Liquid-crystalline supramolecular polymers formed through complementary nucleobase-pair interactions

Article abstract of DOI:10.1002/chem.200500827

We report how the placement of nucleobase units, thymine, or N ⁶-(4-methoxybenzoyl)adenine, onto the ends of a mesogenic core, bis-4-alkoxy-substituted bis(phenylethynyl)-benzene, affects the properties of these materials. We show that addition

Full text of DOI:10.1002/chem.200500827

DOI: 10.1002/chem.200500827
Liquid-Crystalline Supramolecular Polymers Formed through
Complementary Nucleobase-Pair Interactions
Sona Sivakova,[a] Jian Wu,[b] Cheryl J. Campo,[a] Patrick T. Mather,[a] and
Stuart J. Rowan*[a]
Abstract: We report how the place-
ment of nucleobase units, thymine, or
N6-(4-methoxybenzoyl)adenine, onto
the ends of a mesogenic core, bis-4-
alkoxy-substituted bis(phenylethynyl)-
benzene, affects the properties of these
materials. We show that addition of
these bulky polar groups significantly
reduces the range of liquid-crystalline
behavior of these compounds. Howev-
er, mixing two complementary nucleo-
base-containing AA- and BB-type
monomer units together does result in
the formation of stable, thermotropic
liquid-crystalline (LC) phases. Hydro-
gen bonding is shown to play an impor-
tant role in the formation of these LC
phases, consistent with the formation
of oligomeric or polymeric hydrogen-
bonded aggregates. X-ray analyses of
these mixed materials are consistent
with the formation of smectic C phases.
Keywords: hydrogen bonds · liquid
crystals · nucleobases · polymers ·
supramolecular chemistry
Introduction
interesting opportunity offered by such systems is the ability
to self-assemble functional units into processable polymeric
materials, which, for example, exhibit attractive electronic
and/or optical properties.[2] The properties of such noncova-
lently bound aggregates have a strong dependence not only
on their functional core components, but also on the nature
(stability and dynamics) of the supramolecular motifs that
control the self-assembly process. In addition, if the supra-
molecular motif used in the assembly of the polymer is
asymmetric (i.e., it consists of two different complementary
units), then the supramolecular polymer will be formed only
in the presence of both of these complementary units. For
example, a heteroditopic monomer, in which both comple-
mentary units are placed on the same molecule, will result
in a self-assembling (A–B)n polymer (Figure 1a). However,
homoditopic monomers, which have only one of the comple-
mentary units placed on each monomer (e.g., A–A or B–B),
will exhibit polymer-like properties only upon mixing of the
two corresponding monomers (Figure 1b). Therefore, in this
case there exists the potential that the supramolecular poly-
mers formed will feature new functional properties that are
not exhibited by the individual monomers. An example of
such a supramolecular-aided function is the growing area of
supramolecular liquid-crystalline materials.[3–5] In such sys-
tems, molecular-shape anisotropy is important for the for-
mation of a liquid-crystalline (LC) phase. For example, rigid
and semirigid molecules that have a high aspect ratio often
exhibit liquid-crystalline behavior. Therefore, the self-assem-
bly of small (semi)rigid units into a larger linear array, with
Over the last decade, supramolecular polymerization, that
is, the self-assembly of small monomeric units into polymer-
like materials through the use of noncovalent interactions,
has received growing attention.[1] Conceptually, a simple
way to achieve such supramolecular polymers is by the at-
tachment of appropriate supramolecular motifs onto the
ends of a core unit. The backbones of the resulting self-as-
sembled polymeric systems will, therefore, contain noncova-
lent bonds, in addition to covalent bonds, and these collec-
tively impart reversibility upon the system (i.e., a dynamic
degree of polymerization) and temperature sensitivity. This
behavior, in turn, offers the potential to develop polymeric
materials in which the melt viscosity at elevated tempera-
tures is more akin to a monomeric-like state, and thus
allows for the utilization of mild processing conditions. One
[a] Dr. S. Sivakova, C. J. Campo, Prof. Dr. P. T. Mather,
Prof. Dr. S. J. Rowan
Department of Macromolecular Science and Engineering
Case Western Reserve University
2100 Adelbert Road, Cleveland, OH 44106–7202 (USA)
Fax : (+1)216-368-4202
E-mail: stuart.rowan@case.edu
[b] Dr. J. Wu
Department of Chemical Engineering
University of Connecticut, Storrs, Connecticut (USA)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
446
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 446 – 456

FULL PAPER
should exhibit polymer-like properties only upon mixing of
the complementary monomers. Another possible design in-
volves a heteroditopic monomer, in which both complemen-
tary units are placed on the same core, resulting in a self-as-
sembling (AB)n polymer. This manuscript will focus on
(AA–BB)n systems, and studies on a related (AB)n material
will be published elsewhere.
Results and Discussion
To effectively utilize nucleobase binding motifs in the self-
assembly of supramolecular liquid-crystalline polymers, it is
important to consider the nature of the nucleobase interac-
tions. Although nucleobases constrained within DNA se-
quences are some of the most controllable supramolecular
motifs, the individual nucleobases are less predictable.[14] For
example, the purines (adenine and guanine) are able to bind
through two different binding sites (Watson–Crick and
Hoogsteen),[18] and as a result can form multicomponent
complexes. In addition, the individual nucleobases can ho-
modimerize to form even more diverse structures, as in the
case of guanine, which can form linear tapes, sheets, ribbons,
and macrocycles.[19] Thus, the utilization of nucleobases
beyond the constraints of a DNA backbone may require
special modifications to enhance the level of predictability
of the system. To this end, we investigated the binding capa-
bility of N6-anisoyladenine with thymine.
Figure 1. Schematic representation of two different types of supramolec-
ular polymers that are formed by the association of monomers with com-
plementary end-groups. a) Self-assembly of a heteroditopic monomer to
yield an (AB)n supramolecular polymer, b) self-assembly of two comple-
mentary homoditopic monomers to yield an (AA–BB)n supramolecular
polymer. The dynamic nature of these polymers allows them to break
and recombine in response to changes in the environment.
a high axial ratio, can result in the appearance of a liquid-
crystalline phase.[6–9] The induction of an LC phase in such a
case is a macroscopic expression of the molecular recogni-
tion designed into the molecules. Furthermore, the forma-
tion of such an LC phase, with its long-range order, will aid
in the formation of higher-molecular-weight aggregates.[10,11]
Our investigations in this field have, in part, focused on
the use of nucleobase-pair interactions to control the self-as-
sembly of fluorescent, low-molecular-weight monomers into
liquid-crystalline polymeric architectures.[12] The utilization
of the nucleobase interactions in supramolecular chemistry
offers the opportunity to prepare precise self-assembled ar-
chitectures,[13] and allows for the exploitation of different
units, all of which offer various binding motifs.[14] We, and
others, have shown that the placement of a single nucleo-
base at either end of a chain results in dramatic changes in
material properties.[12,15–17] To this end, we have designed ho-
moditopic monomers (Figure 2) in which the nucleobase
thymine (T) and the nucleobase derivative N6-(4-methoxy-
benzoyl)adenine (AAn) are substituted on both ends of an
alkoxy-substituted bis(phenylethynyl)benzene core (BP1aBP
and BP1bBP, BP =nucleobase derivative). Thus, this system
Protection of the adenine moiety by N6-anisoyl effectively
reduces the number of possible thymine binding sites, thus
preventing the formation of 2:1 T2:A complexes. Conse-
quently, the possibility of hydrogen-bond-mediated branch-
ing/crosslinking during the self-assembly process is reduced.
With this in mind, we investigated the effect that such a
modification has on the interaction between N6-anisoylade-
nine and thymine.[20] Two model compounds, 1-dodecylthy-
mine (dodecyl-T) and N6-anisoyl-9-dodecyladenine (dode-
cyl-AAn), were prepared and investigated. NMR titration ex-
periments (in CDCl3), in which the shift of the N6-H on the
adenine was monitored, allowed the binding constant be-
tween these two model compounds to be estimated at
around 22mꢀ1, confirming that AAn and T do indeed interact
with each other through hydrogen bonds. However, this in-
dicates a significant reduction in the strength of the interac-
tion between these two nucleobase moieties compared to
the case of unprotected adenine (KA–T =100mꢀ1 in
CDCl3).[21] Nuclear Overhauser effect (NOE) experiments
were carried out by utilizing a 0.1m solution of 1:1 dodecyl-
T:dodecyl-AAn in CDCl3 to determine through which of the
two purine binding sites (i.e., Hoogsteen or Watson–Crick)
the modified adenine interacts with thymine. Irradiation of
the dodecyl-T NH proton resulted in NOE to both the H8
and H2 protons of the adenine, in a 2:1 ratio. As a conse-
quence of the limitations of this technique, direct correla-
tions cannot be made; however, the results suggest that thy-
mine does have a preference for the Hoogsteen face of the
modified adenine, at least in solution.[20,22]
Figure 2. Structures of the homoditopic (A–A, B–B) nucleobase-termi-
nated bis(phenylethynyl)benzene monomers (BP1aBP and BP1bBP). BP =
nucleobase derivative.
Chem. Eur. J. 2006, 12, 446 – 456
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
447

S. J. Rowan et al.
Pure homoditopic compound properties: Having demon-
strated the existence of a hydrogen-bond-mediated interac-
tion between thymine and N6-anisoyladenine, we next stud-
ied how the placement of these supramolecular binding
units affected low-molecular-weight, mesogenic core mole-
cules. To this end, a series of nucleobase-terminated meso-
genic molecules were prepared. Attention was focused on
systems that were based on the alkoxy-substituted bis(phe-
nylethynyl)benzene core unit. This fluorescent mesogen
with pendant octyloxy chains (and no nucleobases attached)
has been shown to display both nematic (N) and smectic (S)
phases with the following thermal transition temperatures
on heating (8C): K 128.3 S1 167 S2 182 N 218 I.[23] The sym-
metry of this core unit allows
(Figure 3). Liquid-crystallinity is observed for only a narrow
temperature range above Tm, specifically 205–2148C for
T1aT and 206–2158C for T1bT. TMe1aTMe also displays a
liquid-crystalline phase over a narrow temperature range,
but at much lower temperatures, 153–1678C, clearly reveal-
ing the importance of thymine hydrogen-bonding in deter-
mining the primary melting point. Upon cooling from a tem-
perature greater than 2148C, T1aT did not exhibit any ob-
servable LC behavior, but instead crystallized at 1828C
from an isotropic phase. However, both T1bT and TMe1aTMe
do form LC phases upon cooling. Although results of differ-
ential scanning calorimetry (DSC) of TMe1aTMe show only
one broad exothermic peak at 1358C (see Supporting Infor-
quick and easy access to
a
series of homoditopic mono-
mers and the conjugated nature
of the mesogen should impart
fluorescent properties upon the
resulting materials.
Scheme 1 shows the general
synthetic protocol used to
access a range of these homodi-
topic monomers, in which thy-
mine (T), methylthymine (TMe),
or N6-anisoyladenine (AAn) was
placed on both ends of either a
hexyloxy or nonyloxy-substitut-
ed bis(4-alkoxyphenylethynyl)-
Scheme 1. Synthesis of the homoditopic (A–A, B–B) nucleobase-terminated alkoxy-substituted bis(phenyl-
ethynyl)benzene monomers. a) Br(CH2)nBr [n=6,9], K2CO3, T HF; b)BPH, NaH, DMF; c) 1,4-diethynylben-
zene, [Pd(PPh3)4], CuI, diisopropylamine, toluene. BP =nucleobase derivative.
benzene unit. The first step in the synthesis of these nucleo-
base-terminated monomers was the reaction of p-iodophe-
nol (2) with excess 1,6-dibromohexane or 1,9-dibromono-
nane and K2CO3 to yield either 3a or 3b in 57 and 31%
yield, respectively. The synthesis of BP4a, in which BP =AAn
or T, was achieved in moderate yields (20–48%) by reacting
3a, under basic conditions in DMF, with either AAn or T, re-
spectively. By using a similar protocol, BP4b was prepared
from 3b. Finally, the appropriate BP4a (or BP4b) compound
was coupled to 1,4-diethynylbenzene under palladium-cata-
lyzed Sonogashira conditions in toluene/diisopropylamine,
resulting in the desired nucleobase-terminated supramolec-
ular monomers BP1aBP (or BP1bBP) in moderate to good
yield (50–82%). Synthesis of the methylated TMe1aTMe mon-
omer involved the reaction of T4a with methyl iodide in
K2CO3/DMF to yield TMe4a in 88% yield. TMe4a was then
coupled to 1,4-diethynylbenzene in a similar manner as
before, yielding the desired monomer in 75% yield. Struc-
1
tures of all the monomers were confirmed by both H NMR
spectroscopy and MALDI-TOF mass spectrometry.
The effect of nucleobase attachment onto the alkoxy-substi-
tuted bis(phenylethynyl)benzene core: Not surprisingly, de-
rivatization of the ends of the alkoxy-substituted bis(pheny-
lethynyl)benzene cores with thymine units results in a signif-
icant reduction of the temperature range in which these ma-
terials exhibit liquid-crystalline behavior, primarily through
a dramatic increase in the primary melting temperature
Figure 3. DSC thermograms (second heating) of the individual monomers
BP1BP, in which a) six carbons or b) nine carbons separate the nucleobase
from the aromatic core. Heating rate of 58C minꢀ1 was used. The trace
labeled i) corresponds to T-functionalized monomers, ii) to the AAn-func-
!
tionalized monomers, and iii) to the 1:1 mixtures. The in b)ii) indicate
the temperatures of the samples shown in the POM images in Figure 4.
Endothermic heat flow is upward.
448
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 446 – 456

Liquid-Crystalline Supramolecular Polymers
FULL PAPER
mation), results of polarising optical microscopy (POM)
reveal a birefringent phase at this temperature just prior to
crystallization. T1bT, on the other hand, exhibits a much
broader LC temperature range (201–1618C) upon cooling.
Furthermore, within this LC region, a number of small exo-
therms are also present in the DSC thermogram. However,
only subtle differences were observed in the schlieren-like
textures between these transitions, and a definitive assign-
ment of the LC phases that occur at these temperatures was
not possible.
tion observed upon heating at around 1308C does not occur
on cooling, no matter how slow. Furthermore, the recrystal-
lization occurs on heating only if the sample is first cooled
to below 908C, suggesting that the small exotherm at 938C
may correspond to the formation of a nucleation source for
the recrystallization at elevated temperature.
Mixing complementary nucleobase-end-functionalized mon-
omers: Melt mixing of the complementary nucleobase-ter-
minated monomers AAn1aAAn and T1aT results in a dramat-
ic decrease in the melting temperature, along with the con-
current appearance of a viscous birefringent phase, as shown
by the results of POM (Figure 5a) for samples heated to be-
Placement of the N6-anisoyladenine (AAn) onto the sym-
metrical mesogenic units also significantly alters the proper-
ties of these materials. The AAn1aAAn monomer displays a
glass-like transition at Tg =828C upon heating, as well as a
broad crystallization exotherm at 140–1808C before melting
at 1878C. A sharp endothermic transition is superimposed
within the crystallization temperature range, occurring at
1638C. However, the formation of a liquid-crystalline phase
was not observed at any temperature for this compound.
AAn1bAAn behaves slightly differently from AAn1aAAn, ex-
hibiting a glass-like transition at Tg =688C and two exother-
mic peaks, a small one at 938C and a much larger one at
1248C. In addition, unlike AAn1aAAn, AAn1bAAn does exhibit
a liquid-crystalline phase for the range 192–2028C. Cooling
of AAn1bAAn results in the formation of an LC phase at
1668C; however, no subsequent recrystallization is observed,
resulting in a “freezing in” of the texture upon formation of
the glassy solid. Figure 4 shows polarized optical micro-
graphs obtained during the second heating of AAn1bAAn. At
808C, which is above Tg, a metastable birefringent phase
(Figure 4a) is observed. Upon increasing the temperature to
above 1308C, a recrystallization from this metastable LC
phase occurs (Figure 4b and c). This material remains crys-
talline until around 1908C (Figure 4d), before becoming iso-
tropic above 2028C (Figure 4e). Interestingly, the crystalliza-
Figure 5. Polarized optical micrographs of 1:1 mixtures of a) T1aT:
AAn1aAAn at 1358C, b) T1bT:AAn1bAAn at 1458C, c) T1bT:AAn1aAAn at
1258C, and d) T1aT:AAn1bAAn at 1358C.
tween 115 and 1548C. The appearance of the liquid-crystal-
line phase is bounded by endothermic peaks, as observed in
the DSC thermogram (Figure 3a(iii) and inset). Above
1758C, the material behaves as a free-flowing liquid, which
becomes more viscous upon subsequent cooling to below
1758C. At 1538C, a viscous birefringent phase was observed,
which vitrified upon cooling to below 1168C. Notably, these
phenomena are not observed upon the first heating of the
sample obtained by mixing the monomers in a 1:1 ratio in
either solid or solution state, but are consistently present
upon subsequent heating and cooling cycles. Similar obser-
vations were made for mixing of the complementary mono-
mers T1bT and AAn1bAAn, which contain a nine-carbon
chain between the rigid core and the nucleobases. Again,
the formation of a viscous birefringent phase was observed
by both DSC (Figure 3b(iii)) and POM (Figure 5b) analyses,
Figure 4. Polarized optical micrographs of the second heating of
AAn1bAAn at a) 808C; b) 1358C, 0 min; c) 1358C, 2.5 min; d) 1908C; and
e) 2028C. Heating rate 58Cminꢀ1
.
Chem. Eur. J. 2006, 12, 446 – 456
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
449

S. J. Rowan et al.
this time with a broad temperature range spanning 117–
1928C. Furthermore, intermixing of the complementary C6
and C9 monomers in 1:1 ratios also resulted in the formation
of viscous birefringent phases (Figure 5c and d), which were
observed over a temperature range of approximately 608C.
Notably, in some of these mixed-monomer systems, the DSC
thermograms show a number of small endotherms within
the liquid-crystalline region. However, these transitions are
accompanied by little or no difference in liquid-crystalline
texture, as seen in POM images, precluding direct assign-
ment of the LC phases that exist between each transition.
The textures observed in all the samples are best described
as schlieren, which is consistent with the presence of nemat-
ic and/or some smectic phases (e.g., smectic C).[24]
As for conventional liquid crystals, the orientation of our
self-assembled polymers based on complementary nucleo-
base monomers could be macroscopically aligned with an
electric field. A mixture of T1aT:AAn1aAAn was brought to
1328C, 178C above its solid-LC phase transition, at which a
birefringent texture was observed by POM analysis (Fig-
ure 6a). An electric field was then applied to the sample,
Figure 7. Summary of the phase behavior of the individual monomers
and the 1:1 mixtures of complementary monomers on heating. The
dotted horizontal lines correspond to additional endotherms within the
liquid-crystalline range (DSC).
verse-Hoogsteen interaction, resulting in the formation of
long, linear chains. The assembly of longer linear aggregates
increases the effective aspect ratio of the molecules and con-
sequently induces conditions more favorable for LC forma-
tion and stability. Moreover, the supramolecular aggregates
would have more difficulty packing into a crystalline lattice,
thus lowering Tm and effectively broadening the LC window.
To understand more fully the nature of the liquid-crystal
phases observed, wide angle X-ray diffraction (WAXD)
studies were performed on all of the 1:1 complementary
mixtures (Figure 8). The samples were prepared by solution-
mixing the complementary monomers, casting into films,
heating the samples to above 2008C, and subsequently cool-
ing at approximately 58Cminꢀ1. Little or no crystallization
was observed by POM in these samples and the LC struc-
ture that existed prior to vitrification was frozen-in upon
cooling. Furthermore, a glass-like transition is observed in
the DSC thermogram for all such samples; therefore, X-ray
data was recorded at room temperature. The WAXD pat-
terns obtained for all of these mixtures are consistent with
the presence of a smectic phase prior to vitrification, as evi-
denced by a sharp, low-angle ring superposed by much
broader reflections at wider angles. For the T1aT:AAn1aAAn
mixture, in which both components have six carbon atoms
between the nucleobase and the rigid mesogenic unit, the
low-angle ring was characterized by a d-spacing of 37.5 Š.
This distance is slightly, but significantly, less than the supra-
molecular polymer repeating length of 41 Š estimated from
molecular modeling studies of the Hoogsteen hydrogen-
bonded aggregate shown in Scheme 2. Thus, our WAXD ob-
servation implies a smectic C arrangement with mesogens
tilted away from the layer normal at an angle of 248. As ex-
pected, increasing the length of the alkyl space group from
six to nine increased the smectic layer spacing to approxi-
Figure 6. Polarized optical micrographs (magnification ”500) of a 1:1
mixture of T1aT and AAn1aAAn at 1328C a) before application of an elec-
tric field, b) after application of an electric field (300 V), and c) upon re-
moval of an electric field.
causing the LC director to align parallel to the field direc-
tion, yielding darkness under POM viewing conditions (Fig-
ure 6b), consistent with homeotropic alignment. Interesting-
ly, upon removal of the electric field, the director relaxed
and a schlieren texture was again observed, but now with a
finer scale (Figure 6c). The return to a textured orientation
distribution may be due to the lack of homeotropic surface
treatment of the glass surfaces, which would have preserved
orientation upon removal of the field. The relaxation took
several seconds to occur, which is longer than expected for a
small-molecule LC material. We attribute this difference to
supramolecular polymerization and an associated higher vis-
cosity.
Figure 7 summarizes the liquid-crystalline phases ob-
served upon heating for all the materials discussed above.
For all the mixed-monomer systems, it is postulated that the
observed liquid-crystalline phase, which occurs over a much
broader temperature range than is exhibited by the individu-
al monomers, is a consequence of complementary nucleo-
base interactions that result in the formation of larger supra-
molecular aggregates. Scheme 2 shows a proposed model, in
which the nucleobases form hydrogen bonds through the re-
450
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 446 – 456

Liquid-Crystalline Supramolecular Polymers
FULL PAPER
Scheme 2. Proposed nucleobase-induced, self-assembled fluorescent polymer (T1aT:AAn1aAAn).
Figure 8. WAXD data obtained at RTof a)
T1aT:AAn1bAAn, c) T1bT:AAn1aAAn, and d) T1bT:AAn1bAAn
T1aT:AAn1aAAn
, b)
.
mately 40 Š for T1bT:AAn1aAAn and T1aT:AAn1bAAn, and
41.8 Š for T1bT:AAn1bAAn. In each case, the measured d-
spacing is less than that estimated from molecular modeling
(ca. 45 Š for T1bT:AAn1aAAn and T1aT:AAn1bAAn, and ca.
49 Š for T1bT:AAn1bAAn), suggesting that these mixtures
also form a smectic C phase before vitrification, with meso-
gen tilt angles of 27 and 318, respectively.
Figure 9. a) Polarized optical micrograph of a fiber pulled from the melt
of
a a fiber of
T1bT:AAn1bAAn mixture. b) WAXD patterns of
T1bT:AAn1bAAn. The calibrated d-spacings are indicated on the pattern
in Angstroms. The fiber axis was vertical.
Consistent with the formation of supramolecular liquid-
crystalline polymers, fibers could be obtained from the melt
of the 1:1 BP1BP complementary-monomer mixtures. Fig-
ure 9a shows a POM image of one such fiber obtained from
the LC phase of the T1bT:AAn1bAAn mixture at around
1458C. Figure 9b shows the X-ray fiber diffraction pattern,
which confirms modest molecular orientation induced by
drawing. In particular, diffraction rings at d-spacings of 12
and 4.5 Š, visible in unoriented specimens (data not shown),
are split into off-axis reflections, suggesting a tilted arrange-
ment of liquid-crystalline layering. By virtue of the fluores-
cent mesogenic core present in these monomers, the fibers
obtained from the T1bT:AAn1bAAn mixtures are themselves
fluorescent.
The effect of altering the thymine: N6-anisoyladenine stoi-
chiometry: If supramolecular polymers are indeed formed
by the self-assembly of complementary ditopic monomers,
then addition of a mono-adenine derivative, which will act
as a supramolecular-chain terminator, will proportionally de-
Chem. Eur. J. 2006, 12, 446 – 456
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
451

S. J. Rowan et al.
polymerize the supramolecular chains. As a consequence,
modification of the material properties towards those ob-
served for the monomers should occur. To test this idea, do-
ing the ratio of the monomer units, does reduce the solid-
and liquid-state mechanical properties of these materials.
For example, fibers cannot be obtained from the melt of ma-
terials that do not have a 1:1 stoichiometry of the two nucle-
obases, which is in line with the formation of only low-mo-
lecular-weight aggregates in these systems. We note that
conventional solution-based analyses for molecular-weight
determination cannot be applied to these materials; there-
fore, only melt or solid-state properties can be used to infer
effective chain length.
decyl-AAn was added to a 1:1 mixture of T1aT:AAn1aAAn
.
Indeed, addition of even small amounts of dodecyl-AAn
quickly reduced the thermal range of the observed liquid-
crystalline phase. In the absence of dodecyl-AAn, the liquid-
crystal phase spanned a broad 398C. However, addition of
5% dodecyl-AAn resulted in significant destabilization of the
LC phase, with a reduction in the LC range to only 228C
(Figure 10). The addition of 25% of chain terminator result-
ed in the complete abolition of liquid-crystalline properties.
The fact that liquid-crystalline phases were still observed
after addition of up to 20% of chain terminator suggests
that even relatively short oligomeric chains are sufficient to
induce some liquid-crystal formation.
Disruption of the hydrogen-bonding motif: The proposed
self-assembly model outlined in Scheme 2 assumes that hy-
drogen bonding plays an important role in the supramolec-
ular polymerization of these molecules. To test this model,
specifically for the adenine–thymine base-pair, the N3-me-
thylated thymine monomer TMe1aTMe was prepared and in-
vestigated. The placement of a methyl group on the nitrogen
at the 3 position of the thymine effectively disrupts the hy-
drogen bonding between these complementary nucleobase
moieties. As discussed previously, TMe1aTMe itself has a
lower melting temperature (Tm =1538C) than the T1aT
monomer (Tm =2058C), but does exhibit a narrow liquid-
crystalline phase (153–1678C). An equimolar mixture of
TMe1aTMe with AAn1aAAn produces a material that exhibits
three endotherms, as shown by the results of DSC, at 100,
133, and 1828C. Although an LC phase is observed for this
mixture, the LC region occurs at much higher temperatures
than for the hydrogen-bonded mixtures and over a narrower
temperature range (185–1908C). Furthermore, this sample is
less homogenous than the hydrogen-bonded materials and,
from the results of both POM and DSC, it appears that
there are two melting points (ca. 1308C and ca. 1808C), con-
sistent with the presence of two separate crystalline regions.
From POM observations there appears to be no melting at
1008C, suggesting that this endotherm in the DSC thermo-
gram corresponds to a solid-state transition. In addition, the
LC phase that is observed for this mixture is a free-flowing
liquid from which no fibers could be obtained, in stark con-
trast to the highly viscous liquid-crystal phases formed by
the T1aT:AAn1aAAn mixtures.
To further compare the nature of the hydrogen-bond in-
teractions in these materials, infrared spectroscopy of the
annealed films (ca. 2008C) on CaF2 disks was carried out.
Figure 11a shows the FTIR spectrum of the neat materials
TMe1aTMe and AAn1aAAn. Comparison of the spectra ob-
tained from T1aT:AAn1aAAn and TMe1aTMe:AAn1aAAn (Fig-
ure 11b) reveals vast differences in both the NH and carbon-
yl stretching regions. The T1aT:AAn1aAAn mixture exhibits a
broad shoulder at 2800 cmꢀ1, indicative of the presence of
medium-strength hydrogen bonding.[25] In addition, two NH
stretching vibrations are observed: one at 3183 cmꢀ1, consis-
tent with the presence of hydrogen-bonded NH groups, and
a small peak at 3413 cmꢀ1, consistent with the presence of
some free NH groups.[26] Furthermore, there is one very
broad carbonyl peak spanning from 1718–1658 cmꢀ1, sug-
gesting the involvement of the carbonyl units in hydrogen
Figure 10. Phase diagram depicting the effects of monofunctional dode-
cyl-AAn on the liquid-crystalline phase of the T1aT:AAn1aAAn mixture.
Another method of altering the stoichiometry of the two
binding units is to change the ratio of the two complementa-
ry monomers. To obtain the maximum degree of polymeri-
zation, an exact 1:1 molar equivalent of the complementary
monomers is crucial. An excess of one of the monomers will
essentially act as a chain terminator in an analogous way to
the addition of the mono-adenine derivative. Therefore, we
investigated the effects of different stoichiometries of T1bT
and AAn1bAAn in the monomer mixtures on the stability of
the liquid-crystalline phases. Interestingly, the addition of up
to 50% excess equivalents of either monomer did not result
in dramatic changes, as shown by DSC analysis. For exam-
ple, the addition of 1.5 equivalents of T1bT to 1 equivalent
of AAn1bAAn, or vice versa, resulted in only slight shifts rela-
tive to the 1:1 mixture in the Tg, the Tm, as well as in the
clearing transition. These results are qualitatively consistent
with those obtained by addition of a monofunctional chain-
terminator, again suggesting that only low-molecular-weight
aggregates are required to induce and stabilize liquid-crystal
phase formation. However, it should be noted that altering
the stoichiometry of the complementary nucleobases, either
by addition of a monofunctionalized compound or by chang-
452
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 446 – 456

Liquid-Crystalline Supramolecular Polymers
FULL PAPER
meric species that form relatively stable LC phases. Concur-
rent with the formation of the viscous birefringent phases,
the materials also demonstrate the ability to form oriented,
fluorescent fibers. Thus, we have utilized the functionality of
the core unit to aid LC formation and to impart fluorescent
behavior, in conjunction with the self-assembling capability
of the nucleobases, which not only aids LC formation, but
also imparts polymer-like properties, such as fiber forma-
tion, to the material. Further investigations are underway to
examine the possibility of using such weak noncovalent
binding groups to obtain polymeric materials that show high
thermal sensitivity. Such materials will exhibit very low
melt-viscosities and could open the door to facile processing
of functional (conjugated) polymers, cheaper recycling of
plastics, or even thermally repairable materials.[27]
Experimental Section
Materials: T heN6-(4-methoxybenzoyl)adenine was synthesized according
to literature procedures.[28] All reagents and solvents were purchased
from Aldrich. Reagents were used without further purification. Solvents
were distilled from suitable drying agents.
1
Characterization: H and 13C NMR spectra were recorded by using either
a Varian Gemini 200 MHz, a Varian 300 MHz, or a Varian 600 MHz
NMR spectrometer. Differential scanning calorimetry (DSC) experi-
ments were performed by using a Pyris 1 DSC (Perkin–Elmer) under
flowing N2 at a heating rate of 58Cminꢀ1. Polarising optical microscopy
(POM) studies were performed by using an Olympus BX51 microscope
equipped with crossed polarizers, an HCS 402 hot stage from Instec, and
a digital color CCD camera (SPOTInsight Firewire Color Mosaic).
Images were acquired from the CCD camera at selected times and/or
temperatures by using SPOTsoftware (Diagnostic Instruments). Spatial
dimensions were calibrated by using a stage micrometer with 10 mm line
spacing. The samples used for POM analysis were sandwiched between
two glass coverslips. The temperature ramping rates were chosen to be
consistent with DSC experiments for comparison purposes. The hot stage
was equipped with a liquid nitrogen LN2-P 110 VAC cooling unit from
Instec for the accurate control of sample temperature, either isothermally
or during heating and cooling runs. FTIR measurements were performed
by using a Biorad Excalibur Series FTS3000MX spectrometer. Molecular
weights of materials were measured by mass spectrometry on a Bruker
BIFLEX III MALDI-TOF mass spectrometer (matrix: a-cyano-4-hy-
droxycinnamic acid). Two-dimensional X-ray data of fibers drawn in the
vertical direction was obtained by using a Philips PW1830 generator with
Figure 11. IR spectra of the NH stretching region and the carbonyl
region of a) TMe1aTMe vs AAn1aAAn and b) T1aT:AAn1aAAn vs TMe1aT
Me:AAn1aAAn mixtures.
bonding. This data indicates that in the solid-state, T1aT:
AAn1aAAn does indeed form hydrogen-bonded supramolec-
ular aggregates. The TMe1aTMe:AAn1aAAn mixture, on the
other hand, shows no broad shoulder at 2800 cmꢀ1 and a
very weak NH stretching peak centered at 3258 cmꢀ1. This
stretching results from the NH on the AAn1aAAn monomer,
which has a wavenumber of 3245 cmꢀ1 in neat AAn1aAAn
(Figure 11a). In addition, the carbonyl stretching region
shows multiple, sharp carbonyl peaks. Peaks at 1697 and
1667 cmꢀ1 correspond to the two carbonyl groups found in
Ni-filtered CuKa radiation,
a calibrated sample-detector distance of
5.45 cm, and image plates (Model #FDL-UR-V, Fuji Photo Film) with
25 mm pixel size. Wide angle X-ray diffraction measurements on cast
TMe1aTMe
, with presumably the carbonyl peak from
AAn1aAAn overlapping these peaks. The sharpness of the
peaks suggests that each carbonyl moiety exists in its own
distinctive environment and thus unlikely to be involved in
hydrogen bonding. This data is consistent with there being
no significant hydrogen-bond interactions between the two
components in this material.
films were recorded at the University of Connecticut by using
a
GADDS-4 (Bruker AXS); X-ray source: Cr (l=2.291 Š); experimental
conditions: 40 kV, 40 mA, RT.
Binding studies: The binding studies were performed by using deuterated
chloroform predried with potassium carbonate. A mixture of dodecyl-T
(20 mm) and dodecyl-AAn (5 mm) was titrated to a stock solution of dode-
cyl-AAn (5 mm) so that the concentration of dodecyl-T varied from 1 to
16 mm. The shift of the hydrogen on the N6 position of the adenine was
monitored with respect to chloroform and the data was fitted to the
equation for a 1:1 binding.
1H NMR NOE experiments: Experiments were performed by using a
Varian 600 MHz NMR spectrometer at ꢀ308C in CDCl3, pulse width:
39.38, mixing time: 0 sec, acquisition time: 1.5 s. Concentration of both
dodecyl-T and dodecyl-AAn was 100 mm.
Conclusion
We have shown that the binding motif consisting of thymine
and N6-(4-methoxybenzoyl)adenine can be used to control
the aggregation of symmetrically-substituted alkoxy-bis(phe-
nylethynyl)benzene units, resulting in A–A/B–B-type poly-
Field-induced alignment of LC phases: The material was sandwiched be-
tween two indium tin oxide-coated slides. Two 2 mm flexible spacers were
Chem. Eur. J. 2006, 12, 446 – 456
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
453

S. J. Rowan et al.
used to separate the glass slides. Aluminum electrodes were attached to
the cell and the whole assembly was wrapped in Teflon tape to prevent
short circuiting. The electrodes were then connected to a high-voltage
DC power supply (Bertan Associates, Model 2.5). The entire cell was
placed inside a heating stage (Mettler FP85 Hot Stage) and the sample
temperature was elevated to 1328C. Once this temperature was reached,
a potential difference of 300 V (150 Vmmꢀ1) was applied across the
sample. Results of field application were observed by POM and recorded
by using a microscope-mounted digital camera.
(6.63 g, 15.6 mmol). Column chromatography (silica gel; 99:1 CH2Cl2/
MeOH) of the resulting crude product afforded pure T4b as a white solid
(3.52 g, 48%). M.p. 988C; Rf =0.54 (95:5 CH2Cl2/MeOH); 1H NMR
(CDCl3, 200 MHz): d=8.76 (s, 1H; NH), 7.55 (d, 2H; ArH), 6.97 (s, 1H;
T-CH), 6.67 (d, 2H; ArH), 3.90 (t, 2H; OCH2), 3.68 (t, 2H; NCH2), 1.92
(s, 3H; T-CH3), 1.84–1.65 (m, 4H; OCH2CH2 and Br-CH2CH2), 1.62–
1.35 ppm (m, 10H; 5”alkyl CH2); 13C NMR (CDCl3, 75 MHz): d=164.3,
159.2, 150.9, 140.6, 138.4, 117.2, 110.8, 103.6, 82.7, 68.3, 48.8, 29.6, 29.5,
29.3, 26.7, 26.2, 12.6 ppm; MS (MALDI-TOF): m/z: 471 [M+H]+.
General procedure for preparing AAn4a and AAn4b
General procedure for preparing 3a and 3b
N6-(4-Methoxybenzoyl)adenine (1.83 g, 6.80 mmol) was suspended in
DMF (20 mL) and allowed to stir while NaH (0.26 g, 10.9 mmol) was
added in portions. The resulting mixture was stirred for a further 0.5 h,
then 3a or 3b (5.22 mmol) was added in a single aliquot, and the reaction
mixture was left to stir for 24 h at 308C. The reaction was quenched by a
small amount of methanol and the solvent was evaporated. The residue
was dissolved in CH2Cl2 and washed once with water. The organic layer
was separated and any precipitates formed were filtered off. The CH2Cl2
was then evaporated.
A suspension of potassium carbonate (K2CO3) (8.00 g, 57.9 mmol) and
DMF (30 mL) was purged with argon for 15 min and heated to 708C, at
which time 4-iodophenol (2) (5.00 g, 23.0 mmol) was added. The appro-
priate alkyl bromide (46.2 mmol) was then added quickly by means of a
syringe, and the suspension was stirred at 708C under argon sparging for
4 h.
4-(1-Hexyloxy-6-bromo)-1-iodobenzene (3a): The reaction was conducted
according to the general procedure described above, using 1,6-dibromo-
hexane (7.1 mL, 46.2 mmol). The solvent was evaporated and the residue
was dissolved in methylene chloride (CH2Cl2). The organic phase was
washed with water twice and then evaporated, and the residue was left to
dry in a vacuum oven overnight. Column chromatography (silica gel;
99:1 hexane/toluene) of the resulting crude product afforded pure 3a as a
white solid (5.05 g, 57%). M.p. 578C; 1H NMR (CDCl3, 200 MHz): d=
7.55 (d, 2H; ArH), 6.67 (d, 2H; ArH), 3.93 (t, 2H; OCH2), 3.43 (t, 2H;
BrCH2), 1.91 (t, 2H; OCH2CH2), 1.81 (t, 2H; BrCH2CH2), 1.50 ppm (m,
4H; 2”alkyl CH2); 13C NMR (CDCl3, 75 MHz): d=159.2, 138.4, 117.2,
82.8, 68.1, 34.0, 32.9, 29.3, 28.2, 25.6 ppm; MS (MALDI-TOF): m/z: 382
[M+H]+.
4-(1-Hexyloxy-6-(N6-(4-methoxybenzoyl)adenine)-1-iodobenzene
(AAn4a): The reaction was conducted according to the general procedure
described above, using 3a (1.98 g, 5.22 mmol). The residue was dissolved
in a small amount of CH2Cl2 and excess hexane was added, resulting in
the formation of a white crystalline product that was filtered off and
dried under vacuum overnight. Column chromatography (silica gel; 97:3
CH2Cl2/MeOH) of the resulting crude product afforded pure AAn4a as a
white solid (1.30 g, 43.8%). M.p. 1268C; Rf =0.62 (95:5 CH2Cl2/MeOH);
1H NMR (CDCl3, 200 MHz): d=8.9 (s, 1H; NH), 8.79 (s, 1H; adenine
H-8), 8.00 (s, 1H; adenine H-2), 8.02 (d, 2H; anis-ArH), 7.55 (d, 2H;
ArH), 7.00 (d, 2H; anis-ArH), 6.65 (d, 2H; ArH), 4.30 (t, 2H; NCH2),
3.93 (t, 2H; OCH2), 3.90 (s, 3H; OCH3), 2.05–1.88 (m, 2H; NCH2CH2),
1.88–1.65 (m, 2H; OCH2CH2), 1.65–1.35 ppm (m, 4H; 2”alkyl CH2);
13C NMR (CDCl3, 50 MHz): d=165.8, 164.7, 160.3, 154.0, 153.6, 151.3,
144.3, 139.7, 131.6, 127.4, 124.6, 118.4, 115.5, 84.1, 69.2, 57.1, 45.6, 31.5,
30.5, 27.9, 27.1 ppm; MS (MALDI-TOF): m/z: 572 [M+]+.
4-(1-Nonyloxy-6-bromo)-1-iodobenzene (3b): The reaction was conducted
according to the general procedure described above, using 1,9-dibromo-
nonane (8.33 mL, 46.2 mmol). The solvent was evaporated and the resi-
due was dissolved in CH2Cl2. The organic phase was washed with water
twice, separated, and reduced in volume. The unwanted bis-functional-
ized alkane was precipitated out with methanol. The soluble fraction was
then evaporated and the residue was dried in a vacuum oven overnight.
Column chromatography (silica gel; hexane) of the resulting crude prod-
uct afforded pure 3b as a white solid (3.03 g, 31%). M.p. 568C; Rf =0.39
(hexane); 1H NMR (CDCl3, 200 MHz): d=7.55 (d, 2H; ArH), 6.67 (d,
2H; ArH), 3.93 (t, 2H; OCH2), 3.43 (t, 2H; BrCH2), 1.94–1.71 (m, 4H;
OCH2CH2 and BrCH2CH2), 1.35 ppm (m, 10H; 5”alkyl CH2); 13C NMR
(CDCl3, 75 MHz): d=159.3, 138.4, 117.2, 82.7, 68.3, 34.3, 33.1, 29.6, 29.5,
29.4, 28.9, 28.4, 26.2 ppm; MS (MALDI-TOF): m/z: 426 [M+H]+.
4-(1-Nonyloxy-6-(N6-(4-methoxybenzoyl)adenine)-1-iodobenzene
(AAn4b): The reaction was conducted according to the general procedure
described above, using 3b (2.21 g, 5.22 mmol) and afforded a yellow oil
after removal of the CH2Cl2. Column chromatography (silica gel; 90:10
CH2Cl2/MeCN) of the resulting crude product afforded pure AAn4b as a
white solid (1.37 g, 43%). M.p. 1388C; Rf =0.52 (80:20 CH2Cl2/MeCN);
1H NMR (CDCl3, 200 MHz): d=8.95 (s, 1H; NH), 8.79 (s, 1H; adenine
H-2), 8.00(s, 1H; adenine H-8), 8.02 (d, 2H; anis-ArH), 7.55 (d, 2H;
ArH), 7.00 (d, 2H; anis-ArH), 6.65 (d, 2H; ArH), 4.30 (t, 2H; NCH2),
3.93 (t, 2H; OCH2), 3.90 (s, 3H; OCH3), 2.05–1.88 (m, 2H; NCH2CH2),
1.88–1.65 (m, 2H; OCH2CH2), 1.55–1.25 ppm (m, 10H; internal CH2);
13C NMR (CDCl3, 75 MHz): d=164.4, 163.4, 159.2, 152.7, 152.4, 149.9,
143.1, 138.4, 130.3, 126.1, 123.4, 117.2, 114.3, 82.7, 68.3, 55.8, 44.4, 30.3,
29.6, 29.5, 29.4, 29.2, 26.9, 26.2 ppm; MS (MALDI-TOF): m/z: 614
[M+H]+.
General procedure for preparing T4a and T4b
Thymine (4.00 g, 31.7 mmol) was suspended in DMF (30 mL) and al-
lowed to stir while sodium hydride (NaH) (0.41 g, 17.3 mmol) was added
in portions. The mixture was stirred for a further 0.5 h, then 3a or 3b
(15.6 mmol) was added in a single aliquot, and the mixture was left to
stir overnight at 1008C. The resulting mixture was neutralized with 2 mL
of acetic acid and the solvent was evaporated. The residue was dissolved
in CH2Cl2 and washed once with water. The organic layer was separated
and any precipitates formed were filtered off. The CH2Cl2-soluble frac-
tion was then evaporated and the residue was dried overnight in a
vacuum oven.
General procedure for preparing BP1aBP and BP1bBP (BP=T and AAn
)
In
a
glove box, 1,4-diethynylbenzene (0.25 g, 1.98 mmol), BP4
(4.36 mmol), [Pd(PPh3)4] (24.9 mg, 2.16”10ꢀ5 mol), and CuI (4.20 mg,
2.16”10ꢀ5 mol) were added to a 3:1 v/v mixture of toluene/diisopropyl-
amine (90 mL). After the mixture was heated to 758C, a precipitate start-
ed to appear. The mixture was stirred for 18 h at 758C under an argon at-
mosphere. The resulting strongly luminescent suspension was subsequent-
ly dropped slowly into an excess of MeOH (~200 mL), and the precipi-
tate was collected and dried under vacuum.
4-(1-Hexyloxy-6-thymine)-1-iodobenzene (T4a): The reaction was con-
ducted according to the general procedure described above, using 3a
(5.94 g, 15.6 mmol). Column chromatography (silica gel; 99:1 CH2Cl2/
MeOH) of the resulting crude product afforded pure T4a as a white solid
(1.33 g, 20%). M.p. 1398C; Rf =0.5 (99:1 CH2Cl2/MeOH); 1H NMR
(CDCl3, 200 MHz): d=9.11 (s, 1H; NH), 7.55 (d, 2H; ArH), 6.97 (s, 1H;
T-CH), 6.67 (d, 2H; ArH), 3.92 (t, 2H; OCH2), 3.71 (t, 2H; NCH2), 1.92
(s, 3H; T-CH3), 1.84–1.65 (m, 4H; OCH2CH2 and NCH2CH2), 1.61–
1.35 ppm (m, 4H; 2”alkyl CH2); 13C NMR (CDCl3, 50 MHz): d=166.1,
160.4, 152.6, 141.9, 139.7, 118.4, 112.2, 84.1, 69.3, 49.9, 30.6, 30.5, 27.7,
27.2, 13.9 ppm; MS (MALDI-TOF): m/z: 429 [M+H]+.
4-(1-Hexyloxy-6-(thymine))-1-bis(phenylethynyl)benzene (T1aT): The re-
action was conducted according to the general procedure described
above, using T4a (1.86 g, 4.36 mmol) to afford a yellow solid. Column
chromatography (silica gel; 95:5 CH2Cl2/MeOH) of the resulting crude
product afforded pure T1aT as a yellow solid (1.17 g, 82%). 1H NMR
(CDCl3, 600 MHz): d=8.76 (s, 2H; NH), 7.47 (s, 4H; ArH), 7.46 (d, 4H;
ArH), 6.97 (s, 2H; T-CH), 6.86 (d, 4H; ArH), 3.98 (t, 4H; OCH2), 3.71
(t, 4H; NCH2), 1.93 (s, 6H; T-CH3), 1.84–1.65 (m, 8H; OCH2CH2 and
NCH2CH2), 1.60–1.35 ppm (m, 8H; 4”alkyl CH2); 13C NMR (DMSO at
4-(1-Nonyloxy-6-thymine)-1-iodobenzene (T4b): The reaction was con-
ducted according to the general procedure described above, using 3b
454
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 446 – 456

Liquid-Crystalline Supramolecular Polymers
FULL PAPER
708C, 75 MHz): 164.9, 159.9, 151.6, 141.9, 133.7, 132.0, 123.4, 115.7, 114.8,
109.2, 92.3, 88.4, 68.5, 47.9, 29.2, 29.1, 26.3, 25.9, 12.5 ppm; FTIR (an-
nealed film on CaF2): n˜ =3164, 3039, 2935, 2857, 2215, 1684, 1653, 1602,
1518, 1469, 1354, 1283, 1244, 1173 cmꢀ1; UV: lmax =274, 332, 353, 383 nm;
photoluminescence (excitation at 340 nm, CHCl3): 366, 384; DSC (Tm =
2158C); MS (MALDI-TOF): m/z: 726 [M+].
(t, 2H; NCH2), 3.35 (s, 3H; T-NCH3), 1.92 (s, 3H; T-CH3), 1.84–1.65 (m,
4H; OCH2CH2 and NCH2CH2), 1.61–1.35 ppm (m, 4H; 2”alkyl CH2);
13C NMR (CDCl3, 75 MHz): d=159.2, 138.4, 117.2, 82.7, 68.3, 34.3, 33.1,
29.6, 29.5, 29.4, 28.9, 28.4, 26.2. 164.2, 159.1, 151.8, 138.4, 117.1, 109.8,
103.6, 82.8, 68.0, 49.7, 29.3, 29.2, 28.2, 26.5, 25.9, 13.4 ppm; MS (MALDI-
TOF): m/z: 443 [M+H]+.
4-(1-Nonyloxy-6-(N3-methyl)-thymine)-1-bis(phenylethynyl)benzene
(TMe1aTMe): In a glove box, 1,4-diethynylbenzene (0.02 g, 0.15 mmol),
TMe4 (0.08 g, 0.34 mmol), [Pd(PPh3)4] (2.00 mg, 1.71”10ꢀ6 mol), and CuI
(1.00 mg, 1.71”10ꢀ6 mol) were added to a 3:1 v/v mixture of toluene/di-
isopropylamine (7.5 mL). The mixture was then heated to 758C, and stir-
red for 18 h under an argon atmosphere. The solvents of the resulting lu-
minescent suspension were subsequently evaporated. Column chromatog-
raphy (silica gel; 95:5 CH2Cl2/MeOH) of the resulting crude product af-
forded pure TMe1aTMe (0.12 g, 75%). 1H NMR (CDCl3, 300 MHz): d=
7.47 (s, 4H; ArH), 7.46 (d, 4H; ArH), 6.97 (s, 1H; T-CH), 6.86 (d, 4H;
ArH), 3.98 (t, 4H; OCH2), 3.71 (t, 4H; NCH2), 3.37 (s, 6H; T-NCH3),
1.95 (s, 6H; T-CH3), 1.84–1.65 (m, 8H; OCH2CH2 and NCH2CH2), 1.60–
1.35 ppm (m, 8H; 4”alkyl CH2); 13C NMR (CDCl3, 75 MHz): d=164.3,
159.4, 151.8, 138.4, 133.4, 132.6, 131.7, 131.6, 124.9, 123.3, 121.25, 115.7,
114.8, 109.8, 91.5, 88.2, 68.0, 49.7, 29.3, 28.2, 26.5, 25.9, 113.4 ppm; FTIR
(annealed film on CaF2): n˜ =3065, 2933, 2858, 2212, 1698, 1664, 1638,
1602, 1519, 1468, 1356, 1284, 1246, 1174 cmꢀ1; UV: lmax =274, 335, 354,
378 nm; photoluminescence (excitation at 330 nm, CHCl3): 366, 383, 397;
DSC (Tm =1548C); MS (MALDI-TOF): m/z: 754 [M+].
N7-Dodecyl-(N6-(4-methoxybenzoyl)adenine (dodecyl-AAn): p-Anisoyl-
adenine (2.33 g, 8.63 mmol) was dissolved in dry DMF (45.0 mL), and
sodium hydride (1.00 g, 41.6 mmol) was added in portions. The mixture
was allowed to stir for 1 h at RT. Bromododecane (3.98 mL, 16.0 mmol)
was then added dropwise by means of a syringe and the mixture was al-
lowed to stir for 5 h at 408C, followed by ~20 h at 308C. Methanol was
added to deactivate any remaining NaH. The solvent was then evaporat-
ed, the residue was dissolved in CH2Cl2 and the organic phase was
washed two times with water. Hexane was then added to the CH2Cl2
fraction and the product precipitated out upon removal of the CH2Cl2.
The precipitate was filtered and reprecipitated again by using the same
procedure. Column chromatography (silica gel; 99:1 CH2Cl2/MeOH) of
the resulting crude product afforded pure dodecyl-AAn as a white solid
(1.50 g, 40%). M.p. 978C; Rf =0.62 (95:5 CH2Cl2/CH3OH); 1H NMR
(CDCl3, 200 MHz): d=8.92 (brs, 1H; NH), 8.80 (s, 1H; adenine H-2),
8.15 (d, 2H; anis-ArH), 8.00 (s, 1H; adenine H-8), 7.02 (d, 2H; anis-
ArH), 4.27 (t, 2H; alkyl 1CH2), 3.89 (s, 3H; anis-OMe), 2.05–1.82 (m,
2H; alkyl 2CH2), 1.42–1.12 (m, 8H; alkyl (3–11)CH2), 0.80–0.95 (t, 3H;
alkyl CH3); 13C NMR (CDCl3, 75 MHz): d=164.5, 163.4, 152.7, 152.3,
142.9, 130.2, 126.2, 123.2, 114.2, 55.7, 44.4, 32.2, 30.2, 29.8, 29.7, 29.6, 29.5,
29.3, 26.9, 22.9, 14.4 ppm; MS (MALDI-TOF): m/z: 439 [M+H]+.
4-(1-Nonyloxy-6-(thymine))-1-bis(phenylethynyl)benzene (T1bT): The re-
action was conducted according to the general procedure described
above, using T4b (2.05 g, 4.36 mmol), and afforded
a yellow solid.
Column chromatography (silica gel; 100:0, 99:1, 98:2 CH2Cl2/MeOH) of
the resulting crude product afforded pure T1bT as a yellow solid (1.02 g,
64%). 1H NMR (CDCl3, 600 MHz): d=8.32 (s, 2H; NH), 7.47 (s, 4H;
ArH), 7.46 (d, 4H; ArH), 6.97 (s, 2H; T-CH), 6.86 (d, 4H; ArH), 3.98 (t,
4H; OCH2), 3.71 (t, 4H; NCH2), 1.93 (s, 6H; T-CH3), 1.84–1.65 (m, 8H;
OCH2CH2 and NCH2CH2), 1.60–1.35 ppm (m, 20H; 10”alkyl CH2);
13C NMR (CDCl3, 75 MHz): d=164.5, 159.6, 151.1, 140.7, 133.3, 131.6,
123.4, 115.2, 114.9, 110.8, 91.5, 88.2, 68.3, 48.8, 29.6, 29.5, 29.3, 26.6, 26.2,
12.6 ppm; FTIR (annealed film on CaF2) n˜ =3165, 3037, 2927, 2854, 2215,
1683, 1519, 1486, 1486, 1356, 1246, 1173 cmꢀ1; UV: lmax =281, 333, 353,
377 nm; photoluminescence (excitation at 340 nm, CHCl3): 367, 385;
DSC (Tm =2078C); MS (MALDI-TOF): m/z: 810 [M+].
4-(1-Hexyloxy-6-(N6-(4-methoxybenzoyl)adenine)-1-bis(phenylethynyl)-
benzene (AAn1aAAn): The reaction was conducted according to the gener-
al procedure described above, using AAn4a (2.49 g, 4.36 mmol), and af-
forded a yellow solid. Column chromatography (silica gel; 99:1, 98:2!
95:5 CH2Cl2/MeOH) of the resulting crude product afforded pure
AAn1aAAn as a yellow solid (1.14 g, 57%). 1H NMR (CDCl3, 600 MHz):
d=9.09 (brs, 2H; NH), 8.76 (s, 2H; adenine H-8), 8.01(s, 2H; adenine
H-2), 8.00 (d, 4H; anis-ArH), 7.44 (d, 4H; ArH), 7.43 (s, 4H; ArH), 6.98
(d, 4H; anis-ArH), 6.83 (d, 4H; ArH), 4.27 (t, 4H; NCH2), 3.94 (t, 4H;
OCH2), 3.87 (s, 6H; anis-OCH3), 2.05–1.88 (m, 4H; NCH2CH2), 1.88–
1.65 (m, 4H; OCH2CH2), 1.55–1.25 ppm (m, 8H; 4”alkyl CH2);
13C NMR (DMSO at 708C, 75 MHz): d=165.8, 163.3, 159.9, 153.1, 151.9,
150.0, 145.0, 133.7, 132.0, 131.2, 126.7, 125.9, 123.3, 115.7, 114.8, 114.4,
114.5, 92.3, 88.4, 68.5, 56.3, 44.0, 30.2, 29.9, 29.1, 26.5, 25.7 ppm; FTIR
(annealed film on CaF2): n˜ =3411, 3239, 3071, 3041, 2936, 2862, 2214,
1700, 1696, 1685, 1602, 1577, 1517, 1452, 1404, 1306, 1284, 1241, 1171,
1089, 1025 cmꢀ1; UV: lmax =291, 331, 353, 378 nm; photoluminescence
(excitation at 340 nm, CHCl3): 366, 384; DSC (Tm =1888C); MS
(MALDI-TOF): m/z: 1012 [M+].
4-(1-Hexyloxy-6-(N6-(4-methoxybenzoyl)adenine)-1-bis(phenylethynyl)-
benzene (AAn1bAAn): The reaction was conducted according to the gener-
al procedure described above, using AAn4b (2.67 g, 4.36 mmol), and af-
forded a yellow solid. Column chromatography (silica gel; 1:99, 2:98!
4:96 MeOH/CH2Cl2) of the resulting crude product afforded pure
AAn1bAAn as a yellow solid (1.09 g, 50%). 1H NMR (CDCl3, 200 MHz):
d=9.10 (brs, 2H; NH), 8.78 (s, 2H; adenine H-8), 8.00 (d, 4H; anis-
ArH), 7.98 (s, 2H; adenine H-2), 7.45 (d, 4H; ArH), 7.44 (s, 4H; ArH),
6.97 (d, 4H; anis-ArH), 6.85 (d, 4H; ArH), 4.26 (t, 4H; NCH2), 3.95 (t,
4H; OCH2), 3.87 (s, 6H; anis-OCH3), 2.01–1.85 (m, 4H; NCH2CH2),
1.88–1.65 (m, 4H; OCH2CH2), 1.50–1.25 ppm (m, 20H; 10”alkyl CH2);
13C NMR (CDCl3, 50 MHz): d=165.7, 164.8, 160.8, 154.0, 153.6, 151.2,
144.3, 134.7, 134.6, 132.8, 131.5, 127.4, 124.6, 124.5, 116.5, 116.1, 115.8,
115.7, 92.8, 89.5, 69.5, 57.0, 45.7, 31.5, 30.8, 30.7, 30.7, 30.5, 28.2,
27.5 ppm; FTIR (annealed film on CaF2): n˜ =3416, 3257, 3071, 3041,
2927, 2854, 2215, 1699, 1684, 1603, 1577, 1517, 1456, 1312, 1284, 1244,
1172, 1090, 1024 cmꢀ1; UV: lmax =290, 332, 352, 378 nm; photolumines-
cence (excitation at 340 nm, CHCl3): 367, 385; DSC (Tm =2028C); MS
(MALDI-TOF): m/z: 1098 [M+].
4-(1-Hexyloxy-6-(N3-methyl)-thymine)-1-iodobenzene (TMe4a): Methyl
iodide (174 ml, 0.28 mmol) was added in one portion to a stirred solution
of DMF (3 mL), K2CO3 (0.19 g, 0.14 mmol), and T4a (0.30 g, 0.70 mmol)
at 608C. After 16 h, the reaction was poured into water and extracted
with CH2Cl2. The combined organic layers were washed with water and
then the CH2Cl2 was evaporated. Column chromatography (silica gel;
99:1 CH2Cl2/MeOH) of the resulting crude product afforded pure TMe4a
(0.27 g, 88%). M.p. 868C; 1H NMR (CDCl3, 300 MHz): d=7.55 (d, 2H;
ArH), 6.96 (s, 1H; T-CH), 6.65 (d, 2H; ArH), 3.91 (t, 2H; OCH2), 3.73
N3-Dodecyl-thymine (dodecyl-T): Thymine (1.00 g, 7.93 mmol) was
added to dry DMF (14 mL), and NaH (0.38 g, 15.8 mmol) was then
added in portions to the mixture. The solution was thermostated at
1008C for 4 h and then allowed to cool to RT. Bromododecane (4.37 mL,
18.2 mmol) was then added dropwise by means of a syringe, and the mix-
ture was allowed to stir for 48 h at 1008C. The reaction mixture was then
removed from the heat and neutralized with ~2 mL of acetic acid. The
solvent was evaporated in vacuo and the residue was dissolved in chloro-
form. Any remaining undissolved material was filtered off and the solu-
ble fraction was then washed with water (100 mL). The separated chloro-
form was then stirred with sodium sulfate for 15 min and filtered.
Hexane was then added to the CH2Cl2 fraction and the product precipi-
tated out upon removal of the CH2Cl2. Column chromatography (silica
gel; 99:1 CH2Cl2/MeOH) of the resulting crude product afforded pure
dodecyl-T as a white solid (0.72 g, 30.7%). M.p. 122.18C; Rf =0.51 (95:5
CH2Cl2/CH3OH); 1H NMR (CDCl3, 200 MHz): d=8.42 (brs, 1H; NH),
1
6.97 (s, 1H; T-CH), 3.70 (t, 2H; alkyl CH2), 1.95 (s, 3H; T-CH3), 1.75–
1.61 (m, 2H; alkyl 2CH2), 1.40–1.21 (m, 8H; alkyl (3–11)CH2), 1.95–
1.85 ppm (m, 3H; alkyl 12CH3); 13C NMR (CDCl3, 50 MHz): d=166.2,
152.7, 142.0, 112,1, 50.1, 33.4, 31.1, 31.0, 30.9, 30.8, 30.7, 28.0, 24.3, 15.7,
14.0 ppm; MS (MALDI-TOF): m/z: 295 [M+H]+.
Chem. Eur. J. 2006, 12, 446 – 456
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
455

S. J. Rowan et al.
[9] M. J. Brienne, J. Gabard, J.-M. Lehn, I. Stibor, J. Chem. Soc. Chem.
Commun. 1989, 1868–1870.
Acknowledgements
[10] a) J.-M. Lehn, Polym. Int. 2002, 51, 825–839; b) T. Kato, N. Mizoshi-
ta, K. Kanie, Macromol. Rapid Commun. 2001, 22, 797–814;
c) C. M. Paleos, D. Tsiourvas, Liq. Cryst. 2001, 28, 1127–1161; d) A.
Ciferri, Liq. Cryst. 1999, 26, 489–494.
We thank the NIH (NIBIB: EB-001466–01), and the Case School of En-
gineering for helping to fund this work. S.J.R. acknowledges the NSF for
a CAREER Award (CHE-0133164). P.T.M. acknowledges NSF for a
CAREER Award (CTS-00093880). We also thank Professor C. Weder
for useful discussions.
[11] a) T. Gulik-Krzywicki, C. Fouquey, J.-M. Lehn, Proc. Natl. Acad.
Sci. USA 1993, 90, 163–167; b) P. Bladon, A. C. Griffin, Macromole-
cules 1993, 26, 6604–6610; c) M. Kotera, J.-M. Lehn, J.-P. Vigneron,
J. Chem. Soc. Chem. Commun. 1994, 197–199; d) M. Kotera, J.-M.
Lehn, J.-P. Vigneron, Tetrahedron 1995, 51, 1953–1972; e) C. St.
Pourcain, A. C. Griffin, Macromolecules 1995, 28, 4116–4121.
[12] S. Sivakova, S. J. Rowan, Chem. Commun. 2003, 2428–2429.
[13] For examples of the use of DNA sequences in supramolecular poly-
merization processes, see: a) J. Xu, E. A. Fogleman, S. L. Craig,
Macromolecules 2004, 37, 1863–1870; b) F. R. Kersey, G. Lee, P.
Marszalek, S. L. Craig, J. Am. Chem. Soc. 2004, 126, 3038–3039;
c) E. A. Fogleman, W. C. Yount, J. Xu, S. L. Craig, Angew. Chem.
2002, 114, 4198–4200; Angew. Chem. Int. Ed. 2002, 41, 4026–4028;
d) S. M. Waybright, C. P. Singleton, K. Wachter, C. J. Murphy,
U. H. F. Bunz, J. Am. Chem. Soc. 2001, 123, 1828–1833.
[14] For examples, see: a) W. Saenger, Principles of Nucleic Acid Struc-
tures, Springer, New York, 1984; b) S. Sivakova, S. J. Rowan, Chem.
Soc. Rev. 2005, 34, 9–21.
[1] a) L. Brunsveld, B. J. B. Folmer, E. W. Meijer, R. P. Sijbesma, Chem.
Rev. 2001, 101, 4071–4097; b) A. Ciferri, Macromol. Rapid
Commun. 2002, 23, 511–529; c) Supramolecular Polymers (Ed.: A.
Ciferri), Marcel Dekker, New York, 2000.
[2] a) Z. Liang, O. M. Cabarcos, D. L. Allara, Q. Wang, Adv. Mater.
2004, 16, 823–827; b) H. Kosonen, J. Ruokolainen, M. Knaapila, M.
Torkkeli, K. Jokela, R. Serimaa, G. ten Brinke, W. Bras, A. P. Monk-
man, O. Ikkala, Macromolecules 2000, 33, 8671–8675; c) A. Calzo-
lari, R. Di Felice, E. Molinari, A. Garbesi, Phys. E 2002, 13, 1236–
1239; d) G. N. Tew, M. U. Pralle, S. I. Stupp, Angew. Chem. 2000,
112, 527–531; Angew. Chem. Int. Ed. 2000, 39, 517–521; e) X. L.
Chen, S. A. Jenekhe, Macromolecules 2000, 33, 4610–4612; f) J. B.
Beck, S. J. Rowan, J. Am. Chem. Soc. 2003, 125, 13922–13923.
[3] a) T. Kato, Supramol. Science, 1996, 3, 53–59; b) T. Kato, J. M. J.
Frꢁchet, Macromol. Symp. 1995, 98, 311–326; c) C. M. Lee, A. C.
Griffin, Macromol. Symp. 1997, 117, 281–290; d) T. Kato, J. M. J.
Frꢁchet, J. Am. Chem. Soc. 1989, 111, 8533–8534; e) T. Kato in
Handbook of Liquid Crystals (Eds.: D. Demus, J. W. Goodby, G. W.
Gray, H.-W. Spiess, V. Vill), Wiley-VCH, Weinheim, 1998, Vol. 2B,
pp. 969–979; f) “Hydrogen-Bonded Liquid Crystals: Molecular Self-
Assembly for Dynamically Functional Materials”, T. Kato, Struct.
Bonding (Berlin) 2000, 96, 95–146; g) Y. Kamikawa, M. Nishii, T.
Kato, Chem. Eur. J. 2004, 10, 5942–5951.
[4] a) J.-M. Lehn, Makromol. Chem. Macromol. Symp. 1993, 69, 1–17;
b) C. M. Paleos, D. Tsiourvas, Angew. Chem. 1995, 107, 1839–1855;
Angew. Chem. Int. Ed. Engl. 1995, 34, 1696–1711; c) N. Zimmer-
man, J. S. Moore, S. C. Zimmerman, Chem. Ind. (London) 1998, 15,
604–610; d) M. Muthukumar, C. K. Ober, E. L. Thomas, Science
1997, 277, 1225–1232; e) Struct. Bonding (Ed.: M. Fujita), 2000, 96,
whole volume; f) J.-M. Lehn, Supramolecular Chemistry, VCH,
Weinheim, 1995; g) A. Treybig, C. Dorscheid, W. Weissflog, H.
Kresse, Mol. Cryst. Liq. Cryst. 1995, 260, 369–376; h) S. I. Torgova,
A. Strigazzi, Mol. Cryst. Liq. Cryst. 1999, 336, 229–245; i) V. Percec,
J. Heck, G. Johansson, D. Tomazos, M. Kawasumi, P. Chu, G.
Ungar, Mol. Cryst. Liq. Cryst. Sci. 1994, 254, 137–196; j) V. Percec,
Macromol. Symp. 1997, 117, 267–273; k) G. P. Spada, G. Gottarelli,
Synlett 2004, 4, 596–602.
[15] K. Yamauchi, J. R. Lizotte, T. E. Long, Macromolecules 2002, 35,
8745–8750.
[16] S. J. Rowan, P. Suwanmala, S. Sivakova, J. Polym. Sci. Part A:
Polym. Chem. 2003, 41, 3589–3596.
[17] W. H. Binder, M. J. Kunz, C. Kluger, G. Hayn, R. Saf, Macromole-
cules 2004, 37, 1749–1759.
[18] C. M. Paleos, J. Michas, Liq. Cryst. 1992, 28, 773–778.
[19] a) J. T. Davis, Angew. Chem. 2004, 116, 684–716; Angew. Chem. Int.
Ed. 2004, 43, 668–698; b) T. Giorgi, F. Grepioni, I. Manet, P. Maria-
ni, S. Masiero, E. Mezzina, S. Pieraccini, L. Saturni, G. P. Spada, G.
Gottarelli, Chem. Eur. J. 2002, 8, 2143–2152, and references therein.
[20] For other studies of the interactions of N6-substituted adenine deriv-
atives, see: a) J. S. Nowick, T. Cao, G. Noronha, J. Am. Chem. Soc.
1994, 116, 3285–3289; b) J. S. Nowick, J. S. Chen, G. Noronha, J.
Am. Chem. Soc. 1993, 115, 7636–7644; c) J. S. Nowick, J. S. Chen, J.
Am. Chem. Soc. 1992, 114, 1107–1108.
[21] J. Sartorius, H.-J. Schneider, Chem. Eur. J. 1996, 2, 1446.
[22] R. K. Castellano, V. Gramlich, F. Diederich, Chem. Eur. J. 2002, 8,
118–129.
[23] A. R. A. Palmans, M. Eglin, A. Montali, C. Weder, P. Smith, Chem.
Mater. 2000, 12, 472–480.
[24] G. W. Gray, J. W. G. Goodby, Smectic Liquid Crystals: Textures and
Structures, Heyden & Son, Philadelphia, 1984.
[5] a) L. Y. Park, D. G. Hamilton, E. A. McGehee, K. A. McMenimen,
J. Am. Chem. Soc. 2003, 125, 10586–10590; b) K. A. McMenimen
D. G. Hamilton, J. Am. Chem. Soc. 2001, 123, 6453–6454; c) R. K.
Castellano, R. Clark, S. L. Craig, C. Nuckolls, J. Rebek, Proc. Natl.
Acad. Sci. USA 2000, 97, 12418–12421; d) M. L. Bushey, T.-Q.
Nguyen, W. Zhang, D. Horoszewski, C. Nuckolls, Angew. Chem.
2004, 116, 5562–5570; Angew. Chem. Int. Ed. 2004, 43, 5446–5453.
[6] J. M. Lehn, M. Mascal, A. DeCian, J. Fischer, J. Chem. Soc. Chem.
Commun. 1990, 479–481.
[25] J. Y Lee, P. C. Painter, M. M. Coleman, Macromolecules 1988, 21,
954–960.
[26] J. B. Lambert, H. F. Shurvell, D. A. Lightner, R. G. Cooks, Organic
Structural Spectroscopy, Prentice Hall, New Jersey, 1998.
[27] For a thermosensitive reversible polymer at the covalent level, see:
a) X. Chen, M. A. Dam, K. Ono, A. Mal, H. Shen, S. R. Nut, K.
Sheran, F. Wudl, Science 2002, 295, 1698–1702; b) X. Chen, F. Wudl,
A. Mal, H. Shen, S. R. Nut, Macromolecules 2003, 36, 1802–1807.
[28] L. Kosynkina, T. Wei Wang, C. Liang, Tetrahedron Lett. 1994, 35,
5173–5176.
[7] C. Hilger, R. Stadler, Polymer 1991, 32, 3244–3249.
[8] C. He, A. M. Donald, A. C. Griffin, T. Waigh, A. H. Windle, J.
Polym. Sci. Part B: Polym. Phys. 1998, 36, 1617–1624.
Received: July 15, 2005
Published online: October 20, 2005
456
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 446 – 456