl2 1:1): lmax (e)=326 (5043), 312
(6539), 296 (7929), 285 nm (6103mꢁ1 cmꢁ1); MS (ESI): m/z calcd for
C14H19N2O2: 247.14 [M+H]+; found: 247.07; elemental analysis calcd (%)
for C14H18N2O2: C 68.27, H 7.37, N 11.37; found: C 68.36, H 7.38, N
11.44.
Compound D1: A solution of compound 4 (0.8 g, 1.52 mmol) in dry
CH2Cl2 (5 mL) was added to an ice-cold solution of compound 8 (0.15 g,
0.608 mmol) and Et3N (0.4 mL) in dry CH2Cl2 (10 mL). The reaction mix-
ture was then stirred in the ice bath for another 2 h, and then at RT for
12 h. The mixture was diluted with CH2Cl2 (20 mL) and washed with
H2O (3ꢂ30 mL) and brine (1ꢂ30 mL). The organic layer was dried over
anhyd Na2SO4, and the solvent was evaporated to afford the crude prod-
uct, which was purified by using column chromatography with silica gel
as the stationary phase and 5% EtOAc in CH2Cl2 as the eluent to get the
pure product as an off-white solid (5.06 g, 68%). M.p. 1108C; 1H NMR
(300 MHz, CDCl3, TMS): d=7.870 (d, J=8.5 Hz, 2H), 7.365 (t, J=8 Hz,
2H), 6.954 (s, 4H), 6.892 (d, J=7.5 Hz), 6.537 (t, J=6 Hz, 2H), 4.342 (t,
J=5 Hz, 4H), 4.031–3.947 (m, 16H), 1.827–1.692 (m, 12H), 1.465–1.266
(m, 60H), 0.875 ppm (t, J=6.5 Hz, 18H); 13C NMR (CDCl3): d=167.71,
154.17, 153.19, 141.42, 129.32, 126.74, 125.43, 114.46, 106.05, 77.06, 73.55,
69.40, 67.27, 39.69, 31.93, 30.35, 29.54, 26.11, 22.70, 14.11 ppm; UV/Vis
(THF): lmax (e)=326 (5085), 312 (7652), 295 (13677), 283 (15361),
Experimental Section
262 nm (20594mꢁ1 cmꢁ1); MS (ESI): m/z calcd for C76H123N2NaO10
:
Materials and methods: The solvents and reagents were received from
commercial sources and purified by standard methods.[23] For UV/Vis
spectroscopy experiments, spectroscopy-grade solvents were used.
1H NMR spectra were recorded by using a Bruker DPX-300 MHz NMR
spectrometer, and all the spectra were calibrated against TMS. UV/Vis
spectra were recorded by using a Perkin–Elmer Lambda 25 spectrometer
equipped with a Peltier system for variable-temperature experiments.
1246.92 [M+Na+H]+; found: 1247.20; elemental analysis calcd (%) for
C76H122N2O10: C 74.59, H 10.05, N 2.29; found: C 74.63, H 10.04, N 2.35.
Compound A1:[15b] Compound 9 (240 mg, 0.445 mmol) and 1,4,5,8-naph-
thalenetetracarboxylic bisanhydride (41 mg, 0.15 mmol) were placed in a
round-bottomed flask with dry DMF (10 mL) and the reaction mixture
was stirred for 6 h at 1408C under a N2 atmosphere. Then the solution
was allowed to cool to RT and placed in the refrigerator for 30 min to
give an orange solid, which was filtered, and the obtained solid was
washed with MeOH several times. The product was further purified by
column chromatography by using silica gel as stationary phase and
UV/Vis spectroscopic studies: Stock solutions of different chromophores
were made in CHCl3 (2 mm). An aliquot (1 mL) was taken and was
added to an appropriate amount of MCH to adjust the desired solvent
composition and final concentration. The solutions were allowed to equi-
Chem. Eur. J. 2010, 16, 10084 – 10093
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10091

S. Ghosh et al.
CHCl3 as the eluent to obtain the pure product as a yellow solid (0.14 g,
70%). M.p. 1408C; 1H NMR (300 MHz, CDCl3,TMS): d =8.75 (s, 4H),
6.91 (s, 4H), 4.55 (t, 4H), 4.02–3.89 (m, 16H), 1.84–1.71 (m, 12H), 1.30–
1.27 (m, 60H), 0.89–0.87 ppm (m, 18H); 13C NMR (CDCl3): d=166.57,
151.99, 140.09, 130.07, 127.96, 125.85, 125.55, 104.49, 76.25, 75.99, 75.74,
72.45, 68.21, 60.80, 39.22, 38.73, 32.24, 30.88, 30.10, 29.30, 28.68, 28.49,
28.36, 28.34, 28.28, 28.20, 25.04, 21.65, 13.05 ppm; UV/Vis (CH2Cl2): lmax
(e)=382 (26785), 361 (21429), 344 nm (13513mꢁ1 cmꢁ1); MS (ESI): m/z
calcd for C80H120N4NaO12: 1351.88 [M+Na]+; found: 1352.01; elemental
analysis calcd (%) for C80H120N4O12: C 72.25, H 9.10, N 4.21; found: C
72.19, H 9.16, N 4.27.
tography by using silica gel as the stationary phase and CHCl3 as the
eluent to obtain the pure product as an off-white solid (0.520 g, 72%).
M.p. 156–1588C; H NMR (500 MHz, CDCl3,TMS): d=8.75 (s, 4H), 4.19
(t, J=7.5 Hz, 4H), 1.76–1.70 (m, 4H), 1.45–1.41 (m, 4H), 1.37–1.25 (m,
32H), 0.87 ppm (t, J=7 Hz); UV/Vis (CH2Cl2): lmax (e)=382 (26138),
360 (21810), 342 nm (12678mꢁ1 cmꢁ1); MS (ESI): m/z calcd for
C38H55N2O4: 603.41 [M+H]+; found: 603.65; elemental analysis calcd
(%) for C38H54N2O4: C 75.71; H 9.03; N 4.65; found: C 75.82, H 8.97, N
4.72.
1
3,4,5-Tris
(octyloxy)benzohydrazide (10):[24] Compound
2
(615.0 mg,
1.18 mmol) and hydrazine monohydrate (2.9 mL, 59.0 mmol) were placed
in a round-bottomed flask and dissolved in MeOH (10 mL) and THF
(3 mL). The reaction mixture was stirred at 708C for 12 h. The heating
was stopped, the reaction mixture was allowed to cool to RT, the vola-
tiles were removed under reduced pressure, and the product was dis-
solved in CH2Cl2 (25 mL) and washed with H2O (3ꢂ25 mL). The organic
layer was dried over anhyd Na2SO4, and the solvent was evaporated to
get crude product, which was purified by column chromatography by
using silica gel as the stationary phase and 5% MeOH in CH2Cl2 as the
eluent to get the pure product as a colourless waxy material (0.597 g,
96%). 1H NMR (300 MHz, CDCl3, TMS): d=6.92 (s, 2H), 4.06–3.97 (m,
6H), 1.83–1.68 (m, 6H), 1.48–1.45 (m, 6H), 1.31–1.29 (m, 24H), 0.90–
0.86 ppm (m, 9H); UV/Vis (CH2Cl2): lmax (e)=266 nm (7459mꢁ1 cmꢁ1);
MS (ESI): m/z calcd for C31H56N2NaO4: 543.40 [M+Na]+; found: 543.55;
elemental analysis calcd (%) for C31H56N2O4: C 71.49, H 10.84, N 5.38;
found: C 71.44, H 10.89, N 5.43.
Acknowledgements
We thank the Department of Science and Technology (DST), New Delhi,
India, for financial support (Project No: SR/FT/CS-039/2008). S.G.
thanks Prof. S. Ramakrishnan, Department of Inorganic and Physical
Chemistry, Indian Institute of Science, Bangalore, for many valuable sug-
gestions to improve the quality of the manuscript. M.R.M. and A.D.
thank the IACS and CSIR, respectively, for research fellowships.
[1] R. Foster in Organic Charge-Transfer Complexes, Academic Press,
London, 1969.
[2] a) A. Friggeri, O. Gronwald, K. J. C. Van Bommel, S. Shinkai, D. N.
Reinhoudt, J. Am. Chem. Soc. 2002, 124, 10754; b) U. Maitra, P. V.
Kumar, N. Chandra, L. Z. D’Souza, M. D. Prasanna, A. R. Raju,
Chem. Commun. 1999, 595.
[3] P. Talukdar, G. Bollot, J. Marenda, N. Sakai, S. Matile, J. Am. Chem.
Soc. 2005, 127, 6528.
[4] S. A. Vignon, T. Jarrosson, T. Iijima, H-R. Tseng, J. K. M. Sanders,
J. F. Stoddart, J. Am. Chem. Soc. 2004, 126, 9884.
[5] H. Y. Au-Yeung, G. D. Pantos, J. K. M. Sanders, J. Am. Chem. Soc.
2008, 130, 12872.
[6] a) R. S. Lokey, B. L. Iverson, Nature 1995, 375, 303; b) J. Q. Nguyen,
B. L. Iverson, J. Am. Chem. Soc. 1999, 121, 2639; c) A. J. Zych, B. L.
Iverson, J. Am. Chem. Soc. 2000, 122, 8898; d) G. J. Gabriel, B. L
Iverson, J. Am. Chem. Soc. 2002, 124, 15174; e) M. S. Cubberley,
B. L. Iverson, J. Am. Chem. Soc. 2001, 123, 7560.
[7] a) S. Ghosh, S. Ramakrishnan, Angew. Chem. 2005, 117, 5577;
Angew. Chem. Int. Ed. 2005, 44, 5441; b) S. Ghosh, S. Ramakrishn-
an, Angew. Chem. 2004, 116, 3326; Angew. Chem. Int. Ed. 2004, 43,
3264; c) S. Ghosh, S. Ramakrishnan, Macromolecules 2005, 38, 676;
d) S. De, S. Ramakrishnan, Macromolecules 2009, 42, 8599; e) X.
Zhao, M.-X. Jia, X.-K. Jiang, L.-Z. Wu, Z.-T. Li, G.-J. Chen, J. Org.
Chem. 2004, 69, 270; f) S. Burattini, H. M. Colquhoun, J. D. Fox, D.
Friedmann, B. W. Greenland, P. J. F. Harris, W. Hayes, M. E.
Mackay, S. J. Rowan, Chem. Commun. 2009, 6717.
Compound A2: Compound 10 (500 mg, 0.96 mmol) and 1,4,5,8-naphtha-
lenetetracarboxylic bisanhydride (103 mg, 0.384 mmol) were placed in a
round-bottomed flask along with dry DMF (10 mL), and the reaction
mixture was stirred for 6 h at 1408C under a N2 atmosphere. The reaction
mixture was allowed to cool to RT, poured into H2O (100 mL) and then
extracted with CH2Cl2 (2ꢂ20 mL). The combined organic layer was
washed with H2O (2ꢂ50 mL), dried over anhyd Na2SO4, and the solvent
was evaporated to afford the crude product as a brown solid, which was
purified by column chromatography by using silica gel as stationary
phase and CHCl3 as eluent to obtain the pure product as a yellow solid
(0.396 g, 81%). M.p. 138–1408C; 1H NMR (300 MHz, CDCl3,TMS): d=
6.92 (s, 2H), 4.04–4.00 (m, 6H), 1.86–1.73 (m, 6H), 1.49–1.45 (m, 6H),
1.34–1.24 (m, 24H), 0.90–0.87 ppm (m, 9H); 13C NMR (CDCl3,): d=
160.76, 153.29, 142.32, 131.72, 126.82, 125.63, 106.35, 77.27, 77.01, 76.76,
73.59, 69.36, 31.91, 31.84, 30.36, 29.55, 29.38, 29.36, 29.30, 26.09, 22.7,
22.67, 14.09 ppm; UV/Vis (CH2Cl2): lmax (e)=378 nm (20846), 358 nm
(17722), 340 nm (10832mꢁ1 cmꢁ1); MS (ESI): m/z calcd for C76H113N4O12
:
1273.83 [M+H]+; found: 1274.00; elemental analysis calcd (%) for
C76H112N4O12: C 71.67, H 8.86, N 4.40; found: C 71.73, H 8.91, N 4.45.
1, 5-Bis(dodecyloxy)naphthalene (D2):[25] 1,5-Dihydroxynaphthalene
(0.4 g, 2.499 mmol), KOH (0.370 g, 6.607 mmol) and tetrabutyl ammoni-
um bromide (0.510 g, 1.582 mmol) were dissolved in H2O (3 mL) and
stirred at RT for 30 min. Dodecylbromide (3.2 mL, 10.07 mmol) was
added to this solution, and the solution was stirred at 958C for 24 h. The
reaction was stopped, and on cooling to RT the product precipitated as a
yellow solid. The solid was filtered, then washed with H2O and then re-
peatedly washed with hexane to get the pure compound as shiny yellow
solid (0.531 g, 65%). M.p. 69–708C; 1H NMR (300 MHz, CDCl3, TMS):
d=7.84 (d, J=8.3 Hz, 2H), 7.34 (t, J=8.2 Hz, 2H), 6.82 (d, J=7.6 Hz,
2H), 4.12 (d, J=6.4 Hz, 4H), 1.94–1.87 (m, 4H), 1.31–1.27 (m, 36H),
0.88 ppm (t, J=4.4 Hz, 6H); UV/Vis (CH2Cl2): lmax (e)=327 (5345), 313
(5881), 298 (6008), 287 nm (6221mꢁ1 cmꢁ1); MS (ESI): m/z calcd for
C34H57O2: 497.43 [M+H]+; found: 497. 54; elemental analysis calcd (%)
for C34H56O2: C 82.20, H 11.36; found: C 82.27, H 11.32.
[8] a) S. Mahlstedt, D. Janietz, A. Stracke, J. H. Wendorff, Chem.
Commun. 2000, 15; b) W. Pisula, M. Kastler, D. Wasserfallen,
J. W. F. Robertson, F. Nolde, C. Kohl, K. Mꢃllen, Angew. Chem.
2006, 118, 834; Angew. Chem. Int. Ed. 2006, 45, 819; c) W. Kranig,
C. Boeffel, H. W. Spiess, O. Karthaus, H. Ringsdorf, R. Wustefeld,
Liq. Cryst. 1990, 8, 375.
[9] V. Percec, M. Glodde, T. K. Bera, Y. Miura, I. Shiyanovskaya, K. D.
Singer, V. S. K. Balagurusamy, P. A. Heiney, I. Schnell, A. Rapp,
H. W. Spiess, S. D. Hudson, H. Duan, Nature 2002, 417, 384.
[10] For self-sorting in various types of supramolecular assembly, see:
a) A. Wu, L. Isaacs, J. Am. Chem. Soc. 2006, 128, 4831; b) P. Mukho-
padhyay, P. Y. Zavalij, L. Isaacs, J. Am. Chem. Soc. 2006, 128, 14093;
c) A. Heeres, C. V. D. Pol, M. Stuart, A. Friggeri, B. L. Feringa, J.
van Esch, J. Am. Chem. Soc. 2003, 125, 14252; d) N. Mizoshita, H.
Monobe, M. Inoue, M. Ukon, T. Watanabe, Y. Shimizuz, K. Hana-
busa, T. Kato, Chem. Commun. 2002, 428; e) S. Ghosh, X.-Q. Li, V.
Stepanenko, F. Wꢃrthner, Chem. Eur. J. 2008, 14, 8074; f) Y. Rudze-
vich, V. Rudzevich, F. Klautzsch, C. A. Schalley, V. Bçhmer, Angew.
Chem. 2009, 121, 3925; Angew. Chem. Int. Ed. 2009, 48, 3867; g) N.
Tomimasu, A. Kanaya, Y. Takashima, H. Harada, J. Am. Chem. Soc.
2009, 131, 12339.
Compound A3: Dodecylamine (500 mg, 2.7 mmol) and 1,4,5,8-naphthal-
ACHTUNGTRENNUNGenetetracarboxylic dianhydride (322 mg, 1.2 mmol) in dry DMF (15 mL)
were stirred at 1408C for 6 h. The reaction mixture was allowed to cool
to RT, and the solution was placed in the refrigerator for 30 min to get a
cream-coloured solid product. The solution was filtered and washed with
MeOH, and the obtained solid was further purified by column chroma-
10092
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10084 – 10093

Self-Sorted Chromophore Assembly
FULL PAPER
[11] J.-L. Brꢄdas, J. E. Norton, J. Cornil, V. Coropceanu, Acc. Chem. Res.
2009, 42, 1691.
ly very close to zero (Figure 3b) but not exactly 0.0. The very negli-
gible absorbance arises due to scattering, and the intensity of scat-
tering increases towards lower wavelengths. In the calculated spec-
trum, the effect of scattering is actually counted twice in the process
of mathematical summation of the two individual spectra which
might have contributed to the observed minor difference.
[19] To get the sole contribution of D1 to the absorption at l=312 nm,
the spectrum was subtracted from the A1 spectrum.
[12] For orthogonal self-assembly of donor and acceptor chromophores
based on different design strategies, see: a) T. L. Benanti, P. Sae-
jueng, D. Venkataraman, Chem. Commun. 2007, 692; b) K. Sugiya-
su, S-i. Kawano, N. Fujita, S. Shinkai, Chem. Mater. 2008, 20, 2863;
c) J. van Herrikhuyzen, A. Syamakumari, A. P. H. J. Schenning,
E. W. Meijer, J. Am. Chem. Soc. 2004, 126, 10021.
[13] A1 was not fully soluble in pure MCH at RT, thus MCH/CHCl3
(95:5) was used as the solvent for self-assembly.
[20] Molecular modelling was performed in Chem 3D-ultra 8.0 by using
MM2 for energy minimisation.
[14] J. Gawronski, M. Brzostowska, K. Kacprzak, H. Kolbon, P. Skowro-
nek, Chirality 2000, 12, 263.
[21] M. Tomasulo, D. M. Naistat, A. J. P. White, D. J. Williams, F. i. M.
Raymo, Tetrahedron Lett. 2005, 46, 5695.
[15] a) H. Shao, T. Nguyen, N. C. Romano, D. A. Modarelli, J. R. Par-
quette, J. Am. Chem. Soc. 2009, 131, 16374; b) P. Mukhopadhyay, Y.
Iwashita, M. Shirakawa, S-i. Kawano, N. Fujita, S. Shinkai, Angew.
Chem. 2006, 118, 1622; Angew. Chem. Int. Ed. 2006, 45, 1592.
[16] J.-Li. Hou, M-X. Jia, X-K. Jiang, Z-T. Li, G-J. Chen, J. Org. Chem.
2004, 69, 6228.
[22] F. J. M. Hoeben, P. Jonkheijm, E. W. Meijer, A. Schenning, Chem.
Rev. 2005, 105, 1491.
[23] D. D. Perrin, W. L. F. Armarego, D. R. Perrin, Purification of Labo-
ratory Chemicals, 2nd ed., Pergamon, Oxford, 1980.
[24] S. Hagihara, L. Gremaud, G. Bollot, J. Mareda, S. Matile, J. Am.
Chem. Soc. 2008, 130, 4347.
[17] M. Bhattacharya, A. G. Samuelson, P. K. Das, J. Phys. Chem. B
2007, 111, 7122.
[25] H-T. Chang, H-T. Lee, J. Macromol. Sci. Pure Appl. Chem. 2009, 46,
722.
[18] There is a small difference between the sum spectrum and the spec-
trum of the mixture. It is worth noting that even at l=600 nm, at
which the absorbance should ideally be 0.0, the absorbance is actual-
Received: March 7, 2010
Published online: June 25, 2010
Chem. Eur. J. 2010, 16, 10084 – 10093
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10093