STM insight into hydrogen-bonded bicomponent 1D supramolecular polymers with controlled geometries at the liquid-solid interface

Article abstract of DOI:10.1002/anie.200805680

(Chemical Equation Presented) Bicomponent supramolecular polymers, consisting of two alternating molecules bridged through six H-bonds, are observed by STM at the solid-liquid interface. Control of the geometry of the 1D architecture was obtained by using

Full text of DOI:10.1002/anie.200805680

Angewandte
Chemie
DOI: 10.1002/anie.200805680
Supramolecular Polymers
STM Insight into Hydrogen-Bonded Bicomponent 1D Supramolecular
Polymers with Controlled Geometries at the Liquid–Solid Interface**
Artur Ciesielski, Gaꢀl Schaeffer, Anne Petitjean, Jean-Marie Lehn,* and Paolo Samorꢁ*
Self-assembly of molecular species into well-defined multi-
component supramolecular architectures[1] can be obtained
by selective association of complementary building blocks
undergoing molecular recognition events. Among weak
interactions, hydrogen bonding offers a high level of control
over the process of molecular self-assembly, since it combines
reversibility, directionality, specificity, and cooperativity. It
has in particular been implemented for the generation of
supramolecular polymers[2] through the polyassociation of
molecular components bearing complementary recognition
groups. Supramolecular polymers can simultaneously display
tunable material properties with low-viscosity melts.[3] Their
mechanical features result especially from secondary inter-
actions, in particular the strength, reversibility, and direction-
ality of those interactions. Hitherto, H-bonded supramolec-
ular architectures have been self-assembled on solid surfaces
into highly ordered motifs, both under ultrahigh vacuum
(UHV)[4] and at the solid–liquid interface.[5] Whereas for the
former case the formation of multicomponent 1D H-bonded
polymers was recently reported,[6] for the latter, the effort has
to date been primarily devoted to H-bonded architectures
consisting of multicomponent discrete assemblies[7] or to
monocomponent 1D polymers.[8] The generation of ordered
motifs stabilized by hydrogen bonds on a solid surface
requires the fine tuning of the interplay between the
interactions among neighboring molecules and the adsor-
bate–substrate interactions.[9] The controlled formation of
ordered multicomponent architectures at the solid–liquid
interface from a concentrated solution is thermodynamically
unfavored. In fact, among the various components in the
supernatant solution, the component with a greater affinity
for the substrate, that is, offering a minimization of the free
interface energy per unit area, will assemble on its surface,
whereas the others will remain in solution.[10] To immobilize
all the components on the surface, thus to achieve a complete
physisorption of all the components at the solid–liquid
interface, it is necessary to borrow a strategy commonly
employed under UHV, that is, control of the stoichiometry of
the molecules absorbed at surfaces.[11] At the solid–liquid
interface, the number of molecules in the solution applied to
the surface should be lower than that required to form a
monolayer of physisorbed molecules lying flat on the basal
plane of the substrate. However, operating under such
conditions, that is, at low concentration, can lead to poly-
morphism.[12] Although the use of H-bonds to form bicompo-
nent linear supramolecular polymers was introduced over
15 years ago,[2e–f] to date, visualization by scanning tunneling
microscopy (STM) has not been explored at the liquid–solid
interface for supramolecular polymers composed of two
components interacting through multiple hydrogen bonds,
leading to an architecture with a controlled geometry and, in
principle, an infinite length. STM at the solid–liquid interface
has been the method of choice for the visualization of
monolayers as it offers detailed submolecular-scale insight
into the self-assembly of organic molecules on surfaces and its
evolution over time with millisecond resolution.[13]
Herein we present STM visualization at the liquid–solid
interface at room temperature of the formation of supra-
molecular hydrogen-bonded polymers with either a linear or
a zigzag geometry on highly oriented pyrolitic graphite
(HOPG) surfaces. To this end, we used the ditopic molecular
components 1–3 (Scheme 1), bearing complementary hydro-
gen-bonding recognition groups: either a Janus-type cyanuric
wedge 2 or barbituric wedge 3 (ADA-ADA-array) and a
corresponding (DAD-DAD-array) receptor unit 1.[14] These
molecules were selected because they are known to form 1D
supramolecular polymeric strands in solution by polyassoci-
ation through sextuple hydrogen bonding between their
respective recognition sites[15] (Figure 1).
Initially we investigated the homomolecular self-assem-
bled structures obtained by applying a drop of a (6 Æ 1) mm
solution of the molecule 1[16] in 1,2,4-trichlorobenzene (TCB)
on the graphite surface. STM images of the obtained
monolayer (Figure 2) show a polycrystalline structure with
two different domains A and B, which coexist for a several
minutes (see survey STM image in the Supporting Informa-
tion, Figure SI3). The two crystals are characterized by a
different packing motif. The assembly in domain A presents
parallel lamellae (Figure 2a) and has the unit cell: a = (1.78 Æ
0.2) nm, b = (5.62 Æ 0.2) nm, a = (90 Æ 2)8, corresponding to
an area A = (10.0 Æ 1.2) nm2, given that each unit cell contains
one molecule 1. Given its size there is space enough to
accommodate all the aliphatic side groups on the surface,
although, since we have not resolved them, we do not have
unambiguous confirmation of this aspect. It is interesting to
[*] A. Ciesielski, G. Schaeffer, Dr. A. Petitjean,[+] Prof. J.-M. Lehn,
Prof. P. Samorꢀ
ISIS/UMR CNRS 7006, Universitꢁ Louis Pasteur
8 Allꢁe Gaspard Monge, 67000 Strasbourg (France)
E-mail: lehn@isis.u-strasbg.fr
samori@isis-ulp.org
[+] Present address: Chemistry Department, Queen’s University
Kingston, ON, K7L 3N6 (Canada)
[**] This work was supported by the EU through the Marie Curie EST
project SUPER (MEST-CT-2004-008128) and RTN PRAIRIES
(MRTN-CT-2006-035810), as well as by the ERA-Chemistry project
SurConFold, the Universitꢁ Louis Pasteur and CNRS. A.C. and G.S.
thank the French Ministry of Research for a predoctoral fellowship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200805680.
Angew. Chem. Int. Ed. 2009, 48, 2039 –2043
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2039

Communications
Scheme 1. Ditopic molecular components bearing hydrogen-bonding
recognition groups.
Figure 1. Formation of a main-chain supramolecular polymer by poly-
association of molecular monomers of type 1 and 2 or 3 through
sextuple hydrogen bonding between their complementary recognition
units.
point out that the lifetime of domain A was very short, that is,
it disappeared after a few minutes, proving its metastable
nature. The poor stability of domain A was also evidenced by
its high molecular dynamics on a timescale comparable to that
of the tip scanning. For this reason we have not been able to
achieve a high submolecular resolution. Such poor stability is
not surprising. For enthalpy reasons, the system is prone to
form a densely packed assembly, thus minimizing the size of
the unit cell. The more stable domain B features the unit cell:
a = (4.42 Æ 0.2) nm, b = (5.65 Æ 0.2) nm, a = (90 Æ 2)8, leading
to an area A = (24.9 Æ 1.4) nm2, where each unit cell contains
four molecules 1 (Figure 2b). Among them, one molecule
(indicated by a white arrow) seems to be partially desorbed.
Given the area of the darker spots in the STM image As1 =
(11 Æ 2) nm2 (see the Supporting Information), and by
comparing it with the area occupied by a 2D projected
C9H19 chain (AC H = (0.33 Æ 0.02) nm2), it is likely that all
Figure 2. STM images of a monolayer of molecule 1 at the liquid-
graphite interface self-assembled from a solution 1,2,4-trichloroben-
zene: a) Domain A. Tunneling parameters: average tunneling current
(It)=0.5 pA, bias voltage (Vt)=100 mV; b) domain B. Tunneling
parameters: It =15 pA, Vt =300 mV; average area of the darker spots
A
s1 =(11Æ2) nm2; c) model of the molecular components in
domain A; d) model of the molecular components in domain B.
fourteen nonyl side chains are physisorbed on the HOPG
surface, although, owing to their highly dynamic nature, we
are unable to resolve them.
We then extended our studies to heteromolecular non-
covalently engineered main-chain supramolecular polymers
resulting from self-assembly on graphite from a bicomponent
9
19
2040 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 2039 –2043

Angewandte
Chemie
solution containing mole-
cule 1[16] and either mole-
cule 2 or 3.[17] It is important
to note that molecule 1 was
visualized at the HOPG–
solution interface only
upon using 1,2,4-trichloro-
benzene (TCB) as solvent.
Study of this system in dif-
ferent solvents, 1-phenyloc-
tane and tetradecane, did
not produced any ordered
monolayers.
For
this
reason, we continued our
study using TCB as a sol-
vent. Neither molecule 2
nor 3 were found to form
ordered structures in single-
component films at the
liquid–solid interface, high-
lighting a low affinity for
the graphite surface. The
heteromolecular
mono-
layer was formed by apply-
ing a 4 mL drop of the
solution in TCB to the sur-
face. The two bicomponent
solutions were obtained by
mixing 5–10 mm solutions
of 1 + 2 or 1 + 3 in DMSO
and subsequently diluting
them with TCB, to yield
concentrations of (4 Æ
1) mm and (3 Æ 1) mm for
1 + 2 and 1 + 3, respectively.
The procedure for the dep-
osition of the two compo-
nents on the surface is crit-
ical for the formation of the
mixed polymer, as, if one
partner molecule is already
physisorbed on the surface,
its partial desorption is
Figure 3. STM images of monolayers of linear bicomponent supramolecular H-bonded polymer at the liquid–
graphite interface self-assembled from TCB solution; a) linear structure of polymer 1+2; area of darker spots
As =(1.3Æ2) nm2. Tunneling parameters: It =20 pA, Vt =300 mV; b) self-assembled bicomponent polymer
1+3; the area of the darker spots As =(4Æ1) nm2. Tunneling parameters: It =20 pA, Vt =300 mV; c) linear
polymer 1+2; d) linear polymer 1+3.
energetically unfavorable, thereby hindering the emergence
of molecular recognition leading to homogeneous intermixing
on the substrate surface. Therefore both 1 + 2 and 1 + 3
solutions were prepared ex situ. Notably, the formation of
bicomponent polymers was visualized only when the ratio of
the concentrations of the molecules was (30 Æ 5)% of
molecule 1 to (70 Æ 5)% of molecule 2 and 3.
that are stable over several minutes. The proposed packing
motif is in excellent agreement with that suggested by
modeling (Figure 1). The symmetric molecule 1 forms six
hydrogen bonds with each neighboring molecule 2. The
formation of a linear architecture is made possible especially
taking advantage of the conformational flexibility of the
cyanuric wedge 2. The core of each molecule is physisorbed
flat on the surface. The unit cell parameters, a = (2.72 Æ
0.2) nm, b = (3.45 Æ 0.2) nm, and a = (35.8 Æ 2)8, lead to an
area A = (5.5 Æ 0.6) nm2/dimer, where each unit cell contains
one molecule 1 and one molecule 2 (Figure 3a). The proposed
supramolecular motif, featuring a bicomponent linear associ-
ation 1 + 2 (Figure 3c), matches the pattern observed in the
STM image, ruling out the presence of the energetically
equivalent armchair-based motif. Deposition of a small
amount (ca. 4 mL) of molecule 1 and of the bridging molecule
By decreasing the concentration of molecules 2 and 3 or
increasing the concentration of molecule 1, only patterns of
molecule 1 were seen on the HOPG surface. This observation
confirms the greater affinity of molecule 1 for the graphite
surface.
Figure 3a shows a STM image of the linear heteromolec-
ular polymer resulting from deposition of a mixture of
molecules 1 and 2. The monolayer displays a polycrystalline
structure, which consists of tens of nanometer-wide domains
Angew. Chem. Int. Ed. 2009, 48, 2039 –2043
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2041

Communications
3 in a 3:7 stoichiometry results also in the formation of a 1D
introduction of cross-linking components[15] may allow exten-
sion into 2D supramolecular assemblies.
heteromolecular polymer. However, differently from the
linear motif observed for 1 + 2, the bicomponent supramolec-
ular polymer has an intralamellar zigzag geometry. The angle
b between the two adjacent molecular units in the 1D
supramolecular polymer is (90 Æ 2)8. This change in packing
can be ascribed to the greater rigidity of molecule 3 compared
to 2. To enable heteromolecular association of two edges,
molecule 1 needs to adopt a bent conformation, which is
possible thanks to a flexible central -OC3H6O- moiety. The
distance between the two parallel lamellae, that is, between
two adjacent bright structures in the STM image, corresponds
to the length of the C9H19 chains, indicating that, in this case,
all the C9H19 chains of molecule 1 are physisorbed on the
HOPG surface. However, owing to dynamics on a faster
timescale than the tip scanning, we were not able to resolve
them (Figure 3b). The darker areas in the STM current
images (Figure 3a and b) were estimated and compared with
the space that would be occupied by the nonyl chains
physisorbed on the graphite. These areas were found to be
Received: November 20, 2008
Revised: December 21, 2008
Published online: January 29, 2009
Keywords: interfaces · polymers · scanning probe microscopy ·
.
self-assembly · supramolecular chemistry
[1] J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspec-
tives, VCH, New York, 1995.
[2] a) Supramolecular Polymers, 2nd ed. (Ed.: A. Ciferri), Taylor &
Francis, Boca Raton, FL, 2005; b) L. Brunsveld, B. J. B. Folmer,
E. W. Meier, R. P. Sijbesma, Chem. Rev. 2001, 101, 4071; c) J.-M.
Lehn, Polym. Int. 2002, 51, 825; d) J.-M. Lehn, Prog. Polym. Sci.
2005, 30, 814; e) C. Fouquey, J.-M. Lehn, A.-M. Levelut, Adv.
Mater. 1990, 2, 254; f) T. Gulik-Krzywicki, C. Fouquey, J.-M.
Lehn, Proc. Natl. Acad. Sci. USA 1993, 90, 163.
[3] T. F. A. de Greef, E. W. Meijer, Nature 2008, 453, 171.
[4] a) J. V. Barth, Annu. Rev. Phys. Chem. 2007, 58, 375; J. V. Barth,
G. Costantini, K. Kern, Nature 2005, 437, 671; b) F. Rosei, M.
Schunack, Y. Naitoh, P. Jiang, A. Gourdon, E. Laegsgaard, I.
Stensgaard, C. Joachim, F. Besenbacher, Prog. Surf. Sci. 2003, 71,
95.
[5] S. De Feyter, F. C. De Schryver, Chem. Soc. Rev. 2003, 32, 139.
[6] a) A. Llanes-Pallas, M. Matena, T. Jung, M. Prato, M. Stꢀhr, D.
Bonifazi, Angew. Chem. 2008, 120, 7840; Angew. Chem. Int. Ed.
2008, 47, 7726; b) M. E. Caꢁas-Ventura, W. Xiao, D. Wasser-
fallen, K. Mꢂllen, H. Brune, J. V. Barth, R. Fasel, Angew. Chem.
2007, 119, 1846; Angew. Chem. Int. Ed. 2007, 46, 1814.
[7] a) S. De Feyter, A. Miura, S. Yao, Z. Chen, F. Wꢂrthner, P.
Jonkheijm, A. P. H. J. Schenning, E. W. Meijer, F. C.
De Schryver, Nano Lett. 2005, 5, 77; b) K. Eichhorst-Gerner,
A. Stabel, G. Moessner, D. Declerq, S. Valiyaveettil, V.
Enkelmann, K. Mꢂllen, J. P. Rabe, Angew. Chem. 1996, 108,
1599 – 1602; Angew. Chem. Int. Ed. Engl. 1996, 35, 1492; c) A.
Miura, Z. J. Chen, H. Uji-i, S. De Feyter, M. Zdanowska, P.
Jonkheijm, A. Schenning, E. W. Meijer, F. Wꢂrthner, F. C.
De Schryver, J. Am. Chem. Soc. 2003, 125, 14968; d) W.
Mamdouh, R. E. A. Kelly, M. D. Dong, L. N. Kantorovich, F.
Besenbacher, J. Am. Chem. Soc. 2008, 130, 695; e) S. De Feyter,
M. Larsson, A. Gesquiꢃre, H. Verheyen, F. Louwet, B. Groe-
nendaal, J. van Esch, B. L. Feringa, F. De Schryver, ChemPhys-
Chem 2002, 3, 966; f) F. Cicoira, C. Santato, F. Rosei, Top. Curr.
Chem. 2008, 285, 203; g) K. G. Nath, O. Ivasenko, J. A. Miwa, H.
Dang, J. D. Wuest, A. Nanci, D. F. Perepichka, F. Rosei, J. Am.
Chem. Soc. 2006, 128, 4212.
[8] a) G. Gottarelli, S. Masiero, E. Mezzina, S. Pieraccini, J. P. Rabe,
P. Samorꢄ, G. P. Spada, Chem. Eur. J. 2000, 6, 3242; b) A.
Gesquiꢃre, M. M. S. Abdel-Mottaleb, S. De Feyter, F. C.
De Schryver, F. Schoonbeek, J. van Esch, R. M. Kellogg, B. L.
Feringa, A. Calderone, R. Lazzaroni, J. L. Brꢅdas, Langmuir
2000, 16, 10385; c) C. Meier, U. Ziener, K. Landfester, P.
Weihrich, J. Phys. Chem. B 2005, 109, 21015; d) H. Zhou, H.
Dang, J. H. Yi, A. Nanci, A. Rochefort, J. D. Wuest, J. Am.
Chem. Soc. 2007, 129, 13774; e) G. P. Spada, S. Lena, S. Masiero,
S. Pieraccini, M. Surin, P. Samorꢄ, Adv. Mater. 2008, 20, 2433.
[9] J. P. Rabe, S. Buchholz, Science 1991, 253, 424.
A
s1+2 = 0.85 nm2 and As1+3 = 2.45 nm2 for 1 + 2 and 1 + 3
respectively, whereas the modeled area occupied by each
C9H19 chain amounts to AC H = (0.33 Æ 0.02) nm2, which
9
19
suggests that, for the monolayer of 1 + 2, most of the chains
are backfolded in the supernatant solution, whereas for 1 + 3
they are packed on the surface. The unit cell parameters for
1 + 3 are a = (3.93 Æ 0.2) nm, b = (4.27 Æ 0.2) nm, a = (75 Æ
2)8, A = (16.2 Æ 1.2) nm2/dimer. Each unit cell contains two
molecules 1 and two molecules 3 (Figure 3a). In the proposed
packing model of the bicomponent linear polymer 1 + 3
(Figure 3d) each molecule 1 forms six hydrogen bonds with
each neighboring molecule 3.
In summary, by working at a low concentration, in
experimental conditions not susceptible to thermodynami-
cally driven phase segregation between two components on
the solid–liquid interface, we have been able for the first time
to visualize, by STM on the molecular scale, the physisorption
of 1D main-chain bicomponent H-bonded supramolecular
polymers at surfaces, owing to appropriate design of comple-
mentary building blocks linked by the formation of six H-
bonds at each node. By using two different connecting
molecules, 2 and 3, featuring different conformational rigidity,
we were able to control the geometry of the linear supra-
molecular polymer. When a flexible component was used to
bridge adjacent molecules of 1, a linear structure was
obtained, whereas when the bridging molecule was rigid, a
zigzag motif was observed. The visualization of bicomponent
supramolecular polymers at the liquid–solid interface paves
the way towards the understanding of the mechanism of their
formation on surfaces.[3] It also adds further weight to the
initial concept of supramolecular polymers and supramolec-
ular polymer chemistry,[2e–f,18] that has been widely imple-
mented, in particular in bulk materials.[2a–d] Furthermore, the
generation of rigid bicomponent supramolecular polymers,
such as 1 + 3, which present a controlled curvature at the
liquid–solid interface represents the first step towards the
nanopatterning of surfaces with multicomponent functional
architectures for exploitation of vectorial properties. The
[10] C. A. Palma, M. Bonini, T. Breiner, P. Samorꢄ, Adv. Mater. 2009,
DOI: 10.1002/adma.200802068.
[11] C. A. Palma, M. Bonini, A. Llanes-Pallas, T. Breiner, M. Prato,
D. Bonifazi, P. Samorꢄ, Chem. Commun. 2008, 5289.
[12] a) S. Lei, K. Tahara, F. C. De Schryver, M. van der Auweraer, Y.
Tobe, S. De Feyter, Angew. Chem. 2008, 120, 3006; Angew.
2042 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 2039 –2043

Angewandte
Chemie
Chem. Int. Ed. 2008, 47, 2964; b) L. Kampschulte, T. L.
Werblowsky, R. S. K. Kishore, M. Schmittel, W. M. Heckl, M.
Lackinger, J. Am. Chem. Soc. 2008, 130, 8502.
[15] V. Berl, M. Schmutz, M. J. Krische, R. G. Khoury, J.-M. Lehn,
Chem. Eur. J. 2002, 8, 1227.
[16] E. Kolomiets, E. Buhler, S. J. Candau, J. M. Lehn, Macro-
molecules 2006, 39, 1173.
[17] a) Z. Wang, L. Wang, X. Zhang, J. Shen, S. Denzinger, H.
Ringsdorf, Macromol. Chem. Phys. 1997, 198, 573; b) B. S. Jursic,
E. D. Stevens, Tetrahedron Lett. 2003, 44, 2203.
[13] a) S. De Feyter, F. C. De Schryver, J. Phys. Chem. B 2005, 109,
4290; b) M. Surin, P. Samorꢄ, Small 2007, 3, 190.
[14] a) S. K. Chang, D. Van Engen, E. Fan, A. D. Hamilton, J. Am.
Chem. Soc. 1991, 113, 7640; b) S. K. Chang, A. D. Hamilton, J.
Am. Chem. Soc. 1988, 110, 1318.
[18] J.-M. Lehn, Makromol. Chem. Macromol. Symp. 1993, 69, 1.
Angew. Chem. Int. Ed. 2009, 48, 2039 –2043
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2043